Background: Studies on trends in the self-rated health (SRH) of older people have shown conflicting results, which might partly be explained by changing associations between SRH and indicators of other health dimensions over time. Therefore, this study investigates 17-year time trends in older adults' poor SRH, in the context of trends in chronic diseases and disability, between 1992 and 2009. Methods: Data originate from six measurement waves of the Longitudinal Aging Study Amsterdam (N = 4009, ages 60-85 years). SRH was assessed with the question 'How is your health in general?' The presence of lung disease, cardiac disease, peripheral arterial disease, diabetes mellitus, stroke, arthritis and cancer was assessed by self-report. Two severity levels of disability were assessed with six questions on physical functioning. Generalized Estimating Equations (GEE) analysis was applied to assess statistical significance in each time trend. Results: There was a stable trend in the prevalence of poor SRH and severe disability, while the mean number of chronic diseases (1.3-1.8) and the prevalence of mild disability (20.5-32.1%) increased between 1992 and 2009. The association between poor SRH and chronic diseases became weaker, whereas the association between poor SRH and severe disability became stronger over time. Most unfavourable trends were observed in the older old and the lower educated. Conclusion: Our results suggest that the seeming stability of poor SRH hides underlying increases in chronic diseases and disability: over time, people may attach importance to different aspects of health when rating their overall health. © 2012 The Author. Published by Oxford University Press on behalf of the European Public Health Association. All rights reserved.