of Perceived Health in Women Diagnosed with HIV Disease
Kenn M. Kirksey, Mary Jane Hamilton, Mary
Holt-Ashley: Correlates of Perceived Health in Women Diagnosed
with HIV Disease. The Internet Journal of Advanced Nursing
Practice. 2003. Volume 5 Number 2.
The purpose of this descriptive
study was to examine the relationships among
self-efficacy, social support, quality of life, and
overall health perception in a sample of 61 women with a
mean age of 40 years living with HIV disease. Data
collection occurred in a university-based AIDS clinic in
a large metropolitan city in the southern United States.
Each participant completed a sociodemographic
questionnaire, the Sickness Impact Profile, the Norbeck
Social Support Questionnaire, and the Strategies Used by
Patients to Promote Health Questionnaire. The findings
suggest that social support and quality of life were
significantly related to overall perception of health.
Self-efficacy had an indirect impact on overall
perception of health via social support. These findings
support the need for nurses to continue exploring ways
to integrate social support within the domains of
clinical practice of persons with HIV disease.
The sociodemographic characteristics of individuals with
acquired immunodeficiency syndrome (AIDS) have rapidly shifted
during the past decade, as women now comprise the fastest
growing segment of this population. In June 1999, females
accounted for 32% of all adult cases of Human Immunodeficiency
Virus (HIV) infection (CDC, 1999). In addition, the pandemic
continues to disproportionately affect ethnic minorities.
Seventy-seven percent of HIV cases are among African-American
and Hispanic women. Persons aged 13-24 years account for 15% of
reported HIV cases, and women account for 49% of cases in this
age group (CDC,1999).
The two primary modes of transmission among women with AIDS
are injecting drug use (43%) and heterosexual contact (37%)
Research studies have documented different psychological and
sociodemographic characteristics among women and men with AIDS.
Males are reported as being primarily gay, white and college
educated (Kaplan, Marks & Mertens, 1997) and females are
primarily poor, racial minorities, and less educated
(Regan-Kubinski & Sharts-Hopko, 1995). In addition, many of the
women are single mothers who head households with young
children. It has also been suggested that because males are
members of the gay community they have access to more social
support than females (Regan-Kubinski & Sharts-Hopko, 1995).
Although the number of studies in women with AIDS is increasing,
research has focused primarily on prevalence and prevention in
white gay males. Women remain understudied, underreported, and
underdiagnosed. There are insufficient resources and services
targeting the primary prevention needs of women (Stevens, 1995).
of the Literature
As the number of women with HIV disease increases, and with
no cure in immediate sight, understanding how HIV positive women
adapt psychologically and socially has acquired a new importance
(Nannis, Patterson, & Semple, 1997). The psychosocial response
of women to HIV infection may have implications for disease
progression and quality of life. As AIDS treatments become more
effective, concerns about quality of life increase. There is
still too little information about quality of life among women
with HIV disease.
Semple et al. (1996) examined social
and psychological characteristics of 60 HIV –positive women and
gay men, matched on age and stage of HIV infection. A
discriminant function analysis was performed using measures of
life adversity, psychological distress, and psychosocial
resources. The best combination of predictors for distinguishing
between HIV-positive women and gay men was comprised of
satisfaction with emotional support, network size, use of
avoidant coping, depression, and loneliness scores [Chi 2(7) =
14.8, p < 0.05]. Women reported more satisfaction with emotional
support, although gay men had larger social networks. Gay men
used more avoidant coping and had higher loneliness scores as
compared to women, although women reported more depressive
Nannis, Patterson, & Semple (1997) assessed the psychosocial
correlates of 58 HIV-infected women. These women were generally
well educated, employed and therefore did not suffer from
socioeconomic factors associated with inner city living and drug
use. The researcher performed a discriminant function analysis
to determine which of five psychosocial variables could
discriminate between those who strongly self-identified with
each of three coping styles from those with low
self-identification. For problem solving style, the best
combination of predictors (loadings > 0.25) was: strong social
support, lower loneliness, depression, and anger, and a belief
in a chance locus of control. The best combination of predictors
for identification with a hopeless coping style was: loneliness,
depression, anger, lower social support and less belief in a
chance locus of control. Strong identification with a stoic
coping style included belief in a chance locus of control and
lower social support, loneliness, and anger. Results of the
study point to areas of psychosocial functioning which need to
be strengthened or diminished in order to maintain effective
coping for HIV-infected women and enhance their quality of life
(Nannis, Patterson, & Semple (1997).
