Research in Nursing & Health, 2008, 31, 355–369
Testing the Theory of Self-Care
Management for Sickle
Cell Disease
Coretta M. Jenerette,1* Carolyn Murdaugh2**
1
School of Nursing, University of North Carolina at Chapel Hill, Carrington Hall, Campus Box 7460,
Chapel Hill, NC 27599-7460
2
College of Nursing, University of Arizona, Tucson, AZ
Accepted 24 September 2007
Abstract: Factors predicting health outcomes in persons with sickle cell
disease (SCD) were investigated within the framework of the theory of selfcare management for SCD, which proposes that vulnerability factors
negatively affect health care outcomes and self-care management resources
and positively mediate the relationship between vulnerability factors and
health care outcomes. A cross-sectional descriptive design was used to test
the model with a sample of 232 African American adults with SCD. Results
supported the negative effect of vulnerability factors on health outcomes. The
overall model was supported, however, self-care management resources did
not mediate the relationship between vulnerability and health care outcomes.
The findings provide support for interventions to increase self-care management resources to improve health care outcomes. ß 2008 Wiley Periodicals, Inc.
Res Nurs Health 31:355–369, 2008
Keywords: vulnerable populations; sickle cell disease; self-care management;
quality of life
Sickle cell disease (SCD) is an inherited
autosomal recessive genetic disorder that affects 1
in 500 Blacks (Andrews & Mooney, 1994; Godeau
et al., 2001; National Human Genome Research
Institute, 2006). Life expectancy for persons with
SCD has increased from 14 years in 1973 to the mid
to late 40s in 2004, transforming SCD into a longterm chronic illness (Quinn, Rogers, & Buchanan,
2004). This chronic disorder may result in a lifetime
of pain and frequent hospitalizations. Patients with
SCD frequently report dissatisfaction with care
they receive. Dissatisfaction is often related to
inadequate pain relief, which in turn has a negative
effect on health status and quality of life (Bolten,
Kempel-Waibel, & Pforringer, 1998; Osman et al.,
2000; Strickland, Jackson, Gilead, McGuire, &
Quarles, 2001). Individuals may also lack adequate
self-care management resources to be able to access
The authors would like to acknowledge the thoughtful contributions of Dr. Kenneth
Phillips, Dr. Joseph Telfair, and Dr. Joanne Herman to this research. Dr. Stephanie
Milan and Dr. Kristopher Fennie provided statistical guidance.
Contract grant sponsor: Substance Abuse and Mental Health Services Administration
(SAMHSA). Minority Fellowship Program at the American Nurses Association.
Contract grant sponsor: University of South Carolina Summer Dissertation
Fellowship.
Correspondence to Coretta M. Jenerette
*Postdoctoral Fellow.
**Professor and Associate Dean for Research.
Published online 4 February 2008 in Wiley InterScience
(www.interscience.wiley.com) DOI: 10.1002/nur.20261
ß 2008 Wiley Periodicals, Inc.
356
RESEARCH IN NURSING & HEALTH
and navigate the health care system. The purpose of
this research was to describe factors that predict
health outcomes in persons with SCD by testing a
modification of a theory- the theory of self-care
management for sickle cell disease (SCMSCD).
Medical interventions for SCD result in substantial costs. Self-care management is critical to
decrease health care costs as well as improve
the health status and quality of life for persons
living with SCD. Once SCD has been diagnosed,
individuals need to learn to manage symptoms and
maintain control over the course of the disease to
maintain an acceptable quality of life (Watkins
et al., 2000).
The SCMSCD (Fig. 1) shows (a) vulnerability
factors (socio-demographic and health-need factors) have a negative impact on both health
outcomes (health status and quality of life) and
self-care management resources (assertiveness,
self-efficacy, coping behaviors, social support,
self-care ability, self-care actions, and communication skills), and (b) self-care management resources
positively mediate the relationship between vulnerability factors and health outcomes. The SCMSCD
is based on the theory of self-care management
for vulnerable populations, a middle range theory
developed by one of the authors (CMJ) to describe
variables that influence self-care management,
health status, and quality of life among populations
who experience or are at risk for health disparities.
Model relationships have been supported in prior
published research with individuals with SCD
and other chronic illnesses (Dorsey, Phillips, &
Williams, 2001).
Despite extensive research on vulnerable populations, there is no consensus as to what constitutes
vulnerability. Broadly, vulnerability refers to the
likelihood of experiencing poor health and is
determined by a convergence of predisposing
characteristics at both individual and ecological
levels (Shi, 2001). These predisposing characteristics have been described previously (Aday,
1993) and are referred to as vulnerability factors
in the SCMSCD. Vulnerability factors determine
access to health care as well as health service
use, which in turn influence health outcomes,
including health status and quality of life (Shi).
In the SCMSCD vulnerability factors include
socio-demographic characteristics (age, income,
education, employment) and health need factors
(complications and SCD crises). Examining
multiple vulnerability factors is preferred to
studying individual factors, as multiple factors
capture reality and increase our understanding of
characteristics that are related to health outcomes
(Shi).
Age is a vulnerability factor in SCD. Painful
sickle cell crises tend to increase in severity
between ages 15 and 25 in males, progressively
decrease after age 30, and may cease after age 40
(Serjeant, 1997). Pregnancy may increase the
risk of crises for females (Miller et al., 2000).
With age, the potential for pulmonary fibrosis
and glomerular damage progressively increases
FIGURE 1. The theory of self-care management for sickle cell disease.
Research in Nursing & Health
SELF-CARE MANAGEMENT / JENERETTE AND MURDAUGH
(Serjeant). Additionally, SCD crises are inversely
proportional to the concentrations of hemoglobin
and fetal hemoglobin (Platt et al., 1991). The
average lifespan of African Americans with
sickle cell anemia, the most severe form of SCD,
is 42–48 years, compared to healthy African
Americans, who have an average lifespan of
67–75 years (Platt et al., 1994).
Income is another vulnerability factor in adults
with SCD. The negative psychological consequences of poverty, or material deprivation,
account for many of the adjustment problems
associated with SCD (Barbarin, Whitten, Bond, &
Conner-Warren, 1999). Individuals with lower
socio-economic status who have more than one
chronic illness report greater dissatisfaction
with health status than do those with middle incomes (Becker & Newsom, 2003). Additionally,
socio-economic indicators, particularly receipt
of income support such as welfare, (a marker of
poverty), have consistently been associated with
increased odds of poor health outcomes (Grundy
& Sloggett, 2003). In a study of illness beliefs,
concerns about ability for self-management
were predominantly reported in the lowest socioeconomic group (King, 2002).
Education, another vulnerability factor, has
been shown to be related to self-care management.
