WATER AND SANITATION PROGRAM: LEARNING NOTE 64855 Global Scaling Up Handwashing Project Key findings Tanzania: A Handwashing • While designing a behavior change intervention from an evidence base is Behavior Change Journey critical, limited formative research can be analyzed in the FOAM framework October 2011 and information gaps can be filled with small-scale spot research. • Designing a behavior change communication campaign across mutually reinforcing channels may INTRODUCTION prove to be a particularly effective In Tanzania, nearly 30,000 people die rural districts5 with end-of-project tar- approach to changing behaviors. annually due to diarrheal diseases1 and gets that included 1.25 million women an estimated 12.6 percent of children and children practicing improved hand- • Behavior change often takes place suffer from diarrheal diseases.2 There washing behaviors; 14.5 million women through an evolving process; is a widespread and deep-rooted belief and children ages 5 to 14 exposed to programs designed to be multi- that diarrhea is part of growing up and behavior change messaging through phased, with evolving objectives, may cannot be prevented. Working with the the radio; 300,000 women and children help support behavior shifts. government and with non-governmen- reached through Interpersonal Com- tal organizations, the Water and Sanita- munication (IPC) activities; and the at- tion Program (WSP) sought to increase tendance of 170,000 people at Direct rates of handwashing with soap among Consumer Contact (DCC) events. To women and children, especially at criti- analyze the health and poverty im- cal junctures. pacts, intervention areas were selected to produce a representative sample of Initial support started in 2005 with the general population. the development of the Public-Private Partnership for Handwashing3 and This Learning Note documents the de- small-scale formative research which velopment of the project, with a focus showed that while knowledge of the on how it was designed, implemented, reasons to wash hands was high, the and monitored. Challenges and lessons practice was low, with just four percent learned are highlighted to assist program of mothers and five percent of children managers in designing and manag- reporting that they wash hands with ing evidence-based handwashing with soap.4 In 2006, WSP began to scale up soap and/or other hygiene promotion the handwashing program across ten programs. 1 WHO. 2006 Mortality Fact Sheet, http://www.who.int/whosis/mort/profiles/mort_afro_tza_tanzania.pdf 2 Tanzania DHS. 2004-05 Final Report, http://www.measuredhs.com/pubs/pub_details.cfm?ID=566&ctry_ id=39&SrchTp=ctry 3 For more information, see http://www.globalhandwashing.org 4 TMS International. 2006. Understanding the Tanzanian Consumer in Respect to Handwashing with Soap. 5 The ten districts are Masasi, Rufiji, Iringa, Sumbawanga, Mpwapwa, Kiteto, Kondoa, Karagwe, Igunga, and Musoma. 2 Tanzania: A Handwashing Behavior Change Journey Global Scaling Up Handwashing Project Figure 1: FOAM Framework handwashing behavior (Figure 1). Communication objectives were identified based on the identification of determinants Focus pp y Opportunity Ability Motivation (Figure 2) and these were shared with the Government of Access/ Belief and Tanzania and key stakeholders to facilitate engagement Target availability attitudes Knowledge behavior throughout the process. Outcome Product expectations attributes Additional research was conducted to test preliminary cre- Target Social Threat ative concepts developed at a multi-sectoral workshop. In population Social support norms Intention order to identify the “trigger� that would motivate women to improve their handwashing behavior, representatives from the hygiene sector generated several creative concepts based on the results of the formative research, their own experience, ACTION and a field visit. These were tested with rural women at focus Research group sessions led by marketing research specialists who The team’s first steps toward developing an evidence-based worked at a private sector soap company providing tech- behavior change program focused on gathering and con- nical support to the partnership. An important insight was ducting research to better understand the target audience— that mothers felt burdened with responsibility for raising chil- including current understandings, feelings, and actions dren, but were rarely thanked or rewarded, and asking them toward handwashing. to focus on handwashing with soap was felt to be an extra burden. Two ideas emerged from this experience, “Mother is Initial formative research indicated that people would tend to the Pillar of the Home� (“Mama Nguzo�) and “Heroes Make rinse hands when they were visibly dirty, but that the use Heroes.