Features Corless Vol. / Cognitive Functioning in JANAC et al.11, No. 3, May/June 2000 HIV/AIDS Predictors of Perception of Cognitive Functioning in HIV/AIDS Inge B. Corless, RN, PhD, FAAN, Suzanne Bakken, RN, DNSc, FAAN, Patrice K. Nicholas, RN, MPH, DNSc, William L. Holzemer, RN, PhD, FAAN, Chris A. McGibbon, PhD, Jillian Inouye, RN, PhD, Kathleen M. Nokes, RN, PhD, FAAN, Joan G. Turner, RN, DNS, CIC, Gail M. Powell-Cope, ARNP, PhD, Marie-Annette Brown, RN, PhD, FAAN, and Carmen J. Portillo, RN, PhD, FAAN This is a descriptive, correlational study of the pre- Key words: HIV/AIDS, cognition, functioning dictors of perceived cognitive functioning. The conve- nience sample of 728 nonhospitalized persons receiv- Numerous neurological changes, characterized by a ing health care for HIV/AIDS was recruited from seven combination of cognitive, motor, and behavioral alter- sites in the United States. All measures were self- ations, occur across the spectrum of HIV disease reported. Self-perception of cognitive functioning, the (Selnes et al., 1990; Snider et al., 1983). Although neu- dependent variable, was composed of three items from rological change is widely noted in advanced HIV dis- the Medical Outcomes Study HIV scale: thinking, ease, researchers have debated the evidence of cogni- attention, and forgetfulness. Data related to age, gen- tive decline during the asymptomatic stages (Grant, der, ethnicity, education, injection drug use, CD4 Atkinson, Hesselinki, Kennedy, & Richman, 1987; count, and length of time known to be HIV-positive Grassi et al., 1997; Poutiainen, Elovaara, Raininko, were collected on a demographic questionnaire. The Vikki, Lahdeverta, & Iwanainen, 1996; Poutiainen, scale from the Sign and Symptom Checklist for Persons Irvanainan, Elovaara, Valle, Landervirta, 1988; Selnes with HIV Disease was used to measure self-reported et al., 1997; Wilkie et al., 1998). Villa et al. (1996) symptoms. Data were analyzed using hierarchical argued that there are “subtle but significant cognitive multiple regression analysis. Predictors of perception abnormalities” (p. 130). of cognitive functioning explained a total of 36.3% of the variance. Four blocks—person variables (1.5%) Inge B. Corless, RN, PhD, FAAN, Patrice K. Nicholas, RN, (age, gender, education, history of injection drug use), MPH, DNSc, and Chris A. McGibbon, PhD, are at the MGH disease status (2.3%), symptom status (26.5%), and Institute of Health Professions, Boston. Suzanne Bakken, functional status (5.4%)—significantly contributed RN, DNSc, FAAN, William L. Holzemer, RN, PhD, FAAN, statistically to the total variance. Among those indi- and Carmen J. Portillo, RN, PhD, FAAN, are at the Univer- viduals who completed the questions related to depres- sity of California, San Francisco. Jillian Inouye, RN, PhD, is at the University of Hawaii, Manoa. Kathleen M. Nokes, sion (n = 450), 28% of the variance in cognitive func- RN, PhD, FAAN, is at Hunter College, New York, New York. tioning was explained by this variable. The findings in Joan G. Turner, RN, DNS, CIC, is at the University of Ala- this multi-site study indicate that symptom status bama, Birmingham. Gail M. Powell-Cope, ARNP, PhD, is at explained the largest amount of variance in perceived Jack Haley Veterans Hospital, Tampa, Florida. cognitive functioning. Early identification of cognitive Marie-Annette Brown, RN, PhD, FAAN, is at the University impairment can result in appropriate clinical interven- of Washington, Seattle. This research was supported by the tions in remediable conditions and in the improvement of nursing research program of the research and development quality of life. department of Glaxo-Wellcome, Inc. JOURNAL OF THE ASSOCIATION OF NURSES IN AIDS CARE, Vol. 11, No. 3, May/June 2000, 19-26 Copyright © 2000 Association of Nurses in AIDS Care 20 JANAC Vol. 11, No. 3, May/June 2000 Neurologic problems associated with HIV disease biologic and physiologic variables and symptom and include central