Psychology
2011. Vol.2, No.8, 804-810
Copyright © 2011 SciRes. DOI:10.4236/psych.2011.28123
The Influence of Personality and Health Beliefs on Maintaining
Proper Hydration
Stephen M. Patterson, Deborah E. Spinks
Department of Psychology, Ohio University, Athens, Ohio, USA.
Email: patters@ohiou.edu
Received January 26th, 2011; revised August 14th, 2011; accepted September 26th, 2011.
The present study was designed to examine factors that could facilitate or impede adherence to proper hydration.
Forty volunteers (20 male, 20 female) were randomly assigned to one of two groups: Informed Group (n = 20)
and Uninformed Group (n = 20). Bioelectrical impedance was used to measure intracellular (ICW) and extracel-
lular (ECW) body water at Time 1 and 2. Personality, health beliefs, and health behaviors inventories were ad-
ministered at Time 1. A health information brochure on proper hydration and consequences of poor hydration
was given to the Informed Group. All participants were given six 1-liter bottles of water and drank two bottles
per day. Both ICW, F(1,38) = 4.79, p < .05, and ECW, F(1,38) = 10.12, p < .005, significantly increased for
both groups, and females had significantly greater changes than males in ECW, F(1,38) = 4.43, p < .05, and
ICW, F(1,38) = 4.48, p < .05. Health information had no significant effect on female adherence but was a sig-
nificant predictor of male adherence,
= .266, p < .05. Agreeableness, r = .36, p < .05, and social desirability, r
= .33, p < .05, were the only personality factors related to change in ECW for the group as a whole. Health be-
liefs were unrelated to adherence, but general health concern,
= –.053, p < .05, was a significant predictor of
change in ECW for males, although it was an inverse relationship.
Keywords: Hydration, Adherence, Health Beliefs, Health Behaviors, Personality
Introduction
Studies of adherence to positive health behaviors tend to re-
port a broad range of adherence rates, but they all tend to dem-
onstrate that adherence to health enhancement regimens is dif-
ficult to promote as well as assess and continues to be an issue,
especially among “healthy” individuals. Fontaine and Shaw
(1995) assessed attendance at an 8-week aerobics program and
found that 51% of the participants could be labeled as non-
adherent dropouts. Lynch et al. (1992) examined adherence to
an exercise intervention for those with high cholesterol and
found that only 39% of the participants attended 80% or more
of the scheduled sessions. Lynch et al. (2000) measured adher-
ence to a cholesterol-reducing diet in this same sample and
found that 25% of the participants were either “vacillators” or
“dropouts”. In response the lack of adherence to positive health
behaviors, many health professionals have begun to employ
educational strategies for improving adherence (Rapoff, 1998).
Previous research has found health education and psychoeduca-
tion to have significant effects on adherence to health behaviors
associated with diseases such as rheumatoid arthritis, hyperten-
sion, and diabetes (Devine & Reifschneider, 1995; Hammond,
Lincoln, & Sutcliffe, 1999; Ena et al., 2009). Unfortunately,
one of the most difficult obstacles facing health promotion
regimens is the impact of personality and health beliefs on ad-
herence, especially when the benefits or consequences are not
as immediately evident.
The influence of personality traits on adherence to medical
and health regimens has been suggested in the health research
over the last two decades (Leventhal, 1993). Previous research
has indicated that several major personality dimensions (e.g.,
extraversion, psychoticism, and optimism) may be linked to
general health behaviors, such as smoking, drinking, and nutri-
tion. In the last decade, application of the five-factor model of
personality to adherence has been advocated in order to reduce
inconsistencies in this body of literature (Wiebe & Christensen,
1996). These five broad personality dimensions, as assessed
with the NEO-Five Factor Inventory (Costa & McCrae, 1992),
include: neuroticism, extraversion, openness to experience,
agreeableness, and conscientiousness. Within the five-factor
model of personality, the conscientiousness dimension of per-
sonality has been most consistently associated with adherence
and health behaviors. In a two-part study by Booth-Kewley and
Vickers (1994) examined personality and health behaviors in
male U. S. Navy enlisted personnel. Results indicated that all
five personality dimensions were significantly related to at least
one health behavior dimension (i.e., wellness behaviors, acci-
dent control, traffic risk taking, and substance risk taking).
Most results were replicated in the second study. Steptoe et al.
(1994) assessed the relationship between personality variables
and a healthy lifestyle in a large sample of university students
and found that healthier practices were positively associated
with optimism and extraversion but negatively correlated with
neuroticism and psychoticism.
