Factors Associated With Paranoid Symptoms in a Community Sample of Older Adults 1

Transcription

1 Copyright 1996 by The Cerontological Society of America The Cerontologist Vol.36, No. 1,70-75 of paranoia were found in 9.5 of a community sample of older adults in North Carolina. In cross-sectional analyses, these symptoms were associated most strongly with black race, lower income and education, less exercise, and more depressive symptoms. In longitudinal analysis, paranoid symptoms three years following initial interview were predicted by baseline paranoid symptoms, education and depressive symptoms at the initial interview. In blacks, paranoid symptoms may represent an appropriate response to a hostile environment rather than a psychopathic trait. Key words: Community survey, Black race, Sensory impairment, Elderly Factors Associated With in a Community Sample of Older Adults 1 of paranoia are not rare among older adults. Christenson and Blazer (1984) found that 4 of the elderly reported persecutory symptoms in a community sample of elders. Lowenthal (1964) found 2.5 of community-based elders reported suspiciousness and an additional 2 reported paranoid delusions. The estimated prevalence of paranoid psychoses in late life, however, is much lower. In the Epidemiologic Catchment Area studies the current (one-year) prevalence estimate for schizophrenic disorders among persons 65 + years of age was 0.2 (Keith, Regier, & Rae, 1991). Parsons (1964) found 1.7 of persons 65 + experienced late paraphrenia in a Welsh town population sample. In a random sample of 612 elderly Chinese aged 65 and older, living in Singapore, the current prevalence of paranoid disorder was estimated at 0.5 (Kua, 1992). Clinical investigators have found that a variety of factors contribute to late-life paranoid symptoms, including a lifelong propensity toward paranoid thought, late onset schizophrenia-like illness (late paraphrenia or paranoid psychosis), and a maladaptive response to a sense of loss of control over the environment (Eisdorfer, 1980). Among the physical changes associated with aging, most frequently cited as contributing to paranoid symptoms has been sen- 1 These data were collected via contract NO1-AG from the National Institute on Aging in support of Established Populations for Epidemiologic Studies of the Elderly (Duke University). The views expressed here are the authors' and do not necessarily reflect the views or policies of the U.S. Department of Health and Human Services, nor does mention of trade names, commercial products, or organizations imply endorsement. -J. P. Gibbons Professor of Psychiatry and Behavioral Sciences, Box 3005, Duke University Medical Center, Durham, NC Address correspondence to Dr. Dan G. Blazer, Dean of Medical Education. Assistant Research Professor, Department of Psychiatry and Behavioral Sciences, Duke University Medical Center. Senior Fellow, Center for the Study of Aging and Human Development. Epidemiology, Demography and Biometry Program, National Institutes of Health. Dan G. Blazer, MD, PhD, 2 Judith C. Hays, RN, PhD, 3 and Marcel E. Salive, MD, MPH 4 sory impairment. In the study previously cited (Christenson & Blazer, 1984), poor vision and hearing, along with poor self-rated mental health, poor self-rated physical health, poor economic and social resources, and cognitive impairment were associated with persecutory symptoms. Post (1962, 1966) found 30 of persons with late paraphrenia to suffer from deafness compared to 11 of the elderly suffering from depression. Cooper, Day, and Curry (1974) found a higher frequency of deafness among persons with paranoid psychosis than in controls with affective psychoses. In recent years, however, the association between sensory impairment and paranoid symptoms has been questioned. Prager and Jeste (1993) reviewed 27 published studies examining the association between sensory (vision and hearing) impairment and late-life psychosis with paranoid features. Though a majority of the investigations supported the postulated association between hearing impairment and late-onset paranoid disorder, there were methodological limitations in these studies, such as the absence of a longitudinal perspective. These investigators also reported findings from a case-control study in which a blind assessment of visual