Water quality at the source and incidence of water borne diseases in rural households of South India

INTRODUCTION

Access to safe drinking water has long been a central aim of public health and international development policy and water has a profound effect on human health both as a means to reduce disease and as a medium through which disease-causing agents may be transmitted.^1,2 The provision of water was one of the eight essential components of primary health care at Alma-Ata, in 1978.¹ Millennium Development Goals (MDG), declared to halving the proportion of people without sustainable access to safe drinking-water and basic sanitation, by 2015.^1,3 The proposed Sustainable Development goals (SDG) declared in goal 6 to achieve universal and equitable access to safe and affordable drinking-water for all by 2030.⁴

The World Health Organization (WHO) Joint Monitoring Committee (JMC) classified household water sources as improved and unimproved.^2,5 “Use of an improved source” was adopted as an indicator for monitoring access to safe drinking water, presuming that it will not risk health.⁶ Providing disinfected piped water, to each household is the best solution to waterborne disease, which is not available in most of the developing world due to resource constraints.⁷

A systematic review of studies assessing water source contamination published between January 1990 and August 2013 estimated that 1.8 billion people globally used a source of drinking water which suffered from faecal contamination and it was more in rural (41%, CI: 31%–51%) than in urban areas (12%,CI: 8–18).⁶ A report quoting the WHO stated that more people would die from consuming unsafe drinking water and unsanitary conditions by the year 2020 than from AIDS.⁸ So water quality control is critical in reducing the potential for explosive epidemics.^7,9

In India, as per the latest statistics, 85% of the rural households have access to drinking water within or near to their premises and water borne diseases (WBD) account for 10% of the total burden of diseases and claim about 5 million lives.^8,10,11 It is well recognized that failure to protect water sources and inadequate water treatment are the primary reasons for drinking water contamination with bacteria.¹² Two different studies from North India found that 32.1% and 40% samples of water were bacteriologically contaminated, and it was higher in the rainy seasons.^8,13 In the South Indian state of Kerala majority of the population (62%) were using dug well as the drinking water source and its contamination was positively correlated with density of houses and average rain fall.^14-17 Recently along with quality, quantity of water also got equal or greater importance with attributable risk reduction of 39% and 23% respectively.^2,18 Studies from Ecuador and Pakistan reported no association between drinking water quality and the incidence of diarrhea.^18,19 Most of the studies on WBD were cross sectional and reported only point prevalence, with recall bias which tends to under estimate and is not adequate to explain the relation.^8,13,16,17

In this context with the following objectives a longitudinal prospective study was conducted in Kerala, a state of India.

Primary objectives were to assess the incidence of water born disease and to determine the water quality at the source in the study area. Secondary objective was to study the relation between incidence of WBD with water quality at source.

METHODS

The longitudinal study was conducted in South Indian state, Kerala (Kozhikode district) extending for a period of twelve months from July 2015 to August 2016. There are 65.95 lakhs households and 50 lakhs wells enumerated with density ranging from 120-150 wells per square kilometer.^15,20

Study area

Kozhikode district is located in the western coast of Kerala situated between North latitudes 11° 08' and 11° 50' and East longitudes 75° 30' and 76° 08'. According to the 2011 census, Kozhikode district has a population of 3,089,543, population density of 1,318 per square kilometer and 23,445 square kilometer area. Topographically it is divided into sandy coastal area, laterite midland area and rocky high land area. According to the topography (coastal, midland, hilly) three rural areas were selected Chemenchery, Mavoor, Puduppady.

Sample size

According to the previous study the reported household prevalence of WBD was 23%.²⁰ At 95% CI and an error of 20%, the minimum sample size calculated was 300.

Selections of households were done by multistage method. From the selected areas, 3 revenue wards were selected randomly and from each ward, the house list was collected. The first house was selected by simple random method and the consecutive houses were selected till adequate sample size of 100 in each area was attained.

The study protocol was approved by the IEC Medical College, Calicut. Data collection and drawing water samples were done after getting voluntary written informed consent from the head of the household

Data collection

The household data collection was done by 6 adequately trained women health workers (1 per 50 houses) by weekly house visits using a pre tested structured proforma prospectively for 12 months. Data on socio demographic details, housing, environment, sources of drinking water, WBD morbidity and health/sickness behavior were collected. Sanitary inspection was done using a standard check list.

