10.
1177/1055329002250992
ARTICLE
JANAC Vol. 14, No. 2, March/April 2003
Neidig et al. / Aerobic Exercise Training
Aerobic Exercise Training for
Depressive Symptom Management
in Adults Living With HIV Infection
Judith L. Neidig, PhD, RN
Barbara A. Smith, PhD, FAAN
Dale E. Brashers, PhD
Aerobic exercise training may help prevent or
reduce depressive symptoms experienced by persons
living with HIV infection. However, the psychological
effects of aerobic exercise have not been studied exten-
sively. This study evaluated the effects of an aerobic
exercise training program on self-reported symptoms
of depression in HIV-infected adults and examined the
convergent validity of two widely used depressive
symptom scales. Sixty HIV-infected adults partici-
pated in a randomized, controlled trial of a supervised
12-week aerobic exercise training program. As com-
pared to study controls, exercise participants showed
reductions in depressive symptoms on all indices, and
total depressive symptoms scores were highly corre-
lated. Additional study of the psychological effects of
aerobic exercise programs in the target population is
recommended.
Key words: HIV, exercise, depression, symptom
Depression is the most common reason for psychiat-
ric evaluation and treatment of persons living with
HIV infection. Researchers have estimated the preva-
lence of major depression in HIV-infected persons in
clinical settings to be between 22% and 32% (Brown
et al., 1992; Evans et al., 1998; Ferrando et al., 1998;
Rabkin, 1997; Williams, Rabkin, Remien, Gorman, &
Ehrhardt, 1991). These estimated rates are approxi-
mately twice the observed rate in the general commu-
nity population (Kessler et al., 1994). However,
JOURNAL OF THE ASSOCIATION OF NURSES IN AIDS CARE, Vol. 14, No. 2, March/April 2003, 30-40
DOI: 10.1177/1055329002250992
Copyright © 2003 Association of Nurses in AIDS Care
prevalence differences have been less pronounced
when HIV-positive and HIV-negative subpopulations
(gay men, intravenous drug users) were compared in
studies including control groups (Lipsitz et al., 1994;
Rosenberger et al., 1993). In the first nationally repre-
sentative sample of adults receiving care for HIV
infection in the United States, more than one third of
adults screened positive for major depression during
the previous 12 months (Bing et al., 2001).
Although the 1996 probability sample (Bing et al.,
2001) predates the widespread use of more highly
active antiretroviral therapies, persons living with HIV
continue to face considerable psychological chal-
lenges. Distressing physical symptoms (Bonfanti
et al., 2000; Carr et al., 1998) and considerable psy-
chological demands often accompany these lifetime
regimens (Chesney, Morin, & Sherr, 2000).
Researchers who reported a significant reduction in
depressive symptoms with protease inhibitor
antiretroviral therapies also noted that 46% of patients
continued to have significant levels of depressive
symptoms 1 year after treatment with these more
Judith L. Neidig, PhD, RN, is the director of the Office of
Responsible Research Practices at The Ohio State Univer-
sity, Columbus, Ohio. Barbara A. Smith, PhD, FAAN, is a
professor and the Marie L. O’Koren endowed chair in the
School of Nursing at the University of Alabama at Birming-
ham. Dale E. Brashers, PhD, is an associate professor in the
Department of Speech Communication at the University of
Illinois, Urbana-Champaign, Illinois.

effective anti-HIV drug regimens (Low-Beer et al.,
2000). One group reported significant levels of depres-
sive symptoms in 52% of individuals who had not yet
initiated anti-HIV drug regimens (Low-Beer et al.,
2000).
With the lengthening course of HIV illness, the pre-
vention and management of depressive symptoms
have important personal and public health benefits.
Although consensus is lacking, some have reported
more rapid immune decline and accelerated mortality
in persons with HIV infection reporting higher levels
of depressive symptoms (Burack et al., 1993; Mayne,
Vittinghoff, Chesney, Barrett, & Coates, 1996).
Depression is also thought to negatively affect quality
of life more than most medical conditions (Hays,
Wells, Sherbourne, Rogers, & Spritzer, 1995). Among
adults with HIV infection, Sherbourne et al. (2000)
found that those with symptoms of depression or other
probable mood disorders had significantly lower
health-related quality-of-life scores than did those
without depression. Suicide ideation is an important
depressive symptom that must be carefully assessed in
persons with HIV infection. Marzuk et al. (1988) dem-
onstrated dramatic reductions in suicide rates in HIV/
AIDS samples. However, as Levine (2001) astutely
commented, the improved HIV/AIDS therapies that
have reduced suicide among persons living with HIV
infection have also led to new stressors. Requirements
for lifetime adherence to complicated therapies and
HIV risk behavior modification (Rabkin & Ferrando,
1997) as well as the need to renegotiate feelings of
hope and new social roles (Brashers et al., 1999) pres-
ent new challenges for persons once reconciled to a
death from HIV/AIDS.
Beyond the compelling need to manage depressive
symptoms to ameliorate personal suffering, effective
management of depression may also promote HIV
drug adherence and help to minimize HIV transmis-
sion. Depression has been associated with non-
adherence to HIV drug regimens (Gordillo, del Amo,
Soriano, & Gonzalez-Lahoz, 1999; Singh et al., 1999)
and may be associated with increased sexual risk
behaviors, although findings associating negative
affect (depression, anxiety, anger) with sexual risk
behaviors have been mixed (Kalichman, 1999). In a
recent meta-analysis, Crepaz and Marks (2001) con-
cluded there is little evidence that negative affect is
Neidig et al. / Aerobic Exercise Training 31
associated with increased sexual risk behavior. How-
ever, Kalichman and Weinhardt (2001) argued this is
not an unexpected finding when the aggregate litera-
ture is examined given methodological shortfalls in the
original studies. Because partial adherence to complex
HIV drug regimens may lead to drug resistance
(Vanhove, Schapiro, Winters, Merigan, & Blaschke,
1996) and because prevalence of drug resistance in pri-
mary HIV infections is increasing (UK Collaborative
Group on Monitoring the Transmission of HIV Drug
Resistance, 2001), additional research into the impact
of depression on adherence and sexual risk taking is
needed.
