Vous êtes sur la page 1sur 11

637604

research-article2016

HEBXXX10.1177/1090198116637604Health Education & BehaviorGeorge et al.

Article

Behavioral Determinants of Switching to


Arsenic-Safe Water Wells: An Analysis of
a Randomized Controlled Trial of Health
Education Interventions Coupled With
Water Arsenic Testing

Health Education & Behavior


111
2016 Society for Public
Health Education
Reprints and permissions:
sagepub.com/journalsPermissions.nav
DOI: 10.1177/1090198116637604
heb.sagepub.com

Christine Marie George, PhD1, Jennifer Inauen, PhD2,


Jamie Perin, PhD1, Jennifer Tighe, MSPH1, Khaled Hasan, MS3,
and Yan Zheng, PhD4

Abstract
More than 100 million people globally are estimated to be exposed to arsenic in drinking water that exceeds the World
Health Organization guideline of 10 g/L. In an effort to develop and test a low-cost sustainable approach for water arsenic
testing in Bangladesh, we conducted a randomized controlled trial which found arsenic educational interventions when
combined with fee-based water arsenic testing programs led to nearly all households buying an arsenic test for their drinking
water sources (93%) compared with only 53% when fee-based arsenic testing alone was offered. The aim of the present
study was to build on the findings of this trial by investigating prospectively the psychological factors that were most strongly
associated with switching to arsenic-safe wells in response to these interventions. Our theoretical framework was the
RANAS (risk, attitude, norm, ability, and self-regulation) model of behavior change. In the multivariate logistic regression
model of 285 baseline unsafe well users, switching to an arsenic-safe water source was significantly associated with increased
instrumental attitude (odds ratio [OR] = 9.12; 95% confidence interval [CI] = [1.85, 45.00]), descriptive norm (OR = 34.02;
95% CI = [6.11, 189.45]), coping planning (OR = 11.59; 95% CI = [3.82, 35.19]), and commitment (OR = 10.78; 95% CI =
[2.33, 49.99]). In addition, each additional minute from the nearest arsenic-safe drinking water source reduced the odds of
switching to an arsenic-safe well by more than 10% (OR = 0.89; 95% CI = [0.87, 0.92]). Future arsenic mitigation programs
should target these behavioral determinants of switching to arsenic-safe water sources.
Keywords
arsenic, Bangladesh, community health, environmental health, formative evaluation, global health, health communications,
health promotion, water
More than 100 million people in India, Bangladesh, Vietnam,
Nepal, and Cambodia are using drinking water sources with
arsenic concentration exceeding the World Health
Organization (WHO) guideline of 10 g/L (Ahmed etal.,
2006). In Bangladesh alone an estimated 22 million people
are exposed to naturally occurring arsenic concentrations in
groundwater exceeding the Bangladesh arsenic standard of
50 g/L, five times higher than the WHO guideline (Flanagan,
Johnston, & Zheng, 2012; UNICEF, 2009). Exposure to elevated levels of inorganic arsenic is associated with increased
risk for cancers of the lung, bladder, and skin; skin lesions;
respiratory effects; and all-cause and chronic disease mortality (Argos etal., 2010; Argos etal., 2011; Smith etal., 1992;
Smith etal., 2013).

Attempts to reduce arsenic exposure in Bangladesh have


had limited success. Of the initially arsenic exposed
population of 28 to 35 million, 57% remain exposed (Ahmed
1

Johns Hopkins Bloomberg School of Public Health, Baltimore, MD, USA


Swiss Federal Institute of Aquatic Science and Technology, Dbendorf,
Switzerland
3
International Centre for Diarrhoeal Disease Research, Bangladesh,
Dhaka, Bangladesh
4
United Nations Childrens Fund Bangladesh, Dhaka, Bangladesh
2

Corresponding Author:
Christine Marie George, Department of International Health, Johns
Hopkins Bloomberg School of Public Health, 615 North Wolfe Street,
Room E5535, Baltimore, MD 21205, USA.
Email: cgeorg19@jhu.edu

Downloaded from heb.sagepub.com by guest on May 27, 2016

Health Education & Behavior

etal., 2006). The most common form of arsenic mitigation in


Bangladesh is well switching at 29%. This involves a household with an unsafe well relative to the Bangladesh arsenic
standard switching to an arsenic-safe well located in their
community, typically a neighbors well (Ahmed etal., 2006).
However, an important prerequisite to well switching is
knowledge of the arsenic status of wells located in ones
community. Without this information well switching is not
possible. Most recent estimates suggest that nearly half of
wells in arsenic affected areas of Bangladesh are untested for
arsenic, meaning that millions are likely unknowingly
exposed to elevated levels of arsenic in their drinking water
(UNICEF, 2009).
The second most common arsenic mitigation option in
Bangladesh is the use of deep tubewells at 12%. Deep tubewells (>150 m) are generally able to tap low-arsenic aquifers
(Radloff etal., 2011). However, a major barrier to their widespread adoption is the high cost that can be upwards of
US$1,500. Arsenic mitigation options such as rainwater collectors, dugwells, arsenic filters, and pond sand filter are used by
less than 2% of the arsenic-affected population. This is likely
due to the high cost, slow flow rate of the filters, and the lack of
user friendliness of the technologies (Hossain etal., 2005).
Therefore, well switching because of the little or no cost associated with this option has been the most widely adopted public
health approach for arsenic mitigation in Bangladesh.
Even for those households that know their wells are
unsafe for arsenic, and who have access to arsenic-safe wells
in their communities, well switching behavior is often limited (UNICEF, 2009). This is because there are personal,
situational, and social factors that are important determinants
of well switching beyond access to arsenic testing services
(Mosler, 2012). Previous studies have found far distance to
the nearest arsenic-safe well, degree of arsenic contamination in the households well, well ownership, and a higher
proportion of arsenic-unsafe wells in a community (>60%
unsafe wells) to all be significant determinants of the continued use of an arsenic unsafe water source (Aziz, Boyle, &
Rahman, 2006; Chen etal., 2007; George, Geen, etal., 2012;
Hoque etal., 2004; Opar etal., 2007). There are also social
barriers to well switching such as arsenic-safe well owners
refusing to share their well water, and household concerns
about women traveling long distances to collect arsenic-safe
water (George, Geen, etal., 2012; Hoque etal., 2004).
Recently, the risk, attitude, norm, ability, and self-regulation
(RANAS) model of behavior change has been applied to identify the psychological factors that influence the use of arsenic-safe water sources (Inauen, Hossain, Johnston, & Mosler,
2013; Inauen & Mosler, 2013; Inauen, Tobias, & Mosler, 2013,
2014; Mosler, 2012; Mosler, Blchliger, & Inauen, 2010). The
RANAS model combines factors derived from the protection
motivation theory and Ajzens theory of planned behavior
(Ajzen, 1985; Rogers & Prentice-Dunn, 1997). In this model,
psychological factors are divided into risk, attitudinal, normative, ability, and self-regulation factors. A thorough