nervous system infections, neoplasms, functional status. vascular complications, peripheral neuropathies, and myopathies. HIV dementia (HIVD) (also called AIDS dementia complex, subacute encephalitis, HIV Conceptual Framework encephalopathy, and HIV-1-associated cognitive and motor complex), is often identified based on three The conceptual framework for this study was the main categories of symptoms: cognitive, motor, and Wilson and Cleary (1995) model, linking clinical vari- behavioral. The primary symptom is forgetfulness, ables with health-related quality of life. The model which is associated with slowed mental and motor included five sets of variables: physiological and bio- abilities. Loss of balance and leg weakness are early logical factors (including disease status), symptom motor signs. The most commonly observed behavioral status, functional status, general health perceptions, symptoms are apathy and social withdrawal, which are and overall quality of life (Wilson & Cleary, 1995). In often diagnosed as depression. Sometimes, organic addition, selected individual and environmental vari- psychoses, such as acute mania, may be a primary ables are hypothesized to affect model components. manifestation of HIV dementia (Evans & Perkins, For this study, only the first three components and 1990; Simpson & Tagliati, 1994). selected individual characteristics were used. Table 1 The early course of HIV disease may be marked by specifies the study measures by components of the symptoms and signs of dementia too subtle to identify. Wilson and Cleary model and delineates the manner in Several studies suggest that neuropsychological tests which the measures were used in the hierarchical are useful in providing quantitative markers of disease regression analysis to answer the following questions: progression (Derix, de Gans, Stam, & Portigies, 1990; Sidtis, Ornitz, & Price, 1988; Tross et al., 1988). With 1. What are the perceptions of cognitive function- advanced dementia, cognitive impairment becomes ing as reported by persons with HIV on the Medi- more obvious with psychomotor problems and behav- cal Outcomes Study–HIV questionnaire? ioral abnormalities. Wilkie et al. (1998) argued that 2. Do individual characteristics, disease status, “poor neuropsychological test performance prior to symptom status, and functional status explain a the development of AIDS is a proximal predictor of significant amount of the variance in perceptions mortality” (p. 130). Early detection is key to interven- of cognitive functioning as reported by persons ing in this downward trajectory. Factors in addition to with HIV on the Medical Outcomes Study–HIV HIV/AIDS that independently might account for dec- questionnaire? rements in cognition include drug abuse, apathy, and depression (Castellon, Hinkin, Wood, & Yarema, 1998; Grassi et al., 1997). Selnes et al. (1997), how- Method ever, found no significant differences in cognitive per- formance among drug-using and non-drug-using Design HIV-infected persons. Their study suggested that “the natural history of cognitive changes secondary to HIV The study design was descriptive and correlational, infection is similar among HIV-infected IDU’s [injec- focusing on the predictors of perception of cognitive tion drug users] and other risk groups” (p. 223). functioning in HIV/AIDS. Setting Purpose The sample was recruited from seven sites: Bir- The purpose of this study was to examine the rela- mingham, Boston, Honolulu, New York, the San Fran- tionships between perception of cognitive functioning cisco Bay Area of northern California, Seattle, and and other variables. Perception of cognitive function- Tampa. The settings included university-based AIDS ing, the dependent variable, is viewed as influenced by clinics, private practices, public and for-profit Corless et al. / Cognitive Functioning in HIV/AIDS 21 Table 1. Summary of Study Measures by Components of Wil- study were recruited to participate. Inclusion