Another factor that is believed to play a significant role in
health regimen adherence is the health belief model, which is
probably the most discussed motivational construct in the
health literature. The key components to the current health be-
havior model are: 1) perceived susceptibility, 2) perceived se-
verity, 3) perceived benefits, 4) perceived barriers, 5) cues to
action, and 6) self-efficacy. However, the self-efficacy compo-
nent seems to be less incorporated into the health belief model
research. Motivation for health action is hypothesized to be a
product of perceived susceptibility and the severity of a health
threat. Individual responses are a function of the costs and
benefits of each of the actions available for dealing with the
perceived threat (Leventhal, 1993). However, previous research
has provided inconsistent evidence regarding the influence of
the health belief model components on adherence to medical
regimens among individuals with insulin-dependent diabetes
S. M. PATTERSON ET AL.
805
mellitus (Bond, Aiken, & Somerville, 1992; Ena et al., 2009) or
kidney disease (Ghaddar, Shamseddeen, & Elzein, 2009).
The role of the health belief model has rarely been examined
in relation to preventive health behaviors, but studies that have
assessed the influence of the health belief model in this area
appear to support it. Friedman et al. (1995) found that perceived
risk of skin cancer was positively related to intentions to prac-
tice skin self examination regularly, and that reasons for doing
skin self examination (perceived benefits) were positively re-
lated to intention to continue skin-cancer prevention behaviors.
Additionally, Sands, Archer, and Puleo (1998) found that per-
ceived barriers significantly predicted the risk for AIDS, alco-
hol abuse, and poor nutrition to adherence and perceived sever-
ity of the condition. The present study assessed the influence of
an adapted 4-construct health belief model on adherence to a
brief hydration regimen. The 4-construct health belief model
consisted of general health concern/threat, perceived suscepti-
bility, perceived severity, and perceived benefits, and the cue to
action was a brochure describing the benefits of proper hydra-
tion.
In addition to the potential impact that personality and health
beliefs may have on adherence to health promotion regimens,
another problem area within adherence research is the method
used to assess medical regimen adherence. There are numerous
methods to measure adherence, including self-reports, medica-
tion counting and weighing, and physiological measurements,
but all seem to have their shortcomings. However, self-reports
may still provide valuable information when used in combina-
tion with other adherence assessments, such as physiological
measurements (Steele, Jackson, & Gutmann, 1990). The pre-
sent study used both self-report of fluid intake and Bioelectrical
Impedance Assessment, which is a technique that measured
changes in hydration status and total body water (O’Brian,
Young, & Sawka, 2002), to assess adherence to the brief hydra-
tion regimen.
Therefore, the goals of the present study were to: 1) examine
whether health promotion information increased adherence to a
three-day hydration regimen, 2) assess the influence of person-
ality characteristics and health behaviors hydration adherence,
and 3) determine the effects of the health belief model (i.e.,
general health concern/threat, perceived susceptibility, per-
ceived severity, perceived benefits, and cue to action) on ad-
herence to hydration enhancement.
Methods
Participants
Complete data was obtained from 40 participants (20 males
and 20 females) selected from an undergraduate mass screening.
All participants were between 18 and 20 years of age (M =
18.76) and were currently enrolled in an undergraduate psy-
chology class. Requirements for participation in the study were
initially acquired with a mass-screening questionnaire. Potential
participants were telephoned and briefly interviewed with a
health information questionnaire to confirm their health status.
Requirements for participation included 1) being between 18
and 30 years of age, 2) being in good physical health as indi-
cated by the absence of a chronic or acute illness, 3) body
weight no greater than 20% above ideal weight as defined by
the Metropolitan Life Insurance Standards, and 4) report drink-
ing less than 1500 ml of water a day.
Questionnaires
Health Beliefs Questionnaire: The health belief questionnaire
is a 26-item psychosocial measure designed to assess an indi-
vidual’s perceptions about his or her health behavior. This
questionnaire was adapted from the Health Belief Model Ques-
tionnaire (Weissfeld, Kirscht, & Brock, 1990), a 32-item meas-
ure that addresses 6 health belief components: general health
threat, health concern, susceptibility, severity, and medical and
self- help benefits. The items specific to individuals with high
blood pressure were removed, resulting in 4 compressed cate-
gories (health threat/concern, perceived susceptibility, per-
ceived severity, and perceived benefits). Coefficient alpha reli-
abilities for the original questionnaire range from .65 to .89,
and confirmatory factor analysis identified the original six fac-
tors (Weissfeld et al., 1990). Coefficient alpha reliabilities for
the adapted 4-construct questionnaire ranged from .61 to .81.
NEO Five-Factor Inventory (NEO-FFI): The NEO-FFI (9)
(Costa, & McCrae, 1992) consists of 60 items that assess five
broad domains of personality: neuroticism, extraversion, open-
ness to experience, agreeableness, and conscientiousness. Cor-
relations with the full 180-item revised NEO Personality In-
ventory domains have ranged from .75 to .94 across normative
samples. Coefficient alphas have ranged from .68 to .90, and
convergent and discriminant validity is seen in the cross-ob-
server correlations with the NEO PI-R scales (Costa, & McCrae,
1992).
