BACKGROUND: In this study, we investigated primary care physicians' exercise habits, and the association of this variable with their age, specialty, and workplace. METHODS: The population of this cross-sectional study comprised 3,310 medical doctors who graduated from Jichi Medical University in Japan between 1978 and 2012. The study instrument was a self-administered questionnaire mailed in August 2012 to investigate primary care physicians' exercise habits, age, specialty, and workplace. RESULTS: The 896 available primary care physicians' responses to the self-administered questionnaire were analyzed. Their exercise frequency was as follows: daily, 104 (11.6%); at least 2-3 times per week, 235 (26.2%); no more than once a week, 225 (25.1%); no more than once a month, 278 (31.0%); and other, 52 (5.8%). Their exercise intensity was as follows: high (≥6 Mets), 264 (29.5%); moderate (4-6 Mets), 199 (22.2%); mild, (3-4 Mets), 295 (32.9%); very mild (<3 Mets), 68 (7.6%); none, 64 (7.1%); and other, 6 (0.7%). Their exercise volume was calculated to represent their exercise habits by multiplying score for exercise frequency by score for intensity. Multivariate linear regression analyses showed that the primary care physicians' exercise volumes were associated with their age (P<0.01) and workplace (P<0.01), but not with their specialty (P=0.37). Primary care physicians in the older age group were more likely to have a higher exercise volume than those in the younger age groups (50-60 years > older than 60 years >40-50 years >30-40 years >24-30 years). Primary care physicians working in a clinic were more likely to have a higher exercise volume than those working in a university hospital, polyclinic hospital, or hospital. CONCLUSION: Primary care physicians' exercise habits were associated with their age and workplace, but not with their specialty.