Selective non-response to clinical assessment in the longitudinal study of aging: implications for estimating population levels of cognitive function and dementia.

OBJECTIVE: to identify the cognitive outcome of interviewed participants who did not progress to partake in clinical assessments in a longitudinal aging study.
DESIGN: a retrospective study was conducted on participants who were interviewed but who did not complete the clinical assessment (including an extended cognitive assessment) at either Wave 1 or both Wave 1 and Wave 3 of the Australian Longitudinal Study of Ageing. A total of 1947 participants aged 70 and older commenced the study, 246 participants without clinical data at either or both Waves 1 and 3 were identified for the sub-sample followed-up retrospectively. The Informant Questionnaire for Cognitive Decline in the Elderly (IQCODE) was administered to informants and medical records were reviewed.
RESULTS: participants who did not complete the clinical assessment at Wave 3 reported poorer health and had poorer cognitive function at Wave 1 independent of age and gender. Rates of possible dementia or cognitive decline were higher in the group who did not undertake the clinical assessment compared with both those who did the clinical assessment and with population data.
CONCLUSION: selective non-response to clinical assessment in a longitudinal aging study is associated with higher risk of cognitive decline and probable dementia. Longitudinal aging studies may underestimate rates of dementia and population levels of cognitive decline. Copyright 2002 John Wiley & Sons, Ltd.