� Both concepts acknowledged mothers for all they of soap for handwashing was not widespread. Additionally, did—including teaching handwashing with soap—rather than respondents indicated that handwashing with soap was an rebuking them for what was not done. extra burden and many of those surveyed felt they were too busy to wash their hands with soap. Design Findings from the formative research were analyzed using Following testing, a creative brief was developed at a multi-sec- FOAM,6 a framework designed to identify behavioral deter- toral workshop (Figure 3). The brief was provided to an advertis- minants for handwashing, and their relative role in shaping ing agency that ultimately developed the communications idea Figure 2: Mapping Formative Research Findings to FOAM Determinants and Communication Objectives FOAM g Research Finding Determinant Communication Objectives Soap and water are not often located near the Have mothers believe it is easy to ensure that soap Access latrine or cooking area and water are easily available for handwashing Social Everyone routinely uses soap when washing hands and that Low prioritization of using soap while handwashing Norms washing hands without soap is not effective handwashing. Low practice of handwashing with soap Knowledge Have mothers understand the danger of invisible germs after defecation There are no visual cues to make Have mothers construct handwashing stations to act as Intention handwashing habitual reminders to wash hands with soap Social Have mothers believe that by washing their hands, People who do not wash hands are filthy Beliefs they are a good community member 6 See Y. Coombes and J. Devine, Introducing FOAM: A Framework to Analyze Handwashing Behaviors to Design Effective Handwashing Programs. Available at http:// www.wsp.org/wsp/sites/wsp.org/files/publications/WSP_IntroducingFOAM_HWWS.pdf www.wsp.org Global Scaling Up Handwashing Project Tanzania: A Handwashing Behavior Change Journey 3 Figure 3: Summary of Creative Brief The job to be done: To have mothers believe in their hearts that handwashing with soap is a critical part of their role as strong, resourceful caregivers of the family and that it is simple and easy to do What the Audience Does, Thinks, Feels Today What We Want Them to Do, Think, Feel Tomorrow Does If hands are visibly dirty, they will wash hands with water (no soap) Always wash hands with soap after defecation and before or else just wipe hands on clothes handling food Thinks • Handwashing with soap is an extra burden—too busy to do it • Handwashing with soap is simple and quick • Diarrhea is part of growing up • Handwashing with soap makes life easier—saves money • Generations before have survived without soap • Handwashing with soap will have an impact on a child’s wellbeing Feels • They are surviving • Finally, my role as a mother is recognized • Their family trusts them when it comes to managing the • Empowered that handwashing with soap can help care for household and hygiene issues their children • They feel nobody notices their efforts • Feel proud about the role they play • We have found a solution and are less burdened “Mikono Yenye Fahari� (“Hands To Be Proud Of�). This concept health centers—were carried out by Front-Line Activators formed the backbone of the campaign: it tapped into mothers’ (FLAs). Village Executive Officers selected this unpaid, vol- aspirations for recognition, their pride in the work they do for their unteer, community-based cadre of women (and occasionally families, and their role as the central pillar of the family. The de- men) for their self-motivation. Working through FLAs ensured sign went through several iterations and was pre-tested through not only that skills acquired through the training would remain focus groups and key informant interviews. Spot research was in the community but also that messages would be delivered also undertaken to close gaps in the evidence base, particularly by a source respected by the community. The vision for this around beliefs and knowledge about the link between hand- volunteer cadre was that they would serve their immediate washing with soap and acute respiratory infection. communities and take advantage of day-to-day activities to promote handwashing with soap. Because they would be IMPLEMENTATION disassociated from an institutional structure, it would also be Integrated Channels more possible to avoid nepotism and to ensure the hand- Key messages were promoted through multiple integrated chan- washing with soap duties were not lost in the list of duties nels, which has been shown to be effective for behavior change in assigned to village level officials. Working with local govern- other sectors such as HIV/AIDS, malaria, and reproductive health. ment authorities, 45 FLAs were selected and trained in each When an individual receives a consistent message through multi- district, reaching a total of 450 trained. While FLAs are not ple channels it reinforces social norms around the behavior. Mes- part of any institution or government structure, they served saging across three channels, mass media, Direct Consumer as a resource to promote hygiene to