Introduction - Prevalence surveys of nosocomial infections (NIs) conducted in military training hospitals (HIA) since 1995 are part of the national surveillance policy. The objective of this study was to compare the prevalence and risk factors for HIs between two prevalence surveys (1996 and 2006), in the context of the abolition of mandatory military service. Methods - The inclusion criteria and definitions from the 2006 prevalence survey were applied to the 1996 data to the extent possible. For each patient, the presence of a NI and exposure to certain invasive procedures were recorded. The 1996 NI prevalences were standardized to the age structure of the patient population included in the 2006 survey. Prevalence ratios for infected patients, adjusted for the risk factors studied, were calculated. Results - In total, across the nine HIAs, 2,242 patients were included in 1996 and 1,500 in 2006. The mean age of patients was significantly higher in 2006 (58.7 years, median: 62.0 years) than in 1996 (47.8 years, median: 46.0 years) (p = 10⁻⁴). The year of the study (2006 vs. 1996) was associated with a non-significant increase in surgical site infections (SSI) (OR = 1.4; [95% CI: 0.8–2.4]). Discussion - Conclusion - The population of military hospitals is increasingly older, with poorer general health and risk factors for surgical site infections (SSIs) approaching those observed in short-stay public hospitals. The lack of a significant decrease in the prevalence of NIIs during this period should encourage the Armed Forces Health Service to intensify its prevention efforts in the area of hospital hygiene in order to achieve a significant reduction in the incidence of NIIs, particularly SIS. (R.A.)