formative research to scale up a handwashing with soap and … · 2020. 6. 2. · research article...

RESEARCH ARTICLE Open Access

Formative research to scale up ahandwashing with soap and watertreatment intervention for householdmembers of diarrhea patients in healthfacilities in Dhaka, Bangladesh (CHoBI7program)Elizabeth D. Thomas1, Fatema Zohura2, M. Tasdik Hasan2, Md. Sohel Rana3, Alana Teman1, Tahmina Parvin2,Jahed Masud2, Md. Sazzadul Islam Bhuyian2, Md. Khobair Hossain2, Maynul Hasan2, Sanya Tahmina4,Farzana Munmun5, Md. Abul Hashem Khan5, Shirajum Monira2, David A. Sack1, Elli Leontsini1, Peter J. Winch1,Munirul Alam2 and Christine Marie George6*

Abstract

Background: During the time a diarrhea patient presents at a health facility, the household members of thepatient are at higher risk of developing diarrheal diseases (> 100 times for cholera) than the general population. TheCholera-Hospital-based-Intervention-for-7-Days (CHoBI7) is a health facility-initiated water treatment andhandwashing with soap intervention designed to reduce transmission of diarrheal diseases between patients andtheir household members. The present research aimed to (1) develop a scalable approach to integrate the CHoBI7intervention program into services provided at government and private health facilities in Bangladesh; and (2) tailorthe intervention program for the household members of all diarrhea patients, irrespective of the etiology of disease.

Methods: We conducted 8 months of formative research, including 60 semi-structured interviews, 2 groupdiscussions, and a pilot study. Thirty-two interviews were conducted with diarrhea patients and their familycaregivers, government stakeholders, and health care providers both to explore existing WASH and diarrhea patientcare practices in health facilities and to identify considerations for scaling the CHoBI7 program. Fifty-two diarrheapatient households participated in a pilot study of a modified version of the CHoBI7 intervention program fortailoring. Twenty-eight interviews and 2 group discussions were conducted with pilot households to exploreexperiences with and recommendations for intervention delivery.

(Continued on next page)

© The Author(s). 2020 Open Access This article is licensed under a Creative Commons Attribution 4.0 International License,which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you giveappropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate ifchanges were made. The images or other third party material in this article are included in the article's Creative Commonslicence, unless indicated otherwise in a credit line to the material. If material is not included in the article's Creative Commonslicence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtainpermission directly from the copyright holder. To view a copy of this licence, visit http://creativecommons.org/licenses/by/4.0/.The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to thedata made available in this article, unless otherwise stated in a credit line to the data.

* Correspondence: [email protected] Professor, Department of International Health, Program in GlobalDisease Epidemiology and Control, Johns Hopkins Bloomberg School ofPublic Health, 615 N. Wolfe Street, Room E5535, Baltimore, MD 21205-2103,USAFull list of author information is available at the end of the article

Thomas et al. BMC Public Health (2020) 20:831 https://doi.org/10.1186/s12889-020-08727-0

http://crossmark.crossref.org/dialog/?doi=10.1186/s12889-020-08727-0&domain=pdfhttp://orcid.org/0000-0001-9219-0953http://creativecommons.org/licenses/by/4.0/http://creativecommons.org/publicdomain/zero/1.0/mailto:[email protected]

(Continued from previous page)

Results: The intervention program was modified based on formative research findings. Pilot study participantsrecognized the benefits of the CHoBI7 intervention program and made suggestions to improve the acceptabilityand feasibility of the intervention. Modifications included 1) providing additional pictorial modules, cues to action,enabling technologies, and supplies for safe drinking water and handwashing with soap behaviors in the healthfacility; 2) switching out technology prone to breaks and leaks as well as sourcing plastic technologies from a high-quality, local manufacturer; and 3) including instructions discouraging the non-use or misuse of technologies andsupplies. Considerations for scalability include the local availability and marketing of enabling technologies andsupplies, staff for program delivery in health facilities, and potential integration into existing government or healthpromotion programs.

Conclusions: Formative research identified important considerations for the content, delivery, and scalability of theCHoBI7 health facility-initiated WASH intervention program.

Keywords: Diarrhea, Handwashing, Water treatment, Behavior change, Health facilities, Qualitative research,Formative research, Bangladesh

BackgroundDiarrheal disease remains a leading cause of child deathand a major contributor to morbidity across all agegroups [1, 2]. During the time a diarrhea patient pre-sents at a health facility for treatment, the householdmembers of the patient are at much higher risk of devel-oping diarrheal diseases (> 100 times for cholera) thanthe general population [3–7]. This higher risk is likelydue to shared, contaminated environmental sources, likehousehold drinking water, and secondary transmissionfrom poor hygiene practices in both the home andhealth facility, where family members care for patients[8–10]. Access to a safe domestic water source [9] andhandwashing with soap [11] have been shown to reduceenteric diseases. However, acceptable and effective inter-ventions tailored to meet the needs of diarrhea patientsand their household members, both at a health facilityand when they return home, are limited. Health facility-based WASH interventions have been developed in vari-ous settings, though few interventions have reported onformative research conducted for intervention develop-ment [12–15].

The Cholera-Hospital-based-Intervention-for-7-Days(CHoBI7)The original Cholera-Hospital-based-Intervention-for-7-Days (CHoBI7) intervention program was designed forcholera patients and their household members [16]. Totarget a period of high susceptibility to diarrheal disease[3, 17, 18], the intervention was delivered by health pro-moters during the seven-day period after the patient wasadmitted to a health facility for treatment—the high-riskperiod for the transmission of cholera and other diar-rheal diseases [3, 5, 6]. The patient’s hospitalizationserves as a disrupting event for the household, creatingan opportunity to introduce new behaviors [19].

Household members are both counseled on how to pre-vent cholera as well as provided with enabling technolo-gies and supplies in order to facilitate behavior change.The original intervention included a pictorial module

(flipbook), delivered at the patient’s bedside, that depictedhow cholera spreads and can be prevented. The CHoBI7moniker is based on the flipbook, as chobi means ‘picture’in Bangla. After the bedside visit, behavioral recommenda-tions were reinforced via daily household visits during theweek following admission to the health facility. Enrolledhouseholds received an imported, sealed drinking watervessel with tap (Topaz™ [Lion Star Plastic, Jakarta,Indonesia]), chlorine tablets for water treatment, a hand-washing station, and a bottle of soapy water (detergentpowder mixed with water). Target behaviors includeddrinking water treatment with chlorine tablets and hand-washing with soap at key times (e.g. after using the toilet)during the 7 days following admission to a health facility.Households were counseled to continue protective WASHbehaviors after the high-risk period in order to prevent fu-ture enteric infections. The original design of the CHoBI7intervention program was informed by the Risk, Attitudes,Norms, Abilities, and Self-regulation (RANAS) model ofbehavior change [20] and the Integrated Behavioural Modelfor Water, Sanitation, and Hygiene (IBM-WASH) [21].A randomized controlled trial (RCT) of the original

CHoBI7 intervention program showed significantly fewersymptomatic cholera infections and a 47% reduction inoverall cholera infections among those in the interventionarm versus the control arm [16]. Furthermore, interven-tion household members had significantly higher instancesof handwashing with soap and significantly less E.coli instored drinking water when compared to control house-holds 6–12months post-intervention delivery [22]. In ourassessment of the underlying mechanism of behaviorchange, response efficacy was associated with habit

Thomas et al. BMC Public Health (2020) 20:831 Page 2 of 19

formation, and disgust, convenience, and cholera aware-ness were associated with habit maintenance [23]; thispoints to the importance of considering behavior changetheory in the development and implementation of WASHinterventions. Interventions informed by behavior changetheories that target multiple behavioral determinants aremore likely to be successful than interventions that pro-vide information alone [24–26].

Setting and study rationaleAt the request of our partners at the Bangladesh Ministryof Health and Family Welfare, we aimed to develop a scal-able approach to integrate the CHoBI7 intervention pro-gram into services provided at government and privatehealth facilities in Bangladesh. We also aimed to tailor theintervention program for the household members of alldiarrhea patients, irrespective of the etiology of disease.To achieve these aims, formative research and the engage-ment of key stakeholders were required in order to modifythe original intervention program. The formative researchaddressed four research questions: (1) What are existingWASH and diarrhea patient care practices in health facil-ities? (2) What are households’ experiences with and ac-ceptability of CHoBI7 WASH enabling technology and

behavioral recommendations? (3) What are considerationsand modifications for taking the intervention program toscale? (4) What modifications to the original CHoBI7intervention program are needed in order to tailor theintervention for the target population?The goal of the present research was to develop the con-

tent and delivery method of a ‘modified’ CHoBI7 interven-tion program that would require only a single in-personvisit in the health facility and could be integrated into ser-vices provided to diarrhea patients in health facilities inBangladesh. Given the objective of developing an interven-tion program that could be taken to scale, a mobile health(mHealth) program was developed to accompany thehealth facility-initiated program. Formative research forthe development of the CHoBI7 mHealth program is re-ported elsewhere [27]. The modified intervention was re-cently evaluated through a three-arm RCT, which tookplace over 12months, that compared (1) the standard rec-ommendation to diarrhea patients about oral rehydrationsolution (ORS) use; (2) health facility initiation of CHoBI7plus mHealth (mHealth with no home visits); and (3)health facility initiation of CHoBI7 plus two home visitsand mHealth (mHealth with home visits) (Fig. 1) (Georgeet al. 2020, submitted).