understanding of these psychological factors has been shown to


enhance the development of effective arsenic mitigation strategies (Inauen & Mosler, 2013). Previous studies have demonstrated that behavior change interventions using psychological
theory have a higher efficacy than those providing health information alone (Dombrowski etal., 2012; Inauen & Mosler, 2013;
Michie & Johnston, 2012; Michie, Johnston, Francis, Hardeman,
& Eccles, 2008). Cross-sectional studies using the RANAS
model have found that users of arsenic-safe water options have
higher descriptive and injunctive norm, self-efficacy, and a
stronger commitment to use arsenic-safe water options (Inauen,
Hossain, etal., 2013; Inauen & Mosler, 2013; Inauen, Tobias,
etal., 2013). In a recent randomized controlled trial developed
based on an empirical investigation of the RANAS model, it
was found that arsenic interventions that disseminated arsenic
information with commitment-based behavior change techniques were more effective than interventions providing information alone (Inauen & Mosler, 2013).
In an effort to develop and test a low-cost sustainable
approach for water arsenic testing in Bangladesh, we recently
conducted a randomized controlled trial of a fee-based water
arsenic testing program (George, Inauen, Rahman, & Zheng,
2013). In this three-arm trial, we compared a fee-based water
arsenic testing program alone delivered by a community promoter to these same services combined with either a community-level and/or household-level arsenic education program.
Arsenic tests were sold in this trial at a cost of US$0.28 with
the actual cost of the arsenic test being lower at US$0.16.
Strikingly, 93% of households in both educational intervention arms purchased a water arsenic test for their household
well, in comparison to only 53% of households in the water
arsenic testing only arm. Furthermore, well switching was
also significantly higher in the household- and communitylevel education arm compared with the water arsenic testing
only arm (39% vs. 26%, p = .04; George etal., 2013). These
findings suggest that fee-based water arsenic testing when
combined with arsenic education presents a promising
approach to increase access to and use of water arsenic testing services and to encourage the use of arsenic-safe water
sources.
The primary aim of the present study is to investigate
using the RANAS model the psychological factors that were
most influential in predicting the use of arsenic-safe water
sources among baseline unsafe well users in our recent randomized controlled trial. The secondary aim is to determine
if the arsenic interventions implemented significantly
changed the measured psychological factors.

Method
Setting
This trial was conducted in Shibalaya Upazlia in the
Manikganj district of Bangladesh from April to August 2011,
and the trial was conducted in partnership with the Christian

Downloaded from heb.sagepub.com by guest on May 27, 2016

George et al.

Figure 1. Study design.

Commission for Development Bangladesh (CCDB), a nongovernmental organization focused on poverty reduction.

Study Design
The study design was a cluster randomized controlled trial
(Figure 1). The detailed methodology used for the trial
including the eligibility criteria and detailed description of
each intervention arm have been reported elsewhere (George
etal., 2013). Briefly, 9 villages of 50 randomly selected
respondents each were randomized to receive one of the following three interventions by a CCDB promoter: (1) feebased arsenic testing with a household-level arsenic
education program (HE arm), (2) fee-based arsenic testing
with a household-level education arsenic education program
and a community-level local media campaign on arsenic
(HELM arm), and (3) fee-based arsenic testing alone (control arm).

Questionnaire
A structured questionnaire was administered to respondents at baseline and follow-up by personal interviews to
assess psychological factors from the RANAS model and
collect household demographic information (Mosler etal.,
2010). The questionnaire was consistent with the tool used

in previous arsenic-related studies of the RANAS model


(Inauen, Hossain, etal., 2013; Inauen & Mosler, 2013;
Inauen, Tobias, etal., 2013; Mosler etal., 2010). Questions
were administered by trained professional interviewers
from Dhaka, the capital of Bangladesh. In line with the
RANAS model, the following psychological factors were
measured in this analysis: perceived vulnerability and
severity, knowledge, instrumental attitude, affective attitude, descriptive norm, injunctive norm, personal norm,
self-efficacy, coping planning, remembering/forgetting,
and commitment. The definitions, example items, and
internal consistencies for the psychological factors are
described in Table 1. With the exception of knowledge,
which was a composite score of yes/no answers, all items
were assessed using a Likert-type scale on a scale of either
1 to 5 for unipolar items (e.g., 1 = not at all confident to 5
= very confident) or 1 to 9 for bipolar items (e.g., 1 = very
low to 9 = very high). Items were transformed to a scale
from 0 to 1 for analysis of psychological factors to ensure
standardization across factors.

Interventions
Fee-Based Arsenic Testing. Arsenic testers in all study arms
offered arsenic testing services on site using the EconoQuick arsenic test kit (Industrial Test Systems, Inc, Rock

Downloaded from heb.sagepub.com by guest on May 27, 2016

Health Education & Behavior

Table 1. Psychological Factors in Risk, Attitude, Norm, Ability, and Self-regulation (RANAS) Model of Behavior Change and their
Assessment.a
Type of
factor

