Promoting clean hands among children in Uganda: a school-based intervention using ‘tippy-taps’

Summary

Objective

To measure the efficacy of a tippy-tap-based handwashing programme in promoting handwashing rates in elementary schools in rural Uganda.

Methods

Pre-/postintervention surveys were fielded in eight schools (398 students, ages 7–13 years). Four intervention schools were given tippy-taps, soap and educational materials, while four control schools initially received only educational materials.

Results

After 1 month, the intervention schools reported a large increase in daily handwashing rates and absence of stomach pain episodes compared with the control schools. After receiving the intervention, the control schools attained similar handwashing and stomach pain rates.

Conclusions

A school-based tippy-tap handwashing programme improved handwashing rates, and has the potential to reduce the incidence of infectious diarrhoeal disease in children.

Diarrhoeal diseases are the fourth leading cause of child mortality in Uganda,¹ and the second leading cause globally.² Handwashing with soap can reduce the incidence of diarrhoeal disease by almost half.³ Significant challenges in promoting correct handwashing hygiene practices include successfully transforming knowledge into behaviour change among at-risk populations, such as young children, and ensuring affordability and availability of handwashing supplies and infrastructure. The number of child deaths could be cut dramatically if a low-cost and simple handwashing programme were widely promoted and practised. One possible low-cost solution is a ‘tippy-tap’; a simple, economical (US$2) handwashing station constructed from commonly available materials in underdeveloped, rural areas.⁴ Moreover, tippy-taps are suitable for implementation in water-scarce regions, as they only use 40–50 ml of water per handwash on average compared with 500–600 ml of water used in other methods of handwashing.⁵

A handwashing programme modelled on increasing access to low-cost, simple handwashing technologies, coupled with related education programmes and the provision of soap, may help to minimize the spread of diarrhoeal diseases, particularly in low-income and remote areas. This paper summarizes the design and preliminary findings from a comprehensive, child-led tippy-tap handwashing promotion programme implemented in rural schools in Nakigo, Iganga District, Uganda. In Iganga, children under 15 years of age comprise 54% of the total population,⁶ and the child and under-5 mortality ratios in 2004 were 64/1000 and 147.3, respectively (diarrhoeal disease was a leading cause of death⁷).

This study targeted students in Grades 2–5 from eight elementary schools with an average size of 900 students. Ethical approval was obtained from the University Institutional Review Board, and all participants in the study provided informed consent. Participating schools were randomly divided into four intervention schools and four control schools. The intervention schools were provided with tippy-taps, soap and a complementary educational programme designed to promote handwashing behaviour. The control schools initially only received the education programme.

The handwashing programme contains three components: handwashing education, construction of tippy-taps and provision of soap. The education component is centered on instructional lessons about the benefits, proper technique and critical times when handwashing should take place (e.g. after using the toilet, before and after eating, when hands are dirty). This includes poster presentations, a handwashing song, distribution of flyers and discussions with students about handwashing with soap. All educational materials were translated from English into Lusoga, the local language. Tippy-tap construction materials (plastic jerry cans, wood, string, nails and a hammer) were available locally. Five tippy-taps were constructed and installed near boys' and girls' toilets at each intervention school. Students constructed the tippy-taps (under adult supervision) and were assigned maintenance duties by teachers. Only the study participants constructed the tippy-taps, were given the educational programme, and assigned the maintenance duties. Soap was purchased in bulk from local markets (25 bars for US$13), and a 1-month supply (100 bars) was provided for every five tippy-taps.

The data for this study were obtained using pre- and postintervention surveys designed to solicit information on the programme's effectiveness as well as behaviour and attitudes regarding handwashing. At each school, one classroom was selected at random (lottery draw), and 25 boys and 25 girls (Grades 2–5) were selected from that classroom to be given surveys using a systematic random sampling design (every third girl and boy) (398 students participated, with complete data available on 391 students). The same students received the pre-/postintervention surveys.

The data collection waves, three in all, were spaced 1 month apart. At Time 1, baseline surveys were implemented in all schools. The baseline survey was designed to gather attitudinal and behavioural data on personal sanitation at school and home, and self-reported data on stomach pain episodes in the past month. Between Time 1 and Time 2, the intervention schools received tippy-taps, soap and a handwashing education programme, while the control schools only received the education programme. At Time 2, the intervention schools were given the postintervention survey and the control schools repeated the pre-intervention survey. The postintervention survey was used to gather data on handwashing behaviour, use of tippy-taps, stomach pain and student–parent idea dissemination. Between Time 2 and Time 3, the control schools received tippy-taps and soap. Finally, at Time 3, the postintervention survey was administered in control schools. Stata 11 software was used to perform the statistical analyses.⁸ Table 1 reports the distributional characteristics for the five main measures. Statistical comparisons of pairs of group means for the control (Control¹/Control², Control²/Control³) and intervention sites (Intervention¹/Intervention²) were performed using paired Student's t-test, and unpaired Student's t-test was used to compare the control and intervention sites at baseline (Control¹/Intervention¹). As this was a cluster randomized study, where a cluster is defined as a school, design effects were adjusted for in the analyses. Only participants with complete data were analysed (n=191 for control schools and n=200 for intervention schools).

