Water System Publications
interviewing enhances the adoption of water disinfection practices in Zambia
Thevos A, Quick R
and Yanduli V
studies represent the first adaptation of the Motivational Interviewing (MI) behavior
change approach in the developing world, using health workers directly from the
community. The objective was to compare the effectiveness of the standard practice
of health education (comparison group) to MI (experimental group) in initiating
and sustaining safe water treatment and storage behavior. Methods: focus groups
and community surveys were conducted prior to health worker training. The main
outcome variables were detectable disinfectant levels in stored
water (for Field Trial 1) and disinfectant sales (for Field Trial 2). Results:
in Field Trial 1 (n = 185 households), a very high adherence rate was achieved
(range 71.1-94.7%), with no statistical differences between the groups. Field
Trial 2 (n = 427 households) incorporated lessons learned from the previous trial
and resulted in much higher purchase rates of the disinfectant in the MI group,
1(7) = 10.69, p < 0.001, eta2 = 0.94. Conclusion: MI intervention
appears promising for public health initiatives in the developing world. Further
work in this area is indicated.
In the developing
world, diarrheal disease is a leading cause of morbidity and mortality among children
less than 5 years old. Water is an, important vehicle for the transmission of
enteric pathogens; it has been estimated that over one billion people lack access
to safe water (Centers for Disease Control, 1990; Bern et al., 1992). In
response to this
urgent problem, the Centers for Disease Control and Prevention (CDC) and the Pan
American Health Organization developed a simple, inexpensive, easily disseminated,
and effective intervention for safe water treatment and storage at the household
level (Mintz and Tauxe, 1995).
intervention involves household disinfection with a locally produced sodium hypochlorite
solution, and education
on proper storage of treated water and disinfectant solution, the causes of diarrhea,
and the importance of diarrhea prevention. A series of studies in Bolivia demonstrated
that households using this water intervention have significantly better water
quality and a lower incidence of diarrhea than control households (Quick et
al., 1996, 1999). Given the success of the water intervention in Latin America,
two field evaluations extending to the African continent were conducted in two
peri-urban communities of Kitwe, Zambia between March 1998 and January 1999. For
these trials, the education element was augmented by a novel behavior change component,
Motivational Interviewing, to investigate its effectiveness in improving adherence
rates in the use of this safe water treatment and storage system.
in the behavioral sciences has produced efficacious, brief intervention techniques
to produce rapid, internally motivated behavior change. One such approach, developed
by Miller and Rollnick, is termed Motivational Interviewing (Miller and Rollnick,
1991). Motivational Interviewing (MI) is grounded in decision-making theory and
motivational psychology. It incorporates the trans-theoretical model of the stages
of change, which conceptualizes change as a process. People move through stages
with corresponding different levels of readiness (Prochaska and DiClemente, 1984,
1992; Prochaska and Velicer, 1997). The first stage, 'precontemplation', applies
when the person is not ready to consider change or is unaware of any need to change.
'Contemplation', the next stage, implies ambivalence and relates to the person
who both considers change and rejects it. The 'preparation' stage refers to people
open to changing and those who may be preparing to make a change. An individual
who is actually engaging in actions with the intention of bringing about change
is said to be in the 'action' stage.
of stage-based interventions have shown that employing particular strategies for
different stages can be quite effective in the successful changing of health behaviors
(DiClemente et al., 1991; Perz et al., 1996; Prochaska
and Velicer, 1997). Conversely, employing strategies, e.g. advice giving or suggestions
for taking action with a person who is not ready for change can be premature,
inappropriate and ineffective. A more person-centered, stage-based approach, e.g.
MI, with the goal of resolving ambivalence and eliciting a person's own arguments
for change has been shown to be more effective, particularly among those less
ready to change (Heather et al., 1996). The MI practitioner employs motivational
strategies to mobilize a person to decide to choose a change in behavior, and
to use his or her own resources to do so. Its defining feature is a style of communicating
whereby the practitioner takes a partnership role with the client (Rollnick and
Interviewing has been effective in helping clients change behaviors in several
health-related areas, including alcohol and other drug abuse (Bien et al.,
1993; Miller et al., 1993; Saunders
et al., 1995; Heather et al., 1996; Project MATCH Research Group,
1997; Project MATCH Research Group, 1998; Daley et al., 1998), cardiovascular
health (Woollard et a/., 1995), and diabetes (Smith et al., 1997; Trigwell
et ai, 1997). Because MI is both client-centered and effective in producing healthy
changes in a wide range of behaviors, it was a promising approach for use in the
two field trials reported here represent the first application of MI principles
in the developing world using health workers directly from the community. The
objective was to compare the effectiveness of the standard practice of health
education (Ed Only) to Motivational Interviewing (MI) in initiating and sustaining
safe water treatment behavior.