Marlowe-Crowne Social Desirability Scale. The Marlowe-
Crowne Social Desirability Scale (19) (Crowne & Marlowe,
1960) is one of the most widely used measures of need for ap-
proval and is comprised of 33 true-false items. Eighteen items
assess the tendency to attribute positive qualities toward oneself,
and the remaining 15 items assess the tendency to deny nega-
tive qualities in oneself. Construct validity has been established
and coefficient alpha reliabilities have ranged from .73 to .88 in
normative samples (Paulhus, 1991).
Food and Drink Inventory. This inventory was completed at
the beginning of Session 1 and Session 2. Participants were
instructed to write down everything consumed (i.e., food and
drinks) within the last 24 hours, being as specific as possible.
Due to the diuretic effects of alcohol, participants who reported
consuming more than 1100 ml of alcohol (approximately 3
beers) in the previous 24 hours were not included in the analy-
sis.
Health Behavior Questionnaire. This 6-item questionnaire
assessed 6 health behaviors: smoking, drinking, exercising,
unhealthy snacking, fruit consumption, and sleep habits.
Adherence Questionnaire. This brief self-report question-
naire contained 4 questions regarding how much of the 6 liters
was drank over the 3-day regimen; if they drank the instructed
2 liters per day; when they drank the water; and if they intended
to continue drinking more water than they used to after partici-
pating the study.
Physiological Measures
Bioelectrical Impedance Assessment (BIA): Multifrequency
BIA (BODYSTAT LTD Multiscan 5000 model: Douglas,
Isle of Man, UK) measures the distribution of extracellular
body water (ECW) and intracellular body water (ICW). BIA
utilizes the resistance and conductance of a weak electrical
current that is passed through the body between electrodes
placed on the right hand and the right foot. Since variation in
body fluids during the menstrual cycle affects BIA, females
were not scheduled for assessments during menses. Studies
using BIA with healthy subjects have indicated that this is a
valid (Berger, Rousset, MacCormack, & Ritz, 2000; Segal et al.,
S. M. PATTERSON ET AL.
806
1991) and reliable (19, 21) (Berger et al., 2000; Shanholtzer, &
Patterson, 2002) measure of total body water.
Health Promotion Information
During the first session, half of the participants received a
health promotion brochure briefly explaining the benefits of
proper hydration and the consequences of dehydration.
Hydration Fluid
Six 1-liter bottles of commercially available bottled water
were provided to each participant, and each bottle was labeled
with the day it was to be consumed.
Procedure
Initial qualification was assessed at a mass screening of in-
troductory psychology students. Students who estimated drink-
ing less than 1000 milliliters of water each day were contacted
by telephone in order to re-verify their daily liquid consumption
and asked some additional health information. Session 1 and
Session 2 appointments were then scheduled with the partici-
pant. All eligible participants were randomly assigned to one of
two groups: the Informed Group or the Uninformed Group.
Lastly, all participants were instructed to refrain from eating or
drinking in the hour prior to their appointment to control for
short-term dietary effects on body water.
Upon arrival to the psychophysiology laboratory, informed
consent was reviewed and obtained, and each participant was
randomly assigned to either the Hydration Benefits Informed
Group or the Uninformed Group. After being consented and
assigned to one of the two groups, the Food and Drink Inven-
tory, the Health Behavior Questionnaire, the Health Beliefs
Questionnaire, the NEO Five-Factor Inventory, and the Mar-
lowe-Crowne Social Desirability Scale were administered. Next,
height and weight were recorded for the BIA recording, and
participants were subsequently instructed to lie still on a mas-
sage table for a 15-minute rest period prior to the BIA re-
cording. The BIA recording took approximately 5 minutes to
complete.
Following the BIA recording, the participants assigned to the
Informed Group read through a health benefits brochure which
described the psychological and physical health benefits of
keeping well hydrated. Any questions the participants had
about the information in the brochure were answered. Partici-
pants in the Uninformed Group were not given this brochure,
and discussion was minimal. Subsequently, six 1-liter bottles of
water were provided to all participants with instructions to
drink 2 per day (the recommended daily amount) for the next 3
days as labeled on the bottles. Participants were also told they
would complete a final questionnaire and receive their BIA
feedback at the final session. In order to avoid influencing ad-
herence to the regimen, participants were not told that they
would undergo a second BIA recording.
The second session was scheduled for the fourth day at the
same time as the first session. At this session, each participant
completed a brief questionnaire on their adherence to the regi-
men and then had a second BIA recording to measure changes
in their body water. Lastly, the investigator reviewed the body
water readings with all participants and provided the health
benefits brochure to the participants in the Uninformed Group.
All participants were then debriefed and offered the chance to
ask any questions.
Data Reduction
Body water change scores suggested level of adherence to
the hydration regimen. A composite self-reported adherence
score was calculated using the first two questions on the adher-
ence questionnaire (i.e., how much of the 6 liters of water they
drank and if they drank the instructed amount per day) into a
self-report adherence scale. Adherers complied fully with the
regimen, whereas non-adherers were noncompliant with at least
one aspect of the regimen.
