Medication Cost Concerns and Disparities in Patient-Reported Outcomes Among a Multiethnic Cohort of Patients With Systemic Lupus Erythematosus

DISCUSSION

In this study, more than one-quarter of participants in CLUES reported at least 1 medication cost concern, which encompassed financial concerns related to SLE drugs and cost-related medication nonadherence. We found marked differences in PROs among those with medication cost concerns in our cross-sectional analysis at baseline, with worse SLE disease activity, depressive symptoms, and HRQOL among those with drug-related affordability concerns. In this longitudinal cohort, the differences in outcomes we observed at baseline persisted over the course of follow-up. This highlights the urgent need to better understand and ameliorate these concerns among our patients, who often depend on long-term medication use for proper control of their disease.

Our research confirms the high burden of medication cost concerns among individuals with rheumatic diseases and SLE. Prior studies have investigated medication cost concerns from the perspective of cost-related nonadherence, which has been reported in approximately 20% of individuals with SLE⁴ and rheumatoid arthritis (RA).²³ Indeed, persons with SLE often identify medication affordability concerns as a major reason for missing medication doses or discontinuing altogether, along with other reasons such as side effects.^24,25 Qualitative research points to copays, high deductibles, and insurance lapses as prominent cost-related barriers.²⁵ In a study on individuals with lupus nephritis, medication costs also emerged as a leading barrier to medication-related decision making.²⁶ Medication costs can thus have a variety of effects on care trajectories; accordingly, in this study, we defined medication costs concerns broadly to encompass a range of affordability concerns, including cost-related nonadherence, as well as reported difficulties affording medications and behaviors such as requesting lower-cost alternatives from clinicians.

A striking finding in our study was that individuals with SLE and medication cost concerns had worse PRO scores across a variety of domains, such as disease activity, but also including depression, and diverse aspects of HRQOL, even after adjustment for demographic factors, income, insurance, SLE organ damage, and number of immunomodulatory drugs. The disparity in PRO scores was particularly notable for the SLAQ outcome, with the mean patient-reported disease activity score nearly twice as high in those with medication cost concerns as compared to those without those concerns. Participants with medication cost concerns also had worse depression scores and HRQOL, as indicated by the PHQ-8 and PROMIS measures. For the PROMIS measures, the differences were especially pronounced for the fatigue, pain interference, physical function, and social roles/activities domains, exceeding estimates of minimally important differences.²⁷ Interestingly, there were no significant differences for the BILD outcome at baseline; the effects of affordability concerns and other socioeconomic factors, such as poverty, on damage accrual may take longer to manifest.²⁸ These data suggest that clinicians should be alerted to the possibility of affordability concerns when treating patients with either high disease activity or poor HRQOL.

A compelling mechanistic link between the unaffordability of drugs used in SLE and worse health outcomes is nonadherence, which is associated with increased risk of flare and acute care utilization^29,30 and forms the basis of our definition of medication cost concerns. Another potential mediating factor is psychosocial stress. In our longitudinal analysis, we attempted to further examine the association between medication cost concerns and PROs, although we did not observe worsening in PROs in those with medication cost concerns as compared to those without these concerns over follow-up, as hypothesized. This may be because of a variety of reasons. For instance, the observation time of 2 years may have been too limited; prior studies have shown that other sociodemographic variables have an effect on outcomes such as organ damage in SLE, but when observed over longer time intervals.²⁸ Medication cost concerns are likely dynamic over time, although in our study were only measured at baseline. We did observe, though, that the disparities in PROs between those with and without medication cost concerns at baseline persisted over the course of follow-up (Figure 2), suggesting that this subgroup of individuals with affordability concerns continues to have unmet needs over time and may benefit from interventions, either directed at reducing out-of-pocket costs, medication nonadherence, or suboptimal disease outcomes.

A surprising finding was that many sociodemographic factors, such as age, sex, race and ethnicity, and income were not significantly associated with medication cost concerns (Table 1), as has been shown in other studies of chronically ill or older adults.^9,21,22 Self-reported nonadherence did not differ between groups either, although our study was limited in asking participants for the number of missed doses in the last 7 days. This suggests that, in the absence of clear risk factors, potentially any patient with SLE could be susceptible to drug affordability concerns, regardless of their sociodemographic background or recent adherence level. Our results did show a higher proportion of Medicare beneficiaries and current biologic users among those with medication cost concerns. These findings merit further investigation, especially from the context of cost-sharing and drug-related out-of-pockets costs, which have been shown to be important determinants of affordability concerns and subsequent nonadherence in the general population^21,31 and in patients with RA receiving biologic therapy.³²

Although we are still learning about the scope of medication cost concerns in SLE, there are several potential tools that the rheumatology community can use to address this problem. Clinicians can engage patients in discussing their medication costs on a more routine basis, since affordability concerns often affects medication decision making and adherence. Research has documented a willingness on the part of both patients and clinicians to discuss medication costs, although only a fraction ever do in clinical encounters.³³ In a study of patients with RA, clinicians discussed drug costs in only one-third of encounters, despite medication changes being made in more than half of visits.³⁴ Uncovering medication cost concerns may point to solutions such as finding lower-cost alternatives or opportunities for changes in policy that enable dependable healthcare coverage and reduce out-of-pocket drug costs. For instance, implementation of the Medicare Part D prescription drug benefit in 2006 was associated with a decrease in cost-related nonadherence among beneficiaries.³⁵ In parts of the world with universal prescription drug coverage, medication costs are not a major cause of medication nonadherence in patients with SLE.³⁶ Medication coverage and out-of-pocket costs are particularly relevant in the context of biologics and newly approved agents entering the SLE treatment armamentarium, although even relatively low drug cost burdens of common immunosuppressants can exacerbate nonadherence in individuals with SLE.³⁷ Addressing medication cost concerns effectively will thus entail interventions in the clinic as well as in the domain of policy to improve both access and outcomes among patients with SLE.

There are several important strengths to highlight in this study. It is one of the largest investigations of medication cost concerns among individuals with SLE and benefits from the CLUES cohort’s population-based study design and diverse representation of Asian and Hispanic participants. We purposefully employed a broad definition of medication cost concerns to identify a wide variety of financial concerns; we also studied a diverse array of PROs and performed longitudinal analyses to understand the relationship between cost concerns and outcomes.

This study also has several limitations. The questions about medication cost concerns may have lacked precision because no recall time was defined; medication cost concerns were not necessarily current or even recent. In addition, the questions referred to SLE medications in general and not to specific medications or classes of medications. Disease activity and organ damage in this study were assessed by means of patient report and were not verified by physician-assessed measures, which may be less susceptible to noninflammatory symptoms or bias.³⁸ Other factors that likely affect medication affordability concerns, such as out-of-pocket expenses and medical debt, were not measured in CLUES but warrant further investigation. We considered the possibility of reverse causality in our cross-sectional model and for this reason undertook a longitudinal model; additional studies investigating medication cost concerns from a longitudinal perspective with longer follow-up time should be conducted. Future study should focus on the long-term effect of advances in policy, such as the upcoming capping of out-of-pocket expenses for Medicare Part D beneficiaries.³⁹

In summary, medication cost concerns were associated with worse PROs across diverse domains in this multiethnic cohort of people with SLE, even after adjustment for important sociodemographic and clinical variables. These differential outcomes persisted over time in this longitudinal cohort, highlighting the unmet need to address these concerns among people with SLE in general.