Self-care Behaviors and Glycemic Control

777

Social Support, Self-efficacy,

and Outcome Expectations
Impact on Self-care Behaviors and
Glycemic Control in Caucasian and
African American Adults With
Type 2 Diabetes

Purpose
The purpose of this study was to examine the relationships
of psychosocial variables (social support, self-efficacy, Diane Orr Chlebowy, PhD, RN
and outcome expectations) to diabetes self-care behaviors
Bonnie J. Garvin, PhD, RN
and glycemic control in Caucasian and African American
adults with type 2 diabetes.
From the University of Kentucky College of Nursing,
Lexington (Dr Chlebowy), and The Ohio State
Methods University College of Nursing, Columbus (Dr Garvin).

Study participants were scheduled for outpatient visits at Correspondence to Diane Orr Chlebowy, PhD, RN,
1 of 3 clinical sites in the southeastern United States. All University of Kentucky College of Nursing, 315
91 participants completed 4 self-report measures: Social College of Nursing Building, Lexington, KY 40536-
Support Questionnaire (SSQ), Self-efficacy Questionnaire 0232 (dochle0@email.uky.edu).
(SEQ), Outcome Expectancy Questionnaire (OEQ), and
The Diabetes Activities Questionnaire (TDAQ) at the time Acknowledgment: Grant support was provided by the
of the clinic visit. Long-term glycemic control was assessed American Association of Diabetes Educators.
by glycosylated hemoglobin analyses at the time of the
clinic visit. Pearson product–moment correlations were
DOI: 10.1177/0145721706291760
used to determine whether significant relationships existed
between scores on the SSQ, SEQ, OEQ, and TDAQ and
glycosylated hemoglobin values. Two-sample t tests were
used to detect differences in scores on the self-report meas-
ures and glycosylated hemoglobin values between the
2 racial groups.

Chlebowy, Garvin
The Diabetes EDUCATOR
778
Results
In all participants, no significant relationships were found
between (1) social support and self-care behaviors and
(2) self-efficacy and self-care behaviors. Self-care behav-
iors were significantly, positively correlated with outcome
expectancy scores for the total group and for African
Americans. No significant relationships were found
between (1) social support and glycemic control, (2) self-
efficacy and glycemic control, and (3) outcome expecta-
tions and glycemic control. African Americans reported
less social support satisfaction than Caucasians did.
Conclusions
Psychosocial variables investigated in this study were not
related to health outcomes of type 2 diabetes. Caucasians
and African Americans were similar in these variables. It is
important to investigate the relationships between other
variables (eg, age, duration of diabetes, education) and self-
care behaviors and glycemic control. Although African
Americans experience higher rates of diabetes-related com-
plications than Caucasians do, this may possibly be due to
other factors (eg, heredity, financial barriers, inadequate
health care). Additional investigations to study the relation-
ships of these variables to diabetes control are warranted.
T
ype 2 diabetes is associated with relative
insulin deficiency and/or insulin resistance
rather than a total deficit.1 It is estimated that
90% to 95% of the 18.2 million individuals
in the United States with diabetes have type
2
2 diabetes. In the United States, type 2 diabetes is more
common among certain racial groups (African Americans,
Hispanic Latino Americans, American Indians) than other
racial groups.2 Complications related to diabetes include
blindness, amputation, kidney disease, heart disease, hyper-
tension, stroke, nervous system disease, and dental dis-
ease.2 African Americans experience higher rates of 4
complications: blindness, amputation, kidney disease, and
heart disease.3 In the geographic region of this study, the
age-adjusted death rate due to diabetes for African
Americans is more than twice the comparable rate for the
Caucasian population.4 Caucasians and African Americans
with type 2 diabetes continue to experience preventable
diabetes-related life-threatening complications despite
medical advances and health care availability.
Individuals with type 2 diabetes are presented with
the challenges of managing health behavior regimens to
prevent or minimize the diabetes-related complications.
These regimens include (1) following a prescribed diet
plan, (2) administering oral medications and/or insulin,
(3) engaging in daily exercise, and (4) performing blood
glucose monitoring. Health professionals must find ways
to help individuals adapt to health-behavior regimens to
improve their health and prevent life-threatening diabetes-
related complications.
Study Purpose
The purpose of this study was to examine the relation-
ships of psychosocial variables (social support, self-efficacy
[SE], and outcome expectations [OE]) to diabetes self-care
behaviors and glycemic control in Caucasian and African
American adults with type 2 diabetes. Theoretically, social
support from family and friends is expected to enhance
self-care behaviors and improve glycemic control.
Similarly, high levels of SE and OE (beliefs that diabetes-
related behaviors will lead to certain outcomes) are expected
to positively influence self-care behaviors and improve
glycemic control. To examine the relationships of social
support, SE, and OE to self-care behaviors and glycemic
control in Caucasian and African American participants
with type 2 diabetes, the following hypotheses were
tested:
1. Social support, SE, and OE are positively related to self-care
behaviors in the total group and in Caucasians and African
Americans considered separately.
2. Social support, SE, and OE are positively related to
glycemic control in the total group and in Caucasians and
African Americans considered separately.
3. Social support, SE, OE, self-care behaviors, and glycemic
control will differ by race of the participants.
Theoretical Framework
Social support, a multidimensional concept, may influ-
ence the health outcomes of individuals. The availability
and quality of social support may directly affect an indi-
vidual’s ability to adapt to changes associated with chronic
illnesses such as type 2 diabetes. Researchers suggest that
there is a positive relationship between social support
and health outcomes of individuals with type 2 diabetes.5-11
Volume 32, Number 5, September/October 2006