Factors

Definition

Example item

Risk
Perceived
factors vulnerability
Severity

Attitude
factors

Norm
factors

Ability
factors

A persons perception of his/her risk of


contracting arsenicosis (Orbell etal., 2009).
Perception concerning the seriousness of the
consequences of contracting an illness (Floyd,
Prentice-Dunn, & Rogers, 2000)
Arsenic
An understanding of the health implications of
knowledge
arsenic and arsenic mitigation options
Instrumental
Beliefs about the benefits and costs of a behavior
attitude
(Fishbein & Ajzen, 2010)
Affective
Feelings that arise when performing a behavior or
attitude
thinking about it (Trafimow & Sheeran, 1998)
Personal norm Conveys what an individual personally believes she
or he should do (Schwartz, 1977)
Descriptive
Perceptions about which behaviors are typically
norm
performed by others (Cialdini etal., 2006)
Injunctive
Perceptions about which behaviors are typically
norm
approved or disapproved (Schultz, Nolan,
arsenic-safe
Cialdini, Goldstein, & Griskevicius, 2007)
option
Self-efficacy
The belief in ones capabilities to organize and
execute the courses of action required to
manage prospective situations (Bandura, 1997)
Coping
planning

Remembering

Commitment
strength

The presumption of possible barriers and


the invention of ways to overcoming them
(Schwarzer, 2008)
To perform a behavior, it has to be remembered
at the right time/situation (Tobias, 2009)
Actual strength of the decision to display a
behavior (Gollwitzer & Sheeran, 2006)

Cronbach
alpha

How high or low do you feel are the chances


that you get arsenicosis (arsenic disease)?
Imagine that you contracted arsenicosis,
how severely would this impact your life in
general?
Can arsenic be removed by boiling water?

.76

Do you think that collecting water from an


arsenic mitigation option is time-consuming?
How much do you like or dislike the taste of
water from an arsenic mitigation option?
How annoyed do you feel if you forget to collect
water from an arsenic mitigation option?
How many people, excluding your relatives,
collect water from an arsenic mitigation option?
How good or bad would you say is it to
collect water from an arsenic mitigation
option?

.88

How confident are you that you can


resist drinking water from the arsenic
contaminated shallow tubewell even if you
have to walk a long distance to reach the
next safe tubewell?
Do you have a detailed plan regarding what to
do when the arsenic mitigation option gets
broken?
How often does it happen that you forget
to go to collect water from an arsenic
mitigation option?
Do you feel committed to collect water from
an arsenic mitigation option?

.96
.75

.73
b

.63
.64

.95

.89
b

Adapted from Mosler (2012). bCronbach alpha could not be calculated because there was only one item in this factor.

Hill, SC; USA, Part No. 481298; George, Zheng, etal.,


2012). If a study participants primary drinking water source
was found to be unsafe relative to arsenic, assistance to
locate a nearby arsenic-safe drinking water source was provided for all study arms. This was typically done through
offering fee-based arsenic testing to households the participant was willing to share their drinking water with.
Household-Level Arsenic Educational Program. CCDB community promoters in the HE and HELM arms of the study delivered a household-level arsenic educational program to all
households enrolled in these study arms. This was disseminated as a 25- to 30-minute educational module using a
structured script and flash cards with pictorial images of the
health implications of chronic arsenic exposure, safe uses of
arsenic-contaminated water (e.g., bathing), and arsenic mitigation options. At the end of the module, the audience was

asked to pledge their commitment to drink arsenic-safe water


and share arsenic-safe wells with others.
Local Media Campaign. Two promoters in the HELM arm of
the study, different from the CCDB promoters, delivered a
community-level local media campaign on arsenic using a
micro-rickshaw (a rickshaw with microphone equipment),
and conducted 1.5-hour community meetings with arsenic
educational sessions and theater performances. At the end of
community sessions, participants were asked to commit to
only drink water from arsenic-safe options.

Ethical Approval
The study protocol was approved by the Swiss Federal
Institute of Aquatic Science and Technology. Informed consent was obtained from all study respondents.

Downloaded from heb.sagepub.com by guest on May 27, 2016

George et al.

Statistical Analysis
For a comparison of the baseline household- and individuallevel characteristics by study arm, a chi-square test was performed for categorical variables and an ANOVA test for
continuous variables. Pearson correlations were calculated
for changes in psychological factors measured between baseline and follow-up (follow-up baseline). ANOVA tests
were used to compare psychological factors at baseline and
the change in factors between baseline and follow-up by
study arm. Logistic regression models using generalized estimating equations to account for clustering within villages
were performed to estimate the odds of switching to an arsenic-safe well with the change in psychological factors
between baseline and follow-up as predictors. Self-reported
time to an arsenic-safe drinking water option was also
included, since this variable was found previously to be a
significant barrier to the use of an arsenic-safe water source
(George, Geen, etal., 2012; Opar etal., 2007; Van Geen
etal., 2002). Paired t tests were conducted to compare the
change in factors between baseline and follow-up. All analyses were performed using SAS, Version 9.3 (SAS Institute
Inc., Cary, NC).

Results
Out of a total of 579 participants screened for eligibility in the
randomized controlled trial, 452 were enrolled (Figure 1). One
hundred and three were found to be ineligible, and 24 were
unwilling to participate. All respondents were female. There
were no significant differences in age, radio or television ownership, or monthly income between the three study arms
(George etal., 2013). Two hundred and eighty-five study participants were found to have a well with an unsafe arsenic concentration relative to the Bangladesh arsenic standard at
baseline. Thirty-one percent of these participants (89/285)
reported switching to arsenic-safe wells at follow-up. There
were 122 untested well users at baseline who were found be
drinking from an arsenic-safe water source (baseline safe
water users), and 28 untested well users at baseline declined
to receive an arsenic test for their well. No safe well users
reported switching to alternative wells during the study period.
There were no significant differences observed at baseline
across study arms in arsenic knowledge, affective attitude,
descriptive norm, injunctive norm, personal norm, self-efficacy, remembering, or commitment (Table 2). There were
however significant differences observed across study arms
at baseline for perceived vulnerability (p = .02), perceived
severity (p = .01), instrumental attitude (p = .04), and coping
planning (p = .01).
For unsafe well users at baseline, Pearson correlations
were similar across study arms; therefore, we report combined findings for all study arms in Table 3. The largest
Pearson correlations coefficients observed for the change in
psychological factors between baseline and follow-up and the

