Conceived and designed the experiments: RCK ALW PMS. Performed the experiments: RCK LMM LN LM. Analyzed the data: RCK HM LM SRW. Contributed reagents/materials/analysis tools: RCK LMM SRW. Wrote the paper: RCK HM PMS LM ALW SRW.
Ms Lynne Michael was not employed by Bayer at the time that she worked on this study. She was employed afterwards. This does not alter the authors' adherence to all the PLoS One policies on sharing data and materials. Other authors have declared that no competing interests exist.
There is a high prevalence of
A high prevalence of
The ECP reported high levels (28–50%) of human juvenile NCC (which occurs in children) and limited current data for porcine cysticercosis
This study was carried out from February to June 2003, in the six veterinary districts of the Alfred Nzo and Oliver R. Tambo Districts of the ECP. These districts included 124 villages and are served by animal health officers (AHO) and state veterinarians. The role of AHOs is to serve specific villages in a district. Therefore they contacted pig owners and invited them to participate in this study. The criteria for owners to be invited to participate in the study were that they were emerging pig producers, would be available during the days that the research team was scheduled to conduct the survey, and that the villages were accessible by road. All owners who agreed to participate were scheduled to meet on specific days with the research team. Details of the numbers of pigs and the village sampling are reported elsewhere
Data were collected from a standard hardcopy questionnaire that included 42 questions on household identification, information and respondent details, pig breed (South African hut breed, cross bred or other pure breed), pig management and husbandry, possible transmission factors (e.g. source of drinking water, absence of a latrine or toilet) and awareness of taeniosis/cysticercosis in humans and of cysticercosis in pigs. The questionnaire methodology followed
Data from the hardcopy questionnaires were entered into Microsoft Excel
Bivariable models were adjusted for clusters of pigs within owner, while the multivariable model was simultaneously adjusted for clusters of pigs within owner and clusters of owners within district. Porcine infection with cysticercosis was determined using two ELISA tests. B158/B60 Ag-ELISA
The study protocol was approved by the National Department of Agriculture Directorate, South Africa and Animal Ethics Committee of Veterinary Services Eastern Cape Province, Department of Agriculture, ECP, South Africa, village leaders and pig owners. Collection of blood specimens from pigs was performed by a licensed veterinarian according to South African guidelines for animal care.
A definitive account of the results for the true prevalence of porcine cysticercosis as well as operating characteristics of the laboratory tests used can be found in the previous papers
Of the pig owners interviewed, 73% were female; 50% were under 51 years of age; and 79% had less than or equal to standard 8 level of education. Only a few owners (12%) confined their pigs while the remainder (88%) utilized a free range system or semi-intensive system; 90% kept pigs for home consumption. Twenty percent of households obtained drinking water from taps, while 73% obtained their water from rivers, tanks, and boreholes. Forty-eight percent of households had no latrines, and their members defecated in the fields outside the homes.
The majority (93%) of households reported consumption of pork, 98% slaughtered pigs at home and almost all (99%) indicated that meat inspection services were not available. Most owners (74%) had observed cysts in pork, 78% did not know what the cysts were and 83% did not know how pigs acquired cysticercosis. The majority of respondents (69%) had heard of tapeworm infection in humans, though 74% did not know how to recognize a tapeworm if they were individually infected and 87% did not know how infection with a tapeworm occurred. The majority (86%) had heard of persons in their village complaining of epilepsy, 44% of headache and 62% of madness. Seventy-one percent reported that they did not know about the disease or how to treat, prevent or manage the disease.
Overall, the prevalence (95% CI) of porcine infection was 57% (51%, 63%); 41% (35%, 47%) and 54% (48%, 60%) were positive by B158/B60 Ag-ELISA and HP10 Ag-ELISA, respectively. There was some variation between districts, with the prevalence of porcine infection ranging from 29% to 73% (
Risk factor | Category | Infected (n = 146) | Not infected (n = 110) | Crude odds ratio (95% CI) |
p-value |
n (%) |
n (%) |
||||
Veterinary district |
Umzimukulu | 28 (58) | 20 (42) | 1.17 ( 0.53 , 2.59 ) | 0.691 |
Maluti | 17 (49) | 18 (51) | 0.79 ( 0.34 , 1.85 ) | 0.590 | |
Tsolo | 32 (71) | 13 (29) | 2.06 ( 0.89 , 4.78 ) | 0.091 | |
Qumbu | 29 (73) | 11 (28) | 2.21 ( 0.92 , 5.29 ) | 0.075 | |
Lusikisiki | 9 (29) | 22 (71) | 0.34 ( 0.14 , 0.86 ) | 0.022 | |
Mt. Frere | 31 (54) | 26 (46) | Reference | ||
Breed |
Cross bred | 18 (40) | 27 (60) | 0.42 ( 0.22 , 0.83 ) | 0.012 |
Other | 2 (67) | 1 (33) | 1.27 ( 0.08 , 20.97 ) | 0.868 | |
Hut pig | 123 (61) | 78 (39) | Reference | ||
Latrine |
Absent | 82 (66) | 42 (34) | 2.08 ( 1.24 , 3.49 ) | 0.006 |
Present | 63 (48) | 67 (52) | Reference | ||
Husbandry system |
Confined | 23 (56) | 18 (44) | 0.95 ( 0.48 , 1.85 ) | 0.870 |
Free-ranging/semi-intensive | 123 (57) | 91 (43) | Reference | ||
Measles given as a main problem with pig rearing |
Yes | 57 (52) | 53 (48) | 0.67 ( 0.4 , 1.13 ) | 0.131 |
No | 88 (62) | 55 (38) | Reference | ||
Source of drinking water |
Tap water | 27 (54) | 23 (46) | 0.86 ( 0.45 , 1.63 ) | 0.635 |
Natural sources or boreholes | 118 (58) | 86 (42) | Reference | ||
Consumption of pork at home |
Yes | 135 (56) | 104 (44) | Reference | 0.500 |
No | 6 (67) | 3 (33) | 1.54 ( 0.44 , 5.42 ) | ||
Meat inspected by a meat inspector |
Yes | 0 (0) | 0 (0) |
|
- |
No | 145 (57) | 109 (43) | Reference | ||
Pigs slaughtered at home |
Yes | 142 (57) | 109 (43) | Reference | |
No | 3 (100) | 0 (0) |
|
- |
- not determined, due to cells with zero counts.