In a descriptive study (n = 52), Rose &
Clark-Alexander (1996) explored quality of life and coping
methods of HIV positive women with children. The sample answered
an adapted Padilla Quality of Life Index and the Jalowiec Coping
Scale. Among the psychological, physical, and social quality of
life subscales, social quality of life was rated highest and the
psychological scale rated the lowest. Significant relationships
between coping and quality of life were examined.
The authors concluded that it is essential for nurses working
with HIV-infected women to be aware of their common concerns and
provide effective strategies to improve quality of life such as
enhancing decision making and coping skills within the context
of the family unit.
Sarna et al. (1999) described quality of life in a
convenience sample of 44 woman with symptomatic HIV disease.
Changes in physical, psychological, social, sexual dimensions,
and overall quality of life were evaluated. The mean scores of
the global quality of life, physical and psychosocial summary
scales were significantly different over a four-month period as
quality of life improved. The most prevalent disruptions were in
the psychosocial domain, including financial problems, worry
about family, distress about losing others from HIV, and worry
about disease progression. The most prevalent physical
disruptions included reduced energy, difficulties with daily
activities and frequent pain. Married women were more vulnerable
to changes in quality of life over time.
In a similar study, Moneyham et al. (1998) examined the
effectiveness of active and passive coping styles in a sample of
264 women with HIV disease. Active coping strategies that were
examined included social support, managing the illness, and
spiritual activities. Avoidance was used as an indicator of
passive coping. In both models, physical symptoms and emotional
distress were positively and significantly related. The findings
indicated that, whereas current avoidance coping was not related
to emotional distress, current active coping was positively
related to physical symptoms and negatively related to emotional
distress. The findings suggest that interventions that support
attempts to use active coping strategies as physical symptoms
increase may be effective in promoting positive adaptation to
Self-efficacy has been shown to impact health practices as
well as adaptation to illness and treatment (Bandura, 1986).
Bandura (1986) defined self-efficacy expectations as a person's
convictions that he or she can execute the behavior required to
produce a desired outcome. Increased self-efficacy has been
associated with increased adherence to treatment, increased
self-care behaviors, and decreased physical and psychological
symptoms. Converging findings from several areas of research
show that the effects of therapeutic interventions on health
behavior are partly mediated by changes in perceived
self-efficacy (Utz, Shuster, Merwin & Williams, 1994).
Support and Quality of Life
Sowell et al. (1997) evaluated the relationships between
social (family functioning, disclosure, material resources, and
social support) and psychological factors (stigma, fatalism,
avoidance, intrusion, and emotional distress) as predictors of
quality of life in HIV-infected women. The cross-sectional data
were tabulated from interviews of 264 women. There was a wide
range of scores in the sample among quality of life, general
anxiety, limited daily functioning, and HIV symptoms. The data
were analyzed using ANOVA, correlations and hierarchical
multiple regression analysis. Limited daily functioning was
predicted by stigma, fatalism, employment status, and stage of
disease (R2 = 0.179). General anxiety was predicted by emotional
distress, marital status, and intrusion (R2 = 0.503). Reported
HIV symptoms were predicted by race, age, employment status,
intrusion, disclosure, and material resources (R2 = 0.294). The
authors concluded that social and psychological factors are
important in quality of life of HIV-infected women.
In a similar study, Swindells et al. (1999) designed a
multi-center, longitudinal project with a sample of 138
HIV-infected patients. The study aim was to determine whether
the quality of life of HIV-infected patients was influenced by
satisfaction with social support, coping style, and
hopelessness. Quality of life did not correlate with age,
gender, race, HIV risk factor, education or marital status.
Employment (p = 0.0001), higher income (p = 0.03),
satisfaction with social support (p = 0.04), and
problem-focused coping (p = 0.03) were associated with a
significantly better quality of life. Emotion-focused coping (r
= -0.19, p = 0.04), avoidant coping (r = 0.40,
p = 0.0001), hopelessness (r = -0.64, p =
0.0001) and AIDS (p = 0.09) were predictors of poorer
quality of life. The authors concluded that interventions aimed
at alleviating hopelessness, maladaptive coping, and enhancement
of satisfaction with social support may improve overall quality
of life in HIV-infected patients.