Social support may positively affect self-care behaviors
and glycemic control of individuals with type 2 diabetes.
The concept of SE, derived from Bandura’s social cog-
nitive theory, provides a link between self-perceptions and
individual actions.12 Bandura described SE as a cognitive
process involving judgment of one’s ability to perform spe-
cific behaviors required to produce certain outcomes.12,13
Two principle components of Bandura’s theory of SE are
outcome expectations and efficacy expectations.12 Bandura
defined an outcome expectation as “a person’s estimate
that a given behavior will lead to certain outcomes.”12 An
efficacy expectation was defined as “the conviction that
one can successfully execute the behavior required to pro-
duce the outcomes.”12 SE theory has served as a framework
to help investigators understand diabetes-related behaviors
and facilitate behavioral change.14-19
Research Design and Methods
Setting
A 2-group, comparative descriptive design was used in
this study.20 Three different sites in the southeastern United
States were used for data collection to maximize the num-
ber of participants for the study. At site A, all participants
attended 2 all-day sessions taught for 2 consecutive days by
a multidisciplinary health care team (a registered nurse,
clinical dietitian, physician, psychologist, and exercise
physiologist). At site B, participants were scheduled for out-
patient clinic visits. Participants were evaluated and treated
by a physician and a registered nurse and/or a clinical dieti-
tian. Participants were scheduled for outpatient educational
sessions at site C. Each participant was given the opportu-
nity to learn about diabetes management while dialoguing
with a registered nurse and/or a clinical dietitian.
Sample
The convenience sample consisted of 91 adult partici-
pants with type 2 diabetes registered for outpatient visits at
1 of 3 clinical agencies. The inclusion criteria for partici-
pant selection were (1) diagnosed with type 2 diabetes, (2)
older than 18 years, and (3) registered for an outpatient
clinic visit or an educational session. Among the partici-
pants in the study, 64 (70%) were Caucasian and 27 (30%)
were African American. The sample was composed of 51
women and 40 men ranging in age from 19 to 83 years,
with a mean of 55 years (SD = 12.51). Most participants
were married (65%), and 36% were employed. The sample
Chlebowy, Garvin
Self-care Behaviors and Glycemic Control
779
was well educated, with more than 48.3% having some
college or technical school experience, although 17% had
only some high school education or less. For diabetes char-
acteristics, the duration of diabetes ranged from 4 weeks to
28 years, with a mean of 7 years (SD = 6.48). Most partic-
ipants (60%) reported taking oral agents for diabetes man-
agement, with 21% taking oral agents and insulin.
Glycosylated hemoglobin levels ranged from 5% to 16%,
with a mean of 8% (SD = 0.02).
Instrumentation
Four instruments were used for data collection in this
study: the Social Support Questionnaire (SSQ), the Self-
efficacy Questionnaire (SEQ), the Outcome Expectancy
Questionnaire (OEQ), and the Diabetes Activities
Questionnaire (TDAQ).
SSQ. Social support was measured by participants’
responses to the SSQ.21 This 27 item, 2-part instrument
measured the (1) number of social support individuals and
(2) participants’ satisfaction with these individuals. Each
item asked participants to (1) list the individuals to whom
they can rely on in specific situations and (2) indicate how
satisfied they were with these support systems. For exam-
ple, participants answered questions such as, “Whom do
you really count on to be dependable when you need help?”
and “Whom can you really count on to listen to you when
you need help?”
Alpha coefficients for number scores and satisfaction
scores were .97 and .94, respectively.21 The 4-week test-
retest reliabilities for number scores and satisfaction scores
were .90 and .83, respectively.21 In this study, participants
rated each SSQ item by indicating the number of support
persons (N) and their satisfaction with the available social
support (S; maximum = 162). Support satisfaction was
rated on a scale ranging from 1 (very dissatisfied) to 6 (very
satisfied). The sum of N or S scores was divided by 27
(number of items) to determine the overall N and S scores.
High N scores reflected high-perceived availability of
social support persons. High S scores reflected high satis-
faction with the available social support system.22
Cronbach α coefficients in this study were .97 for number
scores and .98 for satisfaction scores.
SEQ. SE was measured with the use of the SEQ.22 The
original SEQ was a 29-item instrument that measured par-
ticipants’ confidence in their abilities to perform a graded
The Diabetes EDUCATOR
780
