Psychosomatics
Journal Home Search Current Issue Past Issues Subscribe All APPI Journals Help Contact Us
 
Quicksearch
Advanced Search
Or Search All APPI Journals
This Article
* Abstract Freely available
* Full Text (PDF)
* Alert me when this article is cited
* Alert me if a correction is posted
* Citation Map
Services
* Email this article to a Colleague
* Similar articles in this journal
* Similar articles in PubMed
* Alert me to new issues of the journal
* Add to My Articles & Searches
* Download to citation manager
* reprints & permissions
Citing Articles
* Citing Articles via HighWire
* Citing Articles via Google Scholar
Google Scholar
* Articles by Barroso, J.
* Articles by Evans, D. N.
* Search for Related Content
PubMed
* PubMed Citation
* Articles by Barroso, J.
* Articles by Evans, D. N.
Related Collections
* AIDS/HIV
* Depression
Psychosomatics 43:317-325, August 2002
© 2002 The Academy of Psychosomatic Medicine

Predicting Fatigue and Depression in HIV-Positive Gay Men

Julie Barroso, Ph.D., A.N.P., C.S., John S. Preisser, Ph.D., Jane Leserman, Ph.D., Bradley N. Gaynes, M.D., Robert N. Golden, M.D., and Dwight N. Evans, M.D.

Received October 4, 2001; revised January 30, 2002; accepted February 13, 2002. From the School of Nursing and the Departments of Biostatistics and Psychiatry, University of North Carolina at Chapel Hill; and the Departments of Psychiatry, Medicine, and Neuroscience, University of Pennsylvania School of Medicine, Philadelphia. Address correspondence and reprint requests to Dr. Barroso, School of Nursing, Carrington Hall, CB# 7460, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-7460; julie_barroso{at}unc.edu (e-mail).


  ABSTRACT

 
TOP
 ABSTRACT
INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
HIV-related fatigue is a prevalent and troubling symptom for HIV-positive people. The purpose of the study was to develop a model for predicting fatigue and depression among HIV-positive gay men as a function of history of fatigue and depression in the previous year and to determine whether psychological and psychosocial variables or physiologic variables better predict fatigue. Data from 96 HIV-positive gay men followed longitudinally for up to 7.5 years were used to develop logistic regression models for predicting fatigue and depression. Fatigue was predicted by both physiologic and psychological risk factors, whereas depression was predicted by only psychological risk factors.

Key Words: Depression • Fatigue • HIV-Positive

  INTRODUCTION

 
TOP
 ABSTRACT
 INTRODUCTION
METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
With the advent of therapies that have been proven to prolong life in people with HIV infection, attention has become focused on symptom management. One of the most problematic of HIV-related symptoms is fatigue. The incidence of fatigue in people with HIV infection ranges from 20% to 85%.1–5 Fatigue is the most common symptom mentioned in both self-reports and provider-reports concerning HIV-positive patients.3,6 Fatigue has been associated with greater functional limitation,6,7 and even with diminished survival.6 Darko et al.8 found that HIV-infected patients were significantly more likely than noninfected subjects to be unemployed, to feel fatigued more hours of the day, to sleep and nap more, and to have diminished midmorning alertness. Seropositive women have reported that fatigue resulted in a lack of stamina, frequent absenteeism, and lowered quality of work; many feared they might lose their jobs as a result.9 Some researchers have speculated that fatigue is related to depression, given the high rate of depressive disorders among seropositive people (22%-45%).10 One task for researchers in this area is to determine the factors contributing to fatigue in people with HIV infection. The principal goal of this study was to develop a model for clinical prediction of fatigue and depression in HIV-positive individuals as a function of the individual's history of fatigue and depression in the previous year. We used observation-driven logistic regression models, and our aim was to develop a model to aid clinicians, not to make inferences about populations.

