Modelling seasonal household variation in harvested rainwater availability: a case study in Siaya County, Kenya.

Rainwater harvesting reliability, the proportion of days annually when rainwater demand is fully met, is challenging to estimate from cross-sectional household surveys that underpin international monitoring. This study investigated the use of a modelling approach that integrates household surveys with gridded precipitation data to evaluate rainwater harvesting reliability, using two local-scale household surveys in rural Siaya County, Kenya as an illustrative case study. We interviewed 234 households, administering a standard questionnaire that also identified the source of household stored drinking water. Logistic mixed effects models estimated stored rainwater availability from household and climatological variables, with random effects accounting for unobserved heterogeneity. Household rainwater availability was significantly associated with seasonality, storage capacity, and access to alternative improved water sources. Most households (95.1%) that consumed rainwater faced insufficient supply of rainwater available for potable needs throughout the year, with intermittencies during the short rains for most households with alternative improved sources. Although not significant, stored rainwater lasts longer for households whose only improved water source was rainwater (301.8 ± 40.2 days) compared to those having multiple improved sources (144.4 ± 63.7 days). Such modelling analysis could enable rainwater harvesting reliability estimation, and thereby national/international monitoring and targeted follow-up fieldwork to support rainwater harvesting.

Quality of maternal healthcare and travel time influence birthing service utilisation in Ghanaian health facilities: a geographical analysis of routine health data.

OBJECTIVES: To investigate how the quality of maternal health services and travel times to health facilities affect birthing service utilisation in Eastern Region, Ghana. DESIGN: The study is a cross-sectional spatial interaction analysis of birth service utilisation patterns. Routine birth data were spatially linked to quality care, service demand and travel time data. SETTING: 131 Health facilities (public, private and faith-based) in 33 districts in Eastern Region, Ghana. PARTICIPANTS: Women who gave birth in health facilities in the Eastern Region, Ghana in 2017. OUTCOME MEASURES: The count of women giving birth, the quality of birthing care services and the geographic coverage of birthing care services. RESULTS: As travel time from women's place of residence to the health facility increased up to two2 hours, the utilisation rate markedly decreased. Higher quality of maternal health services haves a larger, positive effect on utilisation rates than service proximity. The quality of maternal health services was higher in hospitals than in primary care facilities. Most women (88.6%) travelling via mechanised transport were within two2 hours of any birthing service. The majority (56.2%) of women were beyond the two2 -hour threshold of critical comprehensive emergency obstetric and newborn care (CEmONC) services. Few CEmONC services were in urban centres, disadvantaging rural populations. CONCLUSIONS: To increase birthing service utilisation in Ghana, higher quality health facilities should be located closer to women, particularly in rural areas. Beyond Ghana, routinely collected birth records could be used to understand the interaction of service proximity and quality.

A spatiotemporal analysis of cattle herd movement in relation to drinking-water sources: implications for Cryptosporidium control in rural Kenya.

Given the increasing evidence that domestic contact with livestock is a risk factor for child diarrhoea in low- and middle-income countries, there have been calls for greater quantification of human-livestock contact in such countries. This study aimed to quantify seasonality in cattle proximity to domestic water sources and household compounds and develop a preliminary landscape model of faecal deposition by cattle. A total of 120 cattle in smallholder herds in the Asembo area of Siaya County, Kenya, were tracked over 1 week in April 2018 to July 2018 and November 2018 to February 2019 using GPS tracking devices. Dung deposition and behaviour were observed among 33 cattle from these herds over 185.4 hours. Mean cattle home ranges were small at 3.78 km2 and 5.85 km2 in the wet and dry seasons, respectively. There were significant differences between seasons in home range size, distance travelled from the household, and time spent tethered, but not in the time spent at domestic water sources or home range overlap with other herds. On average, 0.76 dung deposition events/hour were observed, with higher frequency in bulls. Variation in cattle proximity to household compounds and water sources did not account for seasonal variation in child diarrhoea in this population. The preliminary landscape model of faecal deposition by cattle could be further developed to inform interventions for safe separation of livestock and people, such as fencing and separate water troughs.