use of arsenic-safe wells were for the following factors: perceived vulnerability (r = 0.53, p < .0001), descriptive norm
(r = 0.43, p < .0001), and commitment (r = 0.46, p < .0001).
In addition, highly significant correlations were observed
between the change in coping planning and self-efficacy (r =
0.76, p < .0001), commitment and self-efficacy (r = 0.63, p <
.0001), and commitment and coping planning (r = 0.51, p <
.0001). For the change in psychological factors between baseline and follow-up for baseline unsafe well users, only arsenic
knowledge (p = .03) was significantly different across study
arms with the mean change in the control arm being the highest (Table 2). All psychological factors except for affective
attitude significantly changed from baseline to follow-up
when all study arms were combined (p = .60; Table 4).
In the multivariate logistic regression model of well
switching, changes in perceived vulnerability were significantly associated well switching (odds ratio [OR] = 0.01;
95% confidence interval [CI] = [0.001, 0.05]; Table 5). Well
switching was also significantly associated with increases in
instrumental attitude (OR = 9.12; 95% CI = [1.85, 45.00]),
descriptive norm (OR = 34.02; 95% CI = [6.11, 189.45]),
coping planning (OR = 11.59; 95% CI = [3.82, 35.19]), and
commitment (OR = 10.78; 95% CI = [2.33, 49.99]). In addition, greater time to the nearest arsenic-safe water source in
minutes was inversely associated with well switching (OR =
0.89; 95% CI = [0.87, 0.92]). Consistent with this finding the
mean time to an arsenic-safe well was 6 minutes for baseline
arsenic-safe well users compared with 31 minutes for baseline arsenic-unsafe well users (p < .0001). As a sensitivity
analysis, we computed the regression again including baseline factors found to be significant across study arms in the
multivariate model, the results remained substantively
unchanged. For safe well users at baseline, the only significant differences observed for the changes in psychological
factors across study arms were for perceived severity and
personal norms (Table 6).

Discussion
This study investigated the importance of psychological factors in the RANAS model in explaining switching to arsenicsafe wells among baseline unsafe well users. We found that
increases in descriptive norm, instrumental attitude, coping
planning, and commitment were all significantly associated
with switching to an arsenic-safe well at follow-up.
Furthermore, all fee-based water arsenic testing program
interventions evaluated were associated with significant
changes in all psychological factors measured from the
RANAS model except for affective attitude. This result suggests that fee-based water arsenic testing itself may have
contributed to these changes in psychological factors irrespective of the educational messages on arsenic delivered in
combination with testing. Future arsenic interventions should
target these identified behavioral determinants of switching
to an arsenic-safe well.

Downloaded from heb.sagepub.com by guest on May 27, 2016

Health Education & Behavior

Table 2. ANOVA Analysis of Psychological Factors by Study Arm for Baseline Unsafe Well Users (N = 285).
Baseline
Factorsa
Perceived vulnerability

Perceived severity

Arsenic knowledge

Instrumental attitude

Affective attitude

Descriptive norm

Injunctive norm

Personal norm

Self-efficacy

Coping planning

Remembering

Commitment

Change (baseline
follow-up)

Follow-up

Study arm

SD

p valueb

SD

p valueb

SD

p valueb

Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm

72
91
122
72
91
122
72
91
122
72
91
122
72
91
122
72
91
122
72
91
122
72
91
122
72
91
122
72
91
122
72
91
122
72
91
122

0.79
0.74
0.74
0.93
0.89
0.92
0.49
0.47
0.51
0.72
0.80
0.72
0.84
0.84
0.87
0.44
0.46
0.49
0.84
0.81
0.84
0.26
0.27
0.38
0.37
0.36
0.47
0.31
0.32
0.37
0.20
0.23
0.23
0.24
0.26
0.38

0.12
0.14
0.15
0.12
0.12
0.11
0.13
0.15
0.16
0.17
0.18
0.19
0.08
0.12
0.10
0.09
0.13
0.14
0.10
0.11
0.09
0.19
0.18
0.27
0.15
0.17
0.23
0.11
0.18
0.17
0.00
0.10
0.09
0.15
0.15
0.26

.02

0.64
0.67
0.64
0.60
0.60
0.60
0.87
0.87
0.87
0.70
0.70
0.70
0.89
0.89
0.89
0.60
0.60
0.60
0.70
0.71
0.71
0.40
0.40
0.40
0.52
0.50
0.60
0.40
0.40
0.60
0.20
0.20
0.20
0.40
0.60
0.80

0.29
0.20
0.20
0.07
0.08
0.07
0.13
0.12
0.14
0.24
0.20
0.22
0.07
0.09
0.08
0.16
0.15
0.14
0.08
0.09
0.08
0.23
0.26
0.25
0.25
0.22
0.22
0.26
0.24
0.24
0.11
0.11
0.13
0.23
0.25
0.24

.29

0.19
0.27
0.23
0.31
0.34
0.34
0.38
0.35
0.31
0.13
0.07
0.08
0.01
0.02
0.01
0.13
0.13
0.10
0.14
0.18
0.17
0.19
0.16
0.10
0.19
0.19
0.11
0.19
0.18
0.15
0.04
0.09
0.06
0.33
0.34
0.27

0.24
0.22
0.25
0.15
0.14
0.14
0.20
0.18
0.19
0.27
0.26
0.28
0.15
0.11
0.13
0.18
0.18
0.21
0.14
0.12
0.13
0.34
0.30
0.38
0.28
0.30
0.34
0.31
0.31
0.29
0.14
0.11
0.17
0.30
0.30
0.37

.10

.23

.03

.24

.12

.49

.11

.13

.14

.69

.20

.31

.01

.42

.04

.20

.08

.11

.13

.17

.01

.05

.13

.76

.19

.17

.20

.40

.80

.38

.52

.64

.55

.07

Note. HE arm = fee-based arsenic testing with a household-level arsenic education program; HELM arm = fee-based arsenic testing with a household-level
education arsenic education program and a community-level local media campaign on arsenic; control arm = fee-based arsenic testing alone.
a
Psychological factors ranged from 0 to 1. bp values compared psychological factors by study arm using an ANOVA test.

