- Email: [email protected]
The dynamics and determinants of household shared sanitation cleanliness in a heterogeneous urban settlement in Southwest Nigeria
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
Available online at www.sciencedirect.com
Public Health journal homepage: www.elsevier.com/puhe
Original Research
The dynamics and determinants of household shared sanitation cleanliness in a heterogeneous urban settlement in Southwest Nigeria O.O. Aluko a,*, E.O. Oloruntoba b, U.A. Chukwunenye a, E.U. Henry a, E. Ojogun a a b
Department of Community Health, College of Health Sciences, Obafemi Awolowo University, Ile Ife, Nigeria Department of Environmental Health Sciences, University of Ibadan, Nigeria
article info
abstract
Article history:
Objectives: The study assessed the behavioural contexts of household shared sanitation
Received 25 March 2018
(SS), the factors that could enhance its cleanliness and sustainable maintenance in the
Received in revised form
study area.
29 August 2018
Study design: The study design was cross-sectional in nature.
Accepted 13 September 2018
Methods: A pretested, interviewer-administered questionnaire with 113 questions provided
Available online 1 November 2018
information on knowledge, attitude and SS maintenance practices from 312 respondents selected through a four-stage sampling technique. Data were analysed by descriptive, chi-
Keywords:
square and logistic regression statistics with attitude and toilet cleanliness as the depen-
Shared sanitation
dent variables, while the sociodemographic characteristics, knowledge and practice,
Toilet cleanliness
including the toilet cleanliness index, were the explanatory variables at P < 0.05.
Behavioural factors
Results: The mean age of respondents was 46.1 ± 16.4 years. The majority were women (62.7%),
Toilet maintenance
tenants (64.7%) and had a secondary education (61.3%). In addition, 27.9% and 2.7% of respondents had shops and gender-segregated SS, respectively. Their knowledge and attitudes were good (66.3%) and positive (72.1%), while the majority (60.3%) of the SS was dirty. Almost two-thirds (62.8%) and 37.2% of respondents, respectively, had one and two or more compartments of unimproved pit toilets (31.6%) and water closets (28.6%), while 1.4% practiced open defaecation, despite having SS. More than two-fifths of SS were cleaned daily (45.4%) by women (52%), using water and soap (33.4%). Most households (93.4%) complied with the SS cleaning routines, with non-compliance attributed to lack of interest (29.5%), apathy when others defaulted (27.9%), lack of time (26.2%) and inadequate water (14.8%). There was a significant association between attitudes and education (P ¼ 0.025), wealth (P ¼ 0.011) and toiletcleaning frequency (P ¼ 0.015). The positive behaviour determinants of SS are factual knowledge (P ¼ 0.039), efforts required to clean (P ¼ 0.019), the ability to remember (P ¼ 0.011) and cleaning commitment (P ¼ 0.040), while freedom of expression among households using SS was a positive predictor of avoiding conflict, among the social dilemma factors. Conclusions: Good knowledge and positive attitudes were reported among users of SS, although with mixed practice and behaviour patterns. Factual knowledge, cleaning
* Corresponding author. E-mail addresses: [email protected] (O.O. Aluko), [email protected] (E.O. Oloruntoba), [email protected] (U.A. Chukwunenye), [email protected] (E.U. Henry), [email protected] (E. Ojogun). https://doi.org/10.1016/j.puhe.2018.09.013 0033-3506/© 2018 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.
126
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
commitments and freedom of expressions predict SS cleanliness, which should be reinforced using periodic messages and structured health promotion strategies. © 2018 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.
Introduction Improved sanitation is one that hygienically separates faeces from human contact1 and provides a safely managed service when used by a single household.2,3 Through Sustainable Development Goals (SDGs), the United Nations seeks equitable sanitation and hygiene (SH) for all and to end open defecation (OD) by 20301 and in Nigeria by 2025.4 The Millennium Development Goals' sanitation target focussed on ending OD5,6 without considering septage management when vaults are full. Improved sanitation offers a limited service when toilet facilities are shared by two or more households.1,7e9 The use of shared sanitation (SS) has increased to 437 million globally between 1990 and 201410 and 600 million people by 2017,1 mostly in Sub-Saharan Africa (SSA; 18%)5 and Eastern Asia (19%).10 The SDG sanitation goal sought the use of private, household-based, improved sanitation, although incremental access and SS are considered as a winewin strategy in many low-income and urban settlements.1,11 When adequately maintained, SS could prevent environmental contamination from sewage, septage or sludge.12 In Nigeria, sanitation is a private service, where programmes focus on the software support (awareness creation and training among primary beneficiaries to enhance the practice of good hygiene) aimed at guiding household decisions on preferences.13 These programmes could help 55.0% of households that use the most basic (33.0%) and limited (22.0%) toilets in Nigeria, while in Osun state, the Multiple Indicator Cluster Survey revealed that 61.8% of household toilets were improved and comprised nonshared (24.7%) and shared (37.1%) toilets.14 In assessing the SS contextual determinants, the risks, attitudes, norms, ability and self-regulation and social dilemma factors were applied.15,16 The risk index explored the perception of the health risks associated with SS, while the attitudes assessed SS cleanliness and maintenance stance.15,17 The descriptive and injunctive norms predict the social pressure of the SS maintenance; the ability factors assessed respondent's confidence in efforts to maintain SS and the belief in the competence to organise and implement cleaning actions. Self-regulation assessed the ability to organise SS maintenance, and the social dilemma context identified the prevailing interactions, rules and willingness to comply with households using SS.15,18
population of 167,254.19 Ile-Ife is approximately 70 km from Ibadan (Oyo State) and one of the rapidly urbanising settlements in Southwest Nigeria. The LGA experiences a tropical climate with annual rainfall of more than 150 cm and a temperature of between 21 C and 38 C. The city lacks a developmental master plan, and its growth does not follow a structured pattern. The living arrangements are such that most buildings have several households using SS.
The study settings and design The study used the cross-sectional design among households with SS.
Target and study population The target population was households living in buildings accommodating multiple households having SS, while the study population was households in selected houses using a SS toilet. The sampling unit was the household, consisting of one or more people who live in the same house and share meals.3
Sample size The minimum sample of 312, including 10% attrition, was determined using Fisher's formula20 with 24% prevalence of households using SS in Nigeria.21
Sampling technique A four-stage sampling method was used to recruit 312 respondents from eight randomly selected streets by equal probability sampling method. Stage one stratified wards into urban (seven) and rural (four), with the four urban wards selected by simple random sampling (SRS). Stage two selected two streets from each ward by SRS, and stage three selected eligible houses by sampling the fifth odd-numbered houses from the first residential building on each street by systematic technique, providing the house accommodated at least two households using the same toilet. In stage four, 39 households were studied from each of the eight randomly selected streets. However, only one household was selected from each building, represented by the household head, the spouse or the most senior adult member.
Data collection instrument
Methods The study area The study was conducted at Ife Central Local Government Area (LGA) [Fig. 1], which comprises 11 wards and has a
The semistructured, household questionnaire was adapted from Tumwebaze and Mosler.18 It has six sections: sociodemographic characteristics, knowledge, attitude, SS maintenance practices, SS maintenance behaviour determinants and the observational checklist (see online supplementary
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
127
Fig. 1 e Map of Nigeria showing the study area. LGA, Local Government Area.
material e questionnaire). The tool was pretested, revised, translated to Yoruba language and back-translated to English for content validity. The instrument was administered by face-to-face interviews and returned 0.6 Cronbach's a value, indicating an acceptable internal consistency score.
Dependent and explanatory variables The dependent variables were the composite attitude and toilet cleanliness recoded to binary attributes. The cleanliness of the SS was defined as clean, well maintained and safe to use now. In the linear regression models, the explanatory variables were sociodemographic characteristics, knowledge and practice, including the toilet cleanliness index.
Data management Data entry and analysis were performed with IBM SPSS, version 20, software. Continuous and categorical variables were presented by summary statistics and frequency (%), respectively. The relationship between categorical variables was assessed by Chi-square test to determine the type and magnitude of associations at P < 0.05. The linear regression models identified predictors of SS cleanliness. Through dimension reduction, the 21 household owned variables were
regressed through factor scores using the principal component analysis technique. The prime loaded factor regression scores were divided by percentile values function into three equal groups, with 0.5621013, 0.2816214 and 0.2816215 outputs and recoded into poor, average and rich wealth categories, respectively.22 The correct and incorrect responses of the knowledge and attitude variables were, respectively, scored ‘1’ and ‘0’. Using the guide provided by Sudeshna et al.,23 the median (interquartile range [IQR]) for knowledge and attitude indices was 6 and 5e6, respectively. The safe and poor practices in the toilet cleanliness index were scored ‘1’ and ‘0’, respectively, with its median of 10 (IQR 7e11). The values were used to dichotomise the three variables, respectively, into ‘good’ and ‘poor’, ‘positive’ and ‘negative’, and ‘clean’ and ‘dirty’ categories. For multivariate analysis, variables with P < 0.25 in bivariate statistics were imputed in the regression models for behaviour determinants of SS cleanliness.24
Ethical considerations The respondents consented and participated in the study in writing, and the study protocol was approved (IPHOAU/12/ 225) by the ethical committee at the Institute of Public Health, Obafemi Awolowo University in Nigeria.
128
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
b
Table 1 e Sociodemographic characteristics of respondents. Description of variables
Frequency (n)
Percentage (%)
Gender (n ¼ 311) Male 116 37.3 Female 195 62.7 House ownership (n ¼ 309) Owner 109 35.3 Tenant 200 64.7 Household size (n ¼ 312) 2 households 47 15.1 3e4 households 115 36.9 5 households 150 48.1 Relationship of respondents to living house (n ¼ 290) Family owned/occupied 18 6.2 Owner/occupied only 4 1.4 Landlordetenants 144 49.7 All tenants only 124 42.8 Age (n ¼ 308) 24 years 23 7.5 25e44 years 139 45.1 45e64 years 90 29.2 65 years 56 18.2 Length of residence in the house (n ¼ 302) 5 years 148 47.4 >5 years 164 52.6 Number of households living in the house of respondents (n ¼ 301) 2 households 36 12.0 3e4 households 115 38.2 5 households 150 49.8 Ethnicity (n ¼ 309) Yoruba 288 92.9 Igbo 15 4.8 Hausa/Fulani 6 1.9 Religion (n ¼ 311) Islam 90 28.9 Christian 211 67.8 10 3.2 Traditionala Marital status (n ¼ 310) Single 64 20.7 Married 206 66.5 Widowed 38 12.3 Other 02 0.6 Highest education (n ¼ 310) No formal education 25 8.1 Primary education 43 13.9 Secondary education 190 61.3 Tertiary education 47 15.2 Vocational education 4 1.3 1 0.3 Quranic educationb Main occupation (n ¼ 303) Self-employed 197 65.0 Paid employment 43 14.2 Farming 14 4.6 Full-time housewife 11 3.6 Pensioner/retired 10 3.3 Unemployed 21 6.9 Other 7 2.3 Wealth category (n ¼ 293)c Poor 97 33.1 Middle 99 33.8 Rich 97 33.1 a
Religious belief and worship of deities indigenous to the respondents through her/her cultural affiliation, belief and practice.
c
Quranic education is an informal education that happen around or in the mosque, that incorporates formal learning in the study of the Quran for those children that focus attention only on the study of the Quran against those exposed to formal, school based education. Through dimension reduction, the 21 household owned variables were regressed through factor scores using the principal component analysis technique. The prime loaded factor regression scores were divided by percentile values function into three equal groups, with 0.5621013, 0.2816214 and 0.2816215 outputs and recoded into poor, average and rich wealth categories, respectively.