Impact of traffic congestion on spatial access to healthcare services in Nairobi.

Background: Geographic accessibility is an important determinant of healthcare utilization and is critical for achievement of universal health coverage. Despite the high disease burden and severe traffic congestion in many African cities, few studies have assessed how traffic congestion impacts geographical access to healthcare facilities and to health professionals in these settings. In this study, we assessed the impact of traffic congestion on access to healthcare facilities, and to the healthcare professionals across the healthcare facilities. Methods: Using data on health facilities obtained from the Ministry of Health in Kenya, we mapped 944 primary, 94 secondary and four tertiary healthcare facilities in Nairobi County. We then used traffic probe data to identify areas within a 15-, 30- and 45-min drive from each health facility during peak and off-peak hours and calculated the proportion of the population with access to healthcare in the County. We employed a 2-step floating catchment area model to calculate the ratio of healthcare and healthcare professionals to population during these times. Results: During peak hours, <70% of Nairobi's 4.1 million population was within a 30-min drive from a health facility. This increased to >75% during off-peak hours. In 45 min, the majority of the population had an accessibility index of one health facility accessible to more than 100 people (<0.01) for primary health care facilities, one to 10,000 people for secondary facilities, and two health facilities per 100,000 people for tertiary health facilities. Of people with access to health facilities, a sub-optimal ratio of <4.45 healthcare professionals per 1,000 people was observed in facilities offering primary and secondary healthcare during peak and off-peak hours. Conclusion: Our study shows access to healthcare being negatively impacted by traffic congestion, highlighting the need for multisectoral collaborations between urban planners, health sector and policymakers to optimize health access for the city residents. Additionally, growing availability of traffic probe data in African cities should enable similar analysis and understanding of healthcare access for city residents in other countries on the continent.

The nexus between improved water supply and water-borne diseases in urban areas in Africa: a scoping review.

Background: The sub-Saharan Africa has the fastest rate of urbanisation in the world. However, infrastructure growth in the region is slower than urbanisation rates, leading to inadequate provision and access to basic services such as piped safe drinking water. Lack of sufficient access to safe water has the potential to increase the burden of waterborne diseases among these urbanising populations. This scoping review assesses how the relationship between waterborne diseases and water sufficiency in Africa has been studied. Methods: In April 2020, we searched the Web of Science, PubMed, Embase and Google Scholar databases for studies of African cities that examined the effect of insufficient piped water supply on selected waterborne disease and syndromes (cholera, typhoid, diarrhea, amoebiasis, dysentery, gastroneteritis, cryptosporidium, cyclosporiasis, giardiasis, rotavirus). Only studies conducted in cities that had more than half a million residents in 2014 were included. Results: A total of 32 studies in 24 cities from 17 countries were included in the study. Most studies used case-control, cross-sectional individual or ecological level study designs. Proportion of the study population with access to piped water was the common water availability metrics measured while amounts consumed per capita or water interruptions were seldom used in assessing sufficient water supply. Diarrhea, cholera and typhoid were the major diseases or syndromes used to understand the association between health and water sufficiency in urban areas. There was weak correlation between the study designs used and the association with health outcomes and water sufficiency metrics. Very few studies looked at change in health outcomes and water sufficiency over time. Conclusion: Surveillance of health outcomes and the trends in piped water quantity and mode of access should be prioritised in urban areas in Africa in order to implement interventions towards reducing the burden associated with waterborne diseases and syndromes.

Effect of Inter-Observer Variation on the Association between Contamination Hazards and the Microbiological Quality of Water Sources: A Longitudinal Study.