series of regimen behaviors (eg, test blood sugar level at
least once per day, test blood sugar level at least twice per
day) in 4 areas: glucose testing (8 items), exercise (8 items),
eating habits (9 items), and medication taking (4 items).22
The SEQ was modified for use in this study. Items were
rewritten using simpler terminology to make completion
easier for participants. Also, the scaling of the instrument
was modified to simplify the scoring of the instrument.
Participants rated each SEQ item according to their per-
ceived ability to perform the behavior on a ranging scale
from 1 (strongly disagree) to 4 (strongly agree) to simplify
the scaling procedure. A subscale score was produced for
each of the 4 regimen areas: (1) glucose testing (SE sugar
testing), (2) exercise (SE exercise), (3) eating habits (SE
eating habits), and (4) medication taking (SE medication
taking). The total SE score (SE overall score) was a sum-
mation of the 4 subscale scores (maximum = 116). A high
subscale score reflected a high perceived ability to per-
form diabetes-related behaviors in a specific area.
Similarly, the total SE score reflected a high perceived
ability to perform diabetes-related behaviors. In this study,
Cronbach α coefficients for the SE subscales ranged from
.83 to .92. The α coefficient for the overall SE score in this
study was .92.
OEQ. Outcome expectations were measured with
the use of the OEQ.22 The original OEQ was a 20-item
instrument that measured individuals’ beliefs that certain
behaviors will lead to specific outcomes.22
The OEQ was modified for use in this study. Items were
rewritten in simpler terms to make completion easier for
participants. Items were placed into 5 areas: (1) OE sugar
testing (4 items), (2) OE exercise (4 items), (3) OE eating
habits (4 items), (4) OE medication taking (4 items), and
(5) OE self-care activities (4 items) to measure beliefs spe-
cific to each regimen area. Also, the scaling of the instru-
ment was modified to simplify the scoring of the
instrument. For each item, subjects chose a response rang-
ing from 1 (strongly disagree) to 4 (strongly agree) to sim-
plify the scaling procedure. Negatively expressed items
were reverse scored. A subscale score was determined for
each of the 5 regimen areas. The total OEQ score (OE
score) was a summation of the 5 subscale scores (maxi-
mum = 80). A high subscale score reflected strong beliefs
that diabetes-related behaviors pertaining to a regime area
will lead to specific outcomes. A high OEQ total score
reflected strong beliefs that diabetes-related behaviors will
lead to specific outcomes. The Cronbach α coefficient for
the OE score in this study was .80.
TDAQ. Adherence to the diabetes regimen was meas-
ured by participants’ responses to TDAQ.23 TDAQ is a 13-
item instrument that measures individuals’ adherence to the
diabetes regimen using 2 subscales: Lifestyle/Monitoring
(8 items) and Treatment (5 items).23 For example, partici-
pants responded to statements such as “I test my blood
sugar as often as suggested by my educator.”
TDAQ scoring was modified for use in this study.
Participants chose a response ranging from 1 (not appli-
cable) to 4 (always). A total TDAQ score (TDAQ score)
was a summation of the 13-item scores (maximum = 52).
A higher score reflected greater adherence to the diabetes
regimen. The Cronbach α coefficient for TDAQ score in
this study was .84.
Glycosylated hemoglobin assay testing. The glycosy-
lated hemoglobin assay (HbA1c) measured the percent-
age of total hemoglobin bound by glucose.24 Since red
blood cells have a life span of approximately 120 days,
the HbA1c was a reliable indicator of the average serum
glucose level over a 120-day period.1,24 The ideal range
for HbA1c was 4.0% to 6.7%.24 A high value indicated
poor glucose control for the past 120 days.
Procedure
The process for participant recruitment varied accord-
ing to the data collection site. At site A, after consent was
obtained, packets containing 4 questionnaires were mailed
to participants. Each participant was asked to return com-
pleted questionnaires in a self-addressed envelope prior to
the first class session. At site B, each participant was asked
to complete the questionnaires at the time of the outpatient
clinic visit. Similarly, at site C, each participant was asked
to complete the questionnaires at the time of the outpatient
educational session. Participants received a thank-you let-
ter and a $5 gift certificate or check in appreciation for their
time and effort.
The process for obtaining glycosylated hemoglobin
assay values varied according to the data collection site.
At sites A and C, values (within the past 4 months) were
obtained from the participants’ medical records. At site B,
assay values were obtained at the time of the participants’
clinic visits.
Volume 32, Number 5, September/October 2006