Results Sociodemographic characteristics of respondents The respondents' mean age was 46.1 ± 16.4 years, and the majority were women (62.7%), tenants (64.7%) and married (66.5%). A large proportion of respondents had a secondary education (61.3%), had lived in the house for more than 5 years (52.6%) and were self-employed (76.0%). The model households per building were four, with close to half (48.1%) sharing the same toilet with more than five households. In addition, one-third (33.1%) of respondents were poor (Table 1).
Knowledge of SS and hygiene In Table 2, correct knowledge was exhibited by most (97.7%) respondents on handwashing before mouth contact, and 92.6% knew that toilets should be cleaned daily. Most respondents (91.2%) knew that dirty toilets could breed pathogens and that toilets should be located at a lower elevation in relation to safe water sources (77.5%). About two-thirds (66.3%) of respondents had a good knowledge of SS.
Attitudes to the cleanliness and maintenance of SS Most respondents (98.7%) believed that SS should be cleaned only when dirty, while 87.5% believed that each household should have a private toilet. Queuing was perceived as a good practice in using SS (67.9%), and its maintenance was considered as a shared responsibility by 92.1% of respondents (Table 2). The majority (72.1%) of respondents had positive attitudes towards the use, cleanliness and maintenance of SS.
Types of sanitation facilities The SS types include various pit latrines (31.6%) and water closets (28.6%), while 1.3% practise OD, despite having SS (see supplementary material e Table S1). Most respondents (62.8%) had one toilet compartment used by 4 households or two compartments (37.2%) used by 5 households.
Physical conditions of the shared toilets Most of the SS were in use (82%) and safe to use (75%) but lack handwashing facilities (81.2%), while 71.9% and 35.8%, respectively, were adequately illuminated in the superstructure during the day and at night.
129
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
Table 2 e Knowledge and attitude of respondents on shared sanitation hygiene variables. Correct response
Description of knowledge variables Handwashing should be performed before all hand-to-mouth contacts (n ¼ 308) Traditional pit latrines are an improved toilet (n ¼ 307) Toilets in the house are required to prevent faeco-oral diseases (n ¼ 309) Toilets should be located downslope to a safe water source (n ¼ 306) Shared toilets are improved if well maintained (n ¼ 309) Toilets should be cited at least 30 m away from water sources (n ¼ 309) Users of toilets could contact infections if dirty (n ¼ 307) Toilets should be cleaned daily (n ¼ 309) Toilets should offer privacy for users (n ¼ 309) Description of attitude variables Cleaning of shared toilets should be performed when dirty (n ¼ 307) Each household should have a private toilet (n ¼ 305) Handwashing can prevent diseases associated with the use of toilet facilities (n ¼ 305) Open defaecation is unavoidable at times, although there is toilet in the house (n ¼ 302) Queuing to use the toilets in houses with a shared toilet is acceptable as a good practice (n ¼ 305) Maintenance of a shared toilet should be a shared responsibility of all resident households (n ¼ 305) Shared toilets should be cleaned and maintained every day (n ¼ 306) It is acceptable to defaecate in the open if someone is pressed but the toilet is being used by someone else (n ¼ 306) Diarrhoeal diseases are prevalent in houses where a toilet is shared (n ¼ 306) Maintenance of a shared toilet is difficult when water is scarce (n ¼ 303)
Perceived attributes of a clean toilet The majority of respondents believed that toilets should not smell (68.8%), should be ventilated (96.3%) and should be devoid of maggots on the superstructure (63.2%). Other important factors were the absence of faeces (27.1%) and urine (44.4%) on its platform, the absence of faeces around the squatting hole (85.5%) and absence of houseflies (62.8%). The toilet cleanliness index showed that only 39.7% of SS were clean.
Maintenance of shared toilets In the study, 27.9% of houses had commercial shops and shared toilets with the shop occupants and their customers. Water was available outside (76.0%) and inside (10.0%) the toilet or absent (14.0%). Shared toilets are cleaned daily in 45.4% of houses, while 14.2% of toilets are cleaned whenever dirty. Toilet cleaning is performed using water and soap (33.4%) primarily by women (52.0%) or anyone (41.8%). Most (93.4%) households abide by the toilet cleaning schedule, with non-compliance mostly attributed to the lack of interest (29.5%) and tiredness (27.9%) [see supplementary material e Table S2].
Determinants of use and cleanliness of shared toilet facilities Table 3 shows a significant association between the attitude and educational status (P ¼ 0.025), wealth (P ¼ 0.011), toilet cleaning frequency (P ¼ 0.015) and toilet compartments (P ¼ 0.046). With cleanliness as the predictor, 15.5% of the
Incorrect response
n
%
n
%
301
97.7
7
2.3
34 283 237 14 195 280 286 244
11.1 91.6 77.5 4.5 63.1 91.2 92.6 79.0
273 26 49 265 113 27 23 65
88.9 8.4 22.5 95.5 36.9 8.8 7.3 21.0
4 267 298
1.3 87.5 97.7
303 38 7
98.7 12.5 2.3
100
33.1
202
66.9
98
32.1
207
67.9
281
92.1
24
7.9
288 208
94.1 68
18 98
5.9 32
164 209
53.6 69
142 94
46.4 31
variation was explained by the regression model (P ¼ 0.001; Table 4). There was no collinearity in the variables (variance inflation factor (VIF) 1.582). The SS cleanliness predictors were factual knowledge (P ¼ 0.039), toilet cleaning efforts (P ¼ 0.019), ability to remember (P ¼ 0.011), commitment (P ¼ 0.040) and cleanliness (P ¼ 0.044). Also, the negative significant factors in the attitudinal affective behaviour, associated with cleaning of SS, implied a dislike for SS maintenance (Table 4). Similarly, the social dilemma factors predict 11.1% of the variance in SS cleanliness (P ¼ 0.002), and the freedom of expression among households on SS cleaning was the only significant (P ¼ 0.011) predictor (Table 5).
Discussion Sociodemographic attributes of respondents In the study, in terms of respondents, the female gender predominates, and most were tenants, in congruence with a study in Uganda.18 Typically, women are more skilled in handling basic household chores in the African setting. The study shows that education influences attitude, and most respondents had a secondary education. Most Nigerians are either poor or vulnerable to poverty25 and earn below the national minimum wage (₦18,000 per month, which is equivalent to approximately US$50), which may be insufficient to sustain life and the provision or upgrade of SH hardware, with most residents being tenants. Besides, the provision of SH facilities is the primary responsibility of house owners.
130
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
Table 3 e Association between attitude with sociodemographic and practice variables. Variable
Composite attitude score Negative attitude
Ethnicity Yoruba Igbo/Ibo Other (Hausa/ Fulani/Igbira) Total Religion Islam Christian Traditional Total Educational statusa Non-formal education Primary Secondary Tertiary Vocational Total
Total
n
%
n
%
n
%
78 7 1
27.1 33.3 100.0
210 14 0
72.9 66.7 0.0
288 21 1
92.9 6.8 0.3
86 27.7 c2 ¼ 2.943
224 72.3 df ¼ 2
310 100.0 P ¼ 0.230
24 26.7 58 27.5 5 50.0 87 28.0 c2 ¼ 3.757
66 153 5 224 df ¼ 3
90 28.9 211 67.8 10 3.2 311 100.0 P ¼ 0.289
13
52.0
9 20.9 50 26.3 14 29.8 0 0.0 86 27.7 a 2 c ¼ 11.114
Monthly income (₦) <₦18,000 ₦18,000 Total
43 34.7 31 25.0 74 29.8 c2 ¼ 2.774 No. of households in houses 2e4 45 29.8 5 39 24.8 Total 84 27.3 c2 ¼ 0.955 Wealth categoriesa Poor 16 16.5 Middle 32 32.3 Rich 33 34.0 Total 81 27.6 c2 ¼ 9.083 Frequency of toilet cleaninga Every day 44 33.1 Others 33 20.5 Total 77 26.2 c2 ¼ 5.968 Toilet cleanliness Not dirty 32 25.8 Dirty 55 29.3 Total 87 27.9 c2 ¼ 0.442 No. of toilet unitsa 1 47 24.0 2 40 34.5 Total 87 27.9 c2 ¼ 3.996 a
Positive attitude
12 34 140 33 05 224 df ¼ 4
73.3 72.5 50.0 72.0
48.0
25
8.1
79.1 73.7 70.2 100.0 72.3
43 13.9 190 61.3 47 15.2 5 1.6 310 100.0 P ¼ 0.025
81 65.3 93 75.0 174 70.2 df ¼ 1
124 50.0 124 50.0 248 100.0 P ¼ 0.096
106 70.2 118 75.2 224 72.7 df ¼ 1
151 49.0 157 51.0 308 100.0 P ¼ 0.328
81 67 64 212 df ¼ 2
83.5 67.7 66.0 72.4
97 33.1 99 33.8 97 33.1 293 100.0 P ¼ 0.011
89 66.9 128 79.5 217 73.8 df ¼ 1
133 45.2 161 54.8 294 100.0 P ¼ 0.015
92 74.2 133 70.7 225 72.1 df ¼ 1
124 39.7 188 60.3 312 100.0 P ¼ 0.506
149 76.0 76 65.5 225 72.1 df ¼ 1
196 62.8 116 37.2 312 100.0 P ¼ 0.046
Significant association.