of Ipusukilo and Luangwa, both low socioeconomic status (SES), peri-urban communities
of Kitwe, Zambia were selected as the study sites for this project because they
lacked a water system and, in a survey conducted by BASICS (a United States Agency
for International Development-funded project), had identified drinking water safety
as a major concern. The communities are served by volunteers in their respective
Neighborhood Health Committees (NHC). NHC members (NHCs) provide health information
to their neighbors and have regular meetings at the local Health Clinics.
the behavior change intervention trials were implemented, Zambian nurses from
the Health Clinics in the two communities conducted a baseline survey of all residents
in the sample, asking about demographic characteristics, and water handling and
sanitary habits. Separate focus groups and individual interviews with community
residents were also conducted by the study team to generate information regarding
the health beliefs of the study community. Stored water from all study households
and source water from a random sample of 25% of households were tested for free
and total chlorine residuals, using the N,N-diethyl-phenylenediamine. (DPD)
method, and for contamination with Escherichia coli, using membrane filtration
(Mates and Shaffer, 1989).
consent in accordance with the requirements of the CDC Institutional Review Board
and the Ethics Committee of the Tropical Diseases Research Centre of Zambia (TDRC)
was obtained from all study participants.
Trial 1: Ipusukilo community
Trial 1 took place from March to June 1998. A microenterprise for the production,
bottling, sale and distribution of the sodium hypochlorite (chlorine) disinfectant
was established at the Ipusukilo Health Clinic using a SANILEC on-site hypochlorite
generator (Exceltech International, Sugarland, TX, USA). Three NHCs were trained
in the operation and maintenance of the SANILEC.
(n = 188) were randomly selected from two zones in Ipusukilo. In the 2
weeks between the baseline survey and the beginning of the Ipusukilo intervention,
three families moved out of the community, leaving 185 households at the start
of the intervention. The behavior change intervention study was designed with
two experimental groups: an intervention group (MI, n = 92) and a comparison
group (Ed Only, n = 93). The nurses from the Ipusukilo Health Clinic selected
the 10 most active NHCs and assigned them to two groups of five to deliver the
intervention and comparison conditions. Households were assigned to NHCs based
on proximity to each NHC's residence. This assignment strategy resulted in households
within the same neighborhood being assigned to different experimental conditions.
In compensation for their participation in the study, each household received
disinfectant free for the duration of Field Trial 1.
the beginning of the study, all 10 of the NHCs
participated in a workshop about the causes of diarrhea, the importance of diarrhea
prevention, and the proper use of chlorine disinfectant and
storage of treated water. This training was provided to the NHCs by the study
team so that the NHCs could, in turn, educate the members of the study households
and reinforce this information as needed at subsequent household visits. NHCs
were given brochures and educational flip charts which presented this information
in simple illustrations and text in English (the official national language) and
Bemba (the local language).
In addition, all of the NHCs received instruction in basic interviewing skills,
the design and rationale of the study, and a review of the general procedures
in the completion and submission of study rating forms which were required to
document household visits. The workshop and informational materials were the only
tools that the five NHCs in the Education Only comparison group received.
The MI intervention group consisted of three local nurses and five NHCs who were
trained in the theoretical model of the stages of change, the basic principles
employed in brief Motivational Interviewing to effect behavior change, and the
task components relevant to these. The principles of MI involve expressing empathy,
developing discrepancy, avoiding argumentation, rolling with resistance, and supporting
self-efficacy. The MI tools are open-ended questions, affirmations, reflective
listening and summarizing (Miller and Rollnick, 1991; Miller et a/., 1992). The
nurses, all of whom were English speaking, were trained first and separately from
the NHCs by the behavioral scientist on the study team. The nurses then developed
and delivered the MI training to the NHCs. A total of 10 h over 4 days were devoted
to the nurses' training.
