Abbott JH1, Wilson R1, MOA Trial Team A2
1University of Otago Medical School, Centre for Musculoskeletal Outcomes Research, Dept of Surgical Sciences, Dunedin, New Zealand, 2University of Otago, Dunedin, New Zealand
Background: There is strong evidence of clinical effectiveness supporting exercise therapy and manual therapy in people with hip and knee osteoarthritis (OA), but treatment effects are assumed to diminish over time. There is very little evidence in the literature relating to their long-term effects and cost-effectiveness.
Purpose: To investigate the 5-year and lifetime net monetary benefit of providing exercise therapy and/or manual therapy, in addition to usual medical care, in people with hip and knee OA.
Methods: We conducted an economic evaluation alongside a RCT to 5-year follow-up. Adults meeting the American College of Rheumatology criteria for hip or knee OA were randomly allocated to receive either exercise therapy; manual therapy; combined exercise therapy and manual therapy; or no trial physioterapy intervention; in addition to usual medical care. Costs and health-related quality of life data were collected at baseline, 6 months, 1, 2 & 5 years. We developed and validated a state-transition simulation model of the clinical course of OA and, from 5-year observed data, simulated the lifetime costs, health-related quality of life (HRQoL), and incremental net monetary benefit (INMB), to the health system and society, in New Zealand (NZ) dollars.
Results: A total of 299 patients were assessed for eligibility, of which 206 met the study eligibility criteria and agreed to participate. Of the 195 participants surviving at 5-year follow-up, 129 (66%) returned completed survey questionnaires.Missing data were replaced using multiple imputation. Exercise therapy showed statistically significant HRQoL gains (0.12, 90% CI 0.07 to 0.15) at 5-year follow-up, relative to usual medical care only. Manual therapy (0.04, -0.00 to 0.09) and combined therapy (0.03, -0.03 to 0.07) did not. INMBs were positive, indicating cost-effectiveness of all three interventions, at all willingness-to-pay thresholds, from both the health system and societal perspectives. Exercise therapy was cost-effective in all analyses, with the largest (and only statistically significant) INMBs. Combined therapy was the least cost-effective treatment. At the NZ population aggregate level, INMB of exercise therapy, at the 1xGDP/capita willingness-to-pay level, of $19.8 billion (health system perspective) and $24.1 billion (societal perspective) over the lifetime of the current NZ adult population. All interventions remained cost-effective relative to usual care at all parameter levels considered in the one-way sensitivity analyses. Assumptions regarding eligibility criteria and intervention uptake rates had modest impacts on estimated INMBs, and differences in the cost of delivering physiotherapistservices had very little impact.
Conclusion(s): The three physiotherapist-delivered, individualised programmes of exercise or manual therapy provided were found to be cost-effective over a 5-year and a lifetime analysis horizon, relative to usual medical care only, for the treatment of hip and knee osteoarthritis, from both the health system and societal cost perspectives. From both perspectives, the exercise therapy programme was the most cost-effective intervention, providing cost savings and significant health gains over usual care only.
Implications: Health systems would likely realize net cost savings, and society net monetary benefits, by providing funded access to a high-quality, individually-supervised course of individualized exercise therapy provided by a physiotherapist.
Keywords: Cost-effectiveness, economic evaluation, computer simulation modelling
Funding acknowledgements: Funded by the Health Research Council of New Zealand (07/200)
Purpose: To investigate the 5-year and lifetime net monetary benefit of providing exercise therapy and/or manual therapy, in addition to usual medical care, in people with hip and knee OA.
Methods: We conducted an economic evaluation alongside a RCT to 5-year follow-up. Adults meeting the American College of Rheumatology criteria for hip or knee OA were randomly allocated to receive either exercise therapy; manual therapy; combined exercise therapy and manual therapy; or no trial physioterapy intervention; in addition to usual medical care. Costs and health-related quality of life data were collected at baseline, 6 months, 1, 2 & 5 years. We developed and validated a state-transition simulation model of the clinical course of OA and, from 5-year observed data, simulated the lifetime costs, health-related quality of life (HRQoL), and incremental net monetary benefit (INMB), to the health system and society, in New Zealand (NZ) dollars.
Results: A total of 299 patients were assessed for eligibility, of which 206 met the study eligibility criteria and agreed to participate. Of the 195 participants surviving at 5-year follow-up, 129 (66%) returned completed survey questionnaires.Missing data were replaced using multiple imputation. Exercise therapy showed statistically significant HRQoL gains (0.12, 90% CI 0.07 to 0.15) at 5-year follow-up, relative to usual medical care only. Manual therapy (0.04, -0.00 to 0.09) and combined therapy (0.03, -0.03 to 0.07) did not. INMBs were positive, indicating cost-effectiveness of all three interventions, at all willingness-to-pay thresholds, from both the health system and societal perspectives. Exercise therapy was cost-effective in all analyses, with the largest (and only statistically significant) INMBs. Combined therapy was the least cost-effective treatment. At the NZ population aggregate level, INMB of exercise therapy, at the 1xGDP/capita willingness-to-pay level, of $19.8 billion (health system perspective) and $24.1 billion (societal perspective) over the lifetime of the current NZ adult population. All interventions remained cost-effective relative to usual care at all parameter levels considered in the one-way sensitivity analyses. Assumptions regarding eligibility criteria and intervention uptake rates had modest impacts on estimated INMBs, and differences in the cost of delivering physiotherapistservices had very little impact.
Conclusion(s): The three physiotherapist-delivered, individualised programmes of exercise or manual therapy provided were found to be cost-effective over a 5-year and a lifetime analysis horizon, relative to usual medical care only, for the treatment of hip and knee osteoarthritis, from both the health system and societal cost perspectives. From both perspectives, the exercise therapy programme was the most cost-effective intervention, providing cost savings and significant health gains over usual care only.
Implications: Health systems would likely realize net cost savings, and society net monetary benefits, by providing funded access to a high-quality, individually-supervised course of individualized exercise therapy provided by a physiotherapist.
Keywords: Cost-effectiveness, economic evaluation, computer simulation modelling
Funding acknowledgements: Funded by the Health Research Council of New Zealand (07/200)
Topic: Musculoskeletal; Orthopaedics; Information management, technology & big data
Ethics approval required: Yes
Institution: New Zealand Ministry of Health
Ethics committee: Lower South Regional Ethics Committee
Ethics number: LRS/07/11/044
All authors, affiliations and abstracts have been published as submitted.