Respondent's knowledge on SS Knowledge is power, and it can moderate attitude and behaviour.26 Most respondents exhibited a good knowledge,
although this is inadequate (<70%) to ensure a herd immunity from related SH diseases in the neighbourhood.11,27e29 Previous studies have shown that knowledge alone could not catalyse a positive behavioural change18,30,31 but is a stepping stone towards good SS practice, where essential hardware and software are sufficiently provided, perhaps through sanitation marketing. Therefore, good SS knowledge should be reinforced in the population through health promotion aimed at reducing the poor knowledge on SS cleanliness in the target population.
Attitudes on SS Attitudes and behaviour are important determinants of SS cleanliness, although one study concluded that good knowledge and positive attitudes are independent of practice.32 A positive attitude signifies the right disposition to SS maintenance. Also, wealth could influence attitudes, and because many respondents were poor, this could have resulted in the perception that SS should be cleaned only when dirty, and prioritisation of a private household toilet would solve the maintenance debacle. This position agrees with findings by Massa et al.,8 where improved sanitation was context specific and varied across sociocultural contexts. The sense of toilet ownership would encourage users to keep them clean.12,33
Operation and maintenance of SS Most SS were single compartments used by multiple households, which becomes challenging if more than one person needs to use it simultaneously. This promotes alternatives, such as OD that could increase stress and expose women to abuse and violation, with the odds increased at night.9,34,35 One-fifth of people with limited sanitation live in SSA,1 although improved sanitation is context specific and differ across sociocultural contexts.8 However, with the increasing trends of SS in urban settlements, a winewin strategy is required to instil a cleaning culture in users.12
Maintenance of shared toilets Although the ultimate goal is for each household to use a private toilet, public health can be enhanced now through sustainable SS cleanliness.8,11 Dirty toilets are repositories for diseases,36,37 and SS is a risk factor for diarrhoea,6 especially when used by households, shop owners and customers, as is the case in about one-third of respondents.6 Most shared toilets were clean, in agreement with findings in Tanzania, where 74.2% of SS were clean,8 an evidence used to argue for SS consideration as a basic sanitation service;11 however, this is in contrast to the results of Simiyu et al., where most SS were dirty,38 a justification for users reverting to OD, despite having SS in Indonesia and Bangladesh.6 Consistent with the findings in Uganda,18 toilet cleaning was achieved with water and soap, and most households abide by the cleaning rosters, perhaps owing to the presence of water around most SS and a positive cleaning attitude, despite a measure of non-compliance6,34 because of the lack of interest, tiredness and time.
Table 4 e Linear regression of respondents' cleaning commitment on RANAS variables. Factor
Variables
Non-standardised coefficients Beta
Vulnerability Severity
Factual knowledge
Affective
Normative belief
Ability beliefs
Action planning Remembering Commitment Cleanliness Cleaning frequency
Std. error
t
P-value
Beta
95% CI Lower
Upper
1.922 0.020
0.278 0.043
0.030
6.925 0.457
0.000 0.648
1.375 0.065
2.469 0.104
0.002
0.036
0.005
0.064
0.949
0.069
0.074
0.052
0.025
0.131
2.073
0.039
0.003
0.101
0.004
0.037
0.007
0.100
0.921
0.077
0.069
0.056 0.019
0.024 0.029
0.160 0.044
2.370 0.650
0.019 0.516
0.009 0.077
0.103 0.039
0.014 0.008
0.031 0.039
0.031 0.015
0.468 0.216
0.640 0.829
0.075 0.068
0.046 0.084
0.040
0.020
0.122
1.955
0.052
0.080
0.000
0.003
0.030
0.007
0.105
0.916
0.062
0.056
0.038
0.022
0.119
1.768
0.078
0.081
0.004
0.007
0.025
0.019
0.282
0.778
0.056
0.042
0.085
0.033
0.175
2.567
0.011
0.020
0.150
0.066 0.079 0.089
0.032 0.039 0.045
0.155 0.148 0.139
2.063 2.030 1.969
0.040 0.044 0.050
0.129 0.156 0.178
0.003 0.002 0.000
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
Instrumental
(Constant) What are the chances that you get sick if you use a dirty toilet? Imagine you contracted a disease such as diarrhoea, how severe would it be on your social life, household and economic situation? What disease(s) is anyone likely to contract as a result of using a dirty toilet?a How time consuming is it for you to clean and maintain a shared toilet? How much effort is it for you to clean a shared toilet?a How do you feel about cleaning a shared toilet with other households? How negative do you think it is to use a dirty toilet? In general, do you think most people approve or disapprove that it is important to clean the toilet shared with other households? Do you feel a form of social pressure to clean your shared toilet? How difficult is it for you to clean a toilet shared with other households? Does your house have a roster on when households are to clean the shared toilet? Is ensuring cleanliness of the shared toilet on the daily routine of your activities? How difficult is it to remember to clean your shared toilet?a Do you feel committed to cleaning your shared toilet?a How dirty is the toilet you share with other households?a How often do you clean your shared toilet?
Standardised coefficients
Dependent variable: Cleanliness: The toilet is clean, well maintained and in good condition to use now. n ¼ 312, R square ¼ 0.16, ANOVA ¼ 2.713 (16, 237), P ¼ 0.001. CI, confidence interval; RANAS, risks, attitudes, norms, ability and self-regulation; Std. standard; ANOVA, analysis of variance. a Significant relationship.
131
132
Table 5 e Linear regression of respondents' cleaning commitment on social dilemma factor variables. Factors
Perceived cleaning frequency Relationship among households Cleaning households
Individual's cooperation
Talking frequency
Talking difficulty
Cleaning exception
Cooperation among households
Cleaning team
(Constant) Do you clean the shared toilet more or less often than the other users? How good or bad is your relationship with the households you share the toilet with, in terms of its cleaning? How many of the toilets sharing households (office/shops owners) participate in its cleaning? How much do you think that keeping the shared toilet clean depends on your cooperation with other user households? I do not clean the shared toilet more because other users do not do the same. How much do you agree with this statement? How often do you talk with other toilet sharing households on the way it is used or managed?a How difficult is it to talk to other households who you share the toilet with? How much of the shared toilet dirt would you ascribe to persons that just could not clean up or cannot be made responsible (e.g. children, elderly, sick)? How confident are you that the households you share a toilet with cooperate in its cleaning? How confident are you that your shared toilet can be kept clean if all households are cooperative? How much do you feel as a team with other households you share a toilet in regard to its cleaning?
Non-standardised coefficients
Standardised coefficients
t
P-value Lower
Upper
0.000 0.827
1.542 0.057
2.726 0.046
Beta
Std. error
2.134 0.006
0.300 0.026
0.014
7.105 0.218
0.034
0.040
0.056
0.853
0.395
0.114
0.045
0.022
0.024
0.063
0.927
0.355
0.025
0.069
0.015
0.027
0.036
0.532
0.595
0.069
0.040
0.016
0.035
0.031
0.458
0.647
0.085
0.053
0.070
0.028
0.162
2.553
0.011
0.124
0.016
0.015
0.038
0.025
0.395
0.693
0.059
0.089
0.028
0.031
0.060
0.906
0.366
0.033
0.088
0.062
0.039
0.119
1.599
0.111
0.139
0.014
0.029
0.053
0.038
0.541
0.589
0.133
0.076
0.043
0.036
0.094
1.205
0.229
0.114
0.027
Dependent variable: Cleanliness: The toilet is clean, well maintained and in good condition to use now. n ¼ 312, R square ¼ 0.11, ANOVA 2.136 (11, 242), P ¼ 0.002. CI, confidence interval; Std. standard; ANOVA, analysis of variance. a Significant relationship.
Beta
95% CI
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
Individual's cleaning
Variables in the model
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
Psychological determinants of shared toilet cleanliness
Funding
Basic sanitation and improved water are fundamental human rights, yet elusive to many households in SSA.3 The drivers of SS cleanliness (vulnerability to illness and the severity of exposure to dirty toilets), expressed through associated diseases, could enhance the maintenance of SS through rotational cleaning, which could be compromised by poor cleaning commitment by users.6 Toilet cleaning was perceived as a time-consuming activity that could determine the cleaning frequency and effectiveness, in agreement with Contzen and Mosler,40 where perceptions, thoughts and beliefs were shown to influence behaviour, exacerbated by social pressure.41 Moreover, the self-regulating beliefs considered toilet cleaning as a daily routine that should be remembered through commitment and mental drives. These perhaps are responsible for the behaviour predictors, driven by good knowledge and emotions.39 Therefore, most households cooperated on SS cleanliness, while social dilemma factors should be mainstreamed for sustained cleanliness outcomes.40 SS has been proposed for inclusion in basic sanitation services in developing countries, especially in clustered settlements,2 and going forward, through the ‘shared toilet models’42 in a system approach.43 The cleanliness of SS facilities could be enhanced through a collective action to ensure a ‘shared commitment’ on schedules,9 motivated through factual knowledge and disseminated through health promotion in social media channels as a policy to improve cleanliness and safety.9
None declared.
Strengths and weaknesses of the study The study was cross-sectional in nature with respondents selected to ensure randomness. Notwithstanding, the study depended partly on self-reported information in addition to observed data. Also, there was a degree of non-response in the data, and the effect of time-dependent seasonal variations was not considered.
Conclusions Users of SS had good knowledge and attitudes, with mixed practice and behaviour patterns. The factual knowledge, commitment, cleaning and interactions among users of SS determines its cleanliness, and they should focus on achieving a positive behaviour towards improving cleanliness.
Author statements Acknowledgements We are grateful to the respondents for their time and space for data collection.
Ethical approval The study protocol was approved (IPHOAU/12/225) by the ethical committee at the Institute of Public Health, Obafemi Awolowo University in Nigeria.
133
Competing interests None declared.
Data statement The article data is confidential but available on reasonable request from the corresponding author.
