WATER AND SANITATION PROGRAM: WORKING PAPER 56945 Global Scaling Up Handwashing Project Practical Guidance for Measuring Handwashing Behavior Pavani Ram, MD March 2010 The Water and Sanitation Program is a multi-donor partnership administered by the World Bank to support poor people in obtaining affordable, safe, and sustainable access to water and sanitation services. By Pavani Ram, MD University at Buffalo, The State University of New York The following individuals, in alphabetical order, have contributed to the thinking presented in this document: Adam Biran, Anna Bowen, Val Curtis, Jacqueline Devine, Stewart Granger, Orlando Hernandez, Steve Luby, Jack Molyneaux, Eddy Perez, and Wolf-Peter Schmidt. Sincere thanks go to them for their constant interest in this topic and their intellectual generosity. Global Scaling Up Handwashing is a Water and Sanitation Program (WSP) project focused on learning how to apply innovative promotional approaches to behavior change to generate widespread and sustained improvements in handwashing with soap at scale among women of reproductive age (ages 15­49) and primary school-aged children (ages 5­9). The project is being implemented by local and national governments with technical support from WSP. For more information, please visit www.wsp.org/scalinguphandwashing. This Working Paper is one in a series of knowledge products designed to showcase project findings, assessments, and lessons learned in the Global Scaling Up Handwashing Project. This paper is conceived as a work in progress to encourage the exchange of ideas about development issues. For more information please email Pavani Ram at wsp@worldbank.org or visit our website at www.wsp.org. 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. WSP reports are published to communicate the results of WSP's work to the development community. The boundaries, colors, denominations, and other information shown on any map in this work do not imply any judgment on the part of the World Bank Group concerning the legal status of any territory or the endorsement or acceptance of such boundaries. © 2010 Water and Sanitation Program Executive Summary Accurate measures of handwashing behavior are critical to Structured Observations have been used frequently in the understanding households' health environment. But it can handwashing literature. These observations require trained be challenging to measure handwashing reliably. This docu- observers to watch and record household handwashing and ment discusses a set of handwashing indicators and recom- related behaviors, and yield details about handwashing at mendations that was prepared to support the Water and critical times, such as after defecation. But they are costly Sanitation Program's six-country scaling-up of handwash- and their validity has been recently questioned because of ing promotion and community-led total sanitation. reactivity on the part of those observed. This preliminary Descriptions of these measures, and the recommendations evidence warrants caution but structured observations for their use (Table 1), should be of interest to a broad remain relevant to handwashing measurement because of audience. the rich details yielded by them. The following handwashing measures are assessed based on Bars of soap with motion sensors provide an objective record their validity, reliability, and efficiency: of the number and timing of soap-use events. The method is still relatively untested, but it is promising in select set- Self-reports are the easiest way to measure handwashing. But tings, despite facing several challenges. The method is they are invalid as measures of handwashing behavior be- relatively expensive, because of specialized hardware and cause individuals often report better handwashing behavior personnel costs. The sensors may only be useful if house- than they display during observation. This exaggeration of holds typically use soap bars for personal hygiene, as true behavior may result from a perceived high social desir- opposed to powder soap or bar soap used for multiple pur- ability of handwashing. However, self-reports remain an poses. The data from soap with motion sensors do not gen- important source of information about handwashing erate respondent-specific information, nor do they inform knowledge and other determinants of handwashing about rates of handwashing with soap at critical times, such behavior. as after defecation. In spite of these caveats, the motion sen- sor yields objectivity and reliability to soap use measure- Rapid household observations include several easily collected ment and, thus, further evaluation is clearly warranted. valid and reliable indicators. These include observations on the availability of soap and water, and the presence of these Combined use of structured observations and motion sensors tools at dedicated handwashing stations. While these indi- permits analysts to link the timing of observed soap-use cators do not directly indicate handwashing behavior, they events to the events recorded by the motion sensors, and to are currently used as surrogate markers because they are reli- extend what is learned to periods outside of the structured able and efficient. But evidence of how well they predict observation. This permits analysts to study context-specific actual handwashing behavior and disease risk is still soap use, while also enabling them to distinguish soap use forthcoming. frequencies during the structured observation from fre- quencies observed during corresponding hours on unob- Microbiological measures of hand contamination are objective served days. But this approach incurs the combined costs of measures of hand contamination, and consequently would the two most costly measures considered. seem desirable. However, this is currently a costly way to assess hand cleanliness. Moreover, hand contamination in- Based on these assessments, the following recommenda- dicators have been found to be unrelated to observed hand- tions are made according to the nature of the study, includ- washing behavior. Furthermore, reliability is a challenge. ing well-funded projects, projects with minimal funding, However, if the cost of such measurement can be decreased and mixed-purpose, large population-based surveys. and reliability improved, hand microbiology may eventu- ally be useful for measuring household environmental For well-funded projects, the most rigorous methods contamination. should be considered. These include the combined use of www.wsp.org iii Practical Guidance for Measuring Handwashing Behavior Executive Summary structured observations and soap with motion sensors, advisable to consider conducting structured observations, along with rapid observations. Questionnaires may also be and soap with motion sensors from small samples of house- used to learn about knowledge and other determinants of holds. Rapid observations and self-reported questionnaires handwashing. Hand contamination measures should be remain the cheapest source of household information. considered, as much is still to be learned of them. These Rapid observations are markers for actual behavior, and well-funded studies should also continue to measure health self-reports may be used to measure knowledge and other outcomes to better document the relationship between possible determinants of handwashing behavior. measured handwashing behaviors and health outcomes. For mixed-purpose, large population surveys, such as the Studies with minimal funding, which need affordable yet Demographic and Health Survey (DHS) or the Multiple reliable methods to monitor handwashing behavior, may Indicator Cluster Survey (MICS), where handwashing is warrant an investment in sample size estimates by a statisti- only one of many behaviors of interest, rapid observations cian or epidemiologist. These investments can frequently are recommended as the most efficient method of measur- pay for themselves, as sample needs are frequently much ing handwashing behavior. lower than expected. As part of these evaluations, it is iv Global Scaling Up Handwashing Table of Contents Executive Summary.................................................................... iii I. Introduction ................................................................................ 