Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2025 Apr 1.
Published in final edited form as: J Subst Use Addict Treat. 2023 Dec 18;159:209281. doi: 10.1016/j.josat.2023.209281

Out-of-pocket spending and health care utilization associated with initiation of different medications for opioid use disorder: Findings from a national commercially insured cohort

Nicole C McCann 1, Marc R LaRochelle 2,3, Jake R Morgan 1
PMCID: PMC10947919  NIHMSID: NIHMS1954143  PMID: 38122988

Abstract

Introduction

Buprenorphine and naltrexone are effective medications for opioid use disorder (MOUD). Naltrexone requires complete detoxification from opioids before initiation while buprenorphine does not, which leads to a differential clinical induction challenge. Few studies have evaluated economic costs associated with MOUD initiation.

Methods

We conducted a retrospective cohort analysis using the 2014–2019 Merative MarketScan database. We included individuals diagnosed with opioid use, abuse, or dependence from 2014 to 2019 who initiated one of three MOUD types: 1) buprenorphine, 2) extended-release naltrexone, or 3) oral naltrexone. We calculated total and monthly out-of-pocket spending, for overall and MOUD-specific claims, for the three months prior through three months after MOUD initiation. We also calculated utilization of detoxification, inpatient, and outpatient services monthly over this period.

Results

Our cohort included 27,133 individuals; 19,536, 1,886, and 5,711 initiated buprenorphine, extended-release naltrexone, and oral naltrexone, respectively. Individuals who initiated naltrexone had the highest out-of-pocket spending over the study period. MOUD-specific spending did not contribute substantially to total out-of-pocket spending. Difference in overall spending by MOUD type was driven by a subset of individuals who initiated naltrexone and had very high out-of-pocket spending in the month prior to MOUD initiation. In this month, mean monthly out-of-pocket spending for high-spenders (above 90th percentile within MOUD type category) was $5,734 (95% confidence interval [CI]: $5,181–$6,286) and $4,622 (95% CI: $4,161–$5,082) for those who initiated oral and extended-release naltrexone, respectively, compared with $1,852 (95% CI: $1,754–$1,950) for those who initiated buprenorphine. In the month prior to MOUD initiation, those who initiated naltrexone also had higher detoxification, inpatient, and outpatient episode/visit frequency. In the month prior to initiation, 28.8% (95% CI: 27.7%–30.0%) and 25.5% (95% CI: 23.6%–27.5%) of individuals who initiated oral and extended-release naltrexone had detoxification episodes, compared with 9.7% (95% CI: 9.3%–10.1%) of those who initiated buprenorphine.

Conclusion

Findings suggest that individuals who initiated naltrexone utilized more intensive health services, including detoxification, in the period prior to MOUD initiation, resulting in significantly higher out-of-pocket spending. Out-of-pocket spending is a patient-centered outcome reflecting potential patient burden. Our results should be considered as part of the shared decision-making process between patients and providers when choosing treatment for OUD.

Keywords: opioid-related disorders [MeSH], expenditures, out-of-pocket [MeSH] patient centered care [MeSH], medications for opioid use disorder

1. Introduction

In 2021, overdose deaths in the United States (US) exceeded 100,000, with 75% of these deaths involving opioids (National Institute on Drug Abuse 2023). About 2.7 million people in the US have opioid use disorder (OUD), putting them at risk for overdose and death (National Institute on Drug Abuse 2021). OUD is a chronic and treatable disease, and medications for opioid use disorder (MOUD) are evidence-based treatments that can reduce overdose and mortality risk (National Institute on Drug Abuse 2021). Increasing uptake of medications for opioid use disorder is a key strategy to address the United States’ opioid overdose crisis (U.S. Department of Health & Human Services 2021).

Buprenorphine and naltrexone are MOUD that can be prescribed in non-specialty medical settings. For opioid-dependent individuals, induction onto the partial opioid agonist buprenorphine is typically attempted once individuals experience mild to moderate withdrawal symptoms after opioid cessation. Conversely, induction onto naltrexone requires opioid cessation for at least one week, often achieved through medically-managed withdrawal, which can be a barrier to its uptake (Oesterle et al. 2019; Blanco and Volkow 2019; Bisaga et al. 2018; Lee et al. 2018). A randomized controlled trial of individuals entering medically-managed opioid withdrawal provided evidence that naltrexone may be harder to initiate due to the requirement for a longer opioid cessation period, posing an induction challenge. In this trial, 72% of individuals randomized to extended-release naltrexone were successfully inducted compared with 94% of those randomized to buprenorphine (Lee et al. 2018).