Ability factors related to identifying strategies to overcome barriers to accessing an arsenic-safe water source (coping planning) and stronger commitment to using an
arsenic-safe water source were all significantly associated
with well switching. Consistent with our present study, coping planning was significantly associated with the use of an
arsenic-safe well in a previous cross-sectional study in rural
Bangladesh (Inauen, Hossain, etal., 2013). This finding suggests that households that switched to arsenic-safe drinking
water sources had more detailed plans on how to overcome

barriers to the use of their arsenic mitigation options. The


significant association we observed between increased commitment and the use of arsenic-safe water sources is consistent with a recent randomized controlled trial of arsenic
interventions in rural Bangladesh (Inauen etal., 2014). In
this trial, it was found that commitment increased by behavior change techniques including reminders and implementation intentions significantly increased well switching. In
contrast to previous investigations, however, we did not
observe a significant association between self-efficacy and

Downloaded from heb.sagepub.com by guest on May 27, 2016

Downloaded from heb.sagepub.com by guest on May 27, 2016

Use of arsenic-safe well


Change in perceived vulnerability
Change in severity
Change in arsenic knowledge
Change in instrumental attitude
Change in affective attitude
Change in descriptive norm
Change in injunctive norm
Change in personal norm
Change in self-efficacy
Change in coping planning
Change in remembering
Change in commitment

*p < .05. **p < .0001.

1.
2.
3.
4.
5.
6.
7.
8.
9.
10.
11.
12.
13.

Variable

0.53**
0.04
0.01
0.04
0.10
0.43**
0.14*
0.31**
0.33**
0.33**
0.05
0.46**

0.07
0.07
0.08
0.04
0.30**
0.12*
0.15
0.18*
0.19*
0.05
0.25**

0.11
0.06
0.16**
0.08
0.16**
0.29
0.21*
0.21*
0.01
0.19*

0.02
0.00
0.08
0.00
0.11
0.06
0.01*
0.12*
0.05

0.02
0.08
0.19**
0.15*
0.17*
0.20*
0.01
0.06

0.01
0.38**
0.21**
0.31**
0.24**
0.13*
0.37**

0.02
0.32**
0.12*
0.13*
0.05
0.32**

0.21**
0.38**
0.29**
0.26**
0.36**

Pearson correlation coefficients

0.44**
0.34**
0.17*
0.49**

0.76**
0.02*
0.63**

10

0.03
0.51**

11

0.04

12

13

Table 3. For All Study Arms, Intercorrelations Between the Use of Arsenic-Safe Wells and the Change in Psychological Factors Between Baseline and Follow-Up for Unsafe Well
Users at Baseline (N = 285).

Health Education & Behavior

Table 4. Change in Psychological Factors Between Baseline and Follow-Up for Baseline Unsafe Well Users (N = 285).
Factorsa
Perceived vulnerability
Perceived severity
Arsenic knowledge
Instrumental attitude
Affective attitude
Descriptive norm
Injunctive norm
Personal norm
Self-efficacy
Coping planning
Remembering
Commitment

Baseline, M (SD)

Follow-up, M (SD)

p valueb

285
285
285
285
285
285
285
285
285
285
285
285

0.75 (0.14)
0.91 (0.12)
0.49 (0.15)
0.74 (0.18)
0.85 (0.10)
0.47 (0.13)
0.83 (0.10)
0.31 (0.23)
0.41 (0.20)
0.34 (0.16)
0.22 (0.08)
0.30 (0.21)

0.53 (0.20)
0.58 (0.08)
0.83 (0.13)
0.65 (0.22)
0.86 (0.08)
0.59 (0.15)
0.67 (0.08)
0.46 (0.25)
0.57 (0.23)
0.51 (0.25)
0.28 (0.12)
0.61 (0.24)

<.0001
<.0001
<.0001
<.0001
.60
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001
<.0001

Psychological factors ranged from 0 to 1. bPaired t test.

Table 5. Multivariate Logistic Regression Model of Switching to an Arsenic-Safe Well (Well Switching) With the Change in
Psychological Factors Between Baseline and Follow-Up as Predictors for Baseline Unsafe Well Users.
Change in factorsa
(follow-up baseline)
Perceived vulnerability
Perceived severity
Arsenic knowledge
Instrumental attitude
Affective attitude
Descriptive norm
Injunctive norm
Personal norm
Self-efficacy
Coping planning
Remembering
Commitment
Reported walking time to
arsenic-safe water source
(minutes)d

Well switching

No well switching

SD

SD

ORb,c

Lower
bound
(95% CI)

89
89
89
89
89
89
89
89
89
89
89
89
89

0.41
0.32
0.34
0.11
0.02
0.24
0.13
0.30
0.31
0.32
0.05
0.53
12.00

0.24
0.14
0.18
0.27
0.12
0.17
0.11
0.33
0.26
0.26
0.13
0.22
6.82

196
196
196
196
196
196
196
194
194
194
195
194
191

0.14
0.33
0.34
0.09
0.01
0.06
0.17
0.07
0.09
0.10
0.07
0.20
23.25

0.19
0.15
0.20
0.28
0.13
0.18
0.13
0.33
0.31
0.29
0.15
0.32
15.01

0.01
0.32
1.96
9.12
0.16
34.02
5.93
1.71
0.70
11.59
0.56
10.78
0.89

0.001
0.06
0.32
1.85
0.02
6.11
0.51
0.31
0.10
3.82
0.02
2.33
0.87

Upper
bound
(95% CI)

p value

0.05
1.84
12.08
45.00
1.10
189.45
69.02
9.42
4.95
35.19
15.19
49.99
0.92

<.0001
.20
.47
.01
.06
<.0001
.16
.54
.72
<.0001
.73
.002
<.0001

Note. OR = odds ratio; CI = confidence interval.


a
Change in psychological factors ranged from 1 to 1. bAll factors are added in a single multivariate model. cLogistic regression models using generalized
estimating equations (GEE). dThis variable is the time to the nearest arsenic-safe water source at follow-up.

the use of arsenic-safe water options (Inauen, Hossain, etal.,


2013; Mosler etal., 2010). This is likely due to the high intercorrelations between self-efficacy and commitment, coping
planning, descriptive norms, and instrumental attitude.
Decreased perceived vulnerability to the health effects of
arsenic exposure, a risk factor in the RANAS model, was significantly associated with the use of arsenic-safe water sources
by baseline unsafe well users. This finding was previously
reported in Inauen etal. and is likely due to reverse causality
(Inauen & Mosler, 2013). Those respondents that switched to
arsenic-safe drinking water sources likely felt less vulnerable
to the health implications of arsenic. Increased knowledge of
arsenic was also not found to be significantly associated with

well switching. Our finding is consistent with Inauen etal.