In a study of adults with heart failure, persons
with more education were more likely to engage
in self-care management than those with less
education (Mathieson, Kronefeld, & Keith, 2002;
Rockwell & Riegel, 2001; Tyson & McElduff,
2003). Among the elderly, higher education
may be an important tool in disease prevention,
progression, health maintenance, and consumer
protection (Hickey, Dean, & Holstein, 1986). Less
education was shown to be a significant predictor
of poor adherence to self-monitoring, a required
self-care component in persons with diabetes
(Karter, Ferrara, Darbinian, Ackerson, & Selby,
2000). Adherence is enhanced by attributes related
to education, such as the ability to comprehend the
prescribed regimen and accommodate it into one’s
lifestyle by making sound judgments (Goldman &
Smith, 2002).
Employment also is considered a vulnerability
factor in adults with SCD. Securing and maintaining employment may be very challenging, and it
has important financial and social implications
(V.J. Thomas & Taylor, 2002). Koshy and Dorn
(1996) described three potential employment
categories for adults with SCD: (a) individuals
with illness-related disability, (b) individuals with
job skills but unable to meet employer work
requirements due to recurrent, unpredictable
Research in Nursing & Health
357
crises, and (c) individuals without job skills. In a
study of beliefs and experiences of persons with
genetic and other clinical conditions, 42% of
respondents with SCD (n ¼ 99) reported they were
not hired for a job due to their medical condition
(Kass et al., 2004). In one study an adult with SCD
stated, ‘‘Once [an employer] finds out [about the
SCD], they think you’re going to miss a lot of days,
and they let you go’’ (Kass et al., p. 265).
Individuals with chronic disorders are concerned
about the prevalence of discrimination and fear the
effects of discrimination on securing employment
and health benefits (Johnson, Kass, & Natowicz,
2005; V.J. Thomas & Taylor). In studies of adults
with SCD, unemployment and disability range
from 26% to 70% (Gil et al., 2004; Kass et al.;
Lenoci, Telfair, Cecil, & Edwards, 2002; Telfair,
Haque, Etienne, Tang, & Strasser, 2003).
Race and ethnicity are also vulnerability
factors. However, SCD is predominately a disease
found in African Americans in the US. All of
our study participants were Black, so race and
ethnicity were not measured as vulnerability
factors in this study.
Health-need vulnerability factors include the
number of SCD complications and crises experienced annually. Frequent hospitalizations in
individuals with SCD lead to early mortality. The
majority of admissions are related to acute painful
crises (Houston-Yu, Rana, Beyer, & Castro, 2003;
Platt et al., 1994; Prasad, Hasan, Castro, Perlin, &
Kim, 2003). Researchers have demonstrated that
adults with SCD who have three or more painful
crises per year have more than a 27% mortality
rate compared to 17% in those with fewer than
three crises (Steinberg et al., 2003). Self-care
management strategies that can mitigate complications and reduce acute painful crises may be
helpful in decreasing hospitalizations.
Self-care management resources (assertiveness,
self-efficacy, coping behaviors, social support, selfcare ability, self-care actions, and communication
skills) and health outcomes (health status and
quality of life) included in the SCMSCD all have
been supported in prior research (C. Edwards et al.,
2006; Gil et al., 2000; Jenerette & Phillips, 2006;
Lenoci et al., 2002; Strickland et al., 2001).
The purpose of this research was to describe
socio-demographic and health need vulnerability
factors and self-care management resources
that are related to health status and quality of
life in persons with SCD. Two hypotheses were
tested:
1. Vulnerability factors will significantly negatively
affect health outcomes in adults with SCD.
358
RESEARCH IN NURSING & HEALTH
2. Self-care management resources will mediate
the relationship between vulnerability and
health outcomes.
METHODS
Sample and Setting
A convenience sample of 232 adults with a
diagnosis of SCD was recruited for a crosssectional descriptive study. Sample size was based
on Kline’s (2005) structural equation modeling
(SEM) guidelines, which indicate that a minimum
sample of 225 subjects is required for stable
parameter estimates of the model (approximately
15 model parameters). Kline defined sample
sizes greater than 200 as large and suggested
10–20 observations per estimated parameter.
Another recommendation for calculating sample
size for SEM is 50 more than 8 times the number
of variables or 15 cases per measured variable
(Garson, n.d.). A calculated sample size based on
the 15 variables measured in this study indicated
that 210 participants ([8 15] þ 50 ¼ 170) were
needed. In the literature, sample sizes for SEM
are commonly 200–400 for 10–15 indicators
(Garson, n.d.). When problems with missing data
or non-normal distributions are not anticipated,
a minimal sample size of 200 is recommended for
any SEM (Weston & Gore, 2006).
Participants in the study had to be able to read,
write, or understand English, be at least 18 years
old, and have a diagnosis of SCD. Participants
were recruited from two SCD clinics in the
southeast United States between September 2003
and April 2004. The university’s Institutional
Review Board and the SCD clinics approved
the study prior to data collection. The PI and a
registered nurse who had been trained recruited
eligible patients and collected data. After giving
written informed consent, participants completed
a questionnaire packet during a clinic visit. A
maximum of 60 minutes was needed to complete
the questionnaires. Five participants needed
assistance, and the PI or nurse read the questions
aloud and recorded their responses. Participants
received $25 for completing the study.
The sample is described in Table 1. Using zip
codes provided by participants, the average
median household income was approximately
Table 1. Demographic Description of Total Sample
Variable
Na
M (SD)
Range
Age (years)
Education (years)
1st SCD crisis (age in years)
Average SCD crises per year (number)
Median annual household income
231
232
212
219
221
34.91 (12.42)
12.36 (1.99)
6.47 (7.54)
2.65 (3.08)
$36,039 ($10,125)
18–73
5–21
1–55
0–25
$17,843–$65,375
Variable
Sex
Employment
Marital status
Living situation
Environment
Categories
Female
Male
Full-time
Part-time
Not-employed
Disabled
Never married
Married
Divorced
Widowed
Separated
Alone
With family
With friends
Rural
Urban
SCD, sickle cell disease.
a
Note: The sample size varies due to missing data.
Research in Nursing & Health
N (%)
143 (61.60)
89 (38.40)
48 (20.70)
24 (10.30)
64 (27.60)
96 (41.40)
143 (61.60)
47 (20.30)
18 (7.80)
12 (5.20)
12 (5.20)
42 (18.30)
174 (75.70)
14 (6.10)
75 (32.90)
153 (67.10)
SELF-CARE MANAGEMENT / JENERETTE AND MURDAUGH
$36,000 (US Census Bureau, 2005). Zip codebased median income, a proxy for socio-economic
position that often is more easily obtained than
personal income data, has been shown to be
a significant independent mortality predictor (A.J.