1 II. Methods of Measuring Handwashing Behavior ......................... 2 Self-Report .............................................................................. 2 Proxy Measures: Measurement of Microbiological Hand Contamination ................................................................ 2 Proxy Measures: Rapid Observations ....................................... 4 Direct Measures: Structured Observations ............................... 5 Direct Measures: Soap with Motion Sensors ............................ 7 Use of Composite Measures .................................................... 8 III. Conclusion................................................................................ 10 IV. Recommendations for Various Scenarios ................................ 13 Well-Funded Handwashing Promotion Programs or Research Studies............................................... 13 Handwashing Promotion Programs with Minimal Funding ..................................................................... 14 References ............................................................................... 15 Table 1: Summary of Strategies to Measure Handwashing Behavior ............................................................................... 11 Boxes 1: Indicators That Could Be Captured by Self-Report.................. 3 2: Indicators That Could Be Tracked Using Rapid Observations of the Household ............................................... 5 3: Indicators That Could Be Tracked Using Structured Observations .......................................................................... 7 4: Indicators That Could Be Tracked Using SmartSoap ............... 8 www.wsp.org v I. Introduction Advocates of promoting handwashing with soap agree that defecation, but not in other critical times, such as before this behavior has important health benefits across the globe.1 feeding a child. Thus, summarizing an individual's overall Efforts, both large-scale and focused, are underway world- handwashing behavior requires taking into account varia- wide to promote handwashing with soap at the community tions in behavior at different critical times. Moreover, an level (www.globalhandwashing.org). Although there is individual may be inconsistent in her behavior, for example broad agreement about the health benefits of promoting washing hands with soap after defecation sometimes but handwashing with soap, there is not similar agreement about not always; such variation in reliability also makes assigning the best ways to measure the behavior that these promotion an individual to a category such as "handwasher" or "non- programs set out to change. The reality is this: there is no handwasher" under-informative. Furthermore, both re- universally applicable method for measuring handwashing ported and observed markers of handwashing behavior behavior that is valid, relevant, affordable, and logistically have been found to be significantly associated with socio- feasible for the various settings in which such behavior might economic status, making adjusting for this important ex- need to be measured. The aim of this document is to de- planatory factor extremely important.3 scribe techniques and to propose strategies for measuring handwashing behavior for a variety of scenarios. Described below are the positive and negative attributes of various commonly applied and novel methods of measur- The measurement of handwashing behavior is relevant to ing handwashing behavior. Both self-reported and observed research studies and public health programs for evaluation measures are described. of the effectiveness of handwashing promotion with respect to behavior change and assessment of the role of hand hy- giene in pathogen transmission and disease burden. As for other outcomes of interest to public health, measurements of handwashing behavior should be scrutinized with respect to the following criteria:2 · Validity: "an expression of the degree to which a measurement measures what it purports to measure" · Reliability: "the degree to which the results obtained by a measurement . . . can be replicated" · Efficiency: "the effects or end results achieved in relation to the effort expended in terms of money, resources, and time" All measurement of handwashing is challenged by the com- plexities of this human behavior. An individual may wash hands with soap in the context of some critical times for pathogen acquisition or transmission, such as after 1 Curtis 2003; Curtis and Cairncross 2003; Rabie and Curtis 2006. 2 Last 2001. 3 Luby and Halder 2008. www.wsp.org 1 II. Methods of Measuring Handwashing Behavior Self-Report related to handwashing behavior. Describing changes in The easiest way to measure handwashing behavior is to use knowledge of appropriate handwashing behavior may be a questionnaire to ask the respondent directly about her be- useful as part of monitoring a handwashing promotion havior (see Box 1). This format is efficient, since the infor- campaign that proposes to increase knowledge of hand- mation can be gathered quickly, among a large number of washing in the target population. households, at relatively low cost. Unfortunately, awareness of the social desirability of handwashing may result in an Proxy Measures: Measurement of individual's overestimation of self-reported handwashing Microbiological Hand Contamination behavior. This overestimation has been demonstrated re- Measurement of microbiological contamination of hands peatedly, when self-reported behavior has been compared to is another proxy measure of handwashing behavior. The observed behavior.4 For example, in a study by Manun'Ebo underlying assumption is that hands that are washed with et al. (1997), the frequency of washing hands with soap and soap will be less contaminated with fecal organisms than water before eating was reported by 14 percent of respon- hands that are not washed with soap. The details of mea- dents but observed for only 2 percent of respondents.5 In suring hand contamination, e.g., by fingertip rinses or Bangladesh, whereas 77 percent of respondents reported hand imprints on semi-solid media, among others, are be- washing hands with soap or ash after defecation, only 32 yond the scope of this paper but are covered in numerous percent were observed to do so.6 Kappa score analysis, a sta- peer-reviewed publications.7 As a proxy measure, hand tistical tool to describe agreement between two different contamination offers a level of objectivity greater than measures of a construct, has been used in several studies; a self-report. In Bangladesh, hands tested immediately after kappa score less than 0.20 is considered to indicate "poor thorough washing with soap have been found to have sub- agreement" in the epidemiological literature. Stanton et al. stantially lower contamination with fecal coliforms than (1987), and Biran et al. (2008), have each shown that there unwashed hands,8 although even this finding is challenged is poor agreement between reported behavior and observed by other research, in which there has been no reduc- behavior, with kappa scores of 0.11 and 0.10, respectively, tion in bacterial contamination between pre- and post- for reported and observed measures of handwashing behav- handwashing measurements.9 ior after fecal contact. These studies, therefore, indicate that self-report is likely an invalid measure of true handwashing In a recent study in Bangladesh, Ram and colleagues com- behavior. On the other hand, questionnaires may be used to pared hand contamination as tested at random to hand con- elicit information relevant to behavioral factors that may tamination tested at critical times when pathogens may be facilitate or impede handwashing; such determinants may passed from hands to a child or to a vehicle such as food.10 include attitudes and beliefs, as well as logistical factors There was no significant correlation between results of hand such as access to adequate quantities of water. One set of contamination testing at random and at critical times. Even determinants that is commonly measured is knowledge though there was a linear relationship between hands tested at 4 Manun'Ebo, et al. 1997; Stanton, et al. 1987; Biran, et al. 2008; Health and Science Bulletin 2008. 