(2014) and Opar etal. (2007), which found that information on
the health implications of arsenic and mitigation options alone
was not sufficient to lead to substantial increases in well
switching.
Perceptions about which behaviors are performed by
others (descriptive norm) were significantly associated
with the use of an arsenic-safe water source. This is consistent with several previous studies in rural Bangladesh
(Inauen, Hossain, etal., 2013; Inauen, Tobias, etal., 2013;
Inauen etal., 2014; Mosler etal., 2010). Increased descriptive norms was likely attributed to the water arsenic testing
program that provided access to arsenic testing and the

Downloaded from heb.sagepub.com by guest on May 27, 2016

George et al.
Table 6. ANOVA Analysis of Change in Psychological Factors
(Follow-Up Baseline) by Study Arm for Baseline Safe Wells
Users (N = 122).
Change in factorsa
Perceived vulnerability

Perceived severity

Arsenic knowledge

Instrumental attitude

Affective attitude

Descriptive norm

Injunctive Norm

Personal norm

Self-efficacy

Coping planning

Remembering

Commitment

Study arm

SD

p valueb

Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control Arm
HE Arm
HELM Arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm
Control arm
HE arm
HELM arm

40
45
21
40
45
21
40
45
21
40
45
21
40
45
21
40
45
21
40
45
21
40
45
21
40
45
21
40
45
21
40
45
21
40
45
21

0.53
0.52
0.45
0.41
0.34
0.39
0.29
0.27
0.22
0.34
0.36
0.38
0.01
0.00
0.04
0.15
0.19
0.13
0.18
0.15
0.11
0.03
0.21
0.02
0.24
0.17
0.26
0.31
0.36
0.27
0.03
0.01
0.00
0.35
0.41
0.33

0.20
0.19
0.20
0.10
0.11
0.09
0.19
0.22
0.21
0.25
0.24
0.28
0.11
0.13
0.25
0.17
0.18
0.22
0.09
0.12
0.16
0.41
0.44
0.39
0.31
0.32
0.37
0.22
0.25
0.32
0.16
0.06
0.15
0.32
0.35
0.31

.27

.02

.50

.82

.46

.36

.08

.03

.47

.40

.74

.58

Note. HE arm = fee-based arsenic testing with a household-level arsenic


education program; HELM arm = fee-based arsenic testing with a
household-level education arsenic education program and a communitylevel local media campaign on arsenic; control arm = fee-based arsenic
testing alone.
a
Change in psychological factors ranged from 1 to 1. bp values compared
psychological factors by study arm using an ANOVA test.

community promoters that assisted households with locating arsenic-safe drinking sources. These activities likely
brought attention to former unsafe well users that were currently practicing well switching.
Situational factors also played an important role in well
switching. Each minute from the nearest arsenic-safe drinking water source reduced the odds of switching to an arsenicsafe well by 10%. This is consistent with findings from Van

Geen etal. (2002) and George, Geen, etal. (2012), which


found distance to the nearest arsenic-safe water source to be
a significant predictor of well switching. These findings
highlight an important structural barrier to well switching.
This approach is only feasible in an environment where arsenic-safe wells are located in close proximity to the homes of
unsafe well users. Consistent with this, a previous study of
arsenic educational interventions in Bangladesh found that
well switching rates were 72% among unsafe well users living in villages where 0% to 60% of wells were arsenic contaminated, compared with only 35% in villages with greater
or equal to 60% of wells arsenic contaminated (George,
Geen, etal., 2012).
Beliefs about the cost or effort of collecting water from an
arsenic-safe option (instrumental attitude) was significantly
associated with well switching. This finding suggests that
individuals that switched to an arsenic-safe drinking water
option at follow-up perceived it to be less time-consuming
and effortful than did persons who did not switch. Perceived
taste, temperature, and color of water from an arsenic-safe
drinking water source (affective attitude) did not change over
the study period nor was it a significant predictor of well
switching. This is in contrast to previous studies that found
that taste influenced the use of certain types of arsenic mitigation options (Inauen, Tobias, etal., 2013; Mosler etal.,
2010). This is likely because households were all using wells
for drinking water at baseline, and not alternative arsenic
mitigation options such as pond sand filters or rain water harvesting that were included in these previous studies.
The RANAS model recommends several interventions
that can be implemented to increase commitment, descriptive norm, coping planning, and instrumental attitude
(Mosler, 2012). To increase commitment previous arsenic
interventions studies in Bangladesh have found that a selfcommitment, an oral or written pledge, that is private or public to practice the target behaviors is effective (Inauen &
Mosler, 2013; Inauen etal., 2014). Descriptive norms can
also be increased by interventions promoting a public commitment of a desired behavior by showing that there are
other people performing the new behavior (Mosler, 2012).
Outcome feedback can be used to increase coping planning
(Mosler, 2012). In Inauen etal. (2014), an intervention targeting implementation intentions using pictograms with
typical tasks during the day when water would need to be
collected was found to be effective in promoting well switching when combined with arsenic information and reminders
(Inauen & Mosler, 2013). To increase instrumental attitude,
persuasive interventions including strong arguments or
peripheral cues can be used (Mosler, 2012).
This study supports the growing body of evidence suggesting that there needs to be a shift in the current paradigm around arsenic mitigation in Bangladesh from one
that focuses almost exclusively on increasing arsenic
knowledge to a more tailored approach to target the identified behavioral determinants of the use of arsenic-safe