While prior research has examined outcomes following MOUD initiation, few studies have evaluated out-of-pocket spending and healthcare utilization leading up to initiation. Because some MOUD require more intensive induction procedures (including full detoxification for naltrexone), it is important to account for this utilization and cost as part of the overall treatment. This could have implications for patients, health systems and insurers. In particular, out-of-pocket costs are a meaningful outcome to consider from a patient-centered perspective, especially if costs pose a burden to patients and differ across treatment options (Patient-Centered Outcomes Research Institute 2022). Thus, we assessed the period immediately before and after MOUD initiation in a cohort of individuals with OUD in a large, commercially insured population and detailed out-of-pocket spending and healthcare utilization during this period.

2. Materials and methods

2.1. Design, population, and data collection

We conducted a retrospective cohort analysis using data from the 2014–2019 Merative MarketScan commercial claims and encounters database. MarketScan is an insurance claims-based dataset that includes outpatient and inpatient visits, laboratory and diagnostic testing, and outpatient pharmacy claims. Unique identifiers link individual across file types (i.e., outpatient, inpatient, and pharmacy) and across years. The database contains a national sample of the US commercially insured population (Merative 2022). MarketScan data contain de-identified information and conform to Health Insurance Portability and Accountability (HIPAA) requirements. Therefore, this study required no institutional review board approval.

Our cohort included individuals diagnosed with OUD from 2014–2019 who initiated MOUD. We identified OUD as having at least one inpatient or two outpatient claims with an International Classification of Diseases ninth or tenth edition (ICD-9 or ICD-10) code for opioid use, abuse, or dependence in medical claims. We considered individuals to initiate MOUD if they had a national drug code (NDC) for a MOUD prescription (oral or extended-release naltrexone or buprenorphine). We based these definitions, including frequency of codes and which codes to include, on expert review and prior literature (eTable 1) (Morgan et al. 2022; Cochran et al. 2014). We did not include methadone because it is not widely captured in commercial claims. We defined individuals’ index date as their first medical claim for opioid use, abuse, or dependence, and included individuals who had an MOUD claim after the index date and who had at least three months of continuous enrollment and a three-month washout period before the index date (Morgan et al. 2019). The study defined the washout period as a period without any MOUD claims, implemented to ensure that an MOUD claim after the index date was true initiation. We followed individuals for three months after MOUD initiation and censored them from analysis if they were disenrolled from their insurance plans during the period analyzed.

2.2. Reported outcomes

Cohort demographic characteristics included age, sex, and region as reported in individuals’ earliest medical claim within the study period. To summarize clinical comorbidities, we calculated Elixhauser scores (eTable 2) (Morgan et al. 2019).

Reported outcomes included total and monthly individual out-of-pocket spending by MOUD type in the three months before and three months after MOUD initiation. The study defined out-of-pocket spending as the sum of inpatient, outpatient, and pharmacy coinsurance, copays, and deductibles. We calculated both total and monthly spending for: 1) overall out-of-pocket spending (all inpatient, outpatient, and pharmacy medical claims) and 2) MOUD-specific out-of-pocket spending (all inpatient, outpatient, and pharmacy medical claims with an NDC code for MOUD [eTable 1]).

To evaluate healthcare utilization in the time period around MOUD initiation, we reported the proportion of individuals with inpatient detoxification episodes and inpatient stays (any type) each month over the study period. The study identified detoxification episodes and inpatient stays from inpatient claim files using unique case identification numbers. Detoxification episodes were further identified using revenue codes for detoxification (eTable 1). We also reported the mean number of outpatient visits, which include doctor’s office, hospital outpatient facility, emergency department, or other outpatient facility visits, monthly over the study period. We identified outpatient visits from outpatient claims files using unique service dates. Additionally, we reported per-episode, per-stay, and per-visit out-of-pocket spending monthly over the study period, defined as the sum of coinsurance, copay, and deductibles associated with each detoxification, inpatient, or outpatient visit.

2.3. Descriptive analyses

We calculated total and monthly out-of-pocket spending for three months prior to three months after MOUD initiation for all claims and for MOUD-specific claims by MOUD type (all MOUD, oral naltrexone, extended-release naltrexone, and buprenorphine). Similar to other cost data, out-of-pocket spending was right-skewed (few people having extraordinarily high costs), so we reported mean and median total and monthly out-of-pocket spending values.