MI content was adapted to be relevant to water and sanitation issues as well as
cultural considerations based on information gained from focus groups, individual
interviews and guidance from the local nurses. The training utilized a participatory
workshop format with didactic instruction and exercises employing role-plays and
material relevant to the MI training was translated into Bemba for distribution
to the NHCs. The series of workshops was designed to be temporally close enough
to allow for skill and information retention. Although 10-12 h of NHC training
were planned, only 5 h were actually accomplished in three separate sessions over
a 2-week period due to illness and competing commitments. The first two trainings
(each was 2 h long) occurred 2 days apart; the third took place 2 weeks after
main outcome variable was whether a household had a detectable level of total
chlorine in its stored water. Zambian health workers from TDRC tested stored water
in all study households for free and total chlorine residuals. Water testing
occurred on four occasions, ~2 weeks apart. Because free chlorine levels decline
over time in stored water, and we could not guarantee that our visits would occur
at a defined period after the household filled their water vessel, we decided
that detectable total chlorine residual was the best indicator of adherence, i.e.
evidence that the household was dosing their water with the chlorine disinfectant.
At week 9, a brief
follow-up interview about water-handling practices, and knowledge of causes and
prevention of diarrhea was conducted with study participants.
Trial 2: Luangwa community
second study was implemented in the community of Luangwa in June 1998 using lessons
learned from the previous trial in Ipusukilo. Two clearly defined zones were selected
by community nurses as treatment zones. Households in one zone received only health
education (Ed Only, n = 200) and households in the other zone received
MI (n = 132). With this assignment procedure, neighbors were not assigned
to two different study conditions and the risk of information sharing was consequently
reduced. In this field trial, the MI training was delivered for 4 h each on two
consecutive days to five NHC members and to a local nurse. The behavioral scientist
had no contact with the comparison group of five NHCs who delivered the Ed Only
treatment. This was done in order to avoid inadvertent teaching of interviewing
skills to this group, thereby diluting the difference between the treatment conditions.
for the sale of sodium hypochlorite solution was established in the Luangwa Health
Clinic. Sodium hypochlorite solution sales to the Ed Only and MI zones were recorded
by the nurses at the Luangwa Health Center. The number of bottles sold divided
by the number of possible buyers (i.e. households) in the community was calculated
for each month (8 months total). A paired samples t-test was conducted
to compare sales ratios for the two treatment groups. Because of funding limitations,
household level testing of free and total chlorine residuals could not be performed.
Some of the households
in the MI zone (n = 95, 72%) participated in weekly diarrheal surveillance
during Field Trial 1 (Group A). Of concern was
whether these households were sensitized to buy more chlorine because of their
inclusion in the diarrheal surveillance sample, and thus inflated the MI group
effect. This concern led us to incorporate a group that did not participate in
the diarrhea surveillance. A paired samples t-test was therefore conducted
to compare sales ratios between the surveillance group (Group A) and the remaining
MI households (Group B) where , monthly sales data were available for each (6
Field Trial 1: Ipusukilo community
Trial 1, sodium hypochlorite utilization rates and water-handling practices were
compared between the MI and Ed Only groups using the t-est. Data from 166
households were available at the end of this 8-week study. During the course of
the study 19 households were excluded: 18
moved outside the study area and one refused to participate further. Results of
the baseline survey indicated that the Ed Only and MI groups were not significantly
different (a = 0.05) on demographic variables and baseline safe water practices.
The mean number of persons/household was 5.8 (SD = 2.7), the median age of respondents
was 33 (range 14-89) and the median annual per capita income was $608 (range $49-3650).
Seventy-seven percent of the respondents were female. All respondents reported
using a shallow well as the source for drinking water. Forty percent of the entire
sample at baseline reported that they had previously treated their household water
to make it safe. Only 9.8% of households had treated their water in the past day
and 7% within the past week. All households used a latrine. Fewer than 4% of households
reported having electricity.
testing for detectable total chlorine residuals revealed a high rate of overall
adherence, ranging from 71.1 to 94.7% of households having positive chlorine readings
over the 8 weeks of sampling (Figure 1). Adherence rates for the MI and Ed Only
groups were not significantly different at any time point.
the follow-up survey, 96.9% of respondents reported that they used chlorine disinfectant
to treat their water. Similarly, direct observation indicated excellent adherence
with instructions for handling the disinfectant, as 99% kept chlorine out of the
sun, 96% kept chlorine out
of the reach of children, 98% knew the correct dose of chlorine for their container
size, and 90% stated an intention to continue to use chlorine. Those who did not
intend to purchase chlorine cited lack of money as the reason. None of these safe
water-handling indicators revealed statistically significant differences between
the MI and Ed Only groups.
