Open defecation free: where do we need to focus ?

Background If not managed properly, human faeces can cause various kinds of communicable diseases, from viral and bacterial to protozoan. Hygienic and proper use of toilets is a step forward in managing those diseases. The burden of these faeco-oral diseases is high in low and middle–income countries (LMICs) like Nepal, where not having access to toilets is considered a major determinant of public health problems, including risk of diarrheal and other water borne diseases [1]. Toilets are a primary barrier to faeco-oral disease transmission and better sanitation practice is recognized as an important parameter for a healthy, dignified and developed society. The goal of achieving adequate and equitable sanitation has been agreed globally, and eliminating Open Defecation (OD) world wide by 2030 is included in the Sustainable Development Goal number six [2]. Despite this, 2.4 billion people around the world lack adequate Received: 1 December 2018

In the context of Nepal, the Constitution has declared access to safe water and sanitation a fundamental right of citizens [5]. The government of Nepal has set a target for universal access to improved sanitation by 2017, for better hygiene, health and environment [6]. Unfortunately, we still have houses, offices and schools without toilets or with poor utilization of toilets. A sanitation promotion program was started in Nepal in the early 1990s as an integral component of water supply projects and since then it has been a working area for governmental and non-governmental agencies. Till date some innovative ideas and concepts have been adopted for sanitation and hygiene, for instance, Open Defecation Free communities, Community Led Total Sanitation [7], school led total sanitation [8], basic sanitation package, school sanitation and hygiene education, national sanitation week [9], global handwash-ing day, world toilet day, Nepal water sanitation and hygiene [6] and much more. All of these have helped achieve the current status where 85 percent of household have toilets (as at the end of 2018) [10]. Households without toilets have various common factors, ranging from socio-culture norms [11] to economic status [12] of the people, creating further challenges for the effectiveness of these programs. Furthermore, despite huge government efforts, a lack of public awareness, human resource constraints, and a lack of proper planning act as obstacles to achieving total sanitation [6,13]. In order to meet government targets, there was national trend for declaring Open Defecation Free (ODF) areas, but some evidence shows inability of those areas to maintain their ODF status; for example, village development committees previously declared ODF were found to be unable to maintain minimum requirements of an ODF area afterwards [14]. Furthermore, although people have constructed toilets with the help of subsidy provided by the government, they do not always use them because of entrenched ethnic and cultural taboos [15], traditional beliefs [16] etc. For example, in some cultures there is a belief that an in-law and a daughter-inlaw cannot use the same toilet. There still are communities where menstruating women cannot use the toilet because of a belief that they are untouchable during their days of menstruation. Similarly, unmanaged urbanization resulting in tightly clustered settlements, lack of space for building toilets, lack of technical support, and poor behavioral attitudes are major challenges for ODF sustainability. The poor, disadvantaged and high-risk groups are outside of the sanitation mainstream, undermining equity, ownership and participation. This acts as a further obstacle to full toilet coverage. Though there has been a shift of approach from the conventional awareness raising approach towards a behavior change approach in pursuit of sustainable ODF, there still is lots of room for improvement, from the individual to the policy making level.

Methodology Study Area and Data collection
This paper uses data from the Nepal Demographic and Health Survey, 2016, a nationally representative sample survey. The primary objective of the 2016 NDHS is to provide up-to-date estimates of basic demographic and health indicators. The information collected through the 2016 NDHS is intended to assist policy makers and program managers in the Ministry of Health and other organizations in designing and evaluating programs and strategies for improving the health of the country's population. The study protocol was approved by the Nepal Health Research Council and the ICF Macro Institutional Review Board in Calverton, Maryland, USA. The sampling frame used for the 2016 NDHS is an updated version of the frame from the 2011 National Population and Housing Census (NPHC). The 2016 NDHS sample was stratified and selected in two stages in rural areas and three stages in urban areas. In rural areas, wards were selected as primary sampling units (PSU), and households were selected from the sample PSUs. In urban areas, wards were selected as PSUs, one enumeration area (EA) was selected from each PSU, and then households were selected from the sample EAs. A total of 11,040 household were sampled. Data was collected via interview, yielding a response rate of 99%. The NDHS report details the methodology used in the survey (MoH, 2017).