To capture differences in the proportion of individuals with high spending across different MOUD-types, we first calculated the 90th percentile overall out-of-pocket spending values for all individuals each month. We then reported the proportion of individuals above these 90th percentile values across each MOUD type. Additionally, to understand differences in spending across MOUD types among people with high overall out-of-pocket spending, we reported mean monthly overall out-of-pocket spending values among individuals above the 90th percentile within their MOUD type category (i.e., “high spenders”). For this analysis, the 90th percentile overall monthly out-of-pocket spending value was calculated separately for each month and within each MOUD type (any MOUD, oral naltrexone, extended-release naltrexone, and buprenorphine) and the out-of-pocket spending was reported within each respective category.

For detoxification and all inpatient utilization, we calculated the proportion of people with one or more detoxification or inpatient episode/stay monthly by MOUD type. We also calculated the mean number of outpatient visits monthly by MOUD type. We reported the mean out-of-pocket spending per-episode/stay for detoxification, inpatient, and outpatient episodes/stays.

For all analyses, we reported 95% confidence intervals (CI) for means (Phillips et al. 2022; du Prel et al. 2009), and interquartile ranges (IQR) for medians. Given skewed out-of-pocket spending data, the Supplemental Appendix reports both mean and median values for all out-of-pocket spending variables. The study team used SAS version 9.4 for data cleaning and STATA version 17.0 for analyses. We conducted the analysis from September 2022 to March 2023.

3. Results

3.1. Cohort characteristics

Our cohort included 27,133 individuals with OUD with new episodes of MOUD from 2014–2019. Of these, 19,536 initiated buprenorphine, 1,886 initiated extended-release naltrexone, and 5,711 initiated oral naltrexone. Of those who initiated oral naltrexone, 591 (10%) switched to extended-release naltrexone within one month. Overall, the cohort was majority male (62.4%) with a median age of 32 years. A plurality of the cohort was from the Southern region (42.2%). Cohort and clinical characteristics overall and by MOUD type are shown in Table 1.

Table 1.

Cohort characteristics overall and by MOUD type

Characteristic1 Any MOUD N=27,133 Oral naltrexone n=5,711 Extended-release naltrexone n=1,886 Buprenorphine n=19,536
Age, years, median (IQR) 32.0 (23.0–44.0) 33.0 (24.0–44.0) 25.0 (22.0–38.0) 20.0 (22.0–44.0)
Sex2, n (%)
 Male 16,925 (62.4) 3,332 (58.3) 1,215 (64.4) 12,378 (63.4)
 Female 10,208 (37.6) 2,379 (41.7) 671 (35.5) 7,158 (36.6)
Region, n (%)
 Northeast 6,383 (23.5) 1,474 (25.8) 540 (28.6) 4,369 (22.4)
 North Central 5,533 (20.4) 1,192 (20.9) 348 (18.5) 3,993 (20.4)
 South 11,457 (42.2) 2,204 (38.6) 731 (38.8) 8,522 (43.6)
 West 3,656 (13.5) 820 (14.4) 260 (13.8) 2,576 (13.2)
 Unknown 104 (0.4) 21 (0.4) 7 (0.4) 76 (0.4)
Elixhauser score 1.39 1.67 1.27 1.31
Open in a new tab

Abbreviations: MOUD: medication for opioid use disorder; n: number; IQR: interquartile range

1

All characteristics reported as recorded at earliest encounter in the study period

2

The MarketScan “sex” variable is defined as the reported gender of the patient; however, since the variable is collected as a binary, we use the variable name “sex”

Values with 95% confidence intervals shown in Supplement, eTable 11

Values may not add up to 100% due to rounding

3.2. Total overall and MOUD-specific out-of-pocket spending by MOUD type

From three months prior to three months after MOUD initiation, mean total overall out-of-pocket spending in the cohort was $1,622 (95% CI: $1,614–$1,711). For those who initiated oral naltrexone, extended-release naltrexone, and buprenorphine, mean total overall out-of-pocket spending over this period was $2,513 (95% CI: $2,349–$2,677), $2,300 (95% CI: $2,066–$2,534) and $1,362 (95% CI: $1,320–$1,403), respectively (mean and median values shown in eTable 3).

Mean total MOUD-specific out-of-pocket spending over the study period was $149 (95% CI: $146–$152) across all MOUD types, and $44 (95% CI: $40–$49), $256 (95% CI: $233–$279), and $169 (95% CI: $166–$172) for oral naltrexone, extended-release naltrexone, and buprenorphine, respectively (mean and median values shown in eTable 4).