Fig. 1 Overview of intervention activities


MethodsSemi-structured interviews and a pilot study for the de-velopment of the modified CHoBI7 intervention programwere conducted in parallel. An outline of the formativeresearch activities is provided in Fig. 2.

Data collectionA total of 60 semi-structured interviews were conductedfrom April to November 2016. Intial semi-structured in-terviews were conducted with a convenience sample ofdiarrhea patients (n = 5), their family caregivers (n = 13),and health providers (n = 2) to explore health facilityWASH and patient care practices (Table 1). Diarrhea pa-tients and their family caregivers were approached in thehospital, and the interviews took place at the patient’sbedside. Interviews with health providers were arrangedover the phone and conducted in the provider’s office.The objective of these interviews was to identify prac-tices and preferences that would necessitate modifica-tions to the original CHoBI7 intervention program.Twelve interviews with 10 key stakeholders (1 research

scientist and 9 health-posting government officials) werealso conducted to explore the scalability of the interven-tion (Table 1). One government stakeholder was

interviewed 3 times. Government stakeholders were pur-posefully selected to include individuals serving in healthservices with decision-making power. Appointments forinterviews with stakeholders were made over the phoneor in-person, and the interviews took place in the inter-viewee’s office.From April through November 2016, trained research

assistants recruited a convenience sample of 65diarrhea patients and their household members toparticipate in a pilot study of the modified CHoBI7intervention program for continued modification andrefinement of the intervention content and delivery.For the pilot study, participants were recruited from 2participating health facilities: icddr,b’s hospital (private)and a government hospital in Dhaka, Bangladesh.Households were selected for participation if they (1)had a household member admitted to either health fa-cility with three or more loose stools over a 24-hperiod, (2) had no basin with running water, (3) had achild under 5 years of age living in the home, (4) had atleast one active mobile phone (for mHealth messagedelivery), (5) planned to reside in Dhaka over the fol-lowing year, and (6) were not involved in another watertreatment program. Possession of at least one active

Fig. 2 Workflow of research design and formative research activities


mobile phone was required for participation in the pilotstudy due to the pre-determined inclusion of mHealthas part of the intervention in order to facilitate futurescaling up of the intervention program.Household members of diarrhea patients were eligible

for enrollment if their meals were prepared in the samecooking pot as the diarrhea patient for the 3 days priorto the diarrhea patient’s admission to a health facilityand if they planned to reside with the diarrhea patientfor the next 12 months.Two households refused to participate after initiating

the hospital-based structured observation (activity dis-tinct from the formative research presented here). Elevenhouseholds were dropped post-consent either becausethe study team could not locate the household or the pa-tient left the hospital before receiving the intervention.A total of 52 households completed the pilot study, and38 households were in the CHoBI7 mHealth arms.The pilot study design included the three pre-

determined arms of the planned RCT (outlined above),with households assigned to either the (1) standard rec-ommendation arm; (2) CHoBI7 mHealth with no homevisits arm; or (3) CHoBI7 mHealth with two home visitsarm. Intervention arms also received the standard ORSmessage. While the intervention arms of the pilot studyhelped to further tailor the intervention program, thestandard recommendation arm served as a comparisongroup to inform operability and is separate from the for-mative research presented here.The modified intervention program included a

Health Facility Flipbook about how to prevent

diarrhea transmission, which was delivered by atrained health promoter to all intervention householdsat the patient’s bedside. Households in the mHealthwith home visits arm received additional counselingfrom health promoters using a separate HouseholdFlipbook during two household visits. All interventionhouseholds received a handwashing station (a redbucket with tap and lid, a bowl to catch rinse water,and a stool), a soapy water bottle (500 mL bottle, witha pierced lid for dispensing liquid, with detergentpowder dissolved in water inside), a safe water storagebucket (a blue bucket with tap and lid and a stool),and a 30-day supply of chlorine tablets for watertreatment (Fig. 3). In the mHealth with no homevisits arm, the hardware was provided in the hospital;the mHealth with home visits arm received hardwareduring the first home visit.Of the 38 households enrolled in the pilot study

CHoBI7 intervention arms, 27 households were visitedfor a follow-up interview (median 6.5 days betweenenrollment and follow-up) to learn about households’acceptability of and experiences with the interven-tion’s enabling technologies and behavioral recom-mendations, and to seek advice for how to improvethe intervention program. One household was visitedtwice, for a total of 28 pilot study interviews (Table 1).Pilot study participants were called consultants forthe CHoBI7 intervention program and were toldabout their role in an effort to reduce courtesy bias.Follow-up interviews were conducted until we reachedsaturation.

Table 1 Formative research activities

Diarrhea patient, family caregiver, and health provider interviews

• Explore WASH and patient care practices in healthfacilities to assist with adapting the intervention fordelivery in public health facilities

DiarrheaPatient,GovernmentHospital

DiarrheaPatient,PrivateHospital

FamilyCaregiver,GovernmentHospital

FamilyCaregiver,PrivateHospital

HealthProvider,GovernmentHospital

Total

Semi-structuredinterviews

4 1 6 7 2 20

Key stakeholder interviews

• Address research questions related to scalability ofthe intervention

Health-Posting GovernmentOfficial

Senior Scientist Total


11 1 12

Pilot household consultant interviews and group discussions

• Learn about households’ experiences with theintervention and promoted behaviors, and seekadvice for how to improve the intervention for theupcoming trial

Standard Message mHealth with no home visits mHealth with2 home visits

Total


- 17 11 28

Groupdiscussions

- - 2 (19Participants)

2


Two group discussions (n = 7 and n = 12) wereconducted with a convenience sample of pilot studyintervention household members to explore the sametopics in a group setting and for additional interven-tion content development. Select findings from initialinterviews or early visits to pilot study householdswere shared and discussed with group discussion par-ticipants to facilitate discussion and request feedbackon intervention improvement.Interviews with diarrhea patients and caregivers,

health providers, and pilot study participants, as wellas the group discussions, were conducted in Banglaby native speakers after extensive training. Ten of the12 stakeholder interviews were conducted in Englishand two were conducted in Bangla. All interviews andgroup discussions were conducted by members of theCHoBI7 research team, who had at least master’slevel education in public health or social sciences,after training in qualitative data collection. Interviewsand group discussions were done with pre-testedguides, to explore research questions; follow-upprobes were used to further explore participants’ re-sponses. Fifty-nine of the 60 interviews and bothgroup discussions were audio recorded. One govern-ment stakeholder declined to have their interview re-corded. Interviews lasted between 12 min (family

caregiver) and 190 min (stakeholder). The group dis-cussions were 127 and 165 min. Extensive field noteswere taken by both the interviewer and any presentobserver to supplement recordings.

Data analysisAnalysis of the interviews and group discussions was com-pleted in three phases (Fig. 4). In the first phase, interviews(n = 24; 13 diarrhea patient/family caregivers and 11 pilotstudy) were followed by extensive debriefing to allow foriterative probing on emerging themes [28]. Second, aselection of interviews (n = 39; 16 diarrhea patient/house-hold member, 14 pilot study, 9 stakeholder) was tran-scribed verbatim in the language in which the interviewswere conducted. Transcriptions were reviewed by at leastone other member of the research team for quality con-trol. Transcriptions were not returned to participants fortheir review. The transcriptions were then used to developa summary template or ‘analysis questionnaire,’ based oncomponents of the intervention and emergent themes.Finally, the analysis questionnaire was completed for allpilot study interviews. Group discussions were summa-rized, and selected quotes were transcribed. Field notesfurther supplemented transcriptions and summaries.Completed analysis questionnaires, transcriptions, andgroup discussion summaries were reviewed to identify and

Fig. 3 CHoBI7 enabling technology


compile key findings and quotes for each interventioncomponent. Based on formative research findings, modifi-cations to the CHoBI7 intervention program were madethroughout the data collection and analysis process.

Data synthesisFor the final synthesis of findings, contextual, psychosocial,and technological factors that emerged from formative re-search were organized using the IBM-WASH framework.

ResultsFindings are presented by research question and includeinsights into WASH and diarrhea patient care practicesin health facilities, pilot study feedback on enabling tech-nologies and behavioral recommendations, and consider-ations for scalability that emerged from discussions withkey stakeholders.