Data Analysis
Pearson product moment correlations were conducted to
evaluate the relationships among the variables. One-way
ANOVAs were performed to examine differences between
groups. Separate 2-between × 2-within repeated measures
ANOVAs were conducted to assess the effects of time, condi-
tion, and sex on the two measures of change in body water, and
a 2-between × 2-between × 2-within repeated measures
MANOVA was conducted to assess Sex by Condition effects.
Additionally, due to the sex differences found on the body wa-
ter measures, several MANCOVAs were performed, with sex
entered as a covariate. Similarly, separate stepwise multiple
regression analyses were performed for males and females, as
well as the sample as a whole. Sex (for the whole-group analy-
sis only) and BMI were forced into Step 1 due to their associa-
tion with body water measurements; the corresponding baseline
body water measure and condition were entered in Step 2; the
current health behaviors and health beliefs were entered in Step
3; and the personality measures were entered in Step 4.
Results
Descriptive Characteristics
An ANOVA indicated that the Informed Group and the Un-
informed Group did not differ on any demographic variables or
baseline body water measure. However, results revealed that
females had significantly lower baseline body water measure-
ments than males: ECW, F(1, 38) = 82.58, p < .0001, and ICW,
F(1, 38) = 178.37, p < .0001. Furthermore, females had sig-
nificantly greater increases on both measures of body water:
ECW, F(1, 38) = 4.43, p < .05, and ICW, F(1, 38) = 4.48, p
< .05, suggesting better adherence (see Table 1).
Adherence to the Hydration Regimen
Correlations revealed that BMI was significantly positively
correlated with baseline measures of ICW, r = .59, p < .001, and
ECW, r = .61, p < .001, but inversely correlated with change in
ECW, r = –.33, p < .05, indicating that those with a larger BMI
had higher baseline measures of body water and experienced
smaller changes in body water by Session 2. Of the participants’
self-reported estimates of liquids consumed daily, caffeinated
coffee and tea were significantly inversely correlated with
change in ECW, r = –.34, p < .05; and water was significantly
inversely correlated with change in ICW, r = –.37, p < .05.
Repeated measures ANOVAs revealed a significant main ef-
fect of Time for ICW, F(1, 38) = 4.79, p < .05, and ECW, F(1,
38) = 10.12, p < .005, indicating that both body water measures
were greater at Session 2. Change in ECW, however, was
greater than change in ICW, but this difference was not signifi-
cant. Additionally, a significant Time by Sex interaction was
found for ICW, F(1, 38) = 4.79, p < .05, indicating that ICW
increased from Session 1 to Session 2 for the females but not
the males.
Repeated measures ANOVAs revealed that Condition (In-
formed vs. Uninformed) had no significant effect on the body
S. M. PATTERSON ET AL.
807
water measurements at Session 1 and Session 2. However, fur-
ther analyses did reveal a significant Sex by Condition interac-
tion for change in ICW, F(1, 38) = 5.38, p < .05 (see Figure 1).
This interaction reflects a decrease in ICW for males in the
Uninformed condition and an increase for males in the In-
formed condition, whereas females in the Uninformed condi-
tion had greater increases in ICW than females in the Informed
condition. Stepwise multiple regression analyses were con-
ducted by sex in order to further examine this interaction (see
Table 2). For males, condition,
= .266, p < .05, was the only
significant predictor of change in ICW, but it accounted for
26.0% of the variance with BMI also in the model.
Table 3 displays the final significant regression equation for
the sample as a whole. Sex,
= .174, p = .069, and BMI,
=
–.023, p = .054, were marginally significant predictors of
change in ECW, whereas low general health concern,
= –.042,
p < .05, and high social desirability,
= .018, p < .05, were
significant predictors of increased ECW hydration. Together,
all four variables accounted for 37.5% of the variance in change
in ECW. Conversely, none of the variables significantly pre-
dicted change in ICW.
Correlations and an ANOVA revealed that there was no sig-
nificant relationship between self-reported non-adherers (n = 6)
and adherers (n = 34) on change in ECW or ICW. Paired-sam-
ples t-tests revealed that both groups had significant increases
in ECW, t (5) = –3.08, p < .05 and t (33) = –2.54, p < .05, re-
spectively, although the adherers also had a marginally signifi-
cant increase in ICW, t (33) = –1.82, p = .08.
Health Behaviors, Health Beliefs, and Adherence
Correlations were calculated to assess the relationship be-
tween general health behaviors and health beliefs and changes
in body water. Current health behaviors were not significantly
related to change in ECW or change in ICW. Of the health
beliefs, only general health concern was significantly inversely
related to change in ECW, r = –.33, p < .05, indicating that
those with greater health concern actually had smaller changes
in ECW. However, ANOVAs revealed that no health behavior
or health belief had a significant effect on change in body wa-
ter.
The final significant regression equations for the sample as a
Table 1.