Effective drug therapy for clinically depressed HIV-
infected persons is available (e.g., Rabkin, Wagner, &
Rabkin, 1999). Selection from among traditional
(tricyclic, selective serotonin reuptake inhibitors) and
alternative agents (dextramphetamine or testosterone
replacement) is often aimed at minimizing cumulative
side effects of antiretroviral and antidepressant regi-
mens and time to clinical response. However, despite
available drugs and psychotherapy, depressive symp-
toms in this population often are not treated
(Sambamoorthi, Walkup, Olfson, & Crystal, 2000).
Providers may view depressive symptoms as a normal
component of the underlying HIV disease or HIV ther-
apy (Rabkin, 1997).
In addition, depressive symptoms consistently cor-
relate with physical symptom reports (Rabkin et al.,
1997), and the overlap between symptoms of the viral
infection, related neuropsychological symptoms, and
depressive symptoms (e.g., fatigue, poor appetite,
weight loss, trouble concentrating) has made selecting
a measure of depressive symptoms difficult. Because
there is no HIV-disease-specific measure of depres-
sion symptoms, many researchers, including those
studying the largest cohorts (Burack et al., 1993;
Mayne et al., 1996; Ostrow et al., 1989), have assessed
depressive symptoms with the Center for Epidemio-
logical Studies–Depression Scale (CES-D) (Radloff,
1977). This scale is thought to be less confounded by
physical symptoms (Drebing et al., 1994). On the
other hand, a considerable number of researchers (e.g.,
Singh, Squier, Sivek, Wagener, & Yu, 1997) have col-
lected self-reported symptoms of depression within
HIV and AIDS studies using the Beck Depression
Inventory (BDI). The BDI is a symptom measure
32 JANAC Vol. 14, No. 2, March/April 2003
originally intended as a clinical index of depression
severity (Beck, Ward, Mendelson, Mock, & Erbaugh,
1961). To overcome symptom confounds, some HIV/
AIDS researchers have developed CES-D subscales
(Burack et al., 1993; Ostrow et al., 1989), whereas oth-
ers have used the total scale and controlled for symp-
toms during analysis (Lyketsos et al., 1993). This
inconsistent use of symptom indices, concern about
symptom confounds, and divergent scoring
approaches have limited interpretation of this body of
research.
Aerobic exercise training is also associated with
decreased depression in health and chronic illness
(Dishman, 1985). In HIV-negative persons, regular
exercise has been shown to be an effective therapy for
mild to moderate depression and an efficacious
adjunct for severe depression (Morgan & Goldston,
1987; Tkachuk & Martin, 1999). In controlled studies
of clinically depressed patients, aerobic exercise train-
ing has been consistently shown to be more effective
than no treatment or placebo control and to have favor-
able efficacy when compared to psychotherapy and
cognitive therapy (Tkachuk & Martin, 1999). As com-
pared to traditional depression treatments, exercise
therapy may be more cost-effective (Greist et al.,
1979). As such, aerobic exercise may be a cost-
effective way for HIV-positive persons to prevent or
reduce mild to moderate symptoms of depression. In
light of potential antiretroviral and antidepressant drug
interactions (Rabkin, 1997), exercise may also be an
important drug adjunct for more severely depressed
persons.
Aerobic exercise is one of the most frequently
reported alternative and complementary activities
used by HIV-infected persons (Greene et al., 1999).
Results from small, controlled studies also suggest
that HIV-infected persons may engage in habitual
moderate-intensity exercise to improve cardiopul-
monary fitness without clinically significant CD4 lym-
phocyte cell decreases (MacArthur, Levine, & Birk,
1993; Rigsby, Dishman, Jackson, Maclean, & Raven,
1992; Smith et al., 2001; Stringer, Berezovskaya,
O’Brien, Beck, & Casaburi, 1998) or undesired increases
in plasma HIV viral load (Smith et al., 2001; Stringer
et al., 1998). Mental health outcomes, however, have not
been studied as extensively, and the effects of chronic
exercise on depressive symptoms have not been isolated.
Using Profile of Mood State (POMS) scores,
LaPerriere and colleagues (1990, 1991) concluded
that aerobic exercise training attenuated stressors
associated with receipt of an HIV-positive test result.
Others (MacArthur et al., 1993) noted a trend to
improvement on an unspecified mental health ques-
tionnaire, and Lox, McAuley, and Tucker (1995)
reported improved mood in HIV-infected exercisers as
compared to study controls. Wagner, Rabkin, and
Rabkin (1998) also reported mood enhancement when
exercise was combined with steroid therapy. Overall,
these findings have been limited by small samples,
limited control for disease stage and concomitant
drugs, variable exercise protocols (Smith et al., 2001),
and the use of diverse mental health measures.
The current study examines the efficacy of aerobic
exercise training for the management of depressive
symptoms experienced by adults living with HIV
infection. The purpose of this study was to (a) deter-
mine the impact of participation in a 12-week aerobic
exercise training program on self-reported depressive
symptoms and (b) to examine the correlations among
participant scores on several widely used measures of
depressive symptoms and depressed mood. The
researchers hypothesized that the experimental group,
as compared to the study controls, would have
improved Week 12 CES-D, BDI, and POMS depres-
sion subscale scores.
Method
Study data were collected between 1995 and 1998
as part of a larger study examining aerobic exercise
training for HIV symptom management (Smith et al.,
2001). Participants were randomly assigned to a 12-
week aerobic exercise training or to a wait list (main-
tain usual activity) condition within a pretest/posttest
control group design. The study received prior
approval by the institutional review board, and
informed consent was obtained. Participants were
exercise tested in a university health and wellness cen-
ter and mainstreamed for training into university staff
and patient fitness programs at the same facility. Other
procedures were performed at the university’s AIDS
Clinical Trials Unit (ACTU). A convenience sample of
60 adults who had laboratory-confirmed HIV

Table 1. Pearson Product-Moment Correlations Between Mean Scale Scores
Week 0
CES-D BDI POMS POMS-D
CES-D .93 .71 .79 .83
BDI .71 .81 .76 .70
POMS .79 .76 .90 .93
POMS-D .83 .70 .93 .92
NOTE: CES-D = Center for Epidemiological Studies–Depression Scale; BDI = Beck Depression Inventory; POMS = Profile of Mood State;
POMS-D = POMS depression subscale. Scale reliabilities are in bold.
infection and met Centers for Disease Control and Pre-
vention (CDC) disease classifications A2 (asymptom-
atic, non-AIDS) or B2 (symptomatic, non-AIDS)
(CDC, 1993) was recruited from the ACTU and the
local community. If participants were taking
antiretrovirals or antidepressants at screening, regi-
mens were stable for at least 8 weeks. During screen-
ing, all participants were examined and cleared medi-
cally by an ACTU physician.
