Cost-effectiveness of the recommended medical intervention for the treatment of dysmenorrhea and endometriosis in Japan

Abstract

Background and objective: This study aims to assess the cost-effectiveness of early physician consultation and guideline-based intervention to prevent endometriosis and/or disease progression using oral contraceptive (OC) and progestin compared to follow-up of self-care for dysmenorrhea in Japan.

Methods: A yearly-transmitted Markov model of five major health states with four sub-medical states was constructed. Transition probabilities among health and medical states were derived from Japanese epidemiological patient surveys and converted to appropriate parameters for inputting into the model. The dysmenorrhea and endometriosis-associated direct costs included inpatient, outpatient visit, surgery, and medication (OC agents, over-the-counter drugs), etc. The utility measure for patients with phase I-IV endometriosis comprised a visual analogue scale. We estimated the cost per quality-adjusted life year (QALY) at a time horizon of 23 years. An annual discount rate at 3% for both cost and outcome was considered.

Results: The base case outcomes indicated that the intervention would be more cost-effective than self-care, as the incremental cost-effectiveness ratio (ICER) yielded 115,000 JPY per QALY gained from the healthcare payers' perspective and the societal monetary value (SMV) was approximately positive 3,130,000 JPY, favoring the intervention in the cost-benefit estimate. A tornado diagram depicting the stochastic sensitivity analysis of the ICER and SMV from both the healthcare payers' and societal perspectives confirmed the robustness of the base case. A probabilistic analysis resulting from 10,000-time Monte Carlo simulations demonstrated efficiency at willingness-to-pay thresholds in more than 90% of the iterations.

Conclusions: The present analysis demonstrated that early physician consultation and guideline-based intervention would be more cost-effective than self-care in preventing endometriosis and/or disease progression for patients with dysmenorrhea in Japan.

Keywords: Cost-effectiveness; Dysmenorrhea; Endometriosis; Guideline-based intervention; Self-care.

Fig. 1

The Markov model used to assess the cost-effectiveness of early medical intervention for dysmenorrhea and endometriosis. A simple Markov model with yearly transmission of five health states (dysmenorrhea, phase I/II endometriosis, phase III/IV endometriosis, cured, and other-cause death) with four sub-medical states (consultation, surgery, recurrence, and stay condition) was constructed based on standard therapeutic and empirical pathways with consensus from gynecologists

Fig. 2

Model calibration and validation through comparison with realistic statistics. The simulation demonstrated the validity of the model in computing prevalence of approximately 210,000 cases with endometriosis, similar to that reported in a national survey (approx. 247,000^a; Terakawa et al. [15]); however, a discrepancy in prevalence for age-groups of 35 years or more existed. ^anumber of patients with endometriosis plus adenomyosis

Fig. 3

Tornado diagram a The incremental cost-effectiveness ratio (ICER) Tornado diagram used to assess the robustness of the base case analysis from the perspective of healthcare payers. A Tornado diagram depicting the results of the stochastic sensitivity analysis for ICER revealed that the cure rate for dysmenorrhea resulting from the guideline-based intervention influenced the base case; however, the robustness of the base case was confirmed. Indices of parameters (a) A: Cure rate for dysmenorrhea (0.643 to 0.957). B: Odds ratio for the development of dysmenorrhea (0.2 to 0.7). C: Utility for endometriosis III/IV (0.15* to 0.557). D: Proportion of visits in patients with endometriosis (0.0 to 0.07). E: Progression to endometriosis I/II (0.179 to 0.189). F: Discount rate (0.01 to 0.05). G: Utility for dysmenorrhea (0.63 to 0.644). H: Recurrence of dysmenorrhea (0.206 to 0.239). J: Cure rate of endometriosis I/II (0.322 to 0.478). K: Utility for endometriosis I/II (0.63 to 0.644). *: To take into account worst case scenario, lower value of the utility measuring for endometriosis III/IV was derived from the external criteria [20]. b The incremental cost Tornado diagram used to assess the robustness of the base case analysis from the societal perspective. A Tornado diagram depicting the results of the stochastic sensitivity analysis for IC revealed that the discount rate resulting from the guideline-based intervention influenced the base case; however, the robustness of the base case was confirmed. Indices of parameters (b) A: Discount rate (0.01 to 0.05). B: Recurrence of dysmenorrhea (0.206 to 0.239). C: Cure rate of endometriosis I/II (0.322 to 0.478). D: Odds ratio for the development of dysmenorrhea (0.2 to 0.7). E: Progression to endometriosis I/II (0.179 to 0.189). F: Cure rate for dysmenorrhea (0.643 to 0.957). G: Proportion of visits in patients with endometriosis (0.0 to 0.07)

Fig. 4

Cost-effectiveness acceptability curve for early medical intervention in dysmenorrhea and endometriosis from the perspective of healthcare payers. The results of a probabilistic analysis with 10,000-time Monte Carlo simulations are illustrated by cost-effectiveness acceptability curves and demonstrated the efficacy of medical intervention at a willingness to pay threshold of less than a million JPY per quality-adjusted life-year gained in more than 90% of the population