The 15 studies selected for inclusion were evaluated for level of evidence, risk of bias, and overall quality. A more detailed summary of the study findings is shown in Table 1. Quality assessment results are shown in Table 2.
Summary of locations, age groups, sample sizes, durations, intervention types, and their outcome measures and overall findings of studies included in review of interventions targeting CFD practices in Asian-Pacific countries
| Authors . | Location . | Age group/inclusion criteria . | Sample size . | Duration of study . | Intervention type: intervention description . | Data collection methods . | Outcome measures . | Impacts on CFD practices . |
|---|---|---|---|---|---|---|---|---|
| Luby et al. (2018) | Rural Bangladesh | Intervention group: Households that had participated in the WASH Benefits Bangladesh trial with pregnant women assigned to the following groups: chlorinated drinking water (water); upgraded sanitation (sanitation); promotion of handwashing with soap (handwashing); combined water, sanitation, and handwashing; counselling on appropriate child nutrition (nutrition); combined water, sanitation, handwashing, and nutrition. Control group: data collection only. | 5,551 pregnant women in intervention groups in the WASH Benefits trial; 1,382 women in the control group. | Evaluated WASH Benefits Bangladesh outcomes at 1- and 2-years post-intervention (intervention delivered over 15 months). | Hardware and Behavioral: WASH Benefits Bangladesh was a cluster-randomized trial that enrolled pregnant women and evaluated outcomes at 1- and 2-years follow-up. All households in groups with sanitation component received latrine installation or repairment, a sani-scoop for removing feces from the compound, and households with children <3 received child potties. Promoters encouraged mothers to teach their children to use the potties, to safely dispose of feces in latrines, and to regularly remove animal and human feces from the compound. | Caregiver-reported data, and the presence of feces on latrine slab or floor. | Comparison to control and baseline % of latrines with functional water seals and the presence of visible feces on latrine slab or floor. | Sanitation groups: 48–56% of households had visible feces on floor at baseline compared to 11–14% at 1 year and 14–18% at 2 years. 26–30% had latrine water seals at baseline compared to 95 and 97% at 1 and 2 years. Control group: 52% of households had visible feces on floor at baseline compared to 40% at 1 year and 44% at 2 years. 31% had functional latrine water seals at baseline compared to 29 and 31% at 1 and 2 years. |
| Parvez et al. (2018) | Rural Bangladesh | Intervention group: Households that had participated in the WASH Benefits Bangladesh trial with pregnant women assigned to the following groups: chlorinated drinking water (water); upgraded sanitation (sanitation); promotion of handwashing with soap (handwashing); combined water, sanitation, and handwashing; counselling on appropriate child nutrition (nutrition); combined water, sanitation, handwashing, and nutrition. Control group: data collection only. | 5,551 pregnant women in intervention groups; 1,382 women in the control group. | Fidelity assessments over 20 months from November 2012 to October 2014. | Hardware and Behavioral: WASH Benefits Bangladesh was a cluster-randomized trial that enrolled pregnant women and evaluated outcomes at 1- and 2-year follow-up. All households in groups with sanitation component received latrine installation or repairment, a sani-scoop for removing feces from the compound, and households with children <3 received child potties. Promoters encouraged mothers to teach their children to use the potties, to safely dispose of feces in latrines, and to regularly remove animal and human feces from the compound. | Implementation fidelity and structured observations. Monthly spot-checks and surveys for technology and behavioral uptake (feces presence in the courtyard as an indicator of sani-scoop use and safe feces disposal). | Observed hygienic latrine, feces presence in the courtyard as an indicator of sani-scoop use and safe feces disposal. | Sanitation groups: 24–38% of households had stool visible on floor (p < 0.01 compared to control). Observed safe disposal of human feces using sani-scoop was moderate (30–38% of events, p > 0.05). Observed child defecation in potty or latrine (37–54% of events, 0.28 < p < 0.81) Control group: 62% had stool visible on floor. Observed safe disposal of human feces (16% of events). Observed child defecation in potty or latrine (32% of events). |
| Rahman et al. (2018) | Bangladesh | Those enrolled in the WASH Benefits Bangladesh trial | 4,169 enrolled households | 6 months | Hardware and Behavioral: Combined water quality, sanitation, handwashing, and child nutrition interventions. Households received free enabling technologies integrated with behavior change promotion. | Monthly household surveys and spot-checks | Intervention fidelity | CFD into latrine met target benchmark (65% of households) at each monthly check. |
