The University of Manchester longitudinal study of cognition in normal healthy old age, 1983 through 2003

This paper describes the participants, design, method and tests used during a 20-year longitudinal study of cognitive changes in increasing age experienced by 6542 healthy residents of Greater Manchester and Newcastle-upon-Tyne, then aged from 42 to 92 years. Information collected and updated includes demographics and health, scores on two, biennially alternated batteries of cognitive tests, repeated administrations of the Beck and Yesavage depression inventories and of self-reports of stressful life events, self-evaluation and locus of control questionnaires and detailed information on lifestyle, hobbies and occupations, physical and social activities, family circumstances and health history. Records have allowed investigation of rates of cognitive changes from 36 months to 20 years preceding death from a variety of causes. Collection of blood and saliva provide, blood chemistry and cortisol levels to analyse associations of rates of cognitive change to genetic factors, blood chemistry and cortisol levels. A random effects analysis confirms marked effects of drop-out and practice due to repeated testing and shows how true rates of change, and of increases in variability between individuals may be ascertained after these have been identified.