Sanitary risk inspection protocols are often used to identify contamination hazards at water sources; however, different observers sometimes struggle to record hazards consistently. This study aimed to assess the effect of inter-observer variation in hazard observations on the strength of relationships between observed hazards and the bacterial contamination of water sources, particularly relationships with animal-related hazards. In a longitudinal study, five surveyors independently recorded hazards at 93 water sources used by 234 households in Siaya County, Kenya, in both wet and dry seasons. One surveyor collected samples from sources for subsequent Escherichia coli and intestinal enterococci testing. The relationship between each surveyor's hazard observations and high bacterial contamination was examined using logistic regression. After controlling for water source type and preceding rainfall; percentage scores for animal-related hazards were significantly related to high contamination with enterococci and E. coli for one surveyor (odds ratio 1.02; 95% confidence intervals 1.00-1.03 for both parameters), but not for the remaining four surveyors. The relationship between observed contamination hazards and the microbiological contamination of water sources is sensitive to variation in hazard recording between surveyors. Sanitary risk protocols should be designed to enable robust and consistent observation of hazards.

An Assessment of Inter-Observer Agreement in Water Source Classification and Sanitary Risk Observations.

Sanitary risk inspection, an observation protocol for identifying contamination hazards around water sources, is promoted for managing rural water supply safety. However, it is unclear how far different observers consistently identify contamination hazards and consistently classify water source types using standard typologies. This study aimed to quantify inter-observer agreement in hazard identification and classification of rural water sources. Six observers separately visited 146 domestic water sources in Siaya County, Kenya, in wet and dry seasons. Each observer independently classified the source type and conducted a sanitary risk inspection using a standard protocol. Water source types assigned by an experienced observer were cross-tabulated against those of his colleagues, as were contamination hazards identified, and inter-observer agreement measures calculated. Agreement between hazards observed by the most experienced observer versus his colleagues was significant but low (intra-class correlation = 0.49), with inexperienced observers detecting fewer hazards. Inter-observer agreement in classifying water sources was strong (Cohen's kappa = 0.84). However, some source types were frequently misclassified, such as sources adapted to cope with water insecurity (e.g. tanks drawing on both piped and rainwater). Observers with limited training and experience thus struggle to consistently identify hazards using existing protocols, suggesting observation protocols require revision and their implementation should be supported by comprehensive training. Findings also indicate that field survey teams struggle to differentiate some water source types based on a standard water source classification, particularly sources adapted to cope with water insecurity. These findings demonstrate uncertainties underpinning international monitoring and analyses of safe water access via household surveys.

A longitudinal study of the association between domestic contact with livestock and contamination of household point-of-use stored drinking water in rural Siaya County (Kenya).

BACKGROUND: Emerging evidence suggests close domestic proximity of livestock and humans may lead to microbiological contamination of hands, objects, food and water supplies within domestic environments, adversely impacting public health. However, evidence quantifying the relationship between livestock, domestic animals, humans and microbiological contamination of household stored water remains limited. AIM: This longitudinal study aimed to examine the relationship between domestic contact with livestock and domestic animals on microbiological contamination of household Point-of-Use (POU) stored drinking water in rural Kenya and assess the influence of choice of faecal indicator on such associations. METHODOLOGY: A survey was performed in 234 households in Siaya county, Kenya, to observe presence of livestock (cattle, goats, poultry) and domestic animals (cats, dogs) in household compounds, alongside other risk factors for contamination of POU stored drinking water such as sanitation, storage conditions and hygiene practices. Samples from water sources (e.g. piped, spring/wells, boreholes, surface and rainwater) and from POU storage containers were tested for E. coli and intestinal enterococci. Livestock-related risk factors for water contamination were examined through multinomial regression, controlling for confounders. RESULTS: Rainwater was the main POU water source and was found to be highly susceptible to contamination. Multivariate analysis showed greater risk of gross (>100 CFU/100 mL) water contamination (with E. coli) for households where goats were observed, and/or where poultry roosted in proximity to stored household water (relative risk RR = 2.71; p = 0.001 and RR = 2.02; p = 0.012 respectively). Presence of a poultry coop was also associated with elevated intestinal enterococci densities (RR = 4.46; p = 0.001). Associations between contamination and livestock risk factors were thus similar for both bacteria groups, but E. coli counts declined more rapidly following collection from surface waters than enterococci counts (p = 0.024). CONCLUSION: The presence of livestock (particularly goats) and poultry within household compounds increases POU water contamination risk, suggesting the need for improved interventions to address cross-contamination within rural domestic settings. Within Siaya county, more effective community education is needed to raise awareness of POU water quality protection, particularly of rainwater.