Definition

Water-borne diseases are diseases caused by the ingestion of water contaminated by human or animal faeces or urine containing pathogens.¹⁸

Diarrhea (ADD) was defined as three or more loose stools during a 24-hour period. A diarrhea episode was marked as a new episode if the person had two or more days without diarrhea. For other diseases the diagnosis was cross checked with medical records.

Water quality sampling: By sub sampling using systematic random method, 10 houses were selected in each ward (10X3=30) and from 30 drinking water sources, water samples were collected and analysis was done thrice corresponding to the seasons (July, December, May) from Centre for Water Research Development Management (CWRDM lab) Calicut using standard Technique.

The samples drawn aseptically were transferred to 10 ml sterile bottles and properly labelled. Two samples were collected from each well after which they were transported to the laboratory in ice cold condition and subjected to bacteriological analysis on the same day. The Most Probable Number (MPN) test was used to enumerate the total number of bacteria in the well water samples. All the media used in this study were prepared and sterilized according to manufacturer’s instructions. The bacteria determined included Total Coliform Count (TCC), Faecal Coliform Count (FCC) or Thermo Tolerant coliforms (TTC) and E. coli. Total coliform and fecal coliform counts were estimated and recorded in Most Probable Number units per 100 ml (MPN/100ml). The pH and Total Dissolved Solids (TDS) were estimated using standard methods. The criteria given by WHO and Bureau of Indian Standards (BIS) for drinking water quality were used as acceptable limits.^1,9,10

The meteorological data like rain fall, temperature, humidity were collected from secondary sources for 12 months.

Statistical analysis

The data analysis was performed using Statistical package of social sciences (SPSS) version 16. According to the type of data, association or correlation was tested, either by chi-square/Pearson and spearman correlation coefficient. The relation of WBD with different qualitative and quantitative variables were analyzed. The morbidity was measured as incidence rates.

RESULTS

From 300 households, total 1459 persons were enrolled as study subjects with an average family size of 4.9 members. The mean age was 30.4±20 years. The sex wise and religion wise distribution are given in Table 1.

The characteristics of water sources and hygiene practices are given in Table 2.

All the houses had water seal toilets. In 85.6% of the houses water was available around the year and in 93.45% electric motor was connected with well. The mean distance from house to water sources was 7.7±12.2 meters.

The mean distance between well and septic tank was 13.8±3.2 meters.

The seasonal wise water quality parameter results – pH, Total dissolved solids, FCC/TTC and E. coli are given in Table 3.

During the 12 month period, 605 episodes of morbidities were reported among the cohorts of which 72 were water borne disease with an incidence of 49/1000 person years. The proportional morbidity due to WBD was 11.9%. The WBD reported were ADD, dysentery and hepatitis A. The details are given in table 4. Twenty-three episodes of WBD (34%) were among children in the age group 0-5 years (n=130, consisted 8.9% of population) with an incidence rate of 177/1000 person years.

The monthly incidence, seasonal pattern of WBD and rain fall are given in Figure 1.

The incidence of WBD had no correlation or relation with E. coli and Fecal coli count in all the seasons (p≥0.05) though there is a negative correlation with distance from septic tank (r =-0.118, p=0.53).

The distance from septic tank to the water source has got significant correlation with E. coli (r=-0.37, p=0.02) and Fecal coli (r=-0.43, p=0.04) during summer season.

DISCUSSION

We conducted a one year longitudinal study in a selected rural area of south India among 300 houses holds with 1459 individuals by directly collecting data by weekly house visits and the water samples from the point sources were analyzed three times according to the seasonality. Among the households 93.3% had dug well as the water source and majority (83.3%) were within the premises and protected. At the state level, 62% sources are dug well. As per the WHO- JMP 2011 criteria it can be included as improved source of water and presumed to be safer.^1,2,9 Though the department of Health and family welfare mandated monthly disinfection of wells by chlorination using bleaching powder by health workers, majority were not practicing it. At the point of use, 95% practiced the physical method - boiling of water as treatment. For storage of water in kitchen >70% were using metallic vessels with lids. This may reduce contamination at storage and consumption points in these households (Table 2).