Village Executive Of- Contact (DCC) and Interpersonal Communication (IPC) was har- ficers, Ward Executive Officers, and district-level officials. monized by the development of a single creative brief. Direct Consumer Contact events complemented IPC ac- Mass media offers constant reminders and cues to action. tivities with a goal of improving their audiences’ knowledge, Radio was selected as the mass media carrier based on re- especially their knowledge of the ease of handwashing with search that indicated that television coverage was low and soap. DCC was chosen because public events have been radio influence high in rural areas. effective in commercial marketing to raise the enthusiasm of participants for a topic and help shifts social norms. Addi- Interpersonal Communication can target building knowledge tionally, these events were a logical channel to improve the and skills. Activities—including visits to homes, schools, and access portion of the opportunity determinant by introducing www.wsp.org 4 Tanzania: A Handwashing Behavior Change Journey Global Scaling Up Handwashing Project the tippy-tap, an enabling technology for handwashing7, on Sustain (Stage 5) began in June 2011 through radio com- which many audience members had not seen before. munication and IPC and will be maintained as the Global Scaling Up Handwashing Project is integrated into Scaling Arrangement and Strategy Up Rural Sanitation, a large-scale initiative that is being im- The implementation arrangement, led by WSP, included gov- plemented by the Government of Tanzania with technical as- ernment representatives, private firms, and community mem- sistance from WSP.8 bers (see Box 1). From the early planning stages, stakeholders were considered part of the team, and their input and buy-in Phase I: Awaken and Inspire was considered crucial. Introducing a behavior-change in- Phase I focused on increasing knowledge about the impor- tervention of this type required stakeholders to adopt a new tance of using soap, critical times for handwashing, and inspir- approach to hygiene interventions. In Tanzania, previous initia- ing women to prioritize handwashing. This phase targeted the tives had been largely didactic, education-driven approaches, knowledge portion of the ability determinant, based on forma- and it required considerable time to get stakeholders to accept tive research that showed key gaps in understanding about the a new approach. Efforts to engage stakeholders throughout importance of handwashing with soap. the process strengthened support, increasing opportunities for both scale-up and sustainability. Mass media, including radio spots, call-in shows, and DJ- mentions,9 focused on when to wash hands with soap. Pro- A marketing team from a private soap company developed a grams aired on national stations and one regional station five-stage, phased approach to help carry the target audience between February and April 2009. Handwashing posters from an initial stage, becoming aware of the importance of were also produced and distributed. handwashing, to building skills to conduct improved handwash- ing, to sustaining handwashing with soap behavior (Figure 4). IPC focused on building knowledge and skills to build tippy- Phase I, including Awaken (Stage 1) and Inspire (Stage 2) taps, managing availability of soap and water, and hand- was implemented from February 2009 to March 2010; Phase washing at critical times. 2, including Empower (Stage 3) and Amplify (Stage 4) was conducted from July 2010 to June 2011. Phase 3, focusing DCC events focused on improving knowledge around the ease of handwashing with soap and introducing the tippy- BOX 1: IMPLEMENTATION ARRANGEMENT tap, which many participants had not seen before. Events were carried out by two teams: “Pathfinders� targeted Activity Implementing Agency smaller communities with a program that included a sound Formative research Research firm truck, master of ceremonies, and dancers; �Handwashers� Research analysis WSP targeted larger villages with a program that included music, dancers, skits, and tippy-tap demonstrations. Events lasted Communications concept Private firm a couple of hours, but were filmed, quickly edited, and then Mass media production and Private firm shown again at night to reach a wider audience. An esti- planning mated 161,000 women and children (95 percent of the tar- DCC development and Private firm get figure) were reached in just seven months, starting in execution August 2009. IPC materials WSP IPC training (Training of WSP and private firm Several learnings emerged during this phase. First, FLAs were Trainers) trained to deliver messages using an emotive approach. Re- IPC delivery Community volunteers ports began to indicate, however, that FLAs were experiencing Monitoring WSP and private firms difficulties conveying knowledge and skills using an emotive approach. Messages were refined to create a more pragmatic 7 For more information on tippy-taps and other enabling technologies for handwashing with soap, see WSP’s Enabling Technologies for Handwashing Database at www.wsp.org/scalinguphandwashing/enablingtechnologies. 