Exercisers were required to attend three supervised
1-hour training sessions per week. Trained exercise
leaders coached participants to exercise a minimum of
30 minutes within an individually prescribed target
heart rate range corresponding to 60% to 80% of maxi-
mum oxygen uptake (VO2 max). Personal trainers
supervised the exercise sessions conducted in a medi-
cal exercise facility with nurse practitioners, exercise
physiologists, and physicians in attendance. Partici-
pants chose an aerobic exercise modality from among
treadmill use, stationary biking, and walking. Further
details of the inclusion/exclusion criteria, exercise
testing and training protocols, and participant moni-
toring are reported elsewhere (Smith et al., 2001).
Measures
Dependent measures were collected at Weeks 0 and
12. Only exercise participants who completed at least
28 of 36 sessions and attended at least 80% of sessions
during the month prior to the graded exercise posttest
were included in these analyses.
CES-D. The CES-D is a 20-item self-report mea-
sure of symptoms commonly associated with depres-
sion. The scale emphasizes affective distress and was
Neidig et al. / Aerobic Exercise Training 33
Week 12
CES-D BDI POMS POMS-D
.93 .71 .90 .86
.71 .89 .78 .67
.90 .78 .91 .91
.86 .67 .91 .96
not intended as a diagnostic measure of clinical
depression (Radloff, 1977). Respondents rate the fre-
quency of symptoms experienced in the past week on a
4-point scale; each item receives a score from 0 to 3,
and total scores range from 0 to 60. Scores greater than
16 indicate a high level of depressive symptoms and
have good sensitivity in predicting a clinical diagnosis
of depression (Weissman, Prusoff, & Newberry,
1975). High internal consistency, concurrent validity
with self-report and clinical interviews, and construct
validity have been established (Radloff, 1977). Reli-
ability (Cronbach’s alpha = .90) has been established
in HIV/AIDS samples (DeGenova, Patton, Jurich, &
Macdermid, 1994); reliability estimates for adminis-
trations of the CES-D and other dependent measures in
this sample are reported in bold in Table 1.
BDI. The BDI is a self-report depressive symptom
index that includes 21 items. Participants rate each
symptom on severity, and individual items are scored
from 0 to 3. Total scores range from 0 to 63, and a score
of 9 is thought to be indicative of at least mild depres-
sion. A score of 15 is designated as a possible depres-
sion cutoff in the general population (Beck & Steer,
1993). Concurrent validity with other measures of
depression, internal consistency, and instrument sta-
bility have been demonstrated (Beck & Steer, 1993);
the BDI has been frequently used by researchers
examining the effects of exercise on depression
(Tkachuk & Martin, 1999).
POMS. The POMS is a 65-item instrument measur-
ing six mood states: tension/anxiety, depression/
dejection, anger/hostility, fatigue/inertia, confusion/
bewilderment, and vigor/activity (McNair, Lorr, &
34 JANAC Vol. 14, No. 2, March/April 2003
Droppleman, 1971). Respondents rate the occurrence
of adjective items during the past week on a 5-point
scale from 0 (not at all) to 4 (extremely). With the
exception of the vigor subscale, higher scores indicate
a negative mood state. Internal consistency estimates
of 0.90 or greater have been reported for all six mood
scales, and test-retest reliabilities have been estimated
at 0.65 to 0.74. Construct and predictive validity using
other measures of depression, anxiety, and distress
have been established (McNair, Lorr, & Droppleman,
1992), and excellent reliability for the entire scale and
the 15-item depression subscale (Cronbach’s alpha =
.93) has been reported in HIV samples (e.g., Perkins
et al., 1995).
Other Variables
A number of potentially confounding or explana-
tory variables were assessed at Weeks 0 and 12 in the
larger exercise study (Smith et al., 2001). These
included measures of physical fitness, physical activ-
ity outside the exercise study, stress, social support,
and HIV-related medical history and self-reported
symptoms.
Physical fitness. Aerobic fitness or peak aerobic
power (VO2 max) was measured during graded exer-
cise testing on a motor-driven treadmill. Testing fol-
lowed guidelines adapted from American College of
Sports Medicine (ACSM) (2000) standards for exer-
cise testing and was done in a controlled laboratory
setting by staff members experienced with exercise
testing in clinical populations.
Physical activity. General physical activity (during
the prior 2 weeks) was measured at baseline and at
Week 12 using an adaptation of an instrument origi-
nally designed by Paffenbarger, Wing, and Hyde
(1978). Participants in both groups were asked to
report the number of hours they spent in light, moder-
ate, or vigorous activity. These data were desired to
identify changes in moderate or vigorous nontraining
physical activity that might influence study outcomes.
Stress. The Perceived Stress Scale 4 (PSS 4)
(Cohen, Kamarck, & Mermelstein, 1983) measures
the degree to which an individual appraises life
situations as stressful. Items are scored from 0 to 5,
with higher scores indicating greater perceived stress.
The PSS 4 is thought adequate when a brief stress mea-
sure is desired (Cohen & Williamson, 1988). In the
general population, a mean of 4.49 (± 2.96) has been
reported. Both the Social Support Questionnaire–6
(SSQ-6) and the PSS 4 have been used in HIV/AIDS
samples; estimated reliabilities in this sample ranged
from 0.82 to 0.84.
Social support. The SSQ-6 (Sarason, Sarason,
Shearin, & Pierce, 1987) is a widely used six-item self-
report that measures perceived access to supportive
others and individual satisfaction with the perceived
support. Higher scores indicate higher levels of per-
ceived access and satisfaction. Across administra-
tions, reliabilities in this sample ranged from 0.77 to
0.91 for the number scale and 0.93 to 0.96 for the satis-
faction scale.