All the households had sanitary toilets of water seal type and zero open defecation. Due to constraints in land availability the mean distance between septic tank and water source was less (13.8±3.2 meters) than as prescribed minimum distance 15 meters by WHO.

During the 12 month period of follow up the incidence of WBD was 49/1000 which was less than previously reported from the state as 84 and 78/1000 per year and showed a declining trend.^20,21 The proportional morbidity due to WBD was 11.9% which was similar to the previous report from the country.¹¹ As given in Table 4, the WBDs mainly reported were ADD (2/3rd), Dysentery and hepatitis (Table 4). As a development paradox, being in the late phase of epidemiological transition, the state now experiences out breaks of hepatitis.²¹ As reported elsewhere compared to other age groups, the children in 0 to 5 years age have high morbidity due to WBD which was 177/1000 with a relative risk of 3.61 (95% CI 2.6- 4.7).

This area receives Southwest and Northeast monsoons during June to August and October to November respectively, where the former consists >80% of the rain. The total rainfall during the year was 3285 mm. Though there were no peaks of seasonality trends, increased number of cases were reported during summer and rainy months, possibly explained by microbial concentration and contamination of water (Figure 1).

As per the water quality parameters given by WHO and BIS, the pH and TDS values of our samples in all the seasons were within the permissible values 6.5 and 500 mg/litre (Table 3).^1,10

Indicator bacteria E. coli was found in upto 30% water sources in summer and winter samples and in more than 60% in rainy samples. Correspondingly Feacal coliform count was >10 MPN/100 ml in all the seasons in more than 60% water sources and the count was markedly increased during monsoon samples.

Cross sectional studies from north Indian states previously reported these contamination rates as 32% and 40%.^8,13 The presence of coliform and E. coli in water is an indication of feacal contamination and has been associated with waterborne diseases among the consumers. But corresponding to the indicators no such out breaks were reported spatially or temporarily from the area except few sporadic incidences.

Our study identified significant correlation between the bacterial burden indicators and proximity of source of water to septic tank in summer (r=0.43, p=0.02 and r=0.37, p=0.04). A study from the district reported presence of Feacal coliforms / TTC in 68% of samples with positive correlation (r=0.47, p=0.01) with distance between the septic tank and well.²²

Both microbial indicators have got significant correlation in all the samples (p=0.01) which point out their common source of origin or pathway of contamination. The increase of bacterial indicators during rainy season in our study especially in areas with sandy soil compared to laterite point out the probability of percolation of bacteria through soil layers. The density of houses, dug wells, proximity of septic tanks which cause pressure on lands along with the climatic conditions may be other environmental factors.

In our study the incidence of WBD had no relation with E. coli and Fecal coliform count in source water in all the seasons (p≥0.05) which point out the weak relation of household source water quality with WBDs. A similar prospective study in the southern Punjab and Pakistan where drinking water quality was monitored weekly among children younger than 5 years in 200 households, found no association between the incidence of childhood diarrhea and the number of E. coli in the drinking water sources.²⁰ Another study which uses chlorination as proxy indicator also reported that it has no relation with WBD.⁶

Since the literacy rate in the area was 99%, the low outcome may be attributed to the increased hygienic practices among the members like storage, boiling before consumption along with hand washing which prevent further contamination with pathogenic microbes. Meta-analysis reported that hand washing with soap at critical points (before eating, after defecating and before handling food), improved sanitation and point of use water treatment are three most effective interventions which reduce the risks by 37%, 34% and 29%.²³ Other possibilities are rather than household sources people are acquiring more infections from outside sources (work place, school etc). Supporting this, study from the state reported that those children eating out side and not washing hands have higher risk of WBD (OR=1.6, OR=2.3).²⁴ Nutritional status, immunologic status, and genetic factors of a person also play a large role in determining disease outcome. First, not all diarrhea disease pathogens are transmitted exclusively via water. They can also be transmitted by food, fomites, personal contact, and in some cases via droplets.¹⁹ From the neighboring country Myanmar, a study reported that at point of use 94% of the water samples contain feacal coliforms and the prevalence of ADD among children were 156/1000, less than our study.²⁵ As reported by many studies from developing countries our study also found that the relation between water quality at the source and WBD was lower than sanitation and treatment at point of use.^5,6,7,19

