RISK ASSESSMENT FOR LISTERIA MONOCYTOGENES IN TRADITIONALLY PROCESSED FISH FROM INFORMAL MARKETS IN ACCRA AND TEMA THIS THESIS IS SUBMITTED TO THE UNIVERSITY OF GHANA, LEGON, IN PARTIAL FULFILMENT OF THE REQUIREMENT FOR THE AWARD OF MPHIL FOOD SCIENCE DEGREE BY KENNEDY BOMFEH JULY 2011 i DECLARATION I hereby declare that I carried out this research under supervision in the Department of Nutrition and Food Science, University of Ghana, Legon. Works referred to have been dully acknowledged. ii DEDICATION To Professor Kwaku Tano-Debrah, for the blessing you have been. iii ACKNOWLEDGEMENTS I am grateful to o God, Who gives me life, and the grace to live it in honour of His Name o my supervisors, Prof. Kwaku Tano-Debrah, Dr. Firibu Kwesi Saalia, and Dr. Betty Bediako-Amoa, for guiding me throughout the work o MYROC Group of Companies for the financial support o Amma Konadu Amponsah, for the invaluable assistance o Genevieve Opoku Frema, Joyceline Duah-Barning and Emelia Appah for facilitating the experimental cooking in the challenge test o Dr. Louise Endemaño Walker of the University of California, Santa Barbara, for sharing her research works on fishing activities in Ghana o Safe Food Fair Food Project, within which framework this study was conducted o Dr. Makita Kohei of the Consultative Group on International Agricultural Research (CGIAR, Kenya) for his thoughts on risk modelling o Kristina Rosel of the International Livestock Research Institute (ILRI) and Dr. Elizabeth R. Turkson, Accra Polytechnic for their comments o Mr. George Amponsah Annor for delivering the @Risk 5.5 Software o Lecturers in the Department of Nutrition and Food Science, University of Ghana, for imparting to me the knowledge applied in this study, and for their inputs in the study design o my classmates for contributing to a memorable MPhil Experience o my family for their kind support iv ABSTRACT Traditionally processed fish contributes significantly to food and nutrition security in Ghana. The processing and handling has however been associated with unsanitary and unhygienic practices with documented occurrence of food-borne pathogens. The products are also mainly sold on informal markets, where earlier studies reported the occurrence of Listeria monocytogenes in products sold therein. This study sought to determine the occurrence of L. monocytogenes in traditionally processed (smoked, dried, salted) fish sold on informal markets and to assess the exposure of consumers to the pathogen and the associated risk of illness. The study was based on the Codex Alimentarius protocol for microbial risk assessment. Surveys were conducted on selected traditional processors and consumers to determine processing practices and consumption patterns (frequency and portion sizes) respectively. Samples of traditionally processed fish were procured from some processors and consumer markets in the survey locations for microbial analysis to determine the occurrence and concentrations of L. monocytogenes in the processed fish. Microbial challenge tests were also done by cooking deliberately-contaminated fish for short and long time intervals to determine the survival of the pathogen during domestic cooking. Data from the survey (quantities of fish often consumed) and the laboratory analyses (microbial load) were used to assess the exposure of consumers to the pathogen, and also fitted to parametric (probability) functions to characterize the dose response using Monte Carlo simulations with the @Risk software (version 5.5, Palisade Corporation). Prevalence of L. monocytogenes in the fish products sampled from the markets was high (40-80%). However, the pathogen was not detected in smoked fish sampled immediately after processing, suggesting that post-processing contamination occurred. The concentrations of the pathogen in the products were generally low v (102-3 CFU/g), and decreased from smoked fish through to sundried fish. The pathogen also survived in fish used for the challenge test. The estimated risks of illness were low, ranging from 1 in 100 to 1 in 100,000,000,000 chances of illness. Higher risks of illness were recorded for consumption of smoked fish than for sundried fish and salted fish, in that order. Consumers with high susceptibility to L. monocytogenes infection (elderly, children and pregnant women) were at a greater risk of illness than low risk individuals (non-pregnant adults aged 18 – 39 years). The findings suggest that consumers are exposed to ingesting L. monocytogenes through consumption of traditionally processed fish on informal markets. However the risk of illness is low. Improvements in hygienic processing and post-processing handling of fish as well as proper cooking of the fish products before consumption are recommended. vi TABLE OF CONTENTS DECLARATION...........................................................................................................i DEDICATION..............................................................................................................ii ACKNOWLEDGEMENT..........................................................................................iii ABSTRACT.................................................................................................................iv CHAPTER ONE .......................................................................................................... x 1.0 INTRODUCTION..............................................................................................1 1.1 Background ......................................................................................................... 1 1.2 Listeria monocytogenes ...................................................................................... 2 1.4 Fish as a vehicle for Listeria monocytogenes transmission ................................ 4 1.5 Problem statement ............................................................................................... 5 1.6 Rationale for the study ........................................................................................ 6 1.7 Study objectives .................................................................................................. 7 1.7.1 Main objective ................................................................................................. 7 1.7.2 Specific objectives........................................................................................... 7 CHAPTER TWO ......................................................................................................... 9 2.0 LITERATURE REVIEW ................................................................................... 9 2.1 The Ghana fishery sector .................................................................................... 9 2.1.1 Economic importance of the fishery sector ..................................................... 9 2.1.1.1 Food security ............................................................................................... 9 2.1.1.2 Foreign exchange ....................................................................................... 10 2.1.1.3 Employment............................................................................................... 11 2.2 Traditional fish processing in Ghana ................................................................ 12 2.2.1 Methods of traditional fish processing .......................................................... 14 2.2.1.1 Smoking ..................................................................................................... 14 2.2.1.2 Salting ....................................................................................................... 16 2.2.1.3 Fermentation .............................................................................................. 17 2.2.1.4 Drying ....................................................................................................... 17 2.3 Consumption of traditionally processed fish in Ghana ..................................... 18 2.4 The informal food sector ................................................................................... 18 2.5 Safety of traditionally processed fish on informal markets .............................. 20 2.6 Bacterial pathogens associated with fish .......................................................... 21 vii 2.7 Listeria monocytogenes ................................................................................... 23 2.7.1 Heat resistance of Listeria monocytogenes ................................................... 26 2.7.2 Occurrence of L. monocytogenes .................................................................. 29 2.7.3 Foods frequently contaminated with L. monocytogenes ............................... 30 2.7.4 Persistence of L.a monocytogenes in foods ................................................... 31 2.7.5 Transmission of L. monocytogenes to humans.............................................. 31 2.7.6 The disease caused by Listeria monocytogenes ........................................... 34 2.7.6.1 Foods implicated in listeriosis outbreaks ................................................... 37 2.7.7 Occurrence of Listeria monocytogenes in fish ............................................. 38 2.7.7.1 Factors influencing initial load of Listeria monocytogenes in fish ........... 39 2.8 Regulation of Listeria monocytogenes in foods.............................................. 39 2.9 Detection of L. monocytogenes in foods ........................................................... 42 2.10 Risk assessment ................................................................................................ 43 2.10.1 Microbial risk assessment ......................................................................... 44 2.10.1.1 Sample microbial risk assessments (MRA) ............................................... 45 CHAPTER THREE 3.0 MATERIALS AND METHODS 3.1 Study design ...................................................................................................... 47 3.2. Overview of survey and sampling sites ............................................................ 48 3.2.1 Survey............................................................................................................ 49 3.2.1.1 Consumer survey ....................................................................................... 49 3.2.1.2 Processor survey ........................................................................................ 50 3.3 Sampling for microbiological analyses ............................................................. 50 3.3.1 Sampling of fish from informal markets ....................................................... 51 3.3.2 Sample collection from processing Sites ...................................................... 51 3.3.3 Sample collection for challenge test.............................................................. 53 3.3.4 Sample collection from food vendors ........................................................... 53 3.4 Laboratory microbiological analyses ................................................................ 53 3.4.1 Sample preparation and enrichment for L. monocytogenes detection and enumeration.................................................................................................................. 53 3.4.2 Plating of enriched cultures ........................................................................... 54 3.4.3 Enumeration .................................................................................................. 54 3.4.4 Presumptive and confirmatory identification ................................................ 55 viii 3.5 Survival of L. monocytogenes in domestic cooking (challenge test) ................ 56 3.5.1 Inoculation of fish ........................................................................................ 56 3.5.2 Soup preparation ........................................................................................... 57 3.6 Quality control for microbiological analyses .................................................... 59 3.7 Microbial Risk Assessment Protocols .............................................................. 59 3.7.1 Hazard identification ..................................................................................... 60 3.7.2 Exposure assessment ..................................................................................... 60 3.7.2.1 Prevalence of L. monocytogenes ............................................................... 60 3.7.2.2 Concentration of L. monocytogenes ......................................................... 60 3.7.2.3 Likely intakes of L. monocytogenes .......................................................... 61 3.7.3 Hazard characterization and dose-response assessment................................ 61 3.7.4 Risk characterization ..................................................................................... 62 3.8 Data analyses .................................................................................................... 62 CHAPTER FOUR ...................................................................................................... 63 4.0 RESULTS AND DATA ANALYSIS ............................................................. 63 4.1 Key findings ...................................................................................................... 63 4.2.1 Demographic characteristics of processors ................................................... 64 4.2.2.1 Procurement of fish ................................................................................... 65 4.2.2.2.1 Hygiene of smoking environment .......................................................... 69 4.2.2.3 Sun-drying ................................................................................................. 71 4.2.2.3.1 Sanitary conditions of processing .......................................................... 71 4.2.2.4 Salting ........................................................................................................ 72 4.2.2.4.1 Momoni processing ................................................................................ 73 4.2.2.4.1.1 Hygienic conditions of processing ......................................................... 73 4.2.2.4.2 Kako processing ..................................................................................... 74 4.3 Occurrence of Listeria monocytogenes and other microorganisms in fish during traditional processing ................................................................................................... 75 4.4 Occurrence of Listeria monocytogenes and other microorganisms in fish sold at informal markets .......................................................................................................... 81 4.5 Detection of Listeria monocytogenes in street food samples ............................ 81 4.6 Challenge test .................................................................................................... 83 4.7 Characterization of isolates ............................................................................... 83 4.8 Demographic characteristics of respondents .................................................... 86 ix 4.9 Consumption of traditionally processed fish .................................................... 86 4.9.1 Jamestown ..................................................................................................... 86 4.9.2 Tema New Town ........................................................................................... 89 4.9.3 Madina ........................................................................................................... 89 4.10 Exposure assessment for Listeria monocytogenes ........................................ 92 4.10.1 Risk pathways and event trees .................................................................. 92 4.10.2 Prevalence of Listeria monocytogenes in the fish products ..................... 96 CHAPTER FIVE ..................................................................................................... 111 5.0 DISCUSSION 5.1 Traditional Fish Processing............................................................................. 111 5.1.1 Sanitation ..................................................................................................... 111 5.1.2 Detection of total coliforms and Escherichia coli in fish during traditional processing .................................................................................................................. 112 5.2 General microbiology of fish on informal markets......................................... 114 5.3.2 Exposure assessment ................................................................................... 117 5.3.3 Hazard Characterization (dose-response assessment) ................................. 120 5.3.3.3 Triangular Distributions for Pill ............................................................... 121 5.3.4 Risk characterization ................................................................................... 122 CHAPTER SIX.........................................................................................................122 6.0 CONCLUSIONS AND RECOMMENDATIONS........................................122 6.1 CONCLUSIONS............................................................................................122 6.2 RECOMMENDATIONS...............................................................................123 REFERENCES.........................................................................................................125 APPENDICES..........................................................................................................135 x LIST OF TABLES Table 1.1: Annual outbreaks of food-borne illnesses from selected pathogens 4 Table 2.1: Some bacterial pathogens implicated in food-borne disease outbreaks caused by fish 23 Table 2.2: Differentiating characteristics of Listeria species 24 Table 2.3: Selected findings on the thermal destruction of Listeria Monocytogenes 27 Table 2.4: Some food-borne listeriosis outbreaks 36 Table 3.1: Sampling points along traditional fish processing chain 51 Table 3.2: Kinds and quantities of ingredients used for challenge test cooking 57 Table 4.1: Purchasing of fresh fish for traditional processing 66 Table 4.2: Average microbial counts (x105 CFU/g) and detection of L. monoctyogenes in samples along smoked fish processing chain in Tema New Town 77 Table 4.3: Average microbial counts (x 105 CFU/g) and detection of L. monoctyogenes in samples along salted fish processing chain in Tema New Town 78 Table 4.4: Average microbial counts (x 105 CFU/g) and detection of L. monocytogenes in samples along sun-drying processing chain in Tema New Town 79 Table 4.5: Average microbial counts (x 105 CFU/g) and detection of L. monoctyogenes in samples along smoked fish processing chain in James Town 80 Table 4.6: Detection and average counts of Listeria monocytogenes and average general microbiological counts in fish samples purchased from informal markets in Accra and Tema 82 Table 4.7: Survival of Listeria monocytogenes NCTC 11994 in fish used in experimental domestic cooking 83 Table 4.8: Characteristics of presumptive Listeria isolates 85 Table 4.9: Summary of demographic characteristics of respondents 87 xi Table 4.10: Summaries of highest frequencies of consumption of traditionally processed fish and highest quantities of the products consumed at an instance in Jamestown 88 Table 4.11: Summaries of highest frequencies of consumption of t traditionally processed fish and highest quantities of the products consumed at an instance in Tema New Town 90 Table 4.12: Summaries of highest frequencies of consumption of t traditionally processed fish and highest quantities of the products consumed at an instance in Madina 91 Table 4.13: Average prevalence of L. monocytogenes in traditionally processed fish purchased from some informal markets in Accra and Tema 96 Table 4.14: Likely numbers of Listeria monocytogenes ingested through consumption of traditionally processed fish in Jamestown 99 Table 4.15: Likely numbers of Listeria monocytogenes ingested through consumption of traditionally processed fish in Tema New Town 100 Table 4.16: Likely numbers of Listeria monocytogenes ingested through consumption of traditionally processed fish in Madina 101 Table 4.17: Probability of illness among consumers in Jamestown 104 Table 4.18: Probability of illness among high risk groups in Tema New Town 105 Table 4.19: Probability of illness among high risk groups in Madina 106 Table 4.20: Summary of ranges of probability of illness among consumers (without regard to communities) 107 xii LIST OF FIGURES Fig. 2.1: Ways by which L. monocytogenes is distributed in the environment 32 Fig. 2.2: Potential routes of transmission of L. monocytogenes to humans 33 Fig. 3.1: Flow diagram for the preparation of light soup 58 Fig. 3.2: Flow diagram for the preparation of groundnut soup 59 Fig. 3.3: General risk assessment framework (CAC, 1998) 60 Fig 4.1: Age of traditional fish processors in Jamestown and Tema New Town 64 Fig 4.2: Highest level of education attained by traditional fish processors in Jamestown and Tema New Town 64 Fig. 4.3: Number of years in traditional fish processing 65 Fig 4.4: Process flow diagram for hot-smoked tuna and mackerel 67 Fig. 4.5: Flow diagram for hot-smoked herrings 68 Fig. 4.6: Metallic ovens 69 Fig. 4.7: Chorkor smoker 69 Fig. 4.8: Sanitation at smoking site 69 Fig. 4.9: Thawing mackerel 70 Fig. 4.10: Washing mackerel 70 Fig. 4.11: Smoked mackerel ready for the market 71 Fig. 4.12: Washing fresh sardines 72 Fig. 4.13: Spreading washed sardines on the ground to dry 72 Fig. 4.14: Sweeping dried sardines off the ground 72 Fig. 4.15: Dried sardines collected in a basket 72 Fig. 4.16: Process flow for the production of momoni 73 Fig. 4.17: Stacks of stale fish being moved from a cold store for momoni processing 74 xiii Fig. 4.18: Gathering sufficiently dried momoni. Note processor stepping on fish 74 Fig. 4.19: Cutting up ray fish 74 Fig. 4.20: Chunks of ray fish being sun-dried after salting 74 Fig. 4.21: Processing flow diagram for kako 75 Fig. 4.22: Fraser broths after 24h incubation 84 Fig. 4.23: Presumptive Listeria monocytogenes colonies on Oxford Agar Plate 84 Fig. 4.24: Escherichia coli isolates on eosin methylene blue agar plate 84 Fig. 4.25: Risk pathway for traditionally smoked fish (tuna, mackerel, herrings) 92 Fig. 4.26: Risk pathway for traditionally salted fish 93 Fig. 4.27: Risk pathway for sundried fish 93 Fig. 4.28: Event tree for risk of ingestion of Listeria monocytogenes through consumption of traditionally smoked fish purchased from informal markets 94 Fig. 4.29: Event tree for risk of ingestion of Listeria monocytogenes through consumption of salted fish purchased from informal markets 95 Fig. 4.30: Event tree for risk of ingestion of Listeria monocytogenes through consumption of sundried fish purchased from informal markets 95 Fig. 4.31: Triangular distribution for probability of illness, Pill, among the elderly in Madina consuming sundried sardines contaminated with Listeria monocytogenes 108 Fig. 4.32: Triangular distribution for probability of illness among respondents in Tema New Town who consumed smoked mackerel contaminated with Listeria monocytogenes 109 Fig. 4.33: Triangular distribution for probability of illness among pregnant women in Jamestown consuming koobi contaminated with Listeria monocytogenes 110 1 CHAPTER ONE 1.0 INTRODUCTION 1.1 Background Traditional fish processing is an enterprise that contributes significantly to efforts at ensuring food and nutrition security in Ghana. In coastal communities, the practice serves not only as a major economic activity for both men and women, but also ensures a continuous supply of their main source of animal protein - fish. Indeed, the entire country is known to consume fish in large quantities; per capita consumption in 2008 was close to twice the world average (Bank of Ghana, 2008). Much of the fish consumed in the country is traditionally processed (smoked, dried, fried, salted and/or fermented) (Nketsia-Tabiri and Sefa-Dedeh, 2000; Adu-Gyamfi, 2006). Although simple and generally inexpensive, traditional fish processing has been characterized by poor quality control and unhygienic processing conditions that compromise the safety of the products (Sefa-Dedeh, 1989; 1993; Nketsia-Tabiri and Sefa-Dedeh, 2000). Additionally, most traditional fish processors sell their products on informal markets. These markets contribute to food and nutrition security by offering physical access to foods at low cost to a majority of Ghanaians. However, studies have shown that foods are generally not handled hygienically in these markets and therefore record high microbial counts. Oppey (2002), Cofie (2003) and Adu- Gyamfi (2006) found that smoked fish sold on such markets in Accra had high counts of coliforms, Staphylococcus aureus, Escherichia coli and other pathogens. It is therefore possible that food-borne illnesses such as staphylococcal poisoning by S. aureus, could result from consumption of the products. Additionally, illnesses could result from the presence of other pathogens such as Listeria monocytogenes known to 2 be associated with fish and introduced either during processing or post-processing handling. 1.2 Listeria monocytogenes L. monocytogenes is a Gram-positive, catalase positive, non-spore forming food-borne bacterial pathogen responsible for a highly fatal disease called listeriosis. It is reported to cause an estimated 2,500 illnesses and 500 deaths annually in the United States of America alone (CDC, 2009), and is also considered the leading cause of death among food-borne bacterial pathogens, recording very high fatality:case ratios (Montville and Matthews, 2005; Jay, 2003). The organism is widely distributed in the environment (ubiquitous), has long survival periods in foods, grows under very low temperatures (-1.5 oC), tolerates high salinity (up to 10 – 12% NaCl), low pH (minimum 4.4), and low water activity (minimum 0.83) (Farber and Peterkin, 1991; Garbutt, 1997; Sutherland and Porritt, 1997; Jay 2003; Montville and Matthews, 2005). As a result of these unique properties, it has the potential to easily contaminate food and, when it does, to survive traditional methods employed to prevent microbial growth in foods, such as reduction in water activity, storage under cold temperatures, salting and acidification (Montville and Matthews, 2005). When ingested, L. monocytogenes has the unique ability to enter and grow in human phagocytes, thereby bypassing the inherent defensive mechanisms of the circulatory system (Montville and Matthews, 2005). Another special feature of this pathogen is that it has an uncommonly long incubation period (the time between ingestion of the 3 pathogen and the appearance of the first symptom of disease), reported to be typically between 1 – 70 days (Garbutt, 1997). Listeriosis, the disease the organism causes, comes with adverse health conditions such as meningitis, encephalitis, corneal ulcer, pneumonia, and septicaemia (Jay 2003). In pregnant women, intrauterine or cervical infections could occur and subsequently lead to spontaneous abortion, pre-mature birth, still birth or prenatal sepsis that could cause neonatal meningitis or death of newborns within a week. It is estimated that generally, 20 – 30% of listeriosis victims die (Montville and Matthews, 2005). 1.3 Listeria monocytogenes and food safety There has been an increasing global concern about L. monocytogenes and its influence on food safety. Several outbreaks of listeriosis have been reported in different parts of the world, with high fatalities. In 1998, a listeriosis occurrence in Finland resulted in 25 illnesses, of which only one person survived. The implicated food was butter. In France, 31 out of 32 infected persons died when an outbreak occurred in the year 2000 through consumption of pork contaminated with the pathogen. Similarly, when a listeriosis outbreak occurred in Canada in 2009 through red meat consumption, close to 50% of the infected persons (20 out of 53 cases) died (CDC, 2009). In September 2011, a listeriosis outbreak claimed 16 lives in 18 states in the USA (CDC, 2011). Cantaloupes from an eastern Colorado farm were implicated. These statistics clearly illustrate the food safety significance of L. monocytogenes. They also show that outbreaks of listeriosis, although sporadic, record very high fatalities (Table 1.1). 4 Table 1.1: Annual outbreaks of food-borne illnesses from selected pathogens Pathogen Cases Illnesses Deaths % Deaths Campylobacter spp. 1,963,141 10,539 99 0.95 Mackerelella non-typhoidal 1,341,873 15,608 553 3.54 E. coli O157:H7 62,458 1,843 52 2.82 E. coli non-O157-STEC 31,229 921 26 2.82 L. monocytogenes 2,493 2,298 499 21.71 Source: (CDC, 2009) Data true for USA The pathogen is largely associated with foods such as ready-to-eat (RTE) meat products, milk and milk products, coleslaw, and fish (particularly vacuum packed and cold-smoked fish) (Jay, 2003; Adam and Moss, 2008). 1.4 Fish as a vehicle for Listeria monocytogenes transmission Studies have shown that fish and fish products are suitable vehicles for the transmission of L. monocytogenes. The pathogen is frequently associated with RTE and heat-treated fishery products (Buchanan et al., 1997; Okutani et al., 2004; Basti et al., 2006). It has been isolated in smoked, salted, and salted and fermented fish (FAO, 1999). The frequent occurrence of L. monocytogenes in fishery products raises a food safety concern since most of such products are not (adequately) heat treated further before consumption. There is therefore the possibility that the pathogen could survive and/or grow in the foods as they move along the distribution chain (FAO, 1999). Three outbreaks of listeriosis that occurred in the 1990s were traced to fish products (FAO, 5 1999). These demonstrate that fish and fish products are suitable vehicles for the transmission of the pathogen to humans. 1.5 Problem statement Recent studies suggest the occurrence of L. monocytogenes in some foods in Ghana (Appiah, 2010; Dobge, 2010). Prior to these studies, published data on the occurrence of the pathogen in foods in Ghana were unavailable. Accordingly, as at 2010, specific national regulatory provisions for the control of the pathogen in Ghana were non- existent. Secondly, fish has been recognized as a vehicle for the transmission of L. monocytogenes. Fish processed by smoking and salting, methods notably employed in Ghanaian traditional fish processing (Adu-Gyamfi, 2006), are of particular interest in this regard (Lindqvist and Westoo, 2000). Interestingly, about 80% of fish landed in Ghana is processed this way (Adu-Gyamfi, 2006; Nketstia-Tabiri and Sefa-Dedeh, 2000). The lack of quality control in traditional fish processing makes contamination of the products highly probable. The processed fish are also sold mainly on informal markets where foods are generally sold under unsatisfactory hygienic conditions. Contamination with L. monocytogenes is therefore probable. Fish also constitutes the bulk of animal protein in diets in Ghana, contributing 60% of the total animal protein in diets throughout the country (Steiner-Asiedu et al., 1991; 6 Plahar et al., 1991). Moreover, fish purchases are estimated to account for 22.4% of household food expenditures (BoG, 2008). Given that more than 80% of the country’s fish landings are traditionally processed, by extension, the country also consumes substantial quantities of traditionally processed fish. A food safety implication of this is that any compromise on the safety of these products could result in potentially devastating outbreaks of preventable food-borne illnesses. 1.6 Rationale for the study The high consumption rates of (traditionally processed) fish in the country and the suitability of the products as vehicles for the transmission of L. monocytogenes, suggest that uncontrolled occurrence of the pathogen in the products could result in outbreaks of listeriosis. Beside its effect on the health of consumers, such contaminations could result in intangible economic losses through loss of consumer confidence in Ghanaian fishery products. It was therefore important to conduct a situational analysis on the extent of contamination of traditionally processed fish sold on informal markets with L. monocytogenes, and to evaluate the exposure of consumers to the pathogen, hence this study. The key research questions the study sought to answer were: i. Is the occurrence of L. monocytogenes in traditionally processed fish a food safety issue in the consumption of the products? 