5 Manun'Ebo, et al. 1997. 6 Health and Science Bulletin 2008. 7 Pinfold 1990; Luby, et al. 2001; Hoque, et al. 1995; Luby, et al. 2007. 8 Hoque, et al. 1995. 9 Larson, et al. 2003. 10 Ram, et al. 2008. 2 Global Scaling Up Handwashing Practical Guidance for Measuring Handwashing Behavior Methods of Measuring Handwashing Behavior BOX 1: INDICATORS THAT COULD BE CAPTURED BY SELF-REPORT Knowledge · Knowledge of importance of washing hands with soap to prevent disease · Knowledge of critical times to wash hands with soap 0 After defecation 0 After contact with the child's stool 0 After going to the toilet 0 Before preparing food 0 Before eating 0 Before feeding a child 0 Before handling water for storage Other possible determinants of handwashing behavior that can be measured by questionnaires relate to the op- portunity to access handwashing tools (e.g., access to soap and water near a latrine), ability (e.g., capacity to ensure access to steady supply of soap), and motivation (e.g., beliefs about the importance of soap). Appropriate measurement of these and other possible determinants can be maximized by reliance on a clearly considered framework for handwashing behavior change. Since this document focuses on measurement of handwashing behavior, a review of frameworks to promote handwashing behavior change is beyond its scope. Behavior ( hevalidityoftheseindicatorsisindoubt) T UnpromptedMeasures · Self-reported handwashing with soap during previous 24 hours (e.g., `Since this time yesterday, did you wash your hands with soap?) · Self-reported handwashing with soap at critical times (e.g., `Under which circumstances did you wash hands with soap?') PromptedMeasures · Frequency of handwashing with soap (e.g., always, often, rarely, never) 0 Frequency of handwashing at critical times After fecal contact events After defecation After cleaning a baby's bottom after the baby has defecated Before food-related events Before eating Before feeding a child Before cooking, cutting, or preparing food Before water-related events Before retrieving water from a wide-mouthed water storage container two different random times, the mean absolute differences in contamination on an individual's hands varies greatly within the actual counts of fecal coliforms and E. coli between the the course of several hours and, thus, reliability is not achieved. two random times confirmed substantial variability in mea- Therefore, the validity of single-point hand contamination as surements of hand contamination. That is, the level of hand a measure of overall handwashing behavior may be poor. www.wsp.org 3 Practical Guidance for Measuring Handwashing Behavior Methods of Measuring Handwashing Behavior It is likely that duration since last handwashing with soap, to ascertain whether soap is readily available for handwash- duration since last fecal contact, and overall fecal contami- ing is to record the amount of time needed for the respon- nation of the environment all impact upon the level of con- dent to bring soap to the interviewer when asked; if less tamination detected on a subject's hands. Also, in some than one minute is required, that could indicate the ready cultures, an individual's two hands may have different levels availability of soap. Since rapid observation measures are of contamination, since left hands may have more fecal objectively recorded and relatively straightforward, validity contact than right hands.11 Currently, measuring hand and reliability are preserved. contamination is relatively expensive; in Bangladesh, laboratory-based microbiological testing for fecal coliforms In a recent study completed in India, the proportion of and E. coli costs ~US$10 per individual, well beyond the households observed to have soap in the household (beside means of most program monitoring and evaluation bud- the latrine or in the yard) was similar to the proportion of gets. Less expensive and field-friendly methods of measur- mothers in those households observed to wash hands prop- ing fecal coliforms and E. coli on hands are being sought erly.12 However, there was only slight agreement, based on and may prove to be more feasible methods of testing for kappa scores, between observation of soap in the home and hand contamination (M. Sobsey, University of North observation of the mother washing hands with soap.13 In Carolina, personal communication). this study, proper handwashing during structured observa- tion was defined as washing both hands with soap after all Currently, substantial variability in the results of serial fecal contact events. While the validity of structured obser- microbiologic testing of hands from the same individual in vation as the basis of comparison may be called into ques- our recent study in Bangladesh suggests that single-point tion, given the potential for reactivity to the observer and hand contamination is a poor measure of handwashing the definition of soap use may be overly restrictive, behavior. Given the relative expense of this approach at these findings do underscore the point that rapid observa- present and challenges to its validity as a measurement of tions are surrogate markers of behavior and, thus, cannot overall handwashing behavior, it is not recommended that indicate with certainty the handwashing behavior of the in- hand contamination tests be built into routine monitoring dividual or the household. The presence of soap and water and evaluation of handwashing promotion programs at this at a designated handwashing place cannot confirm the fre- time. Further study to refine microbiology as a measure of quency or consistency of handwashing with soap for the overall handwashing behavior may enhance the utility of individual or the household, or whether hands are washed this approach in the future. during critical times such as after defecation. Additionally, rapid observations of the household do not provide infor- Proxy Measures: Rapid Observations mation on the handwashing behavior of an individual of Observations of the household environment can be effi- interest, such as the mother of a young child. Instead, they cient means to gather clues about the household's hand- may only provide information about the household as a washing behavior since they can be rapidly collected in a whole, since, in most households, soap is a communal large number of households, and at relatively low cost. resource and not an individual one. Still, since handwashing These rapid observations provide useful information on behavior tends to be socially mediated, household-level whether or not soap is present in the home, whether the measurement may be very useful for describing handwash- household has a designated place for handwashing, and ing behavior of a population. whether the tools required (i.e., soap and water, or mud/ash and water) are simultaneously in place to practice the be- One approach to using rapid observations to obtain clues havior for the individual that chooses to do so. Another way to individual behavior is to ask the individual of interest 11 Hoque, et al. 1995. 