Patient and Caregiver Preferences for Hemophilia Prophylactic Treatments: A Discrete Choice Experiment

Background and Objectives The expansion of hemophilia treatment to include non-factor prophylaxis provides new treatment options for people living with hemophilia (PwH). The objective of this study was to measure PwH and caregiver (CG) preferences regarding hemophilia A (HA) and B (HB) prophylaxis, their willingness to trade-off benefits and risks, and their preferences for administration and device types. Methods A cross-sectional, web-based survey including a discrete choice experiment (DCE) was administered to adult PwH and CGs of children (aged 8 to 17 years) with hemophilia in the US and UK. The DCE design was informed by an evidence review and consultations with a steering committee of clinical and patient advisors. The survey was pre-tested in cognitive interviews with 10 PwH and 10 CGs prior to launch. In the DCE, each participant completed 10 choice tasks, each presenting two hypothetical prophylaxis profiles described by seven attributes. The attributes included two benefits (change in the number of annual bleeds, requirement for a second treatment for breakthrough bleeds), two risks during the next year of treatment (serious side effects, developing inhibitors), and three additional treatment characteristics (administration and device type, dosing frequency, refrigeration requirement). DCE responses were analyzed using a mixed logit model. Hemophilia types and country of residence were pooled for analysis for both PwH and CGs. Relative attribute importance (RAI) scores were calculated to assess the relative impact of improving each attribute from the worst to the best level on overall treatment preference. In addition, the trade-offs participants were willing to accept to change administration and device type were calculated. Results A total of 194 PwH and 169 CGs participated. The mean age of PwH was 38.5 years (range 18-80), the majority (95%) were male, White (78%), and had HA (85%). Over half (59%) were on factor replacement therapy, 43% were on non-factor prophylaxis. The mean age of CGs was 43.3 years (range 22-71), the majority were female (84%). Of the children with hemophilia they cared for, the mean age was 12.6 years (range 8 - 17), the majority were male (97%), White (69%) and had HA (80%). Among the children, 56% were on factor replacement, 47% were on non-factor prophylaxis. The attributes had similar impacts on treatment preferences of both PwH and CGs. Both considered treatment frequency (change from daily to every two to four weeks) to be the most important attribute (RAI: PwH 31.3%; CG 36.6%), followed by change in the number of annual bleeds (change from 2 more bleeds to 3 fewer bleeds; RAI: PwH 23.6%; CG 21.7%). Risk of serious side effects (change from 5% to 0%; RAI: PwH 13.3%; CG 12%), administration and device type (change from intravenous (IV) infusion to subcutaneous (SC) injection via pre-filled pen (PFP); RAI: PwH 12.1%; CG 10.4%), and risk of developing inhibitors (change from 5% to 0%; RAI: PwH 11.8%; CG 11.2%), had a similar impact on preferences, each being approximately one third as important as treatment frequency and half as important as change in annual bleeds. Refrigeration requirements (change from until use to up to 30 days; RAI: PwH 7.7%; CG 6.9%) were less important, although both PwH and CGs preferred to avoid treatments requiring refrigeration until use. The need for a second treatment for breakthrough bleeds (change from required to not required; RAI: PwH 0.3%; CG 1.1%) was unimportant to both PwH and CGs. With regard to administration and device type, both PwH and CGs preferred SC injections to IV infusions (P<0.001 for both groups). SC injection using a pre-filled pen (PFP) was numerically preferred to SC injection using a vial and syringe; however, this finding was not statistically significant. PwH were willing to accept a 2.9 % increase in risk of serious side effects in the next year, or a 3.1% increase in risk of developing inhibitors in the next year, to have SC injection via PFP rather than IV infusion. Conclusions Both PwH and CGs highly valued avoiding daily treatment administration and were willing to accept reduced benefits or increased risks in exchange for SC administration of prophylaxis via a PFP or vial and syringe device instead of IV infusion. Understanding preferences for hemophilia treatments and the trade-offs that PwH and CGs are willing to make between treatment attributes may facilitate shared decision-making when selecting prophylactic options.