| Huda et al. (2012) | Rural Bangladesh | Intervention group: Households enrolled in the Sanitation, Hygiene Education and Water Supply in the Bangladesh program (SHEWA-B). Control: data collection only. | SHEWA-B targeted 20 million rural people. Huda et al. selected 1,000 households for structured observations. | 24 months | Hardware and Behavioral: SHEWA-B targeted improvements in hygiene behaviors while ensuring adequate sanitation and safe water supply. The program engaged local residents to design their own community action plans, including targets for improvements in latrine coverage and usage; access to and use of arsenic-free water; and improved hygiene practices, especially handwashing with soap. Community promoters visited households, facilitated courtyard meetings, and organized social mobilization activities. | Structured observation of CFD of all persons in household (monthly observations for 24 months). | Observed CFD | Intervention group: At baseline, 9.3% of households practiced SCFD, compared to 16% at follow-up (p = 0.84 compared to control). Control group: At baseline, 9% of households practiced SCFD, compared to 15% at follow-up. |
| Sultana et al. (2013) | Three villages of Faridpur district in Bangladesh | Households with children <3 years for intervention, and women were between 20 and 35 years. | 44 women/mothers | 1 week | Hardware: Mini-hoe and mini-shovel designed based on participants’ feedback, used to dispose of child feces. | In-depth group and individual interviews | Perceptions of new tools for safe CFD | Mini-hoe was favored over other tools for ease of disposal |
| Hussain et al. (2017) | Rural subdistrict of Kishoreganj, Bangladesh | Caregivers from households in an easily accessible compound and with children 7–36 months of age | 26 households, 28 children total | 30 days | Hardware and Behavioral: three potty design options were provided to each of the 26 households. Group discussions and individual interviews were conducted throughout a 30-day period to gather data on potty likes and dislikes, feces disposal practices, barriers to use, and recommendations for improvement. | Group and individual interviews | Proportion of households practicing safe and unsafe feces disposal | At baseline: 4% of households practice safe feces disposal, 96% unsafe disposal. Post-intervention: 85% safe feces disposal and 15% unsafe |
| Biswas et al. (2021) | Urban Dhaka, Bangladesh | Caregivers of children <5 in slum areas that were previously enrolled in CHoBI7 mHealth RCT | Phase 1: 50 households (34 treatments and 16 control) Phase 2: 20 households (15 intervention and 5 control) Phase 3: 32 households that had a diarrhea patient (24 receive modified Baby WASH program and 8 control) | July 2018–Dec. 2019 (18 months) | Behavioral: Three-phase modified Baby WASH program with voice and text messaging to promote SCFD, home visits, and cue cards with recommended behaviors. | Exploratory interviews, intervention development through mobile health workshops, pilot studies (three phases) of baby WASH modules. | Practices, perceptions, and barriers related to CFD. | Majority of young children's feces disposed of unsafely, while older children tended to defecate directly into latrine. Text and voice messages promoting SCFD were well received by participants, though some were unable to answer voice calls due to being busy, and some cell phones did not support Bangla script used in text messaging. Participants recommended adding pictures in addition to text. |
| Ashraf et al. (2012) | Rural Bangladesh | Rural households having at least one child under 3 years of age | 104 households at baseline; 96 households at follow-up | 4 months | Hardware and Behavioral: Potties and a customized hoe-like tool (sani-scoop) to dispose of child and animal feces. Community Health Promoters visited households three times during study and used cue cards, images and text messages to encourage sanitation use. | Household level survey at baseline and survey | Baseline and follow-up comparison of CFD practices | Safe CFD was 25% at baseline compared to 73% after 3 months |
| Patil et al. (2014) | Rural Madhya Pradesh, India | Intervention group: Households that had participated in India's Total Sanitation Campaign with at least one child <24 months of age at enrollment Control group: data collection only | 25 households from each of the 80 study villages | Follow-up assessment at 21 months post-implementation | Hardware: Total Sanitation Campaign is a subsidy-based intervention to provide latrines to households below the poverty line. Included several features including flexible technology options for toilets. | Interviews with caregiver-reported safe disposal of feces and presence of visible feces around household | Reported correct disposal of child feces, reported daily open defecation by children, and observed feces in living area around household | Intervention group: At baseline, 16% of households reported SCFD and interviewer observed feces in the living area of 59% of households. At 21 months, 27% of households reported SCFD (p < 0.001 compared to the control group), and interviewer observed feces in 60% of households (p > 0.01). At 21 months, daily open defecation by children was reported by 84% of households (p < 0.05 compared to control) Control group: At baseline, 13% of households reported SCFD and interviewer observed feces in the living area of 62% of households. At 21 months, 18% of households reported SCFD, and interviewer observed feces in 60% of households. At 21 months, daily open defecation by children was reported by 89% of households. |