What are existing WASH and diarrhea patient carepractices in health facilities?Treated drinking water (filtration) was available for con-sumption at icddr,b but not at the government facility.However, in both health facilities, patients and theirfamily members often brought their own drinking water,

such as purchased bottled water or boiled water fromhome. One health provider at the government facilityspecified that family members were counseled to bringin boiled water for patients because the quality of waterat the facility could not be guaranteed. At icddr,b, onecaregiver cited crowding near the facility’s water filterand the filter’s slow flow rate as reasons why he pur-chased bottled water during his family’s stay.Soap was also brought by patients’ families. Icddr,b

provided liquid soap at sinks, though participants saidnot all sinks had soap. Some families brought in soap ordetergent for personal use, such as to wash kanthas(‘blankets’) used by patients. Although handwashing fa-cilities at icddr,b were considered satisfactory, caregiversstill reported barriers to handwashing:

“… Here [icddr,b], the mother can only stay withthe patient, so I will take care of the baby or myself.That’s why [handwashing is] not done properly. But,in [the] home, if someone helps, then we can takecare [of the child] properly. Now, here, do I have totake care of the baby or wash hands? … If someoneis with me, then I can keep my child neat and clean.Here it is not possible.”

Fig. 4 CHoBI7 formative research activities


At the government facility, handwashing and toiletfacilities were considered poor by patients and providers.One patient reported visiting the toilet and not beingable to wash his hands due to lack of soap. Project stafffound that hardware on sinks adjacent to the diarrheaward was broken or missing. One health provider notedthat they did not have a sufficient soap supply to meethandwashing needs, and they had never visited thefacilities for the patients.Health facilities limit the number of caregivers that

can attend a patient. Some mothers said that they couldnot leave to wash their hands or relieve themselves forfear that their child would roll off the bed in their ab-sence. Another concern was that a child might defecatewhile the caregiver was being served food. Indeed, inter-viewers noted missed opportunities for handwashingduring interviews, sometimes because the patient re-quired care at that same moment.Patients and their family caregivers were asked

whether anyone from the health facility had providedhealth-related information. One participant at icddr,bsaid that a nurse had discussed saline intake andiron supplements. Another patient said that no onehad been by because the staff were too busy andthat he preferred to discuss health issues with family.Health providers at the government facility said thatpatients were given hygiene recommendations upondischarge, which may explain why, during our inter-views (before discharge), very few participants wereable to recall their receiving health advice fromhealth facility staff.

What are households’ experiences with and acceptabilityof CHoBI7 WASH enabling technology and behavioralrecommendations?Chlorine tablets for water treatmentPilot participants reported various household watertreatment practices prior to enrollment into the pilotstudy, including consumption of kaccha pani (‘raw, un-treated water’) from municipal water pumps or tubewells and filtering and/or boiling water. For most, use ofchlorine for water treatment was a new practice, thoughsome participants were familiar with a local water purifi-cation tablet called Halotab®. Most participants after re-ceiving CHoBI7 intervention delivery in the healthfacility were able to describe the proper use of theintervention-provided Aquatabs® (chlorine tablets) forwater treatment. However, some participants struggledto recount the correct water-to-tablet ratio or the lengthof time water could remain safe for consumption aftertreatment. A few participants expressed concern for hu-man error, such as taking the tablets as medicine, eventhough the flipbooks included a pictorial warning to notingest the chlorine tablets directly. Still, the chlorine

tablets were often referred to as aushoodh (‘medicine’).One patient attributed his recovery to the chlorinetablets:

“I’m good now only [because of] this chlorine tablet.If I didn’t get this tablet, I would not have beenhealthy.”

Discussions about the chlorine tablets often turned to thetaste and smell of chlorine-treated water. Participants saidtreated water smelled or tasted like “gas” or “medicine.”Though most participants mentioned taste as a concern,many said they were able to overcome it within a few days:

“If they drink this water one or two days, they willget used to it … Perhaps they would feel bad [forthe] first one or two days, [but] they will be okayafter that. I felt bad [for the] first few days, but nowI got used to [it].”

One participant said she had served the chlorine-treated water to employees in her shop, and they liked it.However, some households said they stopped usingchlorine-treated water shortly after returning home fromthe health facility, citing aversion to taste or smell, con-cern that the water was not safe for consumption, or anupset stomach as reasons for discontinuation.For those who did regularly use the chlorine tablets,

the convenience of the tablet compared to boiling drink-ing water was frequently mentioned:

“[The] tablet is a shortcut system and more con-venient ... it takes only 30 minutes. Boiling needsmore than 30 minutes … Sometimes there is no gassupply for the stoves, or sometimes the supply wateris too smelly and dirty to boil it. That’s why I thinktablets are better.”

Challenges with boiling water, such as restricted stoveaccess, gas supply, or time limitations, made chlorinationan appealing alternative. That said, most participantssaid they would boil to treat water after the 30-day sup-ply of chlorine tablets ran out. Several individuals saidthey looked for chlorine tablets in the market; otherssaid it would be best if they were given more tablets. Werecommended using chlorine-treated water to washfruits and vegetables eaten raw, but one participant saidthis would be a waste of treated water.

Safe water storageAt the time of the interviews, most participants were usingthe intervention-provided blue bucket with tap for storageof either chlorine-treated or boiled water for drinking.Participants expressed appreciation for the functionality


and convenience of the safe water storage bucket. For themost part, the bucket was considered good quality anddurable. Several participants preferred a larger bucket (e.g.20l) to accommodate drinking water needs, though oneparticipant noted that people may not be able to liftbuckets if they were larger, and another said that the 10lsize was best to help with measuring water amounts to betreated with chlorine tablets.Households reported using a traditional water storage

vessel, or kolsi, either alongside or prior to the receipt ofthe safe water storage bucket. A preference for the plas-tic bucket over the metal kolsi was common; one reasonwas because plastic would not rust. Both the lid and tapwere also important features of the bucket:

“A tap is attached to this [bucket]. If it were a nor-mal bucket, [people] would say, ‘Why will we usethis bucket instead of a kolsi? [The] kolsi is perfectfor me.’ But, since this bucket has a tap, I can takewater from this bucket through this tap. It is ashortcut with less trouble.”

The lid was seen as a means to keep dirt and hands outof drinking water, something that kolsis did not alwayshave. One concern about the tap was durability, as sometaps initially leaked. One participant did not think shecould replace the tap if she needed to buy a new bucket,saying she would need to return to dipping her hand inthe bucket with a cup. Another said she would pay 200–300 BDT (3.50 USD) for a new tap because she was accus-tomed to it now. The actual cost of the tap in the localmarket is 25–40 BDT (0.30–0.48 USD). One suggestiongiven by a participant was to fix the tap one inch higher tomake it less likely to break. Another issue was that chil-dren would open the tap and spill/waste water; one par-ticipant stored her drinking water in plastic bottles as aresult. On the other hand, the tap was also seen as ameans for children to access safe drinking water:

“I like this bucket because it has a tap. My kidscouldn’t pour water from the storage [container] be-fore, but now they don’t have to pour it. They canjust open the tap and drink from it using a mug.”

Interestingly, three participants mentioned plastic qual-ity, naming the company (a popular plastics company inBangladesh) where project buckets were sourced by nameand saying that this plastic “did not harbor germs” or “hadno typical plastic smell.” One maintenance issue men-tioned was that the bucket would sometimes acquire a“slimy layer” when water was kept in it for too long, mak-ing regular cleaning and drying of the bucket necessary.Two participants mentioned leaving their buckets withrelatives; one reason given was lack of space.

Handwashing station and soapy water bottleMany discussions about the handwashing station (waterstorage bucket with tap, bowl to catch rinse water, stoolto elevate bucket, and soapy water bottle) began with arecitation of the key times for handwashing promoted inthe flipbooks. Several households reported that the fre-quency of handwashing in the family had increased sinceintervention delivery:

“We used to wash our hands after coming from thetoilet and going out for a walk. After coming fromthe walk, we used to eat and feed our children with-out washing hands. But, now, after coming fromoutside, we wash our hands and then feed the chil-dren. The water and the soap are here.”

As noted, one reason mentioned for the increase inhandwashing was that “everything was in one place.”The handwashing station served as a facilitator and a re-minder to wash hands with soap. A key difference fromtypical practice was that handwashing inside or near thehouse was made possible by the handwashing station:

“There’s a bucket, a bowl, and soap there [at thehousehold entrance]. We wash our hands easilywith soap there after coming from outside. But, ifthere is no such arrangement, we have to go to thetube well and press it for water. Moreover, there arewomen there. Men do not always want to go therein the middle of the women.” (Female participant).

