Is the impact of interprofessional primary care teams on health service use mediated by quality of care?

Background: Interprofessional primary care teams, composed of family physicians who collaborate with other health professionals, are considered an effective strategy to counteract the expected burden on the health system due to population aging and the concomitant increased prevalence of multimorbidities. Evidence of the impact of interprofessional primary care teams on quality of care and health service use is mixed, which may be because studies have examined the effect of such teams on each, separately. This study aimed to better understand the mixed evidence by combining all three concepts together to explore the impact of interprofessional primary care teams on health service use, and possible mediation by quality of care.

 

Methods: This was a retrospective longitudinal cohort study using Quebec’s health administrative data (with physician billing code data) linked to the Canadian Community Health Survey (with self-reported sociodemographic information). The target population was community-dwelling adults who were attached to a family physician practicing in an interprofessional primary care team, and who had participated in the 2015 Canadian Community Health Survey.

 

Firstly, we determined whether there was an overall effect of interprofessional primary care teams on health service use indicators (number of nonurgent emergency department visits, total number of emergency department visits, total number of hospitalizations). Then, to better understand whether the impact of interprofessional primary care teams on health service use was mediated by quality of care (access, continuity, coordination), causal mediation analysis was conducted.

 

Causal mediation analysis allowed the decomposition of the total effect of interprofessional primary care teams on health service use into four components. This allowed a better understanding of whether the impact of teams on health service use was mediated by quality of care. The four components were: the effect due to mediation alone, the effect due to interaction alone, the effect due to mediation and interaction, and the effect due to neither mediation nor interaction.

 

Results: Results: for the overall association showed that patient attachment to an interprofessional primary care team was not associated with nonurgent emergency department visits, but was associated with an 11% decrease in emergency department visits and a 21% decrease in hospitalizations.

Causal mediation analysis results showed that continuity and coordination did not mediate the impact of interprofessional primary care teams on health service use. Yet, showed that when access was good, there was a 7% reduction in nonurgent ED visits and an 8% reduction in total ED visits for those attached to an interprofessional primary care team.

 

Implications: This study showed that patient attachment to an interprofessional primary care team does not guarantee an improvement in outcomes. Instead, the effect of interprofessional primary care teams on health service use is conditional on the emphasis on good access within interprofessional primary care teams.

 

Further investigations should consider organizational characteristics of interprofessional teams as moderators of the effect, which could help to better understand their impact on health service use. The results of this study have implications for practice and policy, and results may be generalized to other contexts and settings.