7 ii. To what extent are consumers likely to ingest the pathogen through consumption of traditionally processed fish sold on informal markets? 1.7 Study objectives 1.7.1 Main objective To determine the occurrence (presence and concentration) of Listeria monocytogenes in traditionally processed fish in Ghana, and to estimate the risk of ingesting the pathogen through consumption of traditionally processed fish. 1.7.2 Specific objectives 1. To assess the frequency and quantities of consumption of hot-smoked fish (mackerel, tuna and herrings), salted-dried tilapia (koobi), salted and fermented fish (momoni), chunks of salted and dried ray fish (kako), and sundried sardines among consumers in James Town and Tema New Town 2. To detect the presence and concentration of Listeria monocytogenes in the selected traditionally processed fish purchased from some informal markets in Accra and Tema 3. To detect the presence and concentration of Listeria monocytogenes in fish during traditional processing (from raw materials to finished products) and compare the findings with those obtained for the market samples 4. To evaluate the survival of Listeria monocytogenes in domestic cooking 8 5. To estimate the risk of ingesting Listeria monocytogenes through consumption of traditionally processed fish contaminated with the pathogen 9 CHAPTER TWO 2.0 LITERATURE REVIEW 2.1 The Ghana fishery sector The fisheries sector in Ghana is an important player in the country’s economy. It is estimated to have contributed about 3.9% of the nation’s gross domestic product (GDP) and 11% of the Agricultural GDP in 2008 (Bank of Ghana, 2008). These GDP and AGDP figures stood at 3% and 5% respectively in 1997 (Sarpong, 2008), indicating the significant increases in the contributions of the sector to poverty reduction and provision of sustainable livelihoods over the years. The sector is currently supported by fish supplies from the marine, freshwater and aquaculture fisheries. Marine fisheries are the major suppliers, contributing more than 80% of the total annual fish catch in the country. Freshwater fishes are obtained from the Volta Lake (major supplier), reservoirs, coastal lagoons, and aquaculture (Sarpong, 2008). 2.1.1 Economic importance of the fishery sector The fisheries sector makes tremendous contributions to the economic development of the country through its role in ensuring food security, GDP and foreign exchange earnings, as well as provision of employment (and thus poverty reduction). 2.1.1.1 Food security This refers to the situation in which all people in a given population have enough food to eat at all times to be in good health, and to have assurance that this situation will not change in the future (FAO, 2009). Schmidhuber and Tubiello (2007) identified 10 availability of food, access to food, stability of incomes and food production, and utilization of food as the four crucial factors that must be adequately addressed if the goal of food security for all is to be achieved. Those four elements have accordingly been described as the ‘pillars’ of food security. The fishery sector contributes to food security in the country by ensuring availability of animal protein food. As already indicated, fish protein contributes as much as 60% of the total animal protein consumed in Ghana (Steiner-Asiedu et al., 1991; Plahar et al., 1991). The average per-capita consumption of fish in the country is estimated to be between 20 and 25 kg, about twice the world average of 13 kg. Fish makes up 22.4% of food expenditure in all households and 25.7% in poor households, confirming the significant part it constitutes in Ghanaian diets (Atta-Mills et al., 2004). 2.1.1.2 Foreign exchange Exports of fishery products account for over 50% of the country’s earnings from non- traditional exports (NTE), and are reported to be the second most important NTE after horticultural products (BoG, 2008). In 2006, about 60,000 metric tons of raw and processed fish were exported; earning over US$80 million for the country (BoG, 2008). The Institute of Statistical, Social and Economic Research of Ghana (ISSER) indicates that in just one year, the share of fish and seafood in non-traditional agricultural export products increased from 25% (in 2000) to 33 % (in 2001) (BoG, 2008). 11 2.1.1.3 Employment It is estimated that a total of 500,000 fishermen, fish processors, traders and boat builders are employed in the fisheries sector in Ghana. These ‘employees’ are estimated to form about 10% of the population (Aquay, 1992; Atta-Mills, 2004). A canoe census conducted for the marine fisheries in 2001 placed the number of artisanal fishermen at 120,000 (Bannerman and Cowx, 2002). Apart from the fishermen, processors and traders who generally dominate the sector at the landing sites and market centres, a large number of people also obtain livelihood support through their involvement at different stages of the fish distribution chain. For example, labourers who pack, store, load, unload and transport fresh and processed fisheries products on foot or by trolley for short distances earn some income in the process. Others include export processors, cannery workers, fishmeal manufacturers and their staff, and those engaged in the production of packaging materials for different types of fish products. There are also those who supply production and processing inputs and services such as boat builders, mechanics, timber and fuel wood providers, and food vendors at landing and processing sites (Overa, 2002). The involvement of people in the fish post-harvest chain comes in the form of full- time employment, seasonal involvement, and occasional or opportunistic involvement. This affords people in coastal and lakeshore areas, and all others interested in the fish trade to diversify their livelihood strategies (Atta-Mills et al., 2004). 12 The fishery sector is also one of the few sectors of the economy where there is considerable gender equality in the workforce; men are involved in fishing while women are the key players in on-shore post-harvest activities, undertaking fish processing, storage and trade activities. 2.2 Traditional fish processing in Ghana It has been estimated that more than 80% of fish landed in the country is traditionally processed (Nketsia-Tabiri and Sefa-Dedeh, 2000; Adu-Gyamfi, 2006). Traditional fish processing is thus an important economic activity in Ghana. It serves as a source of income to many and also provides the main form in which fish is consumed. According to Sefa-Dedeh (1989; 1993), traditional fish processing is often characterized by all or some of the following: i. Low capital cost No huge financial inputs are required to start a business in traditional fish processing. The basic requirements are a smoker, firewood as fuel, and fish for smoked fish, salting vat, salt, and fish for salted fish, and fish and sunlight for dried fish. These can be acquired/accessed rather easily with little initial capital. ii. Labour intensive Since the processes are not mechanized, every unit operation requires manual inputs. 13 iii. Time consuming This results from the lack of mechanization in the processing activities. For example, in the production of sundried fish, the duration of drying is controlled by the weather, and can thus be prolonged during rainy seasons. This has implications on the safety of the final products iv. Simple and small scale operations Indigenous processing activities do not require any sophisticated technologies. Fish are also processed in small quantities. In most instances, quantities sufficient for marketing in a day are produced. This could partly be as a result of a lack of good storage facilities. v. Poor quality control No objective methods are employed to monitor processing. The readiness of products for the market is determined by the subjective judgments of the processors. vi. Unhygienic processing conditions Basic rules of personal and environmental hygiene are not satisfactorily practiced during traditional fish processing, thereby compromising the safety of products. vii. Home based Many traditional fish processors operate from their homes. From smoking 14 ovens built in front of homes to compounds in front of homes used for drying, processing areas, materials and activities are not clearly separated from those for households. 2.2.1 Methods of traditional fish processing The methods of traditional fish processing in Ghana are smoking, salting, drying, fermentation, and frying (Nketsia-Tabiri and Sefa-Dedeh, 2000; Neequaye-Tetteh et al., 2002). Among these, smoking is practiced the most; it is estimated that more than 60% of the country’s fish landings are preserved by smoking (Adu-Gyamfi, 2006). Historically, smoked fish has also been the most patronized of all traditionally processed fish in Ghana (Orraca-Tetteh and Nyanteng, 1978; Adu-Gyamfi, 2006). This high level of smoked fish processing and consumption is also true for other West African countries (UNDP, 2002). 2.2.1.1 Smoking The Ghana Standards Board has defined smoked fish as fish which has been exposed to smoke with the intention of deferring spoilage. Traditional fish smoking preserves fish through the combined effects of the following: i. cooking: at high temperatures, the fish are cooked, thereby denaturing native enzymes which could cause deterioration, and kills vegetative microorganisms that could cause spoilage ii. drying: heat from the burning wood contributes to the drying of the fish iii. preservation value of the smoke: compounds such as methanol and phenols in the smoke have bactericidal properties (Ihekoronye and Ngoddy, 1985). 15 Smoked fish are placed into two categories based on the processing temperature at which they are produced. These are cold-smoked and hot-smoked fish (Ihekoronye and Ngoddy, 1985). The processes are accordingly called cold-smoking and hot- smoking respectively. In cold-smoking, the internal temperature of the fish usually does not exceed 35oC. Generally, a range of 30-40oC for 30-60 minutes is typical (Cofie, 2003). It is mostly practiced in technologically advanced countries. Cold-smoked fish are neither well dried nor cooked due to the low temperatures employed. Hence, they have high moisture contents and short shelf-life, typically 3 days (Cofie, 2003). They also require cooking before consumption. In hot-smoking, the processing temperature is usually ≥90oC. The internal temperature of fish typically exceeds 60oC. The products have relatively low moisture content and thus have longer shelf life. Hot-smoked fish are cooked and can therefore be consumed without further heat treatment (Bannerman and Cowx, 2002). Hot-smoking is the method employed in traditional fish smoking in Ghana, and in many developing countries (MOFA, 1999; UNDP, 2002). There are two forms of hot- smoking, namely wet hot-smoking and dry hot-smoking. They differ in their duration and the final moisture content of the products. Whilst wet hot-smoking generally takes 1-2 hours and yields products with moisture contents of 40-55%, dry hot- smoking takes 10-18 hours and yields a product with low moisture contents (10-15%) (UNDP, 2002). 16 2.2.1.2 Salting In salting, fish is preserved by significantly reducing its moisture content through the osmotic effect of common salt (NaCl). The lowered water activity and residual salt in the resulting products discourage the growth of most microorganisms (Essuman, 1982). However, salt tolerant (also called halotolerant) microorganisms such as Staphylococcus aureus and Listeria monocytogenes have the potential to grow in such products. There are two main types of salted fish products in Ghana, namely salted dried fish and salted fermented fish (Nketsia-Tabiri and Sefa-Dedeh, 2000). In salted dried fish, after extraction of water from fish using salt, the fish are further dried under the sun. A product made through this process is koobi (salted dried tilapia) and kako (salted dried ray fish). Products such as momoni are obtained by fermenting fish after the salting operation. There are four methods of salting, namely brining, pickling, kench curing, and Gaspé curing (Horner, 1997). During brining, fish is immersed in a slightly saturated salt solution for a few minutes and removed. This usually serves as a preliminary step for other unit operations such as smoking and drying. The brining step is often done to impart a desirable flavour to the product (Horner, 1997). When fish are salted by immersion in saturated salt solution for long periods, the process is called pickling. This is often employed to preserve fatty fish, since the immersion prevents direct contact with atmospheric oxygen, thus preventing rancidity reactions. 17 In Kench curing, fish are cut open and arranged such that a layer of salt separates any two layers of fish (Horner, 1997). The exudate is drained off as waste. The method is employed in salting non-fatty fish and gives dry products. The only difference between Gaspé curing and Kench curing is that in the former, the exudate serves as a salt solution into which the fish are immersed for a further 2-3 days, after which they are removed and dried (Horner, 1997). The exudate in both Kench and Gaspé curing is called ‘pickle’. When water is extracted from fish so that the moisture content is reduced from 82% to about 54%, the fish is called green cured (Ihekoronye and Ngoddy, 1985). 2.2.1.3 Fermentation In this method, fish is mixed with salt and made to ferment. The method preserves fish by increasing its acidity. Essuman (1992) defined fermented fish as any fishery product resulting from the enzymatic and/or microbial degradation of fish either in the presence or absence of salt. The products are usually used as condiments (Cofie, 2003). 2.2.1.4 Drying The low humidity and ambient temperatures of tropical environments are made use of to dry fish, particularly bony fish such as tilapia. The final products usually have a moisture content of 14-30%. The fish are usually split asymmetrically longitudinally and arranged under the sun to dry (Essuman, 1992). The hygienic conditions of drying are generally unsatisfactory. Artisanal fish processors are known to dry their fish on the bare ground or on mats spread out on the bare ground. 18 2.3 Consumption of traditionally processed fish in Ghana As was mentioned in Section 1.5 (Chapter One), Ghana records high per capita fish consumption. With a value of 20 – 25kg, the country’s per capita fish consumption is nearly twice the world average of 13kg (BoG, 2008). In agreement with these figures, fish has been the preferred and cheapest source of animal protein in Ghana (Steiner- Asiedu et al., 1991; Adu-Gyamfi, 2006). About 75% of total annual fish landings are consumed locally (Sarpong, 2008, BoG, 2008). The high consumption rate is mainly due to high availability and low price of the commodity compared to other sources of animal protein. Given that about 80% of fish landings in Ghana is traditionally processed (smoked, salted, fried, or dried), it can be argued that a greater quantity of the 75% of total annual fish landings consumed in the country is traditionally processed. By extension, it can be hypothesized that traditionally processed fish probably constitutes a greater portion of the 60% animal protein provided by fish in Ghanaian diets, and that a greater portion of the estimated 22.4% household expenditure on fish is made on the traditionally processed fish. Ghana can therefore be said to be a heavy consumer of traditionally processed fish. The products are mostly purchased from informal markets in both urban and rural areas. These informal markets are an essential component of the informal sector in Ghana. 2.4 The informal food sector The informal sector has been defined as employment or production that takes place in 19 small, unregulated and/or unregistered enterprises (Chant, 1999). Generally, the sector is characterized by reliance on indigenous resources, family ownership, small-scale operations and unregulated and competitive markets (Munhande and Makaye, 2008). It offers significant economic benefits globally and in Ghana, it is estimated to contribute as much as 58% GDP (ILO, 2002). The informal sector has several ‘sub-sectors’, each involved in businesses related to a particular service, example barbering shops, unregistered tailors and seamstresses, and street hawkers. Those whose activities are related to food and its distribution form the informal food sector. According to FAO (2007), the informal food sector (IFS) actors include small-scale producers and processors, manufacturing enterprises, traders and service providers who undertake either legal or unrecognized activities related to food. Market women selling livestock and horticultural produce, caterers, street food vendors (including those at fixed kiosks, mobile stands, those who sell from vehicles such as carts and bicycles trucks) are all players in IFS (FAO, 2007). In Ghana, small-scale farmers, persons in markets selling various kinds of fresh, traditionally processed and industrially processed food items, and ready-to-eat street food vendors are included. IFS offers convenience in food purchasing at a low cost to consumers, especially those in the lower and middle classes, the urban poor, office workers and tourists, while providing income to those who sell the foods. It also contributes to making food available in marginal urban districts that are distant from major city commercial centres (FAO, 2007) and thus contributes to food security. 20 The indicated benefits notwithstanding, there are some challenges associated with IFS. Argenti (2000) argued that as a result of the generally unregulated status and lack of formal legal support, the informal food sector lacks the appropriate inputs for improvements in food hygiene. Consumers therefore face food safety risks in purchasing foods from the sector. This is true in Ghana, (as it is in many other developing countries) where the safety of foods sourced from IFS cannot be guaranteed due to poor hygiene control and monitoring. For example, Mensah et al., (2002) isolated enteroaggregative Escherichia coli, Salmonella arizona, and Shigella sonnei in some street foods in Accra.This notwithstanding, many are of the view that the sector should not necessarily be equated to poor quality food (FAO, 2007). 2.5 Safety of traditionally processed fish on informal markets As a result of the generally unhygienic conditions under which fish is traditionally processed and sold, the main safety consideration is the microbiological quality of the products. Adu-Gyamfi (2006) conducted a study on the microbiological quality of smoked fish in some informal markets in Accra and found that the safety of the products was compromised. Microorganisms isolated included Escherichia coli, Klebsiella pneumonia, and Proteus mirabilis. Nyamekye (2000) also isolated Staphylococcus aureus, Escherichia coli, Micrococcus sp and Proteus sp from smoked herrings and mackerel purchased from informal markets in Accra. Oppey (2002) found Aspergillus sp, Penicillium sp, and coliforms in some smoked fish products from informal markets in Accra. 21 In addition to microbial contamination resulting from improper handling practices, processing conditions could also mar the chemical safety of processed fish prior to sales on informal markets. For example, in smoked fish, the deposition of carcinogenic compounds in wood smoke could make the products unsafe. These compounds include polycyclic aromatic compounds (e.g. 3,4-benzopyrene and pyrene) and nitrosamines (Rahman and Perera, 2007). In salted fish, the key concerns are the chemical and microbial quality of the salt used. Additionally, high levels of magnesium and calcium in salt result in bitter and toughened products (Collignan and Raoult-Wack, 1994). In dried fish products, the key concern is possible microbial contamination arising from the practice of drying fish on the bare ground or very close to the same. For fermented fish products, the generally uncontrolled fermentation could result in the growth of potentially dangerous microorganisms in the products (Guizani and Mothershaw, 2007). 2.6 Bacterial pathogens associated with fish Fish has associated bacteria pathogens that could be present in the skin, gills or gut. These organisms have been categorized into two groups: indigenous and non- indigenous pathogenic bacteria (Nickelson and Finne 1992; Huss et al. 1995). The indigenous pathogenic bacteria are commonly found in the aquatic environment; they are present on the live fish and their presence in the final product is probable. They include Listeria monocytogenes, Clostridium botulinum, Aeromonas hydrophila 22 and Vibrio sp (Huss et al. 1995, Nickelson and Finne 1992). Non-indigenous pathogenic bacteria are normally associated with human or warm-blood animals and their faeces, and are not naturally present in fish. They are therefore contaminants. Salmonella sp, Escherichia coli and Staphylococcus aureus are among such pathogens (Huss et al. 1995; Nickelson and Finne 1992). Matte et al., (1994) also categorized the pathogenic bacteria associated with fish (and sea food in general) into three groups as follows: a. bacteria which are normal components of the marine or estuarine environment: Vibrio parahaemolyticus, Vibrio cholera, Vibrio vulnificus, Clostridium botulinum, Aeromonas hydrophilia, and Listeria monocytogenes b. enteric bacteria whose presence is attributed to faecal contamination: Escherichia coli, Shigella sp, Campylobacter sp, and Yersinia enterocolitica c. bacteria that contaminate fish during processing: Listeria monocytogenes, Bacillus cereus, Staphylococcus aureus, and Clostridium perfringens These microorganisms, when present in fish and fish products and not eliminated before consumption, could cause various specific forms of food-borne diseases. Some pathogenic microorganisms implicated in food-borne illnesses arising from fish consumption are listed in Table 2.1. 23 Table 2.1: Some bacterial pathogens implicated in food-borne disease outbreaks caused by fish Pathogen Associated fish Listeria monocytogenes Cold-smoked and hot-smoked fish, salted fish . Salmonella sp Tilapia, carp, prawns, catfish Shigella sp Shellfish Yersinia enterocolitica Aquaculture fish Vibrio cholera Prawns, squid, shellfish Vibrio parahaemolyticus Most finfish, shellfish, Vibrio vulnificus Finfish, mussels, prawns, oysters Campylobacter jejuni Shellfish Source: Fehlduson (1999) It is seen that the classification systems by Huss et al. (1995), Nickelson and Finne (1992), and Matte et al., (1994) all recognize Listeria monocytogenes as a bacterial pathogen associated with fish. 2.7 Listeria monocytogenes L. monocytogenes is one of six species of bacteria in the genus Listeria. The other species are L. ivanovii, L. seeligeri, L. grayi, L. welshimeri, and L. innocua (Prentice and Neaves, 1992). The genus belongs to the Clostridium sub branch with Staphylococcus, Streptococcus, Lactobacillus, and Brochothrix (Montville and Mathews, 2005). The Listeria species are identified by a few biochemical traits. These include tests for acid production from D-xylose, L-rhamnose, methyl-D-mannoside, and D-mannitol (Table 2.2). The ability to lyse red blood cells differentiates L. monocytogenes from the non-pathogenic strains (Jay et al., 2005). 24 Within the genus Listeria, only L. monocytogenes and L. ivanovii are pathogenic; the former, mainly to humans and the latter, to animals, particularly sheep (Prentice and Neaves, 1992). L. monocytogenes is Gram-positive, facultatively anaerobic, non-sporing, rod-shaped motile bacterium. Other cellular shapes, such as palisade, cocci and Y forms have also been observed (Garbutt, 1997). Apparently, the shape occurring at any time depends on the culturing conditions (Garbutt, 1997). It shows a unique tumbling motility at 20- 25oC but not at 35oC (Prentice and Neaves, 1992). The organism is psychrotrophic and grows over a temperature range of 0° to 45°C, with an optimum around 37°C (Jay et al, 2005). It grows slowly at colder temperatures and is generally known to be killed at temperatures >50oC (Black, 1999). L. monocytogenes has been found to grow at water activities (aw) ≥0.92 with sodium chloride (NaCl) as the solute. Generally, however, the organism grows best at aw ≥0.97 (Jay et al., 2005). While the minimum aw for growth for most strains is 0.93, Table 2.2: Differentiating characteristics of Listeria species Acid production from Species β- haemolysis* Glucose α-Methyl-D- mannoside Rhamnose Xylose Mannitol monocytogenes + + + + - - Innocua - + + v - - Ivanovii ++ + - - + - Seeligeri (+)1 + v - + - Welshimeri - + + v + - Grayi - + + - - + *Horse or sheep blood 1Washed sheep blood (+)1 Weak reaction v: different strains give different reactions Source: Prentice and Neaves (1992) 25 some are able to grow at aw 0.90. The organism is able to survive for long periods at aw as low as 0.83 (Montville and Matthews, 2005). Accordingly, L. monocytogenes is recognized as the second food-borne pathogen (after the staphylococci) with the ability to grow at aw values <0.93 (Cofie, 2003). When foods containing L. monocytogenes are heated, the thermal resistance of the organism increases as the water activity of the foods decrease. In one study, when Scott A strain of L. monocytogenes in liquid whole egg was heated at 60 ◦ C for 3.5 minutes, the D value (decimal reduction time, which is the time required to reduce the numbers of microorganisms or their spores by 90% at a specified temperature (Atlas, 1997) was 2.1 minutes. However, the same strain in liquid whole egg to which 10% NaCl was added and heated at 63 ◦ C for 3.5 minutes had a D value of 13.7 minutes (Bartlett and Hawke, 1995). It is seen that although a higher temperature (63oC) was used for liquid whole egg with 10% salt, the D value was about four times that of the liquid whole egg without salt. This was because the 10% NaCl lowered the aw of the product from 0.98 to 0.915 (Linton et al., 1990). L. monocytogenes also exhibits remarkable tolerance to high salt concentrations. It grows to high levels at concentrations of 6.5%, grows considerably at 10-12%, and survives for long periods at higher concentrations of NaCl. In salty food systems, as temperature is lowered, the ability of the pathogen to survive at high salt concentrations increases (Ryser and Marth, 1988). This has implications on the safety of cured products such as salted fish. 26 L. monocytogenes can grow at pH levels between 4.4 and 9.4 in laboratory media. Below pH 4.3, the cells survive, but may not grow. It is reported that organic acids such as acetic, citric and lactic acid at 0.1% concentration inhibit the growth of the organism (Montville and Matthews, 2005). 2.7.1 Heat resistance of Listeria monocytogenes Several studies have been conducted on the effect of heat on the survival and growth of L. monocytogenes in several food items. The decimal reduction time (D value) and z values for the organism in some foods are presented in Table 2.3. 27 Table 2.3: Selected findings on the thermal destruction of L. Monocytogenes Strains tested/state Number of Cells (ml) Heating Menstrum Heating Temp. (oC) D value (sec) z value (oC) Reference Scott A, free suspension ~105 Sterile skim milk 71.7 1.7 6.5 Bradshaw et al, 1987 ~105 Sterile skim milk 71.7 2.0 6.5 Ryser and Marth, 1988 ~105 Sterile skim milk 71.7 0.9 6.3 Bradshaw et al., 1985 Scott A, intracellular ~105 Whole raw milk 71.7 1.9 6.0 Brunning et al., 1986 Scott A, free suspension ~105 Whole raw milk 71.7 1.6 6.1 Brunning et al., 1986 F5069, intracellular ~106 Sterile whole milk 71.7 5.0 8.0 Bruning et al., 1986 F5069, free suspension ~106 Sterile whole milk 71.7 3.1 7.3 Brunning et al., 1986 Scott A, free suspension ~105 Ice cream mix 79.4 2.6 7.0 Bradshaw et al., 1987 ~108 pH 5.9, meat slurry 70.0 13.8 NR Boyles et al., 1990 ~107 Liquid whole egg 72.0 36.0 7.1 Boegeding et al., 1990 Ten strains ~107 Irradiated ground meats 62.0 61.0 4.92 Farber, 1989 Chicken/meat isolate ~105 Beef 70.0 NR 7.2 Mackey et al., 1990 ~105 Minced chicken 70.0 NR 6.7 Mackety et al., 1990 Source: (Jay et al., 2005). D value (decimal reduction time): Time taken to reduce the number of microorganism or spores in a sample by 90% at a specified temperature. Z value: Tempreture increase required to reduce the D value to 10% of its original value (Prescott et al., 1995). NR: Not Reported. 28 For dairy products, the reported D values suggest that high-temperature short-time (HTST) treatment (71.7◦C for 15 seconds) is adequate to reduce normally existing numbers of the pathogen to below detectable levels. Farber et al., (1998) subjected milk that was naturally contaminated with L. monocytogenes at 104 cells/ml to HTST treatment and found no viable cells at processing temperatures of 69 ◦ C or above. Jay et al., (2005) have however indicated that the low-temperature, long-time (LTLT) heat treatment method (62.8 ◦ C for 30 minutes) has a greater lethal effect on the pathogen. Mackey and Bratchell (1989) also concluded that the margin of safety is greater for the LTLT treatment than the HTST treatment. For non-dairy products, the thermal resistance of L. monocytogenes varies with the type and composition of the foods. The biological structure and composition of foods has implications on the thermal resistance of L. monocytogenes; the organism has greater heat resistance in foods with high fat content owing to the heat-shielding effect of the fat. Embarek (1994) studied the heat resistance of two strains of L. monocytogenes in cooked cod and salmon fillets and found that both strains were about four times more heat resistant in salmon than in cod, as a result of the relatively more fatty nature of salmon. D values for liquid whole egg and meat products are also reported to be generally higher than for milk (Jay et al., 2005). The higher protein and fat content in the whole eggs and meat products than in the milk account for the difference. Farber (1989) found that cure ingredients increased the D value for the pathogen. The author found the D value for sausage-type meat at 62 ◦ C to be 61 seconds (Table 2.4). 29 The D value however increased to 426 seconds (7.1 minutes) when cure ingredients were added. This suggests that the components of the cure ingredient (nitrite, dextrose, lactose, corn syrup, and 3% (w/v) NaCl) offered some thermal protection to the organism. Mackey et al., (1990) also found that when 30% fat, 3.5% NaCl, 200 ppm nitrite, and 300 ppm were added to ground beef, the D value approximately doubled. Conflicting results have been obtained for the effect of initial sublethal heating of L. monocytogenes on the resistance of the organism to subsequent heat treatments. While some studies observed no effect (Bradshaw et al., 1985; Bunning et al., 1990), others have reported increased resistance (Fedio and Jackson, 1989; Farber and Brown, 1990; Linton et al., 1990;). Linton et al., (1990) found that when some strains were heat shocked at 48oC for 20 minutes, their heat resistance increased at 55oC. Fedio and Jackson (1989) reported an increased resistance at 60oC after the organisms were heat-shocked at 48oC for 60 minutes. Farber and Brown (1990) also found that when 10 strains at a concentration of 107 cells/g in a sausage mix were heat shocked at 48 ◦ C for 30 or 60 minutes, no significant increase in heat resistance was observed after subsequent exposure to 62o or 64oC. However, in the same study, those shocked for 120 minutes recorded increased resistance at 64oC. The cells also maintained their heat resistance for at least 24 hours when stored at 4oC. 2.7.2 Occurrence of L. monocytogenes The pathogen is highly ubiquitous (Prescott et al., 1995; Garbut 1997; Black, 1999; Jay 2003). It occurs in soil, dust, fresh and salt water, decaying vegetation, silage, slaughter house waste, sewage effluent (Garbutt, 1997; Jay et al., 2005), and as 30 biofilms on food processing equipments (Mauro et al., 2008) and hospital equipments. Jemmi and Keusch (1994) isolated the organism in fish from Swiss freshwater fish farms. L. monocytogenes has the ability to colonize and adapt to various environments as a result of its unique tolerance to a wide range of temperatures, pH and salt concentration (Mauro et al.,2008; Jay et al., 2005), and also because of its ability to form biofilms (Jay et al., 2005). 