Thomas, Eberly, Smith, & Neaton, 2006). The
sample was examined by clinic site to assess
any differences. Participants were an average of
4 years older at one site (33 vs. 37 years), reported
earlier crisis onset (5 vs. 8 years), and had a higher
average median household income ($40,000 vs.
$34,000). No other significant differences were
noted.
In addition to demographic information, participants were asked about their SCD complications. Anemia (69%) and painful crises (77%)
were the most commonly reported complications
in the total sample. Vision problems were also
common (63%). Slightly more that one-fourth
(26%) reported being depressed. Other complications reported were osteonecrosis (19%), kidney
problems (15%), and leg ulcers (15%).
Measures
Cronbach’s alpha reliability coefficients for instruments used in this study are shown in Table 2. All
instruments were assessed with the Flesch-Kincaid
reading level assessment using Microsoft Wordß.
The measures were modified as needed to insure a
reading level at the eighth grade or lower.
Vulnerability
Demographic data/health need vulnerability
questionnaire. The 12-item questionnaire was
developed by the authors to obtain age, sex, education, employment, and marital status. The two
items to assess vulnerability factors were derived
from the literature and supported by prior research
so they were not pretested. The items were, ‘‘On
average, how many sickle cell crises do you have
per year that require being in the hospital?’’ and,
‘‘Circle all of the following complications you have
experienced because of having SCD.’’ The total
number of SCD crises per year and the total number
of complications were the measures of health need
characteristics of vulnerability.
Self-Care Management Resources
Assertiveness, or behavior that enables one to
stand up for one’s rights without infringing on the
rights of others, was measured with the 30-item
Simple Rathus Assertiveness Schedule (SRAS)
(McCormick, 1984). The SRAS is a parallel form
the Rathus Assertiveness Schedule (RAS; Rathus,
Research in Nursing & Health
359
1973). The RAS has established construct validity
in terms of significant correlations with scales
that measure impressions respondents make on
other people and indications of how people would
behave in certain situations. McCormick reported
an internal consistency reliability of .90. The
SRAS correlates .94 with the original RAS. A
total score is obtained by summing responses,
with higher responses indicating higher levels of
assertiveness.
Coping behaviors, cognitive and behavioral
strategies used to master conditions of harm,
threat, or challenge when a normal or routine
response is not available, were measured with the
Family Coping Project Coping Scale (COP; Demi,
Moneyham, Sowell, & Cohen, 1997; Moneyham
et al., 1997). The 54-item Likert-type instrument
contains five subscales (seeking/using social
support, spiritual activities, avoidance, managing
illness, and focusing on others). Construct validity
was demonstrated using exploratory factor
analysis, and internal consistency reliabilities
have ranged from .71 to .90. Subscale scores are
obtained by summing responses, with higher
responses indicating higher levels of the specific
coping behavior.
Self-efficacy, belief about the ability to achieve
a desired health outcome, was measured with
the Sickle Cell Disease Self-Efficacy Scale
(SCSES; Edwards, Telfair, Cecil, & Lenox,
2000). The nine-item Likert-type scale measures
SCD disease-specific perceptions of self-efficacy.
Convergent validity was estimated by significant
correlations between self-efficacy and self-esteem
(.39), sense of mastery (.45), and internal health
locus of control (.41). An internal consistency
reliability of .89 was reported. A total score is
obtained by summing the responses, with higher
scores indicating higher self-efficacy.
Social support, internal perception of interpersonal transactions, including expressions of
positive affect, affirmation of another’s behaviors
or views, or giving symbolic or material aid, was
measured with the Medical Outcomes Study
Social Support Survey (MOS-SSS; Sherbourne
& Stewart, 1991). The 19-item Likert-type
scale measures perceived availability of support
with four subscales (emotional/informational,
affectionate, tangible, and positive social interaction). Construct validity was supported by both
confirmatory and principal components factor
analysis. Both the total scale and subscales
have internal consistency reliabilities above .91
(Sherbourne & Stewart). Responses are summed,
with higher scores indicating higher perceptions
of available support.
Research in Nursing & Health
Present
Without sickle cell
Past
Future
Emotional/informational
Affectionate
Tangible
Positive social interaction
Seeking/using social support
Avoidance
Spiritual activities
Managing the illness
Focusing on others
Subscale
199
141
195
181
208
203
207
222
198
212
222
226
221
192
225
218
224
188
216
214
202
200
N
50–180 (30–180)
56–156 (0–132)
3–36 (0–36)
2–33 (0–36)
2–21 (0–21)
8–27 (0–27)
3–12 (0–12)
12–45 (9–45)
22–95 (19–95)
9–40 (8–40)
3–15 (3–15)
5–20 (4–20)
4–20 (4–20)
24–120
14–22 (8–32)
0–50 (0–50)
5–22 (4–22)
51–280 (70–280)
7–70 (7–70)
7–70 (7–70)
13–70 (7–70)
14–70 (7–70)
Actual range (potential range)
113.74 (20.11)
101.34 (19.74)
20.42 (6.86)
14.82 (5.49)
15.23 (4.16)
20.34 (4.00)
8.34 (2.25)
30.92 (7.16)
75.64 (15.12)
31.00 (7.01)
12.27 (2.82)
15.82 (3.58)
15.37 (3.64)
87.85 (9.02)
28.22 (3.08)
18.12 (10.19)
12.85 (3.50)
223.07 (33.72)
51.77 (10.90)
61.13 (8.98)
49.67 (10.88)
58.37 (10.14)
M (SD)
.85
.90
.85
.83
.72
.73
.56
.87
.95
.92
.82
.80
.87
.75
.72
.82
.77
.95
.87
.87
.86
.90
Cronbach’s alpha
SRAS, Simple Rathus Assertiveness Schedule; ASA, Appraisal of Self-Care Agency; COP, Family Project Coping Scale; J-SAT, Jenerette Self-Care Assessment Tool; SCSES,
Sickle Cell Self-Efficacy Scale; BQ-C, Barriers Questionnaire- Communication; MOS-SSS, Medical Outcomes Study Social Support Survey; CIQOLL, Chronic Illness Quality of Life
Ladder.
Self-Care Ability (ASA) total
Self-Care Action (J-SAT)
Communication Skills (BQ-C)
Health status
Quality of Life (CIQOLL) total
Self-Efficacy (SCSES) total
Social Support (MOS-SSS) total
Assertiveness (SRAS)
Coping Behaviors (COP) total
Scale
Table 2. Self-Care Management Resources and Health Outcomes Results Summary
360
RESEARCH IN NURSING & HEALTH
SELF-CARE MANAGEMENT / JENERETTE AND MURDAUGH
Self-care ability, the capability to engage in
therapeutic behaviors to maintain and/or improve
health status and quality of life, was measured
with the Appraisal of Self-Care Agency Scale
(ASA; Evers et al., 1986). The 24-item scale has
been used in healthy and ill adults from various
ethnic groups across the lifespan The ASA
has established construct validity and a content
validity index of .88 as well as internal consistency
reliabilities ranging from .80 to .86. A total score is
obtained by summing responses, with higher
responses indicating higher levels of self-care
ability.
