P-151

S146 P-148

Posters: P-148 THE RESOURCE UTILIZATION IN DEMENTIA (RUD) INSTRUMENT IS VALID TO ASSESS INFORMAL CARE TIME IN DEMENTIA

Anders Wimo1, Linus Jo¨nsson2, Arthur S. Zbrozek3, 1Karolinska Institutet, Stockholm, Sweden; 2European Health Economics, Stockholm, Sweden; 3Wyeth Pharmaceuticals, Collegeville, PA, USA. Contact e-mail: [email protected] Background: Informal care is a significant component of the societal resource use and costs in dementia care. Thus it is fundamental that assessments of informal care are valid. Objective: This study aimed to analyse the validity of time estimates in the Resource Utilization in Dementia (RUD) instrument in a community setting. Methods: Comparisons between diaries, recall and time observations with agreement statistics. Results: Fourteen married and cohabiting pairs were included, participating in 47 diaries and 30 observation sessions. The agreement between diary time and estimated time was very good for personal ADL (intraclass correlation (ICC) 0.93), good for supervision (ICC 0.87) and total time (ICC 0.91) and lower but acceptable for instrumental ADL (ICC 0.75). Exclusion of outliers with 24 hours of total time estimates in the diaries did not change the ICC figures (PADL 0.94, IADL 0.83, supervision 0.89, total time 0.90). Regarding observed time vs estimated time, the corresponding figures were for personal ADL (ICC 0.81), for instrumental ADL (ICC 0.74), for supervision (ICC 0.78) and for total time (ICC 0.80). Conclusion: Recall, which is the least time demanding method for collecting data on informal care, is with the RUD instrument a valid method to assess the amount of informal care. Assessment of IADL is, however, a bit problematic. P-149

PHYSICAL EXERCISE AT MIDLIFE AND RISK OF DEMENTIA IN A POPULATION BASED STUDY OF SWEDISH TWINS

Ross Andel1, Michael Crowe2, Nancy L. Pedersen3, Laura Fratiglioni3, Boo Johansson4, Margaret Gatz5, 1University of South Florida, Tampa, FL, USA; 2University of Alabama at Birmingham, Birmingham, AL, USA; 3The Karolinska Institute, Stockholm, Sweden; 4Go¨teborg University, Go¨teborg, Sweden; 5University of Southern California, Los Angeles, CA, USA. Contact e-mail: [email protected] Background: Physical exercise has long been recognized as an effective strategy to promote physical health and reduce the risk of cardiovascular and all-cause mortality. Recent animal and human studies have suggested that regular exercise may support cerebrovascular health, reduce oxidative stress and brain tissue loss, and facilitate normal cognitive function. Objective: We explored the association between physical exercise at midlife and subsequent risk of dementia among members of the HARMONY study. Methods: Measures of exercise were obtained by the Swedish Twin Registry an average of 31 years prior to dementia assessment. Dementia was diagnosed using a twostage procedure, screening for cognitive impairment followed by full clinical evaluation. We used two study designs: Case-control analyses included 264 cases with dementia (176 had Alzheimer’s disease) and 2,870 controls; co-twin control analyses included 90 twin pairs discordant for dementia. Results: In case control analyses, controlling for age, sex, education, diet (eating fruits and vegetables), smoking, drinking alcohol, body mass index, and angina, light exercise such as gardening or walking and regular exercise involving sports were associated with reduced odds of dementia compared to hardly any exercise (odds ratio [OR]: 0.63, 95% confidence interval [CI]: 0.43-0.91 for light exercise, OR: 0.34 95% CI: 0.16-0.72 for regular exercise). Hard physical training was not significantly protective (OR 0.70, 95% CI 0.40-1.24). Findings were similar for Alzheimer’s disease alone. In co-twin control analyses, controlling for education, the association between higher levels of exercise and lower odds of dementia approached significance (OR 0.50, 95% CI 0.23-1.06, p ⫽ .072). Conclusions: Exercise at midlife may reduce the odds of dementia in older

adulthood, suggesting that exercise interventions should be explored as a potential strategy for delaying disease onset. Supported by NIA grant No. R01 AG08724, P30 AG17265, and by Alzheimer’s Association/Zenith Fellows Award. P-150