Study Variables
Dependent variables: toilet status is categorized into two different categories; '1' for having toilets and '0' for not having toilets. Independent variables: demographic, socio-economic, and geographical characteristics (As listed in Table 1)

Statistical Analysis
The weighted percentage was calculated. Association between dependent variable (toilet status-having and not having a toilet in the household) of households and independent variables (demographic, socio-economic, and geographical characteristics) was assessed via bivariate analysis using a Chi-square test. Then, a multivariate logistic regression model was used to assess significant predicators for not having a toilet in the household, after controlling for other variables. All the variables were included in the same model and analyzed. Multi-co-linearity between the variables was assessed before analyzing in logistic analysis. The acceptance level of co-linearity was below 0.7. The analysis found that two variables 'Development regions' and 'Province' were highly correlated (r=0.97). Therefore, 'Development regions' was removed from the logistic regression model. The Statistical Package for Social Science (SPSS 20.0 for Windows) software was used to analyze the data.

Results
Out of the total sampled households (11,040) 85% of them had toilet in their household ( Figure 1). Nearly fourth of the respondents belonged to province number 3, more than three-fifths (61%), were urban residents. Respondents were almost equally distributed among different wealth quintiles (Table 1).   Table 2 shows toilet distribution according to the household characteristics. Nearly half (49%) of households in province number 2 had no toilet facility. The proportion of households without toilet was signficantly higher in rural areas (21%) and Terai region (27%) as compared to their respective counterparts. Almost a quarter of respondents belonging to middle and poorer wealth index had no toilet facility in the household. (Table2).   Note *** Significant at p < 0.001; ** = p < 0.01 and * = p < 0.05

Discussion
This study assessed the factors responsible for not having toilets in the context of Nepal, where households that lack a toilet were measured on the basis of their geographical and socio-economic characteristics. Socio-economic and demographic factors are found to be associated with access to portable water and improved sanitation facilities [17]. Furthermore, findings from other studies support the findings of this study on income, household size and region, which are significant predictors for sanitation access [18]. Our findings show that households in the Terai region are less likely to have toilets, which is similar to the findings of the study conducted by the Center Bureau of Statistics [14]. The same study also found that the houses with poor economic status are less likely to have toilets compared to wealthier households. Thus, construction and use of toilets is highly associated with economic status(16). The literature shows that the construction of a toilet is linked with urbanization, which is similar in our study as well, where, the proportion of households having a toilet in an urban area is double to that in rural areas [19]. A study conducted in Myagdi district, three years after it was declared ODF, shows that almost all households had a storage of water facility (98.8%) and hand washing basin near the latrine (99.2%) [15]. Whereas, our study shows that only four out of five households have a fixed place for handwashing. Studies show that the unavailability of water facilities at toilets motivates people for open defecation and disuse of toilets [20]. Similarly, use of unprotected water for drinking is considered a major factor for diarrheal disease (21). Furthermore, mass media like television, radio, print media, internet, etc. play a significant role in spreading information and raising awareness on sanitation issues [13]. They can help trigger positive changes in public opinion and behavior on matters of public health concern. However, this is in contradicts with our findings, where, households having a radio are less like to have toilets in their houses. This study has a number of strengths and some limitations in the interpretation of the results. We used a nationally representative data set that was based on a validated questionnaire and methodology. The findings can be generalized to the whole country as this survey was nationally representative. We still have some limitations that need to be taken into account when interpreting our findings from this study. Because of the survey's cross-sectional design, all of the factors analyzed in the study were measured at a single point of time. Thus, the analysis can only provide evidence of statistical association between those items and the lack of a toilet in the household at that time; it cannot show a cause-effect relationship. However, the findings of this study will be helpful for program implementers and policy makers in suggesting 'what' needs to be focused 'where' , in order to develop effective ODF related programs. Furthermore, our study reflected that socio-economic status of the people is associated with the availability of toilets in their house. However, having a toilet in a house does not mean that people are using toilets, as there are cultural, political, and behavioral factors affecting not just the availability but also their utilization.

Conclusion
Despite some real achievements and progress in trying to achieve ODF status in Nepal, there are still residences without access to adequate sanitation and toilet facilities. Province number two and the Terai region in particular need to be focused on for designing ODF programs to meet national and international sanitation goals.
Abbreviations aOR: Adjusted odds ratio BSP: Basic sanitation package CI: Confidence interval EA: Enumeration Area