8 For more information, see www.wsp.org/scalingupsanitation 9 “DJ-mentions� are a subtle way to convey the handwashing message. For example, DJs invite callers to explain why their mothers are heroes. The caller might mention that their mother made sure they were well dressed every day, or that their mother took care of them when they were sick. The DJ might then segue to the handwashing message by mentioning, “Isn’t in wonderful how mothers do so much for us, like teaching us to wash our hands with soap?� www.wsp.org Figure 4: Campaign Stages Advertising Public Relations Partnerships Experiential Response Consumer Promotions www.wsp.org Purpose Salience Credibility Association Engagement Relationship Conversion Relevance via Engagement via Relationship via Role Persuasion via salience Influence via credibility Conversion via incentive association experience recognition Image Trust Values Interaction Trial Trial Channel Immediacy Rapid Response Content Immersion Loyalty Urgency Use Reassurance News Endorsement Depth Personi�cations Proximity Stature Endorsement Prestige Participation Reward Reward Stage Global Scaling Up Handwashing Project Radio and print on News coverage Promoting buy-in; DCC and IPC on key junctures for of campaign training local, national key junctures and 1 handwashing with in newspapers government on importance of soap Awaken soap and the handwashing with soap importance of soap DCC and IPC— constructing tippy-taps Wall branding and washing hands Radio phone-in with Soap packaging with Radio phone-in with Promotional items DCC and IPC— 2 handwashing with messages handwashing with soap constructing tippy-taps Inspire soap experts experts and washing hands and winning promotional items Radio soap opera IPC and DCC promoting Tippy-taps comic DCC events for mothers 3 mother as heroes ready as heroes ready for Empower for competition Radio soap opera competition Tippy-taps comic Radio sponsorship for News coverage Radio spots of Famous Engagement through Asante Mama through Asante Mama through Asante Mama of Asante Mama people saying why their media, DCC and IPC media, DCC and IPC media, DCC and IPC 4 mothers are heroes as on Asante Mama— Amplify part of Asante Mama. nominating mothers as heroes Handwashing with soap ambassador 5 Radio spots and soap Soap packaging IPC Sustain opera Tanzania: A Handwashing Behavior Change Journey 5 6 Tanzania: A Handwashing Behavior Change Journey Global Scaling Up Handwashing Project approach. Second, the interventions were effective at target- (motivation determinant). Asante Mama was integrated into the ing opportunity determinants (including access, availability, soap opera through the character of a hero-mother who is rec- and product), and ability determinants (knowledge, including ognized for her importance in the family, including handwash- awareness of tippy-taps and the critical times for handwash- ing with soap. Radio spots aired five times daily from July 2010 ing), but were less effective in addressing the motivation de- to June 2011; the soap opera aired two times per week from terminant (intention to wash hands with soap). In early 2010, November 2010 to June 2011. In addition to the radio spots WSP held a learning event to review the intervention and it was and soap opera, sponsorship hours were piloted in which DJs determined that, while IPC messages needed a more prag- asked listeners to call in and tell a story about why their mother matic approach, DCC and mass media should be adjusted to is hero. The DJs were tasked to link that story to handwashing emphasize more emotive messages. For example, a rational with soap. However, after one month, the sponsorships were benefit is that washing hands with soap can save money and cancelled because the DJs focused on rational reasons to prevent disease. An emotional benefit of handwashing with wash hands with soap rather than the Asante Mama concept, soap is to be a good mother and to be recognized as a hero. praising mothers for washing their hands with soap. Phase II: Empower and Amplify IPC included the development and distribution of a ‘comic To increase the emotive impact of the campaign, the team book’ that illustrated, step-by-step, how to build a tippy-tap. developed a creative brief centered on the concept of Asante A graphic approach was selected due to the low rates of lit- Mama, which acknowledged, praised, and thanked mothers for eracy in Tanzania. The storyline incorporated characters from the things they did every day for their family and children, includ- the soap opera, thereby integrating with mass media, and ing washing hands with soap. This phase of the campaign fo- was distributed by FLAs in 30 wards. cused on making handwashing an intrinsic behavior, increasing DCC targeted attitudes and beliefs, outcome expectations, and women’s