General medical and psychiatric comorbidity among HIV-infected veterans in the post-HAART era

General medical and psychiatric comorbidity among HIV-infected veterans in the post-HAART era

Journal of Clinical Epidemiology 54 (2001) S22–S28 General medical and psychiatric comorbidity among HIV-infected veterans in the post-HAART era A.M...

133KB Sizes 0 Downloads 59 Views

Journal of Clinical Epidemiology 54 (2001) S22–S28

General medical and psychiatric comorbidity among HIV-infected veterans in the post-HAART era A.M. Kilbournea,b, A.C. Justicea,b,*, L. Rabeneckc,d, M. Rodriguez-Barradasd, S. Weissmane,†, for the VACS 3 Project Team a

Center for Health Equity Research and Promotion, VA Pittsburgh Healthcare System, University Drive C 11E-124 (130-U), Pittsburgh, PA 15240, USA Veterans Aging Cohort Study (VACS) Center, Division of General Internal Medicine, School of Medicine, University of Pittsburgh, Pittsburgh, PA, USA c VA Health Services Research and Development Center of Excellence, Houston, TX, USA d Department of Medicine, Baylor College of Medicine, Houston, TX, USA e Department of Internal Medicine, Infectious Diseases Section, Yale University School of Medicine and Hospital of Saint Raphael, New Haven, CT, USA Received 24 July 2001; received in revised form 17 August 2001; accepted 17 August 2001

b

Abstract We examined the prevalence of HIV, general medical, and psychiatric comorbidities by age based on a recent multisite cohort of HIV infected veterans receiving care: the Veterans with HIV/AIDS 3 Site Study (VACS 3). VACS 3 includes 881 adult patients with HIV infection enrolled between June 1999 and July 2000. Providers reported their patients’ CDC-defined HIV comorbidities, general medical comorbidities (based on Duke and Charlson comorbidity scales), and psychiatric comorbidity. Mean age of participants was 49 years and 54% were African-American. The most common HIV comorbidities were oral candidiasis (21%), peripheral neuropathy (16%), and herpes zoster (16%). The most common general medical comorbidities included chemical hepatitis (53%), hypertension (24%), and hyperlipidemia (17%). The mean number of HIV and general medical comorbidities experienced by patients were respectively 1.1 and 1.4 (P  .001). Older (50 years) HIV-infected patients experienced a greater number of general medical comorbidities than those 50 years (respectively 1.7 versus 1.2, P  .001). There was no significant difference in mean HIV comorbidity number by age. Based on patient report, 46% had significant depressive symptoms (10 on 10-item CES-D) and 21% reported at-risk drinking (8 on AUDIT). Providers reported 32% of patients had anxiety, 4% mania, 4% schizophrenia, and 11% cognitive impairment/dementia. General medical and psychiatric comorbidities constituted a higher disease burden for HIV-infected veterans than HIV comorbidities. Whether these comorbidities are due to antiretroviral drug toxicity or are age or lifestyle-associated conditions, the substantial prevalence of these “non-HIV” comorbidities suggest an important role for general medical and psychiatric management of HIV-infected patients. © 2001 Elsevier Science Inc. All rights reserved. Keywords: Psychiatric comorbidity; HIV-infected veterans; Post-HAART era

1. Background As HIV enters its third decade, the profile of comorbidities among people with HIV infection is evolving [1]. Some of the previously most common conditions among HIVinfected individuals such as Kaposi’s sarcoma, Mycobacterium avium complex disease, and pneumocystis carinii pneumonia [2–4] are becoming less common. In this article, we define HIV comorbidities as including these and other conditions that were commonly associated with HIV infection early in the epidemic, as enumerated in the 1993 re-

* Corresponding author. Tel.: 412-688-6957; fax: 412-688-6916. E-mail address: [email protected] (A.C. Justice). † S. Weissman is formerly of VA Cleveland Medical Center, Cleveland, OH.