Table 1. Handwashing and stomach pain in control and intervention schools pre- and postintervention.

	Pre-intervention (baseline)		Pre-intervention	Postintervention
	Control1	Intervention1	Control2	Intervention2	Control3	95% CIa
Total	198	200	195	200	193
Do you wash your hands at school?
Always	11 (5.6)	2 (1.0)	12 (6.2)	80 (40.0)	155 (80.3)
Often	48 (24.2)	5 (2.5)	10 (5.1)	120 (60.0)	37 (19.2)
Sometimes	88 (44.4)	175 (87.5)	103 (52.8)	0 (0.0)	1 (0.5)
Never	51 (25.8)	18 (9.0)	70 (35.9)	0 (0.0)	0 (0.0)	ABD
Do you wash your hands before using the latrine/toilet?
Always	9 (4.5)	11 (5.5)	7 (3.6)	130 (65.0)	153 (79.3)
Often	69 (34.8)	55 (27.5)	39 (20.0)	68 (34.0)	37 (19.2)
Sometimes	87 (43.9)	123 (61.5)	137 (70.3)	2 (1.0)	3 (1.6)
Never	33 (16.7)	11 (5.5)	12 (6.2)	0 (0.0)	0 (0.0)	BD
Do you use soap when washing your hands?
Yes	54 (27.3)	27 (13.5)	25 (12.8)	169 (84.5)	193(100.0)
No	144 (72.7)	173 (86.5)	170 (87.2)	31 (15.5)	0 (0.0)	ABD
How many times a day do you wash your hands?
Zero	9 (4.5)	9 (4.5)	18 (9.2)	0 (0.0)	0 (0.0)
One	145 (73.2)	98 (49.0)	133 (68.2)	1 (0.5)	0 (0.0)
Two	24 (12.1)	82 (41.0)	27 (13.8)	13 (6.5)	4 (2.1)
Three or more	20 (10.1)	11 (5.5)	17 (8.7)	186 (93.0)	189 (97.9)	ABD
In the past month, how many times have you experienced stomach pain?
Never	10 (5.1)	14 (7.0)	8 (4.1)	160 (80.0)	142 (73.6)
Once	61 (30.8)	93 (46.5)	80 (41.0)	25 (12.5)	43 (22.3)
Twice	97 (49.0)	50 (25.0)	62 (31.8)	14 (7.0)	8 (4.1)
More than twice	30 (15.2)	43 (21.5)	45 (23.1)	1 (0.5)	0 (0.0)	BD

Data shown as n (%) (total sample n=398).

^aThe following letter codes are used to summarize significant differences between samples at P<0.05 [95% confidence interval (CI]: A=test of Control¹ against Intervention¹ (baseline differences); B=test of Intervention¹ against Intervention² (intervention effect); C=test of Control¹ against Control² (education programme effect in control sites); and D=test of Control² against Ccontrol³ (intervention effect in the control sites). (n for Control¹/Intervention¹=191/200; Intervention¹/Intervention²=200/200; Control¹/Control²=191/191; Control²/Control³=191/191). Stata statistical software was used for all analyses and controlled for design effects of the study (i.e. strata, clusters and weights).

Comparisons of pre- and postintervention data from intervention schools indicated that the frequency of handwashing increased (Comparison B in Table 1). Both handwashing at school and after using the toilet increased after the introduction of tippy-taps. The proportion of students reporting ‘always’ or ‘often’ washing their hands at school increased from 3.5% at baseline to 100.0% at follow-up {t=19.54, P<0.05, 95% confidence interval (CI) 1.21–1.68 in the intervention schools [replicated in control schools by Time 3 (t=12.92, P<0.05, 95% CI 1.48–2.45] – Comparison D in Table 1}. The proportion of students ‘always’ washing their hands after using the toilet increased from 5.5% to 65.0% (t=14.61, P<0.05, 95% CI 1.02–1.58) in the intervention schools [washing hands after using the toilet among students in the control schools increased from 3.6% to 79.3% (t=13.21, P<0.05, 95% CI 1.16–1.90) by Time 3]. Use of soap in the intervention schools increased from 13.5% to 84.5% (t=5.64, P<0.05, 95% CI 0.29–1.04) with even higher proportions reported at control schools at Time 3 (t=298.15, P<0.05, 95% CI 0.86–0.88). In the intervention schools, the proportion of students reporting washing their hands three or more times/day increased from 5.5% to 93.0% (t=9.84, P<0.05, 95% CI 0.98–1.91) after the installation of tippy-taps. Furthermore, the control schools also attained the handwashing rates of the intervention schools (97.9%) by Time 3 (t=18.47, P<0.05, 95% CI 1.42–2.01). Proxy data on the incidence of diarrhoeal disease are indicated by the number of students reporting stomach pain episodes in the previous month. In the intervention schools, the percentage of students reporting no stomach pain episodes increased from 7.0% to 80.0% (t=10.84, P<0.05, 95% CI 0.92–1.68).