criteria son and Cleary (1995) Framework and Block in the were fourfold: that the individuals be at least 18 years Regression Analysis of age, receiving HIV care at their respective facilities, Wilson and English speaking, and able to provide informed consent. Cleary Framework/ Component/Study Study Regression Measure Measure Analysis Procedures Individual characteristics Demographic Race/ethnicity Block 1 The institutional review boards of the coordinating questionnaire site and each data collection site approved the study Individual characteristics protocol. Following completion of informed consent, Demographic Age, gender, high school Block 2 each client completed the set of study instruments. questionnaire diploma, current/past HIV-infected individuals who met the study criteria injection drug use Biological/physiological were approached by either the study site coordinator or (disease status) researcher, their health care provider, or their case Demographic Years known HIV positive, Block 3 managers in seeking their participation. Study person- questionnaire CD4 count, AIDS nel provided assistance if needed for completion of the diagnosis, number of instruments (e.g., reading questions to persons with comorbidities decreased visual acuity). Symptom Status Sign and Symptom SSC–HIV: Diarrhea, Block 4 Checklist–HIV shortness of breath, fever, Study Measures (SSC–HIV) nausea/vomiting, gastrointestinal All measures were self-reported by clients. Each discomfort, fatigue, psychological distress measure is briefly described in the following section. scores Center for CES–D: Total depression Independent Variable Epidemiological score Study–Depression Demographic questionnaire. Data related to indi- Scale (CES–D) vidual characteristics (race and ethnicity, age, gender, Medical Outcomes MOS SF–36: Bodily pain education, intravenous drug use) and physiological/ Study–Short Form score biological factors and disease status (CD4 count, pres- 36 (MOS SF–36) ence of AIDS diagnosis, length of time known to be Functional status HIV positive) were collected on a demographic MOS SF–36 Physical functioning, Block 5 questionnaire. social functioning, role- emotional functioning Functional status Sign and Symptom Checklist–HIV. The Sign and Medical Outcomes Cognitive functioning Depen- Symptom Checklist–HIV (SSC-HIV) was used to Study–HIV dent measure symptom status. Each symptom is rated on a (MOS–HIV) variable 4-point scale, with values ranging from 0 (absent) to 3 (severe). Evidence of construct and concurrent valid- ity, internal consistency reliability, and sensitivity to change over time in the study sample was reported by hospitals, residential and day care facilities, commu- Holzemer, Henry, et al. (1999). The initial scale nity-based organizations, and home care. included 41 signs and symptoms. The following symptoms or symptom clusters were used as predictor Sample variables in the analysis: diarrhea, fatigue, fever, gas- trointestinal discomfort, nausea and vomiting, short- A convenience sample (n = 728) of HIV-infected ness of breath, and psychological distress and men and women who met the inclusion criteria for the depression. 22 JANAC Vol. 11, No. 3, May/June 2000 Center for Epidemiological Studies–Depression 3. Did you have difficulty doing activities involving Scale. The Center for Epidemiological Studies– concentration and thinking? Depression Scale (CES-D) (Radloff, 1977) was designed as a self-report depression scale for research Response options for the three questions are on a in the general population. Responses on the 20-item 6-point scale, with values ranging from all of the time scale range from 0 (never or rarely) to 3 (mostly or all to none of the time. The responses were summed to the time). Total possible scores range from 0 to 60, and generate the perception of cognitive functioning score scores above 15 indicate the need for diagnostic evalu- (dependent variable) for the analysis. ation for major depression. The