vised CDC-Defined Classification for HIV Infection [5]. We define non-HIV comorbidities as all other medical, neurologic, and psychiatric conditions. These “non-HIV” conditions, including general medical comorbidities such as hepatitis [6,7], hyperlipidemia [8,9], cardiac disease [10,11], and diabetes [9,12,13] are becoming more apparent among HIV-infected patients. These conditions are likely associated with antiretroviral therapy, rather than HIV infection [14]. As HIV-infected patients live longer, they may also be susceptible to general medical conditions associated with age [15]. Neurologic conditions and psychiatric comorbidity (including depression and drug/alcohol abuse) are also common among HIV infected individuals receiving care [16,17]. Hence, we determined the prevalence of “HIV” and “non-HIV” comorbidities in a recent multisite cohort of in-

0895-4356/01/$ – see front matter © 2001 Elsevier Science Inc. All rights reserved. PII: S0895-4356(01)00 4 4 3 - 7

A.M. Kilbourne et al. / Journal of Clinical Epidemiology 54 (2001) S22–S28

dividuals with HIV infection receiving care within the VA system. We compared the prevalences of HIV and nonHIV-related comorbidities among older (50 years) HIVinfected veterans to the prevalences among younger (50 years) HIV-infected veterans. In addition, we determined the prevalence of overlapping conditions, such as dually diagnosed depression and drug/alcohol use.

S23

site. We identified CD4 cell count using the lab result obtained on the date nearest to patient survey completion up to 4 weeks (28 days) following the survey date. We obtained CD4 cell count data on 875 VACS 3 participants (99%). 4. Psychiatric comorbidity 4.1. Patient report

2. Methods We evaluated baseline data from the Veterans with HIV/ AIDS 3 site Cohort Study (VACS 3), an ongoing longitudinal study of veterans with HIV from infectious disease (ID) clinics at VA Medical Centers in Houston, Cleveland, and Manhattan. Additional details on VACS 3 are available elsewhere [18]. We collected baseline data from patients and their providers between June 1999 through July 2000. HIV-infected patients currently attending these clinics were approached at the time of their clinic appointments and asked to self-complete a survey that included questions on depression and substance use. Providers were asked to complete a survey on their patients’ severity of illness, including questions pertaining to HIV, general medical, and psychiatric comorbidities. We collected demographic and laboratory data (including CD4 cell count) using the electronic medical record system with assistance from the VA Information Resources Management (IRM) at each site. Informed consent was obtained from all patients participating in this study, and Human Subjects Committees at all participating sites as well as the University of Pittsburgh institutional review board reviewed and approved VACS 3. 3. Comorbidity and CD4 cell count Providers were asked to report which HIV (19 total) and general medical (17 total) comorbidities each patient had experienced at any time (see survey instruments included in Smola et al. in this issue). HIV comorbidities were defined as those conditions included in the 1993 revised CDCDefined Classification for HIV Infection [5]. General medical comorbidity questions were based on the Duke and Charlson comorbidity indices [19–21]. If providers skipped an individual HIV or general medical comorbidity question, the patient was assumed to not have that particular comorbidity. However, data were considered missing for HIV comorbidities if the provider skipped the entire portion of the survey dedicated to HIV comorbidity, and were considered missing for general medical comorbidities if the provider skipped the entire portion of the survey for general medical comorbidities. We determined the overall frequency of each comorbidity, and assessed the total number of HIV and total number of general medical comorbidities among all patients. We collected data on CD4 cell count per cubic millimeter (mm3) from computerized laboratory records at each

