Group-based trajectory modeling to assess adherence to chronic urate-lowering therapies among commercially insured US adults with gout.

Abstract

The benefits of urate-lowering therapies (ULTs) for the long-term management of gout are well established. However, suboptimal adherence remains a significant challenge, resulting in increased gout flares and higher health care utilization. The proportion of days covered (PDC) is commonly used to assess adherence but provides only a single value that fails to distinguish among individuals with differing and dynamic adherence patterns over time. Understanding fluctuations in adherence and their associated characteristics can inform interventions aimed at improving adherence. To identify distinct trajectories of ULT adherence in a commercially insured population and determine the sociodemographic and clinical factors associated with each trajectory. This retrospective cohort study used a 25% random sample from the IQVIA PharMetrics Plus database and included a commercially insured population who had a first index ULT prescription between 2017 and 2020, had at least 1 inpatient or 2 outpatient visits on different dates for gout in the year prior to the index ULT, and maintained continuous medical and prescription coverage for 1 year before and after the index ULT. A group-based trajectory model identified distinct adherence patterns and a multinomial logistic regression identified factors that were associated with adherence trajectory group membership. A total of 9,404 beneficiaries in the analytic sample were categorized into 4 ULT adherence trajectory groups: early decline (PDC = 0 by month 6, 14.97%), high-then-low (PDC = 0 by month 10, 7.95%), intermediate (PDC 0.4-0.6, 16.57%), and continuously high (PDC ≥ 0.8, 60.51%). In general, groups showing intermediate or declining adherence were more likely to be younger than 46 years, be female, reside outside the Eastern United States, have conditions such as peripheral vascular disease or dementia, and be prescribed medications for gout flares in the baseline period compared with the continuously high adherence group. These adherence groups were also less likely to have documented cardiometabolic comorbidities or other arthritic conditions relative to the continuously high adherence group. Nearly 40% of beneficiaries were nonadherent to ULTs during the 1-year follow-up period. Adherence trajectory groups have unique characteristics that could help to target interventions and improve patient care.