CES-D was used as a measure of symptom status (predictor variable) in the Data Analysis analysis. All instruments were scored and scale scores calcu- Medical Outcomes Study–Short Form 36. The Med- lated where appropriate. Descriptive statistics were ical Outcomes Study–Short Form 36 (MOS-SF36) is a performed on all variables. Nominal variables were generic measure of selected components of quality of dummy coded for inclusion in the analysis. A hierar- life, including symptom status, functional status, gen- chical multiple regression analysis with variables, eral health perceptions, and overall quality of life entered in five blocks (see Table 1), was performed, (Ware & Sherbourne, 1992). Scales included physical with perception of cognitive functioning as the functioning, role-physical, role-emotional, body pain, dependent variable. Only participants with data on all social functioning, mental health, general health per- study variables were entered into the regression ception, energy and fatigue, and changes in health analysis. (Stewart, Hays, & Ware, 1988). Scores are trans- formed on a scale of 0 to 100, with 100 meaning opti- mum health. The reliability and validity of the Results MOS-SF36 are well established. Bodily pain was entered into the analysis as a measure of symptom status. Sample Three scales (physical functioning, role-emotional, and social functioning) from the MOS-SF36 were The study sample (n = 728) was 77% male, with an used as predictor variables as measures of functional average age of 39.4 (SD = 7.7). The ethnicity of the status. sample was African American (39.3%), Asian and Pacific Islander (3.7%), Latino (8.9%), Caucasian Dependent Variable (43.1)%, and Other (4.5%). Reported risk factors for HIV infection included the following: male-male sex Medical Outcomes Study–HIV. The Medical Out- (50.2%), male-female sex with bisexual male (5.2%), comes Study–HIV scale (MOS-HIV), formerly male-female sex with intravenous drug user (16%), known as the Health Status Questionnaire, was devel- male-female sex with hemophiliac (1.1%), and oped as an HIV-specific measure of aspects of health male-female sex with transfusion recipient (1.1%). status and quality of life (Wu et al., 1991). Perception The percentage of the sample reporting current or past of cognitive functioning, the dependent variable in this intravenous drug use was 39.3% (see Table 2). study, was composed of the responses to three MOS- The descriptive statistics on the study measures are HIV questions, which included the following: summarized in Table 3. The sample reported mild symptom scores, with 1. How much of the time during the past month did mean scores close to 1.00 (3-point scale). Highest you forget things that happened (e.g., where you means were reported for symptom clusters involving put things, appointments)? psychological distress (anxiety, depression, and fear). 2. Did you have trouble keeping attention on any Participants reported a mean CES-D depression score activity for long? of 21.91 (SD = 12.21) and a mean SF36 body pain Corless et al. / Cognitive Functioning in HIV/AIDS 23 Table 2. Sample Characteristics Table 3. Descriptive Statistics on Study Instruments Variable n % Measure/Variable n M SD Range Gender HIV sign and symptom checklist 696 0.87 0.95 0-3 Male 558 76.6 Diarrhea 699 0.83 0.76 0-3 Female 166 22.8 Shortness of breath 688 0.64 0.75 0-3 Missing 4 0.5 Fever 698 0.70 0.77 0-3 Race/ethnicity Nausea/vomiting 699 0.89 0.78 0-3 Asian/Pacific Islander 27 3.7 Gastrointestinal discomfort 686 1.12 0.99 0-3 Black 286 39.3 Fatigue 695 1.24 0.99 0-3 Latino 65 8.9 Psychological distress 695 1.24 0.99 0-3 White 314 43.1 Other 33 4.5 CES Depression Scale Total 450 21.91 12.21 0-54 Missing 3 0.4 Medical Outcomes Study– AIDS diagnosis Short Form 36 Yes 446 61.3 Bodily pain 722 58.53 25.29 0-90 No 269 37.0 Physical functioning 714 39.23 28.74 0-100 Missing 13 1.8 Role—physical 716 46.18 42.30 0-100 High school diploma Social functioning 