understanding of the tippy-tap, and building skills to intention (all part of the motivation determinant), social norms better manage soap and water. (opportunity determinant), and social support (ability determi- Mass media targeted social support (ability determinant), and nant). A competition was held in ten districts in which audience beliefs and attitudes outcome expectations, and intentions members were invited to go on stage to share why their mother was their hero, with the winner selected by audience response. Each district winner won a trip to participate in Global Hand- Illustration 1: DCC Event in Igunga washing Day. DCC was integrated with IPC and mass media through skits that focused on how to build a tippy-tap; skits based on the radio soap operas; and playing radio jingles at events. At events, district and community officials were invited onstage and recognized, helping to strengthen political buy-in. DCC events reached an additional 166,000 women and chil- dren during this phase (Illustration 1). Wall paintings executed on community buildings depicted a woman teaching her child to wash hands using a tippy-tap. The painting reinforced the theme of Asante Mama to integrate with mass media and DCC, and served as permanent billboards. MONITORING WSP hired four District Coordinators to support district gov- ernments in carrying out the implementation. The District Coordinators supervised and monitored the intervention to make sure that activities were delivered as planned and that messaging stayed on topic. During Phase II, DCC roadshow events included tippy-tap demonstrations to help improve two determinants—opportunity Mass Media was monitored by a local media monitoring firm and ability. Above, women in Nanga, Igunga District use a tippy- that produced monthly reports showing when radio spots tap to wash hands with soap. and soap operas actually aired. www.wsp.org Global Scaling Up Handwashing Project Tanzania: A Handwashing Behavior Change Journey 7 IPC monitoring was done through FLA’s, on a self-reported Figure 5: Knowledge of Key Junctures for basis, using paper forms that were distributed to local gov- Handwashing with Soap ernment authorities. The forms were designed to record the 100 number and type of meetings held (e.g., household, commu- 80 Percentage nity, market, and so forth); who was reached (women ages 60 15–49, women over age 49, men, children); and how many 40 people were reached on a monthly basis. However, perhaps 20 because FLAs were unpaid community volunteers, only 67 0 percent of FLAs submitted forms, and data was not always After the Before Before Before After toilet eating preparing feeding cleaning complete. Additionally, there was a challenge validating re- food a child a child’s ported data. Interestingly, the FLAs who submitted forms bottom Key Junctures to local governments viewed it a means to demonstrate re- Before Event After Event sults of the work, and this helped to motivate them to record data. Additionally, WSP monitored FLA activities through spot-checks. DCC monitoring tracked progress in implementation and the Strengthening the Motivation of FLAs. Since FLAs are quality of the interventions. Implementing agencies also com- not paid, there is also less motivation for preferential selection. pleted a form after each event to record the size and compo- However, without compensation or an institutional structure, sition of the audience, topics covered, type of location, and ongoing volunteerism has lagged. Several challenges have prizes given away. Local government authorities and WSP staff emerged. First, FLAs feel that they should be compensated or monitored activities conducted by firms. Additionally, WSP at least better recognized by local authorities for their efforts. validated event data through spot checks and event impact The government cannot afford to provide the volunteers’ trans- surveys (EISs), developed by WSP, and undertaken to assess portation or reimbursement for their time. Second, volunteers how well the DCC firm transmitted messages. EISs assess the are constrained by transportation and limited geographic influ- impact of live behavior change communication events by mea- ence. The question then becomes how many volunteers are suring reception to the communication among members of the needed to train at scale, and the costs associated with train- target audience and changes in knowledge and intention. ing. More analysis is needed to understand scalability, costs, and what is financially viable for the government. The DCC firm administered the EIS before and after a random sample of 28 DCC events (25 percent of end-of-project target An alternative approach may be to develop a health-product for DCC events) in five districts. Results suggest that the tar- sales force with basic sales training and to provision them geted behavioral determinants, knowledge (ability determinant) with an initial batch of products such as soap, water treat- and intention (motivation determinant), had improved among ment tablets, and condoms. This would help build income those exposed to a DCC event (Figure 5). It is hoped that a gain and could promote FLAs’ status in the community. Another in these determinants will be reflected in improved behavior. option is to find other ways to motivate and provide incen- tives to FLAs. For example, in some villages, FLAs have been CHALLENGES elected as village representative because of their raised pro- Geographic Arrangement of the Intervention. Because file as an FLA. Overall, expectations must be managed up of poor transportation, the dispersed project areas created front about what is expected of a volunteer, and how they will many challenges, from delivering trainings to implementing be recognized for their work DCC events to supervising progress. Monitoring FLAs. Collecting data through the local govern- Developing the Creative Concept. The time needed to ment structure has proven challenging and it is likely that the develop a positive, professional, communications campaign figures reported are an underestimate of work conducted by was underestimated. It took over six months of working with FLAs. An estimated 67 percent of trained FLAs have submit- the advertising firm to develop a concept that met the criteria ted and delivered at least one form to the district government, in the creative brief and established emotive messages. This either directly or through their village and ward leaders. Due delay impacted the development of communication materi- to issues such as transportation, it is possible that an addi- als and timing of intervention across all channels. tional 20 percent of trained FLAs conducted work, but were www.wsp.org 8 Tanzania: A Handwashing Behavior Change Journey Global Scaling Up Handwashing Project Table 1: Project Reach in Tanzania Through December 2010 About the project Reported Global Scaling Up Handwashing is a Intervention Target Reach Percent Water and Sanitation (WSP) project Mass media 14.5 million 14.5 million 100% focused on applying innovative Direct Consumer Contact (DCC) 170,000 327,000 191% behavior change approaches Interpersonal Communication (IPC) 320,000 165,000 52% to improve handwashing with soap behavior among women of not able to submit forms to the district from government officials on their report- reproductive age (ages 15–49) and government. Furthermore, FLAs who ing as well as site visits to see progress. primary school-age children (ages have submitted forms may submit only Without this feedback FLAs are unlikely 5–9). It is being implemented by a small portion of their forms. Anec- to maintain motivation. local and national governments dotal evidence from discussions with with technical support from WSP FLAs suggests that the majority try to Message Creep. During DCC events, in four countries: Peru, Senegal, achieve their target to reach 75 women the implementing firm would naturally Tanzania, and Vietnam. For more a month. To rectify the situation, the start to play for laughs and key messages information, please visit www.wsp team planned a follow-up training with would gradually drop away. This can be .org/scalinguphandwashing. FLAs to review target numbers and ac- mitigated through independent supervi- tual reach. FLAs completed a report on sion by either the local government or Contact us the number of targets reached in the the hiring entity, through Event Impact For more information, please visit past month, and compared this against Survey results, which will show if the firm www.wsp.org or email Yolande previous months. is not delivering the message effectively, Coombes at wsp@worldbank.org and through message checklists. Value of Data. FLA monitoring forms record process indicators such as num- Results ber of people reached, location, and du- As of December 2010, the project has ration of visits but many officials place exceeded project targets for mass a low value on this data. This situation media and DCC, and reached 52 per- impedes the process of collecting the cent of the target for IPC (Table 1). forms at each level, decreasing the likeli- hood that FLA forms will be institution- —By Yolande Coombes alized. FLAs have requested feedback and Nat Paynter WSP is a multi-donor partnership created in 1978 and administered by the World Bank to support poor people in obtaining affordable, safe, and sustainable access to water and sanitation services. WSP’s donors include Australia, Austria, Canada, Denmark, Finland, France, the Bill & Melinda Gates Foundation, Ireland, Luxembourg, Netherlands, Norway, Sweden, Switzerland, United Kingdom, United States, and the World Bank. The findings, interpretations, and conclusions expressed herein are entirely those of the author and should not be attributed to the World Bank or its affiliated organizations, or to members of the Board of Executive Directors of the World Bank or the governments they represent. © 2011 Water and Sanitation Program