Results
Description of the Sample
Analyses were conducted to test for differences in dia-
betes characteristics between the Caucasian and African
American subgroups. Two-sample t tests revealed no sig-
nificant differences in age, duration of diabetes, and gly-
cosylated hemoglobin (see Table 1). The mean age for
Caucasians was 54.8 years (SD = 11.66), with a similar
mean of 55.4 years (SD = 14.57) for African Americans.
For diabetes characteristics, the mean duration of diabetes
for Caucasians was 7.3 years (SD = 6.41), with a mean of
6.6 years (SD = 6.75) for African Americans. The mean
glycosylated hemoglobin level for Caucasians was 8.2%
(SD = 0.02), with a similar mean of 8.1% (SD = 0.02) for
African Americans.
Chi-square analyses revealed 2 significant differences
in demographic characteristics between the 2 subgroups.
Seventy-four percent of the African American sample
was female, while only 48.4% of the Caucasians were
female (P = .04). Only 32% of the African American sam-
ple was married, in contrast to 70.3% of Caucasians who
were married (P = .0001). The 2 groups did not differ on
education, occupation, or medication management.
Relationships Between Variables
Relationship between social support and self-care
behaviors. A Pearson product–moment correlation was
computed between (1) the number of social support per-
sons (SSQ N score) and self-care behaviors and (2) the
satisfaction with social support persons (SSQ S score) and
self-care behaviors for all subjects, Caucasians, and
African Americans. When the group was considered in
total, analyses revealed no significant relationships. The
SSQ N and SSQ S scores were not significantly related to
self-care behaviors of the Caucasian or African American
subgroups.
Relationship between SE and self-care behaviors.
A Pearson product–moment correlation was computed
between (1) SE sugar testing and self-care behaviors,
(2) SE exercise and self-care behaviors, (3) SE eating
habits and self-care behaviors, (4) SE medication taking
and self-care behaviors, and (5) overall SE score and self-
care behaviors for all subjects and Caucasians and African
Americans separately. Analyses revealed no significant
Chlebowy, Garvin
Self-care Behaviors and Glycemic Control
781
relationships between SE and self-care behaviors when
the group was considered in total or by race.
Relationship between OE and self-care behaviors. A
Pearson product–moment correlation was computed
between OE scores and self-care behaviors for all sub-
jects, Caucasians, and African Americans. When the group
was considered in total, analyses revealed a significant
relationship between outcome expectations and self-care
behaviors. The OE score was significantly correlated with
self-care behaviors (r = 0.27, P = .01). For Caucasians,
analyses revealed no significant relationships between
outcome expectations and self-care behaviors (r = 0.24,
P = .06). However, there was a significant relationship
between outcome expectations and self-care behaviors for
African Americans (r = 0.43, P = .03).
Relationship between social support and glycemic
control. A Pearson product–moment correlation was com-
puted between (1) the number of social support persons
(SSQ N score) and glycemic control and (2) the satisfac-
tion with social support persons (SSQ S score) and
glycemic control for all subjects, Caucasians, and African
Americans. When the group was considered in total, analy-
ses revealed no significant relationships. No significant
relationships were found between the SSQ N and SSQ S
scores and glycemic control by race of the participants.
Relationship between SE and glycemic control. A
Pearson product–moment correlation was computed
between (1) SE sugar testing and glycemic control, (2) SE
exercise and glycemic control, (3) SE eating habits and
glycemic control, (4) SE medication taking and glycemic
control, and (5) overall SE score and glycemic control for
all subjects, Caucasians, and African Americans. When the
group was considered in total, no significant relationships
were found between SE and glycemic control. The same
held true when these relationships were considered by race
of the participants.
Relationship between OE and glycemic control. A
Pearson product–moment correlation was computed
between OE score and glycemic control. When the group
was considered in total, analyses revealed an insignifi-
cant relationship between outcome expectations and
glycemic control that also held true for Caucasians and
African Americans.
The Diabetes EDUCATOR

