Wednesday, 20 August 2003
This presentation is part of : Health Economics and Public Health Implications of Alzheimer's Disease with and without Medical Co-Morbidities

S073-003 Factors Associated with AD Disease Progression (Including Comorbidities) to Nursing Home Placement and Mortality

Joel Hay, Department of Pharmaceutical Economics & Policy, University of Southern California, Los Angeles, CA, USA and Danielle Zammit, Pharmaceutical Economics and Policy, University of Southern California, Los Angeles, CA, USA.

Objective: To evaluate Alzheimer's disease survival and nursing home free survival (NHFS); To evaluate comorbidities, cognitive function and other risk factors as predictors of AD patient survival and nursing home admission.

Design: Retrospective panel data analysis.

Methods: Minimum Uniform Data Set (MUDS) cohort data, collected through and maintained by the Alzheimer Research Centers of California (ARCC) were analyzed. Survival and nursing home free survival (NHFS) of 3,150 Alzheimer disease patients enrolled in ARCC between 1992-2000 were evaluated using multivariate survival models. Cox proportional hazard model (PH model) and linear and nonlinear regression models were applied to cognitive function, comorbidities, patient demographics, and other risk factors as multivariate predictors of patient survival and NHFS.

Results: History of open head injury, history of myocardial infarction, history of stroke and history of diabetes were all associated with significantly decreased survival, but had no significant effect on NHFS. COPD history was associated with significantly decreased survival and with decreased survival subsequent to nursing home entry, but not with risk of nursing home entry. History of unintentional weight loss and mobility impairment was associated with significantly decreased survival and decreased NHFS. BMI (body mass index) was associated with significantly decreased NHFS. History of hearing impairment was associated with increased NHFS. Proportional hazards models also showed that females, African-Americans and Hispanics had lower hazard of death. Age at onset, years of education, and history of major psychosis in patient’s primary relatives were associated with higher hazard of death. Marriage and female gender decreased the hazard of nursing home entrance on average, while increased body mass index raised it. The MMSE cognition score was a strongly significant positive predictor of both AD patients’ survival and NHFS in all models.

Conclusions:While several comorbid conditions influence overall Alzheimer patient survival as expected, only COPD, weight and mobility impairment were significantly related to nursing home placement after adjusting for other demographic and medical history characteristics.

Back to S073 Health Economics and Public Health Implications of Alzheimer's Disease with and without Medical Co-Morbidities
Back to The Eleventh International Congress