The conceptual framework for this study was drawn from two areas of research on HIV-related fatigue: the search for psychological and for physiologic correlates of fatigue. Barroso11,12showed that depression is consistently related to fatigue in seropositive subjects. This relationship remained significant even when the analysis accounted for the use of antidepressants and when items measuring somatic symptoms were removed from the instrument used to assess depression. Another study found that increased fatigue was associated with increased depression, but not with HIV disease progression;5 the authors speculated that fatigue in otherwise asymptomatic patients might have psychological rather than physiological causes. Perkins et al.13 examined the relationship of the somatic symptoms of fatigue and insomnia with indicators of both psychiatric disturbance and HIV disease severity during a 6-month period. At study entry, complaints of fatigue and insomnia were associated with dysphoric mood, major depression, and other non-HIV-related symptoms of major depression but not with CD4 cell counts. Increases in levels of fatigue and insomnia over the 6-month follow-up were associated with increases in non-HIV-related symptoms of depression and in severity of dysphoric mood but not with HIV disease progression. The authors concluded that complaints of fatigue and insomnia in otherwise asymptomatic HIV-infected patients are likely to be related to psychological disturbances and possibly major depression, which can be treated. Breitbart et al.1 showed that HIV-positive patients with fatigue had significantly more depressive symptoms than those without fatigue. Despite these findings, the authors concluded that not all fatigued patients with AIDS were depressed and that fatigue was not necessarily a reflection of underlying depression. Others have found that, in advanced HIV illness, fatigue seems to be a chronic symptom that could be associated with depressive symptoms and disorders but was not merely a symptom of depression. Ferrando et al.14 concluded that fatigue in HIV illness may exist with or without a depressive disorder and may require treatment in its own right. For some HIV-positive men, even when depressed mood was alleviated, the problem of fatigue remained.15,16 Thus, the nature of the relationship between fatigue and depression remains unclear. Most of the studies noted here are limited because they were cross-sectional in design, focused on describing relationships among variables in a population, and did not provide a view of changes in fatigue and depression in HIV-infected individuals over time. So the dilemma remains: is the patient depressed and experiencing fatigue primarily as a result of the depression, or is the patient fatigued and then becomes depressed as a result of the impact of fatigue on his or her life? Furthermore, do these temporal relationships exist in analyses that control for other psychological and physiological variables?

Other psychosocial problems may also be related to fatigue. Lack of social support, or dissatisfaction with social support, has been associated with disease progression in people with HIV infection.17,18 For people with HIV-related fatigue, this association may have special significance because they may have no one to assist them (i.e., cook meals, help with parenting) when they are fatigued. Fatigue is also associated with anxiety states and disorders. Major classification systems include fatigue among the criteria for certain anxiety disorders, and numerous studies have shown a relationship between anxiety and fatigue.19 Hopelessness, one of the core characteristics of depression, may also play a role in HIV-related fatigue. Breitbart et al.1 showed that people with HIV-related fatigue had greater hopelessness than those without fatigue.

The psychological factors that may play a role in HIV-related fatigue cannot be properly evaluated without considering the physiologic causes as well. People with HIV infection have exhibited decline in functioning over a 1-year period that was related not only to increasing disease severity but also to prior reports of fatigue and poor self-rated health.7 HIV-related fatigue was associated with significantly poorer physical functioning and was a strong predictor of both limitations of daily activity and number of disability days.1,20,21 Fatigue was also found to be an important factor in the development of physical limitations and disability in ambulatory men with advanced HIV illness.14 In studies involving laboratory measures, some researchers have found no relationship between immune functioning, as measured by CD4 count, and fatigue.1,4,6,11,13,22,23 However, Walker et al.5 found that greater fatigue was associated with lower CD4 counts. Also, Lee et al.24 found that in women with HIV, lower CD4 counts were related to more daytime sleep and higher levels of evening and morning fatigue. Darko et al.8 found a significant inverse correlation between CD4 count and hours of daily fatigue. In the only study, to our knowledge, to have measured both fatigue and HIV viral load, Ferrando et al.14 found no correlation between fatigue and HIV RNA.

Anemia, the most common hematologic abnormality in patients with HIV, may also be related to fatigue. Anemia may occur as a result of HIV-induced ineffective hematopoiesis, opportunistic infections, infiltrative disease of the bone marrow, nutritional deficiencies, hemolysis, or antiretroviral or other therapy.25

The purpose of this study was to examine fatigue and its temporal relationship to psychological and physiological variables in a cohort of HIV-positive gay men. This study was anchored by the previously mentioned study by Perkins et al.,13 which involved the same cohort and reported initial data on these and similar variables. Longitudinal data from the ongoing Coping in Health and Illness Project were collected at 6-month intervals for up to 7.5 years. The first aim was to determine the predictive value of each variable (i.e., fatigue or depression) in relation to the other. A second aim was to determine whether psychological variables (e.g., depression, anxiety, and perceived lack of social support) predicted fatigue better than physiologic variables (e.g., CD4 count, Centers for Disease Control [CDC] clinical status, and hemoglobin and hematocrit levels).