Mapping access to domestic water supplies from incomplete data in developing countries: An illustrative assessment for Kenya.

Water point mapping databases, generated through surveys of water sources such as wells and boreholes, are now available in many low and middle income countries, but often suffer from incomplete coverage. To address the partial coverage in such databases and gain insights into spatial patterns of water resource use, this study investigated the use of a maximum entropy (MaxEnt) approach to predict the geospatial distribution of drinking-water sources, using two types of unimproved sources in Kenya as illustration. Geographic locations of unprotected dug wells and surface water sources derived from the Water Point Data Exchange (WPDx) database were used as inputs to the MaxEnt model alongside geological/hydrogeological and socio-economic covariates. Predictive performance of the MaxEnt models was high (all > 0.9) based on Area Under the Receiver Operator Curve (AUC), and the predicted spatial distribution of water point was broadly consistent with household use of these unimproved drinking-water sources reported in household survey and census data. In developing countries where geospatial datasets concerning drinking-water sources often have necessarily limited resolution or incomplete spatial coverage, the modelled surface can provide an initial indication of the geography of unimproved drinking-water sources to target unserved populations and assess water source vulnerability to contamination and hazards.

Livestock ownership and microbial contamination of drinking-water: Evidence from nationally representative household surveys in Ghana, Nepal and Bangladesh.

BACKGROUND: Current priorities for diarrhoeal disease prevention include use of sanitation and safe water. There have been few attempts to quantify the importance of animal faeces in drinking-water contamination, despite the presence of potentially water-borne zoonotic pathogens in animal faeces. OBJECTIVES: This study aimed to quantify the relationship between livestock ownership and point-of-consumption drinking-water contamination. METHODS: Data from nationally representative household surveys in Nepal, Bangladesh, and Ghana, each with associated water quality assessments, were used. Multinomial regression adjusting for confounders was applied to assess the relationship between livestock ownership and the level of drinking-water contamination with E. coli. RESULTS: Ownership of five or more large livestock (e.g. cattle) was significantly associated with drinking-water contamination in Ghana (RRR=7.9, 95% CI=1.6 to 38.9 for medium levels of contamination with 1-31cfu/100ml; RRR=5.2, 95% CI=1.1-24.5 for high levels of contamination with >31cfu/100ml) and Bangladesh (RRR=2.4, 95% CI=1.3-4.5 for medium levels of contamination; non-significant for high levels of contamination). Ownership of eight or more poultry (chickens, guinea fowl, ducks or turkeys) was associated with drinking-water contamination in Bangladesh (RRR=1.5, 95% CI=1.1-2.0 for medium levels of contamination, non-significant for high levels of contamination). CONCLUSIONS: These results suggest that livestock ownership is a significant risk factor for the contamination of drinking-water at the point of consumption. This indicates that addressing human sanitation without consideration of faecal contamination from livestock sources will not be sufficient to prevent drinking-water contamination.

Integration of population census and water point mapping data-A case study of Cambodia, Liberia and Tanzania.