The need to visit a water pump to wash hands wasviewed as an inconvenience and barrier to handwashingimmediately following a contaminating event (e.g. usingthe toilet); pumps were reported to be crowded and re-quired the use of communal bar soap, which several par-ticipants considered unhygienic:

“In the case of bar soap, many people use the samesoap; germs and disease spread in this way. I thinkthe soapy water is better.”

One participant stated that the soapy water was the“most useful” CHoBI7 intervention component. Soapywater was considered easy to make, affordable, and toproduce a better lather compared to bar soap. Partici-pants said neighbors or relatives inquired about soapywater and later made it themselves. That said, one par-ticipant did not believe soapy water could remove germs,stating “After all, it is laundry detergent.” Another par-ticipant said the soapy water caused a “burning sensa-tion” on her hands.When asked what challenges could come between hand-

washing with soap at key times, participants said that


“laziness,” “busyness,” and the distraction of children werebarriers. Participants also said project staff had motivatedhousehold members, and it was now up to individuals toadopt the recommended behaviors. Additional recom-mendations included promoting regular nail-cutting andhaving multiple handwashing stations (e.g. one for insideand one near the kitchen). Similar to the safe water stor-age bucket, young children would play with and spill thehandwashing water and soapy water. One mother said shehad to keep the bowl stored away when it was not in usebecause she was concerned that her child would play inthe discarded handwashing water and become ill.The stool was seen as a valuable part of the station:

“Since I have this stool, I can easily take water from[the handwashing station]. It would be difficult totake water if there was no stool—our hands mighttouch the ground while washing them. Germs couldcome to our hands from the ground.”

Participants also stated that some of the handwashingstations were leaking. Other challenges included break-ing the bowl and displacement of the tap. Two groupdiscussion participants mentioned trying to fix the issuesby themselves but needing to seek assistance.The color distinction between the safe water storage

bucket (blue) and handwashing station bucket (red) wasconsidered helpful for separating “water for differentuses,” with someone mentioning that red indicated thebucket was not for drinking, like the red-marked tubewells that indicated arsenic:

“The color of the buckets is different, this is good. Ithelps to determine that this bucket is for drinkingwater and this one is for handwashing.”

However, one group discussion participant said thatpeople may be confused and think that the water in thered bucket contained arsenic.Finally, one participant reported that her neighbors

had started to use a bucket like hers to wash their hands“so they could live well and not need to go to the chol-era hospital.”

Cue cards and flipbooksFor the most part, participants felt the cue card design didnot require modifications. Several commented on the im-portance of the pictures on the cards to help illiterate partic-ipants or children understand behavioral recommendations.One participant said the cue cards were superior to oralinstruction. However, others said that people would notunderstand the cards without an accompanying explanation.The cue cards were considered important reminders

in case household members forgot the key times for

handwashing with soap or how to use chlorine tablets.One participant referred to the cue cards as “friends” be-cause they helped her stay healthy:

“These [cards] are like our friends; they are keepingus away from diseases, protecting us.”

Hanging the cue cards in sight or adjacent to the re-lated behavior (e.g. hanging the chlorine treatment cuecard next to the safe water storage bucket) was recom-mended by participants. One participant recommendedhanging cue cards outside so that others could see thecards and learn from them. A few participants reportednot hanging the cards due to busyness or lack of space.One participant commented on the absence of repair/maintenance information on enabling technologies inthe flipbooks, pointing to a concern about guidance forhow to resolve such issues.

What are considerations and modifications for taking theintervention program to scale?Enabling technologies and suppliesGovernment stakeholders had experience with theaforementioned Halotab® tablets, distributed in the pastby the government during outbreaks or national emer-gencies (e.g. floods). Stakeholders were concernedabout the population’s acceptance of the tablet due toits taste and worried that the tablet might be misusedas medicine. One government stakeholder suggestedincorporating flavors, like strawberry, to the tablets toimprove the taste. Important to note is that Halotab®is no longer available on the local market. Some gov-ernment stakeholders advised us to partner with localpharmaceutical companies for marketing chlorine tabletsand to improve the taste of these tablets. A concernwas that local pharmaceutical companies would notdevelop an alternative to Halotab® due to low anticipateddemand and earnings. One stakeholder suggested a part-nership with Non-Governmental Organizations (NGOs)to distribute Aquatabs® to health facilities in the future,thereby avoiding the government-mandated tenderingprocess.In our discussions on hand hygiene, one stake-

holder noted that the current government position onhygiene was that it was the responsibility of theindividual:

“… Hygiene … is now taken by [the] government as‘one individual, one hygiene.’ Your hygiene, it willbe by your own way … Hygiene is now consideredthe individual level, not the state level …”.

The suggestion here was that hygiene interventionsneeded to be sustainable without long-term support


of the government (e.g. continued provision of soap).Both liquid soap and hand sanitizer were said to besupplied in government health facilities, though forhealth professionals, only. To the stakeholders weinterviewed, using detergent to wash hands was anovel concept that aligned with an ongoing govern-ment subsidy program by the Government to keepcosts of detergent low:

“Detergent in Bangladesh is now very cheap. That isthe government action. And people can buy [it] atlow cost and can use [it].”

The same individual was surprised to learn that soapywater was made using the very same ‘non-deluxe’ (i.e.ordinary) detergent subsidized by the government. Onestakeholder recommended we explore other detergentsbeing used in the home to be sure they were suitable forsoapy water and to keep in mind the growing use of li-quid soap in households.

Mass media for promoting CHoBI7 enabling technologiesThe majority of government stakeholders said itwould be difficult to provide a handwashing stationand drinking water container to all diarrhea patientsin health facilities in Bangladesh. They recommendedensuring items were available in the market forpeople to buy and promoting them with a marketingcampaign. Mass media was considered the best av-enue to promote enabling technologies to households.One recommendation was to partner with major plas-tics companies in Bangladesh to market productsusing mass media and to make the handwashing sta-tion and drinking water vessel with tap as attractiveas possible for marketing:

“They [the patients] have money to purchase thishardware, but this hardware should be available …You motivate people to produce this type of hard-ware and they will popularize this hardware to thepeople through electronic media.”

“The manufacturer should be responsible [for messa-ging] because [the hardware] is a commercial project... we should talk with the manufacturer, convincethem … if you develop in this way, and sell it [for]130 Taka, so you are getting a profit [of] 32 Taka.That will give the opportunity to the manufacturer tomake … more profit … If you give the right messagesin the advertisement—that is we’ll ensure you the …safe drinking water...people [will] buy it.”

Television was the communication channel most rec-ommended for delivering CHoBI7 behavioral

recommendations. Radio, newspapers, and electronicmedia were also mentioned. It was recommended thatwe develop up to five short commercials, and to showthese regularly through Bangladesh television channels.A scrolling message on the television was also suggested.The success of companies like Lifebouy (a ‘health soap’marketed by Unilever) was mentioned as a potentialmodel for CHoBI7, with one stakeholder calling Life-buoy the “grand show for handwashing.” Mass mediacampaigns were also encouraged for the marketing ofchlorine tablets.

Delivering the CHoBI7 intervention program in healthfacilitiesStakeholders suggested that the scaling up of theCHoBI7 intervention program could involve healtheducators in government upazila (‘sub-district’) healthcomplexes or community health providers at commu-nity clinics. Community clinics were said to have highcoverage across the country; however, their healthproviders are known to have high workloads. One re-spondent recommended delivering the program atcommunity clinics and hiring individuals to deliverthe intervention. Health educators were considered tohave the advantage of being located in upazila healthcomplexes, where patients often stop first before go-ing to community clinics.A major challenge is the absence of community

clinics and sufficient health educators in urbanBangladesh. For urban settings, it was recommendedthat we partner with local NGOs, urban health cen-ters, government dispensaries, Dhaka Water Supplyand Sewerage Authority, the Local Government En-gineering Department, and Dhaka City Corporation todeliver the intervention program. Providing the hand-washing station and safe water storage bucket directlyto diarrhea patients was not considered a practicaloption for future scaling up of the intervention; onegovernment stakeholder said that, if we gave diarrheapatients these items, the inventory would beexhausted within a month because individuals withoutdiarrhea would come to facilities to collect them. Onesuggestion was for a demonstration handwashing sta-tion and drinking water container to be on display inhealth facilities for patients and family members touse as well as for health providers to recommend pa-tients and their families buy these items afterdischarge:

“If you give … these types of accessories [handwash-ing station and sealed water vessel] to the [health fa-cility] ward … and if it is available in the market,then they can tell the patients that, yes, now you’reusing it, go home and buy it and use it.”


