A cost-effectiveness analysis for total knee arthroplasty telerehabilitation: proof of concept of a decision model

OBJECTIVES: The study aims to assess the cost-effectiveness of telerehabilitation for Total Knee Arthroplasty (TKA) patients in Italy. TKA was performed 64,936 times in Italy in 2012, reasonably leading to the same number of rehabilitation processes. The most recent cost analysis showed rehabilitation to account for Euro 158 millions per year. Therefore, new strategies aiming at optimizing resources and preserve patients’ wellbeing are claimed. METHODS: A four-state Markov model (successful TKA; revision; successful revision; death) forecasted costs and clinical outcome over 10 years (cycle length: 1 year) for 1,000 individuals undergoing usual care rehabilitation (UC) or a mixed UC-telerehabilitation (UC-T) program. Published literature provided transition probabilities and clinical outcome (active knee flexion Range Of Motion-ROM); while UC and telerehabilitation costs were estimated through Italian tariffs and panel of experts. Each surgery was assumed to lead to rehabilitation or telerehabilitation, resulting in direct medical and indirect costs (human capital approach). Results were adjusted applying half-cycle correction method and discount rate of 3%. A Probabilistic Sensitivity Analysis (PSA) described parameters uncertainty and results were reported using Incremental Cost-Effectiveness Ratio (ICER) from society and Italian-NHS perspective. RESULTS: Expected mean health care costs for UC were 1,253.2€/patient over ten years, and UC-T costs were on average 33.7€/patient higher (95% CI €10.8). ROM-degrees for UC and UC-T were respectively 24.5 and 26.8 (mean difference=2.3, 95% CI 0.002). The resulting ICER was 14.5€/ROM-degree (Italian and NHS perspective).Adopting a societal perspective, UC-T was more effective yet appeared cheaper than UC (respectively 1,429 and 1,457€/patient, mean difference -28€/patient (95% CI €10.8)). CONCLUSIONS: Although the preliminary results have shown that UC-T could be a cost-saving procedure if societal perspective is adopted, these findings are uncertain due to the model assumptions. Therefore, further investigations with patients’ level data and generic outcome measures (e.g. QALY) are required to draw definitive conclusions about cost-effectiveness in telerehabilitation.