These findings provide evidence that a tippy-tap promotion programme can potentially serve as a successful, low-cost model for handwashing initiatives in remote, rural and low-resource school settings. The provision of tippy-taps, soap and a simple education programme can increase handwashing rates and soap usage, and reduce stomach pain episodes among elementary school-aged students. The placement of tippy-taps near toilets can effectively disrupt bacterial transmission and cue handwashing behaviour. These data suggest that education alone is insufficient to induce short-term behaviour change (Comparison C effects were not observed in Table 1). A 2012 paper on the evaluation of a large-scale sanitation, hygiene and water improvement programme conducted in rural Bangladesh also found that education did not improve handwashing behaviour.⁹ Without immediate access to a handwashing station and soap, students did not efficiently translate newly acquired knowledge on hygiene into action.

It is important to note that changing the handwashing behaviour of school-aged children may be less challenging than changing the behaviour of adults. To enhance the learning process, this tippy-tap handwashing programme allowed students to lead the construction and maintenance of tippy-taps. This appeared to foster active participation in the health learning process, and may have motivated adoption of healthy behaviours and utilization among students. Lessons on setting up and maintaining their own handwashing station grant school children a sense of self-efficacy, which can accelerate learning.¹⁰ Another possible explanation for the success of the intervention is that students enjoy using tippy-taps because they are fun to use compared with other methods (in the postintervention survey, 98.7% of students reported that using a tippy-tap was fun). Moreover, 99.7% of respondents also reported that they saw friends washing their hands using tippy-taps, suggesting that peer influence may motivate other school children to wash their hands.

The postintervention survey found that students were telling their parents about tippy-taps (98.7%), and research staff directly observed many students mobilizing their families to donate tippy-tap supplies. This suggests that school children can act as agents in the dissemination of health knowledge to their homes. Health knowledge transfer from students to parents is a phenomenon that can improve handwashing techniques in the household while decreasing student absentee rates.¹¹

The implementation of this school-based programme is low cost and not only provided for the 398 study participants but also the entire school population (approximately 7200 students). After the initial construction of five tippy-taps (US$2 each), the monthly maintenance cost per school is equivalent to the market cost of soap ($52/month/900 students; less than US$0.06/student/month. Tippy-taps do break after frequent use; however, the simplicity of tippy-taps makes them easy and cheap to fix.

This study had limitations. The project was low cost and small scale, limiting the range of questions asked, the modes of data collection, and importantly the authors' ability for long-term follow-up. The study relied on student self-report data, and the authors were not able to perform structured observations or collect clinical data. Previous studies have illustrated that health and hygiene questionnaire surveys on sanitation behaviours can be unreliable, and can lead to over-reporting of positive or ‘desirable’ behaviours.^12,13 However, mothers proxy-reporting hygiene behaviours may have a greater tendency to over-report positive behaviours compared with children reporting their own behaviour.¹⁴ The added costs of comprehensive structured observations in the target schools were prohibitive, but supplementing the self-report data with direct observations of handwashing behaviours in the target population will be incorporated into future studies. This study relied on self-reported behaviours; however, significant differences between intervention and control groups at Time 2 suggests that students may not routinely over-report desirable behaviours.

Furthermore, while schools were carefully selected, they are all in one sub-district and the possibility of information transfer and influence between control and intervention populations did exist (pre-intervention data from control schools at Time 1 and 2 suggest that this may be a minor concern, although an education programme was administered in the control sites). Differences in behaviour between control and intervention schools at baseline could be attributed to the presence of an Islamic school in the control group, in which religious practices could account for increased handwashing behaviour and soap usage.

To the best of the authors' knowledge, published literature on the use of tippy-taps has focused on adult use in home settings and most have been qualitative in nature.¹⁵ Furthermore, many handwashing interventions are focused on identifying key determinants,³ and are targeted towards advising members of the household. Few interventions have provided soap and included demonstrations and supervised practice.¹⁶ This study provides quantitative evidence for an effective low-cost handwashing promotion programme specifically targeted towards changing handwashing behaviour in children. Future research goals include educational programme modification, direct handwashing observation, focus group analysis, a more detailed study on differences in handwashing practices between older and younger children, and the design of objective trials to measure the impact of washing hands on diarrhoea. Successful long-term adoption of handwashing behaviours among the children will also require further research on strategies to overcome structural impediments, such as handwashing infrastructure. For example, elucidating cost-effective methods to maintain the provision of soap to low-income communities is critical, as even the minimal costs of supplying soap are a hardship in cash-strapped school systems such as in Uganda. Low-cost solutions that can maintain handwashing infrastructure can increase the scalability and long-term impact of handwashing interventions, and also allow the rural poor to channel resources to other priorities.

Handwashing with soap can save millions of children's lives. Handwashing promotion may become the intervention of choice if handwashing rates are low and if promotion programmes can be low cost and effective. This study reports one such low-cost health promotion programme – tippy taps coupled with education and soap – which can encourage handwashing behaviour and promote health outcomes for children in rural, resource-limited settings.