Evaluating the impact of integrating physiotherapists in primary care teams in Canada: Results of two cluster randomized trials

Background: Models of care that integrate physiotherapists within primary care teams for people with musculoskeletal conditions are becoming more common worldwide as a strategy to provide more accessible, integrated, coordinated, and continuous care. This presentation reports on two studies that examined the effectiveness of integrating physiotherapists within primary care teams for people with low back pain (LBP, study 1) and hip or knee pain (study 2) on patient health outcomes, health service utilization, societal costs, and access capacity of primary care teams.

 

Approach: Two cluster randomized trials with primary care practices (clusters) randomized 1:1 to a physiotherapist-led primary care model or usual physician-led care (usual care) for LBP and hip/knee pain. Our interdisciplinary research team included health service researchers, patients, and health professionals throughout the research process.

 

Results: In study 1, we randomized 20 clusters and recruited 739 participants with LBP. In study 2, we randomized 14 clusters and recruited 647 participants with hip/knee pain. The PT-led model resulted in: improved functioning and quality of life for people with LBP [mean difference (MD) on the Roland Morris Disability Questionnaire = -1.61 (95%CI: -2.55, -0.67); MD on the EQ5D = 0.06 (0.02, 0.09)] and similar functioning and quality of life for people with hip/knee pain [MD on the Lower Extremity Functional Scale = 0.10 (-4.78, 4.97); MD on the EQ5D = -0.01 (-0.05, 0.04)]. In both studies, the PT-led model led to less participants receiving diagnostic imaging [Rate Ratio (RR) = 0.49 (0.36, 0.65) for LBP; RR = 0.54 (0.35, 0.82) for hip/knee pain], with a similar proportion visiting an emergency department [RR = 0.92 (0.53, 1.61) for LBP; RR = 1.83 (0.59, 5.69) for hip/knee pain] or a physician specialist [RR = 0.68 (0.40, 1.18) for LBP; RR = 0.62 (0.26, 1.46) for hip/knee pain]. The PT-led primary care model showed a trend towards fewer visits with primary care physicians or nurse practitioners among participants in the LBP trial (1.7 versus 2.1 visits/person/year), but not in the hip/knee trial (1.7 visits/person/year in both groups). Costs and secondary health outcomes will also be presented.

 

Implications: These two cluster trials provide the most rigorous evidence to-date on the impact of integrating physiotherapists within primary care teams for people with musculoskeletal conditions. The findings have direct policy and practice implications for expanding team-based primary care. Results also highlight the need for detailed process evaluations to understand the differences in outcomes between people with LBP and those with hip/knee pain.