Means and standard deviations for body mass and body water meas-
ures by sex and condition.
Males (n = 10) Females (n = 10) p
Uninformed Group (n = 20)
Body Mass Index (BMI)24.55(2.97) 21.94 (1.67) .026
Session 1
ECW 18.98(1.81) 14.41 (1.01) .000
ICW 22.45(2.49) 13.63 (1.24) .000
Session 2
ECW 19.00(1.89) 14.59 (.97) .000
ICW 22.32(2.56) 13.92 (1.15) .000
Change in
ECW .02 (.26) .26 (.33) ns
ICW –.13(.13) .30 (.48) .014
Informed Group (n = 20)
Body Mass Index (BMI)24.05(2.64) 23.36 (3.16) ns
Session 1
ECW 19.19(1.99) 14.89 (1.29) .000
ICW 22.79(2.58) 14.39 (1.67) .000
Session 2
ECW 19.34(1.89) 15.17 (1.10) .000
ICW 22.92(2.67) 14.52 (1.62) .000
Change in
ECW .15 (.31) .31 (.31) ns
ICW .13 (.31) .12 (.12) ns
Note: ICW = intracellular water; ECW = extracellular water; ns = not significant.
Figure 1.
Sex by condition interacti o n f o r change in intracellular body water (*p < .05).
S. M. PATTERSON ET AL.
808
Table 2.
Predictive value of baseline physiological measures, condition, health
behaviors/beliefs, and personality characteristics on changes in body
water by sex.
Variable
Beta t-value p
Females (n = 20)
ΔICW
BMI –.052 –.380 –1.86 ns
Neuroticism –.026 –.499 –2.45 .025
Males (n = 20)
ΔICW
BMI .012 .124 .59 ns
Condition .266 .506 2.42 .027
ΔECW
BMI –.016 –.149 –.72 ns
Health concern –.053 –.513 –2.49 .023
Note: ICW = intracellular water; ECW = extracellular water; Note: ICW = intra-
cellular water; ECW = extracellular water;
Table 3.
Predictive value of sex, baseline physiological measures, condition,
health behaviors/beliefs, and personality characteristics on changes in
body water (n = 40).
Variable
Beta t–value p
Δ ECW
Sex .174 .282 1.87 ns
BMI –.023–.204 1.99 ns
Health concern –.042–.324 –2.39 .022
Social desirability .018 .329 2.46 .019
Note: ECW = extracellular water; ns = not significant.
whole revealed that current health behaviors and health beliefs
were not significant predictors of change in ICW (see Table 3).
However, low general health concern,
= –.042, p < .05, was
again found to be a significant predictor of greater change in
ECW. Furthermore, additional stepwise multiple regression
analyses conducted by Sex revealed that low general health
concern,
= –.053, p < .05, was a significant predictor of
greater change in ECW for males only, accounting for 32.1% of
the variance with BMI also in the model (see Table 2)
Personality and Adherence
Change in ECW was significantly positively related to
agreeableness, r = .36, p < .05, and social desirability, r = .33, p
< .05, but no other personality measures were significantly
related to changes in body water. However, ANOVAs revealed
that only openness had a significant effect on change in ICW,
F(1, 39) = 2.42, p < .05, after controlling for sex, with those
scoring higher on openness showing smaller changes in ICW.
No other individual personality effects were found.
The regression analyses revealed that none of the personality
characteristics were significant predictors of change in ICW in
the sample as a whole (see Table 3). Furthermore, social desir-
ability,
= .018, p < .05, was the only personality variable that
was a significant predictor of change in ECW. Further regres-
sion analysis by sex revealed that none of the personality vari-
ables were significant predictors of change in ICW or ECW for
males, nor were they significant predictors of change in ECW
for females (see Table 2). Only neuroticism,
= –.026, p < .05,
was a significant predictor of change in ICW for females, al-
though BMI,
= –.052, p = .08, was marginally significant.
Together, BMI and neuroticism accounted for 32.0% of the
variance in change in ICW for females.
Discussion
This study was designed to examine the individual predictors
of adherence to a brief hydration regimen. The predictors ex-
amined in the present study included health education, health
behaviors and beliefs, and personality characteristics. First,
results indicated that individuals who drank more coffee and tea
had smaller increases in ECW, which may suggest possible
diuretic effects of caffeine on decreasing one’s hydration status.
Additionally, participants who reported generally drinking
more water (e.g., between 1000 and 1500 ml per day) may have
had higher ICW at Session 1 or inadvertently attenuated their
usual daily fluid consumption when given the additional 6 liters
and therefore the changes in hydration status for these partici-
pants may not have been as great as it would have had they
truly supplemented their diet with the additional regimen fluid.
In terms of the effects of education on hydration adherence
(Devine, & Reifschneider, 1995; Hammond et al., 1999), the
results of the present study do indicate that participants demon-
strated a small, yet significant increase in both ICW and ECW
from Session 1 to Session 2. However, a Sex by Time interac-
tion revealed that only females significantly increased overall.