724 64.42 27.77 0-100 Yes 547 75.1 Role—emotional 714 45.58 43.98 0-100 No 170 23.4 Missing 11 1.5 NOTE: CES = Center for Epidemiological Studies. Telephone in home Yes 592 81.3 No 131 18.0 Missing 5 0.7 education, history of injection drug use), disease sta- Adequate income to meet daily needs tus, symptom status, and functional status—made sig- Yes 575 79.0 Barely Possible 403 55.4 nificant statistical contributions toward explaining the No 142 9.5 total variance. Person variables explained 1.5% (p = Missing 11 1.5 .027) of the variance in cognitive functioning. Disease status explained an additional 2.3% (p = .005). Symp- N M SD Maximum Minimum tom status explained 26.5% (p < .001) of the variance. Age 713 39.4 7.7 64 20 The block for functional status explained 5.4% (p < CD4 count 535 322.7 244.7 1,500 1 .001) of the variance. There was no significant rela- tionship with race and ethnicity (Block 1) and cogni- tive functioning (p = .145). score of 58.53 (SD = 25.29). The mean SF36 score on Block 2 included person variables, with age, gen- social functioning was 64.42 (SD = 27.77), indicating der, education, and current or past injection drug use relatively high functioning, whereas the role-physical found to be significantly related to higher scores on mean score was 46.18 (SD = 42.30). The physical perception of cognitive functioning. Disease status functioning mean score was 39.23 (SD = 28.74), and explained an additional 2.3% of the variance in per- the role-emotional mean score was 45.58 (SD = ception of cognitive functioning scores. This variance 43.98), indicating a wide variability in these scores. was accounted for by the variables of years known Nearly 40% of the sample (n = 286) indicated a history HIV-positive, CD4 counts, AIDS diagnosis, and of injection drug use. co-morbidities. Symptom status rendered a statisti- cally significant contribution to the explanation of the Regression Analyses total variance (26.5%), primarily accounted for by depression. Depression had a negative Beta weight Predictors of perception of cognitive functioning (–0.158), indicating that lower scores on depression explained a total of 36.3% of the variance (see Table 4). were associated with higher cognitive functioning Four blocks—person variables (age, gender, scores. 24 JANAC Vol. 11, No. 3, May/June 2000 Table 4. Predictors of Cognitive Functioning Discussion Standardized Predictor Final Beta R2 R2∆ p∆ Concerns about cognitive functioning in HIV dis- Block 1: Race/ethnicity 0.015 0.015 .145 ease have arisen with the observation of decreased per- Black formance across the spectrum of HIV disease. Poorer Asian Latino cognitive functioning is found with neurological com- White plications associated with HIV, including crypto- Block 2: Person 0.021 0.015 .027 sporidiosis, toxoplasmosis, and lymphoma. A direct Age effect on neurological and cognitive functioning as a Gender result of HIV infection has also been suspected. In High school diploma Current/past injection addition to these physical causes, symptoms including drug use depression, apathy, and fatigue have also been impli- Block 3: Disease status 0.044 0.023 .005 cated in changes in cognitive functioning (Corless Years known HIV+ 0.030 .447 et al., 1998). The findings in this multisite study indi- CD4 count –0.002 .967 AIDS diagnosis 0.020 .657 cate that symptom status explained the largest amount Comorbidities 0.002 .956 of the variance in perception of cognitive functioning. Block 4: Symptom status 0.309 0.265 < .001 The relationship between cognitive functioning and Diarrhea –0.009 .843 depression in HIV disease requires further study. The Fatigue 0.015 .810 linkage between these variables and the causal mecha- Fever –0.035 .515 Gastrointestinal –0.073 .201 nisms requires further investigation across the spec- Nausea/vomiting 0.013 .800 trum of HIV disease. Furthermore, the impact of inten- Shortness of breath –0.063 .212 sive medication therapies and their potential side Pain –0.081 .116 effects, as well as the relationship with depression and Psychological distress –0.101 .073 Depression –0.158 .009 neurological functioning, should be further examined. Block 5: Functional status 0.363 0.054 < .001 In addition, the mechanism underlying the relation- Physical functioning –0.172 .001 ship between symptom status and cognitive function- Social functioning 0.125 .023 ing requires explication. Emotional role Functional status also was found to be significantly functioning 0.183 .001 related to cognitive functioning, with physical, social, and role-emotional performance contributing to the variance in cognitive functioning. The link between physical, social, and emotional variables and cognitive Functional status, including the variables of physi- functioning requires further study to determine the cal, social, and role-emotional functioning, accounted pathway between progression of disease and cogni- for a total of 5.4% of the additional variance in percep- tion. Impairments in cognition have been associated tion of cognitive functioning. Physical functioning had with decreased adherence to medication, advice and a negative Beta weight (–0.172). Lower scores on instructions, and maintenance of health care provider physical functioning were related to higher scores on appointments (Ware & Sherbourne, 1992). Factors perception of cognitive functioning (e.g., less forget- found by researchers to be associated with decrements ful, less trouble maintaining attention, less difficulty in cognition include the following: neurological, psy- with activities). This indicates that declines in physical chiatric, and psychological factors (Stewart, Hayes, & functioning were not associated with declines in self- Ware, 1998), and apathy (Castellon et al., 1998). In perception of cognitive functioning. Higher scores on turn, cognitive impairment, or “mental slowing,” was social functioning and role-emotional functioning related by some investigators (Bouwman et al., 1998; were associated with higher perception of cognitive Villa et al., 1996) to more rapid disease progression functioning scores. and even to death (Wilkie et al., 1998). Corless et al. / Cognitive Functioning in HIV/AIDS 25 The precise advent of cognitive impairment is a cannot be overstated; and assessment of social and role matter of debate. Villa et al. (1996) found a prevalence functioning may give the health care provider a clue as of 28.2% related to cognitive abnormalities in 78 to the cognitive health status of the client. asymptomatic HIV-seropositive individuals for which Future research should examine subjective and no cause other than their infection was established. objective measures of cognitive functioning and other Other researchers (Poutiainen et al., 1988, 1996; neurological manifestations of HIV disease, because Selnes et al., 1990, 1997) have found no cognitive individuals living with this disease may perceive decline in asymptomatic HIV-infected individuals. impairment in cognitive functioning prior to clinical For those individuals with lower scores on social and observation of impairment. Studies should be under- role-emotional functioning, perceptions of cognitive taken to examine the onset and trajectory of neurologi- functioning scores were lower. Measures of role and cal changes as well as the impact of HAART therapy social factors may be more sensitive indicators than on cognitive functioning. physical functioning in the perception of decline in Clinicians can help move the field forward by shar- cognitive functioning. ing their knowledge and observations regarding cogni- In this study, our relatively healthy cohort (M CD4 = tive function in HIV-positive persons with their 323) experienced substantial cognitive impairment, research colleagues. By using the findings from this perhaps as a later manifestation of disease than as a study, clinicians with a heightened awareness will be decline in physical functioning. A small subset of