We assessed patient-reported depressive symptoms using the 10-item version of the Centers for Epidemiologic Studies Depression Scale (CES-D) [22]. Patients were considered to have significant depressive symptoms if they obtained a standard cut point of 10 or more out of a possible 30 points. This cut point in the 10-item CES-D has been previously shown to be concordant with the CES-D 20-item survey’s standard cut point of 16, with a sensitivity of 96% and specificity of 100%, based on a sample of older adults in primary care [22]. The CES-D 10-item survey has not been directly compared to a DSM-based diagnosis of major depressive disorder. However, a conservative estimate of the sensitivity was 81% (95% CI: 75–86%), and specificity 76% (95% CI: 70–82%). These estimates are based on comparisons of the CESD 10-item with the CES-D 20-item survey [22] and of the CES-D 20-item with DSMIIIR diagnostic interview schedule for major depressive disorder in studies of comparable primary care adult populations [23]. For the purposes of this study, patients who scored 10 on the CES-D 10-item survey were considered to have “significant depressive symptoms.” The Alcohol Use Disorders Identification Test (AUDIT) was used to assess at-risk alcohol use [24], defined as alcohol consumption that places the individual at risk for adverse events, including physical or psychological harm [25]. Patients answered 10 questions pertaining to their past-year quantity and frequency of drinking, drinking behavior, and alcohol-related problems such as guilt or memory loss associated with drinking. We generated a total score by adding the Likert-type responses together from each question. A cutoff score of 8 out of 40 points for the AUDIT has been previously established to have both a sensitivity and specificity of 96% in detecting “at risk alcohol use” [24]. Current alcohol use was also assessed, and information on present drug use was obtained by asking whether the patient is currently taking illegal drugs. The CES-D, AUDIT, and the drug use questions are available from Smola et al. in this issue [18]. 4.2. Provider report Providers were queried on whether their patients had manic depression or bipolar disorder, schizophrenia, anxiety disorder, and depression (response options were yes/no for these conditions). Providers were also asked whether their patient had any cognitive impairment or dementia (not necessarily HIV-related), and response categories were “a great deal,” “some,” “slight,” or “none.” Patients were con-

S24

A.M. Kilbourne et al. / Journal of Clinical Epidemiology 54 (2001) S22–S28

sidered to have dementia if providers responded with “a great deal” or “some.” 4.3. Analyses We included all 19 provider-reported HIV comorbidities in the analyses. We combined the following general medical comorbidities: hepatitis with abnormal liver function test (labeled “chemical hepatitis”), pulmonary disease—not requiring oxygen with pulmonary disease-requiring oxygen, renal insufficiency with renal failure, and diabetes mellitusdiet controlled with diabetes mellitus—requiring medication. Hence, we included 13 provider-reported general medical comorbidities for the analysis. Mean comorbidity counts were stratified by age (50 years, 50 years) and CD4 cell count (50, 50–199, 200– 499, and 500 cells/mm3). Studentized t-tests were used to test for significant differences between HIV and general medical comorbidity counts for each age and CD4 cell count group. Wilcoxon nonparametric tests were employed to determine whether numbers of comorbidities differed between groups of patients classified by age and CD4 counts, given the non-normal distribution of these comorbidity counts. We created a Venn diagram characterizing the overlap of patient-reported depressive symptoms, drug/alcohol use, and provider-reported general medical comorbidities. We assessed the proportion of patients with at least two of the following conditions: significant depressive symptoms (CESD-10 score of 10), substance use problems 8 on the AUDIT or reporting current drug use), and having two or more general medical comorbidities. All analyses were performed using SAS version 8 (SAS Institute; Cary, NC). We also considered potential variation by site regarding comorbidity prevalences. 5. Results 5.1. Demographics

ties: TB (P  .001), herpes zoster ( P  .001), enteric parasites (P  .001), bacterial sepsis (P  .001), and HIV wasting (P  .004). Some general medical comorbidity frequencies also varied by site: chemical hepatitis (P  .001), pancreatitis (P  .001), hypertension (P  .01), hyperlipidemia (P  .001), diabetes (P  .002), myocardial infarction (P  .044), stroke (P  .008), and non-HIVrelated cancer (P  .028). The prevalence of current drinking (patient report) varied by site (P  .001), as well providerreported psychiatric comorbidity: dementia (P  .001), depression (P  .006), and mania (P  .041). The most common HIV comorbidities reported by providers were oral candidiasis (21%), peripheral neuropathy (16%), and herpes zoster (16%) (Fig. 1). The frequencies for pneumocystis carinii pneumonia (PCP) and Kaposi’s sarcoma (KS) were respectively 12 and 3%. Frequencies for HIV and general medical comorbidities were stratified by age groups (50, 50 years). Older (50 years) HIV-infected patients were less likely to have oral candidiasis according to providers than younger HIV infected patients (16 versus 25%, P  .001) (Table 1). There were no other significant differences in HIV comorbidity frequency between age groups. The most common general medical comorbidities were chemical hepatitis (53%), hypertension (24%), and hyperlipidemia (17%) (Fig. 2). Compared to younger HIV-infected patients, those older (50 years) were more likely to have had hypertension (P  .001), hyperlipidemia (P  .001), diabetes (P  .001), myocardial infarction/coronary artery disease (P  .05), non-HIV-related cancers (P  .006), stroke/TIA (P  .001), peripheral vascular disease (P  .02), and congestive heart failure (P  .006) (Table 2). Over a third (39%) of patients experienced 2 general medical comorbidities, while 46% had no HIV comorbidities based on provider report. The mean number of HIV comorbidities experienced was lower (1.1) compared to the mean number of general medical comorbidities (1.4; P  0.001) (Table 3). Those age 50 years had a higher mean general medical comorbidity than HIV comorbidity count