PREVALENCE OF COGNITIVE IMPAIRMENT IN OLDER PEOPLE IN A RURAL POPULATION OF MEXICO

Elva D. Arias-Merino, Elva Merino-Villasen˜or, Martha J. Arias-Merino, Rosa M. Meda-Lara, Universidad de Guadalajara, Cucs, Gerontologia, Guadalajara, Jalisco, Mexico. Contact e-mail: [email protected] Background: In Mexico, during the twentieth century, there were important transformations in the demographic, economic and social aspects. Most activities appeared to be industrial and of services, generating a strong emigration from countryside to the city, modifying the life styles. In our country it is almost unknown the cognitive impairment of the elderly in non-urban areas. Objective(s): To evaluate the cognitive impairment in the elderly of the Municipality of Tecolotla´n, Jalisco, Mexico. Methods: Tecolotla´n is a Mexican rural municipality of 2.305 people of 60 or more years, 13.35% of the total population. The study was made in 571 elders, house by house, the screening intervention consisted of a structured questionnaire that included socio-demographic aspects (age, gender, civil state and education). The Mini-Mental State Examination (MMSE) in Spanish, 30itms, cut off (19/20). The functional state was measured through The Barthel Activities of Daily Living Index (IADL) 10-itms, in Spanish, was classified like independent (80-100) and dependent (0-79) and The Geriatric Depression Scale (GDS) in Spanish, 30-itms, classified with cut-off (10/11) without depression/with depression. The prevalence of cognitive impairment was estimated. The association between the risk factors and the cognitive state was analyzed using odds ratio (ORs) CI 95%. Results: The study sample consisted of 311(54.5%) women and 260(45.5%) men, the mean age was of 72.3 ⫾ 8.5 years , 88% with 0 to 4 education years. 67.1% married and 21% widowers. The mean MMSE was of 23.1⫾ 6.14, GDS 4.8 ⫾ 3.44 and of ADL 94.2 ⫾13. The prevalence of cognitive impairment was of 24.5%, depression 8.2%, and functional impairment 8.6%. The association between the risk factors and the cognitive state were: Age ⱖ75⫽ 3.52(2.31-5.37), female 1.11(0.74-1.67), education of 0 to 4 years 4.91(2.10-12.05), without couple 2.64(1.74-4.01), with depression 2.49(1.29-4.82), and functional impairment 7.22(3.69-14.23). Conclusions: Since there exists a concern to maintain the elderly living in independence in its own community, greater efforts are required to know the environmental risks and the socioeconomic context of the non-urban populations of the country, and to be able to develop programs of prevention and promotion of health. P-151

CO-ADMINISTRATION OF ESTRADIOL AND CORTISOL MODULATES BETA-AMYLOID, IGF ACTIVITY, AND COGNITIVE FUNCTION FOR OLDER POSTMENOPAUSAL WOMEN

Laura D. Baker1, Sanjay Asthana2, Suzanne Craft1, Brenna Cholerton1, George Merriam1, Charles W. Wilkinson1, Stephen R. Plymate1, Mark Fishel1, Pattie S. Green1, G. Stennis Watson1, Mark A. Reger1, 1 University of Washington School of Medicine/VAPSHCS, Seattle, WA, USA; 2University of Wisconsin-Madison/William S. Middleton Veterans Hospital, Madison, WI, USA. Contact e-mail: [email protected] Background: Estradiol has been linked to improved learning and memory in numerous animal studies. However, reports regarding cognitive response by postmenopausal women to estradiol administration are heterogeneous. One potentially influential variable affecting this response is stress; a hypothesis supported by several reports in the animal literature demon-