and hearing impairments was used to explore this postulated association. Due to an overall high level of sensory deficits among older psychiatric patients, no true differences in sensory deficits could be found between persons with paranoid psychoses and those with other psychiatric disorders. There is an area neglected in the study of paranoid symptoms that these so-called symptoms may express an accurate view of the environment for certain populations in society. For example, older persons living in inner city environments, where drug abuse dominates the culture, may have more than ample reason to perceive that environment as hostile. Regardless of the etiology of paranoid symp- 70 The Gerontologist

2 toms, a study of the natural history and correlates of these symptoms can be useful to understanding community-dwelling elders. To our knowledge, there are no longitudinal studies in the literature relating sensory impairment to the onset of paranoid symptoms. We therefore analyzed data from the Piedmont Health Survey of the Elderly (Established Populations for Epidemiologic Studies of the Elderly Project [EPESE/Duke]) in order to determine the influence of self-reported visual and hearing impairment at baseline upon paranoid symptoms three years later in a controlled analysis. We hypothesized that paranoid thoughts at followup would be associated with previously self-reported deficits in hearing and vision, even controlling for physical and cognitive status and paranoid thoughts at baseline. Methods Sample The data for this study derive from the Established Populations for Epidemiologic Studies of the Elderly Project (EPESE) at Duke University Medical Center. The population survey at Duke was part of a multicenter, collaborative epidemiologic investigation of health status in the physical, social, and cognitive functioning of persons 65 years of age and older living in four communities: East Boston, Massachusetts; Iowa and Washington Counties, Iowa; New Haven, Connecticut; and five counties in the north-central Piedmont of North Carolina (Cornoni-Huntley et al., 1990). The North Carolina study sample consists of 4,162 community residents 65 years of age and older selected from Durham, Granville, Vance, Warren, and Franklin counties of North Carolina, with each of the counties contiguous with at least one other county in the catchment area. Durham County is predominantly urban, whereas the remaining four counties are predominantly rural. The population of Durham County is approximately equal to the combined population of the four rural counties. There is a high proportion of blacks in all five counties, ranging at the time of the 1980 U.S. Census, from one-third in Durham County to nearly two-thirds in Warren County. In order to optimize both racial differences (blacks versus non-blacks) and urban/rural differences, the sample was drawn in order to interview approximately equal numbers of subjects from the urban and rural counties and equal numbers of blacks and non-blacks. Forty-eight percent of the sample were from Durham County, and 52 were from the rural counties. Fifty-four percent of the sample were black, and 46 were non-black (almost exclusively white). A four-stage, stratified sampling design was used. In the first stage, primary sampling units consisting of 450 zones of approximately equal population size were selected from each 1980 census block, block cluster, and enumeration district. In the second stage, a land area within each zone was selected based on population density, and all households within that selected land area were enumerated or listed (n = 26,183). The housing units were stratified by race and residence. The third stage involved screening all listed household units for occupants aged 65 or older. At the fourth stage, the Kish procedure was used to select one person aged 65 or older (n = 4,162) from each eligible household (Kish, 1965). Of the 4,162 respondents, 4,000 contributed faceto-face interviews, and 162 contributed abbreviated proxy interviews with a close acquaintance or family member. The proxy interviews did not include the Center for Epidemiologic Studies Depression Scale (CES-D), so proxies were removed from the analyses. An additional 69 respondents did not answer one or both of the paranoia items from the CES-D, and these