A critical review of the use of telephone tests to identify cognitive impairment in epidemiology and clinical research.

We reviewed the use of telephone tests to identify cognitive impairment. We searched PubMed for epidemiological studies and clinical trials reporting the use of telephone tests to identify cognitive impairment. Validation studies and papers published more than 10 years ago were excluded. A total of 132 abstracts were identified, from which 19 epidemiological studies and four clinical trials were selected. Telephone tests were found to reduce selection bias in epidemiology by including people over large areas and facilitating follow-up in longitudinal studies. The most widely used tests were the Telephone Interview for Cognitive Status (TICS) and its modified version, the TICSm. Interviewing a proxy was included in most of the studies to compensate for the unavailability of some participants because of deafness, disease or death. In the epidemiological studies, results of telephone tests were seldom confirmed by a medical examination. Telephone screening for cognitive impairment to identify individuals eligible for clinical trials is impeded by low efficiency and lack of sensitivity for separating early pathological cognitive impairment from dementia and normal ageing.