Author contributions O.O.A. and E.O.O. conceived the study, its design and data analysis and coordinated manuscript drafts. E.O.O. finalised the contents before submission; U.A.C., E.U.H. and E.M.O. supervised field data collection and data analysis and wrote the introduction, methods and results. All authors approved the final draft for submission.
references
1. WHO/UNICEF. Progress on drinking water, sanitation and hygiene: 2017 update and SDG baselines. [Internet]. Launch Ver. Geneva: World Health Organization (WHO) and the United Nations Children’s Fund (UNICEF); 2017. Available from: http://www. who.int/mediacentre/news/releases/2017/launch-versionreport-jmp-water-sanitation-hygiene.pdf. 2. Massa K, Kilamile F, Safari E, Seleman A, Mwakitalima A, Balengayabo JG, et al. Contributing to the debate on categorising shared sanitation facilities as ‘unimproved’: an account based on field researchers' observations and householders' opinions in three regions, Tanzania. PLoS One 2017;12(11):e0185875. 3. Kamara JK, Galukande M, Maeda F, Luboga S, Renzaho AMN. Understanding the challenges of improving sanitation and hygiene outcomes in a community based intervention: a cross-sectional study in rural Tanzania. Int J Environ Res Publ Health 2017;14(6):1e16. 4. FMWR. Making Nigeria open-defecation-free by 2025: a national roadmap. 2015. 5. WHO/UNICEF. Progress on sanitation and drinking water 2013 update. [Internet]. Joint monitoring programme for water supply and sanitation, vol. 1; 2013. Available from: http://whqlibdoc. who.int/publications/2010/9789241563956_eng_full_text.pdf. 6. Nelson KB, Karver J, Kullman C, Graham JP. User perceptions of shared sanitation among rural households in Indonesia and Bangladesh. PLoS One 2014;9(8):1e13. 7. Heijnen M, Routray P, Torondel B, Clasen T. Shared sanitation versus individual household latrines in urban slums: a crosssectional study in Orissa, India. Am J Trop Med Hyg 2015;93(2):263e8. 8. Massa K, Kilamile F, Safari E, Seleman A, Mwakitalima A, Balengayabo JG, et al. Contributing to the debate on categorising shared sanitation facilities as ‘unimproved’: an account based on field researchers' observations and householders' opinions in three regions, Tanzania. [Internet]. PLoS One 2017 Nov 6;12(11):e0185875. Available from: https:// doi.org/10.1371/journal.pone.0185875. 9. Shiras T, Cumming O, Brown J, Muneme B, Nala R, Dreibelbis R. Shared latrines in Maputo, Mozambique:
134
10.
11.
12.
13. 14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
exploring emotional well-being and psychosocial stress. BMC Int Health Hum Rights 2018;18(30):1e12. Heijnen M, Cumming O, Peletz R, Chan GKS, Brown J, Baker K, et al. Shared sanitation versus individual household latrines: a systematic review of health outcomes. PLoS One 2014;9. Buckley RM, Kallergis A. The sustainable development goal for urban sanitation: Africa's statistical tragedy continues? [Internet]. J Urban Health 2018 [cited 2018 Jun 5]; Available from: https://link-springer-com.unesco-ihe.idm.oclc.org/ content/pdf/10.1007%2Fs11524-018-0267-1.pdf. € nder T, Konradsen F, Keraita B, Apoya P, Gyapong M. Rheinla Redefining shared sanitation. [Internet]. Bull World Health Organ 2015;93(7):509e10 [cited 2018 Mar 9] Available from: https://doi.org/10.2471/BLT.14.144980. Federal Ministry of Water Resources. National water sanitation policy. 2004. NBS/UNICEF. Multiple indicator cluster survey 2016e17. [Internet]. National Bureau of Statistics and United Nations Children's Fund; 2017. Available from: https://www.unicef. org/nigeria/NG_publications_mics_201617.pdf. Harter M, Mosch S, Mosler H. How does Community-Led Total Sanitation (CLTS) affect latrine ownership? A quantitative case study from Mozambique. [Internet]. BMC Publ Health 2018;18 [cited 2018 Mar 27] Available from: https://bmcpublichealth. biomedcentral.com/track/pdf/10.1186/s12889-018-5287-y? site¼bmcpublichealth.biomedcentral.com. Mosler H-J. A systematic approach to behavior change interventions for the water and sanitation sector in developing countries: a conceptual model, a review, and a guideline. [Internet]. Int J Environ Health Res 2012 Oct;22(5):431e49 [cited 2017 Oct 17] Available from: http:// www.tandfonline.com/doi/abs/10.1080/09603123.2011. 650156. Mosler HJ, Mosch S, Harter M. Is Community-led total sanitation connected to the rebuilding of latrines? Quantitative evidence from Mozambique. [Internet]. PLoS One 2018;13(5):e0197483 [cited 2018 Jun 26] Available from: http:// dx.plos.org/10.1371/journal.pone.0197483. Tumwebaze IK, Mosler HJ. Shared toilet users' collective cleaning and determinant factors in Kampala slums, Uganda. BMC Publ Health 2014;14(1):1260. National Population Commission. Population and housing census of the Federal Republic of Nigeria. [Internet]. Osun State Prior Tables 2009. Available from: http://population.gov. ng/publications/. Naing L, Winn T, Rusli BN. Practical issues in calculating the sample size for prevalence studies. Arch Orofac Sci 2006;1:9e14. WHO/UNICEF. 25 years progress on sanitation and drinking water: 2015 update and MDG assessment [Internet]. World Health Organization; 2015. Available from: http://apps.who.int/iris/ bitstream/10665/177752/1/9789241509145_eng.pdf?ua¼1. Hirai M, Graham JP, Mattson KD, Kelsey A, Mukherji S, Cronin AA. Exploring determinants of handwashing with soap in Indonesia: a quantitative analysis. Int J Environ Res Publ Health 2016;13(9). Sudeeshna R, Aparajita D. Determinants of menstrual hygiene among adolescent girls: a multivariate analysis. [Internet]. Natl J Commun Med 2012;3(2):294e301 [cited 2018 Feb 14] Available from: http://njcmindia.org/uploads/3-2_294301.pdf. Cronin A, Sebayang S, Torlesse H, Nandy R. Association of safe disposal of child feces and reported diarrhea in Indonesia: need for stronger focus on a neglected risk. Int J Environ Res Publ Health 2016;13(310). Corral P, Molini V, Oseni G. No condition is permanent: middle class in Nigeria in the last decade. Washington, DC. 2015.
26. Powell D, Jacob C, Chapman B. Enhancing food safety culture to reduce rates of foodborne illness. Food Contr 2011;22(817e822). 27. Benjamin-Chung J, Amin N, Ercumen A, Arnold BF, Hubbard AE, Unicomb L, et al. A randomized controlled trial to measure spillover effects of a combined water, sanitation, and handwashing intervention in rural Bangladesh. [Internet]. Am J Epidemiol 2018 Aug 1;187(8):1733e44. Available from: https://academic.oup.com/aje/advance-article/doi/10. 1093/aje/kwy046/4955448. 28. Carter RC. Can and should sanitation and hygiene programmes be expected to achieve health impacts? Waterlines 2017;36(1):92e103. 29. Fuller JA, Eisenberg JNS. Herd protection from drinking water, sanitation, and hygiene interventions. [Internet]. Am J Trop Med Hyg 2016 Nov 2;95(5):1201e10 [cited 2017 Jul 15] Available from: http://www.ajtmh.org/content/journals/10.4269/ajtmh. 15-0677. 30. Bashaar M, Thawani V, Hassali MA, Saleem F. Disposal practices of unused and expired pharmaceuticals among general public in Kabul. BMC Publ Health 2017;17(1):1e8. 31. Aluko OO, Ogundipe A, Odunjo A, Komolafe J, Olatunji I. Knowledge and food handling practices of nurses in a tertiary health care hospital in Nigeria. J Environ Health 2016;78(6):32e8. 32. Aluko OO, Afolabi OT, Olaoye EA, Adebayo AD, Oyetola SO, Abegunde OO. The management of the faeces passed by under five children: an exploratory, cross-sectional research in an urban community in Southwest Nigeria. [Internet]. BMC Publ Health 2017;17(1):178. Available from: http://www. ncbi.nlm.nih.gov/pubmed/28178955%0Ahttp:// bmcpublichealth.biomedcentral.com/articles/10.1186/s12889017-4078-1. 33. Semiyaga S, Okure MAEE, Niwagaba CB, Katukiza AY, Kansiime F. Decentralized options for faecal sludge management in urban slum areas of Sub-Saharan Africa: a review of technologies, practices and end-uses. [Internet]. Resour Conserv Recycl 2015; vol. 104:109e19 [cited 2017 Jun 3] Available from: https://www.scopus.com/inward/record.uri? eid¼2-s2.0-84947615279&partnerID¼40&md5¼ 671c03627014a3249e8f1c137477ace9. 34. Kwiringira J. Barriers to shared sanitation cleaning and maintenance in Kampala Slums, Uganda. In: 40th WEDC International Conference, Loughborough, UK; 2017. p. 1e6. 35. WSUP. High-quality shared toilets can reduce women's feelings of stress due to fear of violence. 2018. 36. Simiyu S, Swilling M, Cairncross S, Rheingans R. Determinants of quality of shared sanitation facilities in informal settlements: case study of Kisumu, Kenya. [Internet]. BMC Publ Health 2017;17(1):1e13. Available from: https://doi.org/10.1186/s12889-016-4009-6. 37. Simiyu S, Swilling M, Cairncross S. Decision-making on shared sanitation in the informal settlements of Kisumu, Kenya. [Internet]. Int J Environ Health Res 2017;1e17. Available from: https://www.tandfonline.com/doi/full/10.1080/ 09603123.2017.1350261. 38. Simiyu S, Swilling M, Rheingans R, Cairncross S. Estimating the cost and payment for sanitation in the informal settlements of Kisumu, Kenya: a cross sectional study. Int J Environ Res Publ Health 2017;14(1). 39. Contzen N, Mosler H. The RANAS behavioral factors. [Internet]. Du¨bendorf, Switzerland. 2015 (Methodological fact sheet). Report No.: 3. Available from: http://www.eawag.ch/en/ department/ess/main-focus/environmental-and-healthpsychology-ehpsy%0APublications. 40. Contzen N, Mosler H. The RANAS model of behavior change. 2015.
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
41. Kwiringira J, Atekyereza P, Niwagaba C, Gu¨nther I. Descending the sanitation ladder in urban Uganda: evidence from Kampala slums. BMC Publ Health 2014;14(1):1e10. 42. Cardone R, Schrecongost A, Gilsdorf R. Shared and public toilets: championing delivery models that work. [Internet]. Washington, DC: The World Bank; 2018 [cited 2018 Aug 28]. Available from: www.worldbank.org/water. nio AM, Biswas R, 43. Rietveld LC, Siri JG, Chakravarty I, Arse Chatterjee A. Improving health in cities through systems approaches for urban water management. [Internet]. Environ
135
Health 2016 Dec 8;15(S1):S31 [cited 2018 Mar 22] Available from: https://link.springer.com/content/pdf/10.1186% 2Fs12940-016-0107-2.pdf.
Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.puhe.2018.09.013.