Quality of life, social support and self-efficacy are omnibus
concepts. The literature cited in MEDLINE/HealthSTAR database
from 1995-2000 is expansive. Because of the extensiveness of the
published literature, this paper only presented select articles
that addressed the predictor variables as correlates of
perceived health status.
The purpose of this study was to describe the relationships
of self-efficacy, social support, quality of life, and
perception of overall health in women with HIV disease. Health
care professionals are challenged to devise interventions based
on psychosocial responses that are culturally appropriate,
sensitive to needs, and gender specific.
A descriptive survey design was used to examine and quantify
the relationships of self-efficacy, social support,
sickness-related dysfunction and perception of overall health in
women with HIV disease. Data collection occurred at an
outpatient facility in Houston, Texas. The clinic treats
approximately 5200 clients diagnosed with HIV or an
AIDS-defining illness. Approximately 44% are African-American
and 15% are Latina. Forty percent of the population is female.
A convenience sample (n = 61) of HIV-infected women
who met the inclusion criteria were recruited to participate.
Inclusion criteria were (1) at least 18 years of age, (2)
ability to understand English, (3) no severe visual impairment,
and (4) absence of stupor, unconscious or confused state.
Persons who met the inclusion criteria were approached either
by the principal investigator, co-principal investigator, or
clinical site coordinator and asked to participate in the study.
The participants were offered $10 gift certificates to a local
grocery store as compensation for their time.
Protection of human subjects was guaranteed by adhering to
the requirements of the Human Subjects Review Committees of
Texas A&M University-Corpus Christi and the clinical facility.
The study was described to potential participants, they were
asked to participate, and, if they agreed, were read the consent
form and given the opportunity to ask questions. They were
assured that failure to participate in the study would not
affect their health care or status at the facility. After
informed consent was obtained, the questionnaires were
administered to the participants. The instrument packet required
approximately 45 minutes to complete. Most questionnaires were
self-administered, but a few participants required the
questionnaires to be read.
The survey questionnaire was composed of four instruments,
which were completed by each participant. Predictor variables in
the sample were assessed by the sociodemographic questionnaire.
Self-efficacy was measured by The Strategies Used by Patients to
Promote Health (Lev & Owen, 1996). Social support was measured
by the Norbeck Social Support Questionnaire (Norbeck, Lindsey &
Carrieri, 1983), and sickness-related dysfunction was measured
by the Sickness Impact Profile (Bergner et al., 1976). The
outcome variable, overall health perception, was represented by
one item on the sociodemographic questionnaire.
The sociodemographic questionnaire was developed by the
authors and pilot tested with the other study instruments on a
sample of similar participants. It assessed age, ethnic group,
personal income, employment, educational level, marital status
and sexual orientation. It also asked participants how healthy
they thought they were. The subjects were also asked when they
were first told that they had HIV disease and about risk
Strategies Used by Patients to Promote Health (SUPPH).
The SUPPH is a 29-item self-report instrument used to measure
self-care/self-efficacy. It was previously used to measure this
variable in both patients with cancer and patients receiving
hemodialysis for end-stage renal disease (ESRD).
Self-care/self-efficacy is defined as a person's confidence to
perform strategies believed to promote health. Exploratory
factor analysis was investigated using 161 subjects; of these,
97 subjects received hemodialysis for ESRD and 64 subjects
received chemotherapy for cancer. Participants were asked to
rate the degree of confidence in carrying out specific self-care
behaviors. Each item was rated on a five-point scale of
confidence from 1 (very little) to 5 (quite a lot). The
instrument was scored by calculating mean responses across all
items for each subscore. Estimates of internal consistency
reliability (Cronbach's a) were estimated at r = 0.94 to
0.96 and validity evidence was reported (Lev & Owen, 1996).
Social Support Questionnaire (NSSQ).
The NSSQ consists of nine questions and measures functional
support; size, stability, and availability of social support;
and the total number of lost relationships over the past year.