persons were also omitted from all analyses. The baseline analyses included 3,931 subjects. At followup three years later, 2,936 subjects, or 88.9 of sample members alive, were interviewed. Missing data varied for different items in the questionnaire, but were minimal except for income status (752 respondents did not provide income data). Measures symptoms were measured by two items from the CES-D, "People were unfriendly" and "I felt that people disliked me" (Radloff, 1977). All questions from the CES-D were included verbatim from the original version of the questionnaire. Response options, however, were collapsed to a yes/no format for reporting the presence or absence of a symptom during the week preceding the interview. The overall correlation of CES-D scores determined by usual methods of scoring and the modified approach was very high (Blazer, Burchette, Service, & George, 1991). In a previous study, the psychometric properties of the CES-D, including its factor structure, were determined in a community sample of married men and women (Ross & Mirowsky, 1984). The factors were, in large part, generalizable across both men and women and included factors of depressed affect, enervation, lack of positive affect, and interpersonal problems. We repeated this analysis and found the factors to be almost identical to those previously reported, with virtually no variability by gender or race. Two items loaded on the "interpersonal problems" factor, i.e., "I felt that people disliked me" and "People were unfriendly." These two items were included in the CES-D to tap the paranoid tendencies in mood disorders. The general perception that "People dislike me" is a phenomenon of one core paranoid symptom, i.e., suspiciousness. The general perception that "People are unfriendly" is a phenomenon of a second core paranoid symptom, i.e., hypersensitivity, which notices any slight, real or imagined. In more recent versions of the psychiatric nomenclature, paranoid symptoms are not considered as central to any but the most severe mood disorders, such as a severe major depression with psychotic features, a disorder which would be extremely rare in community samples of older adults Vol. 36, No. 1,

3 (DSM-IV, 1994). In the Duke EPESE sample, the two "interpersonal problem" items are not highly correlated with the remainder of the CES-D (.30 for "People dislike me" and.29 for "People are unfriendly"), yet they were highly related to each other. Therefore, in these analyses, we used a positive response to either of these items as evidence of paranoid symptoms. The remaining 18 items of the 20-item CES-D scale were used to estimate depression as a control variable. Sensory deficit variables were constructed as follows. Hearing deficit was measured dichotomously and was counted as positive if the respondent answered "no" to the item: "[With/without a hearing aid] can you usually hear and understand what a person says without seeing his face if that person talks in a normal voice to you in a quiet room?" Visual deficit was measured on a continuous scale (range 0-6), as the sum of six items which asked, [When wearing eyeglasses/contact lenses] "can you see well enough to recognize a friend (1) across a street, (2) across a room, (3) who is an arm's length away, (4) if you get close to his face, (5) to read ordinary newspaper print, and (6) to read large print such as newspaper headlines?" The other variables utilized were measured as follows. Age was determined by both self-report of age and report of date of birth. When inconsistencies arose, date of birth was used to measure age. Race was determined by both observation and report of race. Cognitive impairment was determined by using the Short Portable Mental Status Questionnaire (Pfeiffer, 1975). Physical disability was determined by items derived from three physical disability scales, the Katz scale (an activity of daily living scale), the Nagi Scale (a functional limitations scale), and the Rosow and Breslau Scale (a mobility scale; Katz & Akpom, 1976; Nagi, 1976; Rosow & Breslau, 1966). Stressful life events were determined by asking whether a series of stressful events had occurred during the previous year. Characteristics at Wavel Hearing problems 1 Male Female Non-black Black Residence Not married Married Analyses We first