Available online at www.sciencedirect.com
Public Health journal homepage: www.elsevier.com/puhe
Original Research
The dynamics and determinants of household shared sanitation cleanliness in a heterogeneous urban settlement in Southwest Nigeria O.O. Aluko a,*, E.O. Oloruntoba b, U.A. Chukwunenye a, E.U. Henry a, E. Ojogun a a b
Department of Community Health, College of Health Sciences, Obafemi Awolowo University, Ile Ife, Nigeria Department of Environmental Health Sciences, University of Ibadan, Nigeria
article info
abstract
Article history:
Objectives: The study assessed the behavioural contexts of household shared sanitation
Received 25 March 2018
(SS), the factors that could enhance its cleanliness and sustainable maintenance in the
Received in revised form
study area.
29 August 2018
Study design: The study design was cross-sectional in nature.
Accepted 13 September 2018
Methods: A pretested, interviewer-administered questionnaire with 113 questions provided
Available online 1 November 2018
information on knowledge, attitude and SS maintenance practices from 312 respondents selected through a four-stage sampling technique. Data were analysed by descriptive, chi-
Keywords:
square and logistic regression statistics with attitude and toilet cleanliness as the depen-
Shared sanitation
dent variables, while the sociodemographic characteristics, knowledge and practice,
Toilet cleanliness
including the toilet cleanliness index, were the explanatory variables at P < 0.05.
Behavioural factors
Results: The mean age of respondents was 46.1 ± 16.4 years. The majority were women (62.7%),
Toilet maintenance
tenants (64.7%) and had a secondary education (61.3%). In addition, 27.9% and 2.7% of respondents had shops and gender-segregated SS, respectively. Their knowledge and attitudes were good (66.3%) and positive (72.1%), while the majority (60.3%) of the SS was dirty. Almost two-thirds (62.8%) and 37.2% of respondents, respectively, had one and two or more compartments of unimproved pit toilets (31.6%) and water closets (28.6%), while 1.4% practiced open defaecation, despite having SS. More than two-fifths of SS were cleaned daily (45.4%) by women (52%), using water and soap (33.4%). Most households (93.4%) complied with the SS cleaning routines, with non-compliance attributed to lack of interest (29.5%), apathy when others defaulted (27.9%), lack of time (26.2%) and inadequate water (14.8%). There was a significant association between attitudes and education (P ¼ 0.025), wealth (P ¼ 0.011) and toiletcleaning frequency (P ¼ 0.015). The positive behaviour determinants of SS are factual knowledge (P ¼ 0.039), efforts required to clean (P ¼ 0.019), the ability to remember (P ¼ 0.011) and cleaning commitment (P ¼ 0.040), while freedom of expression among households using SS was a positive predictor of avoiding conflict, among the social dilemma factors. Conclusions: Good knowledge and positive attitudes were reported among users of SS, although with mixed practice and behaviour patterns. Factual knowledge, cleaning
* Corresponding author. E-mail addresses: [email protected] (O.O. Aluko), [email protected] (E.O. Oloruntoba), [email protected] (U.A. Chukwunenye), [email protected] (E.U. Henry), [email protected] (E. Ojogun). https://doi.org/10.1016/j.puhe.2018.09.013 0033-3506/© 2018 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.
126
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
commitments and freedom of expressions predict SS cleanliness, which should be reinforced using periodic messages and structured health promotion strategies. © 2018 The Royal Society for Public Health. Published by Elsevier Ltd. All rights reserved.
Introduction Improved sanitation is one that hygienically separates faeces from human contact1 and provides a safely managed service when used by a single household.2,3 Through Sustainable Development Goals (SDGs), the United Nations seeks equitable sanitation and hygiene (SH) for all and to end open defecation (OD) by 20301 and in Nigeria by 2025.4 The Millennium Development Goals' sanitation target focussed on ending OD5,6 without considering septage management when vaults are full. Improved sanitation offers a limited service when toilet facilities are shared by two or more households.1,7e9 The use of shared sanitation (SS) has increased to 437 million globally between 1990 and 201410 and 600 million people by 2017,1 mostly in Sub-Saharan Africa (SSA; 18%)5 and Eastern Asia (19%).10 The SDG sanitation goal sought the use of private, household-based, improved sanitation, although incremental access and SS are considered as a winewin strategy in many low-income and urban settlements.1,11 When adequately maintained, SS could prevent environmental contamination from sewage, septage or sludge.12 In Nigeria, sanitation is a private service, where programmes focus on the software support (awareness creation and training among primary beneficiaries to enhance the practice of good hygiene) aimed at guiding household decisions on preferences.13 These programmes could help 55.0% of households that use the most basic (33.0%) and limited (22.0%) toilets in Nigeria, while in Osun state, the Multiple Indicator Cluster Survey revealed that 61.8% of household toilets were improved and comprised nonshared (24.7%) and shared (37.1%) toilets.14 In assessing the SS contextual determinants, the risks, attitudes, norms, ability and self-regulation and social dilemma factors were applied.15,16 The risk index explored the perception of the health risks associated with SS, while the attitudes assessed SS cleanliness and maintenance stance.15,17 The descriptive and injunctive norms predict the social pressure of the SS maintenance; the ability factors assessed respondent's confidence in efforts to maintain SS and the belief in the competence to organise and implement cleaning actions. Self-regulation assessed the ability to organise SS maintenance, and the social dilemma context identified the prevailing interactions, rules and willingness to comply with households using SS.15,18
population of 167,254.19 Ile-Ife is approximately 70 km from Ibadan (Oyo State) and one of the rapidly urbanising settlements in Southwest Nigeria. The LGA experiences a tropical climate with annual rainfall of more than 150 cm and a temperature of between 21 C and 38 C. The city lacks a developmental master plan, and its growth does not follow a structured pattern. The living arrangements are such that most buildings have several households using SS.
The study settings and design The study used the cross-sectional design among households with SS.
Target and study population The target population was households living in buildings accommodating multiple households having SS, while the study population was households in selected houses using a SS toilet. The sampling unit was the household, consisting of one or more people who live in the same house and share meals.3
Sample size The minimum sample of 312, including 10% attrition, was determined using Fisher's formula20 with 24% prevalence of households using SS in Nigeria.21
Sampling technique A four-stage sampling method was used to recruit 312 respondents from eight randomly selected streets by equal probability sampling method. Stage one stratified wards into urban (seven) and rural (four), with the four urban wards selected by simple random sampling (SRS). Stage two selected two streets from each ward by SRS, and stage three selected eligible houses by sampling the fifth odd-numbered houses from the first residential building on each street by systematic technique, providing the house accommodated at least two households using the same toilet. In stage four, 39 households were studied from each of the eight randomly selected streets. However, only one household was selected from each building, represented by the household head, the spouse or the most senior adult member.
Data collection instrument
Methods The study area The study was conducted at Ife Central Local Government Area (LGA) [Fig. 1], which comprises 11 wards and has a
The semistructured, household questionnaire was adapted from Tumwebaze and Mosler.18 It has six sections: sociodemographic characteristics, knowledge, attitude, SS maintenance practices, SS maintenance behaviour determinants and the observational checklist (see online supplementary
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
127
Fig. 1 e Map of Nigeria showing the study area. LGA, Local Government Area.
material e questionnaire). The tool was pretested, revised, translated to Yoruba language and back-translated to English for content validity. The instrument was administered by face-to-face interviews and returned 0.6 Cronbach's a value, indicating an acceptable internal consistency score.
Dependent and explanatory variables The dependent variables were the composite attitude and toilet cleanliness recoded to binary attributes. The cleanliness of the SS was defined as clean, well maintained and safe to use now. In the linear regression models, the explanatory variables were sociodemographic characteristics, knowledge and practice, including the toilet cleanliness index.
Data management Data entry and analysis were performed with IBM SPSS, version 20, software. Continuous and categorical variables were presented by summary statistics and frequency (%), respectively. The relationship between categorical variables was assessed by Chi-square test to determine the type and magnitude of associations at P < 0.05. The linear regression models identified predictors of SS cleanliness. Through dimension reduction, the 21 household owned variables were
regressed through factor scores using the principal component analysis technique. The prime loaded factor regression scores were divided by percentile values function into three equal groups, with 0.5621013, 0.2816214 and 0.2816215 outputs and recoded into poor, average and rich wealth categories, respectively.22 The correct and incorrect responses of the knowledge and attitude variables were, respectively, scored ‘1’ and ‘0’. Using the guide provided by Sudeshna et al.,23 the median (interquartile range [IQR]) for knowledge and attitude indices was 6 and 5e6, respectively. The safe and poor practices in the toilet cleanliness index were scored ‘1’ and ‘0’, respectively, with its median of 10 (IQR 7e11). The values were used to dichotomise the three variables, respectively, into ‘good’ and ‘poor’, ‘positive’ and ‘negative’, and ‘clean’ and ‘dirty’ categories. For multivariate analysis, variables with P < 0.25 in bivariate statistics were imputed in the regression models for behaviour determinants of SS cleanliness.24
Ethical considerations The respondents consented and participated in the study in writing, and the study protocol was approved (IPHOAU/12/ 225) by the ethical committee at the Institute of Public Health, Obafemi Awolowo University in Nigeria.
128
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
b
Table 1 e Sociodemographic characteristics of respondents. Description of variables
Frequency (n)
Percentage (%)
Gender (n ¼ 311) Male 116 37.3 Female 195 62.7 House ownership (n ¼ 309) Owner 109 35.3 Tenant 200 64.7 Household size (n ¼ 312) 2 households 47 15.1 3e4 households 115 36.9 5 households 150 48.1 Relationship of respondents to living house (n ¼ 290) Family owned/occupied 18 6.2 Owner/occupied only 4 1.4 Landlordetenants 144 49.7 All tenants only 124 42.8 Age (n ¼ 308) 24 years 23 7.5 25e44 years 139 45.1 45e64 years 90 29.2 65 years 56 18.2 Length of residence in the house (n ¼ 302) 5 years 148 47.4 >5 years 164 52.6 Number of households living in the house of respondents (n ¼ 301) 2 households 36 12.0 3e4 households 115 38.2 5 households 150 49.8 Ethnicity (n ¼ 309) Yoruba 288 92.9 Igbo 15 4.8 Hausa/Fulani 6 1.9 Religion (n ¼ 311) Islam 90 28.9 Christian 211 67.8 10 3.2 Traditionala Marital status (n ¼ 310) Single 64 20.7 Married 206 66.5 Widowed 38 12.3 Other 02 0.6 Highest education (n ¼ 310) No formal education 25 8.1 Primary education 43 13.9 Secondary education 190 61.3 Tertiary education 47 15.2 Vocational education 4 1.3 1 0.3 Quranic educationb Main occupation (n ¼ 303) Self-employed 197 65.0 Paid employment 43 14.2 Farming 14 4.6 Full-time housewife 11 3.6 Pensioner/retired 10 3.3 Unemployed 21 6.9 Other 7 2.3 Wealth category (n ¼ 293)c Poor 97 33.1 Middle 99 33.8 Rich 97 33.1 a
Religious belief and worship of deities indigenous to the respondents through her/her cultural affiliation, belief and practice.
c
Quranic education is an informal education that happen around or in the mosque, that incorporates formal learning in the study of the Quran for those children that focus attention only on the study of the Quran against those exposed to formal, school based education. Through dimension reduction, the 21 household owned variables were regressed through factor scores using the principal component analysis technique. The prime loaded factor regression scores were divided by percentile values function into three equal groups, with 0.5621013, 0.2816214 and 0.2816215 outputs and recoded into poor, average and rich wealth categories, respectively.