| Reese et al. (2019) and Bauza et al. (2019) | Odisha, India | Intervention group: 45 villages that had experienced MANTRA intervention 5 years before; households had children <5 years old | 90 villages: 45 intervention and 45 controls | Intervention implemented 5 years ago | Hardware: On-site piped water, and full community sanitation intervention with household latrines and enclosed bathing rooms connected to water supply | Stool samples collected; self-reported outcomes | Comparison with control group of CFD practices | Reese et al. (2019): Combined intervention of piped water connection with community sanitation can decrease open defecation. SCFD was 35% in intervention villages compared to 6% in control villages (p <0.001). Bauza et al. (2019): Feces from children less than 3 years of age was commonly picked up by caregivers but disposed of unsafely with garbage into open areas. Most children's feces that were safely disposed of in a toilet was because of children defecating in the toilet directly. Intervention households without improved sanitation (377): 1.6% SCFD. Intervention households with improved sanitation (2,124): 40.7% SCFD. Control households without improved sanitation (2,243): 0.2% SCFD. Control households with improved sanitation (605): 36.5% SCFD. Overall intervention households: 34.8% SCFD. Overall control households: 7.9% SCFD |
| Freeman et al. (2016) | Odisha, India | Child <4 years of age or women in the third trimester of pregnancy in the house | 1,958 households selected from 100 villages (50 intervention and 50 control) | 3 years | Hardware: Total Sanitation Campaign is a subsidy-based intervention to provide latrines to households below the poverty line. Households constructed their own pour-flush pit latrines. | Household level survey | Baseline and follow-up comparison of CFD practices | Safe disposal was directly related to increases in latrine presence. At baseline, 1.1% of households practiced SCFD either disposing it in a toilet or by burial. The intervention increased SCFD to 10.4% in intervention households, compared to 3.1% in the control households (RR 3.34; 95% CI 1.99–5.59). |
| Majorin et al. (2014) | Odisha, India | Households with at least one child under 5 years old in one of the 20 selected villages where the Total Sanitation Campaign had been implemented. | 136 households and 145 total children | June–July 2012 | Hardware: Total Sanitation Campaign is a subsidy-based intervention to provide latrines to households below the poverty line. Reported building 1 latrine per 10 rural people from 2001 to 2011. | Structured survey, spot-checks of household latrines for indicators of use and presence of human stools in the compound | Defecation sites and feces disposal sites of children under 5 | 81.4% of child feces disposed of unsafely and 18.6% disposed of safely following Total Sanitation Campaign. After restricting the analysis to households owning a latrine, the use of a nappy or potty was associated with safe disposal of feces (OR 6.72, 95% CI 1.02–44.38) though due to small sample size the regression could not adjust for confounders. |
| Caruso et al. (2022) | Rural Odisha, India | Villages where the Total Sanitation Campaign had been previously implemented that had not been declared open defecation free by the Government of India with households with at least one child <5 years. | 3,723 households in the intervention group, and 1,916 households in the control group | July 2017–March 2019 | Hardware and Behavioral: The intervention was required to meet a limit of costing less than US$20 per household and included a folk performance, transect walk, community meeting, recognition banners, community wall painting, mothers’ meetings, household visits, and latrine repairs. Potties and plastic scoops were also given to mothers. | Village-level activities (pre-intervention community visits, performances, community meetings, mothers meetings) and household visits | Latrine use and CFD practices | Intervention group: Latrine use among children was 15.4% at baseline compared to 41.8% at endline. SCFD was 6% at baseline compared to 33.7% at endline. Control group: Latrine among children was 18.2% at baseline compared to 37.4% at endline. SCFD was 3.4% at baseline compared to 10.6% at endline. |
| Lamichhane et al. (2018) | Rural Nepal | Households with a female 15–49 years of age and a child of <5 years | 3,377 children | February–June 2011 (4 months) | Hardware: Improved sanitation was previously provided to households without any behavior change component. Households with improved sanitation and SCFD were matched to households with improved sanitation and unsafe CFD to assess for differences in the prevalence of diarrhea. | Used NDHS data – a nationally representative survey on access to sanitation, CFD practices, and other socioeconomic and demographic factors. | Proportions of households with improved and unimproved sanitation; proportions of safe and unsafe CFD; prevalence of diarrhea | 31.9% of households had improved sanitation – of which 35.7% still practiced unsafe CFD, while 64.3% practiced SCFD. Improved sanitation with safe disposal was significantly associated with a lower prevalence of diarrhea among children <5 years of age by 3.3 (standard error [SE] 0.016) to 6.6 (SE 0.023) percentage points under different methods. |