782

Table 1

Select Demographic and Diabetes Characteristics of All Participants, Caucasians, and African Americans

Participants

Characteristic All Participants (N = 91) Caucasians (n = 64) African Americans (n = 27)

Age, y
x– 54.96 54.75 55.44
SD 12.51 11.66 14.57
Gender, n (%)
Male 40 (44.0) 33 (51.6) 7 (25.9)
Female 51 (56.0) 31 (48.4) 20 (74.1)
Marital status, n (%)
Never married 11 (12.4) 5 (7.8) 6 (24.0)
Married 58 (65.2) 50 (78.1) 8 (32.0)
Separated/divorced 10 (11.2) 5 (7.8) 5 (20.0)
Widowed 10 (11.2) 4 (6.3) 6 (24.0)
Education, n (%)
Less than eighth grade 5 (5.8) 2 (3.3) 3 (11.1)
Some high school 10 (11.5) 6 (10.0) 4 (14.8)
High school graduate 30 (34.5) 19 (31.7) 11 (40.7)
Some college or technical school 18 (20.7) 12 (20.0) 6 (22.2)
College graduate 24 (27.6) 21 (35.0) 3 (11.1)
Occupation, n (%)
Employed 32 (36.0) 27 (42.9) 5 (19.2)
Homemaker 12 (13.5) 8 (12.7) 4 (15.4)
Disabled 14 (15.7) 9 (14.3) 5 (19.2)
Unemployed 12 (13.5) 7 (11.1) 5 (19.2)
Retired 19 (21.3) 12 (19.0) 7 (26.9)
Duration of diabetes, y
x– 7.08 7.26 6.65
SD 6.48 6.41 6.75
Medications, n (%)
None 7 (7.7) 5 (7.8) 2 (7.4)
Oral agents 55 (60.4) 40 (62.5) 15 (55.6)
Oral agents and insulin 19 (20.9) 12 (18.8) 7 (25.9)
Insulin 10 (11.0) 7 (10.9) 3 (11.1)
Glycosylated hemoglobin, %
x– 8.0 8.2 8.1
SD 0.02 0.02 0.02

Volume 32, Number 5, September/October 2006

 Self-care Behaviors and Glycemic Control

783

Table 2

Means and Standard Deviations of Psychosocial, Self-care, and Glycosylated Hemoglobin Variables of 91 Participants
With Type 2 Diabetes in the Southeastern United States Based on Race

Caucasians (n = 64) African Americans (n = 27)

Variable x– SD x– SD P

SSQ number 2.7 1.58 2.3 2.14 .40

SSQ satisfaction 5.4 0.87 4.6 1.86 .05*
SE sugar testing 27.8 4.87 26.0 6.16 .13
SE exercise 24.7 4.41 24.4 4.13 .78
SE eating habits 29.3 4.25 28.5 4.43 .41
SE medication taking 14.0 2.55 13.7 1.82 .57
SE overall score 96.0 11.94 92.2 12.36 .18
OE score 61.7 7.47 61.8 4.28 .92
Self-care behavior 37.8 9.16 36.6 9.46 .58
Glycosylated hemoglobin, % 8.2 0.02 8.1 0.02 .85

SSQ = Social Support Questionnaire; SE = self-efficacy; OE = outcome expectancy.

*P ≤ .05.

Differences in variables according to race. Two-sample
t tests revealed 1 significant difference in the social support
variables between the 2 races. No significant difference was
found by race on the SSQ number score. However, the SSQ
satisfaction score differed by race. Caucasians reported sig-
nificantly greater satisfaction with social support than
African Americans did (see Table 2). There were no signif-
icant differences by race in SE scores, OE scores, self-care
behaviors, and glycosylated hemoglobin (see Table 2).
Discussion
This investigation of factors that may contribute to dif-
ferences in diabetes outcomes has provided data to support
the need for further research. In this study, social support
was not significantly related to self-care behaviors for the
total group or for Caucasians and African Americans.
Existing studies suggest that social support may positively
influence the initiation and maintenance of diabetes self-
care behaviors.6-10,25-28 Differences in study findings may be
due to different operational definitions of social support
and different measurement scales for social support.
In this study, there were no significant relationships
between SE and self-care behaviors for the total group
or for Caucasians and African Americans considered
separately. Existing studies suggest that SE may posi-
tively influence diabetes self-care behaviors.11,14-16,19,29-32
Although SE was defined in this study in a similar way
to all existing studies, differences in findings may be due
to different measurement scales for SE.
In this study, analyses revealed a significant relationship
between OE and self-care behaviors for the total group and
African Americans. Specifically, the OE score was signifi-
cantly, positively correlated with self-care behaviors for
these 2 groups. Therefore, the greater participants’ beliefs
that overall diabetes management will lead to certain out-
comes, the greater participants’ adherence to their diabetes
regimens. This study’s finding is consistent with the find-
ings of existing studies.16,22,29 Outcome expectations were
defined in a similar way in all existing studies. In these
studies, all investigators used the OEQ.
In this study, social support was not significantly
related to glycemic control for the total group, Caucasians,
and African Americans. Previous studies are conflicting,