Downloaded from heb.sagepub.com by guest on May 27, 2016

10

Health Education & Behavior

water. Furthermore, these findings highlight the need for


national policies around arsenic mitigation in Bangladesh
that support the use of several arsenic mitigation options
beyond solely well switching, since well switching does
not seem to be a viable option for households located more
than a few minutes away from an arsenic-safe drinking
water source.
This study had several limitations. First, there was no
control arm that did not receive arsenic testing. Therefore,
we cannot tell if the observed increase in psychological factors was due to secular trends or the interview process itself.
We did not think it was ethical to have a control arm in our
trial given the vast body of literature demonstrating the
health implications of arsenic exposure. Second, well
switching was only measured at baseline and follow-up.
Therefore, we cannot determine if changes in psychological
factors occurred before or after well switching. Future studies should assess the impact of arsenic interventions on well
switching and the measured psychological factors from the
RANAS model more frequently over a longer period of
time. Third, we did not collect information on biomarkers of
arsenic exposure such as urinary arsenic to validate selfreported well switching behavior. However, previous studies have validated the use of self-reported well switching in
comparison with urinary arsenic measurements (Chen etal.,
2007; George, Geen, etal., 2012). Furthermore, household
drinking water samples were found to be highly correlated
with self-reported drinking water source use in Bangladesh
(Inauen& Mosler, 2015).
Our findings based on the use of the RANAS model demonstrated that descriptive norms, instrumental attitudes, coping planning, perceived vulnerability, and commitment were
significant behavioral determinants of switching to an arsenic-safe water source. Future arsenic mitigation programs
should target these behavioral determinants of switching to
an arsenic-safe water source to develop evidence-based theoretically informed arsenic interventions to reduce arsenic
exposure in affected populations globally.
Acknowledgments
We would like to thank the Eawag team, the Swiss Federal Institute
of Aquatic Science and Technology, and the Christian Commission
for Development Bangladesh for their support of our project. We
also would like to thank our promoters, the project team at the
UNICEF, and Columbia University office in Bangladesh, Sheikh
Masudur Rahman, Patricia Portela Souza, Shirin Hussain,
Mojahidul Hossain, Zakir Hossain, and Masud Noor.

Authors Note
All authors have had full access to all data in the study and take
responsibility for its integrity and the accuracy of its analysis.

Declaration of Conflicting Interests


The authors declared no potential conflicts of interest with respect
to the research, authorship, and/or publication of this article.

Funding
The authors disclosed receipt of the following financial support for
the research, authorship, and/or publication of this article: This
study was supported by UNICEF.

References
Ahmed, M. F., Ahuja, S., Alauddin, M., Hug, S. J., Lloyd, J. R.,
Pfaff, A., . . . van Geen, A. (2006). Epidemiology. Ensuring
safe drinking water in Bangladesh. Science, 314, 1687-1688.
doi:10.1126/science.1133146
Ajzen, I. (1985). From intentions to actions: A theory of planned
behavior. In J. Kuhl & J. Beckmann (Eds.), Action control:
From cognition to behavior (SSSP Series; pp. 11-39). Munich,
Germany: Springer.
Argos, M., Kalra, T., Pierce, B. L., Chen, Y., Parvez, F., Islam, T.,
. . . Ahsan, H. (2011). A prospective study of arsenic exposure from drinking water and incidence of skin lesions in
Bangladesh. American Journal of Epidemiology, 174, 185-194.
doi:10.1093/aje/kwr062
Argos, M., Kalra, T., Rathouz, P. J., Chen, Y., Pierce, B., Parvez, F.,
. . . Ahsan, H. (2010). Arsenic exposure from drinking water,
and all-cause and chronic-disease mortalities in Bangladesh
(HEALS): A prospective cohort study. Lancet, 376, 252-258.
doi:10.1016/s0140-6736(10)60481-3
Aziz, S. N., Boyle, K. J., & Rahman, M. (2006). Knowledge of arsenic
in drinking-water: Risks and avoidance in Matlab, Bangladesh.
Journal of Health, Population and Nutrition, 24, 327-335.
Bandura, A. (1997). Self-efficacy: The exercise of control. New
York, NY: Freeman.
Chen, Y., van Geen, A., Graziano, J. H., Pfaff, A., Madajewicz, M.,
Parvez, F., . . . Ahsan, H. (2007). Reduction in urinary arsenic levels in response to arsenic mitigation efforts in Araihazar,
Bangladesh. Environmental Health Perspectives, 115, 917-923.
doi:10.1289/ehp.9833
Cialdini, R. B., Demaine, L. J., Sagarin, B. J., Barrett, D. W.,
Rhoads, K., & Winter, P. L. (2006). Managing social norms for
persuasive impact. Social Influence, 1(1), 3-15.
Dombrowski, S. U., Sniehotta, F. F., Avenell, A., Johnston, M.,
MacLennan, G., & Arajo-Soares, V. (2012). Identifying
active ingredients in complex behavioural interventions for
obese adults with obesity-related co-morbidities or additional
risk factors for co-morbidities: A systematic review. Health
Psychology Review, 6(1), 7-32.
Fishbein, M., & Ajzen, I. (2010). Predicting and changing
behavior: The reasoned action approach. New York, NY:
Psychology Press.
Flanagan, S. V., Johnston, R. B., & Zheng, Y. (2012). Arsenic in
tube well water in Bangladesh: Health and economic impacts
and implications for arsenic mitigation. Bulletin of the World
Health Organization, 90, 839-846.
Floyd, D. L., Prentice-Dunn, S., & Rogers, R. W. (2000). A metaanalysis of research on protection motivation theory. Journal
of Applied and Social Psychology, 30, 407-429.
George, C. M., Geen, A. V., Slavkovich, V. N., Singha, A., Levy, D.,
Islam, T., . . . Liu, X. (2012). A cluster-based randomized controlled
trial promoting community participation in arsenic mitigation efforts
in Singair, Bangladesh. Environmental Health, 11(1), 41.
George, C. M., Inauen, J., Rahman, S. M., & Zheng, Y. (2013). The
effectiveness of educational interventions to enhance the adoption