There was general agreement that individuals wouldbe willing to buy the enabling technologies. It was alsonoted that many households already had these items intheir homes (e.g. bucket, stool, basin) but were not usingthem for WASH purposes.Government stakeholders recommended that doctors,

nurses, and health facility support staff be trained to de-liver the CHoBI7 intervention program:

“It will be more effective if you train our healtheducator(s), because we have health educator(s)in every district and every upazila health com-plex … Every day, they [will] give [the] messageto the outpatient(s) or people coming in thehospitals … First you have to train them … aswell as the doctors and the supporting staff also… and you put this hardware set, one set, inevery hospital.”

The ‘diarrhea corner’ or ‘ORT (Oral RehydrationTherapy) corner’ present at many health facilities,where the use of ORS is promoted, was mentionedas an ideal location for delivering an intervention. Itwas recommended that the Health Facility Flipbookbe delivered to both admitted and outpatient diar-rhea cases and that posters or leaflets be given topatients.Several government stakeholders mentioned that there

would be room for the CHoBI7 intervention program inupcoming, already-funded initiatives. Others said that,for the intervention to be taken to scale, funding wouldneed to be secured. Government stakeholders recom-mended CHoBI7 be part of the upcoming National Op-erational Plan for Communicable Diseases Control(CDC) Department:

“If you include this [the CHoBI7 program] inthe CDC operational plan to train up the healtheducators … the Health Education Bureau, theywill include this soap [and] hand washing [rec-ommendations] ... they [will] include these activ-ities to train-up five hundred to one thousandpeople.”

Integration of the CHoBI7 intervention program withexisting government programs, such as a weekly healthawareness program delivered at upazila health com-plexes, was mentioned as a method to minimize cost. Itwas also recommended that WASH behavioral recom-mendation development be done through the HealthEducation Bureau. One individual stated that, before theprogram could be recommended for scaled-up delivery,evidence that the intervention was effective would needto be published.

What modifications to the original CHoBI7 interventionprogram are needed to tailor the intervention for thetarget population?In preparation for the formative research presented here,revisions to the original CHoBI7 intervention program,enabling technologies, and behavior change communica-tion plan were made ahead of initiating the research ac-tivities presented here. These updates were based onexperiential knowledge of the study setting, previousformative research, and pre-determined changes in pro-gram content (i.e. focus on all diarrhea etiologies). Forexample, the wording of the flipbooks was updated tofocus on all types of diarrhea rather than just on cholera.Data collection to inform the development of the scal-able approach to CHoBI7 was initiated after these pre-liminary updates.The following additional modifications were made to

the intervention program based on formative researchfindings both to tailor the intervention to target house-holds as well as to address considerations for scalability(Table 2).

Chlorine tablets for water treatmentGiven that the taste of water treated with chlorine wasa concern expressed by most, a comparison of chlorineto local foods and plants that are health-protective buthave a bitter taste (e.g. neem plant and bitter gourd)was added to the flipbooks [29]. In addition, duringintervention delivery in health facilities, all family care-givers present were given a taste of chlorine-treatedwater and were probed to discuss their concerns, giv-ing health promoters the opportunity to counsel themon the taste. Based on the finding that many patientsand household members thought that chlorine tabletswere curative, participants were reminded that the tab-lets were not medicine but were a method of watertreatment. Household members were also counseledon how to make the transition from chlorine-treatedwater to boiled water after the 30-day supply of tabletswas finished.

Safe water storageThe original CHoBI7 intervention program had includedan imported safe water storage container. During pilot-ing, project staff met with local plastic suppliers to selecta durable plastic bucket option for the safe water storagebucket and handwashing station and their accompanyingstools. Based on pilot household feedback, the initial tapsfor both buckets were switched out early on in the pilotfor a sturdier option, and extra washers were added tolimit the possibility of leaks. Given households’ concernfor build-up on the bucket related to long-term waterstorage, we also included a recommendation to clean thebucket once per week. Initially, we considered


Table 2 Modifications to the original intervention package based on formative research findings, and considerations for scalability

Original CHoBI7 Intervention Design Formative Research Finding Modified Intervention Design Considerations for Scalability

Health facility-initiated interventiondelivered by health promoters tocholera patients and their householdcontacts over the 7-day high-riskperiod for transmission of disease tohousehold members.

Health facility-initiated interventiondelivered by health promoters todiarrhea patients and their householdcontacts in the health facility over the7-day high-risk period followingadmission to the health facility.

Replacement for study-supportedhealth promoters (e.g. healtheducators or nurses).Utility of mass media channels toalleviate burden of healthcommunication at the health facility.Cost of program materials providedat scale.Market availability of promotedenabling technologies and supplies(e.g. handwashing station, safe waterstorage bucket, and chlorine tablets)Integration of intervention programinto existing government healthprograms.

Pictorial module on how choleraspreads and can be prevented,delivered by a health promoter at thepatient’s bedside.

Pictorial module on how diarrheaspreads and can be prevented,delivered by a health promoter at thepatient’s bedside.

Communication module reinforcedduring daily household visits duringthe one-week intervention period.

Mobile messages reinforced eitherthrough only voice and text messagesweekly for 1 year (mobile health(mHealth) with no home visits), ormHealth and two home visits.

Imported, sealed drinking water vessel. Importance of using locallyavailable items to maximizesustainability and keep costof intervention low.Need for a durable tap.Importance of elevatingwater storage containerfrom the ground forhygiene purposes.Absence of reliable potablewater in some healthfacilities.Recommendation (orpreference) to bring inown source of drinkingwater for health facilitystay.Residue build-up in safewater storage bucket.

Locally made plastic bucket with lidand added durable tap.Provision of a stool to elevate bucket.Provision of chlorine tablets and onebottle of chlorine treated water forhealth facility stay.Messaging around cleaning the safewater storage bucket regularly to avoidbuild-up related to long-term waterstorage.

Low-cost, durable, locally madeoption will need to be madeavailable on the market.Availability of potable water in healthfacilities.

Chlorine tablets for water treatment Dislike of bitter taste ofchlorinated water.Barriers to regular boilingof water.Absence of safe drinkingwater source in healthfacility, or practice ofbringing in own drinkingwater.Trouble distinguishingchlorine as ‘watertreatment agent’ ratherthan ‘medicine’.

Chlorine tablets for water treatment.Messaging relating the taste ofchlorine to local medicinal plants andbitter foods with health benefits.Messaging around chlorine as watertreatment, not curative medicine.Taste tests of chlorinated water whilein the health facility.Messaging on boiling water after the30-day supply of chlorine tablets isexhausted.

Low-cost, locally made option willneed to be available via Bangladesh-based pharmaceutical companies

Handwashing station Handwashing aid neededat bedside.Misuse of enablingtechnologies.Use of handwashingstation by children.

Provision of bottle of soapy waterbottle for duration of health facilitystay.Additional information in thecommunication module aroundimportance of enabling technologiesfor a safe and healthy environment toavoid misuse.Additional messaging around teachingchildren how to use enablingtechnologies.

Low-cost, durable, locally madeoption will need to be madeavailable on the market.


recommending that all patients keep their safe waterstorage bucket in the health facility, but we found thatlimited space and sometimes-soiled floors made use ofthe bucket in the health facility for drinking water achallenge, even with the intervention provided stool. Fol-lowing the finding that family caregivers and patientsoften brought their own water to the health facility dur-ing their stay, we decided to leave a bottle of chlorine-treated water for them to use during the remainder oftheir stay.

Handwashing station and soapy water bottleInterviewers’ observations and interviews with patients,family caregivers, and health professionals necessitatedhandwashing technology at the patient’s bedside. We triedmultiple modifications to the intervention during the pilotstudy period to facilitate health facility-based handwashingwith soap. Ultimately, we decided to recommend that allintervention patients and caregivers keep a soapy waterbottle and a separate bottle of water for rinsing to facilitatehandwashing while in the health facility. We recom-mended that caregivers wash their hands using the soapy

water over the plastic basins that were often available forstorage and/or trash. In the mHealth with no home visitsarm, households were given the handwashing station inthe health facility since no subsequent home visits wereconducted to provide intervention materials. Promotersasked patients and caregivers to use their own detergentpowder for preparing soapy water, as many had broughtthis with them to the health facility. If they did not havetheir own detergent powder, we provided it.Once individuals returned home, we encouraged house-

holds to make multiple soapy water bottles to facilitatehandwashing at key times at different locations (e.g. latrineand kitchen areas). We also encouraged household mem-bers that left the house during the day to take a soapywater bottle with them to facilitate handwashing whileaway from home. We modified the detergent-to-water ra-tio for soapy water, increasing the ratio from four to sixcapfuls of detergent powder per 500mL of water to ensurethat a healthy lather was produced during handwashing.Following feedback about children not using the hand-washing station, we added images and recommendationsabout the importance of teaching children how to use the

Fig. 5 Flipbook modification to include information on caregiver-assisted handwashing with soap for younger children

Table 2 Modifications to the original intervention package based on formative research findings, and considerations for scalability(Continued)

Original CHoBI7 Intervention Design Formative Research Finding Modified Intervention Design Considerations for Scalability

Bottle of soapy water Need for healthy latherfrom soapy water withoutdrying of hands.Need for multiple soapywater bottles.

