Objectives The objective of this study was to evaluate the 2‐year effects of a stepped‐care programme to prevent the onset of a major depressive disorder (MDD) in older people living in residential homes. Methods A 2‐year follow‐up study of a pragmatic randomised controlled trial was conducted in 14 residential homes in the Netherlands. A total of 185 residents (Center for Epidemiologic Studies Depression Scale score >7), who did not meet the diagnostic criteria for MDD, were randomised to a stepped‐care programme (n = 93) or to usual care (n = 92). Stepped‐care participants sequentially underwent watchful waiting, a self‐help intervention, life review and a consultation with the general practitioner. The primary outcome measure was the incidence of MDD during a period of 2 years, according to the Mini International Neuropsychiatric Interview. Results After 2 years, the incidence of MDD was not significantly reduced in the intervention group compared with the control group (incidence rate ratio: 0.98; 95% confidence interval (CI) [0.54, 1.81]). However, in the completer analysis, on the basis of 79 residents who completed the 2‐year measurements, there was a significant difference in favour of the intervention group (incidence rate ratio: 0.53; 95% CI [0.32, 0.87]). Dropout percentages were high (44%), mostly accounted for by illness and death (68%). Conclusion A minority of residents had benefit from the intervention that sustained after 2 years in the completer group. Yet, these findings cannot be generalised as the majority of the residents did not opt for participation in the project and many dropped out. Ways should be sought to motivate residents with depressive symptoms to engage in preventive interventions. Copyright © 2013 John Wiley & Sons, Ltd.