The perspective of women with an increased risk of OHSS regarding the safety and burden of IVF: a discrete choice experiment.

study questionWhat are the preferences of women with an increased risk of ovarian hyperstimulation syndrome (OHSS) for characteristics of IVF treatments? SUMMARY ANSWER: In women with increased risk of OHSS, the chance of OHSS is a strong attribute in determining women's preference for IVF treatment and women are willing to trade off burden (side effects), costs and chance of pregnancy for lower risks of OHSS. WHAT IS KNOWN ALREADY: OHSS is the most serious iatrogenic complication of ovarian stimulation. Polycystic ovaries, high antral follicle count (AFC) and previous OHSS increase the risk of developing OHSS. IVM of oocytes offers great potential for patients with high AFC, since there is no risk of OHSS. With regard to patients' perspectives on fertility treatments, it has been shown that women undergoing IVF place different values on treatment characteristics, such as effectiveness (pregnancy rate), cancellation risk, safety (OHSS risk) and burden (side effects). To our knowledge, the preferences for different IVF treatments in women with increased risk of OHSS have not been studied yet. STUDY DESIGN SIZE DURATION: A multicentre discrete choice experiment (DCE) was performed between 2012 and 2016. The selected attributes offered were chance of OHSS, which represents safety; number of injections; chance of cycle cancellation (the latter two represent burden); chance of pregnancy; and out-of-pocket costs/willingness to pay. A target sample size was calculated by including 20 patients for five attributes resulting in the aim to include 100 women. PARTICIPANTS/MATERIALS SETTING

methodsWe invited subfertile women who were diagnosed with normogonadotrophic ovulation disorder and were undergoing treatment with gonadotrophins and/or had experienced (imminent) OHSS in a previous IVF treatment in the fertility clinic of four hospitals (three teaching and one academic). Women received a printed questionnaire with fictional scenarios and were asked, for each scenario, to choose their preferred treatment. We used a multinominal logit model to determine the preferences of women and investigated heterogeneity in preferences through latent class analysis. The decrease in OHSS risk required for women to accept an increased level of an undesirable attribute, i.e. their willingness to trade off, was calculated. MAIN RESULTS AND THE ROLE OF CHANCE: We distributed 120 questionnaires with a response rate of 79% (95/120). There were 91 questionnaires included in the analysis. All five attributes influenced women's treatment preference. About half of the women considered chance of pregnancy to be more important, while the other half considered prevention of OHSS and lower costs to be more important. Women were willing to trade off cancellation rate, number of injections, chance of pregnancy and costs for lower OHSS chances. We found that women were willing to accept 5% more chance on cycle cancellation if the OHSS rate dropped with 2%. Women were willing to accept one extra treatment for a reduction of 3.9% in OHSS risk. With respect to costs, women were willing to pay €1000 instead of no costs for a decrease in OHSS rate of 5.4% LIMITATIONS REASONS FOR CAUTION: The sample size of our study is relatively small which may limit the generalizability and sensitivity of the study. WIDER IMPLICATIONS OF THE

findingsThe results of this DCE help us to understand the trade-off that women at risk of OHSS make in their preference for characteristics on IVF treatments. This knowledge may be used during the counselling of couples about their treatment options. STUDY FUNDING/COMPETING INTERESTS: B.W.M. is supported by a NHMRC Practitioner Fellowship (GNT1082548). B.W.M. reports consultancy for Merck, ObsEva and Guerbet. J.P.d.B. reports personal fees from the Ferring Medical Advisory Board and grants from Ferring B. V and Merck Serono B. V outside the submitted work. There are no other conflicts of interest to declare. TRIAL REGISTRATION NUMBER: None.