Between June 1999 and July 2000, 1,038 HIV-infected patients were seen at the three VA ID clinics, and of those, 915 patients were approached to be surveyed. Overall, 881 patients completed surveys for VACS 3, representing 85% of the HIV infected clinic population at the three sites. Of the 881 enrolled patients, the mean age was 49 years, 99% were male, 54% African-American, and 87% were currently taking antiretroviral medications. Further details concerning VACS 3 descriptive statistics are available from Smola et al. [18]. 5.2. HIV and general medical comorbidity Seventy-one provider surveys were missing both the HIV and general medical comorbidity sections in their entirety; hence 810 patients were included in the HIV and general medical comorbidity results. Site-specific variations in frequency were apparent for the following HIV comorbidi-

Fig. 1. HIV-related comorbidities (provider-reported; N  810).

A.M. Kilbourne et al. / Journal of Clinical Epidemiology 54 (2001) S22–S28 Table 2 Prevalence of general medical comorbidities by age groups (n  802)a

Table 1 Prevalence of HIV comorbidities by age groups (n  802)a

Percent with comorbidity

Percent with comorbidity Age 50 Oral candidiasis Peripheral neuropathy Herpes zoster Bacterial pneumonia Pneumocystis caninii pneumonia (PCP) HIV wasting Tuberculosis HIVdementia Kaposi’s sarcoma (KS) Bacterial sepsis Mycobacterium avium complex disease MAI/MAC) Crypotcoccus Enteric parasites (e.g., cryptosporidia) CMV retinitis Toxoplasmosis Histoplasmosis Lymphoma Other AIDS-related cancers Coccidioidomycosis

S25

Age 50

P-valueb

25 16 16 14 13 9 4 4 2 2

16 17 15 13 10 8 7 5 4 4

.001 .59 .74 .78 .16 .57 .12 .91 .13 .19

3 3 2 2 1 1 1 1 1

2 2 3 2 1 1 1 1 0

.93 .50 .40 .85 .35 .96 .83 .17 .40

a

Seventy-nine were missing either comorbidity or age data. P-values based on chi-square test—significant difference between age groups. b

(1.7 versus 1.1, P  .001), yet there was no such difference for those 50 years (1.2 versus 1.2, P  1.0). Mean general medical comorbidity counts were higher compared to mean HIV comorbidity counts for those with CD4 counts of 200– 499/mm3 (respectively, 1.4 and 1.0, P  .001), and for those with CD4 counts of 500/mm3 (respectively, 1.5 and 0.7; P  .001). There was no significant difference in the number of HIV comorbidities between age groups (respectively, 1.2 versus 1.1, P  .73). Mean HIV comorbidity counts were higher for those with CD4 counts of 50/mm3 (2.9) compared to those with CD4 counts of 50–199, 200–499, and 500 cells mm3 (respectively, 1.5, 1.0, 0.7; P  .001).