Self-care action, defined as engaging in therapeutic activities and actively accessing resources
to maintain or improve health status and quality of
life, was measured with the eight-item Jenerette
Self-Care Assessment Tool (J-SAT; unpublished).
Construct validity was estimated by a significant
negative correlation with the Centre for Epidemiological Studies Depression Scale (CESD;
Radloff, 1977) and a significant positive correlation with the Functional Status Questionnaire
(FSQ; Jette et al., 1986). The J-SAT had an internal
consistency reliability of .80 in an initial test of
the instrument. Higher summed scores indicate
greater frequency of self-care actions.
Communication skills were defined as ones
ability to share thoughts, feelings, or information.
The Barriers Questionnaire Communication
subscale (BQ-C) assesses one’s communication
skills in pain treatment, as pain management is an
essential component of self-care management in
this population (Wells, Johnson, & Wujcik, 1998).
Validity has been established in prior studies for
the original instrument (Ward et al., 1993). The
10-item communication subscale has a reported
internal consistency reliability of .78. The summed
scores were reversed so that higher scores
indicated higher communication skills and fewer
communication barriers.
361
Murdaugh, Moneyham, Jackson, Phillips, &
Tavakoli, 2006). The 28-item scale uses the
concept of a self-anchoring striving scale (Cantril,
1965) based on the notion of a gap between one’s
current situation and one’s goals and aspirations.
The scale measures life satisfaction in seven
domains: physical status, emotional status, financial status, family and friends, spiritual wellbeing, peace of mind, and overall satisfaction with
life. The instrument uses a 10-stepladder response
format (Cantril; Kilpatrick & Cantril, 1960). The
top rung of the ladder (10) indicates the best
possible quality of life and the bottom rung (1)
indicates the worst possible QOL. Each of the
above seven domains was measured for four time
periods: present QOL; QOL without SCD (i.e., if
the individual did not have a diagnosis of SCD);
QOL 1-year ago; and anticipated future QOL
(1 year from present circumstances). Each time
period represents a subscale. The CIQOLL convergent validity was demonstrated by a significant
correlation with the Functional Status Questionnaire (Jette et al., 1986) and divergent validity
by a significant negative correlation with the
CES-D (Radloff, 1977), and all internal consistency coefficients exceeded the criterion level
of .80 with Cronbach’s a coefficients ranging
from .91 to .95 for the four subscales (Murdaugh
et al.). Higher summed scores indicate higher
QOL. The total score was used in the current
study.
Table 2 shows the descriptive data and reliability
coefficients for measures of self-care management
resources and health outcomes. The reliability of
measures is an important factor in determining
model fit (Jackson, 2003). All measures except one
subscale had reliability coefficients greater than
.70, an acceptable level (Nunnally & Bernstein,
1994).
Data Analysis
Health Outcomes
Health status, one’s perceived physical state,
was measured with a 5-item questionnaire adapted
from the 49-item Health Status scale (Segovia,
Bartlett, & Edwards, 1989). Construct validity of
the original scale has been supported by significant
correlations with measures of health care utilization, but reliability data for the original scale was
not reported. Higher total scores indicate higher
perceived physical health status.
Quality of life, a subjective sense of wellbeing with physical, psychological, and social
dimensions of one’s life, was measured with the
Chronic Illness Quality of Life Ladder (CIQOLL;
Research in Nursing & Health
The SPSS (version 12.0) program was used for
construction of the covariance matrix to be used
in SEM analysis and to check the multivariate
normality of the data. All data were examined both
visually and with the Shapiro–Wilk statistical test
for normality. Transformations were performed
on the data to improve normal distributions.
Not all data, after being transformed, met the
normality assumption via Shapiro–Wilk. Variables with data that did not meet the assumption
of normality either visually or with Shapiro–
Wilk were excluded from the SEM analyses.
These included self-care action, coping behaviors
362
RESEARCH IN NURSING & HEALTH
subscales (5), QOL subscales (4), and social
support subscales (4), for a total of one scale and
13 subscales.
SEM analytical techniques were used to assess
the proposed relationships in the hypothesized
model. Missing data varied across measures (see
Table 2). The AMOS Graphics 4.0 SEM program
uses a procedure known as Full Information
Maximum Likelihood to handle missing data.
This program assumes that data are missing at
random (Byrne, 2001). In order to meet this
assumption, all scales with more than 5% missing
data were evaluated with t-tests to test for
significant differences between participants with
complete data and those with missing data. No
statistically significant differences were found.
Therefore, the assumption that data were missing
at random was met.
First, the measurement models (the mapping
of measures onto theoretical constructs) were
assessed to describe the measurement properties
of latent variables. Next the structural equation
model (the theorized links between the variables),
which specifies direct and indirect relationships
among the latent variables, was obtained to
describe the amount of variance explained by
the model (Kenny, Kashy, & Bolger, 1998). The
chi-square statistic was used to test model fit.
However, because chi-square has no upper bound,
it cannot be standardized, and it is very sensitive to
sample size. For this reason, the chi-square was
divided by the degrees of freedom (w2/df). An
acceptable w2/df ratio is less than 3 (Byrne, 2001;
Kline, 2005).
The model’s overall fit also was assessed using
the following goodness of fit statistics: comparative fit index (CFI), a chi-square estimate
using a maximum likelihood solution, > .90;
normed fit index (NFI), > .90; and the root mean
square error of approximation (RMSEA), < .10.
Data were analyzed using SPSS and AMOS
software.
The latent variables (vulnerability, self-care
management resources, and health outcomes)
along with their respective indicators were
correlated to assess relationships among the latent
variables and the strength of loadings of indicators
on respective latent variables. Higher loadings
indicate a stronger relationship between the
indicator and the latent variable (Kline, 2005;
Pedhazur & Schmelkin, 1991). Indicators were
deleted if standardized loadings were less than .40.
Deleting these indicators improved convergent
validity among each factor’s indicators and
discriminant validity among the latent variables
(Kline).
Research in Nursing & Health
RESULTS
In the first model (Fig. 2), the three measurement
models were correlated, and theorized indicators
of each latent variable were included. In the
vulnerability measurement model, only the health
need vulnerability factor indicators were included
because the socio-demographic vulnerability
factor indicators were controlled for when the
final measurement model was determined. In
the self-care management resources model measures included assertiveness, self-efficacy, coping
behaviors, social support, self-care ability, and
communication skills. Two self-care management
measures that had standardized loadings below
.40, (a) coping behaviors (.31) and (b) the
communications subscale (.33), were deleted,
resulting in Model 2 (Fig. 3). In Model 2, all
standardized loadings were greater than .40 and
correlations among latent variables were less than
.85, so the measurement models were considered
adequate (Kline, 2005). The structural model was
then evaluated to test the two study hypotheses.