  METHODS

 
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
Research Design
Data were collected in North Carolina as part of an ongoing, longitudinal study, the Coping in Health and Illness Project. The study was approved by the University of North Carolina School of Medicine Committee for the Protection of the Rights of Human Subjects. All subjects provided written informed consent. The study group included 96 HIV-infected gay men, recruited from 1990 to 1992 from rural and urban areas of North Carolina. All subjects were clinically asymptomatic (CDC clinical status A)26 at study entry. The men were assessed at 6-month intervals for 7.5 years (15 visits). All subjects had at least two visits. All volunteers were screened to exclude from initial participation those with 1) less than 10 years of education, 2) age less than 18 or greater than 51 years, 3) previous intravenous drug use, 4) significant medical illness (e.g., heart, lung, or kidney disease), 5) preexisting neurological disorder or trauma (e.g., head injury, stroke, seizures), 6) past treatment for alcoholism or current heavy alcohol consumption (60 drinks per month), 7) present or past heavy recreational drug use, 8) use of zidovudine or other antiretroviral medications, and 9) HIV-related symptoms, i.e., those meeting the 1987 CDC criteria for AIDS or ARC (e.g., night sweats, herpes zoster, oral candidiasis, hairy leukoplakia, shingles, unexplained temperature or diarrhea, or unexplained weight loss or fatigue in the presence of other symptoms). These criteria were in place at the initiation of the study.

Procedure
The subjects were evaluated every 6 months in the General Clinical Research Center at the University of North Carolina Hospital in Chapel Hill. The subjects underwent systematic medical, neurological, neuropsychological, and psychiatric assessments performed by specialists in these areas. The subjects were seen by the first author (J.B.) as a nurse practitioner for the last 3 years in the study period. At each visit, blood samples were obtained for measurement of CD4 counts as well as many other physiologic variables.

Measures
Data on several physiologic and psychological variables were collected at each visit. For an initial assessment, we chose from the parent data set those variables that are noted to have a possible effect on fatigue. The physiologic variables were CD4 count, CDC clinical status (clinical status A designates HIV-positive but asymptomatic patients, B designates HIV-positive symptomatic patients, and C designates those with AIDS or stage IV of HIV disease), hemoglobin level, and hematocrit level. The psychological variables were anxiety, hopelessness, social conflict, and satisfaction with support (a negative correlation). Additional variables that were considered but not discussed further because they did not remain in the final model were serum iron level, vitamin B12 level, CD8 count, and level of thyroid functioning.

The two constructs of particular interest, depressed mood and fatigue, were assessed with the corresponding Profile of Mood States (POMS) subscales.27 The POMS, a widely tested and well-validated instrument, is a 65-item self-rating scale measuring six emotional states (tension-anxiety, depression-dejection, anger-hostility, vigor, fatigue, and confusion) and total mood disturbance. Each item is rated on a scale from 0, "not at all," to 4, "extremely." The depression-dejection and fatigue subscales were of primary interest and were treated as dichotomous response variables. Fatigue was defined as present if the POMS fatigue score was at least 11 and was defined as absent otherwise. Depression was defined as present if the POMS depression score was at least 15 and was defined as absent otherwise. The cutoff scores were based on the instrument manual and previous research.28–33 Anxiety was measured with the POMS anxiety subscale. Although other measures of depression were used in the Coping in Health and Illness Project, the POMS depression subscale was used in this analysis for two reasons. First, we wanted a mood measure that was free of somatic symptoms. Second, we were using the fatigue subscale of the POMS as the fatigue measure, and we believed that the POMS depression subscale was the most appropriate and comparable measure.

Hopelessness was measured with the Beck Hopelessness Scale.34 This scale consists of 20 items answered true or false that reflect the subject's negative expectancies. Scores can range from 0 to 20. The social conflict measure was adapted from a questionnaire on coping and change used in the Multicenter AIDS Cohort Study.35 The social conflict scale is made up of seven equally weighted items that ask, for the past month, "How have things been going between you and the people in your personal life—lovers, friends, relatives, etc.?" Examples of scale items include the following questions: 1) "Did you feel that the people in your life let you down by not showing you as much love and concern as you would have liked?" 2) "Have the people in your personal life really gotten on your nerves?" 3) "Did people in your personal life make you feel respected?" Item responses can range from 1, "no" or "never," to 5, "yes" or "all the time." Higher scores indicate more conflict. Cronbach's {alpha} for the social conflict scale was 0.84. Finally, to assess the degree of satisfaction with social support, we administered the Sarason Brief Social Support Questionnaire36 once a year (Cronbach's {alpha} at baseline=0.89). Scale scores can range from 1, very dissatisfied with the support received, to 6, very satisfied. Support scores from the previous yearly interval were used to estimate support scores at each 6-month visit. The correlations between scores obtained 1 year apart ranged from 0.40 to 0.68. The Sarason Brief Support Questionnaire allows for quantification of the number of support persons.