Further analyses revealed a noteworthy Sex by Condition in-
teraction that indicated a decrease in ICW for males in the un-
informed group, but a significant increase in ICW for males in
the informed group. Conversely, the females in the both groups
had an increase in ICW, although, a greater increase was found
in the uninformed group. Thus, the main goal of the present
study was only partially supported, since the effect of the health
information (cue to action) was only significant for the males.
This finding is of particular interest, since intracellular body
water is less sensitive to acute body water changes, in com-
parison to extracellular body water.
The small increases in body water, the gender differences in
changes in body water, and the effect of the health information
are of particular interest in this study, and a number of reasons
may help explain some of these findings. First, the sample con-
sisted of primarily healthy college freshmen, a very restricted
group. Young people of this age have been found to be less
likely to practice health-promoting behaviors than older adults
(Walker, Volkan, Sechrist, & Pender, 1988). Secondly, college
students are also known to be a more unrealistically optimistic
population, which has been related to poorer health behaviors,
such as decreased exercise and poorer health prevention
knowledge (e.g., Davidson, & Prkachin, 1997). Lastly, the
gender effects of patient health information on adherence to the
hydration regimen may warrant further investigation, since such
effects have not been reported in previous health literature of
this kind.
Self-reports of adherence were not associated with any of the
changes in body water. The self-reported adherers and non-
adherers both significantly increased their ECW, but only the
full adherers demonstrated a marginally significant increase in
ICW as well, which suggests better adherence. The lack of a
relationship between self-reported adherence and physiological
measures was expected, since it has been a frequent finding in
the health literature (Rand, & Weeks, 1998).
Personalit y, Health Beha vi ors, and Adher e nc e
The present study examined the relationship between per-
sonality characteristics and adherence. Since previous literature
S. M. PATTERSON ET AL.
809
has found conscientiousness (Christensen, & Smith, 1995;
Lemos-Giraldez, & Fidalgo-Aliste, 1997) to be a common pre-
dictor of adherence and preventive health behaviors, the fourth
hypothesis predicted that conscientiousness would also be
strongly related to better adherence to the hydration regimen.
However, conscientiousness was not related to any body water
changes, and only agreeableness was related to greater changes
in ECW. However, agreeableness was positively related to
better adherence, which corresponds to previous studies (e.g.,
Lemos-Giraldez & Fidalgo-Aliste, 1997).
In previous literature, the other four dimensions of personal-
ity have been variably related to a number of health behaviors.
In the present study, neuroticism was found to be a significant
predictor of change in ICW for females only. Individuals who
score high in neuroticism are generally more susceptible to
psychological distress, which can interfere with impulse control
and adaptive coping. Thus, the females in the present study may
have had more negative feelings about some aspect of the study,
which their adherence. These results partially correspond to
previous studies that found neuroticism to be associated with
fewer wellness behaviors (Booth-Kewley, & Vickers, 1994).
Social desirability, on the other hand, was the only personality
variable found to significantly predict change in ECW for the
sample as a whole. Since increases in ECW register before
changes in ICW, it can be suggested that participants scoring
high in social desirability may have consumed the water closer
to their second session in order to comply with the researcher’s
instructions and, thus, present themselves in a favorable light.
Though the literature on social desirability and adherence is
scarce, the findings were in the expected direction.
In terms of examining the effects of current health behaviors
on health regimen adherence, general health behaviors, as as-
sessed in the present study, were not found to be correlated with
any of the BIA body water measures. This finding was in con-
trast to that of Friedman et al. (1995) who examined skin-can-
cer screening behaviors, but since more comprehensive studies
in this area are lacking, conclusions are provisional.
Health Beliefs, and Adherence
The third goal of the study was to assess whether positive
health beliefs would be related to better adherence. First, the
health information (cue to action), analyzed separately as Con-
dition, was a significant predictor of only the change in ICW
for the males. Analyses of the adapted 4-construct model indi-
cated that health concern (which includes perceived threat) was
significantly inversely related to and a significant predictor of
change in ECW, which suggests that those with greater health
concerns were less likely to adhere to this regimen. Thus, the
third hypothesis is not supported. These findings are compara-
ble to those of Bond et al. (1992), but Wiebe and Christensen
(1997) found that high scores in health beliefs (severity, sus-
ceptibility, and benefits) were associated with better adherence.
Such inconsistencies suggest that further studies in this area are
necessary.
Limitations of the Present Study
Several limitations of the present study should be noted. First,
participants were all young, healthy college students, so the
findings may not generalize to older individuals or less healthy
populations. Second, the sex of the experimenter may have had
some influence on adherence, as suggested by the sex differ-
ences in body water changes. Counterbalancing the sex of the
experimenter and participants may help minimize experimenter
influence. Third, the participants received class credit for par-
ticipating, which may have positively influenced adherence.