Chlebowy, Garvin
The Diabetes EDUCATOR
784
and some suggest that social support may be a predictor of
glycemic control in individuals with diabetes,5,9,33-35
whereas others suggest that social support is not a signifi-
cant predictor of improved glycosylated hemoglobin lev-
els.6,33,36-38 Differences in findings may be due to different
operational definitions of social support and different
measurement scales for social support.
Few studies have explored the relationship between
SE and glycemic control (as measured by glycosylated
hemoglobin assays). In this study, SE was not signifi-
cantly related to glycemic control for the total group or
for Caucasians and African Americans separately.
Similarly, Ludlow and Gein30 found no significant rela-
tionships between (1) exercise SE and glycosylated
hemoglobin levels and (2) medication SE and glycosy-
lated hemoglobin levels. Allison39 found no significant
relationship between SE and glycemic control. However,
Ludlow and Gein30 found that participants with higher
levels of general management SE and diet SE had lower
glycosylated hemoglobin levels.
The relationship between SE and glycemic control is
unclear. Mixed findings exist in the literature. SE was
defined in a similar way in all existing studies and the
present one. Differences in findings may be possibly due
to different measurement scales for SE.
Outcome expectations were not significantly related
to glycemic control in the total group or in the Caucasian
and African American subgroups. Previous studies have
not explored the relationship between these variables.
Therefore, this study yields new knowledge relevant to
the variables affecting diabetes control in Caucasians
and African Americans with type 2 diabetes.
The finding that African Americans in this study were
less satisfied with their social support systems than
Caucasians yields new knowledge that could have a large
impact on health care delivery. African Americans have
identified the need for help and support in managing psy-
chosocial issues related to diabetes.40 No prior studies
have documented racial differences using the SSQ.
There are few existing, published studies that have exam-
ined differences in study variables between the 2 racial
groups. Bailey and Lherisson-Cedano41 found no significant
differences with regard to social support. Bertera42 found
that African Americans scored higher than Caucasians and
Hispanics on social affiliation measures (eg, club meeting
attendance, number of telephone calls with family and
friends). African Americans place more emphasis on their
informal social networks for disease management than
Caucasians do,26 and Via and Salyer43 found no significant
differences with regard to SE. In this study, no significant
differences were found with regard to glycemic control.
Contrary to this finding, Connell et al35 and Bailey and
Lherisson-Cedano41 found that African Americans had sig-
nificantly higher glycosylated hemoglobin levels than
Caucasians did. Differences in findings may be due to the
heterogeneity of the racial groups in each study. In this study,
the 2 racial groups were homogeneous, with few significant
differences in demographic and diabetes characteristics.
Few studies have examined racial differences with
regard to diabetes self-care behaviors. African Americans
identified spirituality as an important factor in diabetes
self-care management.44 Schoenberg and Drungle45 found
that African Americans reported more financial, visual, and
pain barriers to self-care than Caucasians did. However,
Schoenberg and Drungle45 found that both racial groups
reported a reluctance to exercise and perform glucose test-
ing. Similarly, in this study, no significant differences were
revealed with regard to diabetes self-care behaviors.
Previous diabetes education experiences may have
affected the study’s variables. Study participants who had
previous exposure to diabetes education may (1) realize
the importance of diabetes management and perform
the necessary self-care behaviors for glycemic control and
(2) have higher levels of SE and outcome expectations with
a positive influence on self-care behaviors and glycemic
control.
Implications
The results of this study have implications for diabetes
educators and health care providers. One implication is
that outcome expectations affect individuals’ adherence to
their diabetes regimens. The greater individuals’ beliefs
that overall diabetes management will lead to certain out-
comes, the greater their adherence to diabetes regimens.
This evidence can be useful in planning professional inter-
ventions (eg, education, consultation). For example,
during an education session with a client and family mem-
bers, the diabetes educator may wish to emphasize the
short- and long-term benefits of diabetes management to
promote adherence to diabetes regimens. Or, during con-
sultation with the client, the health care provider may wish
to allow additional time for the expression of beliefs about
diabetes management. Perhaps these professional interven-
tions will enhance self-care management in Caucasians
and African Americans with type 2 diabetes.
When considering the study variables, one significant
difference was found between the racial groups. African
Volume 32, Number 5, September/October 2006