Results Sociodemographic characteristics of respondents The respondents' mean age was 46.1 ± 16.4 years, and the majority were women (62.7%), tenants (64.7%) and married (66.5%). A large proportion of respondents had a secondary education (61.3%), had lived in the house for more than 5 years (52.6%) and were self-employed (76.0%). The model households per building were four, with close to half (48.1%) sharing the same toilet with more than five households. In addition, one-third (33.1%) of respondents were poor (Table 1).
Knowledge of SS and hygiene In Table 2, correct knowledge was exhibited by most (97.7%) respondents on handwashing before mouth contact, and 92.6% knew that toilets should be cleaned daily. Most respondents (91.2%) knew that dirty toilets could breed pathogens and that toilets should be located at a lower elevation in relation to safe water sources (77.5%). About two-thirds (66.3%) of respondents had a good knowledge of SS.
Attitudes to the cleanliness and maintenance of SS Most respondents (98.7%) believed that SS should be cleaned only when dirty, while 87.5% believed that each household should have a private toilet. Queuing was perceived as a good practice in using SS (67.9%), and its maintenance was considered as a shared responsibility by 92.1% of respondents (Table 2). The majority (72.1%) of respondents had positive attitudes towards the use, cleanliness and maintenance of SS.
Types of sanitation facilities The SS types include various pit latrines (31.6%) and water closets (28.6%), while 1.3% practise OD, despite having SS (see supplementary material e Table S1). Most respondents (62.8%) had one toilet compartment used by 4 households or two compartments (37.2%) used by 5 households.
Physical conditions of the shared toilets Most of the SS were in use (82%) and safe to use (75%) but lack handwashing facilities (81.2%), while 71.9% and 35.8%, respectively, were adequately illuminated in the superstructure during the day and at night.
129
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
Table 2 e Knowledge and attitude of respondents on shared sanitation hygiene variables. Correct response
Description of knowledge variables Handwashing should be performed before all hand-to-mouth contacts (n ¼ 308) Traditional pit latrines are an improved toilet (n ¼ 307) Toilets in the house are required to prevent faeco-oral diseases (n ¼ 309) Toilets should be located downslope to a safe water source (n ¼ 306) Shared toilets are improved if well maintained (n ¼ 309) Toilets should be cited at least 30 m away from water sources (n ¼ 309) Users of toilets could contact infections if dirty (n ¼ 307) Toilets should be cleaned daily (n ¼ 309) Toilets should offer privacy for users (n ¼ 309) Description of attitude variables Cleaning of shared toilets should be performed when dirty (n ¼ 307) Each household should have a private toilet (n ¼ 305) Handwashing can prevent diseases associated with the use of toilet facilities (n ¼ 305) Open defaecation is unavoidable at times, although there is toilet in the house (n ¼ 302) Queuing to use the toilets in houses with a shared toilet is acceptable as a good practice (n ¼ 305) Maintenance of a shared toilet should be a shared responsibility of all resident households (n ¼ 305) Shared toilets should be cleaned and maintained every day (n ¼ 306) It is acceptable to defaecate in the open if someone is pressed but the toilet is being used by someone else (n ¼ 306) Diarrhoeal diseases are prevalent in houses where a toilet is shared (n ¼ 306) Maintenance of a shared toilet is difficult when water is scarce (n ¼ 303)
Perceived attributes of a clean toilet The majority of respondents believed that toilets should not smell (68.8%), should be ventilated (96.3%) and should be devoid of maggots on the superstructure (63.2%). Other important factors were the absence of faeces (27.1%) and urine (44.4%) on its platform, the absence of faeces around the squatting hole (85.5%) and absence of houseflies (62.8%). The toilet cleanliness index showed that only 39.7% of SS were clean.
Maintenance of shared toilets In the study, 27.9% of houses had commercial shops and shared toilets with the shop occupants and their customers. Water was available outside (76.0%) and inside (10.0%) the toilet or absent (14.0%). Shared toilets are cleaned daily in 45.4% of houses, while 14.2% of toilets are cleaned whenever dirty. Toilet cleaning is performed using water and soap (33.4%) primarily by women (52.0%) or anyone (41.8%). Most (93.4%) households abide by the toilet cleaning schedule, with non-compliance mostly attributed to the lack of interest (29.5%) and tiredness (27.9%) [see supplementary material e Table S2].
Determinants of use and cleanliness of shared toilet facilities Table 3 shows a significant association between the attitude and educational status (P ¼ 0.025), wealth (P ¼ 0.011), toilet cleaning frequency (P ¼ 0.015) and toilet compartments (P ¼ 0.046). With cleanliness as the predictor, 15.5% of the
Incorrect response
n
%
n
%
301
97.7
7
2.3
34 283 237 14 195 280 286 244
11.1 91.6 77.5 4.5 63.1 91.2 92.6 79.0
273 26 49 265 113 27 23 65
88.9 8.4 22.5 95.5 36.9 8.8 7.3 21.0
4 267 298
1.3 87.5 97.7
303 38 7
98.7 12.5 2.3
100
33.1
202
66.9
98
32.1
207
67.9
281
92.1
24
7.9
288 208
94.1 68
18 98
5.9 32
164 209
53.6 69
142 94
46.4 31
variation was explained by the regression model (P ¼ 0.001; Table 4). There was no collinearity in the variables (variance inflation factor (VIF) 1.582). The SS cleanliness predictors were factual knowledge (P ¼ 0.039), toilet cleaning efforts (P ¼ 0.019), ability to remember (P ¼ 0.011), commitment (P ¼ 0.040) and cleanliness (P ¼ 0.044). Also, the negative significant factors in the attitudinal affective behaviour, associated with cleaning of SS, implied a dislike for SS maintenance (Table 4). Similarly, the social dilemma factors predict 11.1% of the variance in SS cleanliness (P ¼ 0.002), and the freedom of expression among households on SS cleaning was the only significant (P ¼ 0.011) predictor (Table 5).
Discussion Sociodemographic attributes of respondents In the study, in terms of respondents, the female gender predominates, and most were tenants, in congruence with a study in Uganda.18 Typically, women are more skilled in handling basic household chores in the African setting. The study shows that education influences attitude, and most respondents had a secondary education. Most Nigerians are either poor or vulnerable to poverty25 and earn below the national minimum wage (₦18,000 per month, which is equivalent to approximately US$50), which may be insufficient to sustain life and the provision or upgrade of SH hardware, with most residents being tenants. Besides, the provision of SH facilities is the primary responsibility of house owners.
130
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
Table 3 e Association between attitude with sociodemographic and practice variables. Variable
Composite attitude score Negative attitude
Ethnicity Yoruba Igbo/Ibo Other (Hausa/ Fulani/Igbira) Total Religion Islam Christian Traditional Total Educational statusa Non-formal education Primary Secondary Tertiary Vocational Total
Total
n
%
n
%
n
%
78 7 1
27.1 33.3 100.0
210 14 0
72.9 66.7 0.0
288 21 1
92.9 6.8 0.3
86 27.7 c2 ¼ 2.943
224 72.3 df ¼ 2
310 100.0 P ¼ 0.230
24 26.7 58 27.5 5 50.0 87 28.0 c2 ¼ 3.757
66 153 5 224 df ¼ 3
90 28.9 211 67.8 10 3.2 311 100.0 P ¼ 0.289
13
52.0
9 20.9 50 26.3 14 29.8 0 0.0 86 27.7 a 2 c ¼ 11.114
Monthly income (₦) <₦18,000 ₦18,000 Total
43 34.7 31 25.0 74 29.8 c2 ¼ 2.774 No. of households in houses 2e4 45 29.8 5 39 24.8 Total 84 27.3 c2 ¼ 0.955 Wealth categoriesa Poor 16 16.5 Middle 32 32.3 Rich 33 34.0 Total 81 27.6 c2 ¼ 9.083 Frequency of toilet cleaninga Every day 44 33.1 Others 33 20.5 Total 77 26.2 c2 ¼ 5.968 Toilet cleanliness Not dirty 32 25.8 Dirty 55 29.3 Total 87 27.9 c2 ¼ 0.442 No. of toilet unitsa 1 47 24.0 2 40 34.5 Total 87 27.9 c2 ¼ 3.996 a
Positive attitude
12 34 140 33 05 224 df ¼ 4
73.3 72.5 50.0 72.0
48.0
25
8.1
79.1 73.7 70.2 100.0 72.3
43 13.9 190 61.3 47 15.2 5 1.6 310 100.0 P ¼ 0.025
81 65.3 93 75.0 174 70.2 df ¼ 1
124 50.0 124 50.0 248 100.0 P ¼ 0.096
106 70.2 118 75.2 224 72.7 df ¼ 1
151 49.0 157 51.0 308 100.0 P ¼ 0.328
81 67 64 212 df ¼ 2
83.5 67.7 66.0 72.4
97 33.1 99 33.8 97 33.1 293 100.0 P ¼ 0.011
89 66.9 128 79.5 217 73.8 df ¼ 1
133 45.2 161 54.8 294 100.0 P ¼ 0.015
92 74.2 133 70.7 225 72.1 df ¼ 1
124 39.7 188 60.3 312 100.0 P ¼ 0.506
149 76.0 76 65.5 225 72.1 df ¼ 1
196 62.8 116 37.2 312 100.0 P ¼ 0.046
Significant association.
Respondent's knowledge on SS Knowledge is power, and it can moderate attitude and behaviour.26 Most respondents exhibited a good knowledge,
although this is inadequate (<70%) to ensure a herd immunity from related SH diseases in the neighbourhood.11,27e29 Previous studies have shown that knowledge alone could not catalyse a positive behavioural change18,30,31 but is a stepping stone towards good SS practice, where essential hardware and software are sufficiently provided, perhaps through sanitation marketing. Therefore, good SS knowledge should be reinforced in the population through health promotion aimed at reducing the poor knowledge on SS cleanliness in the target population.