1: Percent of households
with detectable total chlorine residuals in stored water, by intervention (Motivational
Interviewing) and comparison (Education Only) groups, Field Trial 1, Ipusukilo
community, Kitwe, Zambia, April-May 1998.
Field Trial 2: Luangwa community
of the Luangwa community closely parallel those of Ipusukilo, and there were no
differences between the two treatment groups. The mean number of persons/household
was 6.2 (SD = 3.0),
the median age of respondents was 28 (range 10-79) and the median annual per capita
income was $406 (range $94-3650).
Seventy-two percent of the respondents were female. Fewer than 5% of households
reported having electricity.
the MI zone had significantly higher sales ratios than the Ed Only zone, t(7)
= 10.69, p < 0.001 (Figure 2). The effect size is large, eta2
= 0.94, indicating that 94% of the variance in sales ratios is attributable to
treatment group assignment. Within
the MI zone, Group A (n = 95 households) and Group B (n = 37 households),
did differ significantly, t(5) = 6.49, p = 0.001, although not in the direction
expected. That is, the households who participated in the diarrheal surveillance
had lower sales ratios than the comparison group. While this finding was unexpected,
it provides an even stronger assurance that the mixing of Group A and Group B
households in the MI zone did not confound the data demonstrating a strong association
between use of MI and disinfectant sales.
These field trials
were the first to employ the novel behavior change intervention technique of Motivational
Interviewing to prevent diarrheal diseases in a developing country. In Field Trial
1, despite methodological problems, there was a very high rate of adherence and
community acceptance of the water intervention as evidenced by the high percentage
of water samples with detectable chlorine. In previous Bolivian field trials of
this water quality intervention in which only an education component was used,
were significantly lower (Quick et al., 1999).
2: Magnitude of sales of bottles of sodium hypochlorite solution (no. of bottles
sold/no. of households), by intervention (Motivational Interviewing) and comparison
(Education Only) zones by month. Field Trial 2, Luangwa community, Kitwe, Zambia,
June 1998-January 1999.
The rates of adherence
in Field Trial 1 were equally high in both the MI intervention and the Ed Only
comparison groups, with no statistically significant difference between the groups.
There are several possible explanations for these findings. First, at baseline,
many participants indicated that they had some education about water contamination,
causes of diarrhea and methods of its prevention. Second, the study community
placed a high priority on water quality, which, combined with education about
water contamination and diarrhea, could contribute to high levels of readiness
or motivation to adopt the water intervention. Put another way, it may be that
people were more in the 'action5 stage of change. MI is a strategy which is aimed
at resolving ambivalence towards behavior change. However, in this population,
little ambivalence was present. Further, people in the 'preparation' or 'action'
stages of change often respond to skill building. The educational brochures describing
the proper use of the water intervention may have been all that was needed to
render these skills, and this was provided equally in both of the behavior change
conditions. Third, chlorine disinfectant was available free of charge during the
study. Because no
monetary cost was involved and there was little effort needed for the participants
to receive the product, the ideal circumstances to measure motivation for long-term,
sustained behavior change may not have been present. Fourth, the lack of group
differences may have been related to training efforts. All of the NHCs received
instruction in elementary interpersonal interviewing skills, e.g. exclusively
attending to the respondent, not interrupting, building rapport and listening
carefully. These basic interviewing practices may have influenced the adherence
rates and resulted in a ceiling effect which made group differences undetectable.
To avoid this potentially confounding factor, in Field Trial 2 only the MI group
received interpersonal interviewing instruction. Lastly, problems were encountered
in the MI training during Field Trial 1. Scheduling difficulties and lack of attendance
due to family illnesses greatly hampered efficient progress. On five separate
occasions, NHCs' training sessions had to be canceled and rescheduled. When trainings
did occur, an extensive amount of time was needed to review because so much time
had elapsed since the previous meeting. This made actual practice exercises, which
are critical, difficult to realize. Thus, although the NHCs and nurses were instructed
in the key concepts and principles
of MI, the amount and continuity of trainings, degree of practice within trainings,
as well as time left for supervision and retraining in the field, were less than
ideal. For these reasons it may be that in Field Trial 1, the comparison and intervention
conditions were more similar than different, and therefore there was neither a
true comparison group nor an adequate intervention group which could be compared.