Subjects evaluate social support using a five-point Likert-type
scale, with responses ranging from “not at all” to “a great
The NSSQ measures the variables of total functional, total
network, and total loss. For the first eight questions, the
participant is asked to rate each support person on a five-point
scale. Question 9 is answered with yes or no. These questions
indicate the numbers and quality of losses of social support
(Norbeck, Lindsey & Carrieri, 1981).
Validity and reliability were established at the time of
scale development (Norbeck, Lindsey & Carrieri, 1981) and
through subsequent testing (Norbeck, Lindsey & Carrieri, 1983;
Norbeck & Anderson, 1989). Reliability of the instrument was
established by internal consistency and test-retest. Internal
consistency on three subscale tests was demonstrated to be r
= 0.89, r = 0.96 and r = 0.97.
Impact Profile (SIP).
The SIP is a 136-item questionnaire that measures
sickness-related dysfunction according to twelve areas of
activity. The areas include: communication; social interaction;
alertness behavior; emotional behavior; mobility; recreation and
pastimes; work; home management; eating; ambulation; body care
and movement; and sleep and rest. The SIP was tested for
reliability in terms of internal consistency and reproducibility
on 278 randomly selected subjects from rehabilitation and
audiology/speech pathology units in a large, general hospital.
Internal consistency was r = 0.90 and 24-hour test-retest
reliability was determined to be r = 0.80 for the entire
instrument (Bergner et al., 1976). The SIP was also used in a
prospective study of adjustment to hemodialysis to measure
health status (Lev & Owen, 1998). Internal consistency was
reported as r = 0.75 to 0.92. Evidence of clinical
validity has been reported in several study reports (Bergner et
al., 1981; Bergner, Hudson, Conrad, et al., 1988; Patrick &
Overall health perception was measured using a single item on
the Sociodemographic Questionnaire. The item asked participants
the following question: “How healthy do you think you are?”
Possible responses included: 1) “Very healthy;” 2) “Better than
most people;” 3) “About average;” 4) “Not too healthy;” and 5)
After data cleaning, frequency distributions, descriptive
statistics and correlational analyses were computed for all
variables. Data were analyzed using SPSS Statistical
Package version 9.0 (SPSS Inc., Chicago, IL).
Sociodemographic data for study participants are presented in
Table 1. A selected set of variables explored the contribution
of the predictors to overall health perception. Then, a multiple
regression analysis was done on the outcome variable, overall
health perception, represented by social support,
illness-related dysfunction, and self-efficacy. Means, standard
deviations and ranges for descriptive data for predictor and
outcome variables used for multiple regression analyses are
presented in Table 2. Table 3 presents multiple regression among
the variables; ethnicity, social support, self-efficacy, and
illness-related dysfunction as predictors of perceived health
A total of 61 women with HIV Disease participated in this
study. Of these, 77.0% were African-American, 9.8% were Latina,
9.8% were white, not Hispanic, 1.6% were American Indian/Alaska
Native, and 1.6% classified themselves as other (see Table 1).
In addition, a majority (98.4%) had incomes of less than $15,000
annually and were unemployed (88.5%). Almost half (47.5%) had
completed some high school or college, and the majority (83.6%)
had never been married or were currently separated or divorced.
Participants ranged in age from 22 to 71 years, with a mean of
40.48 and standard deviation of 11.40. An overwhelming majority
(90.2%) self-identified as heterosexual. The mean time in months
since they were first informed of their HIV diagnosis was 51.82
with a standard deviation of 41.25 and range of 1 – 180 months.
Thirteen (21.3) had injected intravenous drugs and 23 (37.7%)
reported having unprotected sexual intercourse with a
The majority of the participants (88.6%) perceived their
health status as “about average” or better. Responses to the
question “How healthy do you think you are?” were “very healthy”
(n = 12, 19.7%), “better than most people” (n =
20, 32.8%), “about average” (n =22, 36.1%), and “not too
healthy” (n = 7, 11.5%). None of the subjects rated
themselves as “very sick.” African-Americans perceived
themselves as being more healthy than other ethnic groups.
Following dummy coding (African-Americans = 1, other races = 0),
the African-American women had a mean score of 2.19 with a
standard deviation of .85 compared to the women of other ethnic
groups who had a mean score of 3.07 with a standard deviation of
.92. Higher scores on this question indicated self-perception of
poorer health status.