12 Biran, et al. 2008. 13 Ibid. 4 Global Scaling Up Handwashing Practical Guidance for Measuring Handwashing Behavior Methods of Measuring Handwashing Behavior to demonstrate usual handwashing practice. Here, too, awareness of social desirability may prompt improved BOX 2: INDICATORS THAT COULD BE TRACKED handwashing practice during the demonstration com- USING RAPID OBSERVATIONS OF THE HOUSEHOLD pared to usual behavior. Of note, in the study from India Hand cleansing agents include soap, ash, or mud, described above, there was a fair degree of agreement depending on the cultural context and the focus of between observation of soap use when the mother was the handwashing promotion program (e.g., soap asked to demonstrate her usual handwashing routine and specifically or any cleansing agent): observation of both hands being washed with soap after · Presence of soap anywhere in the home all fecal contact events witnessed during structured · Procurement of soap in the home within one observation.14 minute of interviewer's request · Presence of a designated place to wash Rapid observations are now being widely used to capture hands handwashing behavior (Box 2). For example, the Rapid · Presence of a designated place to wash CATCH indicators used by the U.S. Agency for Interna- hands with water available at the time of tional Development (USAID) child survival grantees in- inspection clude the measurement of the presence of soap at the · Presence of a designated place to wash location where hands are usually washed.15 hands with a hand cleansing agent, such as soap, available at the time of inspection Direct Measures: Structured Observations · Presence of a designated place to wash As noted above, self-reported handwashing behavior has hands with a hand cleansing agent and been proven invalid when compared with observed behavior. water available at the time of inspection In the literature, this observed behavior has mainly been · Presence of a designated place to wash recorded during continuous structured observation. Such an hands, in or near the sanitation facility, with observation consists of the placement of an observer for sev- a hand cleansing agent and water available eral hours, typically between three and seven hours, in a at the time of inspection household.16 The observer records opportunities for hand- · Presence of a designated place to wash washing, such as feeding a child or visiting the toilet, and the hands, in or near the food preparation area, target respondent's handwashing practices. The benefits of with a hand cleansing agent and water structured observation are the ability to record objective data · Use of soap to wash hands following a re- on handwashing practices and the richness of information quest to demonstrate usual handwashing gathered.17 During structured observation, the observer has behavior the opportunity to record information about numerous indi- viduals of interest, including mothers, young children, non- caregiver males, etc. Additionally, the observer can record handwashing practice, based on observation of behavior dur- detailed information on particular critical times, whether ing multiple opportunities for handwashing.18 hands are washed, whether both hands are washed, the type of cleansing agent used, and the way in which hands are The utility of structured observation in detecting overestima- dried. This richness of details allows for assessment of con- tion of self-reported handwashing behavior has been demon- sistency in handwashing practices. Respondents may be strated. However, the same awareness of social desirability that assigned to categories representing degrees of appropriate likely results in overestimation of self-reported handwashing 14 Ibid. 15 See http://www.childsurvival.com/kpc2000/kpc2008.cfm. 16 Biran, et al. 2008; Bentley, et al. 1994; Curtis, et al. 2001. 17 Bentley, et al. 1994. 18 Biran, et al. 2008. www.wsp.org 5 Practical Guidance for Measuring Handwashing Behavior Methods of Measuring Handwashing Behavior behavior may also result in reactivity to the presence of the handwashing promotion, as evidenced by structured obser- observer during a structured observation.19,20 That is, an indi- vation, with or without reactivity, it would be clear that the vidual may practice better handwashing behaviors while an handwashing promotion intervention did not result in sub- observer is present than when she is unobserved. A key factor stantial behavior change among the target population. in minimizing reactivity to structured observation is the infor- mation given to the target respondent in advance of the obser- The use of structured observation for measurement of hand- vation. During the informed consent process, and in other washing behavior can incur substantial costs in terms of per- verbal and non-verbal communications by the study worker sonnel time. It is preferable to use highly trained staff who are and observer, it is critical not to emphasize that the observation experienced in behavioral observation methods to perform is principally about measuring handwashing behavior. With- observations, or at least train other observers.23 Intensive out violating the respondent's right to understand the nature of standardized training for observers should emphasize details the study, the evaluator may indicate that the observation is such as the time frame within which handwashing would be aimed at understanding "general household practices." considered associated with a particular critical time (e.g., the number of minutes after defecation that handwashing oc- In a study in Bangladesh, the goal of observation was curs) as well as observational techniques, such as the need to described as the measurement of water, sanitation, and utilize neutral body language and avoid judgment or prompt- hygiene practices in the home.21 The observer's training and ing of socially desirable behaviors). Ideally, the time frame for skills with respect to objective data collection may also im- observation would be based on local knowledge of the timing pact reactivity on the part of the individuals being observed. of behaviors of interest. If handwashing after defecation is the With the use of soaps with motion sensors described in behavior of interest, then timing the observation to ensure more detail below, there was substantial reactivity to struc- that the observer is present in the home when most people tured observation, particularly in about one-third of the defecate (early in the morning in many cultures) would be study population.22 The reactive subset was characterized very important. This is not always feasible due to safety or markers of high socioeconomic status. Reactive households logistical concerns, thus necessitating structured observation were also much more likely to have soap available at a des- at other times of day. Over the span of several hours, an ob- ignated handwashing location near the toilet, suggesting server can only complete observation in one household. In that they may, indeed, prioritize hand hygiene more than the Bangladesh study, where five-hour structured observa- non-reactive households. The risk with reactivity, either to tions were compared to 90-minute structured observations, structured observation alone, or to the combination of shorter observation periods resulted in a greater-than-pro- structured observation and the soap with motion sensor is portional loss in observed numbers of defecation-related that the evaluator would overestimate the change in hand- events, suggesting that it is counter-productive to shorten the washing behavior in response to an intervention. If there is observation periods in an attempt to reduce data collection no increase in handwashing behavior following exposure to costs.24 Such extended durations of observation come at a 19 Cousens, et al. 1996. 