Statistical Methods
The goal of the study was to describe quantitatively transitions in states of depression and fatigue in HIV-positive men by using observation-driven models.37,38 We modeled the conditional probabilities of fatigue and depression given past fatigue and depressed states as well as other covariates (CD4 count, CDC clinical status, hemoglobin and hematocrit levels, anxiety, hopelessness, social conflict, and satisfaction with support) measured 6 months earlier. Two logistic regression models with past outcomes considered as covariates were specified:

1) logit[Pr(Ft=1)] = xTt-1ß1 + {alpha}0dt-1 + {alpha}1ft-1 + {alpha}2ft-2 + {alpha}12ft-1 x ft-2, and

2) logit[Pr(Dt=1)] = xTt-1ß2 + {theta}0ft-1 + {theta}1dt-1 + {theta}2dt-2 + {theta}12dt-1 x dt-2

where Ft=1 if the subject had fatigue at the t-th visit, and Ft=0 if he did not have fatigue. Similarly, Dt=1 if the subject had depression at time t, and Dt=0 if he did not have depression. On the right side of the equations, the subscript t-1 indicates covariates, or response variables, measured at the last visit 6 months earlier, and t-2 refers to two visits ago or 1 year earlier. Model 1 specified that the probability of fatigue may be predicted by a subject's depressed state measured 6 months earlier after adjusting for other covariates measured 6 months earlier (denoted xTt-1, where T indicates the transpose of a column vector). Model 1 also conditioned on the fatigue states of 6 and 12 months earlier and their interaction, thereby focusing on transitions from one visit to the next in individuals. Model 2 had similar interpretations, with the roles of fatigue and depression reversed. All of a subject's data on fatigue were modeled, as long as data on fatigue from the two previous visits were available. The intra-subject correlation arising from repeated measures was accounted for by conditional likelihood estimation.39 The second-order Markov assumption required that the two previous visits be treated as explanatory variables. Since many subjects had unusually high levels of fatigue or depression at the first visit, possibly due to the anxiety of entering the study, observations with visit 1 data were excluded. Thus, response data at visits 1, 2, and 3 were not modeled.

Regression coefficients {alpha}0 and {theta}0 were of primary interest. In model 1, exp({alpha}0) expressed the odds of being fatigued versus not being fatigued for subjects who were depressed versus subjects who were not depressed 6 months earlier, adjusting for fatigue history and covariates measured 6 months earlier. Likewise, exp({theta}0) was the odds ratio of being depressed versus not being depressed for subjects who were fatigued versus subjects who were not fatigued 6 months earlier, adjusting for the previous year's depression history and covariates measured 6 months earlier. Although dt-2 and ft-2 did not appear in the models, the statistical significance of these effects in model 1 and model 2, respectively, was also determined. These analyses addressed the question of whether, for example, in model 1 depression 1 year ago significantly improved the prediction of fatigue after accounting for depression 6 months earlier. The advantage of such statistical models is that the values of the predictors can change over time.

Descriptive statistics complementary to the modeling goals were determined. These included summary demographic statistics and assessment of Pearson correlations of fatigue and depression with covariates measured 6 months earlier. Partial correlations were used to control for intra-subject correlation resulting from repeated measures. We report frequency counts of transitions of fatigue and depression to describe the probability of changing from one fatigue or depression state (presence or absence) to another. Regression coefficients in models 1 or 2 were estimated by ordinary logistic regression techniques. This technique gives partial likelihood estimates under the Markovian assumption that {alpha} or {theta} parameters account for the intra-subject correlation of a subject's longitudinal observations.39 We exploited the cross-sectional correlation of the response variables Ft and Dt by employing simultaneous model fitting and robust covariance estimation of regression coefficients40 to increase statistical efficiency. Estimated regression coefficients and their standard errors were determined with SAS PROC GENMOD.41

The goal of the study was to evaluate the relative importance of physiologic compared to psychological predictors of fatigue and depression in HIV-positive gay men. To this end, four models were identified. Model A was fit to characterize the relationship between fatigue and depression without adjustment for other covariates. Next, we identified the best fitting model (model B) that considered only the effect of physiological variables on fatigue and depression. Similarly, we identified the best fitting model (model C) that considered only the effect of psychological variables on each outcome. Last, model D evaluated the relative effects of both physiological and psychological variables on fatigue and depression. Covariates for models B and C, respectively, were identified with a backward selection procedure by using a .05 significance level criterion for variables to stay in the model. Variables with significant partial correlations were treated as candidate predictors. Symmetry was established throughout the analysis by using the same set of covariates for fatigue as for depression.