Fourth, the BIA measurements alone may not have been the
most accurate way to estimate adherence to the hydration regi-
men. Additional factors, such as solid foods eaten and liquid
voided were not taken into consideration in the analyses. Such
individual differences may have affected the body water meas-
urements.
The Health Behaviors Questionnaire used in the present
study was adapted from Lemos-Giraldez and Fidalgo-Aliste
(1997) and may not have been sufficiently comprehensive, as
noted by the original authors. Furthermore, the Health Beliefs
Questionnaire that was adapted from Weissfeld et al. (1990)
may not have been adequate, since most previous measures
have been illness-specific (e.g., Weissfeld et al., 1990; Given,
Given, Gallin, & Condon, 1983). Consequently, the need for a
more reliable and valid measure of these constructs remains.
Conclusion and Directions for Future Research
In summary, the results of the present study suggest that
young adults were not substantially influenced by the health
promotion information they received. In addition, some health
beliefs and personality components were significant predictors
of adherence to the brief hydration regimen, though some of
these findings were not consistent with previous literature.
These incongruities may be due to the inadequacies of existing
measures or to the specific hydration focus of the present study.
Because of the novelty of this study, further studies are being
conducted to better understand the pattern and scale of the BIA
changes associated with changes in hydration status. Future
studies utilizing hydration enhancement regimens should also
assess or control for solid food contributions to body water and
amount of liquid voided. Nonetheless, it is still readily apparent
from the present study, as well as previous studies, that the
numerous and complex predictors of adherence makes the
search for the most effective health promotion approach all the
more complicated.
References
Becker, M. H., Drachman, R. H., & Kirscht, J. P. (1974). A field ex-
periment to evaluate various outcomes of continuity of physician
care. American Journal of Public Health, 64 , 1062-1070.
doi:10.2105/AJPH.64.11.1062
Berger, V. A., Rousset, P., MacCormack, C., & Ritz, P. (2000). Repro-
ducibility of body composition and body water spaces measurements
in healthy elderly individuals. Journal of Nutrition, Health & Aging,
4, 243-245.
Bond, G. G., Aiken, L. S., & Somerville, S. C. (1992). The health belief
model and adolescents with insulin-dependent diabetes mellitus.
Health Psychology, 11, 190-198. doi:10.1037/0278-6133.11.3.190
Booth-Kewley, S., & Vickers, R. R. (1994). Associations between
major domains of personality and health behavior. Journal of Per-
sonality, 62, 281-298. doi:10.1111/j.1467-6494.1994.tb00298.x
Christensen, A. J., & Smith, T. W. (1995). Personality and patient ad-
herence: Correlates of the five-factor model in renal dialysis. Journal
of Behavioral Medicine, 18 , 305-313. doi:10.1007/BF01857875
Costa, P. T., & McCrae, R. R. (1992). Revised NEO Personality In-
ventory and NEO Five-Factor Inventory: Professional Manual.
Odessa, FL: Psychological Assessment Resources.
Crowne, D. P., & Marlowe, D. (1960). A new scale of social desirabil-
ity independent of psychopathology. Journal of Consulting Psychol-
ogy, 24, 349-354. doi:10.1037/h0047358
Davidson, K., & Prkachin, K. (1997). Optimism and unrealistic opti-
mism have an interacting impact on health-promoting behavior and
knowledge changes. Personality & Social Psychology Bulletin, 23,
617-625. doi:10.1177/0146167297236005
Devine, E. C., & Reifschneider, E. (1995). A meta-analysis of the ef-
S. M. PATTERSON ET AL.
810
fects of psychoeducational care in adults with hypertension. Nursing
Research, 44, 237-245. doi:10.1097/00006199-199507000-00009
Ena, J., Casan, R., Lozano, T. Leach, A., Algado, J. T., & Navarro_Diaz,
F. J., (2009). Long-term improvements in insulin prescribing habits
and glycaemic control in medical inpatients associated with the
introduction of a standardized educational approach. Diabetes
Research and Clinical Practice, 85, 159-165.
doi:10.1016/j.diabres.2009.05.011
Fontaine, K. R., & Shaw, D. F. (1995). Effects of self-efficacy and
dispositional optimism on adherence to step aerobic exercise classes.
Perceptual & Motor Skills, 81, 251-255.
doi:10.2466/pms.1995.81.1.251
Friedman, L. C., Weinberg, A. D., Webb, J. A., Cooper, H. P., & Bruce,
S., (1995). Skin cancer prevention and early detection intentions and
behavior. American Journal of Preventive Medicine, 11, 59-65.
Ghaddar, S., Shamseddeen, W., & Elzein, H., (2009). Behavioral
modeling to guide adherence to fluid control in hemodialysis patients.
Journal of Renal Nutrition, 19, 153-160.
doi:10.1053/j.jrn.2008.08.011
Given, C. W., Given, B. A., Gallin, R. S., & Condon, J. W. (1983).