developed a series of bivariate analyses using the presence or absence of paranoid symptoms at baseline and follow-up and tested the significance of differences across the variables of interest using chisquare and t-test statistics. Then a logistic regression analysis using the Statistical Analysis System (SAS) software was performed, which regressed Wave 2 paranoia status on Wave 1 paranoia status on all independent variables and covariates measured at baseline. Results In Tables 1 and 2, the characteristics of the total sample are presented. Differences in the characteristics of individuals demonstrating paranoid symptoms (compared to those without these symptoms) are presented for each wave. Of the 3,931 subjects available for study at Wave 1, 2,936 subjects remained in the sample at Wave 2. symptoms were more frequent in bivariate analyses among persons with hearing problems, among females, among blacks, and among persons not married at Wave 1. These trends continued to Wave 2, except for differences by gender disappearing. However, 69 of the subjects who reported paranoid symptoms at the first interview did not report those symptoms at the follow-up interview. Therefore, paranoid symptoms appear not to persist as a trait over the follow-up interval, even though the prevalence and cross-sectional correlates are similar at both interviews. In Table 2, vision problems, older age, less education, lower income, impaired activities of daily living (Katz, 1976), deficits in function (Nagi, 1976) and mobility (Rosow & Breslau, 1966), smaller social networks, less social interaction, presence of negative life events, depressive symptoms, and cognitive impairment were all associated with paranoid symptoms at both waves. (The Katz activities of daily living [ADL] was not significant at Wave 2, although the trend was similar.) Table 1. Characteristics of Total Sample. Differences in Presence of at Wave 1 and Wave 2 by Sample Characteristics: Weighted Chi-square Analyses No Yes Rural Urban Total Sample N = 3,931 (100) n () 3414 (87.2) 501 (12.8) 1376(35.0) 2555 (65.0) 1822 (46.3) 2109 (53.7) 1732(44.1) 2199 (55.9) 2406(61.2) 1525 (38.8) n = 372 (9.5) (81.5)** (18.5) (29.6)* (70.4) (31.5)*** (68.5) (44.6) (55.4) (68.3)** (31.7) Wavel No n = 3,559 (90.5) (87.8) (12.2) (35.6) (64.4) (47.9) (52.1) (44.0) (56.0) (60.5) (39.5) n = 239(8.1) (82.0)** (18.0) (30.5) (69.5) (32.2)*** (67.8) (47.7) (52.3) (70.3)*** (29.7) Wave 2 No n = 2,697(91.9) (88.9) (11.1) (33.4) (66.6) (47.0) (53.0) (44.6) (55.4) (58.7) (41.3) "Sixteen subjects did not report on hearing problems. *.01 < p ss.05; **.001 < p =s.01; ***p « The Gerontologist

4 Table 2. Characteristics of the Total Sample. Differences in Presence of at Wave 1 and Wave 2 by Sample Characteristics: Weighted f-test Analyses Vision problems Age (years) Education (years) Income ADLs (Katz) Functional limitations (Nagi) Mobility (Rosow-Breslau) Social network Social interaction Negative life events Depressive symptoms (CES-D) Cognitive status (SPMSQ) n a Total Sample Mean (SD) 0.3 (0.8) 73.3 (6.5) 8.6 (4.1) 10.4 (10.0) 0.2 (0.7) 1.1 (1.4) 0.9 (1.1) 13.9 (7.1) 13.9 (6.7) 0.7 (0.9) 2.8 (3.0) 0.1 (0.3) Where totals are less than n =, subjects did not provide data. *.01 <p=.05; **.001 <p=s.01; ***p =.001. In Table 3, paranoid symptoms at Wave 1 are regressed on the variables shown in Table 2, estimating two models, one including hearing problems and one including vision problems. As can be seen, in these cross-sectional models, neither hearing nor vision problems are associated with paranoid symptoms. In addition, many of the variables found to be associated with paranoid symptoms in bivariate analyses were not associated in the multivariate logistic analysis. Depressive symptoms, black race, less education, lower income and lower functioning on the Rosow-Breslau scale remain associated with paranoid symptoms. In Table 4, paranoid symptoms at Wave 2 are regressed on paranoid symptoms at Wave 1 in two models, one containing vision problems and one hearing problems. As can be seen, paranoid symptoms at Wave 1 are a strong predictor of paranoid symptoms at Wave 2, as are education and depressive symptoms. Vision problems at Wave 1 exhibited a weak but positive relationship to paranoid symptoms at Wave 2. In order to control for potential bias, we