Objective Test the feasibility of the modified telephone interview for cognitive status (TICS‐m) as a screening tool to detect cognitive impairment in a population‐based sample of older subjects. Methods Data were collected from 3,578 participants, age 65–94 years, of the KORA‐Age study. We used analysis of covariance to test for significant sex, age and educational differences in raw TICS‐m scores. Internal consistency was analysed by assessing Cronbach's alpha. Correction for education years was undertaken, and participants were divided in three subgroups following validated cut‐offs. Finally, a logistic regression was performed to determine the impact of sex on cognition subgroups. Results Internal consistency of the TICS‐m was 0.78. Study participants needed approximately 5.4 min to complete the interview. Lower raw TICS‐m scores were associated with male sex, older age and lower education (all p < 0.0001). After correction for education years, 2,851 (79%) had a non‐impaired cognitive status (score >31). Male sex was independently associated with having a score equal to or below 27 and 31 (OR = 1.9, 95%CI 1.4–2.5 and OR = 1.5, 95%CI 1.2–1.7, respectively). Conclusions The TICS‐m is a feasible questionnaire for community‐dwelling older adults with normal cognitive function or moderate cognitive impairment. Lower cognitive performance was associated with being a man, being older, and having fewer years of formal education. Copyright © 2012 John Wiley & Sons, Ltd.