Posters: P-152 strating stress-induced amelioration of estradiol effects on cognition. Objective: To evaluate whether estradiol-induced effects on cognitive function will be disturbed when cortisol levels are exogenously elevated in older postmenopausal women, and the effects of this manipulation on blood levels of beta-amyloid (A␤40 & A␤42) and insulin-like growth factors (IGFs). Methods: Forty subjects (56-84 yrs) were enrolled into a placebo-controlled, double blind, parallel-group design study and randomized to receive 0.10 mg/day of transdermal ␤-estradiol or a placebo skin patch for 8 weeks. In the last 4 days of the trial, subjects were also randomized to receive 90 mg/day (30 mg TID) of oral hydrocortisone or matched placebo. Cognitive test scores and blood levels of estradiol, cortisol, insulin-like growth factors, and beta-amyloid (A␤40 A␤42) were quantified at baseline, and at weeks 4 and 8. Results: Estradiol and cortisol levels attained confirmed compliance. Relative to baseline, performance on the Stroop improved when estradiol was administered alone, and worsened when cortisol was administered alone. Co-administration obliterated each of these effects. For tasks of working memory and visual memory, estradiol alone benefited performance while co-administration of estradiol and cortisol had a deleterious effect. Verbal declarative memory improved with estradiol, regardless of whether cortisol was also administered. Cortisol increased plasma levels of IGF-I and A␤42, but only when administered without estradiol. For A␤40, estradiol administered alone increased levels, cortisol administered alone decreased levels and again, the combined regimen altered each of these effects. Conclusions: This is the first clinical study to document that elevated cortisol levels do in fact modulate biological and cognitive response to estradiol. The cognitive findings, with associated changes in beta-amyloid levels, begin to unmask an influential factor that might very well detract from a salutary response to estradiol for older postmenopausal women, and thereby account for inconsistent findings in estradiol intervention trials. P-152

HUMAN BRAIN MYELINATION AND AMYLOID BETA DEPOSITION IN ALZHEIMER’S DISEASE

George Bartzokis, Po H. Lu, Jim Mintz, UCLA, Los Angeles, CA, USA. Contact e-mail: [email protected] Background: The extensive scope of myelination is the single-most unique aspect in which the human brain differs from that of other species. In this “myelin model” of human evolution and development, our brain’s extensive myelination accounts for the high processing speeds underlying our higher cognitive and behavioral functions as well as our unique susceptibility to develop Alzheimer’s Disease (AD). In older age the breakdown of myelin integrity likely underlies both the trajectory of age-related decline in cognitive functions as well as the “age risk factor” for the increasing protein deposits that have been used to define the pathology of AD. Objective: To test the hypothesis that myelin breakdown in vulnerable late-myelinating regions releases oligodendrocyte- and myelin-associated iron thus promoting amyloid beta (A␤) oligomerization, its associated toxicity, and the deposition of oligomerized A␤ and iron in neuritic plaques observed in AD. Methods and Results: The model was tested using published maps of cortical myelination from year 1901 and recent in vivo imaging maps of A␤ deposits in humans. The data show that in AD, radio labeled ligands detect A␤ deposition in a distribution that matches the map of late-myelinating regions. Furthermore, the failure of imaging ligands to bind A␤ in animal models is consistent with their much lower levels of myelin and associated iron levels when compared to humans. Conclusions: The hypotheses derived from the myelin model of the human brain are testable with current imaging methods that can measure myelin integrity, brain iron levels, and protein deposits in vivo. The myelin model of the human brain has important implications for preventive and therapeutic interventions and suggests that such interventions should be expanded to include novel targets such as oligodendrocytes, myelin, and brain iron.

