CHARACTERISTICS OF NONRESPONDERS AND THE IMPACT OF NONRESPONSE ON PREVALENCE ESTIMATES OF DEMENTIA

Background. Differential distributions of sociodemographic characteris tics and cognitive impairment in responders and non-responders may res ult in a biased prevalence estimate of dementia based on responders on ly, Methods. Responders (n = 2191) to a cross-sectional, two-stage com munity study were compared with regard to sociodemographic characteris tics and cognition with three subgroups of non-responders: (A) subject s who refused to participate (n = 369), (B) subjects who were too ill or who had died prior to the screening (n = 72) and (C) subjects who h ad moved out of the study region or were not traceable (n = 23). Preva lence estimates specific for age and housing situation in responders a nd physicians' ratings of cognitive impairment were used to estimate t he prevalence of dementia among non-responders. Results. Group A diffe red from responders in age and housing situation, group B in age, hous ing and cognition, and group C only in age. Separate prevalence estima tes of dementia based on age, housing and cognition yielded figures fo r group A between 4.9% and 7.2%, for group B between 13.1% and 19.1%, and for group C between 2.6% and 4.2%. Joined with the prevalence rate among responders (6.5%) the best possible point estimate of the preva lence of dementia in the target population lies between 6.4% and 6.9%, i.e. within the 95% confidence interval (CI) of the prevalence among responders (5.4-7.5%). Conclusions. Although in this study non-respons e had no important influence on the overall prevalence, the findings a mong the distinct non-response subgroups point to the importance of de scribing non-response sociodemographically as well as in terms of the study objective. The authors recommend that non-responders are categor ized into distinct groups based on the reason for non-response.