HIV medical history. When interviewing partici-
pants about their current medications and other medi-
cal history, trained ACTU research nurses used stan-
dardized data collection tools that are widely used
across AIDS Clinical Trials Group protocols. Self-
reported symptoms were assessed using an HIV symp-
tom checklist (Smith et al., 2001). Demographics, HIV
risk behaviors, and other drug use patterns were also
self-reported.
Data management and analysis. Investigators
reviewed case record forms for missing and aberrant
data. Data were quality controlled using standard
ACTU procedures and were analyzed using SPSS for
Windows computer program (SPSS 10.0) with an
alpha level of .05 specified prospectively. Chi-square
analyses or two-tailed Fisher’s exact tests and
univariate analysis of variance (ANOVA) models were
used to assess pretreatment group equivalence. Means
and standard deviations were calculated, and distribu-
tions were examined for all dependent variables. Anal-
ysis of covariance (ANCOVA) procedures were used
to assess between-group changes after treatment using
pretest measures as the covariate. Finally, Pearson
product-moment coefficients were used to describe
the relationships between mean scores on dependent
measures.
 Neidig et al. / Aerobic Exercise Training 35

Table 2. Baseline Participant Characteristics Table 3. Baseline Depressive Symptoms/Mood State Scores

Sample Exercise Control Sample Exercise Control

(n = 60) (n = 30) (n = 30) (n = 60) (n = 30) (n = 30)
n % n % n % Center for
Female 8 13 4 13 4 13 Epidemiological
Black 11 18 9 30 2 7 Studies–Depression
Male-Sex-Male 49 82 23 77 26 87 Scale
Heterosexual 8 13 4 13 4 13 Range 0 to 45 0 to 45 0 to 35
Intravenous drug use 3 5 2 7 1 3 Mean 13.2 ± 11.2 12.3 ± 11.8 14.1 ± 10.7
Employed 42 70 19 63 23 77 Score 0-14 39 (65%) 22 (73%) 17 (57%)
College or degree 38 63 19 63 19 63 Score 16-60 21 (35%) 8 (27%) 13 (43%)
Income <$20,000/year 30 50 13 43 17 57 Beck Depression
Any antiretrovirals 47 78 25 83 22 73 Inventory
Protease inhibitors 8 13 4 13 4 13 Range 0-21 0-20 0-21
Antidepressants 12 20 5 17 7 23 Mean 8.8 ± 6.0 8.6 ± 6.0 8.9 ± 5.9
Anti-anxiety agents 9 15 3 10 6 20 Score 0-14 47 (80%) 23 (77%) 24 (83%)
Tobacco 29 48 13 43 16 53 Score 15-63 12 (20%) 7 (23%) 5 (17%)
Alcohol 38 63 22 73 16 53 Profile of Mood State
Marijuana 9 15 6 20 3 10 Range –14 to 107.0 –14 to 107.0 –13.0 to 93.0
Other illicit drugs 6 10 3 10 3 10 Mean 32.8 ± 35.9 34.0 ± 39.3 31.5 ± 32.7
Mean depression 10.9 ± 10.2 10.9 ± 11.8 10.9 ± 8.5
Sample Exercise Control
(n = 60) (n = 30) (n = 30)
M SD M SD M SD
participants using the CES-D reported significant lev-
Age (years) 36.0 7 36.6 6 35.4 7
Mean years HIV-positive 4.3 3 4.0 3 4.6 3 els of depressive symptoms (see Table 3). Because
Mean CD4/mm3 347 87 328 69 365 99 study groups were also balanced on a number of poten-
Social support tially confounding variables (e.g., social support, per-
Number 4.57 3.45 3.84 1.61 5.30 4.54 ceived stress, HIV-related symptoms), these variables
Satisfaction 4.50 1.45 4.24 1.66 4.76 1.17
Perceived stress 6.07 3.51 6.03 3.73 6.10 3.34
were not controlled in subsequent analyses.

Participant Attrition/Assessment of Bias

Results
Sample Description
Baseline demographic and clinical participant char-
acteristics are reported in Table 2. Treatment groups
were balanced on depressive symptoms and all demo-
graphic and clinic characteristics except race. Signifi-
cantly more Blacks were assigned to the exercise con-
dition, χ2(1, N = 60) = 5.45, p = .02. The majority of
study participants were Caucasian men. Most partici-
pants had attended college.
The mean CD4 count for the study sample was 347
± 87 cells/mm3, and on average, participants had
known of their HIV infection for more than 4 years.
More than 75% of the participants were taking anti-
HIV drugs, although protease inhibitor use was not yet
common (used by <15% of participants). At baseline,
20% of the participants using the BDI and 35% of
Overall, 48 (80%) of 60 participants completed the
study. All control participants were retained; however,
12 (40%) of 30 exercisers prematurely discontinued
participation for a variety of reasons. To assess for
bias, exercisers who completed the program were
compared to those who prematurely quit (see Table 4).
Significantly more women (Fisher’s exact test, p = .02)
and African Americans (Fisher’s exact test, p = .01)
prematurely discontinued study participation, and sig-
nificantly more smokers than nonsmokers completed
the 12-week exercise program (Fisher’s exact test, p =
.01). Although we suspected that escalating symptom
burden might contribute to participant attrition, exer-
cisers who completed the study were not significantly
different from noncompleters on reported symptom
distress, and the exercise participants did not report
significantly more symptoms at Week 12 as compared
to study controls (data not reported).
36 JANAC Vol. 14, No. 2, March/April 2003

Table 4. Comparison of Baseline Exercise Participant At both administrations, mean CES-D and BDI
Characteristics scores were highly correlated (r = .71, p = .05).
Completers Noncompleters Pearson product-moment correlations among mean
(n = 18) (n = 12) CES-D, BDI, and scores for the sample are shown in
n % n % Table 1.
Female 0 0 4 33
African American 2 11 7 58
Male-Sex-Male 16 89 7 58 Discussion and Implications
Heterosexual 0 0 4 13 for Research and Practice
Intravenous drug use 1 6 1 8
Employed 11 61 8 67
College or degree 14 78 5 42 This is thought to be the first report on the effects of
Income <$20,000/year 6 33 7 58 aerobic exercise training for depressive symptom
Antiretrovirals 15 83 10 83 management in HIV-infected adults. To overcome lim-
Antidepressants 4 22 1 8 itations of previous studies, we controlled for HIV dis-
Anti-anxiety drugs 2 11 1 8
Tobacco 11 61 2 17 ease stage and used a randomized design and a super-
Alcohol 15 83 7 58 vised exercise intervention based on ACSM guidelines
Marijuana 3 17 3 25 (ACSM, 2000). We also examined concomitant medi-
Other illicit drugs 3 17 0 0 cations and a number of other potentially confounding
Center for Epidemiological
Studies–Depression
or explanatory variables.