Age 50 Age 50 P-valueb Chemical hepatitis Hypertension Hyperlipidemia Diabetes—medication or diet controlled Pulmonary disease Renal insufficiency or failure Pancreatitis Myocardial infarction/CAD Non-HIV cancer Stroke/TIA Liver failure/cirrhosis Peripheral vascular disease Congestive heart failure (CHF)

51 33 22 15 12 8 5 5 6 5 2 3 4

.36 .001 .001 .001 .35 .14 .16 .05 .006 .001 .92 .02 .006

a

Seventy-nine were missing either comorbidity or age data. P-values based on chi-square test—significant difference between age groups. b

Older (50 years) HIV-infected patients experienced a greater number of general medical comorbidities than those under 50 years old (respectively, 1.7 versus 1.2, P  .001). There were no differences in mean general medical comorbidity counts by CD4 group (Table 3). 5.3. Psychiatric comorbidity Almost half (46%) of HIV-infected patients had significant depressive symptoms (10 on the CES-D 10-item survey). For substance use, 40% of patients reported current alcohol use, 21% reported at-risk alcohol use (8 on the AUDIT) and 13% reported present drug use (Table 4). There were no significant differences in CES-D score for significant depressive symptoms nor in the frequency of

Table 3 Mean counts for HIV and general medical comorbidities by age group

N Overall Age 50 years 50 years CD4 Cell Count 0–50 cells/mm3 50–199 cells/mm3 200–499 cells/mm3 500 cells/mm3 a

Fig. 2. General medical comorbidities (provider-reported; N  810).

54 18 14 8 10 5 3 2 2 1 2 1 1

810 803 473 329 805 50 165 380 210

HIV comorbidities

General medical comorbidities

P-valuea

1.1

1.4

.001

1.2b 1.1

1.2c 1.7

1.00 .001

2.9d 1.5 1.0 0.7

1.3e 1.4 1.4 1.5

.001 .50 .001 .001

P-values based on Studentized t-tests comparing mean HIV and general medical comorbidity counts. b P  .73 (Wilcoxon nonparametric test). c P  .001; Significant difference within general medical comorbidities based on Wilcoxon nonparametric test. d P  .001; Significant difference within HIV comorbidities based on Wilcoxon nonparametric test. e P  .77 (Wilcoxon nonparametric test).

S26

A.M. Kilbourne et al. / Journal of Clinical Epidemiology 54 (2001) S22–S28

Table 4 Psychiatric comorbidity: patient and provider report Age groups (in years)

Patient Report Significant depressive symptoms Mean (SD) CES-D 10-item score CES-D 10 (%) Substance use Current alcohol use (%) Mean (SD) AUDIT acore AUDIT 8 (%) Presently use illicit drugs (%) Provider Report Depression Drinks too much alcohol Illicit drug use Manic depression/bipolar disorder Schizophrenia Anxiety disorder Dementiac

Na

Overall

50

50

P-valueb

849 849

9.7 (6.7) 46

10.0 (6.8) 47

9.2 (6.6) 44

.10 .43

856 851 851 845

40 4.4 (5.8) 21 13

43 4.5 (5.6) 22 14

36 4.2 (6.1) 18 12

.047 .04 .14 .40

796 791 794 795 798 797 761

34 13 10 4 4 32 11

37 15 12 6 4 34 11

30 9 7 2 3 28 10

.04 .03 .03 .03 .26 .11 .85

Each N takes into account missing data on age (N  14) and for each condition. Based on chi-square test for within age group comparisons; Wilcoxon nonparametric test for CES-D and AUDIT scores. c Defined as “a great deal” or “some” dementia based on provider report. a

b

drug use between older (50 years) and younger HIVinfected patients (Table 4). Younger HIV-infected patients were more likely to report current drinking (43 versus 36%, P  .047), and had higher AUDIT scores (P  .04). Based on provider report, 34% of their patients were depressed, 13% drank too much alcohol, and 10% used illicit drugs (Table 4). Only 47% of patients who screened positive for significant depressive symptoms on the CES-D were recognized by their providers as depressed (data not shown). Although few patients were reported to have mania/bipolar disorder (4%) or schizophrenia (4%), 32% were considered to have anxiety according to provider report. According to providers, 1% of their patients had at least some cognitive impairment or dementia. Provider-reported frequencies were lower for older compared to younger HIV-infected patients for depression (30 versus 37%, P  .04), alcohol use (9 versus 15%, P  .03), drug use (7 versus 12%, P  .03), and mania (2 versus 6%, P  .03).