Sustainable Development Goal (SDG) 6 has expanded the Millennium Development Goals' focus from improved drinking-water to safely managed water services. This expanded focus to include issues such as water quality requires richer monitoring data and potentially integration of datasets from different sources. Relevant data sets include water point mapping (WPM), the survey of boreholes, wells and other water points, census and household survey data. This study examined inconsistencies between population census and WPM datasets for Cambodia, Liberia and Tanzania, and identified potential barriers to integrating the two datasets to meet monitoring needs. Literatures on numbers of people served per water point were used to convert WPM data to population served by water source type per area and compared with census reports. For Cambodia and Tanzania, discrepancies with census data suggested incomplete WPM coverage. In Liberia, where the data sets were consistent, WPM-derived data on functionality, quantity and quality of drinking water were further combined with census area statistics to generate an enhanced drinking-water access measure for protected wells and springs. The process revealed barriers to integrating census and WPM data, including exclusion of water points not used for drinking by households, matching of census and WPM source types; temporal mismatches between data sources; data quality issues such as missing or implausible data values, and underlying assumptions about population served by different water point technologies. However, integration of these two data sets could be used to identify and rectify gaps in WPM coverage. If WPM databases become more complete and the above barriers are addressed, it could also be used to develop more realistic measures of household drinking-water access for monitoring.

A Global Perspective on Drinking-Water and Sanitation Classification: An Evaluation of Census Content.

Following the recent expiry of the United Nations' 2015 Millennium Development Goals (MDGs), new international development agenda covering 2030 water, sanitation and hygiene (WASH) targets have been proposed, which imply new demands on data sources for monitoring relevant progress. This study evaluates drinking-water and sanitation classification systems from national census questionnaire content, based upon the most recent international policy changes, to examine national population census's ability to capture drinking-water and sanitation availability, safety, accessibility, and sustainability. In total, 247 censuses from 83 low income and lower-middle income countries were assessed using a scoring system, intended to assess harmonised water supply and sanitation classification systems for each census relative to the typology needed to monitor the proposed post-2015 indicators of WASH targets. The results signal a lack of international harmonisation and standardisation in census categorisation systems, especially concerning safety, accessibility, and sustainability of services in current census content. This suggests further refinements and harmonisation of future census content may be necessary to reflect ambitions for post-2015 monitoring.

Association of Supply Type with Fecal Contamination of Source Water and Household Stored Drinking Water in Developing Countries: A Bivariate Meta-analysis.

BACKGROUND: Access to safe drinking water is essential for health. Monitoring access to drinking water focuses on water supply type at the source, but there is limited evidence on whether quality differences at the source persist in water stored in the household. OBJECTIVES: We assessed the extent of fecal contamination at the source and in household stored water (HSW) and explored the relationship between contamination at each sampling point and water supply type. METHODS: We performed a bivariate random-effects meta-analysis of 45 studies, identified through a systematic review, that reported either the proportion of samples free of fecal indicator bacteria and/or individual sample bacteria counts for source and HSW, disaggregated by supply type. RESULTS: Water quality deteriorated substantially between source and stored water. The mean percentage of contaminated samples (noncompliance) at the source was 46% (95% CI: 33, 60%), whereas mean noncompliance in HSW was 75% (95% CI: 64, 84%). Water supply type was significantly associated with noncompliance at the source (p < 0.001) and in HSW (p = 0.03). Source water (OR = 0.2; 95% CI: 0.1, 0.5) and HSW (OR = 0.3; 95% CI: 0.2, 0.8) from piped supplies had significantly lower odds of contamination compared with non-piped water, potentially due to residual chlorine. CONCLUSIONS: Piped water is less likely to be contaminated compared with other water supply types at both the source and in HSW. A focus on upgrading water services to piped supplies may help improve safety, including for those drinking stored water.

A cross-sectional ecological study of spatial scale and geographic inequality in access to drinking-water and sanitation.