Downloaded from heb.sagepub.com by guest on May 27, 2016

11

George et al.
of fee-based arsenic testing in Bangladesh: A cluster randomized controlled trial. American Journal of Tropical Medicine and
Hygiene, 89, 138-144. doi:10.4269/ajtmh.12-0664
George, C. M., Zheng, Y., Graziano, J., Hossain, Z., Rasul, S. B.,
Mey, J., & van Geen, A. (2012). Evaluation of an arsenic test kit
for rapid well screening in Bangladesh. Environmental Science
& Technology, 46, 11213-11219. doi:10.1021/es300253p
Gollwitzer, P. M., & Sheeran, P. (2006). Implementation intentions and goal achievement: A meta-analysis of effects and
processes. Advances in Experimental Social Psychology, 38,
249-268.
Hoque, B. A., Hoque, M. M., Ahmed, T., Islam, S., Azad, A. K.,
Ali, N., . . . Hossain, M. S. (2004). Demand-based water options
for arsenic mitigation: An experience from rural Bangladesh.
Public Health, 118(1), 70-77. Retrieved from http://www.scopus.com/inward/record.url?eid=2-s2.0-0346888607&partnerID
=40&md5=1bd9b1ebff89c9447f3e2f1bc53c3a96
Hossain, M. A., Sengupta, M. K., Ahamed, S., Rahman, M.
M., Mondal, D., Lodh, D., . . . Chakraborti, D. (2005).
Ineffectiveness and poor reliability of arsenic removal plants in
West Bengal, India. Environmental Science & Technology, 39,
4300-4306. Retrieved from http://www.scopus.com/inward/
record.url?eid=2-s2.0-20044387956&partnerID=40&md5=
a8737a0d41e8bdb69b9964991abb0705
Inauen, J., Hossain, M. M., Johnston, R. B., & Mosler, H. J. (2013).
Acceptance and use of eight arsenic-safe drinking water
options in Bangladesh. PLoS One, 8(1), e53640. doi:10.1371/
journal.pone.0053640
Inauen, J., & Mosler, H.-J. (2013). Developing and testing theorybased and evidence-based interventions to promote switching to arsenic-safe wells in Bangladesh. Journal of Health
Psychology, 19, 1483-1498. doi:10.1177/1359105313493811
Inauen, J., & Mosler, H. J. (2015). Mechanisms of behavioural
maintenance: Long-term effects of theory-based interventions
to promote safe water consumption. Psychology & Health, 31,
166-183.
Inauen, J., Tobias, R., & Mosler, H. J. (2013). Predicting water
consumption habits for seven arsenic-safe water options in
Bangladesh. BMC Public Health, 13, 417. doi:10.1186/14712458-13-417
Inauen, J., Tobias, R., & Mosler, H. J. (2014). The role of commitment strength in enhancing safe water consumption: mediation
analysis of a cluster-randomized trial. British Journal of Health
Psychology, 19, 701-719. doi:10.1111/bjhp.12068
Michie, S., & Johnston, M. (2012). Theories and techniques of
behaviour change: Developing a cumulative science of behaviour change. Health Psychology Review, 6(1), 1-6.
Michie, S., Johnston, M., Francis, J., Hardeman, W., & Eccles, M.
(2008). From theory to intervention: Mapping theoretically
derived behavioural determinants to behaviour change techniques. Applied Psychology, 57, 660-680.
Mosler, H.-J. (2012). A systematic approach to behavior change interventions for the water and sanitation sector in developing countries: A conceptual model, a review, and a guideline. International
Journal of Environmental Health Research, 22, 431-449.
Mosler, H. J., Blchliger, O. R., & Inauen, J. (2010). Personal,
social, and situational factors influencing the consumption of

drinking water from arsenic-safe deep tubewells in Bangladesh.


Journal of Environmental Management, 91, 1316-1323.
Opar, A., Pfaff, A., Seddique, A. A., Ahmed, K. M., Graziano, J.
H., & van Geen, A. (2007). Responses of 6500 households to
arsenic mitigation in Araihazar, Bangladesh. Health Place,
13(1), 164-172. doi:10.1016/j.healthplace.2005.11.004
Orbell, S., Lidierth, C. J., Geeraert, N., Uller, C., Uskul, A. K., &
Kyriakaki, M. (2009). Social-cognitive beliefs, alcohol, and
tobacco use: A prospective community study of change following a ban on smoking in public places. Health Psychology,
28, 753-661.
Radloff, K. A., Zheng, Y., Michael, H. A., Stute, M., Bostick, B.
C., Mihajlov, I., . . . van Geen, A. (2011). Arsenic migration
to deep groundwater in Bangladesh influenced by adsorption
and water demand. Nature Geoscience, 4, 793-798. doi:http://
www.nature.com/ngeo/journal/v4/n11/abs/ngeo1283.
html#supplementary-information
Rogers, R. W., & Prentice-Dunn, S. (1997). Protection motivation
theory. In D. S. Gochman (Ed.), Handbook of health behavior
research I: Personal and social determinants (pp. 113-132).
New York, NY: Plenum Press.
Schultz, P. W., Nolan, J. M., Cialdini, R. B., Goldstein, N. J., &
Griskevicius, V. (2007). The constructive, destructive, and
reconstructive power of social norms. Psychological Science,
18, 429-434.
Schwartz, S. H. (1977). Normative influence on altruism. In L.
Berkowitz (Ed.), Advances in experimental social psychology
(Vol. 10, pp. 221-279). New York, NY: Academic Press.
Schwarzer, R. (2008). Modeling health behavior change: How to
predict and modify the adoption and maintenance of health
behaviors. Applied Psychology, 57, 1-29.
Smith, A. H., Hopenhayn-Rich, C., Bates, M. N., Goeden, H. M.,
Hertz-Picciotto, I., Duggan, H. M., . . . Smith, M. T. (1992).
Cancer risks from arsenic in drinking water. Environmental
Health Perspectives, 97, 259-267.
Smith, A. H., Yunus, M., Khan, A. F., Ercumen, A., Yuan, Y., Smith,
M. H., . . . Steinmaus, C. (2013). Chronic respiratory symptoms
in children following in utero and early life exposure to arsenic in drinking water in Bangladesh. International Journal of
Epidemiology, 42, 1077-1086. doi:10.1093/ije/dyt120
Tobias, R. (2009). Changing behavior by memory aids: A social
psychological model of prospective memory and habit development tested with dynamic field data. Psychological Review,
116, 408-438.
Trafimow, D., & Sheeran, P. (1998). Some tests of the distinction between cognitive and affective beliefs. Journal of
Experimental Social Psychology, 34, 378-397.
UNICEF. (2009). Bangladesh National Drinking Water Quality
Survey of 2009. Dhaka, Bangladesh: Bangladesh Bureau of
Statistics.
Van Geen, A., Ahsan, H., Horneman, A. H., Dhar, R. K., Zheng,
Y., Hussain, I., . . . Graziano, J. H. (2002). Promotion of wellswitching to mitigate the current arsenic crisis in Bangladesh.
Bulletin of the World Health Organization, 80, 732-737.
Retrieved
from
http://www.scopus.com/inward/record.
url?eid=2-s2.0-0036377841&partnerID=40&md5=95b24e6af
675d9ab151a080d5c5a3d7b

Downloaded from heb.sagepub.com by guest on May 27, 2016