Attitudes on SS Attitudes and behaviour are important determinants of SS cleanliness, although one study concluded that good knowledge and positive attitudes are independent of practice.32 A positive attitude signifies the right disposition to SS maintenance. Also, wealth could influence attitudes, and because many respondents were poor, this could have resulted in the perception that SS should be cleaned only when dirty, and prioritisation of a private household toilet would solve the maintenance debacle. This position agrees with findings by Massa et al.,8 where improved sanitation was context specific and varied across sociocultural contexts. The sense of toilet ownership would encourage users to keep them clean.12,33
Operation and maintenance of SS Most SS were single compartments used by multiple households, which becomes challenging if more than one person needs to use it simultaneously. This promotes alternatives, such as OD that could increase stress and expose women to abuse and violation, with the odds increased at night.9,34,35 One-fifth of people with limited sanitation live in SSA,1 although improved sanitation is context specific and differ across sociocultural contexts.8 However, with the increasing trends of SS in urban settlements, a winewin strategy is required to instil a cleaning culture in users.12
Maintenance of shared toilets Although the ultimate goal is for each household to use a private toilet, public health can be enhanced now through sustainable SS cleanliness.8,11 Dirty toilets are repositories for diseases,36,37 and SS is a risk factor for diarrhoea,6 especially when used by households, shop owners and customers, as is the case in about one-third of respondents.6 Most shared toilets were clean, in agreement with findings in Tanzania, where 74.2% of SS were clean,8 an evidence used to argue for SS consideration as a basic sanitation service;11 however, this is in contrast to the results of Simiyu et al., where most SS were dirty,38 a justification for users reverting to OD, despite having SS in Indonesia and Bangladesh.6 Consistent with the findings in Uganda,18 toilet cleaning was achieved with water and soap, and most households abide by the cleaning rosters, perhaps owing to the presence of water around most SS and a positive cleaning attitude, despite a measure of non-compliance6,34 because of the lack of interest, tiredness and time.
Table 4 e Linear regression of respondents' cleaning commitment on RANAS variables. Factor
Variables
Non-standardised coefficients Beta
Vulnerability Severity
Factual knowledge
Affective
Normative belief
Ability beliefs
Action planning Remembering Commitment Cleanliness Cleaning frequency
Std. error
t
P-value
Beta
95% CI Lower
Upper
1.922 0.020
0.278 0.043
0.030
6.925 0.457
0.000 0.648
1.375 0.065
2.469 0.104
0.002
0.036
0.005
0.064
0.949
0.069
0.074
0.052
0.025
0.131
2.073
0.039
0.003
0.101
0.004
0.037
0.007
0.100
0.921
0.077
0.069
0.056 0.019
0.024 0.029
0.160 0.044
2.370 0.650
0.019 0.516
0.009 0.077
0.103 0.039
0.014 0.008
0.031 0.039
0.031 0.015
0.468 0.216
0.640 0.829
0.075 0.068
0.046 0.084
0.040
0.020
0.122
1.955
0.052
0.080
0.000
0.003
0.030
0.007
0.105
0.916
0.062
0.056
0.038
0.022
0.119
1.768
0.078
0.081
0.004
0.007
0.025
0.019
0.282
0.778
0.056
0.042
0.085
0.033
0.175
2.567
0.011
0.020
0.150
0.066 0.079 0.089
0.032 0.039 0.045
0.155 0.148 0.139
2.063 2.030 1.969
0.040 0.044 0.050
0.129 0.156 0.178
0.003 0.002 0.000
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
Instrumental
(Constant) What are the chances that you get sick if you use a dirty toilet? Imagine you contracted a disease such as diarrhoea, how severe would it be on your social life, household and economic situation? What disease(s) is anyone likely to contract as a result of using a dirty toilet?a How time consuming is it for you to clean and maintain a shared toilet? How much effort is it for you to clean a shared toilet?a How do you feel about cleaning a shared toilet with other households? How negative do you think it is to use a dirty toilet? In general, do you think most people approve or disapprove that it is important to clean the toilet shared with other households? Do you feel a form of social pressure to clean your shared toilet? How difficult is it for you to clean a toilet shared with other households? Does your house have a roster on when households are to clean the shared toilet? Is ensuring cleanliness of the shared toilet on the daily routine of your activities? How difficult is it to remember to clean your shared toilet?a Do you feel committed to cleaning your shared toilet?a How dirty is the toilet you share with other households?a How often do you clean your shared toilet?
Standardised coefficients
Dependent variable: Cleanliness: The toilet is clean, well maintained and in good condition to use now. n ¼ 312, R square ¼ 0.16, ANOVA ¼ 2.713 (16, 237), P ¼ 0.001. CI, confidence interval; RANAS, risks, attitudes, norms, ability and self-regulation; Std. standard; ANOVA, analysis of variance. a Significant relationship.
131
132
Table 5 e Linear regression of respondents' cleaning commitment on social dilemma factor variables. Factors
Perceived cleaning frequency Relationship among households Cleaning households
Individual's cooperation
Talking frequency
Talking difficulty
Cleaning exception
Cooperation among households
Cleaning team
(Constant) Do you clean the shared toilet more or less often than the other users? How good or bad is your relationship with the households you share the toilet with, in terms of its cleaning? How many of the toilets sharing households (office/shops owners) participate in its cleaning? How much do you think that keeping the shared toilet clean depends on your cooperation with other user households? I do not clean the shared toilet more because other users do not do the same. How much do you agree with this statement? How often do you talk with other toilet sharing households on the way it is used or managed?a How difficult is it to talk to other households who you share the toilet with? How much of the shared toilet dirt would you ascribe to persons that just could not clean up or cannot be made responsible (e.g. children, elderly, sick)? How confident are you that the households you share a toilet with cooperate in its cleaning? How confident are you that your shared toilet can be kept clean if all households are cooperative? How much do you feel as a team with other households you share a toilet in regard to its cleaning?
Non-standardised coefficients
Standardised coefficients
t
P-value Lower
Upper
0.000 0.827
1.542 0.057
2.726 0.046
Beta
Std. error
2.134 0.006
0.300 0.026
0.014
7.105 0.218
0.034
0.040
0.056
0.853
0.395
0.114
0.045
0.022
0.024
0.063
0.927
0.355
0.025
0.069
0.015
0.027
0.036
0.532
0.595
0.069
0.040
0.016
0.035
0.031
0.458
0.647
0.085
0.053
0.070
0.028
0.162
2.553
0.011
0.124
0.016
0.015
0.038
0.025
0.395
0.693
0.059
0.089
0.028
0.031
0.060
0.906
0.366
0.033
0.088
0.062
0.039
0.119
1.599
0.111
0.139
0.014
0.029
0.053
0.038
0.541
0.589
0.133
0.076
0.043
0.036
0.094
1.205
0.229
0.114
0.027
Dependent variable: Cleanliness: The toilet is clean, well maintained and in good condition to use now. n ¼ 312, R square ¼ 0.11, ANOVA 2.136 (11, 242), P ¼ 0.002. CI, confidence interval; Std. standard; ANOVA, analysis of variance. a Significant relationship.
Beta
95% CI
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
Individual's cleaning
Variables in the model
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
Psychological determinants of shared toilet cleanliness
Funding
Basic sanitation and improved water are fundamental human rights, yet elusive to many households in SSA.3 The drivers of SS cleanliness (vulnerability to illness and the severity of exposure to dirty toilets), expressed through associated diseases, could enhance the maintenance of SS through rotational cleaning, which could be compromised by poor cleaning commitment by users.6 Toilet cleaning was perceived as a time-consuming activity that could determine the cleaning frequency and effectiveness, in agreement with Contzen and Mosler,40 where perceptions, thoughts and beliefs were shown to influence behaviour, exacerbated by social pressure.41 Moreover, the self-regulating beliefs considered toilet cleaning as a daily routine that should be remembered through commitment and mental drives. These perhaps are responsible for the behaviour predictors, driven by good knowledge and emotions.39 Therefore, most households cooperated on SS cleanliness, while social dilemma factors should be mainstreamed for sustained cleanliness outcomes.40 SS has been proposed for inclusion in basic sanitation services in developing countries, especially in clustered settlements,2 and going forward, through the ‘shared toilet models’42 in a system approach.43 The cleanliness of SS facilities could be enhanced through a collective action to ensure a ‘shared commitment’ on schedules,9 motivated through factual knowledge and disseminated through health promotion in social media channels as a policy to improve cleanliness and safety.9
None declared.
Strengths and weaknesses of the study The study was cross-sectional in nature with respondents selected to ensure randomness. Notwithstanding, the study depended partly on self-reported information in addition to observed data. Also, there was a degree of non-response in the data, and the effect of time-dependent seasonal variations was not considered.
Conclusions Users of SS had good knowledge and attitudes, with mixed practice and behaviour patterns. The factual knowledge, commitment, cleaning and interactions among users of SS determines its cleanliness, and they should focus on achieving a positive behaviour towards improving cleanliness.
Author statements Acknowledgements We are grateful to the respondents for their time and space for data collection.
Ethical approval The study protocol was approved (IPHOAU/12/225) by the ethical committee at the Institute of Public Health, Obafemi Awolowo University in Nigeria.
133
Competing interests None declared.
Data statement The article data is confidential but available on reasonable request from the corresponding author.