Development of scales to measure beliefs of diabetic patients. Re-
search in Nursing and Health, 6, 127-141.
doi:10.1002/nur.4770060306
Hammond, A., Lincoln, N., & Sutcliffe, L. (1999). A crossover trial
evaluating an educational-behavioural joint protection programme
for people with rheumatoid arthritis. Patient Education & Counseling,
37, 19-32. doi:10.1016/S0738-3991(98)00093-7
Lemos-Giraldez, S., & Fidalgo-Aliste, A. M. (1997). Personality dispo-
sitions and health-related habits and attitudes: A cross-sectional
study. European Journal of Personality, 11, 197-209.
doi:10.1002/(SICI)1099-0984(199709)11:3<197::AID-PER283>3.0.
CO;2-H
Leventhal, H. (1993). Theories of compliance, and turning necessities
into preferences: Application to adolescent health action. In N. A.
Krasnegor, L. Epstein, S. B. Johnson, & S. J. Yaffe (Eds.), Develop-
mental aspects of health compliance behavior (pp. 91-124). Hillsdale,
NJ: Lawrence Erlbaum Associates.
Lynch, D., Birk, T., Weaver, M., Gohara, A., Leighton, R., Repka, F.,
& Walsh, M. (1992). Adherence to exercise interventions in the
treatment of hypercholesterolemia. Journal of Behavioral Medicine,
15, 365-377. doi:10.1007/BF00844729
Lynch, D., Repka, F., Nagel, R., Birk, T., Gohara, A., Leighton, R.,
Walsh, M., & Weaver, M., (2000). Prediction of dietary adherence in
cholesterol reduction: Relative contribution of personality variables
and health attitudes. Psychology and Health, 15, 821-828.
doi:10.1080/08870440008405584
Mulkana, S. S., & Hailey, B. J. (2001). The role of optimism in health-
enhancing behavior. American Journal of Health Behavior, 25, 388-
395.
O’Brian, C., Young, A. J., Sawka M. N. (2001). Bioelectrical imped-
ance to estimate changes in hydration status. International Journal of
Sports Medicine, 23, 361-366. doi:10.1055/s-2002-33145
Paulhus, D. L. (1991). Measurement and control of response bias. In J.
P. Robinson, P. R. Shaver, & L. S. Wrightsman (Eds.), Measures of
personality and social psychological attitudes (pp. 17-59). San Diego,
CA: Academic Press.
Rand, C. S., & Weeks, K. (1998). Measuring adherence with medica-
tion regimens in clinical care and research. In S. A. Shumaker, E. B.
Schron, J. K. Ockene, & W. L. McBee (Eds.), The handbook of
health behavior change (Second edition) (pp. 114-135). New York:
Springer Publishing Co.
Rapoff, M. A. (1998). Adherence issues among adolescents with
chronic disease. In S. A. Shumaker, E. B. Schron, J. K. Ockene, & W.
L. McBee (Eds.), The handbook of health behavior change (Second
ed.) (pp. 377-408). New York: Springer Publishing Co.
Sands, T., Archer, J., & Puleo, S. (1998). Prevention of health-risk
behaviors in college students: Evaluating seven variables. Journal of
College Student Devel opment, 39, 331-342.
Segal, K. R., Burastero, S., Chun, A., Coronel, P., Pierson, R. N. Jr.,
Wang, J. (1991). Estimation of extracellular and total body water by
multiple-frequency bioelectrical-impedance measurement. American
Journal of Clinical Nutrition, 54, 26-29.
Shanholtzer. B. A., & Patterson, S. M. (2002). Fluid hydration status
assessment in behavioral medicine research: Seven-day test-retest re-
liability. Annals of Behavioral Medicine, 24 , 120.
Steele, D. J., Jackson, T. C., & Gutmann, M. C. (1990). Have you been
taking your pill? The adherence-monitoring sequence in the medical
interview. The Journal of Family Practice, 30, 294-299.
Steptoe, A., Wardle, J., Vinck, J., Tuomisto, M., Holte, A., & Wich-
strom, L. (1994). Personality and attitudinal correlates of healthy and
unhealthy lifestyles in young adults. Psychology and Health, 9, 331-
343. doi:10.1080/08870449408407492
Walker, S. N., Volkan, K., Sechrist, K. R., & Pender, N. J. (1988).
Health-promoting lifestyles of older adults: Comparisons with young
and middle-aged adults, correlated, and patterns. Advances in Nurs-
ing Science, 11, 76-90.
Weissfeld, J. L., Kirscht, J. P., & Brock, B. M. (1990). Health beliefs in
a population: The Michigan Blood Pressure Survey. Health Educa-
tion Quarterly, 17, 141-155.
doi:10.1177/109019819001700202
Wiebe, J. S., & Christensen, A. J. (1996). Patient adherence in chronic
illness: Personality and coping in context. Journal of Personality, 64,
815-835. doi:10.1111/j.1467-6494.1996.tb00945.x