sensitive to potential impairments of cognitive func- these research participants are past or present injection tioning in their clients. Early identification of cogni- drug users. The role of injection drug use on cognitive tive impairment can result in appropriate clinical inter- functioning in HIV-infected persons remains unclear. ventions in remediable conditions and the improve- Grassi et al. (1997) found a negative effect, whereas ment of quality of life. Selnes et al. (1998) found no effect. The impact of antiretrovirals on cognitive disorders References has been positive. In a study of 154 HIV-infected indi- viduals randomized to receive zidovudine or no Bouwman, F. H., Skolasky, R. L., Hes, D., Selnes, O. A., Glass, antiretrovirals, those receiving zidovudine, regardless J. D., Nance-Sproson, T. E., Royal, W., Dal Pan, G. J., & McArthur, J. C. (1998). Variable progression of HIV-associ- of disease stage, displayed no progression in their ated dementia. Neurology, 50, 1814-1820. dementia, in contrast with those not receiving an active Castellon, S. A., Hinkin, C. H., Wood, S., & Yarema, K. T. (1998). drug (Lopez, Wess, Sanchez, Dew, & Becker, 1998). Apathy, depression and cognitive performance in HIV-1 infec- Other medicinal approaches to the reduction of tion. Journal of Neuropsychiatry and Clinical Neuroscience, HIV-related cognitive impairment included the use of 10(3), 320-329. OPC-14117 (Dana Consortium, 1997), deprenyl and Corless, I. B., Holzemer, W., Turner, J., Nokes, K., Brown, M. A., Powell-Cope, S., & Inouye, J. (1998). Cognition and HIV thioctic acid (Dana Consortium, 1998), and the use of adherence. In XII World Aids Conference (Conference Record Peptide T (Heseltine et al., 1998). OPC-14117 and No. 32387.P.599). Geneva, Switzerland. deprenyl had a positive impact on cognitive impair- Dana Consortium on the Therapy of HIV Dementia and Related ment, but the results require further research. Cognitive Disorders (1997). Safety and tolerability of the anti- This study did not examine the relationship oxidant OPC-14117 in HIV-associated cognitive impairment. Neurology, 49, 142-146. between various medications and cognitive impair- Dana Consortium on the Therapy of HIV Dementia and Related ment, as persons who may be experiencing cognitive Cognitive Disorders (1998). A randomized, double-blind, pla- problems might or might not be reliably identifying cebo-controlled trial of deprenyl and thioctic acid in human their own impairment. immunodeficiency virus–associated cognitive impairment. Recommendations based on the results of this study Neurology, 50, 645-651. Derix, M. M., de Gans, J., Stam, J., & Portigies, P. (1990). Mental are threefold: The use of a self-report, patient-based changes in patients with AIDS. Clinical Neurology and Neuro- assessment of perception of cognitive functioning may surgery, 92, 215-222. be useful for longitudinal assessment; the importance Evans, D. L., & Perkins, D. O. (1990). The clinical psychiatry of of clinician assessment of cognitive functioning AIDS. Current Opinion in Psychiatry, 3, 96-102. 26 JANAC Vol. 11, No. 3, May/June 2000 Grant, I., Atkinson, J.H.M, Hesselinki, J. R., Kennedy, C. J., & Selnes, O. A., Galai, N., McArthur, J. C., Cohn S., Royall III, W., Richman, D. D. (1987). Evidence for early central nervous syn- Esposito, D., & Vlahov, D. (1997). HIV infection and cognition drome (AIDS) and other human immunodeficiency virus in intravenous drug users: Long-term follow-up. Neurology, (HIV) infections. Annals of Internal Medicine, 107, 828-836. 