WHO criteria for safe water accessibility include improved water, equated it as safe.^1,19 There is substantial evidence to demonstrate that improved sources of drinking water can contain E. coli. In a systematic review of microbial drinking water quality, many improved sources including piped water were found to be contaminated with E. coli.^6,24 This has implications in the use of international targets for safe drinking-water access. The selection of indicators and parameters for water quality assessment and analysis should be country and possibly region specific and may also be specific to certain sources of water.^1,2 Guidelines for Drinking-Water Quality recommend that faecal indicator bacteria, preferably E. coli or alternatively thermo tolerant coliform (TTC), should not be detectable in any 100 ml drinking water sample.^1,6,9 In tropical regions, where diarrheal disease poses the biggest health burden, the indicator organisms have been found to naturally occur, bringing into question their utility as indicators of fecal contamination.^6,19 In tropical climate, the proportion of E coli is less in feacal colifoms; therefore they are not useful as an indicator of faecal pathogens and no public health significance.^1,18 It has been argued that the WHO Guideline value of zero faecal coliforms is too restrictive for untreated water sources in rural communities as it is not feasible in practice. Therefore, it is of primary importance that water users are guaranteed a minimum level of achievable standards.^5,18 Our data suggest that such a restrictive standard also lacks adequate justification on epidemiological grounds.

The majority of the previous studies were cross-sectional and do not provide information on temporal variability in water quality and the outcome. Since our study was longitudinal, seasonal temporal effects on variability of WBD, water quality was studied as well as recalls bias was reduced.

We have noticed few limitations for our study. Since the WBD data was collected weekly from households minor episodes from other members of the family may not be passed to the informant so there may be chances of under reporting. Due to resource constraints we did the water quality analysis by taking sub samples and could not analyze the water samples at weekly intervals.

CONCLUSION

Our study found that though majority of people are using protected dug well, as per water quality indicators, the contamination was very high which has got no association with water borne disease. We believe this finding warrants extensive consideration of the use of indicators in studies of waterborne disease. Further studies are necessary to create more precise ways of studying the role of water in the transmission of WBD.

ACKNOWLEDGEMENTS

This study was funded State Health System Resource Centre, Department of health and Family welfare, Govt Of Kerala, India. The authors express thanks to the Staff PHCs of the study areas. Thanks to all staffs at water quality division, CWRDM, Kozhikode, for water quality analysis.

Funding: State Health System Resource Centre, Department of Health and Family Welfare, Government of Kerala.