Increased the ratio from 4 capfuls per500 mL to 6 capfuls per 500 mL.Recommended households makemultiple soapy water bottles forplacement in home, latrine, andkitchen areas.

Current Government detergentsubsidies likely to support soapywater as a low-cost alternative to baror liquid soap.


handwashing station properly as they grew, emphasizingcaregiver-assisted handwashing with soap for youngerchildren and independent handwashing with soap forolder children (Fig. 5).Given the misuse of enabling technologies observed

during the pilot study period, we emphasized the im-portance of using each item, as directed, for a safe andhealthy home environment. Comical photos of

technology misuse, including using the stool as a televi-sion stand and using the basin as a place for dirty dishesor clothing, were shown by the health promoter in thehealth facility, and misuse was discouraged.

Flipbook and cue cardsIn the initial pilot rollout, there were six accompany-ing cue cards. Study staff deemed six too many

Table 3 Key factors related to intervention components as organized by IBM-WASH

Contextual factors Psychosocial factors Technology factors

Structural Government of Bangladesh mandatedtendering process for procurement ofgoods (e.g. chlorine tablets, enablingtechnologies, etc.).Government of Bangladesh policy thathygiene is the responsibility of theindividual.

Government stakeholder interest in, andprioritization of, a health facility-basedWASH intervention.

Past use of chlorine tablets by Bangladeshgovernment during cholera outbreaks orhigh-disease-risk events (e.g. floods).Pharmaceutical company development andprovision of an affordable, locally-availablechlorine tablet.Plastics company development and provisionof an affordable, locally-availablehandwashing station and safe water storagebucket.Government subsidization of laundrydetergent.

HealthFacility/Community

Access to/Condition of latrines, basin,potable water, stove use/fuel in thecompound or community.Access to/Condition of latrines, basin,potable water in public/private healthfacilities.Access to markets and market goods(e.g. soap or detergent).Public/private health facility resourcesand regulations.Informal providers as initial point oftreatment.

Prevalence and expectation of promotedbehaviors within the local and largercommunity.

Position of cue cards to increase visibility ofkey messages for neighbors of enrolledhouseholds.

Household Access to/Condition of, basin, potablewater, stove use/fuel in the household.Responsibility for water collection,treatment, and storage.Responsibility for children under fiveyears of age.Available space in home forhandwashing station and safe waterstorage bucket.Household size.

Desire to develop and maintain a safe andhygienic household environment.

Modelling handwashing station use forchildren to reduce non-use or misuse..

Individual Wealth, education, gender, livelihood ofprimary taker of children under five andprimary decision-maker.Age of children in household.

Awareness of diarrheal disease transmissionwithin the household/compound, and highperceived susceptibility of disease.High perceived benefits of adoptingprotective water, sanitation, and hygienebehaviors.Disgust reaction in response to theinformation that diarrheal disease can be aresult of consuming water or foodcontaminated with feces.Response efficacy that practice ofrecommended WASH behaviors will reducerisk of diarrheal disease.

Dislike of taste/smell of chlorine-treated waterduring.High perceived convenience of handwashingstation, soapy water bottle, safe water storagebucket, and chlorine tablets compared toother options.Strengths/weaknesses of enablingtechnologies for users (e.g. size).

Habitual Environment allows for habit formationand limits habit disruptions.

Existing water collection, treatment andstorage practices.Existing hand hygiene practices.

Ease of routine use of handwashing stationand safe water storage bucket.Visible handwashing station, safe waterstorage bucket, and cue cards to serve ascues to action.


because all available spaces on the walls in house-holds were being covered with project materials. Asa result, we reduced the number of cue cards fromsix to four. We also developed a sticker with a pic-ture of a model family drinking water next to thesafe water storage bucket and the accompanyingmessage: “Germ free hands, safe water, make mehealthy.” We encouraged households to place thesticker on our intervention items, mirrors, or else-where in the home as a reminder to practice the be-havioral recommendations.

Synthesis of findings: IBM-WASH factors of influence forthe CHoBI7 intervention programIn Table 3, we organized the contextual, psychosocial,and technological factors identified from this formativeresearch according to the IBM-WASH framework [21].We were able to identify and modify multiple factors ofinfluence through the formative research activities pre-sented here, and organizing them with IBM-WASHhelps to acknowledge the multiple levels and dimensionsthat influence WASH-related behaviors in this context.For example, interviews with diarrhea patients and theirfamily caregivers, as well as with health care providers,provided important insights into the health facility envir-onment (IBM-WASH community-level, contextual andtechnological dimensions), helping to highlight access-related barriers to the CHoBI7 behavioral recommenda-tions that necessitated modifications to the intervention(e.g. provision of soapy water bottles in the hospital [Ta-bles 2 and 3]). Pilot study participants’ insistence that in-dividuals would become accustomed to the taste ofchlorinated water over time (IBM-WASH individual-level, technological dimension) led us to front-load be-havioral recommendations to help participants throughtheir initial dislike of the taste of chlorinated water. Wealso realized that, as households finished their supply ofchlorine tablets and needed to adopt boiling, we wouldneed supportive mHealth messaging during the transi-tion to alternative methods of water treatment that ad-dressed identified constraints at the IBM-WASHhousehold level. Finally, we identified several higher-level contextual and technological factors that are typic-ally out of the scope of WASH interventions but shouldbe addressed or considered for scale-up, such aspharmaceutical company development and provision ofan affordable, locally-available chlorine tablet.

DiscussionFormative research was critical for tailoring theCHoBI7 intervention program to our target popula-tion and for identifying approaches for scalable pro-gram delivery. Interviews with diarrhea patients andtheir family caregivers, government officials, and

health providers, in addition to the pilot study of theintervention, informed our understanding of factorsthat are important for a household’s adoption andmaintenance of WASH behaviors, modifications need-ing to be made to intervention program materials,and considerations for delivering the CHoBI7 interven-tion program in public and private healthcare facilitiesin Bangladesh. With respect to existing WASH anddiarrhea patient care practices in health facilities, in-terviews provided important insights into the healthfacility environment, including insights into gaps indiarrhea patient and family caregiver access to waterand hygiene while in health facilities that could befilled by the CHoBI7 intervention program.Pilot study feedback on intervention components was

consistent with findings from other settings. For ex-ample, the role of taste with respect to acceptance ofchlorine treated water has been well documented [18],and our modifications to the intervention to mitigatethis concern could be applied in other contexts wheretreated water is part of an intervention. Similar to otherresearch in Bangladesh, when interviewing pilot studyparticipants, we found that having a handwashing stationin the home facilitates handwashing by having all mate-rials required for handwashing in one place; however,durability of the material, cost associated with alternativehandwashing agents (e.g. low cost of laundry detergent),and disgust at sharing soap also play a role in whetheror not the technology will be used [30]. Exploring house-holds’ experiences with and acceptability of CHoBI7WASH enabling technology and behavioral recommen-dations allowed us to tailor the intervention to a specificpopulation.Interviews with government stakeholders provided

valuable insights for scaling up of the CHoBI7 interven-tion program in Bangladesh. We learned that deliveringCHoBI7 enabling technologies to all diarrhea patients inhealth facilities would likely not be feasible at scale.However, stakeholders recommended promising alterna-tives, including partnering with plastics and pharmaceut-ical companies to market enabling technologies andsupplies through mass media. These partnerships shouldbe pursued during the transition to scale. The successfulscale-up campaign for zinc treatment of childhood diar-rhea in Bangladesh that included private-sector partici-pation presents a promising approach for scale-up ofCHoBI7 intervention program components [31]. An-other suggestion was to have a demonstration hand-washing station and drinking water container on displayin health facilities for patients and family members touse and for health providers to recommend their pur-chase. Outside icddr,b’s hospital, shopkeepers sell itemscommonly used in the diarrhea ward (e.g. sippy cupsand bedding). Partnerships with these individuals to sell