Determining the Structural Model
The initial structural model was analyzed to test
the hypothesized relationships among the latent
variables by evaluating the w2/df (< 3), the CFI
( > .90), NFI ( > 90), and the RMSEA (< .10).
Figure 4 depicts the initial structural model. The fit
statistics were: w2/df: 3.3, CFI: .99, NFI: .99, and
RMSEA: .10.
In the final structural model, Figure 5, the sociodemographic vulnerability factors (age, income,
education, and employment) were controlled. The
fit statistics were: w2/df: 3.12, NFI: .98, and
RMSEA: .096. When the socio-demographic
factors were controlled, the theorized relationships in the SCMSCD did not change. Furthermore, the direction of the relationships between
latent variables, statistical significance, and magnitude were either unchanged or improved. For
example, when controlling for socio-demographic
vulnerability factors, vulnerability accounted for
32% of the variance in self-care management
resources, an increase from 13%. The fit statistics
improved slightly from the initial to the final
structural model.
Testing Hypotheses
Once the final structural model was determined,
study hypotheses were addressed. Hypothesis 1
stated that vulnerability factors would significantly
SELF-CARE MANAGEMENT / JENERETTE AND MURDAUGH
363
FIGURE 2. Correlations among measurement models.
negatively affect health outcomes in adults with
SCD. While controlling for socio-demographic
vulnerability factors, the health need vulnerability
factors negatively affected health outcomes (path
coefficient .55, p < .05), so the hypothesis was
supported.
Hypothesis 2 stated that self-care management
resources would mediate the relationship between
vulnerability and health outcomes in adults with
SCD. Using Preacher’s (2006) interactive tool to
test for mediation, results indicated a Sobel
test statistic of 1.90 (p ¼ .057). Therefore, selfcare management resources did not mediate
the relationship between vulnerability and health
outcomes. Hypothesis 2 was not supported.
Factor loadings provide important information
about how well an observed variable is able to
measure a latent variable. The factor loadings are
the regression of the factor on indicators. Therefore, by squaring the factor loadings, one has
the percent of variance that is explained by the
latent variable. Self-care management resources
explained 36% of the variance in social support.
Research in Nursing & Health
Forty-five percent of the variance in self-efficacy
was explained by self-care management resources. Self-care management resources
accounted for 38% of the variance in self-care
ability. Self care resources also accounted for 25%
of the variance in assertiveness. Finally, 87% of
the variance in health outcomes was accounted for
by self-care management resources.
DISCUSSION
The theoretical modeling tested factors expected
to predict health status and quality of life for
persons with SCD. The hypothesis of mediation
was not supported. However, results of model
testing substantiate prior research on other chronic
diseases that has described the benefit of self-care
management resources, such as assertiveness,
social support (Cox, 2002; Dorsey & Murdaugh,
2003; Smith et al., 2005), self-efficacy (R.
Edwards, Telfair, Cecil, & Lenoci, 2001; Farrell,
Wicks, & Martin, 2004; Robinson-Smith & Pizzi,
364
RESEARCH IN NURSING & HEALTH
FIGURE 3. Correlations among measurement models with standardized loadings less than .40 deleted.
FIGURE 4. Initial structural model.
Research in Nursing & Health
SELF-CARE MANAGEMENT / JENERETTE AND MURDAUGH
365
FIGURE 5. Final structural model controlling for socio-demographic factors.
2003), and self-care ability (Dorsey & Murdaugh;
Greene, Yedidia, & The Take Care to Learn
Evaluation Collaborative, 2005; Lenoci et al.,
2002) to health outcomes.
Individuals need to be capable of self-care, be
confident in their self-care abilities, and have
social support in order to enhance their health
outcomes. Results suggest that a combination
of self-care ability, self-efficacy, assertiveness,
and social support, all self-care management
resources, may assist individuals with SCD to
manage the day to day activities required to cope
with a lifelong illness and access needed health
care resources.
Assertiveness training is a potential self-care
management resource strategy for persons with a
chronic illness, such as SCD, who must frequently
access and navigate the healthcare system. Previous research also has identified the role of
assertiveness training as a resource in chronic
illness (Andersen, Abullarade, & Urban, 2005;
Krupat et al., 1999; Richardson, 2000).
Finding support the critical role of self-care
management resources for improving health status
Research in Nursing & Health
and quality of life in chronic illness. These
resources are amenable to interventions, as
interventions can target assertiveness, self efficacy, social support, and self care ability. Self-care
management resources enable persons to gain and
or maintain control over their disease, and to
improve their health status and quality of life.
In future research, the vulnerability predisposing characteristics need to be expanded to
include additional individual factors and ecological factors. As mentioned earlier, although
vulnerability is extensively mentioned in the
literature, measurement is less well described.
In addition, no published measures of the vulnerability predisposing factors for SCD are available.
In the health care utilization framework, Aday
(1993) addressed characteristics that determine
an individual’s ability to use health services,
including demographic variables, social structure
variables, beliefs, enabling characteristics, need
characteristics, and attributes of the community.
Both socio-demographic and need factors were
included in this study. However, additional
relevant vulnerability factors, such as ecological
366
RESEARCH IN NURSING & HEALTH
vulnerability factors need to be explored for their
potential contribution, including residence, social
capital (family structure, religious organizations,
neighborhood connections), availability and accessibility of health care, and community health
indicators. Last, for racial and ethnic minorities
with SCD, additional vulnerability factors not
previously considered may need to be included,
such as stress due to racial discrimination (Lightsey & Barnes, 2007). The limited attention in the
literature to the measurement of vulnerability
as a multidimensional concept, as well as a lack of
reliable and valid measures of this concept,
suggest the need for further research prior to
retesting the model.
Limitations
The results provide important insight about the
needs of adults who live with SCD. However,
several limitations need to be acknowledged.
Although significant differences were noted in
age, onset of first SCD crisis, and median household income between the two recruitment sites, the
sites were combined to achieve the sample size
required for SEM.
The socio-demographic vulnerability factor,
employment, was measured categorically and
had to be excluded from the initial analysis, as
continuous measures were necessary to apply the
powerful statistical functions of the SEM program
used, AMOS 4.0. However, categorical variables
can be controlled for in using AMOS 4.0, as they
were in this study.