Scale score 16-60 5 28 3 25 Only 20% to 35% of participants reported signifi-
Beck Depression Inventory cant levels of depressive symptoms at study entry. Par-
score 15-63 4 22 3 25 ticipants who completed the 12-week aerobic exercise
program reported improvements in depressive symp-
Completers Noncompleters
(n = 18) (n = 12) toms and depressed mood as compared to a control
M SD M SD group that maintained usual activities. Significant
improvements were detected on the less somatic
Age 38 6 37 6
Mean years HIV-positive 4 3 5 3 depressive symptom index (CES-D) and on the POMS
CD4/mm3 332 69 322 73 depression subscale. These findings are consistent
Social support with the improved mood previously observed in HIV-
Number 3.84 1.57 3.85 1.75 positive male exercisers (LaPerriere et al., 1990, 1991)
Satisfaction 4.07 1.70 4.49 1.64
Perceived stress 6.05 4.01 6.00 3.48
and add to the overall body of research linking aerobic
Profile of Mood State exercise and improved mood. Greater improvement on
Mean 39.22 33.61 31.53 32.68 the CES-D scale may be expected because the BDI is
Mean depression 11.16 10.24 10.93 8.57 generally used in clinical treatment trials as opposed to
cohort studies, in which the CES-D is intended as the
first-stage screen in a two-step diagnostic process.
Limited change on the BDI may also reflect scale inad-
Psychological Outcomes equacies in this population (Drebing et al., 1994), low
observed power, or both.
Symptom improvement was noted in the exercise The numbers of women and Blacks in this study
group on both indices. Group means and ANCOVA were modest, and randomization did not yield racially
results for dependent measures are reported in Table 5. balanced treatment groups. Although our participant
As compared to study controls, there was significant retention compared favorably with rates in other exer-
symptom improvement using the CES-D (p = .03) and cise studies enrolling HIV-infected participants
a trend to symptom improvement on the BDI (p = .06). (LaPerriere et al., 1991; MacArthur et al., 1993;
As a group, exercisers also showed significant (p = .01) Stringer et al., 1998), the disproportionate loss of 40%
improvements on overall POMS scores and on the of the exercise participants limits study findings.
depression/dejection subscale (p = .05). Despite significant retention efforts, few Black men
 Neidig et al. / Aerobic Exercise Training 37

Table 5. Group Means and Analysis of Covariance

Exercise Control
Week 0 Week 12 Week 0 Week 12 F p
Center for Epidemiological Studies–Depression Scale
n 30 18 30 30
M ± SD 12.3 ± 11.8 7.2 ± 7.1 14.1 ± 10.7 14.1 ± 11.3 5.16 .028*
Beck Depression Inventory
n 30 18 29 30
M ± SD 8.6 ± 6.0 5.6 ± 6.3 8.9 ± 5.9 8.7 ± 7.1 3.61 .064
Profile of Mood State
n 30 18 30 30
Total (M ± SD) 34.0 ± 39.3 12.2 ± 28.3 31.5 ± 32.7 32.3 ± 40.0 7.04 .011
Depression 10.9 ± 11.8 6.1 ± 8.9 10.9 ± 8.5 10.9 ± 11.2 4.27 .045*

NOTE: Significance indicated with an asterisk.

and no Black women completed the randomized exer-
cise intervention. Individuals who were lost from the
study were often among the working poor and reported
abrupt changes in employment, unreliable transporta-
tion, and increased family responsibilities as reasons
for premature study discontinuation. Most exercisers
who completed the study were Caucasian men who
reported homosexual or bisexual HIV risk factors, and
all had CD4 cell counts ranging from 200 to 500/mm3.
Researchers should budget for personnel and services
that support participant retention in future exercise
studies. Transportation, child care, and other services
may be needed to retain less economically advantaged
participants.
Although it would have been ideal to exclude per-
sons taking antidepressants, only persons on stable
drug regimens were enrolled in this study. The use of
randomization did balance groups on antidepressant
use and should have also controlled for disparate life
events, depressive symptoms that may have spontane-
ously remitted, and differences in concurrent psycho-
therapy. Because social reinforcement may ameliorate
depression, the addition of an attention-only control
group may have helped us to examine the influence of
social interactions. However, it is worth noting that
this study’s exercise intervention did not use a group
format, as was the case in LaPerriere and colleagues’
(1991) trial. Moreover, in this sample, exercisers did
not differ significantly from control participants on
perceived social support. Although exercisers had per-
sonal supervision during each session, the exercise
facility was similar to the “real-life” setting of a health
club or gym. Some sessions were very crowded, but at
other times, exercise participants worked out individu-
ally. Although this and all studies of exercise must nec-
essarily use convenience samples, some may contend
that exercise participants were more motivated to per-
form because they were research participants. Some
participants may have been motivated to continue in
the exercise program longer than they might have if
they had not been involved in a research study.
Findings of psychological benefit should also be
interpreted cautiously because the study attracted few
participants with high levels of depressive symptoms.
Because 65% to 80% of participants had normal
depressive symptom scores at baseline, it is difficult to
evaluate the clinical relevance of these findings. Few
exercise studies have reported pretest BDI and posttest
BDI data; however, the magnitude of CES-D symptom
reduction we observed in this study is comparable to
that observed by Chesney, Folkman, and Chambers
(1996) following a coping effectiveness training pro-
gram. Moreover, any symptom improvement is desir-
able as even low symptom burden represents distress
and suffering. Preventing or reducing depressive
symptom burden over the lengthening course of
chronic HIV infection may be very important for
adherence and quality of life.
As depressive symptom measures, both self-report
measures worked well, and scores were highly corre-
lated providing evidence of convergent validity (r =
.71). The finding that the CES-D and BDI have conver-
gent validity should help readers better integrate the
findings of prior studies that used one of these
38 JANAC Vol. 14, No. 2, March/April 2003
symptom measures. These study findings also provide
preliminary support for using the POMS depression
subscale as a measure of depressed mood in HIV-
infected persons.
