Expanding use of medications for opioid use disorder (MOUD) such as buprenorphine is a major component of the public health response to the opioid overdose epidemic, which caused nearly 50,000 drug overdose deaths in 2019 (Mattson et al., 2021). More recent provisional data suggest that annual deaths continue to rise in 2020 (Centers for Disease Control and Prevention, 2020). Escalating OUD-related mortality has been explicitly tied to an inability to fully deploy proven strategies including MOUD (National Academies of Sciences and Medicine, 2019). Buprenorphine has been available as an evidence-based treatment via prescription since 2002 and legislation in the past two decades has expanded access to buprenorphine (SAMHSA, 2020).

Longer buprenorphine retention (12–18 months with treatment) and better adherence (more days with buprenorphine supply) are associated with less opioid use, fewer costly opioid-related health care events such as inpatient admissions, and fewer overdoses (Ruetsch et al., 2017; Samples et al., 2020; Williams et al., 2020). Still, buprenorphine remains underutilized (Andrilla et al., 2017; Beetham et al., 2019). Less than one third of patients are retained at 180 days and over half of commercially-insured patients discontinue buprenorphine before 12 months (Agbese et al., 2020; American Society of Addiction Medicine, 2020; Olfson et al., 2020). Low utilization has been attributed to the limited number of clinicians authorized to prescribe buprenorphine, as well as a limit on the number of patients that can be treated per clinician (SAMHSA, 2020). In recent years in most US states the number of people with OUD exceeded buprenorphine prescribing capacity (Jones et al., 2015).

However, clinicians have also reported low patient demand for buprenorphine (Jones and McCance-Katz, 2019; Louie et al., 2019). Shorter retention in buprenorphine treatment has been linked to patients’ perceived inflexibility of treatment (Gryczynski et al., 2014), low daily dosage (Lo-Ciganic et al., 2019; Muruganandam et al., 2019), patient stress and emotions (Panlilio et al., 2019), other substance use disorders (Samples et al., 2018), exposure to actively using peers (Velasquez et al., 2019) and stigma surrounding MOUD treatment (Barry et al., 2014). Patients and clinicians also identify insurance factors and cost as inhibiting factors (Jones and McCance-Katz, 2019; Timko et al., 2016). There is some evidence that insurers impose more restrictions and require more cost-sharing at a higher rate for buprenorphine than medication treatment for alcohol use disorder and that some patients lower their buprenorphine fill rate in response to higher out-of-pocket costs (McClellan et al., 2019; Peters and Wengle, 2016).

The average out-of-pocket expenditure for a 30-day buprenorphine prescription among patients with commercial insurance was $32 in 2013–2015 compared to $67 in 2003 (Agbese et al., 2020; Roberts et al., 2018). It is important to contextualize this amount—already intuitively a non-trivial amount for patients to face every month—with evidence that the average total out-of-pocket health care cost among patients with commercial insurance still taking buprenorphine at 12 months was nearly $3,000 in 2013 (Agbese et al., 2020). Given recent changes to address supply-side issues that have hindered prescribing of buprenorphine (SAMHSA, 2020), it is critical to further analyze demand-side issues that may inhibit use of MOUD despite expanded access. This study aimed to measure the impact of out-of-pocket buprenorphine cost on treatment retention and adherence among US commercially insured patients.

This study examined health insurance payments (inpatient, outpatient, outpatient pharmacy dispensing) from IBM MarketScan Commercial Claims and Encounters Database. MarketScan includes commercial health plans and self-insured employers and is one of the largest data sources on the US commercially insured population. Patients were analyzed if they: 1) initiated buprenorphine-based MOUD (identified by National Drug Codes [NDC] excluding buprenorphine pain formulations, Belbuca and Butrans) with ≥7-day total supply (to potentially exclude treatment for short-term opioid withdrawal) during January 1, 2016 to June 30, 2017 (index prescription); 2) had continuous insurance enrollment including prescription drug coverage for 6 months before and 18 months after the index prescription; and 3) had no buprenorphine prescriptions ≤6 months before the index prescription.