performed some additional analyses (data not shown). First, we compared the association of hearing problems with paranoid symptoms (using alternately selfreported and interviewer-rated hearing problems). The regressions were virtually identical. Second, we found that among those subjects interviewed at Wave 2, paranoid symptoms at Wave 1 did not distinguish between subjects lost to follow-up and subjects reinterviewed at Wave 2. Discussion Wavel No Mean (SD) Mean (SD) 0.4 (0.9)* 74.1 (6.9)* 6.8 (3.5)*** 6.7 (5.9)*** 0.3 (0.8)** 1.7 (1.6)*** 1.2 (1.2)*** 12.3 (7.3)*** 11.7 (6.7)*** 1.0 (1.2)*** 5.9 (3.6)*** 0.2 (0.4)** (0.8) (6.4) (4.1) (10.3) (0.7) (1.3) (1.1) (7.0) (6.7) (0.9) (2.7) (0.3) Mean (SD) 0.5 (1.0)*** 73.8 (6.6)** 6.6 (3.4)*** 6.6 (5.9)*** 0.2 (0.6) 1.5 (1.5)*** 1.2 (1.2)*** 12.6 (7.3)*** 12.7 (6.8)*** 0.9 (1.1)*** 4.9 (3.7)*** 0.1 (0.3)* Wave 2 No Mean (SD) Table 3. Wave 1 Regressed on Hearing Problems (Model 1) or Vision Problems (Model 2) Adjusted for Related Wave 1 Characteristics Characteristics Hearing problems Vision problems Age Female Black Married Education Income ADL (Katz) Functional limitations (Nagi) Mobility (Rosow-Breslau) Social network Social interaction Negative life events Depressive symptoms (CES-D) Cognitive impairment (SPMSQ) Constant Model chi-square (df) Model 1 Beta SE *** * -* ** *** (15) (0.6) (6.0) (4.0) (10.5) (0.6) (1.3) (1.3) (7.0) (6.7) (0.9) (2.7) (0.3) Model 2 Beta SE ** * -* * *** (15) Notes. Logistic regression beta coefficients indicate the effect of an individual variable on the log odds of the outcome event with all the remaining variables held constant; the adjusted odds ratio may be calculated as e bela (exponentiated beta); the confidence interval for a given level of confidence (e.g., 95) may be calculated as ei bela±(1xsf»; ADL = activities of daily living; CES-D = Center for Epidemiologic Studies Depression Scale; SPMSQ = Short Portable Mental Status Questionnaire. *.01 <p=s.05; **.001 <p=s.01; ***p *.001. Sample In this study, we provide the first published data assessing the associates of paranoid symptoms at follow-up in a community sample by self-reported problems in a baseline interview and controlling for paranoid symptoms at baseline. Contrary to our hypothesis, we did not find a relationship between baseline hearing impairment and paranoid symp- Vol. 36, No. 1,

5 Table 4. Wave 2 Regressed on Wave 1. Hearing Problems (Model 1) or Vision Problems (Model 2), Adjusted for Related Wave 1 Characteristics Model 1 Model 2 Characteristics Beta SE Beta SE symptoms (WaveD 1.42*** *** 0.37 Hearing problems Vision problems 0.28* 0.13 Age Black Married Education -0.08* ** 0.04 Income - - Functional limitations (Nagi) Mobility (Rosow-Breslau) Social network - - Social interaction Negative life events Depressive symptoms (CES-D) 0.14*** *** 0.04 Cognitive impairment (SPMSQ) Notes. Logistic regression beta coefficients indicate the effect of an individual variable on the log odds of the outcome event with all the remaining variables held constant. The adjusted odds ratio may be calculated as e bela (exponentiated beta); the confidence interval for a given level of confidence (e.g., 95) may be calculated as ei bcla *"- xsf ". *.01 <p=s.05; **.001 <p=s.01; ***p ^.001. toms at follow-up in controlled analyses. A weak but positive association was found between baseline visual impairment and paranoid symptoms at follow-up. As noted earlier, one explanation for paranoid symptoms in late life is a persistent paranoid thought style over time. These data partially substantiate this hypothesis. The strongest predictor in the models of paranoid symptoms at Wave 2 is paranoid symptoms three years earlier at Wave 1. Overall, however, paranoid symptoms did not increase in frequency among those persons followed-up three years later, and two-thirds of those subjects who reported paranoid symptoms at Wave 1 did not report them at Wave 2. Given the relatively short follow-up interval, it is difficult to determine whether age in fact is associated with increased symptoms. In cross-sectional analyses, age did exhibit a significant positive association with paranoid symptoms. As has been demonstrated elsewhere (Christenson & Blazer, 1984), cross-sectional studies usually find an association of paranoid symptoms with cognitive status, psychiatric