Home-based drinking water purification through sunlight : from promotion the health effectiveness

Diarrhoeal diseases constitute a significant illness burden for children living in lowincome
countries. Children under the age of five years suffer about four billion diarrhoea
episodes per year, 90% of which occur in developing countries. Diarrhoeal illness
accounts for more than four percent of the “disability adjusted life years" lost to the
worldwide illness burden. ‘Unsafe water, sanitation and hygiene’ are the main global risk
factors for diarrhoeal diseases.
Current strategies for providing safe water to more than one billion people are being
reshaped. Since the formulation of the seventh Millennium Development Goal (MDG),
the focus is on large-scale and sustainable approaches. Target 10 of the MDGs includes
halving the number of people without access to safe water and sanitation facilities by
2015. Solar water disinfection (SODIS) is a home-based – or ‘point-of-use’ – water
purification and safe storage method that could support the achievement of this goal, by
providing safe drinking water to populations in need. The method consists of exposing
water-filled, transparent PET bottles to full sunlight for about one day.
Our main objective of this project was to measure the effectiveness of solar water
disinfection on the health of children under the age of five years. Based on a Latin
American dissemination programme and further pilot studies in Bolivia (2001) and
Bangladesh (1999/2000) on need assessment approaches in the domain of household
water management, we decided to carry out the research in 10 rural Bolivian communities
situated in the district of Mizque (Department of Cochabamba) from 2001 until 2003.
A case-control study was embedded in a morbidity surveillance scheme and
complemented with cross-sectional surveys, in order to comprehensively describe the
outcome: the impact of the SODIS method on the frequency of childhood diarrhoea. We
developed three SODIS promotion strategies that used various communication channels
to reach the target population: i) monthly community-based workshops; ii) monthly
household visits; and iii) a school campaign in 11 school centres of the district. During
four months of weekly diarrhoea monitoring we interviewed 100 cases, as well as 171
controls; the latter were randomly selected from the entire population. One-hour semistructured
interviews with mothers of selected study children were carried out to assess
relevant risk factors for child diarrhoea.
We then compared the diarrhoea incidence rates in study children between families who
applied the SODIS method with different intensities, and adjusted the results for major
known confounding factors (e.g. age, sex, hand washing). In addition, the quality of
household drinking water was analysed repeatedly, which enabled us to calculate the
efficacy of the SODIS method under natural conditions representing daily Bolivian life in
rural areas. We also repeatedly examined stool samples from community children for the
presence of protozoa and helminths, to identify the main transmission pathways of these
potentially diarrhoea-causing parasites.
The implementation of the SODIS method was challenging, as the target population did
not immediately recognise benefits of using the new method. About 20% drank SODIS
purified water on more than five days per week, and about 40% consumed the SODIS
water less frequently. The individual promotion strategies affected the population in
different ways: e.g. household visits increased adoption of the SODIS method and the
school campaign enhanced awareness about germs and diseases. Such extensive
promotion strategies may not be suitable for large-scale promotion of the SODIS method.
The assessment of motivational messages directed towards tangible benefits for the
population may prove essential to increase compliance.
Once the population was introduced to the SODIS method, we assessed its effect on the
study childrens’ health. The use of the SODIS method averted up to 75% of the diarrhoea
episodes in a rural Bolivian child under five years of age. The impact was less, when
families used the method less frequently (60%), indicating a dose-response relationship
between the intensity of use (or compliance to the SODIS method) and reduction in
diarrhoeal illness. In order to translate individual risk reductions to an impact measure at
population level, we employed population-attributable fraction estimates using an uptake
rate of the SODIS method of 20% in the community. Those calculations indicated that
15% of all child diarrhoeal illnesses in the population could be averted if the SODIS
method would be consistently used.
The high diarrhoea incidences in children under the age of five (about 6 episodes per
year) were not reflected in the infection data. It may therefore be assumed that bacteria
and viruses (which we did not investigate in this study) caused most diarrhoea episodes in
this setting.
We found that children would be re-infected rapidly after treatment for protozoa or
helminth infection – 50% of the children were re-infected in the elapse of two month,
mainly through Giardia lamblia and Entamoeba hist/disp. The age of the child, socioeconomic
status and hygiene indicators in the household were main risk factors for reinfection.
