Cost-utility analysis of total shoulder arthroplasty: a prospective health economic study using real-world data.

BACKGROUND: With increasing health care expenditures, knowledge about the benefit and costs of surgical interventions such as total shoulder arthroplasty (TSA) becomes important for orthopedic surgeons, social insurance programs, and health policy decision makers. We examined the impact of TSA on quality of life (QOL), direct medical costs, and productivity losses and evaluated the cost-utility ratio of TSA compared with ongoing nonoperative management using real-world data. METHODS: Patients with shoulder osteoarthritis and/or rotator cuff tear arthropathy indicated for anatomic or reverse TSA were included in this prospective study. QOL (European Quality of Life 5 Dimensions 5-Level questionnaire) and shoulder function (Constant score; Shoulder Pain and Disability Index; short version of the Disabilities of the Arm, Shoulder and Hand questionnaire; and Subjective Shoulder Value) were assessed preoperatively and up to 2 years postoperatively. Health insurance companies provided all-diagnosis direct medical costs for 2018 in Swiss francs (CHF), where 1 CHF was equivalent to US $1.02. Indirect costs were assessed using the Work Productivity and Activity Impairment Questionnaire. Baseline data at recruitment and the total costs of the preoperative year served as a proxy for nonoperative management. The incremental cost-effectiveness ratio (ICER) was calculated as the total costs to gain 1 extra quality-adjusted life-year (QALY) based on both the health care system perspective and societal perspective. The relationship between QOL and shoulder function was assessed by regression analysis. RESULTS: The mean preoperative utility index for the European Quality of Life 5 Dimensions 5-Level questionnaire of 0.68 for a total of 150 patients (mean age, 71 years; 21% working; 58% women) increased to 0.89 and 0.87 at 1 and 2 years postoperatively, respectively. Mean direct medical costs were 11,771 CHF (preoperatively), 34,176 CHF (1 year postoperatively), and 11,763 CHF (2 years postoperatively). The ICER was 63,299 CHF/QALY (95% confidence interval, 44,391-82,206 CHF/QALY). The mean productivity losses for 29 working patients decreased from 40,574 CHF per patient (preoperatively) to 26,114 CHF at 1 year postoperatively and 10,310 CHF at 2 years postoperatively. When considering these productivity losses, the ICER was 35,549 CHF/QALY (95% confidence interval, 12,076-59,016 CHF/QALY). QOL was significantly associated with shoulder function (P < .001). CONCLUSION: Using real-world direct medical cost data, we calculated a cost-utility ratio of 63,299 CHF/QALY for TSA in Switzerland, which clearly falls below the often-suggested 100,000-CHF/QALY threshold for acceptable cost-effectiveness. In view of productivity losses, TSA becomes highly cost-effective with an ICER of 35,546 CHF/QALY.

Effects of gait variations on grip force coordination during object transport.

In object transport during unimpeded locomotion, grip force is precisely timed and scaled to the regularly paced sinusoidal inertial force fluctuations. However, it is unknown whether this coupling is due to moment-to-moment predictions of upcoming inertial forces or a longer, generalized time estimate of regularly paced inertial forces generated during the normal gait cycle. Eight subjects transported a grip instrument during five walking conditions, four of which altered the gait cycle. The variations included changes in step length (taking a longer or shorter step) or stepping on and over a stable (predictable) or unstable (unpredictable support surface) obstacle within a series of baseline steps, which resulted in altered frequencies and magnitudes of the inertial forces exerted on the transported object. Except when stepping on the unstable obstacle, a tight temporal coupling between the grip and inertial forces was maintained across gait variations. Precision of this timing varied slightly within the time window for anticipatory grip force control possibly due to increased attention demands related to some of the step alterations. Furthermore, subjects anticipated variations in inertial force when the gait cycle was altered with increases or decreases in grip force, relative to the level of the inertial force peaks. Overall the maintenance of force coupling and scaling across predictable walking conditions suggests that the CNS is able to anticipate changes in inertial forces generated by gait variations and to efficiently predict the grip force needed to maintain object stability on a moment-to-moment basis.