  RESULTS

 
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
Of the 96 men in the study, 36 men had data through 7.5 years. Some men were lost to follow-up, while others were late recruits for whom data were still being collected in the parent study. Table 1 shows the demographic and clinical characteristics of the men at entry into the study. In particular, the prevalence of fatigue and depression at entry into the study were 26.0% (the mean fatigue score of 7.1 corresponds to the 46th percentile in a sample of male psychiatric outpatients27) and 31.3% (the mean depression score of 11.9 corresponds to the 43rd percentile in a sample of male psychiatric outpatients27), respectively. Table 2 reports partial Pearson correlations of covariates 6 months earlier to the continuous depression and fatigue POMS scales. Depression was strongly correlated with previous depression (r=0.44), and fatigue with previous fatigue (r=0.43), illustrating the need to account for intra-subject correlation in longitudinal data analysis. Depression at the previous visit was correlated with fatigue (r=0.30), and fatigue at the previous visit with depression scores (r=0.31). CD4 count, CDC clinical status, and hemoglobin and hematocrit levels at the previous visit were slightly correlated with fatigue (r ranged from –0.11 to 0.16). None of the physiologic variables evaluated were significantly related to depression. Anxiety, hopelessness, social conflict, and lack of satisfaction with support were significantly and modestly associated with both fatigue and depression. Age was significantly correlated with both fatigue and depression.


View this table:
[in this window]
[in a new window]
 

TABLE 1. Baseline demographic and clinical characteristics of HIV-positive gay men in a study of HIV-related fatigue (n = 96)




View this table:
[in this window]
[in a new window]
 

TABLE 2. Correlation of depression and fatigue in HIV-positive gay men (n = 96) with risk factors assessed 6 months earlier



Frequency counts of transitions between fatigue and depression are given in Table 3. From months 18 to 84, there were a total of 115 occasions in which fatigue was present and 84 occasions in which depression was present. These data corresponded to 641 observations from 83 subjects who had at least one two-step transition, i.e., three consecutive nonmissing responses for both fatigue and depression. Thus, the estimated proportions of observations from months 18 to 84 with fatigue and depression present were 17.9% and 13.1%, considerably less than the 1-month prevalence levels reported earlier. Subjects with depression 6 months earlier had a 4.5 times greater odds ([434+40][7+31]/[30+22][57+20]) of being fatigued than those who were not depressed 6 months earlier. However, this finding did not take into account previous fatigue as a confounder. With fatigue 6 months earlier either present or absent, the observed odds ratios were 2.82 and 1.78, respectively, illustrating the importance of using observation-driven models. Similarly, with depression 6 months earlier either present or absent, the odds ratios of depression were 3.76 and 1.53, respectively. These descriptive results suggest that fatigue and depression play roles in predicting each other, perhaps to comparable degrees.


View this table:
[in this window]
[in a new window]
 

TABLE 3. Frequency of transitions between fatigue and depression in HIV-positive gay men (n = 83) in study months 18 to 84



Table 4 presents the estimated odds ratios and their 95% confidence intervals for models A, B, C and D. Model A suggested that those with depression 6 months earlier had a 1.97 times greater odds of being fatigued than those who were not depressed 6 months earlier, when the analysis controlled for the previous year's fatigue history. Similarly, model A suggested that those with fatigue 6 months earlier had a 1.99 times greater odds of being depressed than those who were not fatigued 6 months earlier, when the analysis controlled for the previous year's depression history. Thus, at least from model A, it appears that fatigue and depression have similar effects on one another. Model B suggested that, among all the physiologic variables considered, only CDC clinical status was a statistically significant predictor. The estimated odds ratios of 1.97 and 1.99 for the relationship between fatigue and depression in model A changed marginally to 2.11 and 2.10, respectively, when the analysis controlled for CDC clinical status. In model B, CDC clinical status C was a significant predictor of fatigue but not of depression. CDC clinical status A and CDC clinical status B were not significantly different for either outcome. Next, in model C, both anxiety and hopelessness were significant predictors of depression, whereas only anxiety predicted fatigue. When all variables from models B and C were included in model D, similar conclusions were reached. In sum, fatigue had both physiologic and psychological risk factors, whereas depression was found to have only psychological risk factors. Due to colinearity, the effect of depression on fatigue seen in models A and B disappeared when anxiety and hopelessness were added to the model; this result can be explained by the strong cross-sectional correlations, based on all time points, of depression with anxiety (r=0.76) and hopelessness (r=0.49). To a somewhat lesser degree, the effect of fatigue on future depression was accounted for by the strong correlation of fatigue with anxiety (r=0.63) and hopelessness (r=0.42). Finally, the results of model B (compared to model A) and model D (compared to model C) suggested that CDC clinical status was not a confounder for the relationship between fatigue and depression.


