To assess the validity of patient self-reported oral health measures as used in a large multi-country survey for populations aged 50+. Information on people's oral health status is important for assessing oral health needs within populations. However, clinical examination is not always possible. Patient self-reported measures may provide an alternative when time and other resources are scarce. Using oral health items from the Survey of Health, Ageing and Retirement in Europe (SHARE), self-reported measures were collected from 186 patients receiving treatment at Heidelberg University Hospital. Self-reports were compared with subsequent clinical examinations. Analyses were conducted for patients of all age groups and separately for patients aged 50+ (analogous to the SHARE study population). Diagnostic accuracy, agreement and correlation of patient-reported information were examined using descriptive statistics and Bland-Altman plots. Patient-reported presence or absence of a full tooth count was closely related to clinical measurement, both for all age groups (sensitivity: 93%; specificity: 92%) and persons aged 50+ (sensitivity: 100% specificity: 94%). Bland-Altman plots indicate good agreement between patient- and clinical reports of the number of teeth at age 50+ (Concordance Correlation Coefficient=0.95). Discriminatory power of patient-reporting was good regarding presence vs absence of artificial teeth, but less robust regarding partial vs full replacement of missing teeth. Patient self-evaluations provide reasonable estimates of clinical measures and appear sufficiently accurate for examining variations in the number of teeth, including among populations aged 50+. However, patient reports of the extent of replacement of missing teeth may not constitute reliable reflections of clinical conditions.