3.3. Monthly overall and MOUD-specific out-of-pocket spending by MOUD type

Median monthly overall out-of-pocket spending was low for individuals who initiated MOUD, ranging from $0-$49 across all MOUD types (mean and median values shown in eTable 5). However, the proportion of individuals who had monthly out-of-pocket spending above the 90th percentile varied by MOUD type, and differences in total spending over the study period were driven by high spending among these individuals in the time leading up to MOUD initiation.

As shown in Figure 1, in the months leading up to MOUD initiation, individuals who initiated oral and extended-release naltrexone were more likely to be above the 90th percentile for overall monthly out-of-pocket spending compared with individuals who initiated buprenorphine. This difference was greatest in the month immediately prior to MOUD initiation. In the month prior to MOUD initiation, the 90th percentile overall out-of-pocket spending value was $968 (95% CI: $917–$1,015) across all MOUD types; 19.9% (95% CI: 18.9%–20.9%) and 16.4% (95% CI: 14.8%–18.1%) of those who initiated oral and extended-release naltrexone, respectively, had out-of-pocket spending above this value, compared with 6.5% (6.2%–6.8%) of those who initiated buprenorphine (Figure 1, eTable 6). After MOUD initiation, the proportion of individuals in each MOUD category who were above the 90th percentile for monthly overall out-of-pocket spending was similar (Figure 1, eTable 6).

Figure 1. Proportion of individuals above the 90th percentile for overall monthly out-of-pocket spending, by MOUD type.

Figure 1.

Open in a new tab

Figure 1 shows the proportion of individuals above the 90th percentile for overall monthly out-of-pocket spending in each month from three months prior to three months after MOUD initiation. The y-axis shows the proportion of individuals, and the x-axis shows time relative to MOUD initiation. For each month, the 90th percentile out-of-pocket spending value was calculated among all individuals who initiated any MOUD type. The blue line shows the proportion of individuals who initiated oral naltrexone, the green line shows the proportion of individuals who initiated extended-release naltrexone, and the grey line shows the proportion of individuals on buprenorphine. Error bars represent 95% confidence intervals.

Abbreviations: MOUD: medication for opioid use disorder

Mean overall monthly out-of-pocket spending among “high-spender” individuals with out-of-pocket spending above the 90th percentile within their MOUD type category followed a similar pattern. Among these individuals, mean out-of-pocket spending was higher in months leading up to MOUD initiation for those who initiated oral and extended-release naltrexone compared with buprenorphine, with the largest difference in the month immediately prior. In the month prior to MOUD initiation, mean overall out-of-pocket spending for individuals above the 90th percentile out-of-pocket spending values within their MOUD type category was $5,734 (95% CI: $5,181–$6,286 and $4,622 (95% CI: $4,161–$5,082) for those who initiated oral naltrexone and extended-release naltrexone, respectively, compared with $1,852 (95% CI: $1,754–$1,950) for those who initiated buprenorphine (Figure 2, eTable 6). After MOUD initiation, monthly overall out-of-pocket spending among high-spending individuals was similar for all MOUD types (Figure 2, eTable 6).

Figure 2. Mean monthly overall out-of-pocket spending among individuals above the 90th percentile with MOUD type category.

Figure 2.

Open in a new tab

Figure 2 shows mean monthly overall out-of-pocket spending for individuals above the 90th percentile for out-of-pocket spending within their MOUD type category from three months prior to three months after MOUD initiation. The y-axis shows mean out-of-pocket spending in dollars, and the x-axis shows time relative to MOUD initiation. For each month, the 90th percentile out-of-pocket spending value was calculated separately among those who initiated any MOUD type, oral naltrexone, extended-release naltrexone, and buprenorphine. Mean out-of-pocket spending was calculated among individuals above the 90th percentile within each respective category. The black line shows overall out-of-pocket spending among individuals on any MOUD type, the blue line shows out-of-pocket spending among individuals who initiated oral naltrexone, the green line shows out-of-pocket spending among those who initiated extended-release naltrexone, and the grey line shows out-of-pocket spending among those who initiated buprenorphine. Error bars represent 95% confidence intervals. Trends for median values are similar, but lower, shown in eTable 6.

Abbreviations: MOUD: medication for opioid use disorder

While monthly overall out-of-pocket spending was similar across MOUD types after MOUD initiation, MOUD-specific monthly out-of-pocket spending was significantly lower for those who initiated oral naltrexone compared with those who initiated extended-release naltrexone or buprenorphine (eTable 7).