2.7.3 Foods frequently contaminated with L. monocytogenes As a result of its widespread distribution in the environment and hardy nature, L. monocytogenes contaminates a wide range of foods. In fact, some international food regulators have opined that it is impossible to produce foods that are practically free of the organism (Montville and Matthews, 2005). The organism has been found in raw milk and dairy products, fresh and frozen meat and meat products, poultry products, seafood, and on fruits and vegetable products (Jay et al., 2005). Some specific foods in which the pathogen has been isolated include whole milk, skim milk, soft cheese, processed meats, red meat, vacuum packaged beef and poultry products, lettuce, cold-smoked and hot-smoked fish, salted fish, coleslaw, and fried rice (Montville and Matthews, 2005; Jay et al., 2005). These foods can therefore serve as suitable vehicles of transmission for the organism, particularly when they are processed, stored or prepared under conditions that facilitate contamination and encourage the growth of the pathogen. 31 2.7.4 Persistence of L. monocytogenes in foods As a result of its hardy nature (growth over wide range of temperature (0-45oC) and pH (4.1 – 9.6), in high salt concentration and at low water activity), L. monocytogenes is able to survive for long periods in foods. At levels of 104–105/g, the pathogen survived in cottage cheese for up to 28 days when held at 3oC (Ryser and Marth, 1988). It also survived for 130 days in cold-pack cheese stored at 4oC in 0.30% sorbic acid (Ryser and Marth, 1988). Shelef (1989) also found that the count of L. monocytogenes remained unchanged in ground beef and liver for over 30 days, although the standard plate counts (SPC) increased during that same period. Glass and Doyle (1990) challenged eight processed meats with five strains of L. monocytogenes under 12 weeks storage at 4.4oC. The pathogen not only survived on all products, but increased in numbers by 3 to 4 logs in most. The products with high initial pH (such as poultry products) recorded the highest growths. 2.7.5 Transmission of L. monocytogenes to humans L. monocytogenes is widely distributed and therefore has many potential routes to infect humans (Fig. 2.1 and 2.2). The modes of transmission include vertical (mother to child), zoonotic (contact with animal to man), and nosocomial (hospital acquired) (FAO/WHO, 2004). 32 Sludge water, river Aerial contamination Aerial contamination Animal Secretion and Excretion Faeces, uterine and vaginal secretions, placenta, membranes Animals Disease or intestinal colonization Man Disease/intestinal colonization Animal Products Milk, meat, dairy products Fig. 2.1: Ways by which L. monocytogenes is distributed in the environment Source: Jay et al., (2005) Sewage Environment Soil, water, plants, silage, etc Fig. 2.1 shows that humans get infected with L. monocytogenes either through the environment or through food. Humans, once infected, also contribute to the spread of the organism in the environment. Fig. 2.2 focuses on food as the principal means of infection, and also shows how humans have the potential of releasing the pathogen into the environment, with the result that foods ultimately get contaminated and deliver the organism back to humans. 33 It is generally recognized that most cases of human infections involve food-borne transmission, as is apparent in Fig. 2.2. However, this has not always been the accepted knowledge. When the first case of food-borne infection occurred in 1953 in which the stillbirths of twins was linked to consumption by the mother of raw milk from a cow with listerial mastitis (FAO/WHO, 2004), the significance of foods as a mode of transmission was not given much consideration until the 1980s when several large outbreaks of Water Soil Fish Processed foods Milk Food processing, equipment and environment HUMANS Faeces Sewage Plants Silage Ruminants Manure Vegetables Meat Fig. 2.2: Potential routes of transmission of L. monocytogenes to humans Source: Montville and Matthews (2005) 34 listeriosis linked to common foods occurred in North America and Europe. It was then that the significance of foods as the primary route of transmission for human exposure to L. monocytogenes was recognized (Broome et al., 1990; Bille, 1990). Following this, there has been a strong interest in the control of the pathogen in foods (FAO, 2000). Today, contaminated food is estimated to be responsible for about 90% of food-borne illnesses caused by L. monocytogenes. 2.7.6 The disease caused by Listeria monocytogenes Listeria monocytogenes causes a fatal food infection called listeriosis. According to the United States Centre for Disease Control and Prevention (CDC, 2009), listeriosis is clinically defined when the organism is isolated from blood, cerebrospinal fluid, or an otherwise normally sterile site in the body, such as the placenta and foetus. It is therefore only positively diagnosed by culturing the organism from blood, cerebrospinal fluid, or stool. Once infection occurs, the organism is able to enter and multiply in the host’s monocytes, macrophages, or polymorphonuclear leukocytes. This offers it access to the brain and the placenta (Montville and Matthews, 2005). CDC (2009) indicates that the pathogenicity of L. monocytogenes rests on its ability to cause systemic infection (survival and multiplication in body cells). The disease occurs rarely but has very high fatality rates, usually 20 – 30% (Kalliopi et al., 2008). The mortality from meningitis, septicaemia and perinatal/neonatal infection caused by L. monocytogenes are respectively estimated at 70%, 50% and 80% (CDC, 2009). Persons at high risk include pregnant women, the very young (less 35 than 1 year), the very old, and the immunocompromised (Kathariou, 2002). A non- invasive form of the disease characterized by febrile gastroenteritis suggests that persons with no predisposing conditions may be affected (Lorber, 2007). It has been found that some healthy individuals are asymptomatic faecal carriers of the pathogen. These include pregnant women, patients with gastroenteritis, slaughterhouse workers, laboratory workers handling Listeria, food handlers, patients undergoing haemodialysis, and some healthy people (Montville and Matthews, 2005). It is estimated that about 1-10% of humans may be intestinal carries of the pathogen (Montville and Matthews, 2005), and that faecal carriers amplify outbreaks through secondary transmission (i.e. transfer of the organism from the faeces of the first victim to another person) (Montville and Matthews, 2005). Interestingly, some pregnant women who are asymptomatic carriers are able to have normal pregnancy outcomes (Montville and Matthews, 2005). The first human case of listeriosis was reported in 1929 (Jay et al., 2005). Sporadic incidents have followed since then, with high fatalities per occurrence. It is reported that between 1986 and 1988, human listeriosis increased in England and Wales by 150% (Jay et al., 2005). Between 1983–1987, 775 cases were reported in Britain, with 219 (28%) deaths (Jay et al., 2005). Data prospectively collected by the United States Centre for Disease Prevention and Control (CDC, 2009) suggests that at least 1600 cases of listeriosis occur annually in the USA, with 415 deaths per year. Table 2.4 lists some incidents of listeriosis and their attendant fatalities. 36 Table 2.4: Some food-borne listeriosis outbreaks Year Source Cases/Deaths Location 1953 Raw milk 2/1 Germany 1959 Fresh meat/poultry* 4/2 Sweden 1960-1961 Various/unknown 81/? Germany 1966 Milk/products 279/109 Germany 1979 Vegetables/milk** 23/3 Boston 1980 Shellfish 22/6 New Zealand 1981 Cole slaw 41/18 Canada 1983 Pasteurized milk** 49/14 Boston 1983-1987 Vacherin Mont D’Or 122/34 Switzerland 1985 Mexican-style cheese 142/48 California 1986-1987 Vegetables** 36/16 Philadelphia 1987-1989 Pate 366/63 United Kingdom 1987 Soft cheese 1 United Kingdom 1988 Goat milk cheese 1 United Kingdom 1988 Cooked, chilled- chicken 1 United Kingdom 1988 Cooked, chilled- chicken 2 United Kingdom 1988 Turkey franks 1 Oklahoma 1989 Pork sausage 1 Italy 1988 Alfalfa tablets 1 Canada 1989 Salted mushrooms 1 Finland 1989 Shrimp 9/1 United States 1989 Pork sausage 1 Italy 1990 Raw milk 1 Vermont 1990 Pork sausage 1 Italy *Suspected **Epidemiologically linked; organism not found 37 Table 2.4 continued 1990 Pate 11/6 Australia 1991 Smoked mussels 3/0 Australia 1992 Smoked mussels 4/2 New Zealand 1992 Goat meat (California) 1 Canada 1992 Port tongue in jelly 279/85 France 1993 Pork rillettes 39/0 France 1994 Chocolate milk 52/0 USA 1994 Pickled olives 1 Italy 1995 Brie cheese 17/0 France 1998-1999 Wieners 101/21 United States 1998 Butter 25/24 Finland 1999-2000 Pork tongue in jelly 26/7 France 2000 Pork 32/31 France 2000-2001 Homemade Mexican-style cheese 12/0 United States 2002 Deli turkey meat 46/7 10 USA States 2009 Red meat 53/20 Canada *Suspected **Epidemiologically linked; organism not found Source: CDC (2009); Jay et al.,(2005) 2.7.6.1 Foods implicated in listeriosis outbreaks Ready-to-eat (RTE) foods are mostly implicated in listeriosis outbreaks (Mauro et al., 2008). This is because those foods are often not given any further heat treatment before consumption. Among the RTE foods, meat and poultry products are reported to be the most frequently implicated vehicles of transmission (Jay et al., 2005). Other foods implicated in listeriosis outbreaks include red meat (2009, Canada), pork (2000, France), and butter (1998, Finland). Cooked chilled foods, vacuum-packed meat and fish products, and smoked fish (Oroczo, 2000) are also known vehicles for the pathogen (Table 2.5). 38 2.7.7 Occurrence of Listeria monocytogenes in fish Several studies have reported the incidences of Listeria spp. and L. monocytogenes in fish products such as cold-smoked fish, hot-smoked fish, and salted fish (Oroczo, 2000; Heinitz et al., 2000; Rørvick et al. 1995). Lindqvist and Westoo (2000) found the organism in smoked trout in Sweden. Kwiatek (2000), in a study on the occurrence of L. monocytogenes in foods of animal origin, found the pathogen in raw and smoked fish in Poland. Salihu et al. (2008) detected the pathogen in smoked fish in Sokoto, Nigeria. Conflicting results have been reported on the occurrence of L. monocytogenes in marinated fish; some studies have found the organism in the product, whiles others reported its absence (Kwiatek, 2000). Ikeh et al., (2010), in a study to identify the incidence and pathogenicity profile of Listeria sp from food and environmental samples in Nigeria, isolated the organism from both fresh and dried fish. Relative to the other foods found to contain the organism, the fish samples had a low incidence of 40% (beef 80%, poultry 70%, and vegetables 85%). They attributed the low incidence in fish to a possible low level of contamination of the water bodies from which the fish were caught. Prior to recent research, such as those conducted by Salihu et al., (2008) and Ikeh et al., (2010), some studies reported the absence of Listeria sp in tropical fish on the basis of unsuitable environmental conditions (Kamat and Nair, 1994; Karunasagar et al., 1992; Manoj et al., 1991). After detecting L. monocytogenes in 17.2% of finfish and 12.1% of shellfish in India, Jeyasekaram and Karunasagar (1996) attributed the perceived absence of the pathogen in tropical fish to inadequate isolation procedures. 39 2.7.7.1 Factors influencing initial load of Listeria monocytogenes in fish The initial load of Listeria species in fish or seafood is influenced by such factors as origin (wild or farmed), season, fishing technique, handling and storage conditions. For example, a greater incidence of Listeria species including L. monocytogenes has been reported in fresh water fish (81%) than in marine fish (30%) (Colburn et al. 1990). Beumer (1997) indicated that once present in fish (fresh or processed), the ability of L. monocytogenes to grow depends on those same factors that generally affect the growth of microorganisms in food, namely intrinsic factors ( pH, water activity, preservatives in the food), extrinsic factors (storage temperature, atmosphere in the package) and implicit factors (competition with other microorganisms). 2.8 Regulation of Listeria monocytogenes in foods The regulation of L. monocytogenes in foods is a controversial issue under a continued international debate that is anticipated to go on for years. While most countries in the European Union (EU) have set tolerance levels (allowable limits of counts) for the pathogen, the United Kingdom and the United States of America have declared a zero-tolerance for the pathogen (Gallagher et al., 2003). The EU countries argue that since L. monocytogenes is so widespread, it is practically impossible to produce foods that are free of the pathogen. The tolerance levels set by these countries therefore define acceptability and unacceptability of foods based on the counts of the organism in the foods. In some of the EU countries, food products that historically have caused human listeriosis are placed in a special category and are monitored more strictly than those that have never been implicated in the disease. 40 Generally, the tolerance level policy asserts that foods should be L. monocytogenes free if possible, or have the lowest counts possible (Montville and Matthews, 2005). Specifically, foods for high-risk groups (pregnant women, the very young, the elderly, and the immunocompromised) must be Listeria-free. All other foods may contain up to 100CFU/g. In January 2006, Commission Regulation (EC) No. 2073/2005 was implemented in the EU (Kalliopi et al., 2008). This policy defines a tolerance level of 100 CFU/g or ml for ready-to-eat products, and absence in 25g or ml for high-risk individuals. The UK and the USA, which insist on a complete absence of the pathogen in 25g of all foods (zero tolerance policy) argue that since the infectious dose of the pathogen is not known and may vary for different people based on, for example, age and morbidity status, it is not sound to define acceptable levels of the pathogen in foods. They maintain that the infectious dose must be known before any limits can be defined. In the USA, as a result of the zero-tolerance policy, L. monocytogenes is considered an adulterant. Therefore, any food that contains the organism can be considered adulterated and could be seized or recalled (Jay et al., 2005). Two arguments have been raised against the zero-tolerance policy. The first is that the incidence of listeriosis in the USA (~0.7 per 100,000 people) is the same as it is in the EU which allows <100CFU/g (Montville and Matthews, 2005). It appears, therefore, that the zero-tolerance offers no additional protection for consumers. Secondly, microbial specifications must be necessarily harmonized for the advancement of international trade. Therefore, under the zero-tolerance policy, foods that meet the 41 European Union’s <100CFU/g tolerance would be rejected in the USA and the UK. Consequently, these countries could be charged with inhibition of free trade. Another flaw in the zero tolerance policy is that current methods for detecting L. monocytogenes in foods have a 10 to 15% false negative rate (i.e. if 100 food samples containing L. monocytogenes were tested, the pathogen would not be detected in 10 to 15 of the samples) (Montville and Matthews, 2005). Questions have therefore been asked if it is sound to demand zero-tolerance when there is no 100% reliable test for detecting the organism in foods. On country-specific regulations, Great Britain established four quality categories for ready-to-eat foods as follows (Gilbert, 1992): Level in 25g Remarks Not detected Satisfactory <102 cells Fairly satisfactory 102-103 cells Unsatisfactory >103 cells Unacceptable Lammerding and Farber (1994) reported that a 1993 Canadian L. monocytogenes compliance criteria placed ready-to-eat foods into three groups: category 1 included products linked to outbreaks, category 2 included those that had a self-life >10 days, and category 3 included those that either supported growth with a self-life ≤10 days or those that did not support growth. 42 Germany is strongly against the zero-tolerance policy. It asserts that the demands of the policy are unrealistic (Jay et al., 2005). However, the country indicates that products that contain >104 of the pathogen per gram of food must be subjected to automatic recall. In France, foods for high-risk individuals must not contain L. monocytogenes in 25g samples. As generally argued by the EU, the French also assert that it is unrealistic to expect zero counts of L. monocytogenes in raw foods, particularly given the inevitable presence of the pathogen in food processing environments (Tompkin, 2002). The International Commission on Microbiological Specification for Foods (ICMSF, 1996) has concluded that if the counts of L. monocytogenes do not exceed 100CFU/g of food at the point of consumption, the food is considered acceptable for individuals who are not at risk. 2.9 Detection of L. monocytogenes in foods Generally, the detection of L. monocytogenes in foods involves enrichment, culturing on selective media, description of colonial characteristics on solid selective media, observation of cellular morphology, and finally biochemical and confirmatory tests (Prentice and Neaves, 1992). The enrichment step is considered particularly important as the organism is a poor competitor and would therefore not grow well in the presence of other microorganisms. Among the protocols developed for the detection and isolation of L. monocytogenes are the ISO 11290 method and those by the United States Department of Agriculture 43 (USDA), US Food and Drugs Administration (FDA), and the United Kingdom Health Protection Agency (HPA). Whereas the ISO, USDA, and FDA protocols require pre- enrichment (primary and secondary) of samples in either University of Vermont broth (UVM) or Fraser Broth (FDA) for both detection and enumeration, the HPA method allows pre-enrichment for detection only, and not for enumeration (HPA, 2009). Specifically, for detection and isolation, the USDA protocol requires primary enrichment in UVM broth, secondary enrichment in Fraser Broth, and plating on a selective medium. Colonies typical of L. monocytogenes are then taken through confirmatory identification procedures such as general biochemical tests, the Christie, Atkins, Munch-Peterson (CAMP) test, and genetic identification tests. The ISO and FDA protocols are similar in content. 2.10 Risk assessment In the simple terms, risk assessment is an evaluation of the possibility of an undesirable event occurring. The exercise is premised on the prior assumption that the possibility exists for exposure to an injurious event or substance. Technically, the injurious substance is called ‘a hazard’, and the likelihood that it would occur to cause the undesirable outcome is referred to as ‘risk’. According to the Codex Alimentarius Commission (CAC, 2003), a hazard is any biological, chemical or physical agent in, or condition of, food with the potential to cause an adverse health effect. The Commission also defines a risk as a function of the probability of an adverse health effect and the severity of that effect, consequential 44 to a hazard(s) in food. Therefore, risk assessment evaluates the likelihood that exposure to a hazard would result in adverse health effects. Risk assessment involves a systematic determination of the ability of a substance or organism to cause an adverse health effect (hazard identification), evaluation of the probability of consumption of the hazard and the quantities likely to be ingested through food (exposure assessment), description [in quantitative or qualitative terms] of the severity of the effect of the hazard following consumption (hazard characterization), and risk characterization, which is essentially a combination of the information gathered in the previous steps to fully explain the nature of the risk (Lammerding et al., 2001; CAC, 2003). The outcomes of risk assessments are used in risk management (establishment and implementation of appropriate control measures) and risk communication (exchange of information among stakeholders of the risk assessment process) (CAC, 2003). With regard to food safety, the commonest types of risk assessments are chemical and microbial risk assessment, which respectively focus on chemicals and pathogenic microorganisms as the hazards of concern. Regardless of the type, however, there are generally two forms of risk assessment, namely quantitative risk assessment and qualitative risk assessment. Whereas quantitative risk assessment offers numerical estimates of the risk (e.g. 1 in 100 chances of illness), qualitative risk assessments describe the nature of the risk (i.e. high, moderate, or low risk). 2.10.1 Microbial risk assessment The concept of microbial risk assessment (MRA) rests on the determination of the 45 likelihood of consumers falling ill after intake of a food containing a food-borne microbiological hazard. The process offers a description of a given food system, detailing production, commercialization, storage and consumption of the food. It identifies the microbiological hazard(s) associated with the system, and provides information on the possible transfer of the hazard(s) to consumers to cause harm (Lammerding et al., 2001). Lammerding (1998) indicated that MRA has unique features that differentiate it from chemical risk assessments. For example, unlike chemical risk assessments that may consider the cumulative effects of carcinogens and other toxicants causing chronic effects, MRA focuses on outcomes that are primarily the result of single exposures. Each exposure to a pathogen or its toxin represents an independent, non-cumulative event, resulting in outcomes ranging from asymptomatic infection to acute illness, chronic syndromes, or death. The outcomes of MRA serve as tools that are used to address the health and safety challenges presented by the identified food-borne hazard. 2.10.1.1 Sample microbial risk assessments (MRA) A number of major risk assessments have been conducted at country and international levels by individuals, governments and intergovernmental organizations. These include MRA for Listeria monocytogenes in ready-to-eat foods and deli meats, Salmonella enteritidis in eggs and egg products, Vibrio parahaemolyticus in oysters, Enterohemorrhagic Escherichia coli in ground beef, and Fluoroquinoline resistance in Campylobacter (FAO/WHO, 2004). 46 Lindqvist and Westoo (2000) conducted a quantitative risk assessment for L. monocytogenes in smoked salmon and rainbow trout in Sweden and found risks for listeriosis to be 2.0 x 10-3 for low-risk individuals and 1.6 x 10-2 for high-risk individuals. Joint WHO/FAO risk assessments include Salmonella enteritidis in broilers and eggs, Campylobacter sp in broilers, and L. monocytogenes in ready to eat foods (Buchanan, 2003). In the L. monocytogenes risk assessments, the RTE foods examined were milk, ice cream, cold-smoked fish, and fermented meat products. The risks of listeriosis per serving were 5.0 x 10-9 for milk, 1.4 x 10-11 for ice cream, 2.1 x 10-8 for smoked fish, and 2.5 x 10-12 for fermented meats (WHO/FAO, 2004). Smoked fish thus recorded the highest risk per serving. However, relative to the other foods, the consumption of smoked fish was indicated to be modest, hence the estimated total number of listeriosis due to the product was moderate (WHO/FAO, 2004). 47 CHAPTER THREE 3.0 METHODOLOGY 3.1 Study design The study comprised a survey and laboratory microbiological analyses. The survey solicited information on the consumption patterns (frequency and portion sizes) of hot-smoked fish (mackerel, tuna, and herrings), salted-dried tilapia (koobi), salted- dried ray fish (kako), salted and fermented fish (momoni), and sundried sardines among consumers in two coastal and an inland community in the Accra and Tema Metropolitan Assemblies to enable assessment of their exposure to Listeria monocytogenes. A total of 450 consumers, 150 each from Jamestown and Tema New Town (coastal communities) and Madina (inland community), were interviewed with semi-structured questionnaires (Appendix 1). Four (4) processors of each product type were also selected at processing sites and interviewed with semi-structured questionnaires (Appendix 2) on their processing methods and practices. Laboratory microbiological analyses involved detection and enumeration of Listeria monocytogenes as well as determination of general microbial counts (total plate count, total coliform count, and Escherichia coli count) in fish samples collected from processing sites, some informal markets in Accra and Tema, and food vendors1. Additionally, the ability of L. monocytogenes to survive domestic cooking was investigated in a challenge test, in which hot-smoked fish (tuna, herrings and mackerel), sundried sardines, and koobi were inoculated with Listeria monocytogenes NCTC 11994 at ≥108 CFU/ml in buffered peptone water, used to prepare some typical 1 General microbial counts were not determined for samples from food vendors 48 Ghanaian soups and then tested for the presence/absence of the pathogen after cooking. Standard tests (Board et al., 1992) were used to confirm all presumptive L. monocytogenes isolates. Aseptic procedures were followed during sampling and microbiological analyses. Estimation of the risk of consumers ingesting L. monocytogenes through consumption of the traditionally processed fish was done using data on the counts of the pathogen in the fish products and survey data on the portion sizes of fish often consumed at an instance. 3.2. Overview of survey and sampling sites Consumer and processor surveys were conducted in Jamestown and Tema New Town, two coastal fishing communities in the Accra and Tema Metropolitan Assembly respectively, known for their traditional fish processing activities, and in Madina, an inland community with a large central market patronized by several consumers in the Accra Metropolitan Assembly. Fish samples were obtained from three (3) inland informal markets (Madina, Kaneshie and Agbogbloshie markets) and two in the coastal communities, Jamestown and Tema New Town. The markets were selected by convenience from the list of markets to which processors indicated they sent their products. 49 Foods served with smoked fish in soup were also purchased from the University of Ghana Night Market, a major eating joint for students and some staff in the University, to determine the occurrence of the pathogen at the point of consumption. 3.2.1 Survey A total of 462 respondents consisting of 450 consumers and 12 processors were interviewed using pretested semi-structured questionnaires (Appendices 1 and 2). 3.2.1.1 Consumer survey The number of consumers, n, was determined using the following formula from McCabe and Moore (1993): n= (Zα/2/2m) 2 where Z0.025= 1.96, m (margin of error)= 5%, α= 0.05 C.I= 95% Using the indicated margin of error, m, level of significance, α, and confidence interval, CI, n was calculated as: n = [(1.96/0.1)2] n = 19.62 ≈ 202 n = 400 This number (n=400) was increased to 450 in order that 150 respondents each could be drawn from the three communities (Jamestown, Tema New Town, and Madina). The inclusion criteria were that prospective respondents be consumers of traditionally processed fish, be involved in the preparation and/or distribution of food in their households, and be able to estimate the portion sizes of the fish products often consumed by children aged 6 months to 6 years, the elderly (≥60 years), and pregnant women, if they had any of such persons in their households at the time of the survey. 50 To aid the estimation of portion sizes, packaged 50g samples of sundried sardines, smoked fish (tuna, mackerel, and herrings), and 10g samples of kako and koobi were shown to respondents. They were then asked to use those as models to estimate the quantities of the products they often consumed at an instance as multiples or fractions of the sizes shown them. A mould of plasticine (artificial clay) representative of 10g of momoni was used as a model for that product. Where respondents had children, the elderly and/or pregnant women in their households they were asked to make proxy estimations of the frequencies and portion sizes of the traditionally processed fish those individuals often consumed. 3.2.1.2 Processor survey Four (4) processors each of smoked fish, salted fish, and sun-dried sardines were selected by convenience and interviewed with semi-structured questionnaires on their methods of processing and general fish handling practices. The interviews were conducted at the processing sites to enable observation of the methods and practices they describe. 3.3 Sampling for microbiological analyses Four categories of samples were collected for laboratory microbiological analyses: fish from informal market; fish and water from some steps along the processing chains of the various products; freshly processed hot-smoked fish for the challenge test; and hot-smoked fish served with food from food vendors. 51 Samples collected were appropriately labelled and packed into thermos ice chest previously sanitized with 70% ethanol, and transported to the laboratory on ice for immediate analysis 3.3.1 Sampling of fish from informal markets Fifteen (15) samples of each fish product were purchased from Madina, Kaneshie, Agbogbloshie, Jamestown, and Tema New Town markets (total of 105 samples). On each sampling day, fish were purchased from the markets as typically sold to consumers. 3.3.2 Sample collection from processing Sites Fish and water samples (total 64) were collected from some steps along the traditional fish processing chain as indicated in Table 3.1. Gloves were worn in the process to prevent confounding contamination. Table 3.1: Sampling points along traditional fish processing chain Product Sampling point/sample type Number of samples Momoni  Fresh fish before washing 2  Fresh fish after washing 2  Fish after three days soaking in highly saturated salt solution 2  Water before use for washing 2  Water after use for washing 2  Fish after three days sun drying 2 Kako  Fresh ray fish before washing 2  Fresh ray fish after washing 2 52 Table 3.1 continued Kako  Water before use for washing 2  Water after use for washing 2  Ray fish after three days soaking in highly saturated salt solution 2  Fish after three days sun drying 2 Smoked Tuna  Fresh tuna before washing 2  Fresh tuna after washing 2  Water before use for washing 2  Water after use for washing 2  Tuna after smoking 2 Smoked Mackerel  Fresh mackerel before washing 2  Fresh mackerel after washing 2  Water before use for washing 2  Water after use for washing 2  Mackerel after smoking 2 Smoked Herrings  Fresh herring before washing 2  Fresh herring after washing 2  Water before use for washing 2  Water after use for washing 2  Mackerel after smoking 2 Sundried sardines  Fresh sardines before washing 2  Sardines after washing 2  Water before use for washing 2  Water after use for washing 2  Sardines after three days drying 2 Total 64 53 3.3.3 Sample collection for challenge test Hot-smoked tuna, mackerel and herrings were purchased from fish processors in Tema New Town immediately after processing (while still hot). The fish were collected directly from the smokers into sterile sample bags. Koobi and sundried sardines were purchased from the Madina market as typically sold to consumers. 3.3.4 Sample collection from food vendors Hot-smoked tuna, mackerel and herrings were purchased with soup and banku (stiff porridge made from fermented corn dough and cassava dough) from food vendors at the University of Ghana Night Market as normally dished to consumers. The market is a major point for the sale of street foods to students and some staff of the University. 3.4 Laboratory microbiological analyses Samples were analysed for the presence and concentration of L. monocytogenes using the United States Department of Agriculture protocols. Total plate count, total coliform count, and Escherichia coli counts were also determined for each sample except those purchased from food vendors. 3.4.1 Sample preparation and enrichment for L. monocytogenes detection and enumeration Twenty-five grams (25g) of each field sample of fish was weighed into sterile stomacher bags. To this quantity, 225mL of Listeria Enrichment Broth (LEB) was 54 added, and the diluent-fish mixture homogenized with a stomacher blender (Seward Stomacher®400 Circulator) for 60 seconds. The homogenate was incubated at 30oC for 24±2 hours. For water samples, 25mL of water was added to 225mL of LEB. After the incubation period, 0.1mL of the primary enrichment broth was dispensed into 10mL of Fraser broth and incubated at 37oC for up to 48 hours as the secondary enrichment step. 3.4.2 Plating of enriched cultures Using the spread plate technique, 0.1mL of the secondary enrichment broths were plated out on Oxford or Chromogenic agar plates and incubated aerobically at 37oC for up to 48 hours. 