INTRODUCTION: Measuring inequality in access to safe drinking-water and sanitation is proposed as a component of international monitoring following the expiry of the Millennium Development Goals. This study aims to evaluate the utility of census data in measuring geographic inequality in access to drinking-water and sanitation. METHODS: Spatially referenced census data were acquired for Colombia, South Africa, Egypt, and Uganda, whilst non-spatially referenced census data were acquired for Kenya. Four variants of the dissimilarity index were used to estimate geographic inequality in access to both services using large and small area units in each country through a cross-sectional, ecological study. RESULTS: Inequality was greatest for piped water in South Africa in 2001 (based on 53 areas (N) with a median population (MP) of 657,015; D = 0.5599) and lowest for access to an improved water source in Uganda in 2008 (N = 56; MP = 419,399; D = 0.2801). For sanitation, inequality was greatest for those lacking any facility in Kenya in 2009 (N = 158; MP = 216,992; D = 0.6981), and lowest for access to an improved facility in Uganda in 2002 (N = 56; MP = 341,954; D = 0.3403). Although dissimilarity index values were greater for smaller areal units, when study countries were ranked in terms of inequality, these ranks remained unaffected by the choice of large or small areal units. International comparability was limited due to definitional and temporal differences between censuses. CONCLUSIONS: This five-country study suggests that patterns of inequality for broad regional units do often reflect inequality in service access at a more local scale. This implies household surveys designed to estimate province-level service coverage can provide valuable insights into geographic inequality at lower levels. In comparison with household surveys, censuses facilitate inequality assessment at different spatial scales, but pose challenges in harmonising water and sanitation typologies across countries.

Modelling the incidence of Plasmodium vivax and Plasmodium falciparum malaria in Afghanistan 2006-2009.

BACKGROUND: Identifying areas that support high malaria risks and where populations lack access to health care is central to reducing the burden in Afghanistan. This study investigated the incidence of Plasmodium vivax and Plasmodium falciparum using routine data to help focus malaria interventions. METHODS: To estimate incidence, the study modelled utilisation of the public health sector using fever treatment data from the 2012 national Malaria Indicator Survey. A probabilistic measure of attendance was applied to population density metrics to define the proportion of the population within catchment of a public health facility. Malaria data were used in a Bayesian spatio-temporal conditional-autoregressive model with ecological or environmental covariates, to examine the spatial and temporal variation of incidence. FINDINGS: From the analysis of healthcare utilisation, over 80% of the population was within 2 hours' travel of the nearest public health facility, while 64.4% were within 30 minutes' travel. The mean incidence of P. vivax in 2009 was 5.4 (95% Crl 3.2-9.2) cases per 1000 population compared to 1.2 (95% Crl 0.4-2.9) cases per 1000 population for P. falciparum. P. vivax peaked in August while P. falciparum peaked in November. 32% of the estimated 30.5 million people lived in regions where annual incidence was at least 1 case per 1,000 population of P. vivax; 23.7% of the population lived in areas where annual P. falciparum case incidence was at least 1 per 1000. CONCLUSION: This study showed how routine data can be combined with household survey data to model malaria incidence. The incidence of both P. vivax and P. falciparum in Afghanistan remain low but the co-distribution of both parasites and the lag in their peak season provides challenges to malaria control in Afghanistan. Future improved case definition to determine levels of imported risks may be useful for the elimination ambitions in Afghanistan.

Estimation of malaria incidence in northern Namibia in 2009 using Bayesian conditional-autoregressive spatial-temporal models.

As malaria transmission declines, it becomes increasingly important to monitor changes in malaria incidence rather than prevalence. Here, a spatio-temporal model was used to identify constituencies with high malaria incidence to guide malaria control. Malaria cases were assembled across all age groups along with several environmental covariates. A Bayesian conditional-autoregressive model was used to model the spatial and temporal variation of incidence after adjusting for test positivity rates and health facility utilisation. Of the 144,744 malaria cases recorded in Namibia in 2009, 134,851 were suspected and 9893 were parasitologically confirmed. The mean annual incidence based on the Bayesian model predictions was 13 cases per 1000 population with the highest incidence predicted for constituencies bordering Angola and Zambia. The smoothed maps of incidence highlight trends in disease incidence. For Namibia, the 2009 maps provide a baseline for monitoring the targets of pre-elimination.

Accuracy of the H2S test: a systematic review of the influence of bacterial density and sample volume.

