Application of a growth curve approach to modeling the progression of Alzheimer's disease

Yaakov Stern, Xinhua Liu, Marilyn Albert, Jason Brandt, Diane M. Jacobs, Caridad Del Castillo-Castaneda, Karen Marder, Karen Bell, Mary Sano, Fred Bylsma, Ginette Lafleche, Wei Yann Tsai

Research output: Contribution to journalArticlepeer-review

57 Scopus citations


Background. Studies using clinical measures to track AD progression often assume linear declines over the entire course of the disease, which may not be justified. The objective of this study was to model change in measures of the clinical severity of Alzheimer's disease (AD) over time. Methods. We developed a method to apply growth curve models to prospective data and characterize AD patients' functional change over time. Data from the modified Mini-Mental State Examination (mMMSE) and measures of basic and instrumental ADL, administered semiannually for up to 5 years to 236 patients with probably AD, were modeled. Results. The rate of decline in mMMS scores per 6- month interval gradually increased as scores dropped from the maximum of 57 to 20. The rate of decline then decreased as scores approached 0, resulting in an inverse 'S' curve. The rate of increase in instrumental ADL scores per interval attenuated as the scores increased, while that for basic ADL scores across intervals was constant. Conclusions. Differences in the pattern of progression of the three measures is in part a function of their psychometric properties. The progression curves may also reflect content-specific features of the instruments. Superimposition of the modeled decline in these three content areas suggests a hypothetical model of the relative timing of cognitive and functional changes in AD.

Original languageEnglish (US)
Pages (from-to)M179-M184
JournalJournals of Gerontology - Series A Biological Sciences and Medical Sciences
Issue number4
StatePublished - Jul 1996

ASJC Scopus subject areas

  • Aging
  • Geriatrics and Gerontology


Dive into the research topics of 'Application of a growth curve approach to modeling the progression of Alzheimer's disease'. Together they form a unique fingerprint.

Cite this