| Authors . | Location . | Age group/inclusion criteria . | Sample size . | Duration of study . | Intervention type: intervention description . | Data collection methods . | Outcome measures . | Impacts on CFD practices . |
|---|---|---|---|---|---|---|---|---|
| Luby et al. (2018) | Rural Bangladesh | Intervention group: Households that had participated in the WASH Benefits Bangladesh trial with pregnant women assigned to the following groups: chlorinated drinking water (water); upgraded sanitation (sanitation); promotion of handwashing with soap (handwashing); combined water, sanitation, and handwashing; counselling on appropriate child nutrition (nutrition); combined water, sanitation, handwashing, and nutrition. Control group: data collection only. | 5,551 pregnant women in intervention groups in the WASH Benefits trial; 1,382 women in the control group. | Evaluated WASH Benefits Bangladesh outcomes at 1- and 2-years post-intervention (intervention delivered over 15 months). | Hardware and Behavioral: WASH Benefits Bangladesh was a cluster-randomized trial that enrolled pregnant women and evaluated outcomes at 1- and 2-years follow-up. All households in groups with sanitation component received latrine installation or repairment, a sani-scoop for removing feces from the compound, and households with children <3 received child potties. Promoters encouraged mothers to teach their children to use the potties, to safely dispose of feces in latrines, and to regularly remove animal and human feces from the compound. | Caregiver-reported data, and the presence of feces on latrine slab or floor. | Comparison to control and baseline % of latrines with functional water seals and the presence of visible feces on latrine slab or floor. | Sanitation groups: 48–56% of households had visible feces on floor at baseline compared to 11–14% at 1 year and 14–18% at 2 years. 26–30% had latrine water seals at baseline compared to 95 and 97% at 1 and 2 years. Control group: 52% of households had visible feces on floor at baseline compared to 40% at 1 year and 44% at 2 years. 31% had functional latrine water seals at baseline compared to 29 and 31% at 1 and 2 years. |
| Parvez et al. (2018) | Rural Bangladesh | Intervention group: Households that had participated in the WASH Benefits Bangladesh trial with pregnant women assigned to the following groups: chlorinated drinking water (water); upgraded sanitation (sanitation); promotion of handwashing with soap (handwashing); combined water, sanitation, and handwashing; counselling on appropriate child nutrition (nutrition); combined water, sanitation, handwashing, and nutrition. Control group: data collection only. | 5,551 pregnant women in intervention groups; 1,382 women in the control group. | Fidelity assessments over 20 months from November 2012 to October 2014. | Hardware and Behavioral: WASH Benefits Bangladesh was a cluster-randomized trial that enrolled pregnant women and evaluated outcomes at 1- and 2-year follow-up. All households in groups with sanitation component received latrine installation or repairment, a sani-scoop for removing feces from the compound, and households with children <3 received child potties. Promoters encouraged mothers to teach their children to use the potties, to safely dispose of feces in latrines, and to regularly remove animal and human feces from the compound. | Implementation fidelity and structured observations. Monthly spot-checks and surveys for technology and behavioral uptake (feces presence in the courtyard as an indicator of sani-scoop use and safe feces disposal). | Observed hygienic latrine, feces presence in the courtyard as an indicator of sani-scoop use and safe feces disposal. | Sanitation groups: 24–38% of households had stool visible on floor (p < 0.01 compared to control). Observed safe disposal of human feces using sani-scoop was moderate (30–38% of events, p > 0.05). Observed child defecation in potty or latrine (37–54% of events, 0.28 < p < 0.81) Control group: 62% had stool visible on floor. Observed safe disposal of human feces (16% of events). Observed child defecation in potty or latrine (32% of events). |
| Rahman et al. (2018) | Bangladesh | Those enrolled in the WASH Benefits Bangladesh trial | 4,169 enrolled households | 6 months | Hardware and Behavioral: Combined water quality, sanitation, handwashing, and child nutrition interventions. Households received free enabling technologies integrated with behavior change promotion. | Monthly household surveys and spot-checks | Intervention fidelity | CFD into latrine met target benchmark (65% of households) at each monthly check. |
| Huda et al. (2012) | Rural Bangladesh | Intervention group: Households enrolled in the Sanitation, Hygiene Education and Water Supply in the Bangladesh program (SHEWA-B). Control: data collection only. | SHEWA-B targeted 20 million rural people. Huda et al. selected 1,000 households for structured observations. | 24 months | Hardware and Behavioral: SHEWA-B targeted improvements in hygiene behaviors while ensuring adequate sanitation and safe water supply. The program engaged local residents to design their own community action plans, including targets for improvements in latrine coverage and usage; access to and use of arsenic-free water; and improved hygiene practices, especially handwashing with soap. Community promoters visited households, facilitated courtyard meetings, and organized social mobilization activities. | Structured observation of CFD of all persons in household (monthly observations for 24 months). | Observed CFD | Intervention group: At baseline, 9.3% of households practiced SCFD, compared to 16% at follow-up (p = 0.84 compared to control). Control group: At baseline, 9% of households practiced SCFD, compared to 15% at follow-up. |