3.4.3 Enumeration Aliquots (0.1mL) of 10-1 homogenates of fish sample units, prepared as previously described, were plated out using the spread plate technique on Oxford agar plates or Chromogenic agar plates. The plates, after adding the inocula, were allowed to stand for 15 minutes to allow absorption of the inocula into the agar before incubating at 37oC for 24 – 48 hours. The incubated plates were examined and typical colonies of L. monocytogenes (brownish colonies with black halos on Oxford agar and greenish colonies with whitish halo on Chromogenic agar) were counted using a Quebec colony counter. Plates containing up to 150 colonies were considered useful for enumeration. 55 3.4.4 Presumptive and confirmatory identification Typical presumptive L. monocytogenes colonies were purified by streaking on nutrient agar and incubating the plates at 37oC for 24 hours (Board et al., 1992). Purified colonies from the nutrient agar plates were tested for their Gram reaction within 24 hours of visible growth. The Gram staining procedure described by Black (1999) was employed. Colonies identified as pure by their Gram reaction were used for subsequent confirmatory identification tests. 3.4.4.1 Catalase test The catalase test procedure described by Atlas (1997) was used. With a sterilized inoculation loop, pure colonies of L. monocytogenes were smeared on clean glass slides. Drops of 3% hydrogen peroxide were placed on the smears using a capillary pipette. The slides were then observed for gas bubbles. 3.4.4.2 Acid production from carbohydrates The fermentation of D-mannitol, D-xylose, L-rhamnose, and D-glucose by the isolates were determined. One tube each of the four carbohydrates (at 5% concentration in purple broth base) was inoculated with pure isolates. Inoculated broths were incubated at 37oC for up to 72 hours and observed for colour changes. Acid production manifests as a change in broth colour from purple to yellow (Board et al., 1992). 3.4.4.3 -haemolysis test Blood agar plates were streaked with presumptive L. monocytogenes isolates, incubated at 37oC for 18 hours, and observed for clear bands around the lines of streak 56 (Board et al., 1992), (indicative of red blood cell haemolysis). Positive L. monocytogenes isolates showed narrow bands away from lines of growth (Board et al., 1992). 3.4.4.4 Test for umbrella-like growth in semi-solid agar Sulphur Indole Motility (SIM) agar deep tubes were inoculated with presumptive L. monocytogenes by stabbing, incubated at 25oC for up to 48 hours, and observed for a characteristic growth pattern described by Board et al., (1992) as high turbidity about 1cm below the meniscus and away from the stab line such that the turbidity has the appearance of an open umbrella. 3.5 Survival of L. monocytogenes in domestic cooking (challenge test) The ability of L. monocytogenes to survive domestic cooking was tested by inoculating smoked fish (tuna, mackerel and herrings), koobi, and sundried sardines with Listeria monocytogenes NCTC 11994 and using the fish to prepare two kinds of soups commonly consumed in Ghanaian homes: one with a short cooking time (45 minutes; light soup) and another with a longer cooking time (75 minutes; groundnut soup). The composition of the soups (Table 3.2) and duration for cooking were determined through a focus group discussion with volunteers (five women with average age of 24 years) involved in food preparation in their homes. The discussion was held to determine the methods, average cooking times, and regular ingredients often used for preparing light soup and groundnut soup. 3.5.1 Inoculation of fish A broth of L. monocytogenes NCTC 11994 was prepared by incubating microdiscs of the pathogen in buffered peptone water at 37oC for 24 hours to amplify their numbers. 57 After incubation, the count of the pathogen was determined to ensure it was ≥108 CFU/ml. The culture was then poured onto the respective fish products in separate sterile stomacher bags. The broth-fish mixtures were shaken by hand and incubated at 37oC for 24 hours (a stomacher blender was not used to avoid breaking up the fish) to allow the pathogen to enter and grow in the fish tissues. The fish were used for preparing the soups after the incubation period. 3.5.2 Soup preparation Two sets of soups were prepared; experimental set prepared with fish contaminated with L. monocytogenes, and control set prepared with uncontaminated fish. Cooking started at the same time for all the soups. Total cooking time for light soup was forty five (45) minutes and that for groundnut soup was seventy five (75) minutes. The soups were prepared by some respondents from the focus group discussion. The procedures used to prepare the soups are as shown in Figs 3.1 and 3.2. Based on the outcome of the focus group discussion, the kinds and quantities of ingredients in Table 3.2 were used for preparing the soups. The volume of water used for each soup was 1040 mL. The control soups samples enabled evaluation of the sensory appeal of the soups cooked for the indicated durations. The light soups maintained an average temperature of 99.2oC during cooking whilst the groundnut soups maintained an average temperature of 106.5oC. 58 Table 3.2: Kinds and quantities of ingredients used for challenge test cooking Light soup Groundnut soup Ingredient Quantity (g) Ingredient Quantity (g) Tomatoes 74 Tomato paste 35 Garden eggs 171 Groundnut paste 150 Pepper 12 Pepper 5 Salt 7.5 Salt 7.5 Onion 61 Onion 50 Water Simmered tomatoes Wash tomatoes, garden eggs, onions, pepper Shred onions Boil (cook) for 40 minutes Light soup Water Fig. 3.1: Flow diagram for the preparation of light soup Blend garden eggs, pepper and simmered tomatoes Boil garden eggs and pepper till tender Salt Simmer tomatoes, shredded onions and fish for 5 minutes Fish 59 Immediately after cooking, the fish were removed from the soups and prepared for the detection of L. monocytogenes following the methods described in Section 3.4. 3.6 Quality control for microbiological analyses Parallel tests for the identification were conducted on a positive control (Listeria monocytogenes NCTC 11994) and a negative control (Enterococcus faecalis NCTC 775) (HPA, 2009). 3.7 Microbial Risk Assessment Protocols The Codex Alimentarius Commission (CAC, 2003) framework for risk assessment (Fig. 3.4) was used for the study. The hazard identification was done using available literature on the pathogen. Survey and laboratory analyses provided data for fulfilling the requirements of the remaining three steps of the Codex risk assessment protocol. Water (700mL) Water Heat groundnut paste with 200mL of water Wash pepper and onions Salt Boil for 15 minutes Boil for 60 minutes Blend pepper and onions with 140mL water Fish Groundnut soup Fig. 3.2: Flow diagram for the preparation of groundnut soup 60 3.7.1 Hazard identification The identified hazard in this study was Listeria monocytogenes, a Gram positive, microaerophilic, non-spore forming psychrotoph on which epidemiological evidence abounds on its pathogenicity (Lindqvist and Westoo, 2000; Lammerding et. al, 2001). 3.7.2 Exposure assessment The exposure of consumers to L. monocytogenes was assessed using data from the laboratory analyses and field survey to determine the prevalence and concentration of the pathogen in the fish, as well as the likely intakes of the pathogen. 3.7.2.1 Prevalence of L. monocytogenes This was determined as the percentage of fish samples in which the pathogen was detected. 3.7.2.2 Concentration of L. monocytogenes The concentrations of the pathogen were determined as the colony forming units of confirmed L. monocytogenes per gram of fish products. Hazard Identification Exposure Assessment Risk Characterization Dose-response Assessment Hazard Characterization Fig. 3.3: General risk assessment framework (CAC, 1998) 61 Pill = 1 – [1 + (N b)/β ] – α ----------------(2) where Pill = probability of illness N = dose of L. monocytogenes (i.e. likely number ingested, from Equation 1) α, β, b = model parameters α=0.25, b=2.14 (Bemrah et al., 1998) β=1010.98 for high-risk population β=1015.26 for low risk population (Bemrah et al., 1998) 3.7.2.3 Likely intakes of L. monocytogenes The likely numbers of the pathogen in fish at the time of consumption, N, were estimated as 3.7.3 Hazard characterization and dose-response assessment There were two risk outputs in this study: risk of ingestion of the pathogen, and risk of infection with the pathogen. The risk of ingestion gave an indication of the probable intakes of the pathogen through the fish products, whereas the risk of infection provided the probability of occurrence of disease following ingestion of the pathogen. The risk of ingestion was determined using Equation 1 (Section 3.7.2.3). The Weibull-Gamma dose-response model suggested by Farber et al., (1996), and used by Bemrah et al., (1998) to estimate the risk of listeriosis from consumption of soft cheese made from raw milk and Lindqvist and Westoo (2000) to estimate the risk of listeriosis from consumption of smoked salmon was used in the present study to provide estimations of the risk of infection. According to the model, the probability of illness from ingestion of L. monocytogenes is given by N = C x S ----------------------------------(1) where N = likely number of L. monocytogenes cells ingested C = CFU/g of L. monocytogenes in the fish product S = serving size of fish product (Lindqvist and Westoo, 2000) 62 Although there are other models (such as the exponential model), the Weibull-Gamma model has the advantage of being particularly more suitable for risk assessments on Listeria monocytogenes (Lindqvist and Westoo, 2000; Bemrah et al., 1998; Farber et al., 1996) and is thus used more often. For example, a disadvantage of the exponential model not found with Weibull-Gamm is that the former overestimates risks (Lindqvist and Westoo, 2000). 3.7.4 Risk characterization All the qualitative and quantitative information gathered in the previous steps were integrated to provide a scientifically sound description of the risk of ingesting Listeria monocytogenes through traditionally processed fish purchased from informal markets in Accra and Tema. 3.8 Data analyses Survey data were analysed using Microsoft Excel and SPSS v16. The identity of L. monocytogenes isolates were established by comparison of the results of the preliminary and confirmatory tests with those obtained for the control organisms (Listeria monocytogenes NCTC 11994 and Enterococcus feacalis NCTC 775). Distributions for the risk of illness resulting from ingestion of L. monocytogenes were constructed using @Risk 5.5 to make up for uncertainties and provide risk estimations in other possible levels of contamination and fish serving sizes. 63 CHAPTER FOUR 4.0 RESULTS AND DATA ANALYSIS 4.1 Key findings Consumer responses indicated that generally, smoked fish and salted fish were the most and least consumed products, respectively. For each product, consumption was reportedly higher in quantity and more frequent among adults (respondents, the elderly and pregnant women) than among children. The hygienic conditions of processing sites and fish handling practices were generally unsatisfactory. Most sites were located close to areas of unsanitary conditions, including puddles of dirty water. Processors hardly washed their hands before processing, and wash water for raw fish was reused when it ideally needed to be changed. Generally, the microbiological quality of fish improved through the processing steps to the finished products. For example, although Listeria monocytogenes and Escherichia coli were present in samples from the initial processing steps of smoking, they were not detected in samples collected immediately after smoking. However, in salted fish and sundried sardines, E. coli and L. monocytogenes were detected in some of finished product samples. Although the prevalence of L. monocytogenes in the fish products from informal markets was high (40-80%), the counts were generally low (102-3 CFU/g). Accordingly, the risks of ingestion and infection were found to be low. 64 4.2 Traditional fish processing 4.2.1 Demographic characteristics of processors All processors were females aged 20 years or more (Fig. 4.1) who had received some level of formal education (Fig. 4.2). Each processor had been involved in the business for at least six years (Fig. 4.3). , 25% 42% 33% Fig 4.1: Age of traditional fish processors in Jamestown and Tema New Town 20-29 years 30-39 years 40 and above 58% 42% Fig 4.2: Highest level of education attained by traditional fish processors in Jamestown and Tema New Town Primary MiddleSchool/JHS 65 4.2.2 Processing methods Traditional fish processing practices have not changed much over the years; the materials for and the means of processing observed in this study were not different from those reported earlier by Essuman (1982), Yankah (1988), Nketsia-Tabiri (1994), Cofie (2003) and Oppey (2002). The basic requirements for the processing of each fish product were fish, firewood, salt and sunlight. These were employed in different combinations for each product. 4.2.2.1 Procurement of fish Fish were either purchased from fisher folk at the sea shores where fresh fish is landed or from cold stores (Table 4.1). Low premium, old stock and almost-stale fish were also purchased from cold stores for momoni processing. Fresh mackerel and herrings were purchased from cold stores as frozen fish whiles tuna, ray fish for kako and sardines for sun-drying were purchased from fishmongers from the shores. Purchased fish were transported to processing sites by foot, in taxis or by trolleys (Table 4.1). 8% 33% 17% 42% Fig. 4.3: Number of years in traditional fish processing 6-10 years 11-15 years 16-20 years More than 20 years 66 Table 4.1: Purchasing of fresh fish for traditional processing Item Number of processors (%) Source of fish (purchasing point) Fishmongers 12 (100) Cold store 8 (67) Inspection of fish before purchasing 12 (100) Smell 8 (67) Skin surface for sliminess 3 (25) Duration of transportation to processing site Less than 30min 4 (33) 30min – 1 hr 6 (50) 1hr 30mins – 3 hrs 2 (17) Means of transportation By foot 12 (100) Trolleys 7 (58) Commercial vehicles (taxi) 12 (100) Cartons of frozen mackerel and herrings were mostly transported using taxi (or other commercial vehicles) whereas tuna, ray fish and sardines were carried in basins and transported by foot on head loads. 4.2.2.2 Smoking Regardless of the kind of fish (i.e. tuna, mackerel or herrings), the method of smoking was essentially the same. The fish were washed, degutted, cut into three pieces if desired (head, mid-portion, tail), washed, arranged on a smoker and smoked for 2-3 hours for a soft product (i.e. wet hot-smoked product, tuna and mackerel) or ≥12 hours for a smoke-dried product (i.e. dry hot-smoked product, herrings) (Fig.4.4 and 4.5). 67 Fig 4.4: Process flow diagram for hot-smoked tuna and mackerel Fresh tuna or mackerel Degut and wash Arrange on smoker in layers Air-dry for 15 mins Cover smoker with cardboard or slate of plywood Smoke for 2-3hrs Smoked tuna/mackerel Thaw mackerel Cut large fish into 2 or 3 pieces 68 Tuna and mackerel were considered sufficiently smoked when the skin colour was golden-brown and the flesh tender whiles herrings were considered sufficiently smoked when fish were considerably brittle. Soon after processing, finished products were either retailed by hawking or sold in bulk. Entire batches of smoked tuna and salmon were usually sold on the same day of processing as there were no appropriate storage facilities. Smoked herrings (smoke-dried) could however be kept on the smoking racks or in baskets for up to five days or more until sold out. Fig. 4.5: Flow diagram for hot-smoked herrings Fresh herrings Wash Arrange on smoker in layers Air-dry for 15mins Cover smoker with cardboard or slate of plywood Smoke (2-3hrs) Smoked herrings Dry further over embers in oven (≥ 10 additional hours) Thaw 69 Most processors used either metal ovens (Fig. 4.6) or concrete Chorkor smokers (Fig. 4.7). Fig. 4.6: Metallic ovens Fig. 4.7: Chorkor smoker 4.2.2.2.1 Hygiene of smoking environment The hygienic conditions of the smoke sites were generally unsatisfactory. Some of the sites were close to unsanitary shores where human defecation was common, and to puddles of dirty water (Fig. 4.8). There were no sanitary facilities and pipeborne water. Additionally, the grounds were not cemented, facilitating possible transfer of dust and sand into fish handled close to the ground. Fig. 4.8: Sanitation at smoking site Note stagnant water (arrowed) Note dirty gutter behind smoker (arrowed) 70 4.2.2.2.2 Handling practices during and after processing Fish handling practices were also unsatisfactory. Although all processors indicated that they washed their hands before processing, the practice was not observed. Apart from the actual smoking, most of the unit operations were carried out very close to the bare ground which could expose the fish to microbial contaminations. For example, frozen fish were thawed on open cartons on the bare ground (Fig. 4.9). In most instances, some of the fish came into direct contact with the soil on the ground. Additionally, water used for washing fresh fish was not changed as often as it should have (Fig. 4.10). Improper post-processing handling were also unhygienic. In some cases, freshly smoked fish were placed very close to the ground (Fig. 4.11.). Fig. 4.9: Thawing mackerel Note spillage on ground Fig. 4.10: Washing mackerel Note colour of water used 71 Fig. 4.11: Smoked mackerel ready for the market. Note nearness to bare ground 4.2.2.3 Sun-drying Sun-drying was the method of choice for the traditional processing of sardines. The process involved the simplest of unit operations. Landed sardines were simply washed and spread out on the bare ground and left to dry for 3 - 5 days, depending on weather conditions. The dried fish were swept and gathered into heaps, collected in baskets and were ready for marketing. 4.2.2.3.1 Sanitary conditions of processing It was observed that fresh sardines were not thoroughly washed, but were merely dumped into the wash water, scooped into baskets, and spread out on the bare ground. Wash water was obtained from the sea, and was used to wash several batches of fish without changing when it became necessary (Fig. 4.12). The fish were dried on the bare ground which was not zoned off or protected, and were therefore exposed to several contaminants (Fig. 4.13). This was unsatisfactory, particularly since no subsequent step was available in the processing flow to eliminate any pathogenic contaminations that could occur through contact of the fish with the ground. 72 Fig. 4.14 shows how sardines were collected after drying. Note person walking on drying grounds. This was a common practice since the drying grounds were not zoned off. 4.2.2.4 Salting Both kako (salted and dried ray fish) and momoni (salted and fermented fish) were processed essentially the same way. Fish were washed, salted for days under weights in salting vats (during which time some fermentation occurred in momoni processing) and sundried. Fig. 4.12: Washing fresh sardines. Note colour of water Fig. 4.13: Spreading washed sardines on the ground to dry Fig. 4.14: Sweeping dried sardines off the ground Fig. 4.15: Dried sardines collected in a basket 73 4.2.2.4.1 Momoni processing The process flow diagram for momoni is as shown in Fig. 4.16. Stale fish purchased from cold stores were used for momoni (Fig. 4.17). Fig. 4.16: Process flow for the production of momoni 4.2.2.4.1.1 Hygienic conditions of processing As observed in traditional smoking and sun-drying, wash water were in similar unsatisfactory hygienic conditions. Fish were either dried directly on the bare ground, Deteriorated fish Momoni Remove from basket and sundry (period weather dependent) Scale and degut Fill gut and gills with rock salt Place fish in salting vats Overlay fish with more rock salt, cover with synthetic sac and place weights on cover Ferment for 3 days Wash 74 or on nylon nets spread on the ground. The latter means did not seem to offer any prevention of fish contact with the soil. Hygiene of post-processing handling was poor (Fig. 4.18). 4.2.2.4.2 Kako processing Ray fish were cut into chunks (Fig. 4.19), washed, slit and filled with salt, arranged in salting vats and sundried (Fig. 4.20). Fig. 4.21 illustrates the processing of kako. Fig. 4.17: Stacks of stale fish being moved from a cold store for momoni processing Fig. 4.18: Gathering sufficiently dried momoni. Note processor stepping on fish (arrowed) Fig. 4.19: Cutting up ray fish Fig. 4.20: Chunks of ray fish being sun-dried after salting 75 Fig. 4.21: Processing flow diagram for kako 4.3 Occurrence of Listeria monocytogenes and other microorganisms in fish during traditional processing Generally, the microbiological quality of fish improved through the processing stages to the final products (Tables 4.2 – 4.5). In fresh tuna, ray fish, fresh sardines and fish for momoni processing, the average total plate counts ranged from 7.2x106–1.24x107 CFU/g. Average total coliform counts ranged from 3.0x106 – 8.5x106 CFU/g. No microbial growth was recorded for frozen fresh mackerel and herrings. However, L. Ray fish Wash Slit cut portions and fill with rock salt Place portions in salting vats Ferment for 3 days Cut into small pieces Remove from basket and sundry (period weather dependent) Cut Kako Overlay fish with more rock salt, cover with synthetic sac and place weights on cover 76 monocytogenes and the other microorganisms tested for in the study were detected in all other fresh fish samples (Tables 4.2 – 4.5). The microbiological quality of water used in processing was poor, as it contained high counts of total coliforms and E. coli (Tables 4.2 – 4.5). L. monocytogenes was also detected in all water samples. In finished products, the average total plate counts ranged from 0 – 2.7x106 CFU/g, whereas average total coliform and E. coli counts ranged from 0 – 1.90x106 CFU/g and 0-2.6x106 CFU/g, respectively. L. monocytogenes was not detected in any of the smoked fish sampled immediately after processing, but was found in some of the other products (Tables 4.2 – 4.5). 77 Table 4.2: Average microbial counts (x105 CFU/g) and detection of L. monoctyogenes in samples along smoked fish processing chain in Tema New Town Tuna Mackerel Herrings Sample TPC TCC E. coli Lm TPC TCC E. coli Lm TPC TCC E. coli Lm Fresh fish 113 64 24 + nd nd nd nd nd nd nd nd Fish after washing 90 40 19 + 12 13 6 + 21 13 6 + Water before washing 33 30 23 + 34 22 22 + 30 21 12 + Water after washing 54 65 31 + 55 45 33 + 44 24 10 + Fish after smoking nd nd nd nd nd nd nd nd nd nd nd nd nd: not detected TPC: total plate count Lm: Listeria monocytogenes +: detected Note: Counts are averages. Actual values are presented in Appendix 4 78 Table 4.3: Average microbial counts and detection of L. monoctyogenes in samples along salted fish processing chain in Tema New Town Momoni Kako Sample TPC TCC E. coli Lm TPC TCC E. coli Lm Fresh fish 124 85 47 + 129 84 65 + Fish after washing 119 63 29 + 106 63 51 + Water before washing 84 38 18 + 59 33 22 + Water after washing 151 73 23 + 95 98 53 + Fish after salting and fermenting 0 0 0 + 0 0 0 + Fish after sun-drying 16.5 11 8 + 19 5 3 + nd: not detected TPC: total plate count x105 CFU/g or ml TCC: total coliform count x102 CFU/g or ml, E. coli: E. coli count x102 CFU/g or ml Lm: Listeria monocytogenes +: detected Note: Counts are averages. Actual values are presented in Appendix 4 79 Table 4.4: Average microbial counts and detection of L. monocytogenes in samples along sun-drying processing chain in Tema New Town Sample TPC TCC E. coli Lm Fresh fish 72 30 26 + Tuna after washing 57 28 26 + Water before washing 31 18 5 + Water after washing 78 21 14 + Fish after sun-drying 27 19 26 + nd: not detected TPC: total plate count x105 CFU/g or ml TCC: total coliform count x102 CFU/g or ml, E. coli: E. coli count x102 CFU/g or ml Lm: Listeria monocytogenes +: detected Note: Counts are averages. Actual values are presented in Appendix 4 80 Table 4.5: Average microbial counts and detection of L. monoctyogenes in samples along smoked fish processing chain in James Town Tuna Mackerel Herrings Sample TPC TCC E. coli Lm TPC TCC E. coli Lm TPC TCC E. coli Lm Fresh fish 94 45 17 + nd nd nd nd nd nd nd nd Tuna after washing 83 31 16 + 6 3 2 nd 12 6 2 + Water before washing 44 26 15 + 35 18 12 + 31 14 8 + Water after washing 56 48 27 + 43 26 24 + 37 22 10 + Fish after smoking nd nd nd nd nd nd nd nd nd nd nd nd nd: not detected TPC: total plate count x105 CFU/g or ml TCC: total coliform count x102 CFU/g or ml, E. coli: E. coli count x102 CFU/g or ml Lm: Listeria monocytogenes +: detected Note: Counts are averages. Actual values are presented in Appendix 4 81 4.4 Occurrence of Listeria monocytogenes and other microorganisms in fish sold at informal markets Listeria monocytogenes, Escherichia coli and total coliforms were detected at varying levels in fish samples purchased from the markets (Table 4.6). The trend in order of decreasing counts was total plate count > total coliform count > E. coli count > L. monocytogenes count. Generally, the highest counts of L. monocytogenes occurred in salmon and tuna, and the least counts in the salted fish products and sundried sardines. In some fish samples (especially salted fish and sundried fish), L. monocytogenes was not detected by enumeration (direct culturing without preliminary enrichment) but was detected in the same samples plated out after primary and secondary enrichment in Listeria Enrichment Broth and Fraser Broth respectively (Table 4.6). 4.5 Detection of Listeria monocytogenes in street food samples Listeria monocytogenes was not detected in any of the three street food samples (banku with groundnut soup). Food vendors kept soups containing the fish constantly on fire during sales. This was done to ensure that soups were hot to meet consumer preference, as, generally, consumers in the University of Ghana would not patronize the foods if they are not hot. 82 Table 4.6: Detection and average counts of Listeria monocytogenes and average general microbiological counts in fish samples purchased from informal markets in Accra and Tema MKT Count Tuna Mackerel Herrings Kako Koobi Momoni Sundried sardines M a d in a TPC (x105 CFU/g) 87 76 47 26 21 43 76 TCC (x104 CFU/g) 39 30 35 12 11 13 52 E. coli (x103 CFU/g ) 25 24 27 7 4 3 31 Lm (x102 CFU/g) 21 29 8 7 6 3 3 Lm Det + + + + + + + + + + + + + + + + + + + K a n es h ie TPC (x10 5 CFU/g) 89 86 81 58 57 37 97 TCC (x104 CFU/g) 49 35 41 22 20 20 50 E. coli (x103 CFU/g ) 20 22 25 11 11 12 25 Lm (x102 CFU/g) 17 16 0 8 9 5 nd Lm Det + + + + + + + + + + + + + + + A g b o g b lo sh ie TPC (x105 CFU/g) 82 71 48 34 24 22 42 TCC (x104 CFU/g) 31 28 55 12 10 8 89 E. coli (x103 CFU/g ) 30 23 14 1 0 0 80 Lm (x102 CFU/g) 8 10 2 1 0 0 0 Lm Det + + + + + + + + + + + + J a m es to w n TPC (x105 CFU/g) 83 75 68 26 23 27 74 TCC (x104 CFU/g) 23 40 25 20 17 20 47 E. coli (x103 CFU/g ) 19 18 19 9 9 6 18 Lm (x102 CFU/g) 12 9 9 1 2 2 2 Lm Det + + + + + + + + + + + + + T em a N T TPC (x105 CFU/g) 72 79 50 28 15 15 80 TCC (x104 CFU/g) 39 43 23 14 3 11 43 E. coli (x103 CFU/g ) 39 28 17 2 1 0 33 Lm (x102 CFU/g) 2 4 3 1 0 0 3 Lm Det + + + + + nd + + + TPC: total plate count nd: not detected TCC: total coliform count Lm: Listeria monocytogenes count Lm Det: Detection of L. monocytogenes by enumeration +: L. monocytogenes detected in one sample by enrichment, 0 CFU/g recorded in two samples + +: L. monocytogenes detected in two samples by enrichment, 0 CFU/g recorded in one sample + + +: L. monocytogenes detected in three samples by both enrichment and enumeration Tema NT: Tema New Town MKT: Market 83 4.6 Challenge test Table 4.7 shows that Listeria monocytogenes NCTC 11994 survived in koobi, sun- dried fish and in smoked fish (tuna, mackerel and herrings) used in the experimental cooking. Table 4.7: Survival of Listeria monocytogenes NCTC 11994 in fish used in experimental domestic cooking Experiment Soup Fish 1 2 3 Groundnut soup Smoked tuna + + nd Smoked mackerel nd + + Smoked herrings + nd nd Koobi + + np Sun-dried sardines + + nd Light soup Smoked tuna + + + Smoked mackerel nd + nd Smoked herrings + nd + Koobi nd + np Sun-dried sardines + nd nd + : L. monocytogenes detected in fish nd: not detected np: not plated (fish broke up in soup) 4.7 Characterization of isolates The characteristics of the presumptive L. monocytogenes isolates (Brownish/grey colonies with black halo on Oxford Agar Plates (Fig. 4.23) or greenish colonies with pale yellow halo on Chromagar plates) are as summarized in Table 4.8. Colonies with a green metallic sheen on Eosin Methylene Blue incubated at 44.5oC for 24 hours were considered presumptive for E. coli (Fig. 4.24). 84 Fig. 4.24: Escherichia coli isolates on eosin methylene blue agar plate Fig. 4.23: Presumptive L. monocytogenes colonies on Oxford Agar Plate (a), and a control Oxford Agar Plate (b) Fig. 4.22: Fraser broths after 24h incubation. Black tubes may contain Listeria species, yellow tubes do not (a) (b) 85 Table 4.8: Characteristics of presumptive Listeria isolates Sugar Fermentation Umbrella motility β -H Presumptive organism Confirmed organism Isolate Gram reaction, shape Catalase reaction D- Glucose Mannitol Xylose Rhamnose P001 +, rods + + + + - + + L. monocytoge nes L. monocytogenes P002 +, rods + + + - - - - Listeria sp. P003 +, rods + + + - - - - Listeria sp. P004 +, rods + + - + - - - Listeria sp. P005 +, rods + + - - - - - Listeria sp. β – H: β haemolysis +: sugar fermented - :sugar not fermented P001 – 005: Listeria sp. isolates 1 – 5. 86 4.8 Demographic characteristics of respondents Table 4.9 presents a summary of the demographic characteristics of respondents. ‘Respondents’ as used in subsequent text refers to individuals actually interviewed in households visited. In addition to providing information about their own consumption patterns, the respondents made proxy estimations of the frequencies of consumption of traditionally processed fish, and the quantities thereof, by the elderly (≥60 years old), children (6months – 6 years), and pregnant women in their households. 4.9 Consumption of traditionally processed fish Generally, smoked fish products were consumed more often and in greater quantities than sun-dried sardines and the salted fish products (momoni and koobi) in all the communities visited. Sections 4.9.1 – 4.9.2 show the respective consumption patterns in Jamestown, Tema New Town and Madina. 4.9.1 Jamestown In Jamestown, among the smoked fish, mackerel and herrings were the most and least frequently consumed by the respondents, respectively. Frequency of consumption of kako was highest among the salted fish products. The summaries of the frequencies of consumption and quantities most often consumed at an instance are presented in Table 4.10. The respective frequencies and quantities of consumption of the products by respondents and the reported consumption patterns among the elderly, children, and pregnant women are presented in Appendix 3. 