Using the differentiated increase in retirement age across cohorts introduced by the 2010 French pension reform, we estimate the health-consumption effects of a 4-month increase in retirement age. We focus on individuals who were close to retirement age but not retired yet by the time the reform was passed. Using administrative data on individual sick-leave claims and non-hospital health-care expenses, we show that the probability of having at least one sickness absence increases for all treated groups, while the overall number of sick days remains unchanged, conditional on having a sick leave. Delaying retirement does not increase the probability of seeing a GP, except for men in the younger cohorts. In contrast, it raises the probability of having a visit with a specialist physician for all individuals, except men in the older cohorts. Delaying retirement also increases the probability of seeing a physiotherapist among women from the older cohorts. Overall, it increases health expense claims, in particular in the lower part of the expenditure distribution.