The presence/absence hydrogen sulphide test (P/A H2S) is widely used as a low-cost alternative faecal indicator test in remote and resource-poor settings. The aim of the paper is to assess how bacterial density and sample volume affect its accuracy. Based on a systematic search, we identified studies that tested water samples (n = 2,034) using both the P/A H2S test and recognised tests for thermotolerant coliforms (TTC) or Escherichia coli. We calculated P/A H2S test specificity and sensitivity against a range of TTC and E. coli densities. For two studies, we compared this with sensitivity and specificity estimates for simulated 100 and 20 ml presence/absence tests. For most of the 19 included studies, as the threshold used to define contamination increased from 1 to 100 cfu/100 ml, P/A H2S test sensitivity increased but specificity decreased. Similarly, the simulation indicated that increasing test volumes from 20 to 100 ml increased sensitivity but reduced specificity. There was potential for bias, for example from lack of blinding during test interpretation, in most of the studies reviewed. In assessing the P/A H2S test as an alternative to standard methods, careful consideration of likely indicator bacteria levels and sample volume is required.

A spatial analysis of pit latrine density and groundwater source contamination.

This study aims to assess the relationship between chemical and microbial contamination of groundwater sources and a range of potential hazards in two peri-urban areas of Kisumu, Kenya where shallow wells and pit latrines are widely used. From 1998 to 2004, 263 samples were taken from 61 groundwater sources and tested for thermotolerant coliforms. Eighteen of these sources were also tested for chemical contaminants, including nitrate, chloride and fluoride. The locations of all water sources, buildings and pit latrines in the study area were surveyed. Local pit latrine densities were calculated using a geographic information system. Ten out 18 samples were above the World Health Organization guideline values for nitrate, 236 out of 263 were positive for thermotolerant coliforms, and all were above the guideline values for fluoride. There was neither a relationship between thermotolerant coliform levels and daily rainfall patterns nor with sanitary risk inspection scores for samples from shallow wells (r = 0.01, p = 0.91, n = 191). The density of pit latrines within a 100-m radius was significantly correlated with nitrate and chloride levels (r = 0.64, p = 0.004 and r = 0.46, p = 0.05, respectively) but not with thermotolerant coliforms (r = 0.22, p = 0.11). These results illustrate both the public health risks associated with shallow groundwater sources, on-site sanitation and high population density. These findings have implications for current policies that promote latrine construction, especially in peri-urban areas of high population density. More comprehensive studies of larger communities should be commissioned to extend this analysis of the links between latrine density and groundwater contamination and so identify the contingent policy risks.

Public perception of drinking water safety in South Africa 2002-2009: a repeated cross-sectional study.

BACKGROUND: In low and middle income countries, public perceptions of drinking water safety are relevant to promotion of household water treatment and to household choices over drinking water sources. However, most studies of this topic have been cross-sectional and not considered temporal variation in drinking water safety perceptions. The objective of this study is to explore trends in perceived drinking water safety in South Africa and its association with disease outbreaks, water supply and household characteristics. METHODS: This repeated cross-sectional study draws on General Household Surveys from 2002-2009, a series of annual nationally representative surveys of South African households, which include a question about perceived drinking water safety. Trends in responses to this question were examined from 2002-2009 in relation to reported cholera cases. The relationship between perceived drinking water safety and organoleptic qualities of drinking water, supply characteristics, and socio-economic and demographic household characteristics was explored in 2002 and 2008 using hierarchical stepwise logistic regression. RESULTS: The results suggest that perceived drinking water safety has remained relatively stable over time in South Africa, once the expansion of improved supplies is controlled for. A large cholera outbreak in 2000-02 had no apparent effect on public perception of drinking water safety in 2002. Perceived drinking water safety is primarily related to water taste, odour, and clarity rather than socio-economic or demographic characteristics. CONCLUSION: This suggests that household perceptions of drinking water safety in South Africa follow similar patterns to those observed in studies in developed countries. The stability over time in public perception of drinking water safety is particularly surprising, given the large cholera outbreak that took place at the start of this period.