Americans reported less social support satisfaction than
Caucasians did. The availability of social support may
directly affect African Americans’ ability to adapt to
lifestyle changes associated with type 2 diabetes. This find-
ing can be helpful as health care providers care for African
Americans with type 2 diabetes. For example, additional
emphasis could be placed on how the family members
might be involved in planning meals and exercise activities.
Other than social support, no other significant differ-
ences were found between Caucasians and African
Americans. This evidence suggests that with homogeneous
groups (eg, similar age, duration of diabetes, education),
there are no significant differences between the racial
groups. Although African Americans experience higher
rates of diabetes-related complications than Caucasians do,
this difference may not be related to psychosocial vari-
ables. This difference may possibly be due to other factors
(eg, heredity, financial barriers, inadequate health care).
This finding yields new knowledge for diabetes educators
and health care providers and suggests areas for future
research.
The future use of the modifications in the instruments
is supported by this study. The findings were easily inter-
preted, and reliabilities obtained in the sample were
excellent. Future use of easily understood reliable instru-
ments will improve consistency in research findings.
Conclusions
Type 2 diabetes continues to be a major health and
financial concern in the United States. Caucasians and
African Americans with type 2 diabetes continue to expe-
rience preventable diabetes-related life-threatening com-
plications despite medical advances and health care
availability. African Americans are more likely to suffer
from diabetes-related complications and experience a
higher diabetes-related mortality rate than Caucasians do.
African Americans and Caucasians are similar in how the
psychosocial behaviors are related to diabetes outcomes.
Social support may be one area for focus to improve the
health of African Americans with type 2 diabetes.
References
1. Chlebowy DO, Wagner KD. Altered glucose metabolism. In:
Wagner KD, Johnson K eds. High Acuity Nursing. 4th ed. Upper
Saddle River, NJ: Prentice Hall; in press.
2. National Institute of Diabetes, Digestive, and Kidney Diseases.
National diabetes statistics. Available at: www.diabetes.niddk.nih
.gov/about/index.htm. Accessed July 25, 2005.
Chlebowy, Garvin
Self-care Behaviors and Glycemic Control
785
3. American Diabetes Association. Diabetes statistics for African
Americans. Available at: www.diabetes.org/diabetes-statistics/
african-americans.jsp. Accessed July 25, 2005.
4. Kentucky Cabinet for Health and Family Services. Diabetes in
Kentucky African Americans 2005. Available at: www.chfs.ky
.gov/dph/ach/diabetes.htm. Accessed February 17, 2006.
5. Eriksson BS, Rosenqvist U. Social support and glycemic control
in non-insulin dependent diabetes mellitus patients: gender dif-
ferences. Women Health. 1993;20:59-70.
6. Griffith LS, Field BJ, Lustman PJ. Life stress and social support
in diabetes: association with glycemic control. Int J Psychiatry
Med. 1990;20:365-372.
7. Wang CY, Fenske MM. Self-care of adults with non-insulin
dependent diabetes mellitus: influence of family and friends.
Diabetes Educ. 1996;22:465-470.
8. Sousa VD, Zauszniewski JA, Musil CM, et al. Testing a concep-
tual framework for diabetes self-care management. Res Theory
Nurs Pract. 2004;18:293-316.
9. Whittemore R, Melkus GD, Grey M. Metabolic control, self-
management and psychosocial adjustment in women with type 2
diabetes. J Clin Nurs. 2005;14:195-203.
10. Surit P. Health beliefs, social support, and self-care behaviors of
older Thai persons with non-insulin-dependent diabetes mellitus
(NIDDM) [dissertation]. Washington, DC: Catholic University of
America; 2002.
11. Wen LK, Shepherd MD, Parchman ML. Family support, diet, and
exercise among older Mexican Americans with type 2 diabetes.
Diabetes Educ. 2004;30:980-993.
12. Bandura A. Self-efficacy: toward a unifying theory of behavioral
change. Psychol Rev. 1977;84:191-215.
13. Bandura A. Self-efficacy mechanism in human agency. Am
Psychol. 1982;37:122-147.
14. Hurley CC, Shea CA. Self-efficacy: strategy for enhancing dia-
betes self-care. Diabetes Educ. 1992;18:146-150.
15. Kavanagh DJ, Gooley S, Wilson PH. Prediction of adherence and
control in diabetes. J Behav Med. 1993;16:509-522.
16. Skelly AH, Marshall JR, Haughey BP, et al. Self-efficacy and
confidence in outcomes as determinants of self-care practices in
inner-city, African-American women with non-insulin-dependent
diabetes. Diabetes Educ. 1995;21:38-46.
17. Miller CK, Edwards L, Kissling G, et al. Nutrition education
improves metabolic outcomes among older adults with diabetes
mellitus: results from a randomized control trial. Prev Med.