symptoms, social network, and functional impairment. This was true in the bivariate analysis in this study. However, in a multivariate analysis of the Wave 1 data, social support measures, age, gender, marital status, negative life events, and cognitive impairment were no longer associated with paranoid symptoms. The effect of cognitive impairment may be lessened by controlling for depressive symptoms. This attenuation of the relationship with paranoid symptoms may also be true of negative life events and social support. Race continued to be a strong correlate of paranoid symptoms in multivariate analyses of the Wave 1 data, but was not a significant predictor of paranoid symptoms in the longitudinal analyses. Perhaps paranoid symptoms are state phenomena in both blacks and nonblacks, yet phenomena determined by different factors depending upon race. In blacks, paranoid symptoms may represent an appropriate response to a hostile environment rather than a psychopathic trait. There are a number of strengths to this study. First, the sample is a large one and is equally distributed between blacks and non-blacks and urban and rural residents. Follow-up was excellent for a community study, and there were no apparent differences between individuals who were interviewed at follow-up and those lost to follow-up. In addition, a number of factors which could contribute to paranoid symptoms other than the independent variables of interest (difficulty hearing and seeing) were available in the data. These included measures of cognitive impairment, functional impairment, and depressive symptoms. The CES-D scale is useful for exploring the relationship between paranoid symptoms and depressive symptoms, given that the two items within the CES-D used to assess paranoid symptoms load on a separate factor, are highly related to each other, and correlate less with other items in the scale. On the other hand, some problems are inherent in the study design. First, only two items were used to assess paranoid symptoms, i.e., "People were unfriendly" and "I felt that people disliked me." Though these items or similar items are frequently used in scales to assess interpersonal difficulties in paranoid thinking, they also may be based in reality. The prevalence of paranoid symptoms using these items, 9.5 at baseline and 8.1 at follow-up, is somewhat higher than has been recorded for paranoid symptoms in previous community surveys (2-4). The estimate is significantly higher than the community-based estimates of schizophrenic disorders in late life (which are usually well below 1 in community samples in older adults). A second problem is that hearing problems were based on self-assessment. The Kappa value between self-assessment and assessment by the interviewer was moderate (.36). The models did not change, however, when interviewer-rated rather than selfreported hearing problems were used. We recognize that self-reported hearing problems cannot be automatically assumed to represent true difficulties with hearing, nor does lack of report of hearing problems necessarily assure that hearing problems are not present. Another problem might be that assessment of visual problems was based on a scale that included multiple items, rather than simply asking whether the subject suffered from visual impairment or not (in contrast to the assessment of hearing difficulties). Using EPESE data from other sites, Salive et al. (1992) found that, while there was some discordance, selfreports of visual acuity correlated well with the mea- 74 The Gerontologist

6 sured activity. Therefore, the concordance between self-report and interviewer report (or measurement) varies by the sensory modality measured. Finally, the reader must not equate the findings in this paper with the considerable literature on the relationship between perceptual problems and late life paraphrenia (or schizophrenic disorders in late life). Though there clearly is an association between schizophrenia-like disorders and endorsement of items used in this survey, persons responding positively on the items used in this study include many more individuals than would have been identified with a standardized interview for psychiatric disorders. Even so, we believe it important to explore the correlates and temporal relationships of