6. Discussion For HIV comorbidities, we found a higher prevalence of oral candidiasis, peripheral neuropathy, and herpes zoster than PCP and KS. Nonetheless, we found that HIV-infected patients experienced on average a greater number of general medical than HIV comorbidities. Psychiatric disorders were common in this sample as well. A number of HIV patients also reported significant depressive symptoms, at-risk alcohol use, and drug use. A significant number of HIV patients experienced multiple medical and psychiatric problems. Age-related differences were most apparent for general medical and psychiatric comorbidities. General medical comorbidity counts were significantly higher for older (50 years) HIV-infected patients compared to their younger counterparts. Not only were older compared to younger

5.4. Overlapping conditions We created a Venn Diagram demonstrating the overlap of multiple conditions for HIV-infected veterans with significant depressive symptoms, at-risk alcohol or current illicit drug use, and 2 general medical comorbidities. Note that frequencies for conditions listed here differed slightly from the tables due to missing values, as patients with any missing values for depressive symptoms, alcohol/drug use, and comorbidity were excluded from this analysis (N  768). Five percent had all three conditions, 32% had two conditions, and 36% had one condition (Fig. 3). In addition, most HIV-infected patients with substance use problems also experienced significant depressive symptoms.

Fig. 3. Multiple comorbidities (N  768a). a N  768 excludes patients with missing values for drug use (N  23), AUDIT data (N  18), CES-D score (N  19), and comorbidity (N  71).

A.M. Kilbourne et al. / Journal of Clinical Epidemiology 54 (2001) S22–S28

HIV-infected patients more likely to experience comorbidities such as hypertension, CAD, or stroke, but also metabolic-related comorbidities including hyperlipidemia and diabetes. For psychiatric conditions, providers reported a greater prevalence among younger HIV- infected patients, consistent with prior research in the general population [26]. Our findings parallel recent research indicating decreased prevalence for the more common HIV opportunistic infections, which are in part likely due to highly active antiretroviral therapy (HAART) use [27]. Our findings suggest that greater attention should be paid to conditions such as oral candidiasis, peripheral neuropathy, and herpes zoster. Furthermore, our results suggest that “non-HIV” comorbidities (general medical and psychiatric) are highly prevalent. The reasons for the high prevalence of general medical comorbidities are multifaceted, and may be the result of HIV treatment and aging factors. The advent of HAART poses the promise of dramatically increasing survival [28,29]. At the same time, these new medications might be associated with syndromes such as hypertension and hyperlipidemia [8,11,30], both of which were common in our sample. In addition, we found that over half of our sample had chemical hepatitis. Both viral and HIV medication-related hepatitis have contributed to increased morbidity and mortality among HIV infected individuals [6,7,31]. The increased frequency of hepatitis-related adverse outcomes might be due to the combined effects of antiretroviral therapy toxicity [7,31], greater disease susceptibility due to aging, chronic infection with Hepatitis B and/or C [6,31], and chronic alcohol use. There are limitations to this study that warrant consideration. Not all providers completed the questions on general medical comorbidity. We asked providers to report the lifetime prevalence of both HIV and general medical comorbidities. Some HIV comorbidities such as oral candidiasis or MAC tend to be more episodic and, while recorded by the provider, the patient may have no traces of current infection at time of the survey. In contrast, general medical comorbidities such as diabetes or hyperlipidemia tend to be chronic conditions. However, the distinction between HIV and general medical comorbidity frequencies and counts would have been more pronounced if only current conditions were included. Finally, this study was based on data from patients enrolled at three VA infectious disease clinics, and site-specific variation was apparent for some conditions. Hence, our findings may not be generalizable to the rest of the VA patient population. More sites are needed to determine a more generalizable estimate of the prevalence of these conditions. Nevertheless, comparing our HIV comorbidity data with 1996–1997 baseline comorbidity data from a national probability sample of HIV-infected adults in care, the HIV Costs and Services Utilization Study, revealed similar findings adjusted for age (Dr. David Zingmond, personal communication). Further, comparing our results with data from a national survey of eligible veterans, the 1999 Large Health