CHoBI7 technologies could also be explored. In our re-cent RCT of the modified intervention (George et al.,2020 submitted), several neighbors of enrolled house-holds made their own enabling technology using mate-rials they obtained from the local market, whichsuggests that a model display in the hospital could facili-tate uptake. Government stakeholder recommendationsfor integration of the CHoBI7 intervention program intothe services provided in government upazila healthcomplexes and community clinics are encouraging. Fu-ture research is needed to pilot the intervention programin these health facilities.It is important to note that some recommendations

made by government stakeholders were not possible toincorporate in this iteration of intervention adaptation(e.g. mass media campaign or selling of enabling tech-nologies and supplies), given the intention to test theeffectiveness of the intervention through an RCT. A keyobjective of this formative research was to define amore scalable approach for CHoBI7 program delivery.Developing a scalable intervention is an iterativeprocess. The first iteration of the CHoBI7 interventionprogram was not scalable due to its intensive homevisits. In the second iteration, presented here, we re-duced the cost of enabling technologies, reduced thenumber of home visits in one arm, and eliminatedhome visits altogether in the mHealth with no homevisits arm. We also added a mobile message componentto reinforce the key behaviors promoted by the pro-gram over time. Our next step is to scale this programin partnership with the Ministry of Health and FamilyWelfare, incorporating the suggestions made by gov-ernmental stakeholders.Mangham and Hanson identify four major challenges

to scaling up interventions in international health: costsassociated with scaling up coverage, scaling constraints,concerns for equity and quality, and issues related toservice delivery [32]. Challenges with the reach, sustain-ability, and scalability of the CHoBI7 intervention pro-gram remain. First, a challenge to delivering the CHoBI7intervention program in government health facilities isvariability in the availability of WASH infrastructure,such as running water. Second, human resources in gov-ernment health facilities in Bangladesh are often limited[33]. Third, in Bangladesh, informal health providers, in-cluding village doctors and pharmacists, may be visitedbefore care is sought at a health facility [34, 35]. Futurestudies should assess approaches to include informalhealth providers.The formative research presented here includes several

strengths. One strength of the design was the involvementof pilot study participants as consultants to improve theintervention. This practice follows similar research [36]that suggests that the direct involvement of beneficiaries

and an iterative process for intervention design comprise avaluable approach to tailoring interventions. A secondstrength was the iterative nature of the formative research,which allowed us to tailor the intervention as we learnedand to explore emergent themes that further deepenedour understanding of households’ experiences with theintervention and behavioral recommendations. Finally, ourpartnership with key government stakeholders allowed usto consider pathways to scale early in the process and de-velop support for a scalable approach to the intervention.Diarrhea remains a top reason for seeking care at a

health facility in Bangladesh [37]. There is a high risk oftransmission of enteric disease within the health facilityand home for household members of diarrhea patients[3–5, 7]. The CHoBI7 intervention program presents animportant opportunity to intervene during this high-riskperiod, when perceived susceptibility of illness is high,perceived benefits of protective WASH behaviors arealso likely to be high [18], and the environment is favor-able for habit formation [19]. Other studies have alsoprovided evidence that WASH interventions in healthfacilities can facilitate sustained behavior change [12,15]. The CHoBI7 intervention program differs frommore traditional, community-based WASH interven-tions, given that it is health facility-initiated and facili-tates changes in behavior both in the health facility andonce the patient and caregivers return home. Futurework should prioritize WASH interventions for those athigh risk of enteric diseases that can be taken to scale inhealth facilities—and sustained in households—to reducefuture enteric disease episodes.One limitation is that our interviews with pilot study

participants were mostly with women. Although muchof this imbalance was due to the availability of partici-pants, the interviews we were able to conduct with pilothousehold men helped us to better understand how toincorporate their needs and schedules into our behav-ioral recommendations. Additional interviews with menmay provide important insight into how to better bal-ance household responsibilities related to WASH behav-iors. Finally, future formative research should includeinterviews with health facility staff at multiple levels ofcare and management (e.g. doctors, nurses, and janitorialstaff) to better inform our understanding of the practicesand resources in health facilities that may impact deliv-ery of the CHoBI7 intervention program.

ConclusionFormative research identified existing WASH and diar-rhea patient care practices, target population experienceswith and acceptability of a health facility-initiatedWASH intervention, and facilitated modifications to theCHoBI7 intervention program in an effort to identify ascalable approach to deliver this program in Bangladesh.


AbbreviationsCHoBI7: Cholera-Hospital-based-Intervention-for-7-Days; WASH: Water,sanitation, and hygiene; IBM-WASH: Integrated Behavioural Model for Water,Sanitation, and Hygiene; RCT: Randomized controlled trial; ORS: Oralrehydration solution; icddr,b: International Centre for Diarrhoeal DiseaseResearch, Bangladesh; NGOs: Non-Governmental Organizations;CDC: Communicable Diseases Control

AcknowledgementsThe authors would like to thank all CHoBI7 study participants, theparticipating health facilities, supporting members of the Ministry of Health &Family Welfare of Bangladesh, and other local partners that have made thiswork possible. The authors would also like to thank the larger CHoBI7 studyteam for their support. ICDDR, B thanks the Governments of Bangladesh,Canada, Sweden, and the United Kingdom for core/unrestricted support.

Authors’ contributionsEDT supported data collection, conducted data analysis and interpretation, andwrote the original draft of this manuscript. CMG conceived of the study designand directed the study. CMG and EDT designed the interview and focus groupdiscussion guides. Interviews were conducted by MTH, TP, MKH, MH, FZ, andCMG. Group discussions were led by MTH and FZ. Data summaries andtranscriptions were completed by MSR, MTH, TP, FZ, AT, MKH, MH, and EDT. FZ,AT, and JM supported intervention development and delivery. EDT, MSR, andPJW developed the data analysis questionnaire. MSR conducted data analysisand interpretation. MSIB contributed to data collection and led datamanagement. ST, FM, MAHK, CMG, EL, and PJW critically reviewed and revisedthe original manuscript. DS, MA, and MS contributed to the design of theCHoBI7 study. CMG, MA, and SM directed data collection for the study. Allauthors provided support and suggestions during the interventiondevelopment process and/or subsequent writing and review of this manuscript.All authors read and approved the final manuscript.

FundingThis study received financial support from the United States Agency forInternational Development (USAID-OAA-F-15-00038). The funder did nothave a role in the design, implementation, or evaluation of the study.

Availability of data and materialsThe datasets generated and/or analyzed during the current study are notpublicly available due to identifiable content.

Ethics approval and consent to participateStudy procedures were explained to all participants by trained researchassistants who collected written informed consent and/or parental permissionfrom all participants ahead of study enrollment for the pilot study or standalonesemi-structured interviews and group discussions. Participants were informed ofthe study objectives as part of the consent process. Prior to beginning inter-views and group discussions, interviewers/moderators introduced themselvesand their current position in the study team. The study was reviewed and ap-proved by the Johns Hopkins Bloomberg School of Public Health InstitutionalReview Board and the Ethical Review Committee of icddr,b.

Consent for publicationNot applicable.

Competing interestsChristine Marie George is an Editorial Board Member of BMC Public Health.The other authors declare no competing interests.

Author details1Department of International Health, Johns Hopkins Bloomberg School ofPublic Health, Baltimore, MD, USA. 2Infectious Diseases Division, InternationalCentre for Diarrhoeal Disease Research, Bangladesh, Dhaka, Bangladesh.3Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA.4Department of Communicable Diseases, Ministry of Health and FamilyWelfare, Dhaka, Bangladesh. 5Community Based Health Care, DirectorateGeneral of Health Services, Dhaka, Bangladesh. 6Associate Professor,Department of International Health, Program in Global Disease Epidemiologyand Control, Johns Hopkins Bloomberg School of Public Health, 615 N. WolfeStreet, Room E5535, Baltimore, MD 21205-2103, USA.

Received: 8 December 2019 Accepted: 17 April 2020

References1. Naghavi M, Abajobir AA, Abbafati C, Abbas KM, Abd-Allah F, Abera SF, et al.

Global, regional, and national age-sex specific mortality for 264 causes ofdeath, 1980–2016: a systematic analysis for the global burden of diseasestudy 2016. Lancet. 2017;390(10100):1151–210.

2. Liu L, Oza S, Hogan D, Chu Y, Perin J, Zhu J, et al. Global, regional, andnational causes of under-5 mortality in 2000-15: an updated systematicanalysis with implications for the sustainable development goals. Lancet.2016;388(10063):3027–35.

3. Weil AA, Khan AI, Chowdhury F, Larocque RC, Faruque AS, Ryan ET, et al.Clinical outcomes in household contacts of patients with cholera inBangladesh. Clin Infect Dis. 2009;49(10):1473–9.

4. Echeverria P, Seriwatana J, Taylor DN, Yanggratoke S, Tirapat C. Acomparative study of enterotoxigenic Escherichia coli, Shigella, Aeromonas,and Vibrio as etiologies of diarrhea in northeastern Thailand. Am J TropMed Hyg. 1985;34(3):547–54.

5. George CM, Ahmed S, Talukder KA, Azmi IJ, Perin J, Sack RB, et al. Shigellainfections in household contacts of pediatric shigellosis patients in ruralBangladesh. Emerg Infect Dis. 2015;21(11):2006–13.

6. Black RE, Merson MH, Rowe B, Taylor PR, Abdul Alim AR, Gross RJ, et al.Enterotoxigenic Escherichia coli diarrhoea: acquired immunity andtransmission in an endemic area. Bull World Health Organ. 1981;59(2):263–8.