Means, standard deviations, and ranges were calculated for
the predictor variables (see Table 2). A multiple regression was
calculated on the components (self-efficacy, quality of life,
and social support) predicting overall health status (see Table
3). Overall, the combination of predictors explained 25.5% of
the variance in the outcome variable, perceived health status.
Ethnicity (i.e., African-American women) accounted for 11.4% of
this explained variance. Figure 1 presents a model depicting the
relationships among the predictor and outcome variables. In the
analysis, social support and quality of life (psychosocial
dimension on the Sickness Impact Profile) had a direct effect on
overall health perception. Self-efficacy had an indirect impact
on overall health perception via social support.
These findings support other studies that have documented a
positive relationship between social support and quality of life
(overall health perception) (Nannis, Patterson, & Semple, 1997;
Swindells et al., 1999; Sowell et al., 1997). Social support and
the psychosocial dimension of sickness-related dysfunction were
significant predictors of overall health perception. Because
social support plays such a significant role in one's progress
through a life-threatening illness (Moser, 1994; White &
Frasure-Smith, 1995), strategies should be explored that might
enhance women's perception of having that support.
Ethnicity was identified as an exogenous variable in this
study. This is in contrast to findings by Swindells et al (1999)
that quality of life did not correlate with race. However,
Davidson et al. (1998) found similar characteristics in their
sample to those in this study. The participants in both studies
were primarily women of color who generally had little formal
Due to the study's design, causal relationships cannot be
inferred among social support, self-efficacy, quality of life
and overall health perception. In addition, generalizability is
limited given the sample size of 61 women with HIV infection.
In addition to contributing to the field of nursing knowledge
regarding the relationship between social support and quality of
life (health perception), this study on social support,
self-efficacy, and quality of life as multiple implications for
nursing care and the delivery of care. This research
demonstrated that social support and quality of life correlated
significantly with overall health perception. The implications
for nursing practice are to identify and implement innovative
means of fostering support in this population.
Future research efforts might replicate the study with a heterogeneous
sample of men and women. The relationships between social
support and quality of life, and overall health perception could
be examined in various populations. Other social support
instruments might be used to determine if different
conceptualizations of social support are better predictors of
the expected positive relationships. A grounded theory
methodology could be used to obtain data about social support
and overall health perception from the perspective of persons
living with HIV disease.
This research was supported by grants from Texas Nurses
Association District 17 and Texas A&M University-Corpus Christi
School of Nursing. Dr. Kirksey was supported during the
preparation of this manuscript by a postdoctoral fellowship from
the National Institute of Nursing Research, NIH, HIV/AIDS
Nursing Care and Prevention (T32 NRO7081), William L. Holzemer,
RN, PhD, FAAN, Project Director. The authors are grateful to
Mrs. Carolyn Barrett, the clinic staff, and clients for their
participation in this project. Thank you to Dr. William Holzemer
for his thoughtful review and editorial assistance in the
preparation of this article. The authors would also like to
thank Dr. Steven Paul for his assistance with statistical
Kenn M. Kirksey, PhD, APRN, BC
Associate Professor of Nursing
California State University, Fresno
Department of Nursing
Fresno, CA 93740-8031
1. Bandura, A. (1986). Social foundations of thought and
action: A social cognitive theory. Englewood Cliffs, NJ:
2. Bergner, M., Bobbitt, R., Pollard, W., Martin, D., &
Gilson, B. (1976). The Sickness Impact Profile: Validation of a
Health Status Measure. Medical Care, 14(1), 57-67.
3. Bergner, M., & Gilson, B. (1981). The Sickness Impact
Profile: The Relevance of Social Science to Medicine. In:
Eisenberh, L., and Kleinman, A. (eds). The Relevance of of
Social Science to Medicine. D. Riedel Publishing Co., Dordrecht:
4. Bergner, M., Hudson, L., Conrad, D., et al. (1988). The
Cost and Efficacy of Home Care for Chronic Lung Patients.
Medical Care, 26, 566-579.
5. CDC (1999). HIV/AIDS Surveillance Report: U.S. HIV and
AIDS cases reported through June 1999. 11(1):3-4.
