Key points are not available for this paper at this time.
Background: Patients with severe hemophilia are at risk for musculoskeletal bleeding that can result in arthropathy and disability. Prophylactic treatment of hemophilia prevents bleeding and joint damage and is recommended as the standard of care for persons with severe hemophilia. Non-adherence to a prophylactic regimen can limit treatment effectiveness and result in adverse health outcomes. Emicizumab, a self-administered subcutaneous treatment, is available for prophylactic therapy of persons with hemophilia A with inhibitors and without inhibitors. This therapy is expected to enhance treatment satisfaction and adherence in hemophilia A. Aims: We completed a pilot study on the initial development and preliminary psychometrics of a new self-report adherence scale for emicizumab titled VERITAS-NEXGEN, modeled on the VERITAS-Pro hemophilia clotting factor adherence scale. In this current study, we investigated a broader set of psychometrics supporting this novel measure of emicizumab adherence in a larger sample. We also sought to describe self-reported adherence using this measure in the larger sample of patients with hemophilia. Methods: The VERITAS-NEXGEN scale consisted of five 5-item subscales (dosing, injection, timing, planning, and bleeds) and an aggregate total score. Items were rated on a 1-9 scale (higher values indicate greater adherence), and subscale and total scores were derived as average scores of constituent items. Participants (N=69 males) with factor VIII deficiency from two hemophilia treatment centers and an active prescription for emicizumab for at least four months were then recruited to complete the scale for psychometric analysis and investigation of self-reported adherence. A subset of the participants (N=20) completed a second adherence scale within one month to provide an estimate of test-retest reliability. The adherence scale was validated against two global patient-ratings of adherence: an established 1-10 rating scale of global adherence (global adherence rating; GAR) and a new 1-10 rating scale of familiarity with treatment. Results: Using the VERITAS-NEXGEN scale, total self-reported adherence was high (average total item score of 8.3 on a 1-9 scale; 9% of patients reporting perfect mean adherence scores of 9, and 87% of patients reporting strong mean adherence scores of 7-8). Based on subscale scores, dosing and injection domains (mean scores of 8.6 and 8.7, respectively; 97% of patients reporting scores of 7 or greater) had greater adherence than the timing and planning domains (mean scores of 7.8 and 8.4, respectively; 14% and 7% of patients reporting scores under 7, respectively). Adherence related to Bleeds also was lower than dosing and injections (mean score=8.1; 14% of patients reporting scores under 7). Internal consistency values for the dosing, injection, bleeds, and total scores were excellent (alpha 0.81-0.91), while alpha values for planning (alpha=0.73) and timing (alpha=0.57) fell in a fair to moderate range. Test-retest reliability for all subscales was good (Planning, r=0.61) to excellent (all other subscales and Total r=0.72-0.83). Dosing, injection, planning, and total scores correlated significantly with global ratings of adherence and treatment familiarity. Timing scores also correlated significantly with global ratings of adherence. Conclusions: Results from this study supported the reliability and validity of subscale and total scores from this new emicizumab adherence scale and provide new insights into adherence strengths and weaknesses in a sample of patients with hemophilia. Overall, self-reported emicizumab adherence was excellent in the sample, with 86-97% of the sample reporting strong to excellent adherence depending on adherence domain. In addition to providing a rapid, easy-to-obtain estimate of patient-rated adherence in the clinical setting, this new adherence scale will allow for investigation of specific adherence domains that are most challenging for patients taking emicizumab, allowing for targeted interventions.
Okolo et al. (Tue,) studied this question.