| Sultana et al. (2013) | Three villages of Faridpur district in Bangladesh | Households with children <3 years for intervention, and women were between 20 and 35 years. | 44 women/mothers | 1 week | Hardware: Mini-hoe and mini-shovel designed based on participants’ feedback, used to dispose of child feces. | In-depth group and individual interviews | Perceptions of new tools for safe CFD | Mini-hoe was favored over other tools for ease of disposal |
| Hussain et al. (2017) | Rural subdistrict of Kishoreganj, Bangladesh | Caregivers from households in an easily accessible compound and with children 7–36 months of age | 26 households, 28 children total | 30 days | Hardware and Behavioral: three potty design options were provided to each of the 26 households. Group discussions and individual interviews were conducted throughout a 30-day period to gather data on potty likes and dislikes, feces disposal practices, barriers to use, and recommendations for improvement. | Group and individual interviews | Proportion of households practicing safe and unsafe feces disposal | At baseline: 4% of households practice safe feces disposal, 96% unsafe disposal. Post-intervention: 85% safe feces disposal and 15% unsafe |
| Biswas et al. (2021) | Urban Dhaka, Bangladesh | Caregivers of children <5 in slum areas that were previously enrolled in CHoBI7 mHealth RCT | Phase 1: 50 households (34 treatments and 16 control) Phase 2: 20 households (15 intervention and 5 control) Phase 3: 32 households that had a diarrhea patient (24 receive modified Baby WASH program and 8 control) | July 2018–Dec. 2019 (18 months) | Behavioral: Three-phase modified Baby WASH program with voice and text messaging to promote SCFD, home visits, and cue cards with recommended behaviors. | Exploratory interviews, intervention development through mobile health workshops, pilot studies (three phases) of baby WASH modules. | Practices, perceptions, and barriers related to CFD. | Majority of young children's feces disposed of unsafely, while older children tended to defecate directly into latrine. Text and voice messages promoting SCFD were well received by participants, though some were unable to answer voice calls due to being busy, and some cell phones did not support Bangla script used in text messaging. Participants recommended adding pictures in addition to text. |
| Ashraf et al. (2012) | Rural Bangladesh | Rural households having at least one child under 3 years of age | 104 households at baseline; 96 households at follow-up | 4 months | Hardware and Behavioral: Potties and a customized hoe-like tool (sani-scoop) to dispose of child and animal feces. Community Health Promoters visited households three times during study and used cue cards, images and text messages to encourage sanitation use. | Household level survey at baseline and survey | Baseline and follow-up comparison of CFD practices | Safe CFD was 25% at baseline compared to 73% after 3 months |
| Patil et al. (2014) | Rural Madhya Pradesh, India | Intervention group: Households that had participated in India's Total Sanitation Campaign with at least one child <24 months of age at enrollment Control group: data collection only | 25 households from each of the 80 study villages | Follow-up assessment at 21 months post-implementation | Hardware: Total Sanitation Campaign is a subsidy-based intervention to provide latrines to households below the poverty line. Included several features including flexible technology options for toilets. | Interviews with caregiver-reported safe disposal of feces and presence of visible feces around household | Reported correct disposal of child feces, reported daily open defecation by children, and observed feces in living area around household | Intervention group: At baseline, 16% of households reported SCFD and interviewer observed feces in the living area of 59% of households. At 21 months, 27% of households reported SCFD (p < 0.001 compared to the control group), and interviewer observed feces in 60% of households (p > 0.01). At 21 months, daily open defecation by children was reported by 84% of households (p < 0.05 compared to control) Control group: At baseline, 13% of households reported SCFD and interviewer observed feces in the living area of 62% of households. At 21 months, 18% of households reported SCFD, and interviewer observed feces in 60% of households. At 21 months, daily open defecation by children was reported by 89% of households. |