7. George CM, Hasan K, Monira S, Rahman Z, Saif-Ur-Rahman KM, Rashid MU,et al. A prospective cohort study comparing household contact and waterVibrio cholerae isolates in households of cholera patients in ruralBangladesh. PLoS Negl Trop Dis. 2018;12(7):e0006641.

8. Islam MS, Luby SP, Sultana R, Rimi NA, Zaman RU, Uddin M, et al. Familycaregivers in public tertiary care hospitals in Bangladesh: risks andopportunities for infection control. Am J Infect Control. 2014;42(3):305–10.

9. Clasen T. Household water treatment and safe storage to prevent diarrhealdisease in developing countries. Curr Environ Health Rep. 2015;2(1):69–74.

10. Zohura F, Bhuyian SI, Monira S, Begum F, Biswas SK, Parvin T, et al.Observed handwashing with soap practices among cholera patients andaccompanying household members in a hospital setting (CHoBI7 trial). AmJ Trop Med Hyg. 2016;95(6):1314–8.

11. Curtis V, Cairncross S. Effect of washing hands with soap on diarrhoea risk inthe community: a systematic review. Lancet Infect Dis. 2003;3(5):275–81.

12. Sreenivasan N, Gotestrand S, Ombeki S, Oluoch G, Fischer T, Quick R.Evaluation of the impact of a simple hand-washing and water-treatmentintervention in rural health facilities on hygiene knowledge and reportedbehaviours of health workers and their clients, Nyanza Province, Kenya,2008. Epidemiology & Infection. 2015;143(4):873–80.

13. Wood S, Foster J, Kols A. Understanding why women adopt and sustainhome water treatment: insights from the Malawi antenatal care program.Soc Sci Med. 2012;75(4):634–42.

14. Sheth AN, Russo ET, Menon M, Wannemuehler K, Weinger M, Kudzala AC,et al. Impact of the integration of water treatment and handwashingincentives with antenatal services on hygiene practices of pregnant womenin Malawi. Am J Trop Med Hyg. 2010;83(6):1315–21.

15. Parker AA, Stephenson R, Riley PL, Ombeki S, Komolleh C, Sibley L, et al.Sustained high levels of stored drinking water treatment and retention ofhand-washing knowledge in rural Kenyan households following a clinic-based intervention. Epidemiol Infect. 2006;134(5):1029–36.

16. George CM, Monira S, Sack DA, Rashid M-u, Saif-Ur-Rahman K, Mahmud T,et al. Randomized controlled trial of hospital-based hygiene and watertreatment intervention (CHoBI7) to reduce cholera. Emerg Infect Dis. 2016;22(2):233.

17. Rosenstock IM. Historical origins of the health belief model. Health EducMonogr. 1974;2(4):328–35.

18. Figueroa ME, Kincaid DL. Social, cultural and behavioral correlates ofhousehold water treatment and storage. Center Publication HCI 2010–1:Health Communication Insights; 2010.

19. Neal D, Vujcic J, Hernandez O, Wood W. The science of habit: creatingdisruptive and sticky behavior change in handwashing behavior.Washington DC: USAID/WASHplus Project; 2015.

20. Mosler HJ. A systematic approach to behavior change interventions for thewater and sanitation sector in developing countries: a conceptual model, areview, and a guideline. Int J Environ Health Res. 2012;22(5):431–49.


21. Dreibelbis R, Winch PJ, Leontsini E, Hulland KR, Ram PK, Unicomb L, et al.The integrated Behavioural model for water, sanitation, and hygiene: asystematic review of behavioural models and a framework for designingand evaluating behaviour change interventions in infrastructure-restrictedsettings. BMC Public Health. 2013;13(1):1015.

22. George CM, Jung DS, Saif-Ur-Rahman KM, Monira S, Sack DA, Mahamud-urR, et al. Sustained uptake of a hospital-based handwashing with soap andwater treatment intervention (cholera-hospital-based intervention for 7 days[CHoBI7]): a randomized controlled trial. Am J Trop Med Hyg. 2016;94(2):428–36.

23. George CM, Biswas S, Jung D, Perin J, Parvin T, Monira S, et al. Psychosocialfactors mediating the effect of the CHoBI7 intervention on handwashingwith soap: a randomized controlled trial. Health Educ Behav. 2017;44(4):613–25.

24. Briscoe C, Aboud F. Behaviour change communication targeting four healthbehaviours in developing countries: a review of change techniques. Soc SciMed. 2012;75(4):612–21.

25. Curtis V, Schmidt W, Luby S, Florez R, Touré O, Biran A. Hygiene: new hopes,new horizons. Lancet Infect Dis. 2011;11(4):312–21.

26. Inauen J, Mosler HJ. Developing and testing theory-based and evidence-based interventions to promote switching to arsenic-safe wells inBangladesh. J Health Psychol. 2014;19(12):1483–98.

27. George CM, Zohura F, Teman A, Thomas E, Hasan T, Rana S, et al. Formativeresearch for the design of a scalable mobile health program water,sanitation, and hygiene: CHoBI7 mobile health program. BMC Public Health.2019;19(1):1028.

28. McMahon SA, Winch PJ. Systematic debriefing after qualitative encounters:an essential analysis step in applied qualitative research. BMJ Glob Health.2018;3(5):e000837.

29. Qadri F, Ali M, Chowdhury F, Khan AI, Saha A, Khan IA, et al. Feasibility andeffectiveness of oral cholera vaccine in an urban endemic setting in Bangladesh:a cluster randomised open-label trial. Lancet. 2015;386(10001):1362–71.

30. Hulland KR, Leontsini E, Dreibelbis R, Unicomb L, Afroz A, Dutta NC, et al.Designing a handwashing station for infrastructure-restricted communitiesin Bangladesh using the integrated behavioural model for water, sanitationand hygiene interventions (IBM-WASH). BMC Public Health. 2013;13(1):877.

31. Larson CP, Koehlmoos TP, Sack DA. Scaling up of zinc for young childrenproject T. scaling up zinc treatment of childhood diarrhoea in Bangladesh:theoretical and practical considerations guiding the SUZY project. HealthPolicy Plan. 2012;27(2):102–14.

32. Mangham LJ, Hanson K. Scaling up in international health: what are the keyissues? Health Policy Plan. 2010;25(2):85–96.

33. Ahmed SM, Hossain MA, Rajachowdhury AM, Bhuiya AU. The healthworkforce crisis in Bangladesh: shortage, inappropriate skill-mix andinequitable distribution. Hum Resour Health. 2011;9(1):3.

34. Das SK, Nasrin D, Ahmed S, Wu Y, Ferdous F, Farzana FD, et al. Health care-seeking behavior for childhood diarrhea in Mirzapur, rural Bangladesh. Am JTrop Med Hyg. 2013;89(1 Suppl):62–8.

35. Ahmed SM. Exploring health-seeking behaviour of disadvantagedpopulations in rural Bangladesh: Institutionen för folkhälsovetenskap/Department of Public Health Sciences; 2005.

36. Harvey SA, Paredes Olortegui M, Leontsini E, Ramal Asayag C, Scott K,Winch PJ. Trials of improved practices (TIPs): a strategy for making long-lasting nets last longer? Am J Trop Med Hyg. 2013;88(6):1109–15.

37. Sultana M, Mahumud RA, Sarker A. Emerging patterns of mortality andmorbidity in district level hospitals in Bangladesh. Ann Public Heal Res.2015;2:2–4.

Publisher’s NoteSpringer Nature remains neutral with regard to jurisdictional claims inpublished maps and institutional affiliations.


AbstractBackgroundMethodsResultsConclusions

BackgroundThe Cholera-Hospital-based-Intervention-for-7-Days (CHoBI7)Setting and study rationale

MethodsData collectionData analysisData synthesis

ResultsWhat are existing WASH and diarrhea patient care practices in health facilities?What are households’ experiences with and acceptability of CHoBI7 WASH enabling technology and behavioral recommendations?Chlorine tablets for water treatmentSafe water storageHandwashing station and soapy water bottleCue cards and flipbooks

What are considerations and modifications for taking the intervention program to scale?Enabling technologies and suppliesMass media for promoting CHoBI7 enabling technologiesDelivering the CHoBI7 intervention program in health facilities

What modifications to the original CHoBI7 intervention program are needed to tailor the intervention for the target population?Chlorine tablets for water treatmentSafe water storageHandwashing station and soapy water bottleFlipbook and cue cards

Synthesis of findings: IBM-WASH factors of influence for the CHoBI7 intervention program

DiscussionConclusionAbbreviationsAcknowledgementsAuthors’ contributionsFundingAvailability of data and materialsEthics approval and consent to participateConsent for publicationCompeting interestsAuthor detailsReferencesPublisher’s Note

formative research to scale up a handwashing with soap and … · 2020. 6. 2. · research article...

Documents