87 Table 4.9: Summary of demographic characteristics of respondents Community Age Gender Elderly (≥60 yrs) present in household Pregnant woman present in household Children (6mo-6yrs) present in household a b c d e Male Female Yes No Yes No Yes No Tema New Town 1 44 29 11 15 26 74 70 30 33 67 64 36 Jamestown 6 39 33 13 9 37 63 68 32 39 61 62 38 Madina 4 27 35 13 21 23 77 53 47 31 69 75 25 yrs = years old mo = months old a = Less than 20 yrs b = 20 – 29 yrs c = 30 – 39 yrs d = 40 – 40 yrs e = 50 yrs or more All values represent percentages of respondents 88 Table 4.10: Summaries of highest frequencies of consumption of traditionally processed fish and highest quantities of the products consumed at an instance in Jamestown Product Highest Frequency (%) Most frequently consumed quantities (%) * R es p o n d en ts Smoked tuna 2-3 times a week (70) More than 200g (63.3%) Smoked mackerel 2-3 times a week (60) 151-200g (40) Smoked herrings Once a week (50.7) 151-200g (58.7) Sundried sardines Once a month 101-150g (40) Kako Once a month 5-10g (53.7) Momoni Once a week (48.7) 5-10g (52) Koobi Once a month (65.3) 41-50g (45.2) E ld er ly Smoked tuna Daily (52) 151-200g (35.6) Smoked mackerel 2-3 times a week (51.9) 151-200g (69.2) Smoked herrings Once a week (76.3) 101-150g (44.2) Sundried sardines Once a month (43.3) 101-150g (49) Kako Never (68.3) 5-10g (51.5) Momoni Never (66) 5-10g )51) Koobi Never (63.5) 21-30g (60.5) P re g n a n t w o m en Smoked tuna Daily (58.6) More than 200g (79.3) Smoked mackerel 2-3 times a week (60.3) 151-200g (32.8) Smoked herrings Once a week (58.6) 101-150g (48.3) Sundried sardines Once a week (58.6) 151-200g (55.1) Kako Once a week (69) 5-10g (52) Momoni Once a week (93.1) 5-10g (38.9) Koobi Once a month (77.6) More than 50g (48.3) C h il d re n Smoked tuna 2-3 times a week (76.3) 101-150g (73.2) Smoked mackerel 2-3 times daily (66) 101-150g (63.9) Smoked herrings Once a week 50-100g (74.2) Sundried sardines Once a month (84.5) 50-100g (100) * figures in parenthesis represent percentages of consumers 89 4.9.2 Tema New Town Similar to the consumption pattern in Jamestown, generally, smoked fish were consumed more frequently and in greater quantities than the dried and salted fish products. Table 4.11 summarizes the highest frequencies and quantities of consumption of the respective traditionally processed fish in Tema New Town. Individual differences in the actual frequencies and quantities of consumption of the products among respondents and values of the same reported for the elderly, children, and pregnant women are presented in Appendix 3. 4.9.3 Madina On the whole, Madina recorded the least frequencies of consumption of the traditionally processed fish. The pattern of consumption was however similar to those reported for Jamestown and Tema New Town; smoked fish was consumed more frequently and in greater quantities whilst salted fish was the least consumed (Table 4.12). The respective quantities and frequencies of consumption of the fish products are shown in Appendix 3. 90 Table 4.11: Summaries of highest frequencies of consumption of traditionally processed fish and highest quantities of the products consumed at an instance in Tema New Town Product Highest Frequency (%) Most frequently consumed quantities (%) * R es p o n d en ts Smoked tuna 2-3 times a week (76) More than 200g (51) Smoked mackerel 2-3 times a week (62) More than 200g (45) Smoked herrings Once a week (51) 101-150g (43) Sundried sardines Once a month (53) 101-150g (37) Kako Once a month (32) 5-10g (40) Momoni Once a week (57) 11-20g (35) Koobi Once a month (54) 41-50g (37) E ld er ly Smoked tuna 2-3 times a week (76) 151-200g (42) Smoked mackerel 2-3 times a week (59) 151-200g (58) Smoked herrings Once a week (49) 101-150g (44) Sundried sardines Once a week (41) 101-150g (57) Kako Never (65) 5-10g (50) Momoni Once a month (33) 5-10g (52) Koobi Never (65) 21-30g (63) P re g n a n t w o m en Smoked tuna 2-3 times a week (68) More than 200g (60) Smoked mackerel 2-3 times a week (72) More than 200g (54) Smoked herrings Once a week (56) 101-150g (52) Sundried sardines Once a week (54) 101-150g (46) Kako Once a week (74) 5-10g (78) Momoni Once a week (96) 5-10g (44) Koobi Once a month (52) More than 50g (48) C h il d re n Smoked tuna 2-3 times a week (76) 101-150g (73) Smoked mackerel 2-3 times a week (66) 101-150 (64) Smoked herrings Once a week (76) 50-100 (74) Sundried sardines Once a month (84) 50-100g (100) * figures in parenthesis represent percentages of consumers 91 Table 4.12: Summaries of highest frequencies of consumption of traditionally processed fish and highest quantities of the products consumed at an instance in Madina Product Highest Frequency (%) Most frequently consumed quantities (%)* R es p o n d en ts Smoked tuna Once a week (46) More than 200g (70) Smoked mackerel 2-3 times a week (48) More than 200g (47) Smoked herrings Once a month (63) 101-150g (58) Sundried sardines Once a month (83) 101-150 (43) Kako Once a month (47) 5-10g (51) Momoni Once a month (63) 41-50g (36) Koobi Never (45) 5-10g (67) E ld er ly Smoked tuna Once a week (66) 151-200g (46) Smoked mackerel Once a week (71) 151-200g (68) Smoked herrings Once a week (64) 101-150g (48) Sundried sardines Once a month (51) 101-150g (46) Kako Never (77) 11-20g (66) Momoni Never (59) 21-30g (58) Koobi Never (73) 5-10g (57) P re g n a n t w o m en Smoked tuna Once a week (76) More than 200g (78) Smoked mackerel 2-3 times a week (72) 151-200g (48) Smoked herrings Once a week (100) 151-200g (41) Sundried sardines Once a month (74) 50-100g (70) Kako Once a month (48) 5-10g (53) Momoni Once a month (91) More than 50g (72) Koobi Once a week (50) 5-10g (67) C h il d re n Smoked tuna Once a week (44) 50-100g (66) Smoked mackerel Once a week (71) 50-100g (75) Smoked herrings Once a month (63) 50-100g (79) Sundried sardines Once a month (51) 50-100g (97) * figures in parenthesis represent percentages of consumers 92 4.10 Exposure assessment for Listeria monocytogenes 4.10.1 Risk pathways and event trees The points along the fish processing and/or distribution chain at which the Listeria monocytogenes could occur were determined to establish the possible route through which the pathogen is transferred to consumers (Figs. 4.25 – 4.27, in which Blue text: possible point of contamination Green text: possible point of pathogen elimination Red text: risk of ingestion) Raw fish Washing, cutting Arrangement on smokers Overlay with cardboards Smoke for 2-3 hours) Retail Domestic cooking Consumption as is [Risk of ingestion] Fig. 4.25: Risk pathway for traditionally smoked fish (tuna, mackerel, herrings) 93 Raw fish Washing Soaking in brine Sun-drying on bare ground Retail Domestic cooking Fig. 4.26: Risk pathway for traditionally salted fish Raw fish Washing Sun-drying on bare ground Retail Domestic cooking Consumption as is is (possible elimination of pathogens [Risk of ingestion] during cooking) Fig. 4.27: Risk pathway for sundried fish 94 Following the establishment of the risk pathways, an event tree, which is a sequence of events that could result in ingestion of the hazard, was constructed for each product to enable determination of the likelihood of ingesting the hazard (Fig. 4.28 – 4.30). Event tree analysis is based on binary logic, in which an event either has or has not happened and is useful in determining the risk of ingesting L. monocytogenes. Yes Risk of ingestion Yes No Ingestion not likely Fish eaten as is (without heating) Fish contaminated with Lm on informal markets No Yes Lm survives in fish during cooking N = CxS Ϯ Risk of ingestion N = CxS Ϯ Ingestion not likely No ϮEquation 1, Section 3.7.2.3, Chapter Three Fig. 4.28: Event tree for risk of ingestion of Listeria monocytogenes through consumption of traditionally smoked fish purchased from informal markets N=number of cells C=CFU of L. monocytogenes S=Serving size of fish 95 Yes No Fish contaminated with Lm on informal markets Ingestion not likely No Yes Lm survives in fish during cooking Risk of ingestion N = CxS Ingestion not likely Fig. 4.29: Event tree for risk of ingestion of Listeria monocytogenes through consumption of salted fish purchased from informal markets o 96 4.10.2 Prevalence of Listeria monocytogenes in the fish products As indicated in Section 3.7.2 (Chapter Three), the prevalence and concentration of L. monocytogenes in the fish products, as well as the likelihood of ingestion of the pathogen through the products were used as the basis for the exposure assessment. Table 4.13 shows the number of fish samples from markets in which the pathogen was detected. The values for individual prevalence of the pathogen in each product from the respective markets are presented in Appendix 5. Table 4.13: Average prevalence of L. monocytogenes in traditionally processed fish purchased from some informal markets in Accra and Tema Product Number of samples purchased Number of samples positive for L. monocytogenes Prevalence of L. monocytogenes (%) Smoked tuna 15 12 80 Smoked mackerel 15 14 93 Smoked herrings 15 10 67 Sundried sardines 15 9 60 Koobi 15 6 40 Kako 15 8 53 Momoni 15 8 53 4.10.3 Concentration of Listeria monocytogenes in the fish products The concentrations of L. monocytogenes in the respective traditionally processed fish have been presented in Table 4.6. The counts were generally low, ranging from 102 to 10 3 CFU/g. 97 4.10.4 Ingestion of L. monocytogenes through the fish products For all products, the estimation of likely numbers of the pathogen ingested was based on the following general assumptions: i. either traditionally processed fish is consumed as purchased from informal markets OR traditionally processed fish purchased from informal markets is not heat-treated to an extent that guarantees elimination of L. monocytogenes cells initially present in the product (i.e. the pathogen survives domestic cooking) ii. all strains of L. monocytogenes in the fish products are virulent For kako and momoni, an additional assumption was that the quantities reported by the respondents are actually consumed whole, although the products are used as condiments and usually break up in soups and are thus not available for direct consumption. This enabled determination of likelihood of ingestion in a worst case scenario. From Equation 1 (Section 3.9.2.3, Chapter Three), the likely numbers of L. monocytogenes ingested through consumption of the respective traditionally processed fish was calculated as Therefore, using the average concentrations of L. monocytogenes in each product (Table 4.6) and the highest portion sizes of each product most often consumed at an instance N = C x S ----------------------------------(1) where N = likely number of L. monocytogenes cells ingested C = CFU/g of L. monocytogenes in the fish product S = serving size of fish product (Lindqvist and Westoo, 2000) 98 (Tables 4.9 – 4.12), the likely numbers of the pathogen ingested through consumption of each product were calculated for the communities in which the surveys were conducted (Jamestown, Tema New Town and Madina) and are presented in Tables 4.14 – 4.16. Generally, in all three communities, consumers were likely to ingest more L. monocytogenes cells through the smoked fish products than through dried fish and salted fish (in that order). This was because the smoked fish products were consumed more frequently and in greater quantities than the dried fish and salted fish. Additionally, the concentration of L. monocytogenes were higher in the smoked fish products. 99 Table 4.14: Likely numbers of Listeria monocytogenes ingested through consumption of traditionally processed fish in Jamestown Product Average Lm CFU/g, (C) Quantity (g) most often consumed at an instance (S) Likely number of Lm ingested, N=(C).(S) R es p o n d en ts Smoked tuna 1.20x103 > 200 > 2.40x105 Smoked mackerel 9.00x102 151-200 1.36x105 – 1.80x105 Smoked herrings 9.00x102 151-200 1.36x105 – 1.80x105 Sundried sardines 2.00x102 101-150 2.02x104 – 3.00x104 Kako 1.00x102 5-10 5.00x102 – 1.00x103 Momoni 2.00x102 5-10 1.00x103 – 2.00x103 Koobi 2.00x102 41-50 8.20x103 – 1.00x104 E ld er ly Smoked tuna 1.20x103 151-200 1.81x105 – 2.40x105 Smoked mackerel 9.00x102 151-200 1.36x105 – 1.80x105 Smoked herrings 9.00x102 101-150 9.09x104 – 1.35x105 Sundried sardines 2.00x102 101-150 2.02x104 – 3.00x104 Kako 1.00x102 5-10 5.00x102 – 1.00x103 Momoni 2.00x102 5-10 1.00x103 – 2.00x103 Koobi 2.00x102 21-30 4.20x102 – 6.00x103 P re g n a n t w o m en Smoked tuna 1.20x103 > 200 >2.40x105 Smoked mackerel 9.00x102 151-200 1.36x105 – 1.80x105 Smoked herrings 9.00x102 101-150 9.09x104 – 1.35x105 Sundried sardines 2.00x102 151-200 3.02x104 – 4.00x104 Kako 1.00x102 5-10 5.00x102 – 1.00x103 Momoni 2.00x102 5-10 1.00x103 – 2.00x103 Koobi 2.00x102 > 50 > 1.00x104 C h il d re n Smoked tuna 1.20x103 101-150 1.21x105 – 1.80x105 Smoked mackerel 9.00x102 101-150 9.09x104 – 1.35x105 Smoked herrings 9.00x102 50-100 4.50x104 – 9.00x104 Sundried sardines 2.00x102 50-100 1.00x104 – 2.00x104 100 Table 4.15: Likely numbers of Listeria monocytogenes ingested through consumption of traditionally processed fish in Tema New Town Product Average Lm CFU/g, (C) Quantity (g) most often consumed at an instance (S) Likely number of Lm ingested, N=(C).(S) R es p o n d en ts Smoked tuna 2.00x102 > 200 > 4.00x104 Smoked mackerel 4.00x102 151-200 6.04x104 – 8.00x104 Smoked herrings 3.00x102 151-200 4.53x104 – 6.00x104 Sundried sardines 3.00x102 101-150 3.03x104 – 4.50x104 Kako 1.00x102 5-10 5.00x102 – 1.00x103 Momoni 0 5-10 0 Koobi 0 41-50 0 E ld er ly Smoked tuna 2.00x102 151-200 3.02x104 – 4.00x104 Smoked mackerel 4.00x102 151-200 6.04x104 – 8.00x104 Smoked herrings 3.00x102 101-150 3.03x104 – 4.50x104 Sundried sardines 3.00x102 101-150 3.03x104 – 4.50x104 Kako 1.00x102 5-10 5.00x102 – 1.00x103 Momoni 0 5-10 0 Koobi 0 21-30 0 P re g n a n t w o m en Smoked tuna 2.00x102 > 200 > 4.00x104 Smoked mackerel 4.00x102 151-200 6.04x104 – 8.00x104 Smoked herrings 3.00x102 101-150 3.03x104 – 4.50x104 Sundried sardines 3.00x102 151-200 4.53x104 – 6.00x104 Kako 1.00x102 5-10 5.00x102 – 1.00x103 Momoni 0 5-10 0 Koobi 0 > 50 0 C h il d re n Smoked tuna 2.00x10 2 101-150 2.02x102 – 3.00x104 Smoked mackerel 4.00x102 101-150 4.04x104 – 6.00x104 Smoked herrings 3.00x102 50-100 1.50x104 – 3.00x104 Sundried sardines 3.00x102 50-100 1.50x104 – 3.00x104 101 Table 4.16: Likely numbers of Listeria monocytogenes ingested through consumption of traditionally processed fish in Madina Product Average Lm CFU/g, (C) Quantity (g) most often consumed at an instance (S) Likely number of Lm ingested, N=(C).(S) R es p o n d en ts Smoked tuna 2.10x103 > 200 > 4.20x105 Smoked mackerel 2.90x103 151-200 4.38x105 - 5.80x105 Smoked herrings 8.00x102 151-200 1.21x104 - 1.60x104 Sundried sardines 3.00x102 101-150 3.03x104 – 4.50x104 Kako 7.00x102 5-10 3.50x103 – 7.00x103 Momoni 3.00x102 5-10 1.50x103 – 3.00x103 Koobi 6.00x102 41-50 2.50x104 – 3.00x104 E ld er ly Smoked tuna 2.10x103 151-200 3.20x105 – 4.20x105 Smoked mackerel 2.90x103 151-200 4.40x105 – 5.80x105 Smoked herrings 8.00x102 101-150 8.08x104 – 1.20x105 Sundried sardines 3.00x102 101-150 3.03x104 – 4.50x104 Kako 7.00x102 5-10 3.50x103 – 7.00x103 Momoni 3.00x102 5-10 1.50x103 – 3.00x103 Koobi 6.00x102 21-30 1.26x104 – 1.80x104 P re g n a n t w o m en Smoked tuna 2.10x103 > 200 > 4.20x105 Smoked mackerel 2.90x103 151-200 4.40x105 – 5.80x105 Smoked herrings 8.00x102 101-150 8.08x104 – 1.20x105 Sundried sardines 3.00x102 151-200 4.53x104 – 6.00x104 Kako 7.00x102 5-10 3.50x103 – 7.00x103 Momoni 3.00x102 5-10 1.50x103 – 3.00x103 Koobi 6.00x102 > 50 > 3.00x104 C h il d re n Smoked tuna 2.10x10 3 101-150 2.12x105 – 3.15x105 Smoked mackerel 2.90x103 101-150 2.93x105 – 4.35x105 Smoked herrings 8.00x102 50-100 4.00x104 – 8.00x104 Sundried sardines 3.00x102 50-100 1.50x104 – 3.00x104 102 4.11 Hazard Characterization and Dose Response Assessment The assumptions employed for the evaluation of the nature of the possible adverse health effect resulting from consumption of traditionally processed fish contaminated with L. monocytogenes were: i. high risk individuals (children, elderly, pregnant women) must consume fish containing more than 104 CFU/g of L. monocytogenes to suffer listeriosis (Buchanan et al., 1997) ii. low risk individuals (non-pregnant, apparently healthy individuals) must consume fish containing more than 109 CFU/g of L. monocytogenes to get ill (Schlech, 1999). Based on the foregoing assumptions, none of the consumers (neither high risk nor low risk individuals) in all three communities surveyed was at risk of suffering listeriosis from consumption of traditionally processed fish, since the order of the count of the pathogen in the products was 102-3 CFU/g in Jamestown and Madina, and 102 CFU/g in Tema New Town, levels less than the indicated limits of >104 CFU/g for high risk individuals and >109 CFU/g for low risk individuals. However, since the pathogen was detected in the fish products and the products were consumed frequently in all surveyed communities, the probabilities of illness, Pill, resulting from ingesting doses (N) of the pathogen in the products were calculated using the Weibull-Gamma model (Section 3.9.3, Chapter Three). The assumption employed for this calculation was that repeated exposure to the doses of L. monocytogenes in Tables 4.14 – 4.16 (i.e. 103-5 cells per eating instance) could result in illness. Therefore, if those 103 doses could cause illness, the probabilities of occurrence of the disease would be as presented in Tables 4.17 – 4.19. Among high risk consumers (children, the elderly and pregnant women), the probability of illness increased as the likely dose reached ≥104 cells. The general ranges of probability of illness for this group were 10-4 (for dose 103 cells) – 0.9850 (for dose 105 cells) in Madina, 10-6 (for dose 102 cells) – 0.4529 (for dose 104 cells) in Tema New Town, and 10-5 (for dose 103 cells) – 0.9583 (for dose 105 cells) for Jamestown (Tables 4.17 – 4.19). This supports the assumption that doses ≥104 CFU/g result in illness among high risk individuals (Buchanan et al., 1997). For the low risk group (respondents, majority of who were aged 20-39yrs), the probabilities of illness were even lower, as expected. Values were 10-9 (for dose 103 cells) – 10-3 (for dose 105 cells) in Madina, 10-10 (for dose 102 cells) – 10-5 (for dose 104 cells) in Tema New Town, and 10-9 (for dose 103 cells) – 10-4 (for 105 cells) (Tables 4.17– 4.19). This also supports the assumption that doses ≥ 109 CFU/g cause illness in low risk individuals (Schlech, 1999). 104 Table 4.17: Probability of illness among consumers in Jamestown Consumers Product Lower N Pill Upper N Pill C h il d re n Smoked tuna 1.21x105 0.7120 1.80x105 0.2301 Smoked mackerel 9.09x104 0.5334 1.35x105 0.1588 Smoked herring 4.50x104 0.1766 9.00x104 0.0837 Sundried sardines 1.00x104 0.0081 2.00x104 0.0041 E ld er ly Smoked tuna - - 2.40x105 0.3102 Smoked mackerel 1.36x105 0.7763 1.80x105 0.2301 Smoked herrings 1.36x105 0.7763 1.80x105 0.2301 Sundried sardines 2.02x104 0.0357 3.00x104 0.0097 Kako 5.00x102 1.34x10-5 1.00x103 6.89x10-6 Koobi 8.20x103 0.0053 1.00x104 0.0009 Momoni 1.00x103 5.89x10-5 2.00x103 3.03x10-5 P re g n a n t w o m en Smoked tuna - - 2.40x105 0.3102 Smoked mackerel 1.36x105 0.7763 1.00x105 0.1001 Smoked herrings 9.09x104 0.5334 1.35x105 0.1588 Sundried sardines 3.02x104 0.0814 4.00x104 0.0177 Kako 5.00x102 1.34x10-5 1.00x103 6.89x10-6 Koobi - - 1.00x104 0.0009 Momoni 1.00x103 5.89x10-5 2.00x103 3.03x10-5 R es p o n d en ts Smoked tuna - - 2.40x105 4.48x10-5 Smoked mackerel 1.36x105 0.0001 1.80x105 2.42x10-5 Smoked herrings 1.36x105 0.0001 1.80x105 2.42x10-5 Sundried sardines 2.02x104 1.92x10-6 3.00x104 5.24x10-7 Kako 5.00x102 7.02x10-10 1.00x103 3.61x10-10 Koobi 8.20x103 2.79x10-7 1.00x104 4.99x10-8 Momoni 1.00x103 3.09x10-9 2.00x103 1.59x10-9 Lower N, Upper N: lower and higher values of doses in Tables 4.13-4.15. Pill: probability of illness as computed from Weibull-Gamma model (Equation 2, Section 3.9.3) -: no lower value 105 Table 4.18: Probability of illness among high risk groups in Tema New Town Consumers Product Lower N Pill Upper N Pill C h il d re n Smoked tuna 2.02x102 2.25x10-7 3.00x104 0.0097 Smoked mackerel 4.04x104 0.0180 6.00x104 0.0397 Smoked herring 1.50x104 0.0023 3.00x104 0.0097 Sundried sardines 1.50x104 0.0023 3.00x104 0.0097 E ld er ly Smoked tuna 3.02x104 0.0099 4.00x104 0.0177 Smoked mackerel 6.04x104 0.0402 8.00x104 0.0680 Smoked herrings 3.03x104 0.0099 4.50x104 0.0224 Sundried sardines 3.03x104 0.0099 4.50x104 0.0224 Kako 5.00x102 1.56x10-6 1.00x103 6.89x10-6 Koobi 0 0 0 0 Momoni 0 0 0 0 P re g n a n t w o m en Smoked tuna - - 4.00x104 0.0177 Smoked mackerel 6.04x104 0.0402 8.00x104 0.0680 Smoked herrings 3.03x103 7.38x10-5 4.50x104 0.0224 Sundried sardines 4.53x104 0.0227 6.00x104 0.0397 Kako 5.00x102 1.56x10-6 1.00x103 6.89x10-6 Koobi 0 0 0 0 Momoni 0 0 0 0 R es p o n d en t Smoked tuna - - 4.00x104 9.69x10-7 Smoked mackerel 6.04x104 2.34x10-6 8.00x104 4.27x10-6 Smoked herrings 4.53x104 1.26x10-6 6.00x104 2.31x10-6 Sundried sardines 3.03x104 5.35x10-7 4.50x104 1.25x10-6 Kako 5.00x102 8.20x10-11 1.00x104 4.99x10-8 Koobi 0 0 0 0 Momoni 0 0 0 0 Lower N, Upper N: lower and higher values of doses in Tables 4.13-4.15. Pill: probability of illness as computed from Weibull-Gamma model (Equation 2, Section 3.9.3) -: no lower value 106 Table 4.19: Probability of illness among high risk groups in Madina Consumers Product Lower N Pill Upper N Pill C h il d re n Smoked tuna 2.12x105 0.2750 3.15x105 0.3877 Smoked mackerel 2.93x105 0.3672 4.35x105 0.4754 Smoked herring 4.00x104 0.0177 8.00x104 0.0680 Sundried sardines 1.50x104 0.0023 3.00x104 0.0097 E ld er ly Smoked tuna 3.20x105 0.3921 4.20x105 0.4662 Smoked mackerel 4.40x105 0.4783 5.80x105 0.5462 Smoked herrings 8.08x104 0.0693 1.20x105 0.1337 Sundried sardines 3.0-3x104 0.0099 4.50x104 0.0224 Kako 3.50x103 0.0001 7.00x103 0.0004 Koobi 1.26x104 0.0016 1.80x104 0.0033 Momoni 1.50x103 1.64x10-5 3.00x103 7.23x10-5 P re g n a n t w o m en Smoked tuna - - 4.20x105 0.4662 Smoked mackerel 4.40x105 0.4783 5.80x105 0.5462 Smoked herrings 8.08x104 0.0693 1.20x105 0.1337 Sundried sardines 4.53x104 0.0227 6.00x104 0.0397 Kako 3.50x103 0.0001 7.00x103 0.0004 Koobi - - 3.00x104 0.0097 Momoni 1.50x103 1.64x10-5 3.00x103 7.23x10-5 R es p o n d en t Smoked tuna - - 4.20x105 0.0001 Smoked mackerel 4.38x105 0.0002 5.80x105 0.0003 Smoked herrings 1.21x104 7.50x10-8 1.60x104 1.36x10-7 Sundried sardines 3.03x104 5.35x10-7 4.50x104 1.25x10-6 Kako 3.50x103 5.28x10-9 7.00x103 2.33x10-8 Koobi 2.50x104 3.54x10-7 3.00x104 5.24x10-7 Momoni 1.50x103 8.61x10-10 3.00x103 3.79x10-9 Lower N, Upper N: lower and higher values of doses in Tables 4.13-4.15. Pill: probability of illness as computed from Weibull-Gamma model (Equation 2, Section 3.9.3). -: no lower value 107 Considered without regard to the communities, the probabilities of illness from ingesting the doses (N) of L. monocytogenes (Tables 4.14-4.15) among the different consumer groups ranged from 1 in 100 chances of illness (order of 10-1) to 1 in 1011 chances, as summarized in Table 4.20. Table 4.20: Summary of ranges of probability of illness among consumers (without regard to communities) Ranges of probability of illness Low Risk Group High Risk Group Product Respondents Elderly Children Pregnant women Smoked Tuna 10-4 -10-7 10-1 -10-3 10-1 -10-7 10-1 -10-2 Smoked mackerel 10-3 -10-6 10-1 -10-2 10-1 -10-2 10-1 -10-2 Smoked herrings 10-6 -10-8 10-1 -10-2 10-1 -10-3 10-1 -10-5 Sundried sardines 10-6 -10-7 10-2 -10-3 *10-3 *10-2 Kako 10-8 -10-11 10-4 -10-6 - 10-4 -10-6 Koobi 10-7 -10-8 10-3 -10-4 - 10-7 -10-8 Momoni 10-9 -10-10 10-5 -10-9 - 10-5 -10-9 * same order recorded for both lower dose and upper dose - not determined since data on consumption patterns for products were not collected for consumer category 4.11.1 Monte Carlo simulations The risk estimates (probabilities of illness) presented in Tables 4.17 – 4.19 are point estimates, the use of which to determine risks is considered unsatisfactory (Lindqvist and Westoo, 2000) and has been criticized for its tendency to give errors (Cassin et al., 1996). Therefore, the point estimates were converted into distributions using @Risk 5.5 108 Software. The distributions are shown in Appendix 6. A few typical distributions are show in Fig. 4.31 – 4.33 as examples. Fig. 4.31 shows that the minimum and maximum probabilities of illness are 0.0109 and 0.0802 (marked *), between which 90% of the probabilities of illness fell. The figure shows that about 2 in 100 chances of falling ill is the highest likelihood. Beyond this, the chances of infection diminish steadily. 0.0109* 0.0802* 5.0% 5.0% 0 5 10 15 20 25 0 0 .0 1 0 .0 2 0 .0 3 0 .0 4 0 .0 5 0 .0 6 0 .0 7 0 .0 8 0 .0 9 0 .1 0 .1 1 Fig. 4.31: Triangular distribution for probability of illness, Pill, among the elderly in Madina consuming sundried sardines contaminated with Listeria monocytogenes Pill P ro b ab il it y 90% 109 Among respondents in Jamestown consuming smoked mackerel, Fig. 4.32 shows that the likely tendency is an increase in the probability of infection from 2 in 1,000,000 to about 5 in 1,000,000. Beyond this range, lower probabilities of illness are expected. 1.51 8.29 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 1 0 1 1 Pill (x10 -6 ) Fig. 4.32: Triangular distribution for probability of illness among respondents in Tema New Town who consumed smoked mackerel contaminated with Listeria monocytogenes P ro b ab il it y 90% 110 The distribution in Fig. 4.33 shows that the most likely probability of illness among pregnant women in James town as a result of consumption of koobi contaminated with L. monocytogenes is 8 in 10,000. Lower probabilities are expected beyond this level. 0.208 0.900 5.0% 5.0% 0 5 10 15 20 25 0 0 .1 0 .2 0 .3 0 .4 0 .5 0 .6 0 .7 0 .8 0 .9 1 1 .1 Pill (x10 -3) Fig. 4.33: Triangular distribution for probability of illness among pregnant women in Jamestown consuming koobi contaminated with Listeria monocytogenes P ro b ab il it y 90% 111 CHAPTER FIVE 5.0 DISCUSSION 5.1 Traditional Fish Processing The methods of traditional fish processing have not changed over the years. Practices during processing observed in this study were similar to those reported by earlier researchers (Essuman, 1982; Yanka, 1988; Nketstia-Tabiri, 1994; Coffie, 2002). The means by which fresh fish were transported to the processing sites made them susceptible to contamination. The use of public transport for the fish could result in contaminations with pathogens that could survive processing and pose a food safety risk to consumers. 5.1.1 Sanitation The processing environments were generally unsanitary (Fig.4.6, 4.11 and 4.16). The processors did not conform to the acceptable conditions for processing premises (buildings, hygienic facilities and water quality programme) and general hygiene (sanitation programmes and handling practices) stipulated by the Ghana Standards Board Code of Practices (GS 235:1997). Fish handling practices were generally poor and unhygienic among the processors. Thawing of frozen mackerel and herrings close to the ground with spillage on the same (Fig. 4.12 – 4.13), cutting of fresh ray fish on the bare ground (with only wet cardboard separating fish and soil, Fig. 4.29) and use of same bowl of water to wash several fish 112 (Fig. 4.14, 4.17) compromised the hygienic conditions of fish handling during traditional processing. These made contaminations very probable, and could explain the high microbial counts recorded for some of the fresh fish (Table 4.2 – 4.5). Observations made at the processing sites also suggest that post-processing contaminations could start from the “packaging” of fish for the market. For example, in smoked fish, processors did not wash their hands before taking the finished product off the oven. While taking fish off the oven, other processors offered to help without considerations on the hygienic status of their hands or clothing (Fig. 4.10). Additionally, smoked fish on trays very placed close to the bare ground (Fig. 4.15) possibly enhanced susceptibility to post-processing microbial contamination. In processing sundried sardines, the practice of sweeping the dried fish into heaps to be collected in baskets (Fig. 4. 19) was unsatisfactory. In addition to promoting contamination with soil microflora, physical hazards such as stones could also be introduced into the fish. This is particularly important from a food safety perspective as the fish do not go through any additional cleaning processes before being sent to the market. Similarly, the practice of stepping on salted fish (Fig. 4.27) while gathering them was not satisfactory, as contaminations could result. 5.1.2 Detection of total coliforms and Escherichia coli in fish during traditional processing The results suggest that the microbial counts of fish decreased from the raw through to the processed fish (Tables 4.2 – 4.5). The absence of growth for all fresh mackerel and 113 herrings suggest adherence to hygienic codes of fish packaging in the frozen companies from which the fish were purchased. Oppey (2002), however, reported counts of 1.3x103 CFU/g for total plate count in frozen mackerel Accra. The microbiological quality of the water used for washing was unsatisfactory (Tables 4.2 – 4.5, Fig. 4.2 – 4.5). The total plate counts of wash water before and after use were in the order of 106 CFU/ml. E. coli and total coliform counts were also in the order of 103 CFU/ml. This violates the requirement that water considered ideal for food processing operations must not have any coliforms (ICMSF, 1996). Cofie (2003) recorded similar findings on the microbiological quality of water used in traditional smoking of mackerel in Accra and Tema. The absence of microbial growth in all smoked fish (tuna, mackerel and herrings) samples collected immediately after processing suggest that the time-temperature combination of smoking was sufficient to eliminate the microorganisms. Cofie (2003) also did not record any growths for total plate count and E. coli in smoked fish sampled immediately after smoking. These findings emphasize the smoking stage as a critical control point in the smoked fish process flow. In salted fish, the absence of growth for total plate count, total coliform count and E. coli and L. monocytogenes could be explained by the high concentrations of salt used (Fig. 4.30). Growths recorded after at least three (3) days sun-drying on the ground could be a result of contamination from the soil (Fig. 4.31). 114 In sundried sardines, the occurrence of microbial growth at every stage of processing (Table 4.4) could be due to the lack of any step in the processing chain that had the potential to eliminate microorganisms. Beyond reduction in water activity of the fish, no heating or osmotic stress strategies are employed to eliminate microorganisms. Additionally, the direct exposure of the fish to soil for days (with strong winds blowing dust over the fish), and mode of collecting fish after drying (Fig. 4.19) contributed to the generally high microbial counts obtained for this product. 5.2 General microbiology of fish on informal markets The microbiological status of the traditionally processed fish on all the five informal markets surveyed was generally unsatisfactory. As shown in the Table 4.6, the counts were in the order of 106 CFU/g for total plate counts, 105 CFU/g for total coliforms, 103 CFU/g for E. coli, and 103 for L. monocytogenes. Contamination of foods with coliforms in general and E. coli in particular mostly results from unhygienic handling of foods (Jay et al., 2005; Hobbs and Roberts, 1987), suggesting that hygienic handling during fish sales on informal markets is unsatisfactory. Additionally, although most coliforms are not pathogenic (Prescott et al, 1995; Montville and Matthews, 2005) they are indicator organisms (Brock and Madigan, 1991). Therefore, an additional safety concern is that their presence in the fish products suggests the likelihood that other pathogenic microorganisms (possibly including heat-resistant strains) were also present. Other studies on the microbiological safety of fish on informal markets in Ghana reported similar findings (Adu-Gyamfi, 2006; Cofie, 2003; Oppey, 2002). 115 Counts obtained after processing were generally lower than those obtained for market samples. Example, whereas no counts were obtained for smoked fish sampled immediately after processing (Tables 4.2 and 4.5), counts for smoked fish on informal markets were high 103 - 106 CFU/g (Table 4.6). This suggests that post-processing handling practices either caused or contributed significantly to the contamination of the fish products. L. monocytogenes has the potential to contaminate fish at any point between harvesting and consumption (FAO, 1999). 5.3 Risk assessment for Listeria monocytogenes from the consumption of traditionally processed fish 5.3.1 Hazard identification Listeria monocytogenes, the hazard considered in this study, causes listeriosis, a relatively rare but highly fatal disease. The pathogen is opportunistic and particularly affects segments of the population that are immunocompromised, pregnant women, unborn or newly borne infants and the elderly. Detailed information on the organism is presented in Section 2.7 (Chapter Two). 5.3.1.1 Detection of Listeria monocytogenes during processing Listeria monocytogenes was not detected in fresh mackerel and herrings before washing. After washing, however, it was detected, suggesting that the wash water could be a source of contamination. This was confirmed when the pathogen was detected in water samples used for washing (Tables 4.2-4.5). The next unit operation after washing of the 116 fish was smoking. Immediately after smoking, the pathogen was no longer detectable in the smoked fish, suggesting that the heat process was adequate to eliminate L monocytogenes in the product. In fresh ray fish and sardines, the occurrence of L. monocytogenes could be because the pre-processing operations were done close to the ground under unsanitary conditions. Additionally, contamination may have resulted from contact of the fish with soil during cutting (Fig. 4.29). In stale fish used for momoni, the presence of the pathogen was expected as the fish were deteriorated and heavily contaminated (Figure 4.22). 