Missing data were also a potential issue. The
AMOS Graphics 4.0 SEM program fits models
to incomplete raw data within a single sample
without the need for case deletion or imputation of
missing values (Kline, 2005). Although SEM with
AMOS can accommodate incomplete data, missing data may have been an issue in testing the
model.
Missing data, although apparently missing at
random, may have resulted from response burden,
an important consideration in research for persons
with a chronic illness such as SCD (Shadish,
Cook, & Campbell, 2002). Respondents completed
11 questionnaires ranging in length from 5 to
54 items each. A balance is needed between longer
instruments that have higher reliability and require
smaller sample sizes, yet increase response burden,
and shorter instruments that require larger sample
sizes but impose less response burden.
Use of self-report data is considered a limitation
by some authors. However, self-report data are
Research in Nursing & Health
essential to behavioral and healthcare research
and practice (Baldwin, 2000). For this reason self
report measures and data collection procedures
need to be as robust as possible, as they are influenced by wording, format, and context
(Schwartz & Oyserman, 2001). Strategies to decrease potential issues were implemented. All of
the measures had sound psychometric properties,
and measures were at the appropriate level and in
a language the participants understood. The data
collectors, both African American, established
rapport with the participant prior to administering
the measures.
A potential validity issue was that all of the
model concepts were measured with one method.
An individual’s score is a component of both the
concept (trait) measured by the scale items, as well
as the method used to measure the items. Because
only one measurement method (self report) was
used, it is not possible to measure the effect of
methods variance on the path coefficients in the
model. This issue can be addressed in future
research by using multiple, independent methods
to measure each concept or trait to be able to
demonstrate convergence, a minimal requirement
in testing for methods versus trait variance
(Campbell & Fiske, 1959).
The goodness of fit criterion of a w2/df less than
3 (Kline, 2005) was not met. This may cast some
doubt on the parameter estimates of the final
model, although the final model is plausible. There
is a lack of agreement on the appropriate value of
w2/df. Some researchers allow w2/df values as high
as 5 to consider a model adequate fit, while others
insist on a ratio of 2 or less (Bollen, 1989; Garson,
2007).
Finally, to meet the assumptions of normality as
well as criteria for retaining variables for the SEM
analysis, several theorized self-care management
resource variables (coping behaviors and self-care
activities) were deleted. In order to have confidence in the final structural model, the analysis
needs to be repeated with another sample and
including the full set of theorized variables.
CONCLUSIONS
The Theory of Self Management for SCD was
tested using SEM. In this first test of the
model, results indicate that almost all theorized
relationships were supported. Additional research
is needed to expand the vulnerability factors
(predisposing characteristics). The model can
then be retested to examine further the hypothesized mediation effects. In spite of lack of support
SELF-CARE MANAGEMENT / JENERETTE AND MURDAUGH
for mediation in this study, results suggest
potential interventions to increase self-care management resources to improve health status and
life quality.
REFERENCES
Aday, L.A. (1993). Indicators and predictors of health
care utilization. In S.J. Williams & P.R. Torrens
(Eds.), Introduction to health services (4th ed.,
pp. 46–70). Albany, NY: Demar.
Andersen, M.R., Abullarade, J., & Urban, N. (2005).
Assertiveness with physicians is related to women’s
perceived roles in the medical encounter. Women and
Health, 42, 15–33.
Andrews, M.M., & Mooney, K.H. (1994). Alterations in
hematologic function in children. In K.L. McCance &
S.E. Huether (Eds.), Pathophysiology: The biologic
basis for disease in children (pp. 908–939). St. Louis,
MO: Mosby-Yearbook.
Baldwin, W. (2000). Information no one else knows:
The value of self-report. In A.S. Stone, J.S. Turkkan,
C.A. Bachraach, J.B. Jobe, H.S. Kurtzman, & V.S.
Carin (Eds.), The science of self-report: Implications
for research and practice (pp. 3–8). Mahwah, NJ:
Erlbaum.
Barbarin, O.A., Whitten, C.F., Bond, S., & ConnerWarren, R. (1999). The social and cultural context of
coping with sickle cell disease: II. The role of
financial hardship in adjustment to sickle cell disease.
Journal of Black Psychology, 25, 294–315.
Becker, G., & Newsom, E. (2003). Socioeconomic
status and dissatisfaction with health care among
chronically ill African Americans. American Journal
of Public Health, 93, 742–748.
Bollen, K.A. (1989). Structural equations with latent
variables. New York: Wiley.
Bolten, W., Kempel-Waibel, A., & Pforringer, W.
(1998). Analysis of the cost of illness in backache
[in German]. Medizinische Klinik, 93, 388–393.
Byrne, B. (2001). Structural equation modeling with
AMOS: Basic concepts, applications, and programming. Mahwah, NJ: Erlbaum.
Campbell, D.T., & Fiske, D.W. (1959). Convergent
and discriminant validation by the multitraitmultimethod matrix. Psychological Bulletin, 56,
81–105.
Cantril, H. (1965). The patterns of human concerns.
New Brunswick, NJ: Rutgers University Press.
Cox, L.E. (2002). Social support, medication compliance and HIV/AIDS. Social Work in Health Care, 35,
425–460.
Demi, A., Moneyham, L., Sowell, R., & Cohen, L.
(1997). Coping strategies used by HIV infected
women. OMEGA—Journal of Death and Dying, 35,
377–391.
Dorsey, C.J., & Murdaugh, C.L. (2003). The theory of
self-care management for vulnerable populations.
The Journal of Theory Construction and Testing, 7,
43–49.
Research in Nursing & Health
367
Dorsey, C., Phillips, K., & Williams, C. (2001). Adult
sickle cell patients’ perceptions of nurses’ caring
behaviors. The Association of Black Nursing Faculty
Journal, 12, 95–100.
Edwards, R., Telfair, J., Cecil, H., & Lenoci, J. (2000).
Reliability and validity of a self-efficacy instrument
specific to sickle cell disease. Behavior Research and
Therapy, 38, 951–963.
Edwards, R., Telfair, J., Cecil, H., & Lenoci, J. (2001).
Self-efficacy as a predictor of adult adjustment to
sickle cell disease: One-year outcomes. Psychosomatic Medicine, 63, 850–858.
Edwards, C., Whitfield, K., Sudhakar, S., Pearce, M.,
Byrd, G., Wood, M., et al. (2006). Parental substance
abuse, reports of chronic pain and coping in adult
patients with sickle cell disease. Journal of the
National Medical Association, 98, 420–428.
Evers, G., Isenberg, M., Phillipsen, G., Brouns, G.,
Halfens, R., & Smeets, H. (1986). The ‘appraisal of
self-care agency’ ASA-scale: Research program to
test reliability and validity. Proceedings of the
International Nursing Research Conference New
Frontiers in Nursing Research (p. 130). Edmonton,
Canada: University of Alberta, Canada.
