Primary Care Visit Rates Among Canadian Veterans in Ontario: A Retrospective Cohort Study of Sex- and Length of Service-stratified Comparisons With Nonveterans

Introduction Research comparing primary care (PC) use among veterans and nonveterans has not widely considered the impact of sex and length of service on the association between veteran status and PC use. We calculated relative differences in the rate of PC visits between Canadian Armed Forces and Royal Canadian Mounted Police veterans and nonveterans overall and by sex and length of service. Materials and Methods We conducted a matched, retrospective cohort study of Canadian veterans and nonveterans residing in Ontario, Canada between 1990 and 2019 using routinely collected linked administrative health care data held at ICES (formerly known as the Institute for Clinical Evaluative Sciences). We compared PC visit rates using multivariable Andersen-Gill (AG) recurrent event regression models. Effect measure modification by sex and length of service was investigated using statistical interaction terms. Results Overall, veterans had a higher adjusted relative rate (aRR) of PC visits compared to nonveterans (aRR 1.06, 95% CI 1.04-1.07). Male veterans had an aRR of 1.07 (95% CI, 1.05-1.09), while females had an aRR of 1.31 (95% CI, 1.26-1.36). Veterans who served for <5 years had a significantly higher rate of PC visits relative to nonveterans (RR 1.09, 95% CI 1.03-1.15), while veterans who served for >= 30 years had comparable rates to nonveterans (RR 1.00, 95% CI 0.97-1.02). Conclusions Veterans had an overall higher rate of PC visits compared to nonveterans, and the effect of veteran status appeared stronger among females and veterans with fewer years of service. The observed differences in rates of PC use could be the result of increased need, increased access to PC, or proactive health care-seeking behaviors retained from military service.