S27

Survey of Veteran Enrollees [32], demonstrated greater frequencies for general medical comorbidities among this larger (primarily HIV negative) sample. This divergence is likely due to differences in age distribution between the two cohorts, as the mean age for the Veterans survey participants was 60 years. The mean age among VACS 3 participants was 49 years. These age differences underscore the need for further research on comorbidity and outcomes that includes patients with and without HIV disease of similar age distribution. Longitudinal research is required to determine (1) whether HIV infection and HIV treatment impact general medical and psychiatric comorbidities, (2) whether general medical and psychiatric comorbidities affect outcomes including survival among older HIV infected patients, and if so, (3) whether the effect on outcomes is mediated by adequate treatment for general medical and psychiatric comorbidities [33]. As HIV-infected patients live longer, general medical comorbidities are becoming increasingly important in their medical management. As HIV infection requires increasingly intensive and complex medical management, patients may require equally intensive management of psychiatric comorbidity to fully participate in HIV care. The burden of these combined “non-HIV” comorbidities suggests a substantial role for general medical and psychiatric management in HIV infected patients. Acknowledgments The primary funding sources were from a National Institute of Aging (NIA) Career Development Award (Grant #K23 AG 00826-03), a Robert Wood Johnson Faculty Scholar Award, and an Interagency Agreement between NIA, National Institute of Mental Health, and the Department of Veteran Affairs (Dr. Justice, PI). Dr. Justice is a staff physician at the VA Pittsburgh Healthcare System. Unrestricted Educational Grants: GlaxoSmithKline and Agouron (Dr. Justice, PI). Dr. Rabeneck is the recipient of a VA Health Services Research and Development Advanced Research Career Development Award. References [1] Reiter GS. Comprehensive clinical care: managing HIV as a chronic disease. AIDS Clin Care 2000; 12:13–9. [2] Chan IS, Neaton JD, Saravolatz LD, Crane LR, Osterberger J, for the Community Programs for Clinical Research on AIDS. Frequencies of opportunistic diseases prior to death among HIV-infected persons. AIDS 1995;9:1145–51. [3] Jones J, Hanson D, Dworkin M, et al. Surveillance for AIDS-defining opportunistic illnesses, 1992–1997. Morb Mortal Wkly Rep 1999;48:1–22. [4] Munoz A, Schrager LK, Bacellar H, et al. Trends in the incidence of outcomes defining acquired immunodeficiency syndrome (AIDS) in the Multicenter AIDS Cohort Study: 1985–1991. Am J Epidemiol 1993;137:423–38. [5] CDC. 1993 revised classification system for HIV infection and expanded surveillance case definition for AIDS among adolescents and adults. Morb Mortal Wkly Rep 1992;41(RR-17).

S28

A.M. Kilbourne et al. / Journal of Clinical Epidemiology 54 (2001) S22–S28

[6] Bonacini M, Puoti M. Hepatitis C in patients with human immunodeficiency virus infection: diagnosis, natural history, meta-analysis of sexual and vertical transmission, and therapeutic issues. Arch Intern Med 2000;160:3365–73. [7] den Brinker M, Wit FW, Wertheim-van Dillen PM, et al. Hepatitis B and C virus co-infection and the risk for hepatotoxicity of highly active antiretroviral therapy in HIV-1 infection. AIDS 2000;14:2895–902 [8] Graham NM. Metabolic disorders among HIV-infected patients treated with protease inhibitors: a review. J Acquir Immune Defic Syndr 2000;25(Suppl. 1):S4–11. [9] Kotler DP. Nutritional alterations associated with HIV infection. J Acquir Immune Defic Syndr 2000;25(Suppl. 1):S81–7. [10] Friedl AC, Jost CH, Schalcher C, et al. Acceleration of confirmed coronary artery disease among HIV-infected patients on potent antiretroviral therapy. AIDS 2000;14:2790–2. [11] Rerkpattanapipat P, Wongpraparut N, Jacobs LE, et al. Cardiac manifestations of acquired immunodeficiency syndrome. Arch Intern Med 2000;160:602–8. [12] Carr A, Samaras K, Thorisdottir A, et al.. Diagnosis, prediction, and natural course of HIV-1 protease-inhibitor-associated lipodystrophy, hyperlipiddaemia, and diabetes mellitus: a cohort study. Lancet 1999; 353:2093–9. [13] The Forum for Collaborative HIV Research. Metabolic abnormalities in HIV disease and treatment: the need for an interim case definition and a study to examine prevalence. The George Washington University Medical Center, The Forum for Collaborative HIV Research. (GENERIC), 1000, p. 1–20. [14] Department of Health and Human Services and the Henry J. Kaiser Family Foundation. Panel on Clinical Practices for the Treatment of HIV Infection. Guidelines for the Use of Antiretroviral Agents in HIV Infected Adults and Adolescents, February 2001. [15] Justice AC, Whalen CC. Aging in AIDS; AIDS in aging. J Gen Intern Med 1996;11:645–7. [16] Selwyn PA, Goulet JL, Molde S, et al. HIV as a chronic disease: implications for long-term care at an AIDS-dedicated skilled nursing facility. J Urban Health 2000;77:187–203. [17] Sherbourne CD, Hays RD, Fleishman JA, et al. Impact of psychiatric condition on health-related quality of life in persons with HIV infection. Am J Psychiatry 2000;157:248–54. [18] Smola S, Justice AC. Veterans with HIV/AIDS 3 Site Study (VACS 3): overview and description. J Clin Epidemiol, submitted. [19] Charlson ME, Pompei P, Ales KL, et al. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chron Dis 1987;40:373–20. [20] Parkerson GR Jr, Broadhead WE, Chiu-Kit JT. The Duke severity of illness checklist (DUSOI) for measurement of severity and comorbidity. J Clin Epidemiol 1993;46:379–93.