Symptom management is an essential but often
neglected component of HIV care. Because persons
living with HIV infection often experience depressive
symptoms over the course of their chronic HIV illness,
these symptoms warrant systematic study to deter-
mine the best management strategies. This study pro-
vides preliminary evidence that a program of moder-
ately intense aerobic exercise might be an effective
approach to preventing or reducing prevalent symp-
toms of depression in persons with HIV infection.
Additional research is needed to determine whether
exercise can benefit HIV-infected persons with signifi-
cantly higher levels of depressive symptoms. Future
researchers should also consider including a social
attention control group (in addition to a no-treatment
control group).
The differential loss of participants from the exer-
cise treatment group reinforces the general finding that
exercise adherence is very difficult for many people. In
general, a 50% drop-out rate within 6 months of start-
ing a physician-prescribed or self-selected supervised
exercise program is common (Dishman, 1987).
Because of the considerable personal effort required
for exercise regimen adherence, it is important for
researchers to determine the optimal frequency, inten-
sity, and duration for promoting health in this popula-
tion. By examining the relationships among heart rates
during exercise training and exercise attendance,
researchers may be able to more completely evaluate
possible mediators of improved psychological
outcomes.
A growing literature suggests that regular exercise
is safe and does not significantly change CD4 cell
counts or HIV-RNA copies. Therefore, nurse clini-
cians can recommend habitual exercise for their HIV-
infected patients after a health appraisal as is sug-
gested for all persons before beginning an exercise
program (ACSM, 2000). Our study enjoyed strong
community interest, but our trainers were impressed at
how excellent recruitment and retention rates waned
dramatically after the introduction of more effective
anti-HIV drug cocktails. Perhaps more recent con-
cerns about HIV-related fat redistribution, metabolic
changes, and potential cardiac risks (Carr et al., 1998;
Henry, Melroe, Huebsch, Kopaczewski, & Simpson,
1998) will lead to renewed enthusiasm for exercise
training among some persons on anti-HIV drug regi-
mens. Nurses may find the current HIV treatment
environment may present excellent opportunities to
encourage health-promoting behaviors that include
regular physical activity.
Acknowledgments
This research was funded by the National Institutes
of Health, National Institute of Nursing Research (NR
03974), the National Institute of Allergy and Infec-
tious Diseases (Adult AIDS Clinical Trials Group,
A125924), and The Ohio State University College of
Nursing. The authors gratefully acknowledge the
efforts of the study participants and thank Michael
McDonald, Dylan Wessman, Jeremy Johnson, the
ACTU staff, and Dr. Jennie Nickel for their important
contributions. In remembrance we also acknowledge
Dr. Robert J. Fass for his encouragement and thought-
ful manuscript reviews.
References
American College of Sports Medicine. (2000). ACSM’s guidelines
for exercise testing and prescription (6th ed.). Philadelphia:
Lippincott, Williams & Wilkins.
Beck, A. T., & Steer, R. A. (1993). Beck depression inventory man-
ual. San Antonio, TX: Psychological Corporation, Harcourt
Brace & Co.
Beck, A. T., Ward, C. H., Mendelson, M., Mock, J., & Erbaugh, J.
(1961). An inventory for measuring depression. Archives of
General Psychiatry, 4, 561-571.
Bing, E. G., Burnam, A., Longshore, D., Fleishman, J. A.,
Sherbourne, C. D., London, A. S., et al. (2001). Psychiatric dis-
orders and drug use among human immunodeficiency virus-
infected adults in the United States. Archives of General Psy-
chiatry, 58, 721-728.
Bonfanti, P., Valsecchi, L., Parazzini, F., Carradori, S., Pusterla, L.,
Fortuna, P., et al. for the Coordinamento Italiano Studio
Allergia E Infezione Da HIV (CISAI) Group. (2000). Inci-
dence of adverse reactions in HIV patients treated with protease
inhibitors: A cohort study. Journal of Acquired Immune Defi-
ciency Syndromes, 23, 236-245.
Brashers, D. E., Neidig, J. L., Cardillo, L. W., Dobbs, L. K., Rus-
sell, J. A., & Haas, S. M. (1999). “In an important way, I did die:
Uncertainty and revival in persons living with HIV or AIDS.”
AIDS Care, 11, 201-219.
Brown, G. R., Rundell, J. R., McManis, S. E., Kendall, S. N.,
Zachary, R., & Temoshok, L. (1992). Prevalence of psychiatric

disorders in early stages of HIV infection. Psychosomatic Med-
icine, 54, 588-601.
Burack, J. H., Barrett, D. C., Stall, R. D., Chesney, M. A., Ekstrand,
M. L., & Coates, T. J. (1993). Depressive symptoms and CD4
lymphocyte decline among HIV-infected men. Journal of the
American Medical Association, 270, 2568-2573.
Carr, A., Samaras, K., Burton, S., Law, M., Freund, J., Chisholm,
D. J., et al. (1998). A syndrome of peripheral lipodystrophy,
hyperlipidaemia, and insulin resistance in patients receiving
HIV protease inhibitors. AIDS, 12, F51-58.
Centers for Disease Control and Prevention. (1993). Revised clas-
sification system for HIV infection and expanded surveillance
case definition for AIDS among adolescents and adults. Mor-
bidity and Mortality Weekly Report, 41.
Chesney, M., Folkman, S., & Chambers, D. (1996). Coping effec-
tiveness training for men living with HIV: Preliminary find-
ings. International Journal of STD & AIDS, 7, 75-82.
Chesney, M. A., Morin, M., & Sherr, L. (2000). Adherence to HIV
combination therapy. Social Science & Medicine, 50, 1599-
1605.
Cohen, S., Kamarck, T., & Mermelstein, R. (1983). A global mea-
sure of perceived stress. Journal of Health and Social Behavior,
24, 385-396.
Cohen, S., & Williamson, G. M. (1988). Perceived stress in a prob-
ability sample of the United States. In S. Spacapan & S.
Oskamp (Eds.), The social psychology of health (pp. 31-67).
Thousand Oaks, CA: Sage.