View this table:
[in this window]
[in a new window]
 

TABLE 4. Adjusted odds ratios and 95% confidence intervals in four logistic regression models predicting fatigue and depression in HIV-positive gay men (n = 83)




  DISCUSSION

 
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
Some limitations must be considered in interpreting the study results. We did not know when the study subjects became HIV-positive; however, most clinicians do not have this information. Therefore, we made assumptions about transitions in fatigue and depression from one visit to the next that apply regardless of the time since seroconversion. Despite the absence of this information, the observation-driven models were well suited for making clinical predictions. The study subjects were all gay men; however, men who have sex with other men constitute the largest group of people with HIV infection, categorized by type of exposure.42 This was also the largest group by exposure category when the Coping in Health and Illness Project began. Similarly, we did not have information on HIV viral load from most of the visits, because the test for viral load was not available until the study had been underway for several years. Finally, our data did not permit us to predict whether fatigue or depression occurred first for the subjects; however, clinical experience has taught us that most patients are themselves unable to distinguish which came first. Perhaps the most important issues are how depression and fatigue predict one another and how they are predicted by other factors.

The similarity in the estimated odds ratios (1.97 and 1.99) for the relationship between fatigue and depression suggests that the roles of fatigue and depression in predicting one another in HIV-positive gay men are comparable. The only physiologic variable that predicted fatigue was CDC clinical status C. Depression no longer predicted fatigue when anxiety and hopelessness were added to the model because of colinearity among the three concepts. Clark and Watson43 have proposed that depression and anxiety represent a single construct of negative affectivity.

The results suggest that clinicians should make assessments for depression in patients who present with HIV-related fatigue. Depression should be treated aggressively, in the effort to affect any accompanying fatigue. Clinicians should also consider the unfortunate longevity of these conditions in people with HIV, as past history of either depression or fatigue predicted a continuation of these states. A few intervention studies have examined the relationship between fatigue and depression in patients with HIV; one study found dextroamphetamine to be a potentially effective, fast-acting treatment for both depression and fatigue.44 Another study that used DHEA to treat depressed mood and fatigue found that both improved after 8 weeks of treatment.45

Additional research examining HIV-related fatigue is needed. Besides the toll in terms of human suffering, the economic costs associated with HIV-related fatigue are likely to be large, given the many people who are affected and the extent of the disability associated with fatigue. When people with HIV have to stop working due to fatigue, they often lose their health insurance, and many must turn to public assistance to meet health care and daily living expenses, including housing expenses. Therapy to reduce fatigue could give HIV-positive patients improved quality and vigor of life, and reduce societal costs, by enhancing their ability to maintain employment.46 New approaches to managing HIV disease must include strategies to deal with fatigue and other symptoms. Our data indicated that HIV-related fatigue has both psychological and physiologic risk factors, and researchers should continue to search for causes in both of these areas. Different causes of HIV-related fatigue may be operating in different people; for some, the fatigue may have primarily psychological correlates and others for whom they are primarily physiologic. In the quest to find a way to slow or stop HIV, quality of life for some has been sacrificed in exchange for longevity. Now that health care providers can extend length of life for HIV-positive persons, it is time to deal with the symptoms that affect the quality of that life.


  ACKNOWLEDGMENTS

 
This study was funded by the National Institute of Mental Health (MH-44618 and MH-33127) and the National Institutes of Health (RR-00046 and HD-37260).


  REFERENCES

 
TOP
 ABSTRACT
 INTRODUCTION
 METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 