| Reese et al. (2019) and Bauza et al. (2019) | Odisha, India | Intervention group: 45 villages that had experienced MANTRA intervention 5 years before; households had children <5 years old | 90 villages: 45 intervention and 45 controls | Intervention implemented 5 years ago | Hardware: On-site piped water, and full community sanitation intervention with household latrines and enclosed bathing rooms connected to water supply | Stool samples collected; self-reported outcomes | Comparison with control group of CFD practices | Reese et al. (2019): Combined intervention of piped water connection with community sanitation can decrease open defecation. SCFD was 35% in intervention villages compared to 6% in control villages (p <0.001). Bauza et al. (2019): Feces from children less than 3 years of age was commonly picked up by caregivers but disposed of unsafely with garbage into open areas. Most children's feces that were safely disposed of in a toilet was because of children defecating in the toilet directly. Intervention households without improved sanitation (377): 1.6% SCFD. Intervention households with improved sanitation (2,124): 40.7% SCFD. Control households without improved sanitation (2,243): 0.2% SCFD. Control households with improved sanitation (605): 36.5% SCFD. Overall intervention households: 34.8% SCFD. Overall control households: 7.9% SCFD |
| Freeman et al. (2016) | Odisha, India | Child <4 years of age or women in the third trimester of pregnancy in the house | 1,958 households selected from 100 villages (50 intervention and 50 control) | 3 years | Hardware: Total Sanitation Campaign is a subsidy-based intervention to provide latrines to households below the poverty line. Households constructed their own pour-flush pit latrines. | Household level survey | Baseline and follow-up comparison of CFD practices | Safe disposal was directly related to increases in latrine presence. At baseline, 1.1% of households practiced SCFD either disposing it in a toilet or by burial. The intervention increased SCFD to 10.4% in intervention households, compared to 3.1% in the control households (RR 3.34; 95% CI 1.99–5.59). |
| Majorin et al. (2014) | Odisha, India | Households with at least one child under 5 years old in one of the 20 selected villages where the Total Sanitation Campaign had been implemented. | 136 households and 145 total children | June–July 2012 | Hardware: Total Sanitation Campaign is a subsidy-based intervention to provide latrines to households below the poverty line. Reported building 1 latrine per 10 rural people from 2001 to 2011. | Structured survey, spot-checks of household latrines for indicators of use and presence of human stools in the compound | Defecation sites and feces disposal sites of children under 5 | 81.4% of child feces disposed of unsafely and 18.6% disposed of safely following Total Sanitation Campaign. After restricting the analysis to households owning a latrine, the use of a nappy or potty was associated with safe disposal of feces (OR 6.72, 95% CI 1.02–44.38) though due to small sample size the regression could not adjust for confounders. |
| Caruso et al. (2022) | Rural Odisha, India | Villages where the Total Sanitation Campaign had been previously implemented that had not been declared open defecation free by the Government of India with households with at least one child <5 years. | 3,723 households in the intervention group, and 1,916 households in the control group | July 2017–March 2019 | Hardware and Behavioral: The intervention was required to meet a limit of costing less than US$20 per household and included a folk performance, transect walk, community meeting, recognition banners, community wall painting, mothers’ meetings, household visits, and latrine repairs. Potties and plastic scoops were also given to mothers. | Village-level activities (pre-intervention community visits, performances, community meetings, mothers meetings) and household visits | Latrine use and CFD practices | Intervention group: Latrine use among children was 15.4% at baseline compared to 41.8% at endline. SCFD was 6% at baseline compared to 33.7% at endline. Control group: Latrine among children was 18.2% at baseline compared to 37.4% at endline. SCFD was 3.4% at baseline compared to 10.6% at endline. |
| Lamichhane et al. (2018) | Rural Nepal | Households with a female 15–49 years of age and a child of <5 years | 3,377 children | February–June 2011 (4 months) | Hardware: Improved sanitation was previously provided to households without any behavior change component. Households with improved sanitation and SCFD were matched to households with improved sanitation and unsafe CFD to assess for differences in the prevalence of diarrhea. | Used NDHS data – a nationally representative survey on access to sanitation, CFD practices, and other socioeconomic and demographic factors. | Proportions of households with improved and unimproved sanitation; proportions of safe and unsafe CFD; prevalence of diarrhea | 31.9% of households had improved sanitation – of which 35.7% still practiced unsafe CFD, while 64.3% practiced SCFD. Improved sanitation with safe disposal was significantly associated with a lower prevalence of diarrhea among children <5 years of age by 3.3 (standard error [SE] 0.016) to 6.6 (SE 0.023) percentage points under different methods. |
Quality assessment of studies included in review of interventions targeting CFD practices in Asian-Pacific countries
| Authors . | Issue or potential intervention . | Study design . | Hardware or behavioral intervention . | Levela . | Risk of biasb . | Overall certainty ratingc . | Impact on safe CFD (SCFD) . |
|---|---|---|---|---|---|---|---|