paranoid symptoms in community samples, irrespective of diagnosis. References American Psychiatric Association. (1994). Diagnostic and statistical manual of mental disorders (4th ed.). (DSM-IV). Washington, DC: APA. Blazer, D. C, Burchette, B., Service, C, & George, L. K. (1991). The association of age and depression among the elderly: an epidemiologic exploration. Journal of Gerontology: Medical Sciences, 46, M210-M215. Christenson, R., & Blazer, D. (1984). Epidemiology of persecutory symptoms in an elderly population in the community. American journal of Psychiatry, 141, Cooper, A. F., Day, D. W. K., & Curry, B. R. (1974). Hearing loss in paranoid and affective psychoses of the elderly. Lancet, 2, Cornoni-Huntley, J., Blazer, D. C, Lafferty, M. E., Everett, D. F., Brock, D. B., & Farmer, M. E. (1990). Established populations for epidemiologic studies of the elderly. Resource Data Book (Vol. 2). (NIH Publ. No ). Bethesda, MD: National Institutes of Health. Eisdorfer, C. (1980). and schizophrenic disorders in later life. In E. W. Busse & D. C. Blazer (Eds.), Handbook of geriatric psychiatry (pp ). New York: Van Nostrand. Katz, S., & Akpom, C. A. (1976). A measure of primary sociobiological functions. International journal of Health Services, 6, Keith, S. J., Regier, D. A., & Rae, D. S. (1991). Schizophrenic disorders. In L. N. Robbins & D. A. Regier (Eds.), Psychiatric disorders in America (pp ). New York: The Free Press. Kish, L. (196S). Survey sampling. New York: John Wiley & Sons. Kua, E. H. (1992). A community study of mental disorders in elderly Singaporean Chinese using the CMS-ACCAT package. Australian and New Zealand journal of Psychiatry, 26, Lowenthal, M. F. (1964). Lives in distress. New York: Basic Books. Nagi, S. Z. (1976). An epidemiology of disability among adults in the United States. Milbrank Memorial Fund Quarterly, 54, Parsons, P. L. (1964). Mental health of Swansea's old folk. British journal of Preventive and Social Medicine, 19, Pfeiffer, E. (1975). A short portable mental status questionnaire for the assessment of organic brain deficit in elderly patients, journal of the American Geriatrics Society, 23, Post, F. (1962). The significance of affective symptoms in old age. London: Oxford University Press. Post, F. (1966). Persistent persecutory state of the elderly. Oxford: Pergamon Press. Prager, S., & jeste, D. V. (1993). Sensory impairment in late-life schizophrenia (Review). Schizophrenia Bulletin, 79, Radloff, L. S. (1977). The CES-D Scale: A self-report depression scale for research in the general population. Applied Psychological Measures, 1, Rosow, l.,& Breslau,N. (1966). ACuttman health scale for the aged. Journal of Gerontology, 21, Ross, C. A., & Mirowsky, J. (1984). Components of depressed mood in married men and women: the Center for Epidemiologic Studies Depression Scale. American journal of Epidemiology, 119, Salive, M. E., Curalnik, J. M., Christen, W., Glenn, R. J., Colsher, P., & Ostfeld, A. M. (1992). Functional blindness and visual impairment in older adults from three communities. Ophthalmology, 99, Received April 28, 1995 Accepted October 10, 1995 Chief of Geriatrics Department of Internal Medicine The University of California, Davis, School of Medicine is recruiting for a full-time academic position as Chief of the Section of Geriatrics at the Associate Professor or Professor level in the Department of Internal Medicine. The individual must possess an M.D. degree, be board certified/eligible in Internal Medicine and Geriatrics and be eligible for licensure in the State of California. Direction of ongoing home care programs as well as an inpatient Geriatric Consultation Service is expected. The appointee will help develop quality assurance and utilization outcomes analysis for the new primary care physician network as it relates to geriatric issues. Additional objectives will be leadership in organizing and supervising an interdepartmental Geriatrics/Gerontology teaching and research program at the Davis and Sacramento campuses. Please forward CA to: Renee Korte, Department of Internal Medicine, University of California, Davis, 4150 V Street, Suite 3100, Sacramento, CA This position is open until filled, but no later than June 30, The University of California is an affirmative action/equal opportunity employer. Vol.36, No. 1,