20 It is certainly possible that the reactivity was not solely as a result of the structured observation but, rather as a result of the combination of introducing both the soap with motion sensor and the structured observation into the household. That is, when the structured observation began, respondents may have remembered that the soap with motion sensor had been given to them several days prior. Since the observer was connected with the study that introduced the soap with motion sensor into the home, the respondents may have utilized the soap more while the observer was present than before the observer had been present. 21 Ram, Halder, et al. 2008. 22 Ibid. We detected this reactivity based on increases in the number of soap use events detected during the structured observation, using soap with a motion sensor, compared to the same time period on days preceding the structured observation. 23 Bentley, et al. 1994. 24 We compared observations of 5 hours duration to those of 90 minutes duration in order to examine data loss resulting from reducing the observation duration. Essentially, we sought to understand whether similar information could be gained by reducing the observation duration, which might have allowed for multiple households to be observed by an observer in a single day. Compared to those in 5-hour observation households, observers in 90-minute observation households were significantly less likely to observe one or more defecation-related events. The ability to witness at least one observation of other types of events, such as feeding a child, was not significantly different between the 90-minute and 5-hour observation households. Perhaps, it took several hours for the observed individual to become comfortable enough with the presence of the observer that she engaged in her usual toileting behavior. To our knowledge, this question has only been tested in Bangladesh and our findings may not be representative of other cultural settings. Still, our results do inform about the greater depth of information provided by prolonged structured observations. 6 Global Scaling Up Handwashing Practical Guidance for Measuring Handwashing Behavior Methods of Measuring Handwashing Behavior cost since each observer can only conduct one observation dimensions. Based on the movement patterns of the soap, per household per day. From a practical standpoint, it would the number of times soap is used in a given time period can be difficult to carry out two 5-hour (or even 4-hour) observa- be counted. The soap can be left in a household for several tions in a single day without risking substantial interviewer days, allowing for observation of soap use behavior over a fatigue and compromise in the quality of observed data. much longer period of time than would be feasible by struc- Moreover, handwashing, bathing, and toileting behavior may tured observation. The SmartSoap's ability to detect consis- differ according to the time of day, possibly rendering morn- tency in soap use behavior has been demonstrated in ing observations incomparable to afternoon observations. Bangladesh, where the number of times soap was used in a Thus, depending on the required sample size to demonstrate household was remarkably consistent across each of eight project outcomes of interest, structured observations would days. It is possible that, overall, households increase the require substantial numbers of trained individuals, or a pro- number of times they use soap when SmartSoap is in the longed data collection period, either of which might be ex- home, compared to when SmartSoap is not in the home. pensive for the program being monitored. There was no evidence that the households used SmartSoap more frequently on the first day of having the SmartSoap in Despite these caveats, several groups have successfully com- the home and then reduced use as they became accustomed pleted structured observations on the scale of hundreds of to its presence. A study is underway now in Bangladesh to households without substantial difficulty. They provide a understand whether households reduce overall soap use wealth of detail regarding handwashing behavior at critical once the novelty of SmartSoap wanes, after it has been in times of interest, including defecation, feeding, eating, and the home for several weeks. cooking (Box 3). It is likely best to replace bar soap existing in the household Direct Measures: Soap with Motion Sensors with SmartSoap, rather than introducing bar soap in the A technology-based method of objective recording of hand- form of SmartSoap to a household that is not in the practice washing behavior is the SmartSoap, a Unilever-developed of using bar soap daily. Also, it is likely best to suggest that all technology. The SmartSoap consists of a motion sensor household members that use the existing bar soap use the embedded in ordinary-appearing Lifebuoy® soap. The SmartSoap provided in its place. This would minimize bias motion sensor tracks movement of the soap in three that might result from asking only the mother of young chil- dren, or other such targeted respondents, to use the Smart- BOX 3: INDICATORS THAT COULD BE TRACKED Soap. However, such a strategy allows only for soap use USING STRUCTURED OBSERVATIONS measurement at the household level and not at the level of a · All household members, or specifically single respondent of interest, such as the mother of the primary caregivers of young children youngest child. Using the household size, the analyst may 0 Any use of soap for handwashing convert the number of soap use events into per capita soap 0 Any use of another cleansing agent, such use events but this would remain an estimate of individual as ash or mud, for handwashing soap use behavior. Additionally, since the soap use is detected 0 Proportion of all critical times observed by the motion sensor, it would not be possible in many cir- during which one and/or both hands are cumstances to detect the circumstances in which hands are washed with soap and water being washed with soap. For example, it would not be pos- 0 Proportion of specific critical times ob- sible to detect whether a mother is washing hands before served during which hands are washed feeding the child. Unilever has devised one solution to this with soap and water, e.g., proportion of by attaching motion sensors to items that are closely associ- defecation-related events after which ated with defecation. In South Asia, it is common to reserve hands are observed to be washed with one water vessel for washing oneself after defecating. Unile- soap and water ver has worked in India and Bangladesh to attach motion sensors to these water vessels, thereby allowing for detection www.wsp.org 7 Practical Guidance for Measuring Handwashing Behavior Methods of Measuring Handwashing Behavior potentially problematic, since the bar soap in which the BOX 4: INDICATORS THAT COULD BE TRACKED motion sensor has usually been embedded has been a USING SMARTSOAP beauty-soap (Lifebuoy®). To date, there is no publicly avail- · Total number of soap use events in a speci- able or published information on whether data from fied time period, e.g., 24 hours or during the motion sensors embedded in multipurpose bar soaps is duration of the structured observation complicated by the fact that such multipurpose bar soaps · Number of per capita soap use events in a are used for a variety of purposes, and not handwashing or specified time period bathing alone. The movement signatures for handwashing · Proportion of