The main outcome measures were buprenorphine retention and adherence. Retention was defined as ≥80 % days with buprenorphine supply (observed through patients’ prescription fill date and days’ supply) and ≤60 days’ gap (no buprenorphine possession) using three retention thresholds: 180, 360, and 540 days. Overlapping buprenorphine prescriptions were separately counted. Adherence was defined by the number of days with buprenorphine supply among retained patients during each analyzed retention threshold.

Patients’ out-of-pocket buprenorphine cost was the sum of the associated co-pay, co-insurance, and deductible. Patients’ average daily out-of-pocket buprenorphine cost was calculated as the total out-of-pocket buprenorphine cost divided by the total days supplied. Patient sex, age, and residential Census region were analyzed as reported in the MarketScan enrollment file. Patients’ other substance use disorder diagnoses, psychiatric diagnoses, and pain diagnoses (inpatient or outpatient) ≤3 months before the index prescription were identified by ICD-10-CM diagnosis codes (Supplemental Table 1). Patients’ buprenorphine dosage was converted to Suboxone dosage (Supplemental Table 2). Patients’ outpatient prescriptions for four other drug classes—antidepressants, opioid analgesics, benzodiazepines, and stimulants—≤3 months before index prescription were identified by NDC.

A multivariable logistic regression model estimated the association between patients’ average daily out-of-pocket buprenorphine cost and treatment retention for each analyzed retention threshold, controlling for patient demographic and clinical characteristics. A multivariable Poisson regression model estimated the association between patients’ average daily out-of-pocket buprenorphine cost and treatment gaps (or, days patients were without buprenorphine supply) for each analyzed retention threshold. Sensitivity analysis examined results using: 1) retention defined by ≥70 % and ≥90 % of days with buprenorphine supply, 2) restricting analysis to patients with continuous insurance enrollment in 2016, 2017, and 2018; and 3) restricting analysis to patients with ≥1 OUD clinical diagnosis ≤3 before the index prescription. This study used publicly available data and no human subjects. Analysis was conducted using SAS Software Version 9.4.

The total number of patients identified as initiating buprenorphine treatment was 6,439. The likelihood that patients were retained in MOUD treatment decreased as the retention threshold increased, with 49 % of analyzed patients retained in treatment at 180 days, 41 % at 360 days, and 37 % at 540 days (Table 1). Over 94 % of patients faced some out-of-pocket buprenorphine cost but patients retained in treatment had lower average daily out-of-pocket costs than those who discontinued treatment for all three analyzed retention thresholds. The average daily out-of-pocket buprenorphine cost was $1.91/day for patients retained in treatment at the 180-day threshold compared to $2.63/day for those who discontinued treatment within 180 days, $1.76/day compared to $2.38/day at 360 days, and $1.74/day compared to $2.27 at 540 days (p < 0.05 for all) (Table 1). Patients that discontinued treatment during all analyzed retention thresholds also had a lower average buprenorphine dose, were younger (aged 18–34 years), more often resided in the South or West region of the country, more often had other substance use disorder diagnoses (alcohol, cannabis, sedatives, cocaine, stimulants), psychiatric diagnoses (bipolar disorder, major depression, post-traumatic stress disorder), more often dispensed a prescription stimulant, and were less likely to have an osteoarthritis diagnosis or an opioid prescription prior to starting buprenorphine (p < 0.05 for all) (Table 1).

In multivariable analysis, a $1 increase in patients’ average daily out-of-pocket buprenorphine cost was associated with a 12 %, 14 %, and 13 % decrease (p < 0.05 for all) in the odds of treatment retention at 180, 360, and 540 days, respectively, controlling for patients’ demographic characteristics, other prescriptions, and SUD, psychiatric, and pain diagnoses (Table 2). For patients retained in treatment, a $1 increase in patients’ average daily out-of-pocket buprenorphine cost was associated with a 5%, 8%, and 8% increase (p < 0.05 for all) in the number of days without buprenorphine supply (roughly equivalent to missing 1.5 weeks, 1 month, and 1.5 months) during retention periods at the 180-, 360-, and 540-day thresholds, respectively, again controlling for patients’ demographic and clinical characteristics (Table 2 and Supplemental Table 3 [complete model estimates]). The direction and magnitude of base case results was not substantively changed in the sensitivity analyses (data available upon request).

