Research methodology for developing efficient handwashing options: an example from Bangladesh.

Handwashing has been universally promoted for health interventions, but it is essential that the factors related to behaviour are understood in order to develop appropriate handwashing messages. We found the study of behaviour complex and had to combine several methods: in-depth interviewing, questionnaire; observational and bacteriological. Here we present our experiences in developing efficient handwashing options for rural Bangladesh. The components of handwashing practices after defecation of 90 rural women were studied (phase 1). During phase 1 an in-depth interview was used to design the observational and questionnaire surveys. Behaviour was observed using a semi-structured record form and the effectiveness of the acts was measured by means of bacteriological tests. A questionnaire survey was undertaken on socioeconomic and water sanitation-related variables since they influence behaviour. Then, to develop efficient handwashing options, an experimental phase (phase 2) tested the bacteriological efficacy of the components found appropriate in phase 1. The effectiveness of the handwashing practices is believed to be poor since the bacteriological counts were found to be high (faecal coliform count of the left hand 1995, and of the right hand 1318 colony forming units/hand). The practice comprised several components: use of an agent, handedness, frequency of rubbing, source and volume of rinsing water, and drying of the hands. Seventy-five per cent of the women reported that they could not afford soap. The experimental trials showed that soap, ash and soil give similar results under similar conditions of handwashing (faecal coliform counts of left hands: 195 (soap), 98 (ash), 129 (soil) and of right: 112 (soap), 54 (ash) and 89 (soil) cfu/hand). The use of multi-method techniques in the study helped to understand and develop efficient handwashing options.