5.3.1.2 Occurrence of Listeria monocytogenes in traditionally processed fish on the markets Listeria monocytogenes was detected in 53 – 80% of the fish samples obtained from the five markets in the study (Table 4.13). This was a rather high prevalence, as most studies report low values. Kwiatek (2004) found the organism in only 4% of 451 smoked fish samples, and 8% of 633 raw fish samples in Poland. In Mexico, Oroczo (2000) also recorded a prevalence of 28% in 14 samples of smoked salmon. Mahmood et al., (2003) detected the pathogen in 24% of 320 samples of some poultry products. The FAO (2009) also asserts that prevalence of L. monocytogenes in foods is generally low. Salihu et al. (2008) reported an incidence of 25% in 115 samples of smoked fish in Sokoto, Nigeria. The closest prevalence to the finding in the present study was the value reported by Ikeh et al., (2010), who detected the organism in 40% of 15 fish samples from some informal markets in Nigeria. 117 The absence of the pathogen in fish sampled immediately after smoking and detection in samples on the markets suggests that post-processing contamination was responsible for the occurrence. This is further supported by the observation that fish products that had antimicrobial properties such as koobi, kako and momoni because of their high salt concentrations had the lowest prevalences of 40%, 53% and 53%respectively. The combination of high salt content and dry nature (i.e. low water activity) probably did not support the survival of L. monocytogenes. It is plausible to suggest that water activity may have had a role in the prevalence rate of L. monocytogenes, even though it was not determined for the products. The lower prevalence in the sundried fish (60%) than the smoked fish (67 – 80%) could be a result of the relatively lower water activity in the former. Moreover, among the smoked fish samples, the lower prevalence in herrings (67%) than tuna and mackerel (80% and 97% respectively) could be attributed to the relatively dry nature of the smoked herrings. Detection in the dried fish could be attributed to the unsanitary practice of drying the fish on the bare ground and using brooms to sweep the products into heaps (Fig. 4.19). Contamination from soil was therefore very likely. 5.3.2 Exposure assessment The primary objective of the exposure assessment in this study was to estimate the level of exposure of consumers to Listeria monocytogenes based on frequency of consumption, and the pathogen load ( i.e. extent of contamination) of fish consumed. Detection of L. 118 monocytogenes in all the fish products suggests that consumers are exposed to ingestion of the pathogen if it is not eliminated before the products are consumed. Therefore, from a food safety perspective, the most important point of the processing and distribution chains of the traditionally processed fish examined in this study was the point of consumption (Fig. 4.50 and 4.51). Some consumers indicated consuming smoked fish and sundried fish as purchased from the market, without (much) further heat processing. Such consumption practices enhance the exposure to the pathogen. Of all the fish products, exposure through salted fish would generally be minimal as the count for samples in which the pathogen was detected was low (102CFU/g). Additionally, the salted fish products are generally not consumed frequently, and when used, are usually in small quantities (portion sizes of koobi could however be more than 50g, Appendix 3), and often heat treated. Food service centres, both formal (registered, audited and regulated food businesses) and informal could be likely places for consumer exposure to the pathogen. However, samples obtained from food vendors showed non-detectable counts of the pathogens (Section 4.5). The absence of the pathogen in samples from food vendors even though reassuring, does not necessarily imply that cooking completely eliminates the pathogen from foods so that no risk could occur at the point of consumption. Non-detection of the pathogen could be because food vendors (as generally practiced) kept the soups on fire 119 throughout sales to ensure that they were hot, as cold or lukewarm food would normally not be patronized. The focus group discussions revealed that street foods stay on fire longer than in domestic cooking, where for instance light soup was cooked for approximately 45 minutes and groundnut soup for 1 hour 15 minutes. However, sometimes, street vended foods are not kept hot throughout sales. In such instances, consumers could be at risk of ingestion if the pathogen is present in the foods. Based on the estimated cooking times of popular products such as light soup and groundnuts soup, the assumptions at Section 4.9.4 were formulated and the exposure of consumers to the pathogen determined. 5.3.2.1 Event trees The event tree analyses (Fig. 4.28 - 4.30) were used to evaluate the consequences arising from the event of detection of Listeria monocytogenes leading to consumer exposure. The consequences of detection of the pathogen were followed through a series of possible paths (that may or may not eliminate the hazard) during handling. The conditions under which consumers were at risk of ingesting L. monocytogenes are shown in the event trees in Figs. 4.52 – 4.55. Under those conditions, consumers were exposed to ingesting up to 105 cells through smoked tuna, mackerel and herrings, up to 104 cells through sundried sardines, and up to 105 cells through the salted fish products (Tables 4.14 – 4.16). 120 5.3.3 Hazard Characterization (dose-response assessment) The dose response relationship was modelled using the triangular probability distribution where the ‘normal’ could be predicted knowing the worst and best case scenarios. 5.3.3.1 Host susceptibility Granted that the number of cells in Tables 4.14 – 4.16 (103 to 105 cells) were consumed at an instance by the consumers, the likelihoods of suffering listeriosis (infection), Pill, would be as presented in Tables 4.17 – 4.19. From the Pill values in Tables 4.17 – 4.19, it is seen that the risk of infection was generally lower among the low risk groups (respondents) (Pill of 10 -4 to 10-11, i.e. 1 in 10,000 to 1 in 100,000,000,000 chances of illness) than the high risk groups (10-1 – 10-6 i.e. 1 in 10 to 1 in 1,000,000 chances of illness). Although higher doses (>109 CFU/g, Schlech, 1999) are required to cause illness in low risk groups than in high risk groups (104 CFU/g, Buchanan et al., 1997), the results of the present study suggests that host susceptibility is a more important factor in determining probability of illness than the dose of the pathogen ingested. At equal doses, high risk groups recorded higher risk of illness (infection) than low risk individuals. For example, in Tema New Town, although the dose in smoked tuna was the same for pregnant women and respondents (104 cells in each case) the respective Pill were 1.77x10 - 2 and 9.69x10-11, about nine orders less in the low risk group. 5.3.3.2 Matrix effects The three factors that affect the dose-response relationship include the environment (i.e. the food matrix), the pathogen (virulence characteristics), and the host (susceptibility or 121 immune status factors). The assumption in this work is that L. monocytogenes detected in the study belong to the most virulent strains. Different sub groups of the populations have been identified as least susceptible (low risk) and most susceptible (high risk). Among the fish products, the highest Pill were recorded for the smoked fish, and the lowest for salted fish (Table 4.20). Although differences in the counts of the pathogen in the respective products were important, the key factor affecting the size of Pill was the portion sizes of the products consumed at an instant. For example, although the counts of L. monocytogenes in sundried sardines and koobi were in the same order (102) for the elderly in Jamestown (Table 4.14), the Pill value in sundried sardines was one order greater (10-1) than that for koobi (10-2) due to larger portion sizes consumed at an instance in the former. 5.3.3.3 Triangular Distributions for Pill The probability distributions (Fig. 4.55 – 4.57 and Appendix 6) show that the chances of recording a Pill value beyond the most likely Pill were generally low. Therefore, the chances of a larger number of consumers than recorded in Table 4.20 falling ill following ingestion of the doses of L. monocytogenes in Tables 4.14 – 4.16 is low. In describing the distributions of Pill using the @Risk 5.5 software, the minimum values were set at one order less than the Pill values for Upper N (Tables 4.17-4.19) whiles the maximum values were set at one order greater. The actual Pill values or approximations of the same were used as the most likely values. 122 5.3.4 Risk characterization Given that the average counts of Listeria monocytogenes in the traditionally processed fish were generally low (102 – 103 CFU/g), on the whole, consumers are at a low risk of ingestion. Consumers who cook the products and consume them while hot are likely to further lower their risk of ingesting the pathogen. As a result of the low counts, the number of cells that would survive cooking is expected to be less. Additionally, those that survive are expected to be injured/stressed as not to pose threats to the health of consumers. Those who consume the products as purchased from informal markets increase both their risk of ingesting the pathogen and their risk of infection. 123 CHAPTER SIX 6.0 CONCLUSION AND RECOMMENDATIONS 6.1 CONCLUSION The sanitary conditions of traditional fish processing in Accra and Tema were unsatisfactory. This could result in the occurrence of coliforms in general and Escherichia coli in particular (and possibly other pathogens) along some points during processing. Although the microbial counts of the products generally decreased after processing, improper post-processing handling resulted in contamination of the processed fish. Listeria monocytogenes occurs in traditionally processed fish on informal markets, suggesting that the products could be vehicles for the transmission of the pathogen to consumers. Depending on the kind of traditionally processed fish consumed, consumers are exposed to ingesting 102 to 105 cells of L. monocytogenes, which could result in a 1 in 10 to 1 in 100,000,000 chances of illness. Consumers with high susceptibility to L. monocytogenes infection (elderly, children and pregnant women) were at a greater risk of illness than low risk individuals (non-pregnant adults aged 18 – 39 years). Although the estimated risks of ingestion and infection were generally low, individuals who either consume traditionally processed fish purchased from informal markets as is or do not heat-treat the products sufficiently increase their risks of ingestion and infection, and vice versa. 124 6.2 RECOMMENDATIONS Further studies should be conducted on  the occurrence of Listeria monocytogenes in other (traditionally processed) foods with high consumption rates in Ghana to determine the exposure of consumers to those products, and estimate the associated risks of listeriosis.  screening of placental smears for and molecular typing of L. monocytogenes to confirm occurrence of listeriosis. This will help to establish if the strains isolated from foods could be conclusively implicated in the occurrence of listeriosis or its symptoms.  consumption patterns of various foods in Ghana to aid risk assessments. Information on portion sizes and frequency of consumption of foods are essential for determining the exposure of consumers to food-borne hazards. Without this information, a comprehensive risk assessment cannot be conducted. 125 REFERENCES Adam, M.R. and Moss, M.O. (2008). Food Microbiology, 3rd Ed. RSC Publishing, Cambridge. pp 229, 224 – 231. Adu-Gyamfi, A. (2006). Studies on microbiological quality of smoked fish in some markets in Accra, Ghana. Ghana J. Sci. 46, 67 – 75. Appiah, J. (2010). Risk assessment of consuming milk and milk products contaminated with Listeria monocytogenes from informal markets in Ashiaman, Accra. MPhil Thesis. Department of Nutrition and Food Science, University of Ghana. Unpublished. Aquay, J.K. (1992). Implications of structural adjustment for Ghana’s marine fisheries policy. Fisheries Research, 14(1): 59-70. Argenti, O. 2000. Food for the cities: food supply and distribution policies to reduce urban food insecurity. “Food into cities” Collection, DT/43-00E. FAO, Rome. Atlas, R.M. (1997). Principles of Microbiology, 2nd Edition. Wm. C. Brown Publishers, Boston. Atta-Mills, J.E., Alder, J. and Sumaila, U.R. (2004). The decline of a regional fishing nation: The case of Ghana and West Africa. Natural Resources Forum 28 (2004) 13-21. Bank of Ghana. (2008). The Fishing Sub-sector and Ghana’s Economy. Research Department, Bank of Ghana. Bannerman, P.O. and Cowx, I.G. (2002). Stock Assessment of the big-eye grunt (Brachydeuterus auritus) fishery in Ghanaian Coastal Waters. Fisheries Research 59 (2002), 197-207. Bartlett, R.M., and Hawke, A.E.(1995). Heat resistance of Listeria monocytogenes Scott A and HAL 957El in various liquid egg products. J. Food Protect. 58: 1211-1214. Basti, A.A., Misaghi, A., Salehi, T.Z., and Kamkar, A.(2006). Bacterial pathogens in fresh, smoked and salted Iranian fish. Food Control 17 183 – 188. Bemrah, N., Sanaa, M., Cassin, M.H., Griffiths, M.W. and Cerf, O. (1998). Quantitative risk assessment of human listeriosis from consumption of soft cheese made from raw milk. Prev. Vet. Med. 37, 129-145. 126 Beumer, R. 1997. Listeria monocytogenes detection and behaviour in food and in the environment. Thesis work. University of Wageningen. ISBN 90-5485-6327 Bille, J. (1990). Epidemiology of listeriosis in Europe, with special reference to the Swiss outbreak. pp. 71-74, in: Miller, Smith & Somkuti. Black, J.G. (1999). Microbiology: Principles and Explorations. 4th Ed. Prentice Hall, New Jersey. Pp 386, 600. Board, R.G., Jones, D., and Skinner, F.A. (1992). Identification Methods in Applied and Environmental Microbiology. Blackwell Scientific Publications, Oxford. pp193, 197. Bradshaw, J.G., Peeler, J.T., and Corwin, JJ. (1985). Thermal resistance of Listeria monocytogenes in milk. J. Food Protect. 48:743-755. Brock, T.D. and Madigan, M.T. (1991). Biology of Microorganisms. 6th Ed. Prentice Hall, New Jersey. p 553 Broome, C.V., Gellin, B. & Schwartz, B. (1990). Epidemiology of listeriosis in the United States. CRC Press, New York. pp. 61–65. Buchanan, R.L., Damert, W.G., Whiting, R.C., and van Schothorst, M. (1997). Use of epidemiologic and food survey data to estimate a purposefully conservative dose- response relationship for Listeria monocytogenes levels and incidence of listeriosis. J. Food Protect. 8, 918 – 922. Bunning, V.K., Crawford, R.G., and Tierney, J.T. (1990). Thermotolerance of Listeria monocytogenes and Salmonella typhimurium after sublethal heat shock. Appl. Environ. Microbiol. 56:3216-3219. Cassin, M.H., Paoli, G.M., McColl, R.S., and Lammerding, A.M.(1996). A comment on hazard assessment of Listeria monocytogenes in the processing of bovine milk. J. Food Protect. 59, 341-343. Chant, S. (1999). Informal sector activity in the third world city. In Pacione, M. (ed). Applied Geography: Principles and Practice, London, Routledge Company. Codex Alimentarius Commission, CAC. (2003). Principles and guidelines for the conduct of microbiological risk assessment (CAC/GL-30, 1999). pp 53 – 62. FAO. 127 Cofie, A. (2003). Processing Environment and the microbial quality of smoked fish in the Accra and Tema Municipalitiees. MPhil Thesis. Department of Environmental Science, University of Ghana. Colburn, K. G., Kaysner, C. A., Abeyta, C. and Wekeel, M. (1990). Listeria species in a California Coast Estuarine Environment. Applied and Environmental Microbiology. 56(7):2007-2011. Collignan, A. and Raoult-Wack A. L. (1994). Dewatering and salting of cod by immersion in concentrated sugar/salt solutions. Food Sci. Technol. 27:259 (1994). Dobge, E. (2010). Risk of Listeria monocytogenes ingestion in consuming coleslaw from food vendors in the Accra metropolis. M.Phil Thesis. Department of Nutrition and Food Science, University of Ghana. Unpublished. Embarek, B.P.K. (1994). Presence, detection and growth of Listeria monocytogenes in sea foods: a review. Int. J. Food. Microbiol. 23: 17-34. Essuman, K.M. (1982). The Effects of hot smoke-curing on the protein and organoleptic quality of West African herrings (Sardinella eba) processed in improved traditional ovens. MSc. Thesis, Department of Nutrition and Food Science, University of Ghana. FAO (2007). Promises and challenges of the informal food sector in developing countries. A joint publication by the Rural Infrastructure and Agro-Industries Division, and the Nutrition and Consumer Protection Division of the Agriculture and Consumer Protection Department of FAO. FAO/WHO.(2004). Risk assessment for Listeria monocytogenes in ready-to-eat foods, Interpretative Summary. Microbiological Risk Assessment Series 4 and 5. Rome. Farber, J.M. and Peterkin, P.I. (1991). Listeria monocytogenes, a food-borne pathogen. Microbiological Reviews 55, 476-511. Farber, J.M., and B.E. Brown. (1990). Effect of prior heat shock on heat resistance of Listeria monocytogenes in meat. Appl. Environ. Microbiol. 56: 1584-1587. Farber, J.M., Ross, W.H., Harwig, J. (1996). Health risk assessment for Listeria monocytogenes in Canada. Int. J Food Microbiol. 30, 145-156. 128 Farber, J.M., Wang, S.L., and Cai, Y.(1998). Changes in populations of Listeria monocytogenes inoculated on packaged fresh-cut vegetables. J. Food Protect. 61: 192- 195. Fedio, W.M. and Jackson, H. (1989). Effect of tempering on the heat resistance of Listeria monocytogenes. Lett. Appl. Microbiol. 9:157-160. Fehlduson, F. (1999). Foodborne bacterial pathogens. Marcel Dekker, New York. Food and Agriculture Organization (FAO). (1999). Expert consultation on the trade impact of Listeria in fish products. FAO Fisheries Report No. 604. Amherst, MA. http://fao.org/WAICENT/FAOINFO/FISHERY/FISHERY.HTM. Food and Agriculture Organization (FAO). (2000). The state of world fisheries and aquaculture. FAO Fisheries Department, Rome. Gallagher, D. L., Ebel, E.D., and Kause, J.R. (2003). Food Safety and Inspection Service (FSIS) risk assessment for Listeria monocytogenes in Deli Meats. United States Department of Agriculture. Garbutt, J. (1997). Essentials of Food Microbiology. Arnold, London. pp 162 – 164. Ghana Standards Board Code of Practice for processing of dried and smoked fish (GS 235:1997). Gilbert, R.J. (1992). Provisional microbiological guidelines for some ready-to-eat foods sampled at point of sale: Notes for PHLS Food Examiners. Public Health Serv. Lab. Q. 9:98-99. Glass, K.A., and Doyle, M.P. (1990). Fate of Listeria monocytogenes in processed meat products during refrigerated storage. Appl. Environ. Microbiol. 55:1565–1569. Guizani, N. and Mothershaw, A. (2007). Fermentation as a method for food preservation. In Rahman M.S. (ed). Handbook of food preservation.2nd ed. CRC Press, Boca Raton. p215 Heinitz, M.L., Ruble, R.D., Wagner, D.E., and Tatini, S.R. (2000). Incidence of Salmonella in fish and seafood. J. Food Protect. 63:579–592. Hobbs, B.C. and Roberts, D. (1987). Food Poisoning and Food Hygiene. 5th Ed. Edward 129 Arnold, London. p 217 Horner, W.F.A. (1997). Preservation of fish by curing (drying, salting and smoking). In Hall, G.M. (ed). Fish Processing Technology. Academic and Professional, London. Huss, H. H., Embarek, P. K. B. and Jeppesen, V. F. 1995. Control of biological hazards in cold smoked salmon production. Food Control 6(6):335-340. Ihekoronye, A.I. and Ngoddy, P.O.(1985). Integrated food science and technology for the tropics. Macmillan Publishers, New York. p339. Ikeh, M.A.C., Obi, S.K.C., Ezeasor, D.N., Ezeonu, I.M., and Moneke, A.N. (2010). Incidence and pathogenecity profile of Listeria sp. isolated from food and environmental samples in Nsukka, Nigeria. African Journal of Biotechnology Vol 9(30), pp. 4776-4782. International Commission on Microbiological Specification of Foods (ICMSF). (1996). Microorganisms in Foods. 5th ed., Aspen Publishers, Inc., Gaithersburg. International Labour Organization (ILO). (2002). Women and men in the informal sector. Geneva, Switzerland. Jay, J.M. (2003). Modern Food Microbiology, 6th Ed. Aspen Publishers, Inc., Gaithursburg. pp 485 – 505. Jay, J.M., Loessner, M.J. and Golden, D.A. (2005). Modern food microbiology. 7th ed. Springer Science+Business Media, Inc. p591. Jemmi, T. and Keusch, A. (1994). Occurrence of Listeria monocytogenes in freshwater fish farms and fish-smoking plants. Food Microbiology, Volume 11, Issue 4, pp 309-316. Jeyasekaram, G. and Karunasagar, I. (1996). Incidence of Listeria species in tropical fish. Int. J. Food Microbiol. 31: 333-340. Kalliopi, R., Valentina, A., Rosalinda U., Paola, D., and Luca C. (2008). Detection, quantification and vitality of Listeria monocytogenes in food as determined by quantitative PCR. International Journal of Food Microbiology 121 (2008) 88 – 105. Kamat, A.A and Nair, P.M. (1994). Incidence of Listeria species in Indian sea foods and meat. J. Food Saf. 14: 117-1130. 130 Kamat, A.A. and Nair, P.M. (1994). Incidence of Listeria species in Indian seafoods and meat. J. Food Saf. 14: 117-1130 Karunasagar, I., Segar, K. and Goebel, W.(1992). Incidence of Listeria species in tropical sea foods. In: Listeria 1992. Abstract no. 155. Eleventh International Symposium on Problems of Listeriosis (ISOPOL XI). May 11th-14th. Copenhagen, Denmark. Kathariou, S. (2002). Listeria monocytogenes virulence and pathogenicity, a food safety perspective. J. Food Protect., 65: 1811-1829. http://www.ncbi.nlm.nih.gov/pubmed/12430709 Kwiatek, K. (2000). Occurrence of Listeria monocytogenes in selected foods of animal origin. Bull. Vet. Inst. Pulawy 48, 269-272. Kwiatek, K. Occurrence of Listeria monocytogenes in selected foods of animal origin. Bull Vet Inst Pulawy 48, 269 – 272 2004. Lammerding, A.M. (1998). An overview of microbial food safety risk assessment. J. Food Prot. 60:1420. Lammerding, A.M., and J.M. Farber. (1994). The status of Listeria monocytogenes in the Canadian food industry. Dairy Food Environ. Sanit. 14:146-150. Lammerding, A.M., Fazil, A.M. and Paoli, G.M. (2001). Microbial Food Safety Risk Assessment. In Downs, F.P. and Ito, K. (eds). Compendium of methods for the microbiological examination of foods. American Public Health Association, Washington DC. pp 267 – 277. Lindqvist, R. and Westoo, A. (2000). Quantitative risk assessment for Listeria monocytogenes in smoked or gravad salmon and rainbow trout in Sweden. International Journal of Food Microbiology 58 181 – 196. Linton, R.H., Pierson, M.D., and Bishop, R.. (1990). Increase in heat resistance of Listeria monocytogenes Scott A by sublethal heat shock. J. Food Protect. 53:924-927. Lober, B. (2007). Listeriosis. In Goldfine, H. and Shen, H. Listeria monocytogenes: pathogenesis and host response. Springer Science+Business Media, LLC. p 13 Mackey, B.M., and N. Bratchell. (1989). The heat resistance of Listeria monocytogenes. 131 Lett. Appl. Microbiol. 9:89-94. Mackey, B.M., Pritchet, C., and Norris. P. (1990). Heat resistance of Listeria: Strain differences and effects of meat type and curing salts. Lett. Appl. Microbiol. 10:251-255. Mahmood, M.S., Ahmed, A.N. and Hussain, I. (2003). Prevalence of Listeria monocytogenes in poultry meat, poultry meat products and other related inanimates at Faisalabad. Pakistan Journal of Nutrition 2 (6): 346-349. Manoj, Y.B., Rosalind, G.M., and Karanasagar, I.(1991). Listeria species in fish and fish handling areas. Mangalore, India. Asian Fish. Sci. 4: 119-122. Matte, E., Sprenger, A.K. and Reddy, P. (1994). Markets for frozen small pelagic fish. FAO Fisheries Technical Paper No. 221. Rome. Mauro, C., Pierluigi, D. C., Domenico, M., Zanardi, E., Festino, A.R., Ghidini, S., Vergara, A., and Ianieri, A. (2008). Sources and tracking of Listeria monocytogenes in a cold-smoked processing plant. Ann. Fac. Medic. Vet. di Parma Vol. XXVIII 97-104 Mensah, P., Yeboah-Manu, D., Owusu-Darko, K., and Ablordey, A. (2002). Street foods in Accra, Ghana: how safe are they? Bulletin of the World Health Organization 2002;80:546- 554. Ministry of Food and Agriculture, Ghana (MOFA).(1999). A manual on training agricultural extension agents in fish processing. Fish Processing and Marketing Section, Fisheries Directorate, MOFA, Accra, Ghana. Montville, T.J. and Matthews, K.R. (2005). Food Microbiology: An Introduction. ASM Press, Washington D.C. pp 159 – 171. Munhande, C. and Makaye, P. (2008). The contribution of informal food trading to urban food security in Zimbabwe: the case of Gweru 200-2007. Journal of Sustainable Development in Africa, Vol. 10 No. 2. Neequaye-Tetteh, G. A., Dassah, A. L. and Quarshiesam, S. J. (2002). Effect of fuel wood type on the quality of smoked fish–Chrysichthys auratus. Ghana Jnl agric. Sci. 35. 95-101. Nickelson II, R. and Finne, G. (1992). Fish, Crustaceans, and Precooked Seafood. In Vanderzant, C. and Spilttstoesser D.F. (eds). Compendium of Methods for the 132 Microbiological Examination of Foods. 3rd Ed. Washington D.C.: American Public Health Association. Nketsia-Tabiri, J. and Sefa-Dedeh, S. (2000). Quality attributes and utilization of cured fish products in Ghana. Journal of Applied Science and Technology 5:1 148 – 155. Nketstia-Tabiri, J. (1994). Traditional Fish Processing: Technology, Quality Development and Evaluation. PhD Thesis, Department of Nutrition and Food Science, University of Ghana. Nyamekye, S.K. (2000). Microbiology of smoked fish sold on the Ghanaian market. Higher Diploma in Microbiological Techniques Dissertation. Institute of Science and Technology, London. Okutani, A., Okada, Y., Yamamoto, S., and Igimi, S. Nationwide survey of human Listeria monocytogenes infection in Japan. Epidemiol. Infect. (2004), 132, 769 – 772. Oppey, D. (2002). Determination of microbial hazards and critical control points of handling and processing of smoked mackerel (Scomber Japonicus). MPhil Thesis. Department of Nutrition and Food Science, University of Ghana. Oroczo, L.N. (2000). The occurrence of Listeria monocytogenes and microbiological quality of cold smoked and gravid fish on the Icelandic retail market. Undergraduate thesis, United Nations University. Orraca-Tetteh, R. and Nyanteng, V.K. (1978). Consumer attitudes to cured fis in Ghana. Report submitted to FAO. Part 1 and 2. Overa, R., (2002), When sisters become competitors: Coastal women’s access and utilisation of trawlers’ by-catch in Ghana. In: Neis, B., Binkley, M., Gerrard, S. and Maneschy, C. (Eds.), Gender, Globalization and Fisheries. Fernwood Publishing, Halifax, Canada. Plahar, W.A., Pace, R.D. and Lu, J. Y. (1991). Effect of storage conditions on the quality of smoke-dried herrings (Sardinella eba). J. Sci. Food. Agric 57, 597 – 610. Prentice, G.A. and Neaves, P. (1992). The Identification of Listeria Species. In Board, R.G., Jones, D. and Skinner, F.A. (eds). Identification Methods in Applied and Environmental Microbiology. Blackwell Scientific Publications, New York. pp 283-296. 133 Prescott, L.N., Harley, J.P. and Klein, D.A. (1995). Microbiology. 3rd ed. Wm. C. Brown Publishers, Dubuque. p850. Rahman, M.S. and Perera, C.O. (2007). Drying and food preservation. In Rahman M.S. (ed). Handbook of food preservation.2nd ed. CRC Press, Boca Raton. p414 Rørvik, L. M., Caugant, D. A. and Yndestad, M. (1995). Contamination pattern of Listeria monocytogenes and other Listeria spp. in a salmon slaughterhouse and smoked salmon processing plant. International Journal of Food Microbiology 25:19-27. Ryser, E.T. and Marth, E.H. (1988). Survival of Listeria monocytogenes in cold-pack cheese food during refrigerated storage. J. Food Protect. 51:615-621. Salihu, M.D., Junaidu, A.U., Manga, S.B., Gulumbe, M.L., Magaji, A.A., Ahmed, A., Adamu, A.Y., Shitty, A. and Balarabe, I. (2008). Occurrence of Listeria monocytogenes in smoked fish in Sokoto, Nigeria. African Journal of Biotechnology Vol 7 (17). pp 3082-3084. Sarpong, D.B. (2008). Draft Report on Harvest and Postharvest Baseline Study in Ghana. Department of Agriculture, Economics and Agribusiness, College of Agriculture and Consumer Sciences, University of Ghana. Pp 43 – 51. Schlech W.F. (1999). Pathogenesis and immunology of Listeria monocytogenes. Pathol Biol (Paris) 44: 775–782 Schmidhuber, J. and Tubiello, F.N. (2007). Global food security under climate change. National Academy of Sciences, USA. 104, 50:1970319708 http://www.pnas.org_cgi_doi_10.1073_pnas.0701976104 Sefa-Dedeh, S. (1989). Harnessing traditional food technology for development. In Sefa- Dedeh, S. and Orraca-Tetteh, R. (eds). Proceedings of the Kellogg International Fellowship Programme in Food Systems workshop, Accra. Sefa-Dedeh, S. (1993). Traditional Food Technology. In Macrae, R., Robinson, R., and Sadler, M.(eds). Encyclopaedia of Food Science, Food Technology and Nutrition. Academic Press, New York. pp 4600-4606. Shelef, L.A. (1989). Survival of Listeria monocytogenes in ground beef or liver during storage at 4oC and 25oC. J. Food Protect. 52:379–383. 134 Steiner-Asiedu, M., Asiedu, D. and Njaa, L.R. (1991). Effect of local processing methods (cooking, frying, smoking) on three fish species from Ghana: Part 2 – Amino acids and protein quality. Food Chemistry. 41 227 – 236. Sutherland, P.S. and Porritt, R.J. (1997) Listeria monocytogenes. In: Food-borne microorganisms of public health importance, 5th Edition. AIFST (NSW Branch), Sydney. pp 333-378. Swaminathan, B. (2001). In Doyle, M. P., Beuchat, L.R., and Montville, T.J. (eds). Food Microbiology: Fundamentals and Frontiers, 2nd Ed. ASM Press, Washington D.C. pp 383-409. The United Kingdom Health Protection Agency (HPA). (2009). Detection and enumeration of Listeria monocytogenes and other Listeria species. Issue No. 3.1. Standards Unit, Department for Evaluations, Standards and Training. Reference No. F19i3.1 The United States Centre for Disease Control and Prevention (CDC). (2009). Emerging pathogens and their estimated annual outbreaks of foodborne illnesses in the USA. The United States Centre for Disease Control and Prevention (CDC). (2011). Outbreak of listeriosis Linked to Cantaloupe. http://www.cdc.gov/Features/Archive/2011.html Tompkin, R.B. (2002). Control of Listeria monocytogenes in the food-processing environment. J. Food Protect. 65:709–725. UNDP. (2002). Fish smoking with the Chorkor smoker: Ghana. http:www.undp.org/tcdc/bestprac/agri/cases/ghana15/html United States Centre for Disease Control and Prevention (CDC). (2009). Annual outbreaks of food-borne illnesses. http://www.cdc.gov/Features/dsFoodborneEstimates Yanka, V.V. (1988). Studies on momoni, a Ghanaian fermented fish product. BSc. Thesis. Department of Nutrition and Food Science, University of Ghana. 135 APPENDICES Appendix 1: Consumer Questionnaire DEPARTMENT OF NUTRITION AND FOOD SCIENCE UNIVERSITY OF GHANA, LEGON RISK ASSESSMENT FOR LISTERIA MONOCYTOGENES IN TRADITIONALLY PROCESSED FISH IN GHANA Dear respondent, this questionnaire seeks to solicit some information on the consumption of traditionally processed fish in Ghana, as part of an MPhil Food Science Thesis on the topic above. The information you provide in this document will be treated as confidential and used for academic purposes only. Thank you. Date: _______________________ Area: _______________________ Kindly tick (√) the responses that apply to you. Where appropriate, write out your own responses in the spaces provided. A: BACKGROUND INFORMATION RESPONSE [For interviewer use only] 1. Sex: 1=Male 2=Female 2. Age: 1=Less than 20 years 2=20 – 29 years 3=30 – 39 years 4=40 – 49 years 5=50 years and above 3. Highest level of education 1=None 4=Secondary 2=Primary 5=Tertiary 3=Middle School/JHS 6=Other, specify........................ B: FISH CONSUMPTION PATTERNS 4. Do you consume traditionally processed fish? 1 = Yes 2 = No 136 5. If yes to 4, which of the following traditionally processed fish do you consume? 1=Smoked fish 3=Smoke-dried fish 2=Salted fish 4=Fried fish 5=Other, specify........................................................................... 