| Luby et al. (2018) | WASH Benefits Bangladesh trial | RCT | Both | A | Low (large sample size, use of random number generator for randomization, geographical matching, all outcomes reported) | High | 11–18% of intervention households had visible feces compared to 40–44% in control households |
| Parvez et al. (2018) | WASH Benefits Bangladesh trial | RCT | Both | A | Low (large sample size, observers had no connection to intervention, all outcomes reported) | High | Observed safe disposal of human feces using sani-scoop was moderate at 30–38% (p > 0.05) Observed child defecation in potty or latrine was 37–54% of events (0.28 < p < 0.81) |
| Rahman et al. (2018) | WASH Benefits Bangladesh trial | RCT | Both | A | Intermediate-High (no control arm for fidelity checks, qualitative investigation) | Intermediate | Qualitative assessment; SCFD met target benchmark of 65% at each fidelity check |
| Huda et al. (2012) | Sanitation, Hygiene Education and Water Supply in Bangladesh program (SHEWA-B) | RCT | Both | A | Low (community monitors not aware of hypothesis, calculated sample size for 80% power and 95% confidence) | High | At baseline, 9.3% of households practiced SCFD, compared to 16% at follow-up (p = 0.84 compared to control) |
| Sultana et al. (2013) | Mini-hoe and mini-shovel tools for feces removal | Pilot study | Hardware | D | High (small sample, potential for courtesy bias) | Low | Qualitative only |
| Hussain et al. (2017) | Three potty design options for children | Formative research | Both | D | High (small sample, potential for courtesy bias, self-reported CFD) | Low | At baseline: 4% SCFD. Post-intervention: 85% SCFD |
| Biswas et al. (2021) | Baby WASH program | Formative research | Behavioral | D | High (convenience sampling, small sample) | Low | Qualitative only |
| Ashraf et al. (2012) | Potties and a customized hoe-like tool (sani-scoop) to dispose of child and animal feces | Prospective Cohort | Both | B | High (self-reported CFD, lack of follow-up) | Low | SCFD was 25% at baseline compared to 73% after 3 months (p < 0.001) |
| Patil et al. (2014) | Total Sanitation Campaign | RCT | Hardware | A | Low (randomization via lottery, included other outcome indicators in household surveys to mitigate reporting bias) | High | SCFD increased from 16 to 27% in the intervention group (p < 0.001) |
| Reese et al. (2019) | Movement and Action Network for the Transformation of Rural Areas (MANTRA program) | Matched cohort study | Hardware | B | Low (large samples, multivariate matching to achieve balance across cohorts, included other outcome indicators in household surveys to mitigate reporting bias, adjusted for confounding) | High | SCFD was 35% in intervention villages compared to 6% in control villages (p < 0.001) |
| Bauza et al. (2019) | Movement and Action Network for the Transformation of Rural Areas (MANTRA program) | Retrospective cohort study | Hardware | B | Low (large sample, accounted for potential confounders and clustering) | High | Intervention households without improved sanitation (377): 1.6% SCFD. Intervention households with improved sanitation (2,124): 40.7% SCFD. Control households without improved sanitation (2,243): 0.2% SCFD. Control households with improved sanitation (605): 36.5% SCFD. Overall intervention households: 34.8% SCFD. Overall control households: 7.9% SCFD |
| Freeman et al. (2016) | Total Sanitation Campaign | Randomized control trial | Hardware | A | Intermediate (large sample, but self-reported CFD, only one data collection point) | High | SCFD was 10.4% in intervention households compared to 3.1% in control households (RR 3.34, 95% CI [1.99–5.59]). |
| Majorin et al. (2014) | Total Sanitation Campaign | Cross-sectional study | Hardware | B | Intermediate (self-reported CFD subject to courtesy and recall bias) | Intermediate | 18.6% of intervention households practiced SCFD following Total Sanitation Campaign. Among households owning a latrine, the use of a nappy or potty was associated with safe disposal of feces (OR 6.72, 95% CI [1.02–44.38]). |
| Caruso et al. (2022) | Total Sanitation Campaign | RCT | Hardware | A | Low (large samples, similar pattern of missingness in both study groups and within covariate strata, results of sensitivity analyses showed no evidence of bias in effect estimates) | High | SCFD improved by 20 percentage points in the intervention group compared to the control group |
| Lamichhane et al. (2018) | Improved sanitation access | Quasi-experimental | Hardware | B | Intermediate (large samples, but chance of social desirability bias and use of propensity score matching to estimate treatment effects of improved sanitation with and without safe disposal may increase imbalance) | Intermediate | Of households with improved sanitation, 64% practiced SCFD and 36% unsafe CFD. Only 31.9% of households had improved sanitation access |
| Authors . | Issue or potential intervention . | Study design . | Hardware or behavioral intervention . | Levela . | Risk of biasb . | Overall certainty ratingc . | Impact on safe CFD (SCFD) . |
|---|---|---|---|---|---|---|---|
| Luby et al. (2018) | WASH Benefits Bangladesh trial | RCT | Both | A | Low (large sample size, use of random number generator for randomization, geographical matching, all outcomes reported) | High | 11–18% of intervention households had visible feces compared to 40–44% in control households |