defecation-related events that may be difficult to distinguish from those of washing are followed by soap use, within a desig- clothes, washing dishes, or playing with the soap. In many nated period of time countries, powder detergent or liquid soap are the most 0 This is applicable if a motion sensor can commonly used forms of soap. Replacing powder or liquid be attached to an item closely associated with a bar may allow for substantial bias, in that the ob- with defecation, e.g., water vessel used served household is given a novel and `special' way of wash- to cleanse oneself after defecation or a ing hands that it has not previously used. roll of toilet paper Further experience with SmartSoaps in the hands of various researchers, and publication in peer-reviewed journals on positive and negative experiences, are needed to understand of defecation events (Granger, Unilever R&D, personal fully the utility of this novel technique. Currently, the larg- communication).25 Pairing that information with data from est drawback to the use of the SmartSoap for monitoring the SmartSoap allows for detection of soap use during the handwashing promotion programs are capacity limitations several minutes following defecation, which would be robust of staff associated with Unilever to deploy, extract, and ana- information indeed (Box 4). If the water vessel is used for lyze data from the soaps with motion sensors. Transfer of purposes other than cleansing oneself after defecation, it technical capacity is imperative if SmartSoaps are to be- would be difficult to pair water vessel use information with come more widely adopted for monitoring handwashing soap use information. Ultimately, if a program proposes to behavior. However, such transfer clearly comes at the risk of increase soap use in general, the SmartSoap would serve as a lack of assurance of the various steps contributing to data useful tool to detect that outcome of interest. quality. Additionally, the cost of individual motion sensors is projected at approximately US$120, which may be pro- There are a few caveats to the use of SmartSoap at present. hibitive if a program actually had to purchase large num- The experience reported above is based on one study of bers of the sensors. SmartSoap conducted in Bangladesh, although additional studies are underway. The consistency of soap use from day Use of Composite Measures to day in households is a comforting marker of accuracy, Several studies have attempted to combine information but such accuracy should be confirmed in multiple studies from different methods of measuring handwashing behav- in a spectrum of cultures and handwashing practices. ior.26 These composite measures have included information SmartSoap provides household-level soap-use data and, obtained purely from demonstration of handwashing, as thus, if individual soap use behavior is of interest, Smart- well as information from self-report. In our review of the Soap would not be the measurement technique of choice. literature, it appears that the same composite measure has In many countries, households often use multipurpose bar not been repeatedly tested in different geographic areas, soaps for washing hands. Replacing multipurpose bar soaps, which might provide information on its utility across cul- which may also be used for washing laundry or dishes, is tural settings. There has been no comparison of these 25 Biran et al. 2009. 26 Hoque, et al. 1995; Yalcin, et al. 2004; Sandora, et al. 2005. 8 Global Scaling Up Handwashing Practical Guidance for Measuring Handwashing Behavior Methods of Measuring Handwashing Behavior composite indicators to other objective measures, such as observation data, from those households shown not to be structured observation or soaps with motion sensors. reactive to the presence of the observer, can then be useful for examining context-specific handwashing practices, such Based on our review of existing literature and our own ex- as washing hands after defecation or before feeding a child. perience, combining information from structured observa- There is no guarantee that the observed individual (from a tion with data from the soap with motion sensor may household that is shown not to be reactive) would behave in provide the strongest information on a household's soap use the same way during the structured observation as she behavior. The soap with motion sensor can provide infor- would during her usual practice. But, the removal of struc- mation on household-level soap use on days preceding tured observation data from obviously reactive households structured observation and can facilitate the identification would certainly enhance the validity of data obtained via of households that are reactive to the presence of the structured observation. observer during the structured observation. Structured www.wsp.org 9 III. Conclusion This document attempts to give a balanced view of each of Demographic and Health Survey (DHS) and every multi- the methods of measuring handwashing behavior, includ- indicator cluster survey (MICS) undertaken in resource- ing both routinely used and novel methods. Positive and poor countries. While these self-reported measures likely negative attributes of each method have been described misrepresent true practices and health conditions, they do (Table 1). The attempt to be balanced may lead to skepti- provide insights into trends in these behaviors over time cism about the utility of measuring handwashing behavior and important predictors of child morbidity and mortality. at all. However, there are few perfect measurements avail- Given these caveats, described below are potential ap- able for outcomes of human behavior or health, apart from proaches to measuring handwashing behavior for a variety cadaver autopsy for some health conditions. Thus, health of program types and settings. Reference to health out- researchers and public health practitioners must frequently comes as proxy measures of handwashing behavior has been accept the limitations of the measures that are available to intentionally minimized, since, almost universally, mea- them but not be paralyzed by those limitations. Examples surement of health outcomes such as diarrhea incidence or of imperfect measures that still provide useful and necessary prevalence is very costly. As indicated below, additional data information are self-reported breastfeeding or self-reported on the relationship between other measures of handwash- use of oral rehydration therapy for diarrhea treatment, as ing behavior and health outcomes is clearly needed and well as caregiver reported symptoms of childhood diarrhea should be sought where possible. and other illnesses. These measures are undertaken in every 10 Global Scaling Up Handwashing TABLE 1: SUMMARY OF STRATEGIES TO MEASURE HANDWASHING BEHAVIOR Measurement strategy Requirements for use Advantages Disadvantages Appropriate setting for use www.wsp.org Self-report: using closed · Questionnaire · Is efficient · Shown to overesti- · Self-report is not recommended or open-ended questions · Training · Can be incorporated mate handwashing for measurement of handwashing to ask respondents about into multipurpose behavior behavior their own handwashing surveys · Questionnaires may be useful behavior for tracking knowledge, monitor- ing the reach of a handwashing promotion program, or identifying barriers to handwashing Microbiological hand · Supplies for hand rinse / · Is objective · Is not reliable · Well-funded handwashing pro- contamination: hand imprint collection · Reflects individual · Is expensive motion program evaluations or Practical Guidance for Measuring Handwashing Behavior (aproxymeasure) · Microbiology laboratory hand contamination · Serves as a proxy research studies may be able to or equipment