  1. Breitbart W, McDonald MV, Rosenfeld B, et al: Fatigue in ambulatory AIDS patients. J Pain Symptom Manage 1998; 15:159-167[CrossRef][Medline]
  2. Cunningham WE, Shapiro MF, Hays RD, et al: Constitutional symptoms and health-related quality of life in patients with symptomatic HIV disease. Am J Med 1998; 104:129-136[CrossRef][Medline]
  3. Fontaine A, Larue F, Lassauniere JM: Physicians' recognition of the symptoms experienced by HIV patients: how reliable? J Pain Symptom Manage 1999; 18:263-270[CrossRef][Medline]
  4. Vogl D, Rosenfeld B, Breitbart W, et al: Symptom prevalence, characteristics, and distress in AIDS patients. J Pain Symptom Manage 1999; 18:253-262[CrossRef][Medline]
  5. Walker K, McGown A, Jantos M, et al: Fatigue, depression, and quality of life in HIV-positive men. J Psychosoc Nurs and Ment Health Serv 1997; 35:32-40
  6. Justice AC, Rabeneck L, Hays RD, et al: Sensitivity, specificity, reliability, and clinical validity of provider-reported symptoms: a comparison with self-reported symptoms. J Acquir Immune Defic Syndr 1999; 21:126-133
  7. Fleishman JA, Crystal S: Functional status transitions and survival in HIV disease. Med Care 1998; 36:533-543[CrossRef][Medline]
  8. Darko DF, McCutchan JA, Kripke DF, et al: Fatigue, sleep disturbance, disability, and indices of progression of HIV infection. Am J Psychiatry 1992; 149:514-520[Abstract/Free Full Text]
  9. Semple SJ, Patterson TL, Temoshok LR, et al: Identification of psychobiological stressors among HIV-positive women. Women Health 1993; 20:15-36[CrossRef][Medline]
  10. Penzak SR, Reddy S, Grimsley SR: Depression in patients with HIV infection. Am J Health Syst Pharm 2000; 57:376-386[Abstract/Free Full Text]
  11. Barroso J: "Just worn out:" a qualitative study of HIV-related fatigue, in Key Aspects of Preventing and Managing Chronic Illness. Edited by Funk SG, Tornquist EM, Leeman J, et al. New York, Springer, 2001, pp 183-194
  12. Barroso J: "Biomarkers associated with HIV-related fatigue." Poster presented at the 13th National Conference of the Association of Nurses in AIDS Care. November 13-14, 2000. San Juan, PR
  13. Perkins DO, Leserman J, Stern RA, et al: Somatic symptoms and HIV infection: relationship to depressive symptoms and indicators of HIV disease. Am J Psychiatry 1995; 152:1776-1781[Abstract/Free Full Text]
  14. Ferrando S, Evans S, Goggin K, et al: Fatigue in HIV illness: relationship to depression, physical limitations, and disability. Psychosom Med 1998; 60:759-764[Abstract/Free Full Text]
  15. Rabkin JG, Wagner GJ, Rabkin R: Fluoxetine treatment for depression in patients with HIV/AIDS: a randomized, placebo-controlled trial. Am J Psychiatry 1999; 156:101-107[Abstract/Free Full Text]
  16. Rabkin JG, Rabkin R, Harrison W, et al: Effects of imipramine on mood and enumerative measures of immune status in depressed patients with HIV illness. Am J Psychiatry 1994; 151:516-523[Abstract]
  17. Leserman J, Jackson BS, Pettito JM, et al: Progression to AIDS: the effects of stress, depressive symptoms, and social support. Psychosom Med 1999; 61:397-406[Abstract/Free Full Text]
  18. Leserman J, Pettito JM, Golden RN, et al: Impact of stressful life events, depression, social support, coping, and cortisol on progression to AIDS. Am J Psychiatry 2000; 157:1221-1228[Abstract/Free Full Text]
  19. Wessely S, Hotopf M., Sharpe M: Chronic Fatigue and Its Syndromes. Oxford, UK, Oxford University Press, 1998
  20. Cleary PD, Fowler FJ, Weissman J, et al: Health-related quality of life in persons with acquired immune deficiency syndrome. Med Care 1993; 31:569-580[CrossRef][Medline]
  21. O'Dell MW, Hubert HB, Lubeck DP, et al: Pre-AIDS physical disability: data from the AIDS Time-Oriented Health Outcome Study. Arch Phys Med Rehabil 1998; 79:1200-1205[CrossRef][Medline]
  22. De Boer JB, van Dam FSAM, Sprangers MAG, et al: Longitudinal study on the quality of life of symptomatic HIV-infected patients in a trial of zidovudine versus zidovudine and interferon-alpha. AIDS 1993; 7:947-953[Medline]
  23. Vlahov D, Munoz A, Solomon L, et al: Comparison of clinical manifestations of HIV infection between male and female injecting drug users. AIDS 1994; 8:819-823[Medline]
  24. Lee KA, Portillo CJ, Miramontes H: The fatigue experience for women with human immunodeficiency virus. J Obstet Gynecol Neonatal Nurs 1999; 28:193-200[CrossRef][Medline]
  25. Groopman JE: Fatigue in cancer and HIV/AIDS. Oncology (Huntingt) 1998; 12:335-341[Medline]
  26. Centers for Disease Control and Prevention (CDC):1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adult males. MMWR Morbid Mortal Wkly Rep 1992; 41(RR-17):1-19
  27. McNair D, Lorr M, Droppleman L: Manual for the Profile of Mood States. San Diego, Educational and Industrial Testing Service, 1971
  28. Akechi T, Kugaya A, Okamura H, et al: Fatigue and its associated factors in ambulatory cancer patients: a preliminary study. J Pain Symptom Manage 1998; 17:42-48
  29. Cella DF, Tross S, Orav EJ, et al: Mood states of patients after the diagnosis of cancer. J Psychosoc Oncology 1989; 7(1/2):45-54
  30. Cunningham AJ, Lockwood GA, Edmunds CV: Which cancer patients benefit most from a brief, group, coping skills program? Int J Psychiatry Med 1993; 23:383-398[Medline]
  31. Kugaya A, Akechi T, Okamura H, et al: Correlates of depressed mood in ambulatory head and neck cancer patients. Psychooncology 1999; 8:494-499[CrossRef][Medline]
  32. Nyenhuis DL, Yamamoto C, Luchetta T, et al: Adult and geriatric normative data and validation of the Profile of Mood States. J Clin Psychol 1999; 55:79-86[CrossRef][Medline]
  33. Mohr DC, Likosky W, Bertagnolli A, et al: Telephone-administered cognitive-behavioral therapy for the treatment of depressive symptoms in multiple sclerosis. J Consult Clin Psychol 2000; 68:356-361[CrossRef][Medline]
  34. Beck AT, Weinnman A, Lester D, et al: The measurement of pessimism: the Hopelessness Scale. J Consult Clin Psychol 1974; 42:861-865[CrossRef][Medline]
  35. O'Brien KO, Wortman CB, Kessler RC, et al: Social relationships of men at risk for AIDS. Soc Sci Med 1993; 36:1161-1167
  36. Sarason IG, Sarason BR, Shearin EN, et al: A brief measure of social support: practical and theoretical implications. J Soc Pers Relat 1987; 4:497-510[Abstract]
  37. Cox DR: Statistical analysis of time series: some recent developments. Scand J Stat 1981; 8:93-115
  38. Zeger SL, Qaqish B: Markov regression models for time series: a quasi-likelihood approach. Biometrics 1988; 44:1019-1031[CrossRef][Medline]
  39. Cox DR: The Analysis of Binary Data. London, Chapman and Hall, 1970
  40. Diggle PJ, Liang K-Y, Zeger SL: Analysis of Longitudinal Data. Oxford, UK, Clarendon Press, 1994
  41. SAS/STAT Software: Changes and Enhancements through Release 6.12. Cary, NC, SAS Institute, 1997
  42. Centers for Disease Control and Prevention (CDC), Divisions of HIV/AIDS Prevention: Basic statistics through June 2001. http://www.cdc.gov/hiv/stats.htm#exposure (accessed September 18, 2001)
  43. Clark LA, Watson D: Tripartite model of anxiety and depression: psychometric evidence and taxonomic implications. J Abnorm Psychol 1991; 100:316-336[CrossRef][Medline]
  44. Wagner GJ, Rabkin R: Effects of dextroamphetamine on depression and fatigue in men with HIV: a double-blind, placebo-controlled trial. J Clin Psychiatry 2000; 61:436-440[Medline]
  45. Rabkin JG, Ferrando SJ, Wagner GJ, et al: DHEA treatment for HIV+ patients: effects on mood, androgenic and anabolic parameters. Psychoneuroendocrinology 2000; 25:53-68[CrossRef][Medline]
  46. Darko DF, Mitler MM, Henriksen SJ: Lentiviral infection, immune response peptides and sleep. Adv Neuroimmunol 1995; 5:57-77[CrossRef][Medline]