Conflict of interest: None declared

Ethical approval: The study was approved by the Institutional Ethics Committee

REFERENCES

1. 1. World Health Organization. Rapid assessment of drinking-water quality: a handbook for implementation. World Health Organization. 2012. Available at: https://apps.who.int/iris/handle/10665/331485. Accessed on 10 June 2021.
2. 2. UNICEF Handbook on Water Quality. United Nations Children's Fund (UNICEF), New York. 2008;160. Available at: https://www.unicef.org/documents/2008-unicef-handbook-water-qualityhttps://jalshakti-. Accessed on 21 May 2021
3. 3. Bain RE, Gundry SW, Wright JA, Yang H, Pedley S, Bartram JK. Accounting for water quality in monitoring access to safe drinking-water as part of the Millennium Development Goals: lessons from five countries. Bulletin of the World Health Organization. 2012;90:228-35.
4. 4. Payden. Towards sustainable access to safe drinking water in South-East Asia. WHO South-East Asia Journal of Public Health. 2015;4:1-2.
5. 5. Bain R, Cronk R, Wright J, Yang H, Slaymaker T, Bartram J. Fecal contamination of drinking-water in low-and middle-income countries: a systematic review and meta-analysis. PLoS Med. 2014;11(5):e1001644.
6. 6. Bain R, Cronk R, Hossain R, Bonjour S, Onda K, Wright J et al. Global assessment of exposure to faecal contamination through drinking water based on a systematic review. Tropical Medicine & International Health. 2014;19(8):917-27.
7. 7. Zin T, Mudin KD, Myint T, Naing DK, Sein T. Influencing factors for household water quality improvement in reducing diarrhoea in resource-limited areas. WHO South-East Asia journal of public health. 2013;2(1):6-11.
8. 8. Malhotra S, Sidhu SK, Devi P. Assessment of bacteriological quality of drinking water from various sources in Amritsar district of northern India. The Journal of Infection in Developing Countries. 2015;9(08):844-8.
9. 9. Kumar M, Puri A. A review of permissible limits of drinking water. Indian journal of occupational and environmental medicine. 2012;16(1):40.
10. 10. Ministry of Drinking Water and Sanitation (MDWS). Uniform Drinking Water Quality Monitoring Protocol. Available at: ddws.gov.in/sites/default/files/UniformDrinkingWaterQualityMonitoringProtocol.pdf. Accessed on 10 May 2021.
11. 11. Report. Study on Disease Burden Due to Inadequate Water & Sanitation Facilities in India. Sulabh International Academy of Environmental Sanitation Palam, New Delhi. 2007.
12. 12. Ye B, Yang L, Li Y, Wang W, Li H. Water sources and their protection from the impact of microbial contamination in rural areas of Beijing, China. International journal of environmental research and public health. 2013;10(3):879-91.
13. 13. Taneja N, Singh M, Rao P, Biswal M, Priya S, Chander R, Sharma M. Fecal contamination of drinking water supplies in and around Chandigarh and correlation with acute gastroenteritis. The Journal of communicable diseases. 2011;43(3):193-9.
14. 14. State Environment Report. Government of Kerala. India. 2007. Available at: http://moef.gov.in/wp-content/uploads/2018/04/Kerala_SoE2007-V1_0.pdf. Accessed on 10 May 2021.
15. 15. Jayakrishnan Thayyil. Status of Drinking water sources and water borne disease in Kerala. Science and Technology in Transforming Women's Lives. 2013;1. 2-6.
16. 16. Rejith PG, Jeeva SP, Vijith H, Sowmya M, Hatha AM. Determination of groundwater quality index of a highland village of Kerala (India) using geographical information system. Journal of Environmental Health. 2009;71(10):51-61.
17. 17. Anjali, Rajeena, Harikumar. Assessing ground water vulnerability and risk from on –site sanitation at Calicut city. Kerala Environment Congress Proceedings. Center for Environment and Development. 2008;135-9.
18. 18. Jensen PK, Jayasinghe G, van der Hoek W, Cairncross S, Dalsgaard A. Is there an association between bacteriological drinking water quality and childhood diarrhoea in developing countries? Tropical medicine & international health. 2004;9(11):1210-5.
19. 19. Levy K, Nelson KL, Hubbard A, Eisenberg JN. Rethinking indicators of microbial drinking water quality for health studies in tropical developing countries: case study in northern coastal Ecuador. The American journal of tropical medicine and hygiene. 2012;86(3):499-507.
20. 20. Jayakrishnan T, Thomas B. Status of Household drinking water source, water use and it’s correlation with Water borne disease in rural area of Kerala. Proceedings Kerala Environment Congress. 2009;216-23.
21. 21. Jayakrishnan T, Jeeja MC. Disease Burden of Kerala. Research Report. Thrissur: India. Society for Social Health Action and Research. 2007;27.
22. 22. Megha PU, Kavya P, Murugan S, Harikumar PS. Sanitation mapping of groundwater contamination in a rural village of India. Journal of Environmental Protection. 2015;6(01):34.
23. 23. Satapathy BK. Safe drinking water in slums: from water coverage to water quality. Economic and Political Weekly. 2014;50-5.
24. 24. Thankappan KR. Diarrhoea morbidity among under-five children: A comparative study of two villages. Centre for Development Studies. 2002.
25. 25. Myint SL, Myint T, Aung WW, Wai KT. Prevalence of household drinkingwater contamination and of acute diarrhoeal illness in a periurban community in Myanmar. WHO South-East Asia journal of public health. 2015;4(1):62-8.