Physical activity (PA) can improve cardiorespiratory status, strength, body composition and quality of life for patients infected with HIV. Evidence from HIV-uninfected populations also shows that PA is associated with a lower risk of mortality, primarily death due to cardiovascular causes. There is, however, a lack of data on how physically active HIV-infected patients are. In this study, we assessed levels of self-reported PA over time in patients enrolled in the Swiss HIV Cohort Study, a large multicentre prospective observational cohort study. We included a total of 10,540 patients who completed at least one report of PA between December 2009 and November 2014 during routine clinical follow-up (scheduled every 6 months). In the first year after December 2009 there was a higher rate of non-response so these data are of a lesser reliability. Over the next four years, the percentage of patients reporting no free-time PA at all declined from 49% to 44%. In contrast, in two "Sport Switzerland" surveys of the general population in 2008 and 2014, the percentage of individuals reporting no sports activities at all was considerably lower and relatively stable over time (27% in 2008; 26% in 2014). In our analysis, the percentage of patients reporting sedentary activity at work increased from 23% to 26% over the four years. Subgroup findings suggest differences between women and men and between patients classified by their age, stage of infection and CD4 cell count. Integrating PA counselling into the routine care of HIV-infected patients and promoting PA among this population has the potential to improve the general state of health and quality of life for HIV-infected patients and reduce their risk of cardiovascular disease.