Determined using two enzyme-linked immunosorbent assays (B158/B60 Ag-ELISA
Analyses were performed using the survey logistic procedure of the SAS/STAT® software to adjust for clusters of pigs within one owner.
(%) refers to row percentages.
CI = Confidence interval.
Type 3 wald statistic had a
Type 3 wald statistic had a
Had missing values for n = 2.
Had missing values for n = 1.
Had missing values for n = 3. Measles are cysts, from the larvae of
Had missing values for n = 2.
Had missing values for n = 8.
Had missing values for n = 2.
Had missing values for n = 2.
Note: no significant associations (
Risk factor | Category | Adjusted odds ratio (95% CI) |
|
Breed |
Cross bred | 0.54 (0.26, 1.12) | 0.099 |
Other | 1.80 (0.10, 31.38) | 0.687 | |
Hut pig | Reference | ||
Latrine |
Absent | 1.89 (1.07, 3.35) | 0.028 |
Present | Reference |
Determined using two enzyme-linked immunosorbent assays (B158/B60 Ag-ELISA
Analyses were performed using the surveylogistic procedure of the SAS/STAT® software to simultaneously adjust for clusters of pigs within one owner and for clusters of owners within district of residence.
CI = Confidence interval.
Had missing values for n = 7.
Had missing values n = 2. Other includes other pure bred breeds.
This is the first community-based survey in an emerging pig producing and endemic area in South Africa. The objective of this study was to identify risk factors for porcine cysticercosis. As mentioned earlier in this paper, the true prevalence of porcine cysticercosis in the ECP was previously estimated, using Bayesian analysis, as 64.6%
Both the Bayesian and frequentist methods suggest a very high prevalence of this disease, especially when compared to other prevalence studies, using Ag-ELISA, in Zambia (30–52%) and Tanzania (3–48%)
Households lacking a latrine (or toilet) were associated with an 89% increase in the odds of porcine infection, and this contributes to our understanding of the transmission of this parasite in this study area. Our study agrees with those carried out in Zambia and Tanzania where the prevalence of porcine cysticercosis was considerably higher in pigs reared in households lacking latrines
It is essential to consider the results of the current study in the context of the socioeconomic characteristics of the study population. Of the nine provinces in South Africa, the ECP continues to be the most challenged in terms of infrastructure. For example, when 2007 provincial and national figures are compared, the ECP reported the highest proportion (48%) of households with no toilets (or latrines) as compared to the provincial and national figures of 25% and 9% , respectively. This province also has the lowest proportion of homes with access to piped water (70% vs 89%). In addition, unemployment is second highest with 25% as compared to the national average of 21%. Levels of unemployment, the proportion of households without toilets and homes with access to piped water have improved across the country, but little has changed in this province since 2003
The presence of
Mr. D Stockdale and Dr. K Lucas of the United States Agency for International Development (USAID) Mission in South Africa, and RUSVM are thanked for support; Dr. L Mrwebi, Dr. R Hyde, Mr. M Lephallo and other staff of the Veterinary Services Eastern Cape Province; the Cysticercosis Working Group of Eastern and Southern Africa for technical inputs to field collections; Mr. TE Krecek of Krecek and Krecek CC, and Mr. B Weaver, Mr. B Ho-Thanh, Mr. M McMillan, Mr. D John-King, J Heller, J Moore and A Hefter of RUSVM for technical assistance; Dr. LJS Harrison, Dr. P Dorny and Dr. N Praet for inputs with the diagnostic test analysis; Ms. MF Smith of the Agricultural Research Council for statistical support; the emerging pig producers who cooperated so willingly throughout this study; and several anonymous reviewers for valuable comments that helped improve previous versions of this manuscript.