[21] Parkerson GR Jr, Broadhead WE, Chiu-Kit JT. Health status and severity of illness as predictors of outcomes in primary care. J Clin Epidemiol 1995;33:53–66. [22] Andresen EM, Carter WB, Malmgren JA, et al. Screening for depression in well older adults: evaluation of a short form of the CES-D. Am J Prev Med 1994;10:77–84. [23] Mulrow CD, Williams JW Jr, Gerety MB, et al. Case-finding instruments for depression in primary care settings. Ann Intern Med. 1995; 122:913–21. [24] Isaacson JH, Butler R, Zacharek M, et al. Screening with the Alcohol Use Disorders Identification Test (AUDIT) in an inner-city population. J Gen Intern Med 1994;9:550–3. [25] Saunders JA, Aasland OG, Babor TF, et al. Development of the Alcohol Use Disorders Identification Test (AUDIT): project on early detection of persons with harmful alcohol consumption-II. Addiction 1993;88:791–804. [26] Kessler RC, McGonagle KA, Zhao S, et al. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States. Arch Gen Psychiatry 1994;51:8–19. [27] Kaplan JE, Hanson D, Dworkin MS, et al. Epidemiology of human immunodeficiency virus-associated opportunistic infections in the United States in the era of highly active antiretroviral therapy. Clin Infect Dis 2000;30(Suppl. 1):S5–14. [28] Palella FJ, Delaney KM, Moorman AC, et al. Declining morbidity and mortality among patients with advanced human immunodeficiency virus infection. HIV Outpatient Study Investigators. N Engl J Med 1998;338:853–60. [29] The CASCADE Collaboration. Concerted action on seroconversion to AIDS and death in Europe. Survival after introduction of HAART in people with known duration of HIV-1 infection. Lancet 2000;355: 1158–9. [30] Koppel K, Bratt G, Eriksson M, et al. Serum lipid levels associated with increased risk for cardiovascular disease is associated with highly active antiretroviral therapy (HAART) in HIV-1 infection. Int J Stud AIDS 2000;11:451–5. [31] Sulkowski MS, Thomas DL, Chaisson RE, et al. Hepatotoxicity associated with antiretroviral therapy in adults infected with human immunodeficiency virus and the role of hepatitis C or B virus infection. JAMA 2000;283:74–80. [32] Veterans Health Administration (VHA) and Veterans Health Administration Office of Quality Performance and the Center for Health Quality, Outcomes and Economic Research. Health status and outcomes of Veterans: physical and mental component summary scores veterans SF-36; 1999 large health survey of veteran enrollees executive report, May 2000. [33] Adler WH, Nagel JE. Acquired immunodeficiency syndrome in the elderly. Drugs Aging 1994;4:410–6.