Author contributions O.O.A. and E.O.O. conceived the study, its design and data analysis and coordinated manuscript drafts. E.O.O. finalised the contents before submission; U.A.C., E.U.H. and E.M.O. supervised field data collection and data analysis and wrote the introduction, methods and results. All authors approved the final draft for submission.
references
1. WHO/UNICEF. Progress on drinking water, sanitation and hygiene: 2017 update and SDG baselines. [Internet]. Launch Ver. Geneva: World Health Organization (WHO) and the United Nations Children’s Fund (UNICEF); 2017. Available from: http://www. who.int/mediacentre/news/releases/2017/launch-versionreport-jmp-water-sanitation-hygiene.pdf. 2. Massa K, Kilamile F, Safari E, Seleman A, Mwakitalima A, Balengayabo JG, et al. Contributing to the debate on categorising shared sanitation facilities as ‘unimproved’: an account based on field researchers' observations and householders' opinions in three regions, Tanzania. PLoS One 2017;12(11):e0185875. 3. Kamara JK, Galukande M, Maeda F, Luboga S, Renzaho AMN. Understanding the challenges of improving sanitation and hygiene outcomes in a community based intervention: a cross-sectional study in rural Tanzania. Int J Environ Res Publ Health 2017;14(6):1e16. 4. FMWR. Making Nigeria open-defecation-free by 2025: a national roadmap. 2015. 5. WHO/UNICEF. Progress on sanitation and drinking water 2013 update. [Internet]. Joint monitoring programme for water supply and sanitation, vol. 1; 2013. Available from: http://whqlibdoc. who.int/publications/2010/9789241563956_eng_full_text.pdf. 6. Nelson KB, Karver J, Kullman C, Graham JP. User perceptions of shared sanitation among rural households in Indonesia and Bangladesh. PLoS One 2014;9(8):1e13. 7. Heijnen M, Routray P, Torondel B, Clasen T. Shared sanitation versus individual household latrines in urban slums: a crosssectional study in Orissa, India. Am J Trop Med Hyg 2015;93(2):263e8. 8. Massa K, Kilamile F, Safari E, Seleman A, Mwakitalima A, Balengayabo JG, et al. Contributing to the debate on categorising shared sanitation facilities as ‘unimproved’: an account based on field researchers' observations and householders' opinions in three regions, Tanzania. [Internet]. PLoS One 2017 Nov 6;12(11):e0185875. Available from: https:// doi.org/10.1371/journal.pone.0185875. 9. Shiras T, Cumming O, Brown J, Muneme B, Nala R, Dreibelbis R. Shared latrines in Maputo, Mozambique:
134
10.
11.
12.
13. 14.
15.
16.
17.
18.
19.
20.
21.
22.
23.
24.
25.
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
exploring emotional well-being and psychosocial stress. BMC Int Health Hum Rights 2018;18(30):1e12. Heijnen M, Cumming O, Peletz R, Chan GKS, Brown J, Baker K, et al. Shared sanitation versus individual household latrines: a systematic review of health outcomes. PLoS One 2014;9. Buckley RM, Kallergis A. The sustainable development goal for urban sanitation: Africa's statistical tragedy continues? [Internet]. J Urban Health 2018 [cited 2018 Jun 5]; Available from: https://link-springer-com.unesco-ihe.idm.oclc.org/ content/pdf/10.1007%2Fs11524-018-0267-1.pdf. € nder T, Konradsen F, Keraita B, Apoya P, Gyapong M. Rheinla Redefining shared sanitation. [Internet]. Bull World Health Organ 2015;93(7):509e10 [cited 2018 Mar 9] Available from: https://doi.org/10.2471/BLT.14.144980. Federal Ministry of Water Resources. National water sanitation policy. 2004. NBS/UNICEF. Multiple indicator cluster survey 2016e17. [Internet]. National Bureau of Statistics and United Nations Children's Fund; 2017. Available from: https://www.unicef. org/nigeria/NG_publications_mics_201617.pdf. Harter M, Mosch S, Mosler H. How does Community-Led Total Sanitation (CLTS) affect latrine ownership? A quantitative case study from Mozambique. [Internet]. BMC Publ Health 2018;18 [cited 2018 Mar 27] Available from: https://bmcpublichealth. biomedcentral.com/track/pdf/10.1186/s12889-018-5287-y? site¼bmcpublichealth.biomedcentral.com. Mosler H-J. A systematic approach to behavior change interventions for the water and sanitation sector in developing countries: a conceptual model, a review, and a guideline. [Internet]. Int J Environ Health Res 2012 Oct;22(5):431e49 [cited 2017 Oct 17] Available from: http:// www.tandfonline.com/doi/abs/10.1080/09603123.2011. 650156. Mosler HJ, Mosch S, Harter M. Is Community-led total sanitation connected to the rebuilding of latrines? Quantitative evidence from Mozambique. [Internet]. PLoS One 2018;13(5):e0197483 [cited 2018 Jun 26] Available from: http:// dx.plos.org/10.1371/journal.pone.0197483. Tumwebaze IK, Mosler HJ. Shared toilet users' collective cleaning and determinant factors in Kampala slums, Uganda. BMC Publ Health 2014;14(1):1260. National Population Commission. Population and housing census of the Federal Republic of Nigeria. [Internet]. Osun State Prior Tables 2009. Available from: http://population.gov. ng/publications/. Naing L, Winn T, Rusli BN. Practical issues in calculating the sample size for prevalence studies. Arch Orofac Sci 2006;1:9e14. WHO/UNICEF. 25 years progress on sanitation and drinking water: 2015 update and MDG assessment [Internet]. World Health Organization; 2015. Available from: http://apps.who.int/iris/ bitstream/10665/177752/1/9789241509145_eng.pdf?ua¼1. Hirai M, Graham JP, Mattson KD, Kelsey A, Mukherji S, Cronin AA. Exploring determinants of handwashing with soap in Indonesia: a quantitative analysis. Int J Environ Res Publ Health 2016;13(9). Sudeeshna R, Aparajita D. Determinants of menstrual hygiene among adolescent girls: a multivariate analysis. [Internet]. Natl J Commun Med 2012;3(2):294e301 [cited 2018 Feb 14] Available from: http://njcmindia.org/uploads/3-2_294301.pdf. Cronin A, Sebayang S, Torlesse H, Nandy R. Association of safe disposal of child feces and reported diarrhea in Indonesia: need for stronger focus on a neglected risk. Int J Environ Res Publ Health 2016;13(310). Corral P, Molini V, Oseni G. No condition is permanent: middle class in Nigeria in the last decade. Washington, DC. 2015.
26. Powell D, Jacob C, Chapman B. Enhancing food safety culture to reduce rates of foodborne illness. Food Contr 2011;22(817e822). 27. Benjamin-Chung J, Amin N, Ercumen A, Arnold BF, Hubbard AE, Unicomb L, et al. A randomized controlled trial to measure spillover effects of a combined water, sanitation, and handwashing intervention in rural Bangladesh. [Internet]. Am J Epidemiol 2018 Aug 1;187(8):1733e44. Available from: https://academic.oup.com/aje/advance-article/doi/10. 1093/aje/kwy046/4955448. 28. Carter RC. Can and should sanitation and hygiene programmes be expected to achieve health impacts? Waterlines 2017;36(1):92e103. 29. Fuller JA, Eisenberg JNS. Herd protection from drinking water, sanitation, and hygiene interventions. [Internet]. Am J Trop Med Hyg 2016 Nov 2;95(5):1201e10 [cited 2017 Jul 15] Available from: http://www.ajtmh.org/content/journals/10.4269/ajtmh. 15-0677. 30. Bashaar M, Thawani V, Hassali MA, Saleem F. Disposal practices of unused and expired pharmaceuticals among general public in Kabul. BMC Publ Health 2017;17(1):1e8. 31. Aluko OO, Ogundipe A, Odunjo A, Komolafe J, Olatunji I. Knowledge and food handling practices of nurses in a tertiary health care hospital in Nigeria. J Environ Health 2016;78(6):32e8. 32. Aluko OO, Afolabi OT, Olaoye EA, Adebayo AD, Oyetola SO, Abegunde OO. The management of the faeces passed by under five children: an exploratory, cross-sectional research in an urban community in Southwest Nigeria. [Internet]. BMC Publ Health 2017;17(1):178. Available from: http://www. ncbi.nlm.nih.gov/pubmed/28178955%0Ahttp:// bmcpublichealth.biomedcentral.com/articles/10.1186/s12889017-4078-1. 33. Semiyaga S, Okure MAEE, Niwagaba CB, Katukiza AY, Kansiime F. Decentralized options for faecal sludge management in urban slum areas of Sub-Saharan Africa: a review of technologies, practices and end-uses. [Internet]. Resour Conserv Recycl 2015; vol. 104:109e19 [cited 2017 Jun 3] Available from: https://www.scopus.com/inward/record.uri? eid¼2-s2.0-84947615279&partnerID¼40&md5¼ 671c03627014a3249e8f1c137477ace9. 34. Kwiringira J. Barriers to shared sanitation cleaning and maintenance in Kampala Slums, Uganda. In: 40th WEDC International Conference, Loughborough, UK; 2017. p. 1e6. 35. WSUP. High-quality shared toilets can reduce women's feelings of stress due to fear of violence. 2018. 36. Simiyu S, Swilling M, Cairncross S, Rheingans R. Determinants of quality of shared sanitation facilities in informal settlements: case study of Kisumu, Kenya. [Internet]. BMC Publ Health 2017;17(1):1e13. Available from: https://doi.org/10.1186/s12889-016-4009-6. 37. Simiyu S, Swilling M, Cairncross S. Decision-making on shared sanitation in the informal settlements of Kisumu, Kenya. [Internet]. Int J Environ Health Res 2017;1e17. Available from: https://www.tandfonline.com/doi/full/10.1080/ 09603123.2017.1350261. 38. Simiyu S, Swilling M, Rheingans R, Cairncross S. Estimating the cost and payment for sanitation in the informal settlements of Kisumu, Kenya: a cross sectional study. Int J Environ Res Publ Health 2017;14(1). 39. Contzen N, Mosler H. The RANAS behavioral factors. [Internet]. Du¨bendorf, Switzerland. 2015 (Methodological fact sheet). Report No.: 3. Available from: http://www.eawag.ch/en/ department/ess/main-focus/environmental-and-healthpsychology-ehpsy%0APublications. 40. Contzen N, Mosler H. The RANAS model of behavior change. 2015.
p u b l i c h e a l t h 1 6 5 ( 2 0 1 8 ) 1 2 5 e1 3 5
41. Kwiringira J, Atekyereza P, Niwagaba C, Gu¨nther I. Descending the sanitation ladder in urban Uganda: evidence from Kampala slums. BMC Publ Health 2014;14(1):1e10. 42. Cardone R, Schrecongost A, Gilsdorf R. Shared and public toilets: championing delivery models that work. [Internet]. Washington, DC: The World Bank; 2018 [cited 2018 Aug 28]. Available from: www.worldbank.org/water. nio AM, Biswas R, 43. Rietveld LC, Siri JG, Chakravarty I, Arse Chatterjee A. Improving health in cities through systems approaches for urban water management. [Internet]. Environ
135
Health 2016 Dec 8;15(S1):S31 [cited 2018 Mar 22] Available from: https://link.springer.com/content/pdf/10.1186% 2Fs12940-016-0107-2.pdf.
Appendix A. Supplementary data Supplementary data to this article can be found online at https://doi.org/10.1016/j.puhe.2018.09.013.