6. Which species of traditionally processed fish do you consume? 1 = Mackerel 2 = Tilapia 3 = Herrings 4 = Other, specify......................... How often do you consume the fish products? 7. Smoked tuna 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify............................................................................ 8. Smoked mackerel 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify............................................................................ 9. Smoked herrings 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify............................................................................ 10. Koobi 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify............................................................................ 11. Kako 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify............................................................................ 137 12. Momoni 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify............................................................................ 13. Sundried sardines 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify............................................................................ How much do you consume at an instance? 14. Smoked tuna 1=50-100g 2=101-150g 3=151 -200g 4=More than 200g 15. Smoked mackerel 1=50-100g 2=101-150g 3=151 -200g 4=More than 200g 16. Smoked herrings 1=50-100g 2=101-150g 3=151 -200g 4=More than 200g 17. Koobi 1= 10-20g 2= 21-30g 3= 31-40g 4= 41-50g 5= >50g 18. Kako 1= 5-10g 4= 31-40g 2= 11-20g 5= >40g 3= 21-30g 138 19. Momoni 1= 5-10g 4 = 31-30g 2= 11-20g 5 = > 50g 3= 21-30g 20. Sundried sardines 1=50-100g 2=101-150g 3=151 -200g 4=More than 200g C: CONSUMPTION BY HIGH RISK GROUPS I. CHILDREN 21. Do you have children (6mo to 6years) in your home? 1 = Yes 2 = No If No to Q21, go to Q29 How often do they consume traditionally processed fish? 22. Smoked tuna 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify............................................................................ 23. Smoked mackerel 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify............................................................................ 24. Smoked herrings 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify........................................................................... 139 How much do they consume at an instance? 25. Smoked tuna 1=50-100g 2=101-150g 3=151 -200g 4=More than 200g 26. Smoked mackerel 1=50-100g 2=101-150g 3=151 -200g 4=More than 200g 27. Smoked herrings 1=50-100g 2=101-150g 3=151 -200g 4=More than 200g 28. Sundried sardines 1=50-100g 2=101-150g 3=151 -200g 4=More than 200g II. THE ELDERLY 29. Do you have elderly people (≥60y) in your home> 1 = Yes 2 = No If No to Q29, go to Q 37 How often do they consume traditionally processed fish? 30. Smoked tuna 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify............................................................................ 140 31. Smoked mackerel 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify............................................................................ 32. Smoked herrings 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify........................................................................... How much do they consume at an instance? 33. Smoked tuna 1=50-100g 2=101-150g 3=151 -200g 4=More than 200g 34. Smoked mackerel 1=50-100g 2=101-150g 3=151 -200g 4=More than 200g 35. Smoked herrings 1=50-100g 2=101-150g 3=151 -200g 4=More than 200g 36. Sundried sardines 1=50-100g 2=101-150g 3=151 -200g 4=More than 200g III. PREGNANT WOMEN 37. Do you have a pregnant woman/women in your home? 1 = Yes 2 = No If No to Q 37, go to Q 45 141 How often do they consume traditionally processed fish? 38. Smoked tuna 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify............................................................................ 39. Smoked mackerel 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify............................................................................ 40. Smoked herrings 1=Daily 2=Once a month 3=2 – 3 times a week 4=Once a week 5=Other, specify........................................................................... How much do they consume at an instance? 41. Smoked tuna 1=50-100g 2=101-150g 3=151 -200g 4=More than 200g 42. Smoked mackerel 1=50-100g 4 = More than 200g 2=101-150g 3=151 -200g 43. Smoked herrings 1=50-100g 2=101-150g 3=151 -200g 4=More than 200g 142 44. Sundried sardines 1=50-100g 2=101-150g 3=151 -200g 4=More than 200g D: FISH HANDLING PRACTICES 45. Where do you buy your traditionally processed fish? 1 = Informal market 2 = Processing sites 3 = House to house vendors 4 = Other, specify.................................................................... 46. How much traditionally processed fish do you often purchase? 1 = Enough quantity for one meal 2 = Enough quantity for three days 3 = Enough quantity for one week 4 = Enough quantity for one month 47. How do you store unused purchased traditionally processed fish? 1 = In a refrigerator 2 = Under the sun during the day, and room temperature at dusk 3 = In polythene bags at room temperature 4 = Other, specify 48. In what form(s) do you consume traditionally processed fish? 1 = Consumed as is 2 = Cooked in soups 3 = Cooked in stews 4 = Roasted 5 = Washed and broken into fresh pepper sauce THANK YOU 143 Appendix 2: Processor Questionnaire DEPARTMENT OF NUTRITION AND FOOD SCIENCE UNIVERSITY OF GHANA, LEGON RISK ASSESSMENT FOR LISTERIA MONOCYTOGENES IN TRADITIONALLY PROCESSED FISH IN GHANA Dear respondent, this questionnaire seeks to solicit some information on traditional fish processing in Ghana, as part of an MPhil Food Science Thesis on the topic above. The information you provide in this document will be treated as confidential and used for academic purposes only. Thank you. Date: _______________________ Area: _______________________ Processor Code:_______________ Kindly tick (√) the responses that apply to you. Where appropriate, write out your own responses in the spaces provided. A: BACKGROUND INFORMATION RESPONSE [For interviewer use only] 49. Sex: 1=Male 2=Female 50. Age: 1=Less than 20 years 2=20 – 29 years 3=30 – 39 years 4=40 – 49 years 5=50 years and above 51. Highest level of education received 1=None 4=Secondary 2=Primary 5=Tertiary 3=Middle School/JHS 6=Other, specify........................ 52. How long have you been in the fish processing business? 1= 1-5 years 2= 6-10 years 3= 11-15 years 4= 16-20 years 5= More than 20 years 144 53. What kind of fish products do you process? Tick as many as apply to you. 1=Smoked fish 2=Salted Fish 3=Dried fish 4=All the above 5=Other, specify.............................................................. B: RAW MATERIAL ACQUISITION 54. What kind of fish do you process? 1 = Marine fish 2 = Freshwater fish 55. Where do you get your raw fish from? 1 = Fishermen 2 = Fishmongers 3 = Cold Store 3 = Open market 4 = Other, specify....................................................... 56. What species of fish do you process? 1=Salmon 2=Tuna 3=Tilapia 4=Other, specify............................................................ 57. Do you inspect fresh fish before purchasing? 1=Yes 2=No 58. If yes to 9, what do you look out for? 1=Colour of eyes 2=Colour of gills 3=Skin surface (smooth or slimy) 4=Other, specify..................................................................... D: TRANSPORTATION OF RAW FISH 59. How long does it take to transport raw fish to the processing site? 1=Less than 30 minutes 2=30mins – 1 hour 3=More than 1 hour, less than 10 hours 4=More than 10 hours, less than 24 hours 5= More than 24 hours 145 60. How do you transport the raw fish to the processing site? 1=By foot 2=Public transport 3=Private transport 4=Refrigerated truck/van 5=Other, specify..................................................................... 61. What containers do you use to carry the raw fish during transportation? 1=Basket 2=Basin 3=Ice chest 4=Other, specify E: PROCESSING OF FISH 62. Do you wash your hands before starting processing? 1=Yes 2=No 63. What do you use to wash your hands? 1= Only water 2=Water and soap 3=Other, specify........................................................ 64. How long do you keep the fish before starting processing? 1=Less than 30 minutes 2=30mins – 1 hour 3=More than 1hour, less than 1 day 4=More than 1 day, less than 1 week? 65. How do you keep raw fish before starting processing? 1=At room temperature 2=In a fridge 3=In a freezer 4=Other, specify....................................................... 66. Describe how you process your fish. 146 DESCRIPTION OF PROCESSING METHODS (Space for interviewer use only) 67. How do you know when raw fish is adequately processed? 68. How much fish do you process at a time/what constitutes a batch? 1= Less than 1 carton 2= 1 – 5 cartons 3= 6 – 10 cartons 4= More than 10 cartons 69. What do you do to keep raw fish from spoiling when processing is delayed? F: HANDLING AND STORAGE OF PROCESSED FISH 70. Where do you store processed fish? 1= Regular room 147 2= Wooden shed 3= In refrigerator 4= In deep freezer, freezer compartments of refrigerators 4= Other, specify.......................................................... 71. How are the processed fish stored? 1 = In basket/sacks 2 = In perforated boxes 3 = In solid boxes (not perforated) 4 = Arranged on wooded trays 5 = Other, specify............................................... 72. For how long after processing do you store fish before selling? 1= Less than 1 day 2= 1 – 3 days 3= More than 3 days, less than 1 week 4= 1 week – 1 month 5 = More than a month G. TRANSPORTATION OF PROCESSED FISH 73. Approximately how long does it take to transport processed fish from the storage/processing site to the market? 1= Less than 30 minutes 2= 30 mins – 2 hours 3= 3 – 6 hours 4= 4 – 12 h 5= More than 12 hours 74. How do you transport processed fish to the market? 1= By foot 2= Public transport 3= Private transport 4= Refrigerated truck/van 5= Other, specify....................................................... 75. Which markets do you send your processed fish to? THANK YOU 148 Appendix 3: Responses from Consumer Survey I. Data on Respondents Table 1: Gender distribution of respondents Jamestown Tema New Town Madina Gender Number % Number % Number % Male 55 36.7 49 32.7 34 22.7 Female 95 63.3 101 67.3 116 77.3 Total 150 100 150 100 150 100 Table 2: Age distribution of respondents Jamestown Tema New Town Madina Age group (yrs) Number % Number % Number % Less than 20 9 6 2 1.3 6 4 20-29 59 39.3 66 44 41 27.3 30-39 49 32.7 44 29.3 53 35.3 40-49 20 13.3 16 10.7 19 12.7 50 and above 13 8.7 22 14.7 31 20.7 Total 150 100 150 100 150 100 Table 3: Highest level of education attained by respondents Jamestown Tema New Town Madina Level Number % Number % Number % None 7 4.7 8 5.3 5 3.3 Primary 40 26.7 46 30.7 35 23.3 Middle School/JHS 52 34.7 59 39.3 45 30 Secondary 51 34 37 24.7 53 35.3 Tertiary 0 0 0 0 12 8 Total 150 100 150 100 150 100 Table 4: Frequency of tuna consumption among respondents Jamestown Tema New Town Madina Frequency Number % Number % Number % Daily 38 25.3 22 14.7 40 26.7 2-3 times a week 105 70 114 76 70 46.7 Once a week 7 4.7 14 9.3 40 26.7 Total 150 100 150 100 150 100 149 Table 5: Quantities of smoked tuna often consumed at an instance by respondents Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 101-150 9 6 10 6.7 13 8.7 151-200 46 30.7 63 42 32 21.3 More than 200 95 63.3 77 51.3 105 70 Total 150 100 150 100 150 100 Table 6: Frequency of consumption of smoked salmon among respondents Jamestown Tema New Town Madina Frequency Number % Number Percentage Number % Daily 50 33.3 48 32 72 48 2-3 times a week 90 60 93 62 67 44.7 Once a week 10 6.7 9 6 11 7.3 Total 150 100 150 100 150 100 Table 7: Quantities of smoked salmon often consumed at an instance by respondents Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 101-150 16 10.7 20 13.3 12 8 151-200 60 40 62 41.3 67 44.7 More than 200 74 49.3 68 45.3 71 47.3 Total 150 100 150 100 150 100 Table 8: Frequency of consumption of smoked herring among respondents Jamestown Tema New Town Madina Frequency Number % Number % Number % Daily 7 4.7 1 0.7 0 0 2-3 times a week 40 26.7 53 35.3 0 0 Once a week 76 50.7 77 51.3 55 36.7 Once a month 27 18 19 12.7 95 63.3 Total 150 100 150 100 150 100 150 Table 9: Quantities of smoked herring often consumed at an instance by respondents Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 50-100 13 8.7 16 10.7 29 19.3 101-150 39 26 65 43.3 87 58 151-200 88 58.7 61 40.7 34 22.7 More than 200 10 6.7 8 5.3 0 0 Total 150 100 150 100 150 100 Table 10: Frequency of consumption of momoni among respondents Jamestown Tema New Town Madina Frequency Number % Number % Number % Daily 0 0 0 0 0 0 2-3 times a week 6 4 3 2 0 0 Once a week 73 48.7 86 57.3 34 22.7 Once a month 24 16 25 16.7 48 32 Never 47 31.3 36 24 68 45.3 Total 150 100 150 100 150 100 Table 11: Quantities of momoni often consumed at an instance by respondents Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 5-10g 53 35.3 48 32 55 36.7 11-20 41 27.3 52 34.7 27 18 21-30 9 6 14 9.3 0 0 Not applicable 47 31.3 36 24 68 45.3 Total 150 100 150 100 150 100 151 Table 12: Frequency of koobi consumption among respondents Jamestown Tema New Town Madina Frequency Number % Number % Number % Daily 0 0 0 0 0 0 2-3 times a week 0 0 0 0 0 0 Once a week 48 32 63 42 10 6.7 Once a month 98 65.3 81 54 95 63.3 Never 4 2.7 6 4 45 30 Not applicable 0 0 0 0 0 0 Total 150 100 150 100 150 100 Table 13: Quantities of koobi often consumed at an instance by respondents Jamestown Tema New Town Madina Quantities (g) Number % Number % Number % 10-20 9 6 14 9.3 5 3.3 21-30 44 29.3 24 16 24 16 31-40 10 6.7 17 11.3 11 7.3 41-50 66 44 56 37.3 38 25.3 More than 50 17 11.3 33 22 27 18 Not applicable 4 2.7 6 4 45 30 Total 150 100 150 100 150 100 Table 14: Frequency of consumption of kako among respondents Jamestown Tema New Town Madina Frequency Number % Number % Number % Daily 0 0 0 0 0 0 2-3 times a week 19 12.7 16 10.7 0 0 Once a week 43 28.7 46 30.7 15 10 Once a month 46 30.7 48 32 70 46.7 Never 42 28 40 26.7 65 43.3 Not applicable 0 0 0 0 0 0 Total 150 100 150 100 150 100 Table 15: Quantities of kako often consumed at an instance by respondents Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 5-10g 58 38.7 60 40 43 28.7 11-20 30 20 26 17.3 29 19.3 21-30 20 13.3 24 16 13 8.7 Not applicable 42 28 40 26.7 65 43.3 Total 150 100 150 100 150 100 152 Table 16: Frequency of consumption of sundried sardines among consumers Jamestown Tema New Town Madina Frequency Number % Number of respondents % Number % Daily 0 0 0 0 0 0 2-3 times a week 17 11.3 24 16 0 0 Once a week 34 22.7 46 30.7 9 6 Once a month 99 66 80 53.3 124 82.7 Never 0 0 0 0 17 11.3 Not applicable 0 0 0 0 0 0 Total 150 100 150 100 150 100 Table 17: Quantities of sundried sardines often consumed at an instance by consumers Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 50-100 35 23.3 41 27.3 49 32.7 101-150 60 40 55 36.7 57 38 151-200 55 36.7 54 36 27 18 More than 200 0 0 0 0 17 11.3 Total 150 100 150 100 150 100 Table 18: Purchasing sites for traditionally processed fish Jamestown Tema New Town Madina Number % Number % Number % Processing Site 150 100 73 48.7 0 0 Informal market 91 60.7 150 100 150 100 House to House Vendors 146 97.3 149 99.3 136 90.7 Table 19: Quantities of traditionally processed fish often purchased at an instance Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % Enough for one meal 132 88 147 98 148 98.7 Enough for three days 18 12 3 2 71 0 Enough for a week 0 0 0 0 0 153 Table 20: Use of traditionally processed fish among respondents Jamestown Tema New Town Madina How used Number % Number % Number % Used in stews 150 100 150 100 150 100 Used in soups 150 100 150 100 150 100 Broken into fresh, unheated pepper sauce 146 97.3 133 88.7 52 34.7 II. Data on Children (6months to 6years old) Table 21: Respondents with children aged 6months to 6years in their household Jamestown Tema New Town Madina Number % Number % Number % Yes 97 64.7 97 64.7 113 75.3 No 53 35.3 53 35.3 37 24.6 Total 150 100 150 100 150 100 Table 22: Frequency of consumption of smoked tuna among children Jamestown Tema New Town Madina Frequency Number % Number % Number % Daily 10 6.7 10 6.7 30 20 2-3 times a week 74 49.3 13 8.7 50 33.33 Once a week 13 8.7 74 49.3 33 22 Not applicable 53 35.3 53 35.3 37 24.6 Total 150 100 150 100 150 100 Table 23: Quantities of smoked tuna often consumed at an instance by children Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 50-100 26 17.3 26 17.3 75 50 101-150 71 47.3 71 47.3 38 25.3 Not applicable 53 35.3 53 35.3 37 24.6 Total 150 100 150 100 150 100 154 Table 24: Frequency of consumption of smoked salmon among children Jamestown Tema New Town Madina Frequency Number % Number % Number % Daily 32 21.3 32 21.3 50 33.3 2-3 times a week 64 42.7 64 42.7 53 35.3 Once a week 1 0.7 1 0.7 10 6.7 Not applicable 53 35.3 53 35.3 37 24.7 Total 150 100 150 100 150 100 Table 25: Quantities of smoked salmon often consumed at an instance by children Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 50-100 29 19.3 29 19.3 62 41.3 101-150 62 41.3 62 41.3 48 32 151-200 6 4 6 4 3 2 Not applicable 53 35.3 53 35.3 37 24.7 Total 150 100 150 100 150 100 Table 26: Frequency of consumption of smoked herring among children Jamestown Tema New Town Madina Frequency Number % Number % Number % 2-3 times a week 9 6 9 6 39 26 Once a week 74 49.3 74 49.3 62 41.3 Once a month 14 9.3 14 9.3 37 24.7 Not applicable 53 35.3 53 35.3 12 8 Total 150 100 150 100 150 100 Table 27: Quantities of smoked herring often consumed at an instance by children Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 50-100 72 48 72 48 80 53.3 101-150 25 16.7 25 16.7 21 14 Not applicable 53 35.3 53 35.3 49 32.7 Total 150 100 150 100 150 100 155 Table 28: Frequency of consumption of sundried sardines among children Jamestown Tema New Town Madina Frequency Number % Number % Number % Once a week 15 10 15 10 2 1.3 Once a month 82 54.7 82 54.7 87 58.7 Never 0 0 0 0 37 24 Not applicable 53 35.3 53 35.3 24 16 Total 150 100 150 100 150 100 Table 29: Quantities of sundried sardines often consumed at an instance by children Jamestown Tema New Town Madina Quantity (g) Frequency % Frequency % Number % 50-100g 97 64.7 97 64.7 86 58.7 101-150g 0 0 0 0 3 1.3 Not applicable 53 35.3 53 35.3 61 40 Total 150 100 150 100 150 100 III. Data on the Elderly (≥60years old) Table 30: Respondents in James Town with the elderly (≥60years old) in their households Jamestown Tema New Town Madina Number % Number % Number % Yes 104 69 93 62 80 53 No 46 31 57 38 70 47 Total 150 100 150 100 150 100 Table 31: Frequency of consumption of smoked tuna among the elderly Jamestown Tema New Town Madina Number % Number % Number % Daily 54.0 36.0 40.0 26.7 2.0 1.3 2-3 times a week 48.0 32.0 45.0 30.0 25.0 16.7 Once a week 2.0 1.3 8.0 5.3 53.0 35.3 Not applicable 46.0 30.7 57.0 38.0 70.0 46.7 Total 150.0 100.0 150.0 100.0 150.0 100.0 156 Table 32: Quantities of smoked tuna often consumed at an instance by the elderly Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 101-150 35.0 23.3 25.0 16.7 18.0 12.0 151-200 37.0 24.7 39.0 26.0 37.0 24.7 More than 200 32.0 21.3 29.0 19.3 25.0 16.7 Not applicable 46.0 30.7 57.0 38.0 70.0 46.7 Total 150.0 100.0 150.0 100.0 150.0 100.0 Table 33: Frequency of consumption of smoked salmon among the elderly Jamestown Tema New Town Madina Frequency Number % Number % Number % Daily 31.0 20.7 24.0 16.0 12.0 8.0 2-3 times a week 54.0 36.0 55.0 36.7 57.0 38.0 Once a week 19.0 12.7 14.0 9.3 11.0 7.3 Not applicable 46.0 30.7 57.0 38.0 70.0 46.7 Total 150.0 100.0 150.0 100.0 150.0 100.0 Table 34: Quantities of smoked salmon often consumed at an instance by the elderly Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 101-150 29.0 19.3 29.0 19.3 21.0 14.0 151-200 72.0 48.0 54.0 36.0 54.0 36.0 More than 200 3.0 2.0 10.0 6.7 5.0 3.3 Not applicable 46.0 30.7 57.0 38.0 70.0 46.7 Total 150.0 100.0 150.0 100.0 150.0 100.0 Table 35: Frequency of consumption of smoked herrings among the elderly Jamestown Tema New Town Madina Frequency Number % Number % Number % Daily 3.0 2.0 18.0 12.0 2.0 1.3 2-3 times a week 18.0 12.0 23.0 15.3 6.0 4.0 Once a week 70.0 46.7 46.0 30.7 51.0 34.0 Once a month 13.0 8.7 6.0 4.0 21.0 14.0 Not applicable 46.0 30.7 57.0 38.0 70.0 46.7 Total 150.0 100.0 150.0 100.0 150.0 100.0 157 Table 36: Quantities of smoked herring often consumed at an instance by the elderly Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 50-100 19.0 12.7 13.0 8.7 31.0 20.7 101-150 46.0 30.7 41.0 27.3 38.0 25.3 151-200 39.0 26.0 39.0 26.0 11.0 7.3 Not applicable 46.0 30.7 57.0 38.0 70.0 46.7 Total 150.0 100.0 150.0 100.0 150.0 100.0 Table 37: Frequency of consumption of momoni among the elderly Jamestown Tema New Town Madina Frequency Number % Number % Number % Once a week 5.0 3.3 2.0 1.3 3.0 2.0 Once a month 30.0 20.0 31.0 20.7 18.0 12.0 Never 69.0 46.0 60.0 40.0 59.0 38.0 Not applicable 46.0 30.7 57.0 38.0 70.0 48.0 Total 150.0 100.0 150.0 100.0 150.0 100.0 Table 38: Quantities of momoni often consumed at an instance by the elderly Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 5-10 18.0 12.0 17.0 11.3 12.0 8.0 11-20 17.0 11.3 16.0 10.7 9.0 6.0 Not applicable 115.0 76.7 117.0 78.0 129.0 86.0 Total 150.0 100.0 150.0 100.0 150.0 100.0 Table 39: Frequency of consumption of kako among the elderly Jamestown Tema New Town Madina Frequency Number % Number % Number % Once a month 33.0 22.0 32.0 21.3 18.0 12.0 Never 71.0 47.3 61.0 40.7 62.0 41.3 Not applicable 46.0 30.7 57.0 38.0 70.0 46.7 Total 150.0 100.0 150.0 100.0 150.0 100.0 158 Table 40: Quantity of kako often consumed at an instance by the elderly Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 5-10 16.0 10.7 16.0 10.7 8.0 5.3 11-15 17.0 11.3 16.0 10.7 10.0 6.7 Not applicable 117.0 78.0 118.0 78.7 132.0 88.0 Total 150.0 100.0 150.0 100.0 150.0 100.0 Table 41: Frequency of consumption of koobi among the elderly Jamestown Tema New Town Madina Frequency Number % Number % Number % Once a month 38.0 25.3 32.0 21.3 33.0 22.0 Never 66.0 44.0 61.0 40.7 47.0 31.3 Not applicable 46.0 30.7 57.0 38.0 70.0 46.7 Total 150.0 100.0 150.0 100.0 150.0 100.0 Table 42: Quantities of koobi often consumed at an instance by the elderly Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 10-20 3.0 2.0 1.0 0.7 3.0 2.0 21-30 23.0 15.3 20.0 13.3 19.0 12.7 31-40 11.0 7.3 9.0 6.0 10.0 6.7 41-50 1.0 0.7 2.0 1.3 1.0 0.7 Not applicable 112.0 74.7 118.0 78.7 117.0 78.0 Total 150.0 100.0 150.0 100.0 150.0 100.0 Table 43: Frequency of consumption of sundried sardines among the elderly Jamestown Tema New Town Madina Frequency Number % Number % Number % 2-3 times a week 30.0 20.0 30.0 20.0 0.0 0.0 Once a week 29.0 19.3 38.0 25.3 39.0 26.7 Once a month 45.0 30.0 25.0 16.7 41.0 27.3 Not applicable 46.0 30.7 57.0 38.0 70.0 46.0 Total 150.0 100.0 150.0 100.0 150.0 100.0 159 Table 44: Quantities of sundried sardines often consumed at an instance by the elderly Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 50-100 23.0 15.3 15.0 10.0 36.0 24.0 101-150 51.0 34.0 53.0 35.3 37.0 24.7 151-200 30.0 20.0 25.0 16.7 7.0 5.3 Not applicable 46.0 30.7 57.0 38.0 70.0 46.0 Total 150.0 100.0 150.0 100.0 150.0 100.0 IV. Data on Pregnant Women Table 45: Respondents with pregnant women in their household Jamestown Tema New Town Madina Number % Number % Number % Yes 58 38.7 50 33.3 46 30.67 No 92 61.3 100 66.7 104 69.33 Total 150 100 150 100 150 100 Table 46: Frequency of consumption of smoked tuna among pregnant women Jamestown Tema New Town Madina Frequency Number % Number % Number % Daily 34 22.7 16 10.7 4.67 4.67 2-3 times a week 24 16 34 22.7 23.33 23.33 Once a week 0 0 0 0 0.00 0.00 Once a month 0 0 0 0 2.67 2.67 Not applicable 92 61.3 100 66.7 69.33 69.33 Total 150 100 150 100 100 100 Table 46: Quantities of smoked tuna often consumed at an instance by pregnant women Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 101-150 0 0 2 1.3 1 0.67 151-200 12 8 18 12 9 6 More than 200 46 30.7 30 20 36 24 Not applicable 92 61.3 100 66.7 104 69.3 Total 150 100 150 100 150 100 160 Table 48: Frequency of consumption of smoked salmon among pregnant women Jamestown Tema New Town Madina Frequency Number % Number % Number % Daily 23 15.3 14 9.3 7 4.67 2-3 times a week 35 23.33 36 24 33 22 Once a week 0 0 0 0 0 0 Once a month 0 0 0 0 6 4 Not applicable 92 61.33 100 66.7 104 69.33 Total 150 100 150 100 150 100 Table 49: Quantities of smoked salmon often consumed at an instance by pregnant women Jamestown Tema New Madina Quantity (g) Number % Number % Number % 101-150 10 6.7 4 2.7 2 1.33 151-200 19 12.7 19 12.7 22 14.67 More than 200 29 19.3 27 18 22 14.67 Not applicable 92 61.3 100 66.7 104 69.33 Total 150 100 150 100 150 100 Table 50: Frequency of consumption of smoked herring among pregnant women Jamestown Tema New Town Madina Frequency Number % Number % Number % 2-3 times a week 24 16 22 14.7 0 0 Once a week 34 22.7 28 18.7 46 30.67 Not applicable 92 61.3 100 66.7 104 69.33 Total 150 100 150 100 150 100 Table 51: Quantities of smoked herring often consumed at an instance by pregnant women Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 50-100 7 4.7 4 2.7 9 6 101-150 28 18.7 26 17.3 17 11.33 151-200 22 14.7 18 12 19 12.67 More than 200 1 0.7 2 1.3 1 0.67 Not applicable 92 61.3 100 66.7 104 69.33 Total 150 100 150 100 150 100 161 Table 52: Frequency of consumption of momoni among pregnant women Jamestown Tema New Town Madina Frequency Number % Number % Number % Once a week 54 36 48 32 23 15.33 Once a month 0 0 0 0 17 11.33 Never 4 2.7 2 1.3 104 69.33 Not applicable 92 61.3 100 66.7 6 4 Total 150 100 150 100 150 100 Table 53: Quantities of momoni often consumed at an instance by pregnant women Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 5-10g 21 14 21 14 27 18 11-20 17 11.3 21 14 12 8 21-30 16 10.7 6 4 1 0.67 31-40 0 0 102 68 0 0 Not applicable 96 64 0 0 110 73.33 Total 150 100 150 100 150 100 Table 54: Frequency of consumption of kako among pregnant women Jamestown Tema New Town Madina Frequency Number % Number % Number % Once a week 40 26.7 37 24.7 12 8 Once a month 10 6.7 8 5.3 22 14.67 Not applicable 92 61.3 5 3.3 104 69.33 Never 8 5.3 100 66.7 12 8 Total 150 100 150 100 150 100 162 Table 55: Quantities of kako often consumed at an instance by pregnant women Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 5-10 26 17.3 35 23.3 18 12 11-20 4 2.7 2 1.3 7 4.67 21-30 16 10.7 6 4 6 4 31-40 4 2.7 2 1.3 3 2 Not applicable 100 66.7 105 70 116 77.33 Total 150 100 150 100 150 100 Table 56: Frequency of consumption of koobi among pregnant women Jamestown Tema New Town Madina Frequency Number % Number % Number % Once a week 13 8.7 24 16 4 2.67 Once a month 45 30 26 17.3 42 28 Not applicable 92 61.3 100 66.7 104 69.33 Total 150 100 150 100 150 100 Table 57: Quantities of koobi often consumed at an instance by pregnant women Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 10-20 6 4 2 1.3 3 2 21-30 14 9.3 12 8 2 1.33 31-40 10 6.7 12 8 8 5.33 More than 40 28 18.7 24 16 33 22 Not applicable 92 61.3 100 66.7 104 69.33 Total 150 100 150 100 150 100 Table 58: Frequency of consumption of sundried sardines among pregnant women Jamestown Tema New Town Madina Frequency Number % Number % Number % 2-3 times a week 5 3.3 16 10.7 0 0 Once a week 34 22.7 27 18 12 8 Once a month 19 12.7 7 4.7 34 22.67 Not applicable 92 61.3 100 66.7 104 69.33 Total 150 100 150 100 150 100 163 Table 59: Quantities of sundried sardines often consumed at an instance by pregnant women Jamestown Tema New Town Madina Quantity (g) Number % Number % Number % 50-100 11 7.3 5 3.3 32 21.33 101-150 15 10 23 15.3 6 4 151-200 32 21.3 22 14.7 8 5.33 Not applicable 92 61.3 100 66.7 104 69.33 Total 150 100 150 100 150 100 164 Appendix 4: Microbiology of fish during traditional fish processing I. Jamestown Table 60: Microbiology of samples along tuna processing chain TPC TCC E. coli LM Sample C1 C2 Av G.Av C1 C2 Av G.Av C1 C2 Av G.Av Fresh tuna S1 88 92 90 94 53 46 49.5 45 17 13 15 17 + S2 92 103 97.5 42 39 40.5 21 17 19 + Fresh tuna after washing S1 72 89 80.5 83 30 37 33.5 31 11 14 12.5 16 + S2 77 94 85.5 21 34 27.5 21 15 18 + Water before washing S1 39 41 40 44 22 27 24.5 26 15 11 13 15 + S2 44 53 48.5 25 31 28 13 19 16 + Water after washing S1 55 60 57.5 56 44 39 41.5 48 27 33 30 27 + S2 47 61 54 52 57 54.5 21 25 23 + Tuna after smoking S1 0 0 0 0 0 0 0 0 0 0 0 0 - S2 0 0 0 0 0 0 0 0 0 - TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x102 CFU/g or ml, E.coli x102 CFU/g or ml, Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S2: samples 1 and 2 165 Table 61: Microbiology of samples along mackerel processing chain TPC TCC E. coli LM Sample C1 C2 Av G.Av C1 C2 Av G.Av C1 C2 Av G.Av Fresh mackerel S1 0 0 0 0 0 0 0 0 0 0 0 0 - S2 0 0 0 0 0 0 0 0 0 - Mackerel after washing S1 5 9 7 6 3 0 2 3 2 1 2 2 + S2 7 0 4 5 1 3 1 1 1 + Water before washing S1 36 32 34 35 18 20 19 18 11 14 12.5 12 + S2 39 31 35 21 14 17.5 13 10 11.5 + Water after washing S1 44 36 40 43 20 27 23.5 26 21 27 24 24 + S2 51 42 46.5 23 33 28 19 30 24.5 + Mackerel after smoking S1 0 0 0 0 0 0 0 0 0 0 0 0 - S2 0 0 0 0 0 0 0 0 0 - TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x102 CFU/g or ml, E.coli x102 CFU/g or ml, Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S2: samples 1 and 2 166 Table 62: Microbiology of samples along herrings processing chain TPC TCC E. coli LM Sample C1 C2 Av G.Av C1 C2 Av G.Av C1 C2 Av G.Av Fresh herrings S1 0 0 0 0 0 0 0 0 0 0 0 0 - S2 0 0 0 0 0 0 0 0 0 - Herrings after washing S1 11 9 10 12 5 1 3 6 0 2 1 2 + S2 13 15 14 8 9 8.5 1 3 2 + Water before washing S1 39 34 36.5 31 13 15 14 14 9 5 7 8 + S2 21 28 24.5 17 11 14 10 7 8.5 + Water after washing S1 45 39 42 37 21 26 23.5 22 11 14 12.5 10 + S2 35 28 31.5 19 22 20.5 5 9 7 + Herring after smoking S1 0 0 0 0 0 0 0 0 0 0 0 0 - S2 0 0 0 0 0 0 0 0 0 - TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x102 CFU/g or ml, E.coli x102 CFU/g or ml, Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S2: samples 1 and 2 167 II. Tema New Town Table 63: Microbiology of samples along tuna processing chain TPC TCC E. coli LM Sample C1 C2 Av G.Av C1 C2 Av G.Av C1 C2 Av G.Av Fresh tuna S1 122 111 117 113 71 69 70 64 21 27 24 24 + S2 113 106 110 64 53 58.5 23 25 24 + Tuna after washing S1 91 101 96 90 47 41 44 40 21 18 19.5 19 + S2 88 79 83.5 33 39 36 19 16 17.5 + Water before washing S1 31 35 33 33 31 28 29.5 30 20 25 22.5 24 + S2 29 37 33 37 23 30 26 21 23.5 + Water after washing S1 44 60 52 54 62 67 64.5 65 31 32 31.5 31 + S2 58 52 55 59 71 65 37 25 31 + Tuna after smoking S1 0 0 0 0 0 0 0 0 0 0 0 0 - S2 0 0 0 0 0 0 0 0 0 - TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x102 CFU/g or ml, E.coli x102 CFU/g or ml, Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S2: samples 1 and 2 168 Table 64: Microbiology of samples along mackerel processing chain TPC TCC E. coli LM Sample C1 C2 Av G.Av C1 C2 Av G.Av C1 C2 Av G.Av Fresh mackerel S1 0 0 0 0 0 0 0 0 0 0 0 0 + S2 0 0 0 0 0 0 0 0 0 + Mackerel after washing S1 16 10 13 12 15 8 11.5 13 7 4 5.5 6 + S2 11 9 10 17 12 14.5 8 3 5.5 + Water before washing S1 42 33 37.5 34 23 18 20.5 22 23 17 20 22 + S2 35 27 31 28 21 24.5 27 22 24.5 + Water after washing S1 48 63 55.5 55 40 47 43.5 45 31 43 37 33 + S2 57 53 55 43 51 47 28 30 29 + Mackerel after smoking S1 0 0 0 0 0 0 0 0 0 0 0 0 - S2 0 0 0 0 0 0 0 0 0 - TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x102 CFU/g or ml, E.coli x102 CFU/g or ml, Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S2: samples 1 and 2 169 Table 65: Microbiology of samples along herrings processing chain TPC TCC E. coli LM Sample C1 C2 Av G.Av C1 C2 Av G.Av C1 C2 Av G.Av Fresh herrings S1 0 0 0 0 0 0 0 0 0 0 0 0 - S2 0 0 0 0 0 0 0 0 0 - Herrings after washing S1 19 24 21.5 21 10 13 11.5 13 5 2 3.5 6 + S2 18 22 20 12 15 13.5 9 7 8 + Water before washing S1 39 32 35.5 30 21 25 23 21 15 11 13 12 + S2 21 27 24 18 21 19.5 10 12 11 + Water after washing S1 47 44 45.5 44 27 23 25 24 11 13 12 10 + S2 40 46 43 22 25 23.5 6 9 7.5 + Mackerel after smoking S1 0 0 0 0 0 0 0 0 0 0 0 0 - S2 0 0 0 0 0 0 0 0 0 - TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x102 CFU/g or ml, E.coli x102 CFU/g or ml, Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S2: samples 1 and 2 170 Table 66: Microbiology of samples along kako processing chain TPC TCC E. coli LM Sample C1 C2 Av G.Av C1 C2 Av G.Av C1 C2 Av G.Av Fresh fish S1 132 137 135 129 82 75 78.5 84 57 63 60 65 + S2 125 121 123 91 86 88.5 71 68 69.5 + Fish after washing S1 98 