P-153

S147 ANTIOXIDANT SUPPLEMENT USE AND RISK OF DEMENTIA AND ALZHEIMER’S DISEASE IN OLDER ADULTS

John C. S. Breitner1,2, Shelly L. Gray3, Melissa L. Anderson4, Paul K. Crane5, Wayne McCormick5, James Bowen6, Linda Teri7, Eric B. Larson4, 1GRECC (S-182) VA Puget Sound Health Care System, Seattle, WA, USA; 2Dept of Psychiatry and Behavioral Sciences, University of Washington, Seattle, WA, USA; 3University of Washington, Seattle, WA, USA; 4Center for Health Studies, Group Health Cooperative, Seattle, WA, USA; 5Dept of Medicine, University of Washington, Seattle, WA, USA; 6Dept of Neurology, University of Washington, Seattle, WA, USA; 7Dept of Psychosocial & Community Health, University of Washington, Seattle, WA, USA. Contact e-mail: [email protected] Background: Oxidative pathways may have an important role in the etiology of Alzheimer’s disease (AD) and vascular dementia. Studies of dietary or supplemental antioxidant vitamins as potential interventions for prevention of dementia have produced conflicting results. Objective(s): To examine whether use of vitamins C or E, alone or in combination, is associated with reduced incidence of dementia and AD in a community-based prospective cohort study. Methods: 2969 members of the Group Health Cooperative, Seattle, Washington, aged 65 or older, were recruited as participants in the Adult Changes in Thought study. Participants were without evidence of cognitive impairment at enrollment. They were then examined biennially to identify incident dementia and Alzheimer’s disease diagnosed by standard criteria. At each examination participants were asked whether they had taken vitamins C, E, or multivitamins (MVI) for at least one week during the preceding month (“users”). Results: During a mean follow-up of 6.4 (SD 2.3) years, 405 participants developed dementia (289 attributed to Alzheimer’s disease). Use of vitamin E was not associated with dementia (adjusted hazard ratio [aHR] 0.98, 95% confidence interval [CI] 0.77-1.25) or AD (aHR 1.04, CI 0.78-1.39). Comparable results were noted for vitamin C alone (dementia aHR 0.90, CI 0.71-1.13; AD aHR 0.95, CI 0.72-1.25) and for concurrent use of vitamins C and E (dementia aHR 0.93, CI 0.72-1.20; AD aHR 1.00, CI 0.73-1.35) or E with MVI (dementia aHR 1.01, CI 0.57-1.81; AD aHR 1.03, CI 0.44-2.39). Similar results were found when we excluded participants with low screening scores at baseline (who might therefore have begun using vitamins to slow progression of prodromal symptoms). Conclusions: These data provide no evidence that use of supplemental vitamins E or C, either alone or in combination, is associated with reduced risk of dementia or AD over 6 years of follow-up. The contrast of these results with others may reflect different population nutritional intake, or dosage of supplements, or methods of exposure classification. P-154

DEPRESSION AS A RISK FACTOR FOR DEMENTIA

Jessica A. Brommelhoff1, Margaret Gatz1,2, Nancy L. Pedersen2,1, 1 University of Southern California, Los Angeles, CA, USA; 2Karolinska Institutet, Stockholm, Sweden. Contact e-mail: [email protected] Background: Studies have shown that depression and dementia frequently coexist, but whether depression is a risk factor or prodrome of dementia remains controversial. Objective: This study tested whether history of depression was associated with significantly increased risk of developing dementia, and whether those with a greater time difference between the age of onset of depressive episodes and onset of dementia were at greater risk for developing dementia compared with those who had a first depressive episode occurring closer in time to dementia onset. Methods: Dementia was clinically diagnosed by an assessment team using DSM-IV criteria, with age of onset estimated by using both informant reporting and medical records. Depression information, including occurrence and age of onset, was collected from four sources: 1) a national inpatient discharge registry, 2) a national psychiatric hospital discharge registry, 3) medical history for cases and co-twin controls, as reported by an informant, and 4) medical records for cases and co-twin controls. This study used two designs_case-control and co-twin control. In the case-control design, twins diagnosed with dementia were compared