Farrell, K., Wicks, M.N., & Martin, J.C. (2004). Chronic
disease self-management improved with enhanced
self-efficacy. Clinical Nursing Research, 13, 289–
308.
Garson, G.D. (2007). Structural equation modeling:
Introduction. Retrieved July 4, 2007, from http://
www2.chass.ncsu.edu/garson/pa765/structur.htm.
Garson, G.D. (n.d.). Structural equation modeling, from
statnotes: Topics in multivariate analysis. Retrieved
May 30, 2007, from http://www2.chass.ncsu.edu/
garson/pa765/statnote.htm.
Gil, K.M., Carson, J.W., Sedway, J.A., Porter, L.S.,
Schaeffer, J.J., & Orringer, E. (2000). Follow-up of
coping skills training in adults with sickle cell
disease: Analysis of daily pain and coping practice
diaries. Health Psychology, 19, 85–90.
Gil, K.M., Carson, J.W., Porter, L.S., Scipio, C.,
Bediako, S.M., & Orringer, E. (2004). Daily mood
and stress predict pain, health care use, and work
activity in African American adults with sickle-cell
disease. Health Psychology, 23, 267–274.
Godeau, B., Noel, V., Habibi, A., Schaeffer, A., Bachir,
D., & Galacteros, F. (2001). Sickle cell disease in
adults: Which emergency care by the internists? La
Revue de Medecine Interne, 22, 440–451.
Goldman, D.P., & Smith, J.P. (2002). Can patient selfmanagement help explain the SES health gradient?
Proceedings of the National Academy of Sciences of
the United States of America, 99, 10929–10934.
Greene, J., Yedidia, M.J. & the Take Care to Learn
Evaluation Collaborative. (2005). Provider behaviors
contributing to patient self-management of chronic
illness among underserved populations. Journal of
Health Care for the Poor and Underserved, 16, 808–
824.
Grundy, E., & Sloggett, A. (2003). Health inequalities in
the older population: The role of personal capital,
368
RESEARCH IN NURSING & HEALTH
social resources and socio-economic circumstances.
Social Science and Medicine, 56, 935–947.
Hickey, T., Dean, K., & Holstein, B.E. (1986). Emerging
trends in gerontology and geriatrics: Implications for
the self-care of the elderly. Social Science and
Medicine, 23, 1363–1369.
Houston-Yu, P., Rana, S.R., Beyer, B., & Castro, O.
(2003). Frequent and prolonged hospitalizations: A
risk factor for early mortality in sickle cell disease
patients. American Journal of Hematology, 72, 201–
203.
Jackson, D.L. (2003). Revisiting sample size and
number of parameter estimates: Some support for
the N:Q hypothesis. Structural Equation Modeling,
10, 128–141.
Jenerette, C.M., & Phillips, R.C.S. (2006). An examination
of differences in intra-personal resources, self-care
management, and health outcomes in older and younger
adults with sickle cell disease. Southern Online Journal
of Nursing Research, 7, 1–24. Retrieved, September 21,
2007, from http://www.snrs.org/publications/SOJNR_
articles/iss03vol07. pdf.
Jette, A.M., Davies, A.R., Cleary, P.D., Calkins, D.R.,
Rubenstein, L.V., Fink, A., et al. (1986). The
Functional Status Questionnaire: Reliability and
validity when used in primary care. Journal of
General Internal Medicine, 1, 143–149.
Johnson, S., Kass, N.E., & Natowicz, M. (2005).
Disclosure of personal medical information: Differences among parents and affected adults for genetic
and nongenetic conditions. Genetic Testing, 9, 269–
280.
Karter, A.J., Ferrara, A., Darbinian, J.A., Ackerson,
L.M., & Selby, J.V. (2000). Self-monitoring of
blood glucose: Language and financial barriers in a
managed care population with diabetes. Diabetes
Care, 23, 477–483.
Kass, N.E., Hull, S.C., Natowicz, M.R., Faden, R.R.,
Plantinga, L., Gostin, L.O., et al. (2004). Medical
privacy and the disclosure of personal medical
information: The beliefs and experiences of those
with genetic and other clinical conditions. American
Journal of Medical Genetics Part A, 128, 261–270.
Kenny, D., Kashy, D.A., & Bolger, N. (1998). Data
analysis in social psychology. In D. Gilbert, S.T.
Fiske, & G. Lindzey (Eds.), Handbook of social
psychology (4th ed., Vol. 1, pp. 233–268). New York:
Oxford University Press.
Kilpatrick, F.P., & Cantril, H. (1960). Self-anchoring
scaling: A measure of individuals’ unique reality
worlds. Washington, DC: Brookings Institution.
King, R. (2002). Illness attributions and myocardial
infarction: The influence of gender and socioeconomic circumstances on illness beliefs. Journal
of Advanced Nursing, 37, 431–438.
Kline, R.B. (2005). Principles and practice of structural
equation modeling (2nd ed.). New York: The
Guilford Press.
Koshy, M., & Dorn, L. (1996). Continuing care for adult
patients with sickle cell disease. Hematology—
Oncology Clinics of North America, 10, 1265–1273.
Research in Nursing & Health
Krupat, E., Irish, J.T., Kasten, L.E., Freund, K.M.,
Burns, R.B., Moskowitz, M.A., et al. (1999). Patient
assertiveness and physician decision-making among
older breast cancer patients. Social Science and
Medicine, 49, 449–457.
Lenoci, J.M., Telfair, J., Cecil, H., & Edwards, R.F.
(2002). Self-care in adults with sickle cell disease.
Western Journal of Nursing Research, 24, 228–
245.
Lightsey, O.R., & Barnes, P.W. (2007). Discrimination,
attributional tendencies, generalized self-efficacy,
and assertiveness as predictors of psychological
stress among African Americans. Journal of Black
Psychology, 33, 27–50.
Mathieson, K.M., Kronenfeld, J.J., & Keith, V.M.
(2002). Maintaining functional independence in
elderly adults: The roles of health status and financial
resources in predicting home modifications and use of
mobility equipment. Gerontologist, 42, 24–31.
McCormick, I. (1984). A simple version of the Rathus
Assertiveness Schedule. Behavioral Assessment, 7,
95–99.
Miller, S.T., Sleeper, L.A., Pegelow, C.H., Enos, L.E.,
Wang, W.C., Weiner, S.J., et al. (2000). Prediction of
adverse outcomes in children with sickle cell disease.
The New England Journal of Medicine, 342, 83–89.
Moneyham, L., Demi, A., Mizuno, Y., Sowell, R., Seals,
B., & Guillory, J. (1997). Developing and testing of a
contextually and culturally relevant measure of
coping for use with women infected with HIV.