In Field Trial
2, some of these issues were addressed. Training in MI was improved. Specifically,
the extended length of the study allowed for valuable in-field instruction and
guided practice opportunities specifically tailored to individual needs. Also,
contact by the behavioral scientist with the comparison condition NHCs was eliminated.
The intervention was aimed at encouraging households to purchase chlorine, rather
than simply to utilize free chlorine. Chlorine sales may be a measure that is
more sensitive to and representative of intrinsic behavior change as it is a more
realistic measure of motivation and adherence with the water intervention, particularly
in a low SES group. Unfortunately, because of a lack of adequate funding, we were
unable to test residual chlorine levels, which is a more objective and precise
measure of adherence.
findings may provide support for using theory-driven behavioral change interventions
in the developing world (Stanton et al., 1992; Stanton,
1997). Results from Field Trial 2 are remarkably promising for the potential of
MI to effect long-term behavior change in at-risk populations. Rates of purchase
of sodium hypochlorite solution were much higher in the MI group than the Ed Only
group and were sustained during the 8-month course of the trial. This effect is
remarkable considering that the residents had to change their daily behavior as
well as expend extremely limited resources to purchase the disinfectant. Most
importantly, an efficacy study conducted concurrently with Field Trial 1 showed
that the sodium hypochlorite point-of-use disinfection method reduced diarrhea
incidence (Quick, Centers for Disease Control and Prevention, unpublished 7 data),
which is consistent with previous work accomplished in Latin America (Quick etai,
on results from these trials, Motivational Interviewing deserves more thorough
scientific investigation to determine its efficacy in preventing disease in the
developing world. There is also a need to refine implementation methods. For example,
weekly individualized visits may not be practical for most behavioral change
are implemented on a community level. The Zambian infrastructure provided by the
NHCs network is exceptional and very useful in this regard. More research is needed
to attempt to identify the salient features of MI and/or other brief interventions
to allow for less frequent, briefer contacts and those that can be applied to
whole communities, with special emphasis on the major health problems faced by
Funding for this
project was provided by the Office of Health Communication, National Center for
Infectious Diseases, Centers for Disease Control and Prevention and the United
States Agency for International Development. The authors would like to acknowledge
Dr Akiko Kimura, the Tropical Diseases Research Centre of Zambia, the Kitwe City
Council, Mr Peter Kalenga, Mr Promise Kaminsa, Dr Cleto Chashi, and the Urban
Health Program of BASICS for technical support, and the nurses who participated
in the trial: Margaret Bowa, Catherine Gondwe, Dalley Kafwimbi, Rachel Lungu,
Irene Musungu and Joyce Ndhlovu. The Neighborhood Health Committee workers, who
were central to this endeavor, are also gratefully acknowledged.
Bern, C, Marlines, J., de Zoysa, I. and Glass, R. (1992) The
magnitude of the global problem of acute diarrheal disease:
a ten-year update. Bulletin of the World Health Organization:
Bien, T. H, Miller, W. R. and Tonigan, J. S. (1993) Brief interventions
for alcohol problems: a review. Addiction: 88:
for Disease Control (1990) Waterborne disease outbreaks,
1986-1988. Morbidity and Mortality Weekly Report:
D. C., Salloum, I. M., Zuckoff, A., Kirisci, L. and Thase,
M. E. (1998) Increasing treatment adherence among
outpatients with depression and cocaine dependence:
results of a pilot study. American Journal of Psychiatry:
DiClemente, C. C, Prochaska, J. 0., Fairhurst, S. K., Velicer, W. R, Velasquez,
M. M. and Rossi, J. S. (1991) The processes of smoking cessation: an analysis
of precontem-plation, contemplation, and preparation stages of change. Journal
of Consulting and Clinical Psychology: 59: 295-304.
Heather, N., Rollnick, S., Bell, A. and Richmond, R. (1996) Effects of brief counselling
among male heavy drinkers identified on general hospital wards. Drug and Alcohol
Review: 15: 29-38.