48, 223-230. Grassi, M. P., Perin, C., Clerici, F., Zochetti, C., Borella, M., Selnes, O. A., Miller, E., McArthur, J., Gordon, B., Munoz, A., Cargnel, A., & Mangoni, A. (1997). Effects of HIV Sheridan, K., Foz, R., Saah, A. J., & the Multicenter AIDS seropositivity and drug abuse on cognitive function. European Cohort Study (1990). HIV-1 Infection: No evidence of cogni- Neurology, 37, 48-52. tive decline during the asymptomatic stages. Neurology, 40, Heseltine, P.N.R., Gordkin, K., Atkinson, J. H., Vitiello, B., 204-208. Rochon, J., Heaton, R. K., Eaton, E. M., Wilkie, F. L., Sobel, E., Sidtis, J. J., Ornitz, A. H., & Price, R. W. (1988). Neuropsycho- Brown, S. J., Feaster, D., Schneider, L., Goldschmidts, W. L., & logical and neurological characterization of the AIDS dementia Stover, E. S. (1998). Randomized, double-blind placebo- complex. Journal of Clinical Experimental Neuropsychology, controlled trial of Peptide T for HIV-associated cognitive 10, 76. impairment. Archives of Neurology, 55, 41-51. Simpson, D. M., & Tagliati, M. (1994). Neurologic manifestations Holzemer, W. L., Corless, I. B., Nokes, K. M., Turner, J. G., of HIV infection. Annals of Internal Medicine, 121(10), 769- Brown, M. A., Powell-Cope, G. M., Inouye, J., Henry, S. B., 785. Nicholas, P. K., & Portillo, C. J. (1999). Predictors of self- Snider, W. D., Simpson, D. M., Nielsen, S., Gold, J. W., Metroba, reported adherence in persons living with HIV disease. AIDS C. D., & Posner, J. B. (1983). Neurological complications of Patient Care, 13(3), 185-197. acquired immune deficiency syndrome: Analysis of 50 Holzemer, W. L., Henry, S. B., Nokes, K. M., Corless, I. B., Brown, patients. Annals of Neurology, 14, 403-418. M. A., Powell-Cope, G. M., Turner, J. G., & Inouye, J. (1999). Stewart, A. L., Hays, R. D., & Ware, J. E., Jr. (1988). The MOS Validation of the Sign & Symptom Check-List for Persons with short-form general health survey: Reliability and validity in a HIV Disease (SSC-HIV). Journal of Advanced Nursing, 30(5), patient population. Medical Care, 26, 724-735. 1041-1049. Tross, S., Price, R. W., Navia, B. A., Thaler, H. T., Gold, J., Hirsch, Lopez, O. L., Wess, J., Sanchez J., Dew, M. A., & Becker, J. T. D. A., & Sidtis, J. J. (1988). Neuropsychological characteriza- (1998). Neurobehavioral correlates of perceived mental and tion of the AIDS dementia complex: A preliminary report. motor slowness in HIV infection and AIDS. Journal of AIDS, 2, 81-88. Neuropsychiatry and Clinical Neuroscience, 10, 343-350. Villa, G., Solida, A., Moro, E., Tavalozza, M., Antinori, A., Poutiainen, E., Elovaara, I., Raininko, R., Vilki, J., Lahdevirta, J., & DeLuca, A., Murri, R., & Tamburrini, E., (1996). Cognitive Iivanainen, M. (1996). Cognitive decline in patients with symp- impairment in asymptomatic stages of HIV infection. Euro- tomatic HIV-1 infection. No decline in asymptomatic infec- pean Neurology, 36, 125-133. tion. Acta Neurologica Scandinavia, 93, 421-427. Ware, J., & Sherbourne, C. (1992). The MOS-36 Short-Form Poutiainen, E., Iivanainen, M., Elovaara, I., Valle, S. L., & Health Survey (SF-36). Medical Care, 30, 473-483. Lahdevirta, J. (1988). Cognitive changes as early signs of HIV Wilkie, F. L., Goodkin, K., Eisdorfer, C., Fesster, D., Morgan, R., infection. Acta Neurologica Scandinavia, 78, 49-52. Fletcher, M. A., Blaney, N., Baum, M., & Szapocanik, J. Radloff, L. S. (1977). The CES-D scale: A self-report depression (1998). Mild cognitive impairment and risk of mortality in scale for research in the general population. Applied Psycho- HIV-1 infection. Journal of Neuropsychiatry, 10(2), 125-132. logical Measures, 1, 385-411. Wilson, I. B., & Cleary, P. D. (1995). Linking clinical variables Selnes, O. A., Galai, N., Bacellar, M. A., Miller, E. N., Becker, with health-related quality of life: A conceptual model of J. T., Wesch, J., Van Gorp, W., & McArthur, J. C. (1998). Cog- patient outcomes. Journal of the American Medical Associa- nitive performance after progression to AIDS: A longitudinal tion, 273, 59-65. study from the Multicenter AIDS Cohort Study. Neurology, 45, Wu, A. W., Rubin, H. R., Matthews, W. C., Ware, Jr., J. E., Byrck, 267-275. L. T., Hardy, W. D, Bazzette, S. A., Spector, S. A., & Richman, R. D. (1991). A health status questionnaire using 30 items from the Medical Outcomes Study: Preliminary validation in per- sons with early HIV infection. Medical Care, 291, 786-798.