Water-borne transmission of these protozoa was not dominant in this setting.
The high efficacy of the SODIS method in producing pathogen-free drinking water in the
field is the foundation for a high effectiveness on people’s health. SODIS-purified
drinking water contained 90% less faecal coliform contamination than untreated drinking
water. Families that left their water exposed for at least two days, achieved a significantly
better purification effect, almost reaching the WHO recommendations of zero-tolerance of
indicator bacteria in drinking water. These results support our findings of a high
effectiveness of the SODIS method on child health due to the purification of their
drinking water. Increased purification efficacy of SODIS due to prolonged exposure time
further indicates that SODIS user instructions could be further revised to guarantee best
efficacy under field conditions.
In rural Bangladesh, people were forced to switch from arsenic- to microbiologically
contaminated drinking water sources. Diarrhoea rates were similar between intervention
families that switched water sources, with people drinking groundwater, indicating that
the SODIS method was efficacious enough to maintain water quality. We also learnt
important aspects on the use (e.g. bottle scarcity in rural areas) and determinants for
uptake and possible sustainable use of the method (e.g. acceptable alternative water
source). Specific community selection criteria, including normative and perceived needs,
were formulated for a subsequent need assessment in the Bolivian setting.
Since the start of our activities, we were challenged with developing and validating
indicators to classify families according to their use of the SODIS method, as no standards
existed. In Bolivia, the combination of three indicators for the uptake of SODIS may best
estimate the use of the method in families during a one-time evaluation visit: reported use
(sensitivity: 73%), observed use (specificity: 82%) and frequency of drinking SODIS
water in the last week (positive predictive value: 85%). The indicators can be measured
rapidly and easily through especially appointed staff during programme evaluation.
The most precise indicator is the repeated observation of SODIS purified water at the
home during unannounced visits.
The finding of a significant impact of the SODIS method on child health is consistent
with our other findings of a dose-response relationship, high efficacy under field
conditions and the efficacious protection from diarrhoeal diseases in Bangladesh.
Future research should confirm these findings under a multitude of environmental,
geographical and cultural settings and study designs, to produce reliable evidence of the
methods’ effectiveness in improving the health of populations. The current research raised
issues regarding the implementation of and compliance of the population to the SODIS
method in combination with the possibility to guarantee its water-purification ability. This
also raised questions on the costs in relation to the benefits of the SODIS method from the
individual to the programme – and planners’ level; and the likelihood that the planner may
no longer perceive the method as an intermediate but rather as a permanent solution to
provide safe drinking water.
In the future, investigations should:
(i) define tangible benefits for target populations, that can later also be applied in
social marketing strategies for the broader promotion and higher acceptance of
the method in the population. In this context, user instructions should be
adapted to guarantee water quality during large-scale promotion activities,
(ii) assess the costs of the SODIS method at individual-, programme- and
planners’ level. This will allow decision making at regional level, and
comparison with other point-of-use methods at policy level,
(iii) consider in the planning process with the local people that the SODIS method
should not replace future permanent and durable solutions for drinking water
(“SODIS is only a valuable means to the end”).
This is the first research that evaluated various levels of the solar disinfection method
(efficacy, promotion, compliance, use, health effect) in different settings. Also, this study
estimated the effectiveness of the method on the health of young children at population
level applying an innovative population-based approach.
It demonstrated with confidence that the method is efficacious in reducing the diarrhoea
burden in a child population. Due to its simplicity and almost ubiquitous applicability,
solar disinfection is applicable in various settings, but the long-term use of the method
also depends on the political will and the availability of subsidies (e.g. for motivational
campaigns, or bottle provision). This project, with its multiplicity of findings served to
inform and support a randomised control trial on the effectiveness of solar water
disinfection in a rural area of Bolivia, and current endeavours in the national SODIS
dissemination programme. At regional level, we mostly increased awareness about the
application of the SODIS method that we hope will stimulate regional development. The
ultimate decision-maker will always be the consumer and potential beneficiary.