Crepaz, N., & Marks, G. (2001). Are negative affective states asso-
ciated with HIV sexual risk behaviors? A meta-analytic review.
Health Psychology, 20, 291-299.
DeGenova, M. K., Patton, D. M., Jurich, J. A., & Macdermid, S. M.
(1994). Ways of coping among HIV-infected individuals. Jour-
nal of Social Psychology, 134, 655-663.
Dishman, R. K. (1985). Medical psychology in exercise and sport.
Medical Clinics of North America, 69, 123-143.
Dishman, R. K. (1987). Exercise adherence and habitual physical
activity. In W. P. Morgan & S. E. Goldston (Eds.), Exercise and
mental health (pp. 57-83). Washington, DC: Hemisphere.
Drebing, C. E., Van Gorp, W. G., Hinkin, C., Miller, E. N., Satz, P.,
Kim, D. S., et al. (1994). Confounding factors in the measure-
ment of depression in HIV. Journal of Personality Assessment,
62, 68-83.
Evans, S., Ferrando, S., Sewell, M., Goggin, K., Fishman, B., &
Rabkin, J. (1998). Pain and depression in HIV illness.
Psychosomatics, 39, 528-535.
Ferrando, S., Evans, S., Goggin, K., Sewell, M., Fishman, B., &
Rabkin, J. (1998). Fatigue in HIV illness: Relationship to
depression, physical limitations and disability. Psychosomatic
Medicine, 60, 759-764.
Gordillo, V., del Amo, J., Soriano, V., & Gonzalez-Lahoz, J.
(1999). Sociodemographic and psychological variables influ-
encing adherence to antiretroviral therapy. AIDS, 13, 1763-
1769.
Greene, K. B., Berger, J., Reeves, C., Moffat, A., Standish, L., &
Calabrese, C. (1999). Most frequently used alternative and
complementary therapies and activities by participants in the
Neidig et al. / Aerobic Exercise Training 39
AMCOA study. Journal of the Association of Nurses in AIDS
Care, 10, 60-73.
Greist, J. H., Klein, M. H., Eischens, R. R., Faris, J., Gurman, A. S.,
& Morgan, W. P. (1979). Running as treatment for depression.
Comprehensive Psychiatry, 20, 41-53.
Hays, R. D., Wells, K. B., Sherbourne, C. D., Rogers, W., & Sprit-
zer, K. (1995). Functioning and well-being outcomes of
patients with depression compared with chronic general medi-
cal illnesses. Archives of General Psychiatry, 52, 11-19.
Henry, K., Melroe, H., Huebsch, J., Kopaczewski, J., & Simpson,
J. (1998). Astorvastatin and gemfibrozil for protease-inhibitor
related lipid abnormalities. Lancet, 352, 1031-1032.
Kalichman, S. C. (1999). Psychological and social correlates of
high-risk sexual behavior among men and women living with
HIV/AIDS. AIDS Care, 4, 415-428.
Kalichman, S. C., & Weinhardt, L. (2001). Negative affect and sex-
ual risk behavior: Comment on Crepaz and Marks (2001) [Edi-
torial comment]. Health Psychology, 20, 300-301.
Kessler, R., McGonale, K., Zhao, S., Nelson, C., Hughes, M.,
Eshleman, S., et al. (1994). Lifetime and 12-month prevalence
of DSM-III-R psychiatric disorders in the United States:
Results from the National Comorbidity Survey. Archives of
General Psychiatry, 51, 8-19.
LaPerriere, A. R., Antoni, M. H., Schneiderman, N., Ironson, G.,
Klimas, N., Caralis, P., et al. (1990). Exercise intervention alle-
viates emotional distress and natural killer cell decrements fol-
lowing notification of positive serologic status for HIV-1. Bio-
feedback and Self-Regulation, 15, 229-242.
LaPerriere, A., Fletcher, M. A., Antoni, M. H., Klimas, N. G.,
Ironson, G., & Schneiderman, N. (1991). Aerobic exercise
training in an AIDS risk group. International Journal of Sports
Medicine, 12, S53-S57.
Levine, J. M. (2001). Psychiatric aspects for HIV care. AIDS Clini-
cal Care, 13, 101-108.
Lipsitz, J. D., Williams, J. B., Rabkin, J. G., Remien, R. H.,
Bradbury, M., el Sadr, W., et al. (1994). Psychopathology in
male and female intravenous drug users with and without HIV
infection. American Journal of Psychiatry, 151, 1662-1668.
Low-Beer, S., Chan, K., Yip, B., Wood, E., Montaner, J. S. G.,
O’Shaughnessy, M. V., et al. (2000). Depressive symptoms
decline among persons on HIV protease inhibitors. Journal of
Acquired Immune Deficiency Syndromes, 23, 295-301.
Lox, C. L., McAuley, E., & Tucker, R. S. (1995). Exercise as an
intervention for enhancing subjective well-being in an HIV-1
population. Journal of Sport & Exercise Psychology, 17, 345-
362.
Lyketsos, C. G., Hoover, D. R., Guccione, M., Senterfitt, W., Dew,
M. A., Wesch, J., et al. (1993). Depressive symptoms as predic-
tors of medical outcomes in HIV infection. Journal of the
American Medical Association, 270, 2563-2567.
MacArthur, R. D., Levine, S. D., & Birk, T. J. (1993). Supervised
exercise training improves cardiopulmonary fitness in HIV-
infected persons. Medicine and Science in Sports & Exercise,
25, 684-688.
Marzuk, P. M., Tierney, H., Tardiff, K., Gross, E. M., Morgan,
E. B., Hsu, M. A., et al. (1988). Increased risk of suicide in
40 JANAC Vol. 14, No. 2, March/April 2003
persons with AIDS. Journal of the American Medical Associa-
tion, 259, 1333-1337.
Mayne, T. J., Vittinghoff, E., Chesney, M. A., Barrett, D. C., &
Coates, T. J. (1996). Depressive affect and survival among gay
and bisexual men infected with HIV. Archives of Internal Medi-
cine, 156, 2233-2238.
McNair, D., Lorr, M., & Droppleman, L. (1971). Profile of mood
states. San Diego, CA: EdITS/Educational and Industrial
Testing Service.
McNair, D., Lorr, M., & Droppleman, L. (1992). EdITS manual for
the profile of mood states (Rev. ed.). San Diego, CA: EdITS/
Educational and Industrial Testing Service.