Mates, A. and Shaffer, M. (1989) Membrane filtration differentiation of E.
coli from coliforms in the examination of water. Journal of Applied Bacteriology:
Miller, W. R., Benefield, R. G. and Tonigan, J. S. (1993) Enhancing motivation
for change in problem drinking: a
controlled comparison of two therapist styles. Journal of Consulting and Clinical
Psychology: 61: 455-461.
Miller, W. R. and Rollnick, S. (1991) Motivational Interviewing: Preparing People
to Change Addictive Behavior. The Guilford Press, New York.
10. Miller, W. R.,
Zweben, A., DiClemente, C. C. and Rychtarik, R. G. (1992) Motivational enhancement
therapy (MET): a clinical research guide for therapists treating individuals with
alcohol abuse and dependence (DHHS Publication No. ADM 92-1894). US Government
Printing Office, Washington, DC.
Mintz, E. D. and Tauxe, R. V. (1995) Safe water treatment and storage in the home:
a practical new strategy to prevent waterborne disease. Journal of the American
Medical Association: 273: 948-953.
Perz, C. A., Di Clemente, C. C. and Carbonari, J. P. (1996) Doing the right thing
at the right time? The interaction of stages and processes of change in successful
smoking cessation. Health Psychology: 15: 462-468.
Prochaska, J. 0. and Di Clemente, C. C. (1992) Stages of change in the modification
of problem behaviors. Progress in Behavior Modification: 28: 183-218.
J. 0. and DiClemente, C. C. (1984) Toward a comprehensive model of change. In
Miller, W. R. and Heather, N. (eds) Treating Addictive Behaviors: Processes
of Change. Plenum Press, New York, pp. 3-27.
Prochaska, J. 0. and Velicer, W. F. (1997) The trans-theoretical model of health
behavior change. American Journal of Health Promotion: 12: 38-48.
16. Project MATCH
Research Group (1997) Matching alcoholism treatments to client heterogeneity:
Project MATCH posttreatment drinking outcomes. Journal of Studies on Alcohol:
Project MATCH Research Group (1998) Matching alcoholism treatments to client heterogeneity:
Project MATCH three-year drinking outcomes. Alcoholism: Clinical
and Experimental Research: 22: 1300-1311.
Quick, R. E., Venczel, L. V, Gonzalez, O., Mintz, E. D., Highsmith, A. K., Espada,
A., Damiani, E., Bean, N. H., De Hannover, E. H. and Tauxe, R. V. (1996) Narrow-mouthed
water storage vessels and in situ chlorination in a Bolivian community: a simple
method to improve drinking water quality. American Journal of Tropical Medicine
& Hygiene: 54: 511-516.
Quick, R. E., Venczel, L. V, Mintz, E. D., Soleto, L., Aparicio, J. and Gironaz,
M. (1999) Diarrhea prevention through point-of-use water treatment and storage:
a promising new strategy.
Epidemiology and Infection: 122: 83-90.
Rollnick, S. and Miller, W. R. (1995) What is motivational interviewing? Behavioural
and Cognitive Psychotherapy: 23: 325-334.
Saunders, W, Wilkinson, C. and Phillips, M. (1995) The impact of a brief motivational
intervention with opiate users attending a methadone program. Addiction:
Smith, D. E., Heckemeyer, C. M., Kratt, P. P. and Mason, D. A. (1997) Motivational
interviewing to improve adherence to a behavioral weight-control program for older
obese women with NIDDM. A pilot study. Diabetes Care: 20: 52-54.
23. Stanton, B.
(1997) Editorial: good news for everyone? American Journal of Public Health:
Stanton, B., Black, R., Engle, P. and Pelto, G. (1992) Theory-driven behavioral
intervention research for the control of diarrheal diseases. Social Science
and Medicine: 35: 1405-1420.
Trigwell, P., Grant, P. J. and House, A. (1997) Motivation and glycemic control
in diabetes mellitus. Journal of Psychosomatic Research: 43: 307-315.
Beilin, L., Lord, T, Puddey, I., MacAdam, D. and Rouse, I. (1995) A controlled
trial of nurse counselling on lifestyle change for hypertensives treated in general
practice: preliminary results. Clinical and Experimental Pharmacology and Physiology:
A, Quick R and Yanduli V. Motivational Interviewing enhances the adoption of water
disinfection practices in Zambia. Health Promotion International 2000: 15:207-214.