This article has been cited by other articles:


Home page
 Psychosom. Med.Home page
S. Reid and J. Dwyer
Insomnia in HIV Infection: A Systematic Review of Prevalence, Correlates, and Management
Psychosom Med, March 1, 2005; 67(2): 260 - 269.
[Abstract] [Full Text] [PDF]

Home page
 Biol Res NursHome page
K. D. Phillips, R. L. Sowell, M. Rojas, A. Tavakoli, L. J. Fulk, and G. A. Hand
Physiological and Psychological Correlates of Fatigue in HIV Disease
Biol Res Nurs, July 1, 2004; 6(1): 59 - 74.
[Abstract] [PDF]

This Article
* Abstract Freely available
* Full Text (PDF)
* Alert me when this article is cited
* Alert me if a correction is posted
* Citation Map
Services
* Email this article to a Colleague
* Similar articles in this journal
* Similar articles in PubMed
* Alert me to new issues of the journal
* Add to My Articles & Searches
* Download to citation manager
* reprints & permissions
Citing Articles
* Citing Articles via HighWire
* Citing Articles via Google Scholar
Google Scholar
* Articles by Barroso, J.
* Articles by Evans, D. N.
* Search for Related Content
PubMed
* PubMed Citation
* Articles by Barroso, J.
* Articles by Evans, D. N.
Related Collections
* AIDS/HIV
* Depression

Get information about faster international access.

Privacy Policy

Copyright © 2002 Academy of Psychosomatic Medicine. All rights reserved.

Home | Search | Current Issue | Past Issues | Subscribe | All APPI Journals | Help | Contact Us

American Psychiatric Publishing, Inc. Academy of Psychosomatic Medicine
1000 Wilson Boulevard, Suite 1825, Arlington, VA 22209-3901 * 800-368-5777 * appi at psych.org