Morgan, W. P., & Goldston, S. E. (1987). Exercise and mental
health. Washington, DC: Hemisphere.
Ostrow, D. G., Monjan, A., Joseph, J., Vanraden, M., Fox, R.,
Kingsley, L., et al. (1989). HIV-related symptoms and psycho-
logical functioning in a cohort of homosexual men. American
Journal of Psychiatry, 146, 737-742.
Paffenbarger, R. S., Jr., Wing A. L., & Hyde, R. T. (1978). Physical
activity as an index of heart attack risk in college alumni. Amer-
ican Journal of Epidemiology, 108, 161-174.
Perkins, D. O., Leserman, J., Stern, R. A., Baum, S. F., Liao, D.,
Golden, R. N., et al. (1995). Somatic symptoms and HIV infec-
tion: Relationship to depressive symptoms and indicators of
HIV disease. American Journal of Psychiatry, 152, 1776-1781.
Rabkin, J. G. (1997). Meeting the challenge of depression in HIV.
GMHC Treatment Issues: Newsletter of Experimental AIDS
Therapies, 11(10), 6-7.
Rabkin, J., & Ferrando, S. (1997). A “second life” agenda: Psychi-
atric research issues raised by protease inhibitor treatments for
people with human immunodeficiency virus or the acquired
immunodeficiency syndrome. Archives of General Psychiatry,
54, 1049-1053.
Rabkin, J. G., Johnson, J., Lin, S. H., Lipsitz, J. D., Remien, R. H.,
Williams, J. B., et al. (1997). Psychopathology in male and
female HIV-positive and negative injecting drug users: Longi-
tudinal course over 3 years. AIDS, 11, 507-515.
Rabkin, J. G., Wagner, G. J., & Rabkin, R. (1999). Fluoxetine treat-
ment for depression in patients with HIV an AIDS: A random-
ized, placebo-controlled trial. American Journal of Psychiatry,
156, 101-107.
Radloff, L. S. (1977). The CES-D scale: A self-report depression
scale for research in the general population. Applied Psycho-
logical Measurement, 1, 385-401.
Rigsby, L. W., Dishman, R. K., Jackson, A. W., Maclean, G. S., &
Raven, P. B. (1992). Effects of exercise training on men sero-
positive for the human immunodeficiency virus-1. Medicine
and Sciences in Sports and Exercise, 24(1), 6-12.
Rosenberger, P. H., Bornstein, R. A., Nasrallah, H. A., Para, M. F.,
Whitacre, C. C., Fass, R. J., et al. (1993). Psychopathology in
human immunodeficiency virus infection: Lifetime and current
assessment. Comprehensive Psychiatry, 34, 150-158.
Sambamoorthi, U., Walkup, J., Olfson, M., & Crystal, S. (2000).
Antidepressant treatment and health services utilization among
HIV-infected Medicaid patients diagnosed with depression.
Journal of General Internal Medicine, 15, 311-320.
Sarason, I. G., Sarason, B. R., Shearin, E. N., & Pierce, G. R.
(1987). A brief measure of social support: Practical and theo-
retical implications. Journal of Social and Personal Relation-
ships, 4, 497-510.
Sherbourne, C. D., Hays, R. D., Fleishman, J. A., Vitiello, B.,
Magruder, K. M., Bing, E. G., et al. (2000). Impact of psychiat-
ric conditions on health-related quality of life in persons with
HIV infection. American Journal of Psychiatry, 157, 248-254.
Singh, N., Berman, M., Swindells, S., Justis, J. C., Moha, J. A.,
Squier, C., et al. (1999). Adherence of human immunodefi-
ciency virus-infected patients to antiretroviral therapy. Clinical
Infectious Diseases, 29, 824-830.
Singh, N., Squier, C., Sivek, C., Wagener, M. W., & Yu, V. L.
(1997). Psychological stress and depression in older patients
with intravenous drug use and human immunodeficiency virus
infection: Implications for intervention. International Journal
of STD & AIDS, 8, 251-255.
Smith, B. A., Neidig, J. L., Nickel, J. T., Mitchell, G. L., Para,
M. F., & Fass, R. J. (2001). Aerobic exercise: Effects on param-
eters related to fatigue, dyspnea, weight and body composition
in HIV-infected adults. AIDS, 15, 693-701.
SPSS for Windows (Version 10.0). [Computer software]. Chicago,
IL: SPSS, Inc.
Stringer, W. W., Berezovskaya, M., O’Brien, W. A., Beck, C. K., &
Casaburi, R. (1998). The effect of exercise training on aerobic
fitness, immune indices, and quality of life in HIV+ patients.
Medicine and Science in Sports and Exercise, 30, 11-16.
Tkachuk, G. A., & Martin, G. L. (1999). Exercise therapy for
patients with psychiatric disorders: Research and clinical
implications. Professional Psychology: Research and Practice,
30, 275-282.
UK Collaborative Group on Monitoring the Transmission of HIV
Drug Resistance. (2001). Analysis of prevalence of HIV-1 drug
resistance in primary infections in the United Kingdom. British
Medical Journal, 322, 1087-1088.
Vanhove, G. F., Schapiro, J. M., Winters, M. A., Merigan, T. C., &
Blaschke, T. F. (1996). Patient compliance and drug failure in
protease inhibitor monotherapy [Letter]. Journal of the Ameri-
can Medical Association, 276, 1955-6.
Wagner, G., Rabkin, J., & Rabkin, R. (1998). Exercise as a media-
tor of psychological and nutritional effects of testosterone ther-
apy in HIV+ men. Medicine and Science in Sports and Exer-
cise, 30, 811-817.
Weissman, M. M., Prusoff, B., & Newberry, P. (1975). Compari-
son of the CES-D with standardized depression rating scales at
three points in time. Yale University Technical Report. New
Haven, CT: Yale University and the National Institute of Men-
tal Health.
Williams, J. B., Rabkin, J. G., Remien, R. H., Gorman, J. M., &
Ehrhardt, A. A. (1991). Multidisciplinary baseline assessment
of homosexual men with and without human immunodefi-
ciency virus infection, II: Standardized clinical assessment of
current and lifetime psychopathology. Archives of General
Psychiatry, 48, 124-130.