OMEGA—Journal of Death and Dying, 36, 359–
374.
Murdaugh, C., Moneyham, L., Jackson, K., Phillips, K.,
& Tavakoli, A. (2006). Predictors of quality of life in
HIV-infected rural women: Psychometric test of the
chronic illness quality of life ladder. Quality of Life
Research, 15, 777–789.
National Human Genome Research Institute. (2006).
Learning about sickle cell disease. Retrieved
December 12, 2006, from http://www.genome.gov/
10001219.
Nunnally, J.C., & Bernstein, I.H. (1994). Psychometric
theory. New York: McGraw-Hill.
Osman, L., Calder, C., Robertson, R., Friend, J.A.,
Legge, J.S., & Douglas, J.G. (2000). Symptoms,
quality of life, and health service contact among
young adults with mild asthma. American Journal of
Respiratory and Critical Care Medicine, 161, 498–
503.
Pedhazur, E.J., & Schmelkin, L.P. (1991). Measurement, design, and analysis: An integrated approach.
Hillsdale, NJ: Erlbaum.
Platt, O.S., Thorington, B.D., Brambilla, D.J., Milner,
P.F., Rosse, W.F., Vichinsky, E., et al. (1991). Pain in
sickle cell disease. Rates and risk factors. The New
England Journal of Medicine, 325, 11–16.
Platt, O.S., Brambilla, D.J., Rosse, W.F., Milner, P.F.,
Castro, O., Steinberg, M.H., et al. (1994). Mortality in
sickle cell disease. Life expectancy and risk factors
for early death. The New England Journal of
Medicine, 330, 1639–1644.
SELF-CARE MANAGEMENT / JENERETTE AND MURDAUGH
Prasad, R., Hasan, S., Castro, O., Perlin, E., & Kim, K.
(2003). Long-term outcomes in patients with sickle
cell disease and frequent vaso-occlusive crises.
American Journal of the Medical Sciences, 325,
107–109.
Preacher, K.J. (2006). An interactive calculation tool for
mediation tests. Retrieved May [22,] 2007, from
http://www.psych.ku.edu/preacher/sobel/sobel.htm.
Quinn, C.T., Rogers, Z.R., & Buchanan, G.R. (2004).
Survival of children with sickle cell disease. Blood,
103, 4023–4027.
Radloff, L.S. (1977). The CES-D scale: A self-report
depression scale for research in the general population. Applied Psychological Measurement, 1, 385–
401.
Rathus, S. (1973). A 30-item schedule for assessing
assertive behavior. Behavior Therapy, 4, 398–406.
Richardson, S.S. (2000). The effects of an assertiveness
training intervention on clients’ level of assertiveness, locus of control, and ways of coping with
serious illness. Unpublished doctoral dissertation,
Louisiana State University Medical Center, New
Orleans.
Robinson-Smith, G., & Pizzi, E.R. (2003). Maximizing
stroke recovery using patient self-care self-efficacy.
Rehabilitation Nursing, 28, 48–51.
Rockwell, J.M., & Riegel, B. (2001). Predictors of selfcare in persons with heart failure. Heart and Lung, 30,
18–25.
Schwartz, N., & Oyserman, D. (2001). Asking questions
about behavior: Cognition, communication, and
questionnaire construction. American Journal of
Evaluation, 22, 127–160.
Segovia, J., Bartlett, R.F., & Edwards, A.C. (1989). An
empirical analysis of the dimensions of health status
measures. Social Science and Medicine, 29, 761–
768.
Serjeant, G.R. (1997). Sickle-cell disease. Lancet, 350,
725–730.
Shadish, W.R., Cook, T.D., & Campbell, D.T. (2002).
Experimental and quasi-experimental designs for
generalized causal inference. Boston: Houghton
Mifflin.
Sherbourne, C.S., & Stewart, A.L. (1991). The MOS
social support survey. Social Science and Medicine,
32, 705–714.
Shi, L. (2001). The convergence of vulnerable characteristics and health insurance in the US. Social
Science and Medicine, 53, 519–529.
Smith, W.R., Bovbjerg, V.E., Penberthy, L.T., McClish,
D.K., Levenson, J.L., Roberts, J.D., et al. (2005).
Research in Nursing & Health
369
Understanding pain and improving management of
sickle cell disease: The PISCES study. Journal of the
National Medical Association, 97, 183–193.
Steinberg, M.H., Barton, F., Castro, O., Pegelow, C.H.,
Ballas, S.K., Kutlar, A., et al. (2003). Effect of
hydroxyurea on mortality and morbidity in adult
sickle cell anemia: Risks and benefits unto 9 years of
treatment. JAMA, 289, 1645–1651.
Strickland, O., Jackson, G., Gilead, M., McGuire, D.B.,
& Quarles, S. (2001). Use of focus groups for pain and
quality of life assessment in adults with sickle cell
disease. Journal of the National Black Nurses
Association, 12, 36–43.
Telfair, J., Haque, A., Etienne, M., Tang, S., & Strasser,
S. (2003). Rural/urban differences in access to and
utilization of services among people in Alabama with
sickle cell disease. Public Health Reports, 118, 27–
36.
Thomas, V.J., & Taylor, L.M. (2002). The psychosocial
experience of people with sickle cell disease and its
impact on quality of life: Qualitative findings from
the focus groups. British Journal of Health Psychology, 7, 345–363.
Thomas, A.J., Eberly, L.E., Smith, G.D., & Neaton, J.D.
(2006). Zip-code-based versus tract-based income
measures as long-term risk-adjusted mortality predictors. American Journal of Epidemiology, 164,
586–590.
Tyson, M.J., & McElduff, P. (2003). Self-bloodpressure monitoring—a questionnaire study: Response, requirement, training, support-group popularity and recommendations. Journal of Human
Hypertension, 17, 51–61.
US Census Bureau. (2005). United States Census 2000.
Retrieved January 6, 2006, from http://www.census.gov/main/www/cen2000.html.
Ward, S.E., Goldberg, N., Miller-McCauley, V., Mueller, C., Nolan, A., Pawlik-Plank, D., et al. (1993).
Patient-related barriers to management of cancer
pain. Pain, 52, 319–324.
Watkins, K., Connell, C.M., Fitzgerald, J.T., Klem, L.,
Hickey, T., & Ingersoll-Dayton, B. (2000). Effect of
adults’ self-regulation of diabetes on quality of life.
Diabetes Care, 23, 1511–1515.
Wells, N., Johnson, R.L., & Wujcik, D. (1998).
Development of a short version of the Barriers
Questionnaire. Journal of Pain and Symptom
Management, 15, 294–297.
Weston, R., & Gore, P.A., Jr. (2006). A brief guide
to structural equation modeling. The Counseling
Psychologist, 34, 719–751.