for field- measure of hand- improve the utility of microbio- level quantitation of hand washing behavior logical hand contamination as a Conclusion contamination · Requires some struc- marker of handwashing behavior · Training and quality control tured observation for · Not recommended for small or · At least one staff mem- measurement at criti- minimally funded handwashing ber with microbiology cal times (e.g., after promotion programs experience defecation) · Not recommended for nation- ally or regionally representative multipurpose surveys (e.g., DHS, MICS) Rapid observations: · Checklist · Is efficient · Serves as a proxy · Well-funded handwashing recording presence of soap · Training · Is objective measure promotion program evaluations and water at a designated · Can be incorporated · Cannot confirm or research studies handwashing station into multipurpose the frequency or · Evaluations of small or minimally (aproxymeasure) surveys consistency of funded handwashing promotion · Shown to have internal handwashing programs validity with structured · Does not reflect · Nationally or regionally represen- observation individual-level tative multipurpose surveys (e.g., behavior DHS, MICS) (Continued) 11 12 TABLE 1: SUMMARY OF STRATEGIES TO MEASURE HANDWASHING BEHAVIOR (Continued) Measurement strategy Requirements for use Advantages Disadvantages Appropriate setting for use Structured observations: · Structured format to · Is objective · Shown to result in · Well-funded handwashing pro- recording observed fre- capture details regarding · Reflects individual reactivity, because of motion program evaluations or quency of handwashing critical times of interest and behavior the presence of the research studies with soap at critical times, handwashing behaviors · Captures rich detail on human observer · Sample size calculations, and e.g., after defecation · Several hours of observa- handwashing behavior · Is costly in terms of consultation with persons with tion, preferably at the same person-time epidemiological or statistical ex- time of day in all house- · Requires highly pertise, may indicate feasibility holds, and at times that trained staff of this approach for evaluation of capture critical times of even small handwashing promo- interest and that are locally tion programs Practical Guidance for Measuring Handwashing Behavior acceptable · Training and quality control · Preferable to have staff Conclusion with experience in behav- ioral observation Soap with motion sensors: · Equipment, including · Is objective · Does not reflect · Well-funded handwashing pro- recording the number of soaps with motion sensors · Does not require the individual-level motion program evaluations or times soap is used during and microwave presence of a human behavior research studies an observation period · Training in the preparation, observer · Cannot provide infor- · Sample size calculations, and initialization, deployment, mation on handwash- consultation with persons with data download, and data ing at critical times epidemiological and statistical analysis · Requires highly expertise, may indicate feasibility trained staff of this approach for evaluation of · Is costly in terms of even small handwashing promo- equipment needs tion programs · May not be useful in settings in which bar soap is used for multiple purposes, or in which liquid or powder soap is com- monly used Global Scaling Up Handwashing IV. Recommendations for Various Scenarios The recommendations below are made on the basis of ease For measurement of changes in knowledge or attitudes, or of data collection and potential cost to the program or exposure to handwashing promotion programs or specific study. The focus here is on the measurement of handwash- messages, questionnaires may prove useful. As noted above, ing behavior, although I comment on the utility of collect- the use of questionnaires for measurement of handwashing ing data regarding knowledge, attitudes, and health behavior is not recommended, since self-reported hand- outcomes in some of these contexts. washing behavior overestimates observed behavior. Well-Funded Handwashing Promotion At present, random or critical-time measurement of hand Programs or Research Studies contamination is also not recommended as a measure of Ideally, such studies should strive to use the most rigorous handwashing behavior, given the substantial variability de- methods to measure handwashing behavior. Specifically, tected in several studies described above. But, as detailed the use of soaps with motion sensors and structured obser- below, well-funded programs or research studies may serve vations is recommended. Soaps with motion sensors can be as opportunities for improving upon this measure. Future used to track overall soap use in target households. The soap studies should address the utility of indicator organisms use measurement is not context specific, not tied to specific other than E. coli, whether variability in hand contamina- critical times. However, overall soap use may be expected to tion is evident in other laboratories, and the relationship increase if there is an increase in handwashing with soap in between hand contamination and health outcomes. response to handwashing promotion. This recommenda- tion is made for those sites in which bar soap is the pre- Several questions of import may be answered in the context dominant form of soap utilized. of well-funded public health program evaluations and re- search studies. The soaps with motion sensors are also useful for detecting · There is a fundamental gap in the literature on the a household's reactivity to structured observation. This is relationship between the various measures of hand- critical in order to elucidate which data is useful (i.e., from washing behavior and health outcomes. At pres- households that are not reactive to the presence of the ob- ent, it is not clear whether changes in handwashing server) and which data may be compromised as a result of behavior, as measured by the techniques described substantial reactivity to the presence of the observer. above, are correlated with changes in risk of health outcomes of interest, particularly diarrhea and respi- Structured observation data can elucidate handwashing be- ratory infections. It is strongly recommended that havior for specific household members, e.g., primary care- well-funded research studies and programs include givers of young children, and/or during particular critical measurement of both behavioral outcomes and times, for example, after defecation. The use of soap with health outcomes in the same study populations, pref- motion sensors may elucidate the degree to which a particu- erably in a longitudinal fashion, in order to examine lar household is reactive. The program will need to decide these relationships in detail. upon acceptable levels of reactivity and the extent to which · At present, there is still a paucity of published obtaining context-specific information is important enough effectiveness data regarding the impact of public to warrant collection of structured observation data. health programs on behavioral and health outcomes. Rapid observations, which are proxies, should also be in- It is strongly recommended that effectiveness data cluded among measures of handwashing behavior used in (positive, negative, and neutral) be published in well-funded programs. These observations provide useful peer-reviewed literature in order to inform the pub- information on the facilitating environment found in the lic health community, policy makers, and funding home for good handwashing behavior. agencies. www.wsp.org 13 Practical Guidance for Measuring Handwashing Behavior Recommendations for Various Scenarios · Well-funded programs and research studies may be designs, and