107 103 106 68 57 62.5 63 61 57 59 51 + S2 117 102 110 55 71 63 49 38 43.5 + Water before washing S1 64 50 57 59 38 35 36.5 33 25 18 21.5 22 + S2 58 65 61.5 27 32 29.5 21 23 22 + Water after washing S1 97 91 94 95 91 86 88.5 98 55 47 51 53 + S2 101 89 95 111 103 107 51 58 54.5 + Fish after salting S1 0 0 0 0 0 0 0 0 0 0 0 0 + S2 0 0 0 0 0 0 0 0 0 + Fish after sundrying S1 17 23 20 19 9 5 7 5 0 3 1.5 3 + S2 20 15 17.5 7 0 3.5 1 7 4 + TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x102 CFU/g or ml, E.coli x102 CFU/g or ml, Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S2: samples 1 and 2 171 Table 67: Microbiology of samples along momoni processing chain TPC TCC E. coli LM Sample C1 C2 Av G.Av C1 C2 Av G.Av C1 C2 Av G.Av Fresh fish S1 121 132 127 124 79 86 82.5 85 42 37 39.5 47 + S2 119 123 121 83 90 86.5 51 59 55 + After washing S1 108 111 110 119 77 54 65.5 63 38 26 32 29 + S2 131 127 129 63 59 61 29 22 25.5 + Water before washing S1 85 73 79 84 51 47 49 38 18 16 17 18 + S2 91 87 89 31 25 28 21 18 19.5 + Water after washing S1 161 142 152 151 79 62 70.5 73 19 13 16 23 + S2 155 147 151 82 71 76.5 32 27 29.5 + Fish after salting and fermenting S1 0 0 0 0 0 0 0 0 0 0 0 0 + S2 0 0 0 0 0 0 0 0 0 + Fish after sundrying S1 17 11 14 16.5 8 11 9.5 11 4 9 6.5 8 + S2 20 18 19 10 14 12 7 11 9 + TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x102 CFU/g or ml, E.coli x102 CFU/g or ml, Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S2: samples 1 and 2 172 Table 68: Microbiology of samples along sundried fish processing chain TPC TCC E. coli LM Sample C1 C2 Av G.Av C1 C2 Av G.Av C1 C2 Av G.Av Fresh fish S1 64 72 68 72 37 24 30.5 30 20 33 26.5 26 + S2 81 69 75 28 31 29.5 28 22 25 + Fish after washing S1 68 66 67 57 27 21 24 28 27 20 23.5 26 + S2 59 35 47 32 32 32 24 31 27.5 + Water before washing S1 28 21 24.5 31 17 12 14.5 18 5 11 8 5 + S2 41 32 36.5 19 21 20 0 2 1 + Water after washing S1 83 59 71 78 23 19 21 21 10 16 13 14 + S2 92 78 85 16 24 20 20 12 16 + Fish after sundrying S1 19 23 21 27 25 18 21.5 19 19 24 21.5 26 + S2 37 29 33 13 21 17 27 31 29 + TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x102 CFU/g or ml, E.coli x102 CFU/g or ml, Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S2: samples 1 and 2 173 Table 69: Microbial counts and detection of Listeria monocytogenes in traditionally processed fish purchased from Jamestown Market Smoked Tuna TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 87 91 89 83 32 27 29.5 23 20 21 20.5 19 17 20 18.5 12 + S2 71 69 70 24 22 23 19 17 18 15 11 13 + S3 87 92 89.5 18 14 16 13 24 18.5 1 8 4.5 + Smoked Mackerel TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 72 76 74 75 41 33 37 40 17 20 18.5 11 8 9.5 9 + S2 74 68 71 58 44 51 22 19 20.5 18 6 6 6 + S3 82 79 80.5 35 29 32 11 21 16 9 13 11 + Smoked Herrings TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 85 79 82 68 25 25 25 25 20 21 20.5 19 14 23 18.5 9 + S2 55 68 61.5 30 27 28.5 19 17 18 0 0 0 - S3 63 59 61 21 19 20 13 24 18.5 5 11 8 + Sundried Sardines TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 98 83 90.5 58 47 52.5 47 15 11 13 18 4 9 6.5 2 + S2 75 67 71 74 50 45 47.5 17 21 19 0 0 0 - S3 69 52 60.5 42 38 40 20 23 21.5 0 0 0 - TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x104 CFU/g or ml, E.coli x103 CFU/g or ml, LM: Listeria monocytogenes x102 Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S3: samples 1,2,3 LM DET: detection of Listeria monocytogenes APPENDIX 5: OCCURRENCE OF L. MONOCYTOGENES AND OTHER MICROOGANISMS IN FISH ON INFORMAL MARKETS 174 Table 69 continued Kako TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 22 29 25.5 26 19 21 20 20 10 13 11.5 9 4 2 3 1 + S2 24 27 25.5 21 17 19 9 0 4.5 0 0 0 - S3 31 25 28 20 23 21.5 10 10 10 0 0 0 - Koobi TPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 23 27 25 23 17 19 18 17 9 12 10.5 9 0 0 0 2 - S2 19 25 22 11 16 13.5 0 7 3.5 4 9 6.5 + S3 26 19 22.5 19 21 20 13 10 11.5 0 0 0 - Momoni TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 37 29 33 27 21 23 22 20 0 3 1.5 6 1 0 0.5 2 + S2 25 23 24 21 19 20 8 14 11 2 5 3.5 + S3 27 22 24.5 16 21 18.5 0 11 5.5 0 0 0 - TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x104 CFU/g or ml, E.coli x103 CFU/g or ml, LM: Listeria monocytogenes x102 Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S3: samples 1,2,3 LM DET: detection of Listeria monocytogenes 175 Table 70: Microbial counts and detection of Listeria monocytogenes in traditionally processed fish purchased from Tema New Town Market Smoked Tuna TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 75 61 68 72 45 37 41 39 18 22 20 39 0 0 0 2 - S2 67 73 70 30 43 36.5 30 57 43.5 0 0 0 - S3 80 73 76.5 39 33 36 47 60 53.5 5 8 7 + Smoked Mackerel TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 77 68 72.5 79 33 37 35 43 25 21 23 28 0 0 0 4 - S2 70 83 76.5 41 61 51 31 27 29 5 7 6 + S3 84 91 87.5 47 36 41.5 33 28 30.5 3 6 5 + Smoked Herrings TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 58 50 54 50 27 18 22.5 23 12 18 15 7 11 9 3 - S2 61 53 57 31 27 29 14 11 12.5 17 0 0 0 - S3 47 33 40 22 14 18 11 33 22 0 0 0 - Sundried Sardines TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 71 83 77 80 44 38 41 43 36 47 41.5 33 0 0 0 3 - S2 89 92 90.5 51 47 49 32 28 30 5 7 6 + S3 76 70 73 38 41 39.5 27 33 30 3 5 4 + TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x104 CFU/g or ml, E.coli x103 CFU/g or ml, LM: Listeria monocytogenes x102 Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S3: samples 1,2,3 LM DET: detection of L. monocytogenes 176 Table 70 continued Kako TPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 29 33 31 28 12 15 13.5 14 5 9 7 2 0 0 0 1 - S2 26 29 27.5 17 22 19.5 0 0 0 0 0 0 - S3 31 19 25 11 9 10 0 0 0 4 2 3 + Koobi TPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 12 9 10.5 15 0 0 0 3 0 0 0 1 0 0 0 0 - S2 19 11 15 11 7 9 5 0 3 0 0 0 - S3 21 17 19 0 0 0 0 0 0 0 0 0 - Momoni TPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 12 7 9.5 15 10 17 14 11 0 0 0 0 0 0 0 0 - S2 17 15 16 15 22 19 0 0 0 0 0 0 - S3 21 19 20 0 0 0 0 0 0 2 3 2 + TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x104 CFU/g or ml, E.coli x103 CFU/g or ml, LM: Listeria monocytogenes x102 Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S3: samples 1,2,3 LM DET: detection of Listeria monocytogenes 177 Table 71: Microbial counts and detection of Listeria monocytogenes in traditionally processed fish purchased from Madina Market Smoked Tuna TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 75 49 62 87 34 39 36.5 39 27 19 23 25 18 25 21.5 21 + S2 101 86 93.5 42 36 39 21 24 22.5 27 32 29.5 + S3 97 113 105 37 39 38 32 29 30.5 8 13 10.5 + Smoked Mackerel TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 93 81 87 76 27 31 29 30 18 24 21 24 15 47 31 29 + S2 73 77 75 33 37 35 27 31 29 33 51 42 + S3 66 70 68 28 24 26 20 22 21 18 9 13.5 + Smoked Herrings TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 36 42 39 47 31 29 30 35 17 25 21 27 14 10 12 8 + S2 49 57 53 45 34 39.5 38 27 32.5 0 0 0 + S3 51 49 50 37 32 34.5 23 31 27 18 7 12.5 + Sundried Sardines TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 73 85 79 76 51 58 54.5 52 31 31 31 31 3 5 4 3 + S2 80 69 74.5 44 39 41.5 28 35 31.5 1 7 4 + S3 67 79 73 56 63 59.5 26 33 29.5 0 0 0 + TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x104 CFU/g or ml, E.coli x103 CFU/g or ml, LM: Listeria monocytogenes x102 Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S3: samples 1,2,3 LM DET: detection of Listeria monocytogenes 178 Table 71 continued Kako TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 27 21 24 26 10 13 11.5 12 9 4 6.5 7 5 11 8 7 + S2 28 23 25.5 18 10 14 3 10 6.5 6 18 12 + S3 25 31 28 12 10 11 7 11 9 0 0 0 + Koobi TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 19 20 19.5 21 11 14 12.5 11 3 3 3 4 15 19 17 6 + S2 16 24 20 9 10 9.5 4 7 5.5 0 0 0 + S3 25 21 23 13 11 12 3 4 3.5 0 0 0 - Momoni TPC TCC E. coli LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 28 34 31 43 10 14 12 13 0 4 2 3 11 3 7 3 + S2 51 46 48.5 13 19 16 1 3 2 1 4 2.5 + S3 38 59 48.5 11 16 13.5 4 5 4.5 0 0 0 - TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x104 CFU/g or ml, E.coli x103 CFU/g or ml, LM: Listeria onocytogenes x102 Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S3: samples 1,2,3 LM DET: +detection of Listeria monocytogenes 179 Table 72: Microbial counts and detection of Listeria monocytogenes in traditionally processed fish purchased from Kaneshie Market Smoked Tuna SPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 78 91 85 89 50 43 46.5 49 15 21 18 20 21 33 27 17 + S2 92 102 97 61 55 58 20 17 19 16 11 14 + S3 84 88 86 45 39 42 23 25 24 9 14 12 + Smoked Mackerel SPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 93 87 90 86 37 26 31.5 35 11 14 13 22 8 11 9.5 16 + S2 76 73 75 38 31 34.5 24 24 24 3 0 1.5 + S3 96 90 93 43 32 37.5 30 29 30 29 37 33 + Smoked Herrings SPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 76 79 78 81 30 34 32 41 20 23 22 25 0 0 0 0 - S2 84 87 86 51 42 46.5 38 27 33 0 0 0 + S3 79 83 81 48 39 43.5 19 22 21 0 0 0 + Sundried Sardines SPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 112 103 108 97 77 72 74.5 50 28 21 25 25 0 0 0 0 - S2 90 97 94 34 28 31 24 19 22 0 0 0 + S3 94 91 93 45 49 47 32 27 30 0 0 0 + 180 Table 72 continued Kako SPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 67 53 60 58 23 20 21.5 22 13 7 10 11 0 0 0 6 - S2 69 49 59 19 25 22 10 17 14 11 25 18 + S3 55 57 56 21 27 24 9 11 10 0 0 0 - Koobi SPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 63 50 56.5 57 20 22 21 20 11 12 12 11 21 29 25 9 + S2 51 63 57 19 26 22.5 9 10 9.5 0 0 0 - S3 69 48 58.5 12 19 15.5 13 11 12 1 5 3 + Momoni SPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 30 37 33.5 37 17 15 16 20 10 12 11 12 19 12 16 5 + S2 45 49 47 13 21 17 11 10 11 0 0 0 - S3 29 31 30 23 28 25.5 14 12 13 0 0 0 + TPC: Total plateTPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x104 CFU/g or ml, E.coli x103 CFU/g or ml, LM: Listeria monocytogenes x102 Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S3: samples 1,2,3 LM DET: detection of L. monocytogenes count x105 CFU/g or ml, TCC: Total coliform x104 CFU/g or ml, E.coli x103 CFU/g or ml, LM: Listeria monocytogenes x102 Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S3: samples 1,2,3 LM DET: detection of Listeria monocytogenes 181 Table 73: Microbial counts and detection of Listeria monocytogenes in traditionally processed fish purchased from Agbogbloshie Market Smoked Tuna SPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 79 73 76 82 31 28 30 31 35 27 31 30 10 14 12 8 + S2 96 87 91.5 34 29 32 25 21 23 12 9 10.5 + S3 82 77 79.5 29 33 31 32 40 36 0 0 0 - Smoked Mackerel SPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 58 69 63.5 71 22 29 26 28 19 21` 19 23 5 3 4 10 + S2 84 78 81 33 28 31 31 23 27 15 21 18 + S3 61 73 67 34 21 28 20 25 23 7 11 9 + Smoked Herrings SPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 47 32 39.5 48 52 47 50 55 21 18 20 14 3 1 2 1 + S2 53 40 46.5 67 71 69 17 12 15 0 0 0 - S3 61 53 57 55 38 47 10 8 9 1 0 0.5 + Sundried Sardines SPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 35 27 31 42 80 93 87 89 77 81 79 80 0 0 0 0 - S2 50 44 47 97 103 100 72 93 83 0 0 0 - S3 38 56 47 73 88 81 83 76 80 0 0 0 + TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x104 CFU/g or ml, E.coli x103 CFU/g or ml, LM: Listeria monocytogenes x102 Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S3: samples 1,2,3 LM DET: detection of Listeria monocytogenes 182 Table 73 continued Kako SPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 27 31 29 34 12 7 9.5 12 0 0 0 1 2 3 2.5 1 + S2 40 37 38.5 11 9 10 4 2 3 0 0 0 - S3 38 32 35 15 19 17 0 0 0 1 0 0.5 + Koobi SPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 19 25 22 24 11 4 7.5 0 0 0 0 0 0 0 0 - S2 22 17 19.5 7 16 12 10 0 0 0 0 0 0 - S3 31 29 30 6 13 9.5 0 0 0 0 0 0 + Momoni SPC TCC E COLI LM LM DET C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av C1 C2 Av. G.Av S1 21 23 22 22 9 8 8.5 8 0 0 0 0 0 0 0 0 - S2 24 19 21.5 5 8 6.5 0 0 0 0 0 0 + S3 28 20 24 12 4 8 0 0 0 0 0 0 - TPC: Total plate count x105 CFU/g or ml, TCC: Total coliform x104 CFU/g or ml, E.coli x103 CFU/g or ml, LM: Listeria monocytogenes x102 Av: Average G.Av: Grand average C1-C2: counts for plate 1 and 2; S1- S3: samples 1,2,3 LM DET: +detection of Listeria monocytogenes 183 183 14.9 82.8 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0 1 0 0 1 1 0 P ro b a b ili ty Pill Values in Millionths Fig. 1: Triangular distribution for probability of illness among respondents in Jamestown who consumed smoked tuna contaminated with Listeria monocytogenes 10.7 80.1 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0 10 0 11 0 P ro b ab ili ty Pill Values in Millionths Fig. 2: Triangular distribution for probability of illness among respondents in Jamestown who consumed smoked mackerel contaminated with Listeria monocytogenes 10.7 80.1 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 10 20 30 40 50 60 70 80 90 10 0 11 0 P ro b a b ili ty Pill Values in Millionths Fig 3: Triangular distribution for probability of illness among respondents in Jamestown who consumed smoked herrings contaminated with Listeria monocytogenes 0.166 0.843 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 P ro b ab ili ty Pill Values in Millionths Fig. 4: Triangular distribution for probability of illness among respondents in Jamestown who consumed sundried sardines contaminated with Listeria monocytogenes 0.130 0.814 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 P ro b ab ili ty Pill Values in Millionths Fig. 5: Triangular distribution for probability of illness among respondents in Jamestown who consumed kako contaminated with Listeria monocytogenes 16.4 84.1 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 10 20 30 40 50 60 70 80 90 10 0 P ro b ab ili ty Pill Values in Billionths Fig.6: Triangular distribution for probability of illness among respondents in Jamestown who consumed koobi contaminated with Listeria monocytogenes 1.07 8.01 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 10 11 P ro b a b ili ty Pill Values in Billionths Fig. 7: Triangular distribution for probability of illness among respondents in Jamestown who consumed momoni contaminated with Listeria monocytogenes Appendix 6: Triangular distributions for probability of illness, Pill , in Jamestown, Tema New Town and Madina (a) Jamestown Respondents 184 184 0.1632 0.4367 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 1 2 3 4 5 6 0. 05 0. 1 0. 15 0. 2 0. 25 0. 3 0. 35 0. 4 0. 45 0. 5 0. 55 Fig. 8: Triangular distribution for probability of illness among the elderly in Jamestown who consume smoked tuna contaminate d with Listeria monocytogenes 0.1332 0.2700 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 2 4 6 8 10 12 0 .0 5 0 .1 0 .1 5 0 .2 0 .2 5 0 .3 0 .3 5 Fig. 9: Triangular distribution for probability of illness among the elderly in Jamestown who consume smoked mackerel contaminated with Listeria monocytogenes 0.1406 0.3466 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 1 2 3 4 5 6 7 0. 05 0. 1 0. 15 0. 2 0. 25 0. 3 0. 35 0. 4 0. 45 Fig. 10: Triangular distribution for probability of illness among the elderly in Jamestown who consume smoked herrings contaminated with Listeria monocytogenes 2.90 9.33 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 10 11 Pill Values in Thousandths Fig. 11: Triangular distribution for the probability of illness among the elderly in Jamestown who consume sundried sardines contaminated with Listeria monocytogenes 0.181 0.859 5.0% 5.0% 90.0% 90.0% .0% 5.0% 0 0.5 1 1.5 2 2.5 0 0 .1 0 .2 0 .3 0 .4 0 .5 0 .6 0 .7 0 .8 0 .9 1 1 .1 P ro b ab ili ty Pill Values in Millionths Fig. 12:Triangular distribution for the probability of illness among the elderly in Jamestown who consume kako contaminated with Listeria monocytogenes 0.217 0.923 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 1. 1 Pill Values in Thousandths Fig. 13: Triangular distribution for the probability of illness among the elderly in Jamestown who consume koobi contaminated with Listeria monocytogenes 12.3 81.3 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 20 40 60 80 10 0 12 0 V al u es x 1 0^ 4 Values in Millionths Fig. 14: Triangular distribution for the probability of illness among the elderly in Jamestown who consume momoni contaminated with Listeria monocytogenes Pill P ro b ab ili ty P ro b ab ili ty Pill` P ro b ab ili ty Pill` P ro b ab ili ty P ro b ab ili ty (b) Jamestown Elderly 185 185 0.1316 0.2684 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 2 4 6 8 10 12 0. 05 0. 1 0. 15 0. 2 0. 25 0. 3 0. 35 Fig. 15: Triangular distribution for probability of illness among children in Jamestown who ingest Listeria monocytogenes through consumption of smoked tuna 0.0534 0.2518 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 1 2 3 4 5 6 7 0 0. 05 0. 1 0. 15 0. 2 0. 25 0. 3 0. 35 Fig. 16: Triangular distribution for probability of illness among children in Jamestown ingesting Listeria monocytogenes through consumption of smoked mackerel 0.0250 0.0866 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 01 0. 02 0. 03 0. 04 0. 05 0. 06 0. 07 0. 08 0. 09 0. 1 0. 11 Fig. 17: Triangular distribution for probability of illness among children in Jamestown ingesting Listeria monocytogenes through consumption of smoked herrings 1.591 3.988 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 1 2 3 4 5 6 0. 5 1 1. 5 2 2. 5 3 3. 5 4 4. 5 5 Pill Values in Thousandths Fig. 18: Triangular distribution for probability of illness among children in Jamestown ingesting Listeria monocytogenes through consumption of sundried sardines P ro b ab ili ty Pill Pill P ro b ab ili ty Pill Pill P ro b ab ili ty P ro b ab ili ty (c) Jamestown Children 186 186 0.102 0.506 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 3 3.5 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 Fig. 19: Triangular distribution for probability of illness among pregnant women in Jamestown who consumed smoked tuna contaminated with Listeria monocytogenes 0.0392 0.1692 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 2 4 6 8 10 12 0 0. 02 0. 04 0. 06 0. 08 0. 1 0. 12 0. 14 0. 16 0. 18 0. 2 0. 22 Fig 20" Triangular distribution for probability of illness among pregnant women in Jamestown who consumed smoked mackerel contaminated with Listeria monocytogenes 0.0566 0.2549 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 1 2 3 4 5 6 7 0 0. 05 0. 1 0. 15 0. 2 0. 25 0. 3 0. 35 Fig. 21: Triangular distribution for probability of illness among pregnant women in Jamestown who consumed smoked herrings contaminated with Listeria monocytogenes 0.0107 0.0801 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 01 0. 02 0. 03 0. 04 0. 05 0. 06 0. 07 0. 08 0. 09 0 .1 Fig. 22: Triangular distribution for probability of illness among pregnant women in Jamestown who consumed sundried sardines contaminated with Listeria monocytogenes 1.81 8.59 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 10 P ro b a b ili ty PillValues in Millionths Fig. 23: Triangular distribution for probability of illness among pregnant women in Jamestown who consumed kako contaminated with Listeria monocytogenes 0.208 0.900 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 1. 1 Pill Values in Thousandths Fig. 24: Triangular distribution for probability of illness among pregnant women in Jamestown who consumed koobi contaminated with Listeria monocytogenes 13.0 81.4 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 10 20 30 40 50 60 70 80 90 10 0 11 0 P ro b ab ili ty Pill Values in Millionths Fig 25: Triangular distribution for probability of illness among pregnant women in Jamestown who consumed momoni contaminated with Listeria monocytogenes P ro b ab ili ty Pill Pill P ro b ab ili ty P ro b ab ili ty Pill P ro b ab ili ty Pill (d) Jamestown Pregnant Women 187 187 0.226 0.950 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 1. 1 P ro b ab ili ty Pill Values in Millionths Fig. 26: Triangular distribution for probability of illness among respondents in Tema New Town who consumed smoked tuna contaminated with Listeria monocytogenes 1.51 8.29 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 10 11 P ro b ab ili ty Pill Values in Millionths Fig. 2Triangular distribution for probability of illness among respondents in Tema New Town who consumed smoked mackerel contaminated with Listeria monocytogenes 1.09 8.02 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 1 0 1 1 P ro b ab ili ty Pill Values in Millionths Fig. 28: Triangular distribution for probability of illness among respondents in Tema New Town who consumed smoked herrings contaminated with Listeria monocytogenes 0.84 7.91 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 10 11 P ro b ab ili ty Pill Values in Millionths Fig. 29: Triangular distribution for probability of illness among respondents in Tema New Town who consumed sundried sardines contaminated with Listeria monocytogenes 16.6 84.3 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 10 20 30 40 50 60 70 80 90 10 0 11 0 P ro b ab ili ty Pill Values in Billionths Fig. 30: Triangular distribution for probability of illness among respondents in Tema New Town who consumed kako contaminated with Listeria monocytogenes (e) Tema New Town Respondents 188 188 0.0093 0.0795 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 01 0. 02 0. 03 0. 04 0. 05 0. 06 0. 07 0. 08 0. 09 0. 1 0. 11 Fig. 31: Triangular distribution for probability of illness among the elderly in Jamestown who consumed smoked tuna contaminated with Listeria monocytogenes 0.0188 0.0868 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 01 0. 02 0. 03 0. 04 0. 05 0. 06 0. 07 0. 08 0. 09 0. 1 0. 11 Fig. 32: Triangular distribution for probability of illness among the elderly in Jamestown who consumed smoked mackerel contaminated with Listeria monocytogenes 0.0109 0.0802 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 0 1 0. 0 2 0. 0 3 0. 0 4 0. 0 5 0. 0 6 0. 0 7 0. 0 8 0. 0 9 0 .1 0. 1 1 Fig. 33: Triangular distribution for probability of illness among the elderly in Jamestown who consumed smoked herrings contaminated with Listeria monocytogenes 0.0109 0.0802 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 01 0. 02 0. 03 0. 04 0. 05 0. 06 0. 07 0. 08 0. 09 0. 1 0. 11 Fig. 34: Triangular distribution for probability of illness among the elderly in Jamestown who consumed sundried sardines contaminated with Listeria monocytogenes 1.88 8.68 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 1 0 1 1 P ro b a b il iy Piill Values in Millionths Fig. 35: Triangular distribution for probability of illness among the elderly in Jamestown who consumed kako contaminated with Listeria monocytogenes Pill P ro b ab ili ty P ro b ab ili ty Pill P ro b ab ili ty Pill Pill P ro b ab ili ty (f) Tema New Town Elderly 189 189 2.20 9.30 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 10 11 Pill Values in Thousandths Fig. 36: Triangular distribution for probability of illness among children in Jamestown who consumed smoked tuna contaminated with Listeria monocytogenes 0.0140 0.0820 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 01 0. 02 0. 03 0. 04 0. 05 0. 06 0. 07 0. 08 0. 09 0. 1 0. 11 Fig. 37: Triangular distribution for probability of illness among children in Jamestown who consumed smoked mackerel contaminated with Listeria monocytogenes 2.20 9.30 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 1 0 1 1 Pill Values in Thousandths Fi.g 38: Triangular distribution for probability of illness among children in Jamestown who consumed smoked herrings contaminated with Listeria monocytogenes 2.20 9.30 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 1 0 1 1 Pill Values in Thousandths Fig. 39: Triangular distribution for probability of illness among children in Jamestown who consumed sundried sardines contaminated with Listeria monocytogenes P ro b ab ili ty P ro b ab ili ty Pill P ro b ab ili ty P ro b ab ili ty (g) Tema New Town Children 190 190 0.0093 0.0795 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 01 0. 02 0. 03 0. 04 0. 05 0. 06 0. 07 0. 08 0. 09 0. 1 Fig. 40: Triangular distribution for probability of illness among pregnant women in Tema New Town who consumed smoked tuna contaminated with Listeria monocytogenes 0.0188 0.0868 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0 .0 1 0 .0 2 0 .0 3 0 .0 4 0 .0 5 0 .0 6 0 .0 7 0 .0 8 0 .0 9 0 .1 0 .1 1 Fig. 41: Triangular distribution for probability of illness among pregnant women in Tema New Town who consumed smoked mackerel contaminated with Listeria monocytogenes 0.0109 0.0802 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 01 0. 02 0. 03 0. 04 0. 05 0. 06 0. 07 0. 08 0. 09 0. 1 0. 11 Fig. 42: Triangular distribution for probability of illness among pregnant women in Tema New Town who consumed smoked herrings contaminated with Listeria monocytogenes 0.0140 0.0821 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 01 0. 02 0. 03 0. 04 0. 05 0. 06 0. 07 0. 08 0. 09 0. 1 Fig. 43: Triangular distribution for probability of illness among pregnant women in Tema New Town who consumed sundried sardines contaminated with Listeria monocytogenes 1.88 8.68 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 1 0 1 1 P ro b a b il it y Pill Values in Millionths Fig. 44: Triangular distribution for probability of illness among pregnant women in Tema New Town who consumed kako contaminated with Listeria monocytogenes P ro b ab ili ty Pill Pill Pill Pill P ro b ab ili ty P ro b ab ili ty Pill P ro b ab ili ty (h) Tema New Town Pregnant Women 191 191 0.077 0.789 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 Pill Values in Thousandths Fig. 45: Triangular distribution for probability of illness among respondents in Madina who consumed smoked salmon contaminated with Listeria monocytogenes 0.119 0.807 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 1. 1 Pill Values in Thousandths Fig. 46: Triangular distribution for probability of illness among respondents in Madina who consumed smoked mackerel contaminated with Listeria monocytogenes 0.080 0.790 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 1. 1 P ro b a b ili ty Pll Values in Millionths Fig. 47: Triangular distribution for probability of illness among respondents in Madina who consumed smoked herrings contaminated with Listeria monocytogenes 0.80 7.90 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 10 11 P ro b ab ili ty Pill Values in Millionths Fig. 48: Triangular distribution for probability of illness among respondents in Madina who consumed sundried sardines contaminated with Listeria monocytogenes 10.9 80.2 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 10 20 30 40 50 60 70 80 90 10 0 11 0 P ro b ab ili ty Pill Values in Billionths Fig. 49: Triangular distribution for probability of illness among respondents in Madina who consumed kako contaminated with Listeria monocytogenes 0.166 0.843 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 0 .1 0 .2 0 .3 0 .4 0 .5 0 .6 0 .7 0 .8 0 .9 1 1 .1 P ro b a b il it y Pill Values in Millionths Fig. 50: Triangular distribution for probability of illness among respondents in Madina who consumed koobi contaminated with Listeria monocytogenes 1.40 8.21 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 10 11 P ro b ab ili ty Pill Values in Billionths Fig. 51: Triangular distribution for probability of illness among respondents in Madina who consumed momoni contaminated with Listeria monocytogenes P ro b ab ili ty P ro b ab ili ty (i) Madina Respondents 192 192 0.157 0.827 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 Fig. 52: Triangular distribution for probability of illness among the elderly in Madina who consumed smoked tuna contaminated with Listeria monocytogenes 0.166 0.837 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 0 .1 0 .2 0 .3 0 .4 0 .5 0 .6 0 .7 0 .8 0 .9 1 Fig. 53: Triangular distribution for probability of illness among the elderly in Madina who consumed smoked mackerel contaminated with Listeria monocytogenes 0.093 0.787 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 Fig. 54: Triangular distribution for probability of illness among the elderly in Madina who consumed smoked herrings contaminated with Listeria monocytogenes 0.0109 0.0802 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 01 0. 02 0. 03 0. 04 0. 05 0. 06 0. 07 0. 08 0. 09 0. 1 Fig. 55: Triangular distribution for probability of illness among the elderly in Madina who consumed sundried sardines contaminated with Listeria monocytogenes 0.140 0.821 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 Pill Values in Thousandths Fig. 56: Triangular distribution for probability of illness among the elderly in Madina who consumed kako contaminated with Listeria monocytogenes 1.30 8.14 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 10 11 Pill Values in Thousandths Fig. 57: Triangular distribution for probability of illness among the elderly in Madina who consumed koobi contaminated with Listeria monocytogenes 19.5 87.8 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 0 2 0 3 0 4 0 5 0 6 0 7 0 8 0 9 0 10 0 11 0 P ro b ab ili ty Pill Values in Millionths Fig. 58: Triangular distribution for probability of illness among the elderly in Madina who consumed momoni contaminated with Listeria monocytogenes Pill Pill P ro b ab ili ty P ro b ab ili ty P ro b ab ili ty P ro b ab ili ty Pill Pill P ro b ab ili ty P ro b ab ili ty (j) Madina Elderly 193 193 0.118 0.590 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 3 0 0 .1 0 .2 0 .3 0 .4 0 .5 0 .6 0 .7 0 .8 Fig. 59: Triangular distribution for probability of illness among children in Madina who consumed smoked tuna contaminated with Listeria monocytogenes 0.0188 0.0868 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0 .0 1 0 .0 2 0 .0 3 0 .0 4 0 .0 5 0 .0 6 0 .0 7 0 .0 8 0 .0 9 0 .1 Fig. 60: Triangular distribution for probability of illness among children in Madina who consumed smoked herring contaminated with Listeria monocytogenes 0.165 0.835 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 Fig. 61: Triangular distribution for probability of illness among children in Madina who consumed smoked mackerel contaminated with Listeria monocytogenes 2.26 9.50 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 10 Pil Values in Thousandths Fig. 62: Triangular distribution for probability of illness among children in Madina who consumed sundried sardines contaminated with Listeria monocytogenes P ro b ab ili ty P ro b ab ili ty P ro b ab ili ty Pill Pill Pill P ro b ab ili ty (k) Madina Children 194 194 0.150 0.821 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 Fig. 63: Triangular distribution for probability of illness among pregnant women in Madina who consumed smoked tuna contaminated with Listeria monocytogenes 0.166 0.837 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 Fig. 64: Triangular distribution for probability of illness among pregnant women in Madina who consumed smoked mackerel contaminated with Listeria monocytogenes 0.080 0.782 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 Fig. 65: Triangular distribution for probability of illness among pregnant women in Madina who consumed smoked herrings contaminated with Listeria monocytogenes 0.0149 0.0828 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0 .0 1 0 .0 2 0 .0 3 0 .0 4 0 .0 5 0 .0 6 0 .0 7 0 .0 8 0 .0 9 0 .1 0 .1 1 Fig. 66: Triangular distribution for probability of illness among pregnant women in Madina who consumed sundried sardines contaminated with Listeria monocytogenes 0.140 0.820 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 5 10 15 20 25 0 0. 1 0. 2 0. 3 0. 4 0. 5 0. 6 0. 7 0. 8 0. 9 1 1. 1 Fig. 67: Values in Thousandths Triangular distribution for probability of illness among pregnant women in Madina who consumed kako contaminated with Listeria monocytogenes 2.25 9.45 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 1 2 3 4 5 6 7 8 9 10 11 Pill Values in Thousandths Fig. 68: Triangular distribution for probability of illness among pregnant women in Madina who consumed koobi contaminated with Listeria monocytogenes 19.5 87.8 5.0% 5.0% 90.0% 90.0% 5.0% 5.0% 0 0.5 1 1.5 2 2.5 0 10 20 30 40 50 60 70 80 90 10 0 11 0 P ro b ab ili ty Pill Values in Millionths Fig. 69: Triangular distribution for probability of illness among pregnant women in Madina who consumed momoni contaminated with Listeria monocytogenes P ro b ab ili ty Pill P ro b ab ili ty P ro b ab ili ty Pill P ro b ab ili ty P ro b ab ili ty Pill Pill P ro b ab ili ty (l) Madina Pregnant Women