6. Davidson, A., Bertram, S., Lezotte, D., Marine, W.,
Rietmeijer, C., Hagglund, B., & Cohn, D. (1998). Comparison of
health status, socioeconomic characteristics, and knowledge and
use of HIV-related resources between HIV-infected women and men.
Medical Care, 36(12), 1676-1684.
7. Fowler, M., Melnick., & Mathieson, B. (1997). Women and
HIV: Epidemiology and global overview. Obstetrics and Gynecology
Clinics of North America, 24(4), 705-729.
8. Kaplan, M., Marks, G., & Mertens, S. (1997). Distress and
coping among women with HIV infection: Preliminary findings from
a multiethnic sample. American Journal of Orthopsychiatry, 6(1),
9. Lev, E., & Owen, S. (1996). A Measure of Self-care
Self-efficacy. Research in Nursing & Health, 19(5), 421-429.
10. Lev, E., & Owen, S. (1998). A Prospective Study of
Adjustment to Hemodialysis. American Nephrology Nurses'
Association Journal, 25(5), 495-506.
11. Moneyham, L., Hennessy, M., Sowell, R., Demi, A., Seals,
B., & Mizuno, Y. (1998). The effectiveness of coping strategies
used by HIV-seropositive women. Research in Nursing and Health,
12. Moser, D. (1994). Social support and cardiac recovery.
Journal of Cardiovascular Nursing, 9(1), 27-36.
13. Nannis, E., Patterson, T., & Semple, S. (1997). Coping
with HIV disease among seropositive women: psychosocial
correlates. Women and Health, 25(1), 1-22.
14. Norbeck, J., & Anderson, N. (1989). Psychosocial
predictors of pregnancy complications in low income black,
hispanic, and white women. Nursing Research, 38, 204-209.
15. Norbeck, J., Lindsey, A., & Carrieri, V. (1981). The
development of an instrument to measure social support. Nursing
Research, 30, 264-269.
16. Norbeck, J., Lindsey, A., & Carrieri, V. (1983). Further
development of the Norbeck Social Support Questionnaire:
Normative data and validity testing. Nursing Research, 32, 4-9.
17. Patrick, D., & Deyo, R. (1989). Generic and Disease:
Specific Measures in Assessing Health Status and Quality of
Life. Medical Care, 27, 217-232.
18. Regan-Kubinski, M., & Sharts-Hopko, N. (1995). Illness
Cognition of HIV-infected Mothers. Issues in Mental Health
Nursing, 16(4), 327.
19. Rose, M., & Clark-Alexander, B. (1996). Quality of life
and coping styles of HIV-positive women with children. Journal
of the Association of Nurses in AIDS Care, 7(2), 28-34.
20. Sarna, L., van Servellen, G., Padilla, G., & Brecht, M.
(1999). Quality of life in women with symptomatic HIV/AIDS.
Journal of Advanced Nursing, 30(3), 597-605.
21. Semple, S., Patterson, T., Straits-Troser, K., Atkinson,
J., McCutchan, J., & Grant, I. (1996). Social and psychological
characteristics of HIV-infected women and gay men: HIV
Neurobehavioral Research Center Group. Women and Health, 24(2),
22. Sowell, R., Seals, B., Moneyham, L., Demi, A., Cohen,
L., & Brake, S. (1997) Quality of life in HIV-infected women in
the south-eastern United States. AIDS Care, 9(5), 501-512.
23. Stevens, P. (1995). Impact of HIV/AIDS on women in the
United States: challenges of primary and secondary prevention.
Health Care for Women International, 16(6), 577-595.
24. Swindells, S., Mohr, J., Justis, J., Berman, S., Squier,
C., Wagener, M., & Singh, N. (1999). Quality of life in patients
with human immunodeficiency virus infection: impact of social
support, coping style and hopelessness. International Journal of
STD and AIDS, 10(6), 383-391.
25. Utz, S., Shuster, G., & Williams, B. (1994). A
Community-based Smoking Cessation Program: Self-Care Behaviors
and Success. Public Health Nursing, 11(5),
26. White, R., & Frasure-Smith, N. (1995). Uncertainty and
psychologic stress after coronary angioplasty and coronary
bypass surgery. Heart & Lung, 24(1), 19-27.