program goals and targets. However, because of opportunities to improve upon the utility of micro- the "longitudinal" nature of data collected from each of biological testing of hand contamination, as well these data sources, required sample sizes to measure impact as other innovative and/or composite measures. It of the program may indeed be smaller than one might would be useful to examine, for example, whether expect. rates and levels of contamination several hours fol- lowing a supervised, thorough handwashing, change Rapid observations, which are proxies, are certainly recom- over the course of a handwashing promotion pro- mended as efficient measures of handwashing behavior in gram. Also, it would be useful to examine whether not-so-well-funded public health programs. mean levels of hand contamination, as obtained Questionnaires may be useful for measurement of knowl- from serial measurements from the same individual, edge, attitudes, and program exposure. Again, questionnaires are associated with other objective measures of hand- are not recommended for measurement of self-reported washing behavior, such as soaps with motion sen- handwashing behavior. sors, or with health outcome data. · The utility of principal components analysis, or other such means of assigning participating house- Nationally Representative Surveys holds or participants to categories of handwashing The Demographic and Health Surveys are described as behavior should be further examined, particularly in "nationally-representative household surveys that provide relationship to health outcomes, and as a means of data for a wide range of monitoring and impact evaluation identifying important explanatory factors associated indicators in the areas of population, health, and nutri- with strata of handwashing behavior. tion". The multi-indicator cluster surveys (MICS) are con- ducted by UNICEF and may be described similarly. DHS Handwashing Promotion Programs with and MICS surveys are conducted every 3-5 years in most Minimal Funding low- and middle-income countries. Handwashing is only Ideally, these programs, as better-funded programs, would one of a myriad number of topics covered in these surveys obtain objective measurement of handwashing behavior and, thus, measurement of handwashing behavior is with soaps with motion sensors and structured observations. necessarily restricted to the most efficiently administered Cost is the primary limiting factor. Program evaluation staff questions. Therefore, the use of rapid observations, which are strongly encouraged to consult statistical and/or epide- are proxies, for measuring handwashing behavior is miologic expertise in order to determine required sample recommended. sizes for measurement of handwashing behavior with soaps with motion sensors and structured observations. Universal It is not feasible to do more intensive measurements, such recommendations regarding sample sizes cannot be made as structured observations or soaps with motion sensors, in here, given the diversity in program types, evaluation the context of these large nationally representative surveys. 14 Global Scaling Up Handwashing Practical Guidance for Measuring Handwashing Behavior References References Bentley, M., Boot, M., Gittelsohn, J., Stallings R. 1994. The Use Last, J.M., ed. 2001. A Dictionary of Epidemiology. Fourth ed. of Structured Observations in the Study of Health Behaviour. New York: Oxford University Press. IRC International Water and Sanitation Centre. The Hague, Luby, S.P., Halder, A.K. 2008. Associations among handwash- The Netherlands. ing indicators, wealth, and symptoms of childhood respi- Biran, A., Rabie, T., Schmidt, W., Juvekar, S., Hirve, S., ratory illness in urban Bangladesh. Trop Med Int Health Curtis, V. 2008. Comparing the performance of indicators of 3(6):835­44. hand-washing practices in rural Indian households. Trop Med Luby, S.P., Agboatwalla, M., Billhimer, W., Hoekstra, R.M. Int Health 13(2): 278­85. 2007. Field trial of a low cost method to evaluate hand Biran, A., Schmidt, W-P, Wright, R., Jones, T., Seshadri, M., cleanliness. Trop Med Int Health 12(6):765­71. Isaac, P., Nathan, NA, Hall, P., McKenna, J., Granger, S., Luby, S.P., Agboatwalla, M., Raza, A., Sobe, I. J., Mint, E.D., Bidinger, P., and Curtis, V. The effect of a soap promotion Baier, K., et al. 2001. Microbiologic effectiveness of hand and hygiene education campaign on handwashing behavior washing with soap in an urban squatter settlement, Karachi, in rural India: a cluster randomized trial. Trop Med Int Health Pakistan. Epidemiol Infect 127(2):237­44. 14 (10): 1303-1314. Manun'Ebo, M., Cousens, S., Haggerty, P., Kalengaie, Cousens, S., Kanki, B., Toure, S., Diallo, I., Curtis, V. M., Ashworth, A., Kirkwood, B. 1997. Measuring 1996. Reactivity and repeatability of hygiene behaviour: hygiene practices: a comparison of questionnaires with structured observations from Burkina Faso. Soc Sci Med direct observations in rural Zaire. Trop Med Int Health 43(9):1299­308. 2(11):1015­21. Curtis, V. 2003. Talking dirty: how to save a million lives. Int J Pinfold, J.V. 1990. Faecal contamination of water and fingertip- Environ Health Res 13 Suppl 1:S73-9. rinses as a method for evaluating the effect of low-cost water Curtis, V., Cairncross, S. 2003. Effect of washing hands with supply and sanitation activities on faeco-oral disease trans- soap on diarrhea risk in the community: a systematic review. mission. A case study in rural north-east Thailand. Epidemiol Lancet Infectious Disease 3:275­281. Infect 105(2):363­75. Curtis, V., Kanki, B., Cousens, S., Diallo, I., Kpozehouen, A., Rabie, T., Curtis, V. 2006. Handwashing and risk of respiratory Sangare, M., et al. 2001. Evidence of behavior change following infections: a quantitative systematic review. Trop Med Int a hygiene promotion programme in Burkina Faso. Bull WHO Health 11(3):258­67. 79(6):518­527. Ram, P., Halder, A., Granger, S., Hall, P., Jones, T., Halder, A.K, Luby, S.P., Bhuiya, A., Tronchet, C., Akhter, S., Hitchcock, D., Nygren, B., Islam, M., Molyneaux, J., Johnston, R. 2008. Handwashing behavior in rural Bangla- Luby, S., 2008. Use of a novel method to detect reactivity desh. Health and Science Bulletin 6(3). to structured observation for measurement of handwashing Hoque, B.A., Mahalanabis, D., Alam, M.J., Islam, M.S. 1995. behavior. American Society of Tropical Medicine and Post-defecation handwashing in Bangladesh: practice and Hygiene. New Orleans, LA. efficiency perspectives. Public Health 109(1):15­24. Ram, P., Jahid, I., Halder, A., Nygren, B., Islam, M., Granger, S., Larson, E., Aiello, A., Lee. L.V., Della-Latta, P., Gomez- et al. 2008. Variability in measurement of hand contamina- Duarte, C., Lin, S. 2003. Short- and long-term effects of tion: implications for evaluating hand hygiene promotion handwashing with antimicrobial or plain soap in the programs. American Society of Tropical Medicine and Hygiene. community. J Community Health 28(2):139­50. New Orleans, LA. www.wsp.org 15 Practical Guidance for Measuring Handwashing Behavior References Sandora, T.J., Taveras, E.M., Shih, M.C., Resnick, E.A., Lee, questionnaires, and direct observations of sanitary prac- G.M., Ross-Degnan, D., et al. 2005. A randomized, con- tices: a comparative study. Bull World Health Organ trolled trial of a multifaceted intervention including alcohol- 65(2):217­22. based hand sanitizer and hand-hygiene education to reduce Yalcin, S.S., Yalcin, S., Altin, S. 2004. Hand washing and illness transmission in the home. Pediatrics 116(3):587­94. adolescents. A study from seven schools in Konya, Turkey. Stanton, B.F., Clemens, J.D., Aziz, K.M., Rahman, M. 1987. Int J Adolesc Med Health 16(4):371­6. Twenty-four-hour recall, knowledge-attitude-practice 16 Global Scaling Up Handwashing