This study’s primary contribution is directly quantifying the association between patients’ average daily out-of-pocket buprenorphine cost and retention and adherence. Controlling for observable factors previously identified to be associated with buprenorphine use, this study found that higher average out-of-pocket buprenorphine cost is associated with reduced buprenorphine retention and increased gaps in treatment coverage. For example, a $1 dollar increase in the daily out-of-pocket cost for buprenorphine decreased the likelihood of retention at 360 days by 14 % and increased the number of days without coverage by roughly one month for those still in treatment at 360 days.

These findings are important in the context of recent policymaking to expand access to office-based outpatient buprenorphine treatment as well as clinical recommendations on the minimum retention of 180 days for desired treatment outcomes (National Quality Forum, 2017). Longer buprenorphine retention and better adherence are linked to less opioid use, fewer costly opioid-related health care events, and fewer overdoses (Ruetsch et al., 2017; Samples et al., 2020; Williams et al., 2020). This study supports other recent research in concluding that demand-side factors inhibiting buprenorphine use, including patient out-of-pocket cost, must be addressed to fully realize the benefits of MOUD. The results of this study could be applied in cost-benefit analysis among commercial insurance providers—that is, weighing revenue from patient buprenorphine cost-sharing against the consequences of high-cost healthcare events that are more likely among patients with shorter retention in buprenorphine treatment (Samples et al., 2018.). The cost of a single ED visit or inpatient admission with principal diagnosis of opioid-related disorder is approximately $1,700 and $9,000 respectively (2017 USD) (Peterson et al., 2021), whereas this study’s estimated average out-of-pocket cost for buprenorphine treatment for 360 days is $765.

This study’s finding that out-of-pocket buprenorphine cost is negatively associated with buprenorphine use is supported by analytic methods presented here that addressed both the extensive (treatment retention) and intensive margins (treatment adherence during retention). Previous research on the price elasticity of demand for pharmaceuticals based on the intensive margin has demonstrated that patients may respond differently to price changes based on drug class (Gatwood et al., 2014). Previous research on buprenorphine cost-sharing demonstrated a statistically significant reduction in buprenorphine prescription fills during 2011–2015 associated with increased cost sharing only among some patient subgroups (patients age 45–64 years, those with rural or South region residence, and those enrolled in a health maintenance organization) (McClellan et al., 2019). The present study’s finding with more recent data that average out-of-pocket buprenorphine cost is associated with shorter retention in buprenorphine treatment and less adherence among patients overall suggests that even though the average out-of-pocket buprenorphine price is lower now than 20 years ago, more patients with OUD are less willing or able to sustain that cost and continue buprenorphine treatment.

This study is subject to several limitations. First, this study was not able to control for patients’ OUD duration and severity, which might affect MOUD initiation and continuation. Second, this study was not able to examine socioeconomic differences that might account for some patients’ greater sensitivity to out-of-pocket costs. Third, this study examined prescription fills but could not observe actual use. Fourth, this study did not examine concurrent services such as behavioral therapy that might be associated with patients’ MOUD retention and adherence. Finally, this study is limited to commercially insured patients and results may not be generalizable to different health insurance.

Recent policies have addressed supply-side inhibitors of MOUD treatment by making it easier for clinicians to prescribe buprenorphine to a greater number of patients. This study highlights cost as a demand-side barrier, suggesting that public health and insurer strategies aimed at reducing patient cost-sharing could improve buprenorphine treatment retention. Although the average out-of-pocket buprenorphine cost is lower than two decades ago, even at current levels such costs decrease retention and adherence among commercially insured patients. Efforts to address demand-side barriers, such as reducing patient cost-sharing, could help fully maximize the health and social benefits of buprenorphine-based MOUD.