2002;34:252-259.
18. Miller CK, Edwards L, Kisssling G, et al. Evaluation of a theory-
based nutrition intervention for older adults with diabetes melli-
tus. J Am Diet Assoc. 2002;102:1069-1081.
19. Williams KE, Bond MJ. The roles of self-efficacy, outcome
expectancies and social support in the self-care behaviours of dia-
betics. Psychol Health Med. 2002;7:127-141.
20. Cook TD, Campbell DJ. Quasi-experimentation: Design and
Analysis Issues for Field Settings. Chicago, Ill: Rand McNally;
1963.
21. Sarason IG, Levine HM, Basham RB, et al. Assessing social sup-
port: the Social Support Questionnaire. J Pers Soc Psychol. 1983;
44:127-139.
22. Glasgow RE, Toobert DJ, Riddle M, et al. Diabetes-specific
social learning variables and self-care behaviors among persons
with type II diabetes. Health Psychol. 1989;8:285-303.
The Diabetes EDUCATOR
786
23. Hernandez CA. The development and pilot testing of The Diabetes
Activities Questionnaire (TDAQ): an instrument to measure adher-
ence to the diabetes regimen. Appl Nurs Res. 1997;10:202-211.
24. Fischbach F. A Manual of Laboratory and Diagnostic Tests. 7th
ed. Philadelphia, Pa: Lippincott; 2004.
25. Garay-Sevilla ME, Nava LE, Malacara JM, et al. Adherence to
treatment and social support in patients with non-insulin depend-
ent diabetes mellitus. J Diabetes Complicat. 1995;98:81-86.
26. Ford ME, Tilley BC, McDonald PE, et al. Social support among
African-American adults with diabetes, part 2: a review. J Natl
Med Assoc. 1998;90:425-432.
27. McDonald PE, Wykle ML, Misra R, et al. Predictors of social sup-
port, acceptance, health-promoting behaviors, and glycemic control
in African-Americans with type 2 diabetes. J Black Nurses Assoc.
2002:13:23-30.
28. Wilson W, Ary DV, Biglan A, et al. Psychosocial predictors of
self-care behaviors (compliance) and glycemic control in non-
insulin-dependent diabetes mellitus. Diabetes Care. 1986;9:614-622.
29. Kingery PM, Glasgow RE. Self-efficacy and outcome expecta-
tions in the self-regulation of non-insulin-dependent diabetes
mellitus. Health Educ. 1989;20:13-19.
30. Ludlow AP, Gein L. Relationships among self-care, self-efficacy
and HbA1C levels in individuals with non-insulin dependent dia-
betes mellitus. Can J Diabetes Care. 1995;19:10-15.
31. Aljasem LI, Peyrot M, Wissow L, et al. The impact of barriers and
self-efficacy on self-care behaviors in type 2 diabetes. Diabetes
Educ. 2001;27:393-404.
32. Senecal C, Nouwen A, White D. Motivation and dietary self-
care in adults with diabetes: are self-efficacy and autonomous
self-regulation complementary or competing constructs? Health
Psychol. 2000;19:452-457.
33. Heitzmann CA, Kaplan RM. Interaction between sex and social
support in the control of type II diabetes mellitus. J Consult Clin
Psychol. 1984;52:1087-1089.
34. Schwartz LS, Coulson LR, Toovy D, et al. A biopsychosocial
treatment approach to the management of diabetes mellitus. Gen
Hosp Psychiatry. 1991;13:19-26.
35. Connell CM, Fisher EB, Houston CA. Relationships among social
support, diabetes outcomes, and morale for older men and women.
J Aging Health. 1992;4:77-100.
36. Kaplan RM, Hartwell SL. Differential effects of social support
and social network on physiological and social outcomes in men
and women with type II diabetes mellitus. Health Psychol. 1987;
6:387-398.
37. Murphy DJ, Williamson PS, Nease DE. Supportive family mem-
bers of diabetic adults. J Fam Pract Res. 1994;14:323-331.
38. O’Connor PJ, Crabtree BF, Abourizk NN. Longitudinal study of a
diabetes education and care intervention: predictors of improved
glycemic control. J Am Board Fam Pract. 1992;5:381-387.
39. Allison OCP. The relationship of metabolic control to hardiness,
self-efficacy, and perceived medication adherence in adult with dia-
betes mellitus [dissertation]. Richmond: Virginia Commonwealth
University; 2003.
40. Anderson RM, Barr PA, Edwards GJ, et al. Using focus groups to
identify psychosocial issues of urban black individuals with dia-
betes. Diabetes Educ. 1996;22:28-33.
41. Bailey BJ, Lherisson-Cedeno D. Diabetes outcomes and prac-
tices: comparison of African Americans and Caucasians. J Natl
Black Nurses Assoc. 1997;9:66-75.
42. Bertera EM. Psychosocial factors and ethnic disparities in dia-
betes diagnosis and treatment among older adults. Health Social
Work. 2003;28:33-43.
43. Via PS, Salyer J. Psychosocial self-efficacy and personal charac-
teristics of veterans attending a diabetes education program.
Diabetes Educ. 1999;25:727-737.
44. Samuel-Hodge CD, Headen SW, Skelly AH, et al. Influences on
day-to-day self-management of type 2 diabetes among African-
American women. Diabetes Care. 2000;23:928-933.
45. Schoenberg NE, Drungle SC. Barriers to non-insulin dependent
diabetes mellitus (NIDDM) self-care practices among older
women. J Aging Health. 2001;13:443-466.
Volume 32, Number 5, September/October 2006