| Parvez et al. (2018) | WASH Benefits Bangladesh trial | RCT | Both | A | Low (large sample size, observers had no connection to intervention, all outcomes reported) | High | Observed safe disposal of human feces using sani-scoop was moderate at 30–38% (p > 0.05) Observed child defecation in potty or latrine was 37–54% of events (0.28 < p < 0.81) |
| Rahman et al. (2018) | WASH Benefits Bangladesh trial | RCT | Both | A | Intermediate-High (no control arm for fidelity checks, qualitative investigation) | Intermediate | Qualitative assessment; SCFD met target benchmark of 65% at each fidelity check |
| Huda et al. (2012) | Sanitation, Hygiene Education and Water Supply in Bangladesh program (SHEWA-B) | RCT | Both | A | Low (community monitors not aware of hypothesis, calculated sample size for 80% power and 95% confidence) | High | At baseline, 9.3% of households practiced SCFD, compared to 16% at follow-up (p = 0.84 compared to control) |
| Sultana et al. (2013) | Mini-hoe and mini-shovel tools for feces removal | Pilot study | Hardware | D | High (small sample, potential for courtesy bias) | Low | Qualitative only |
| Hussain et al. (2017) | Three potty design options for children | Formative research | Both | D | High (small sample, potential for courtesy bias, self-reported CFD) | Low | At baseline: 4% SCFD. Post-intervention: 85% SCFD |
| Biswas et al. (2021) | Baby WASH program | Formative research | Behavioral | D | High (convenience sampling, small sample) | Low | Qualitative only |
| Ashraf et al. (2012) | Potties and a customized hoe-like tool (sani-scoop) to dispose of child and animal feces | Prospective Cohort | Both | B | High (self-reported CFD, lack of follow-up) | Low | SCFD was 25% at baseline compared to 73% after 3 months (p < 0.001) |
| Patil et al. (2014) | Total Sanitation Campaign | RCT | Hardware | A | Low (randomization via lottery, included other outcome indicators in household surveys to mitigate reporting bias) | High | SCFD increased from 16 to 27% in the intervention group (p < 0.001) |
| Reese et al. (2019) | Movement and Action Network for the Transformation of Rural Areas (MANTRA program) | Matched cohort study | Hardware | B | Low (large samples, multivariate matching to achieve balance across cohorts, included other outcome indicators in household surveys to mitigate reporting bias, adjusted for confounding) | High | SCFD was 35% in intervention villages compared to 6% in control villages (p < 0.001) |
| Bauza et al. (2019) | Movement and Action Network for the Transformation of Rural Areas (MANTRA program) | Retrospective cohort study | Hardware | B | Low (large sample, accounted for potential confounders and clustering) | High | Intervention households without improved sanitation (377): 1.6% SCFD. Intervention households with improved sanitation (2,124): 40.7% SCFD. Control households without improved sanitation (2,243): 0.2% SCFD. Control households with improved sanitation (605): 36.5% SCFD. Overall intervention households: 34.8% SCFD. Overall control households: 7.9% SCFD |
| Freeman et al. (2016) | Total Sanitation Campaign | Randomized control trial | Hardware | A | Intermediate (large sample, but self-reported CFD, only one data collection point) | High | SCFD was 10.4% in intervention households compared to 3.1% in control households (RR 3.34, 95% CI [1.99–5.59]). |
| Majorin et al. (2014) | Total Sanitation Campaign | Cross-sectional study | Hardware | B | Intermediate (self-reported CFD subject to courtesy and recall bias) | Intermediate | 18.6% of intervention households practiced SCFD following Total Sanitation Campaign. Among households owning a latrine, the use of a nappy or potty was associated with safe disposal of feces (OR 6.72, 95% CI [1.02–44.38]). |
| Caruso et al. (2022) | Total Sanitation Campaign | RCT | Hardware | A | Low (large samples, similar pattern of missingness in both study groups and within covariate strata, results of sensitivity analyses showed no evidence of bias in effect estimates) | High | SCFD improved by 20 percentage points in the intervention group compared to the control group |
| Lamichhane et al. (2018) | Improved sanitation access | Quasi-experimental | Hardware | B | Intermediate (large samples, but chance of social desirability bias and use of propensity score matching to estimate treatment effects of improved sanitation with and without safe disposal may increase imbalance) | Intermediate | Of households with improved sanitation, 64% practiced SCFD and 36% unsafe CFD. Only 31.9% of households had improved sanitation access |
aLevel A represents RCTs, Level B represents peer-reviewed research studies (cross-sectional studies, quasi-experimental studies, cohort studies), and Level D represents pilot studies or formative research.
bLow risk of bias indicates no limitations that could compromise study findings. Intermediate risk of bias includes studies with minor limitations that would not compromise study findings. A high risk of bias indicates studies with several limitations that may compromise study findings.
cOverall certainty ratings imply confidence that the true effect lies close to the estimated effect determined in the study. Studies at Level A were initially given a high certainty, studies at level B were rated intermediate, and studies at Level D were initially rated as low. Overall certainty was then moved either higher or lower based on risk of bias estimates.