3.4. Monthly healthcare utilization prior to MOUD initiation by MOUD type

Higher overall out-of-pocket spending among individuals who initiated oral or extended-release naltrexone prior to MOUD initiation was associated with higher utilization of detoxification, inpatient, and outpatient services compared with individuals who initiated buprenorphine, particularly in the month immediately prior to MOUD initiation (Table 2). In this month, the proportion of individuals with a detoxification episode was 14.8% (95% CI: 14.4%–15.2%) for all MOUD types, 28.8% (95% CI: 27.7%–30.0%) for those who initiated oral naltrexone, 25.5% (95% CI: 23.6%–27.5%) for those who initiated extended-release naltrexone, and 9.7% (95% CI: 9.3%–10.1%) for those who initiated buprenorphine. Additionally, among those who did have detoxification episodes, the mean out-of-pocket spending associated with these episodes were higher for those who initiated oral naltrexone compared with those who initiated buprenorphine (oral naltrexone: $785 [95% CI: $716-$853] per episode, extended-release naltrexone: $650 [95% CI: $532-$768] per episode, buprenorphine: $580 [95% CI: $527-$633] per episode). Patterns of health service utilization frequency and per-visit out-of-pocket spending were similar for inpatient and outpatient visits (eTables 8–10).

Table 2.

Healthcare utilization and associated out-of-pocket spending in the month prior to MOUD initiation, overall and by MOUD type

Utilization by MOUD type Proportion with at least one episode/stay, % (95% CI) Mean per-episode/stay out-of-pocket spending, $ (95% CI)
All MOUD
 Detoxification episode 14.8 (14.4–15.2) 672 (632–712)
 Any inpatient stay1 26.0 (25.5–26.5) 1,014 (970–1,057)
Oral naltrexone
 Detoxification episode 28.8 (27.7–30.0) 785 (716–853)
 Any inpatient stay1 51.6 (50.5–53.1) 1,251 (1,173–1,329)
Extended release naltrexone
 Detoxification episode 25.5 (23.6–27.5) 650 (532–768)
 Any inpatient stay1 43.6 (41.4–45.8) 1,007 (878–1,136)
Buprenorphine
 Detoxification episode 9.7 (9.3–10.1) 580 (527–633)
 Any inpatient stay1 16.7 (16.2–17.3) 802 (750–853)
Mean number of outpatient visits2, no (95% CI) Mean per-outpatient2 visit out-ofpocket spending, $ (95% CI)
All MOUD 5.3 (5.2–5.4) 375 (357–393)
Oral naltrexone 7.9 (7.8–8.1) 656 (589–723)
Extended release naltrexone 7.9 (7.5–8.1) 590 (518–662)
Buprenorphine 4.2 (4.1–4.2) 261 (250–273)
Open in a new tab

Abbreviations: MOUD: medication for opioid use disorder; CI: confidence interval

1

Any inpatient stay includes detoxification episodes, as well as inpatient stay encounters without revenue codes for detoxification

2

Outpatient visits include visits to: doctor’s office, hospital outpatient facility, emergency department, or another outpatient facility

Mean and median out-of-pocket spending values for all months (three months prior to three months after MOUD initiation) shown in eTables 8–10

4. Discussion

We found that while most individuals with OUD who initiated MOUD had low out-of-pocket spending, a substantial portion of those who initiated naltrexone were most likely to have very high out-of-pocket spending. This difference was largely driven by health service utilization, particularly inpatient detoxification services, in the month leading up to MOUD initiation. In fact, after MOUD initiation, individuals who initiated all three MOUD types had similar overall monthly out-of-pocket spending. We also determined that total out-of-pocket spending over the study period was not driven by MOUD drug costs: oral naltrexone had the lowest out-of-pocket spending for MOUD specifically, but the highest overall. While extended-release naltrexone and buprenorphine resulted in the highest MOUD-specific out-of-pocket spending, this spending was low relative to the total overall out-of-pocket amount from three months prior through three months after MOUD initiation.

Prior work has investigated costs and cost-effectiveness of different MOUD types. A retrospective study of commercially-insured individuals with OUD found that those who initiated extended-release naltrexone had stable healthcare costs from 12 months prior to 12 months after MOUD initiation, while those who initiated buprenorphine incurred increased costs overtime (Shah et al. 2018). However, out-of-pocket costs were not reported in this study. Meanwhile, a model-based analysis found that from both a health sector and societal perspective, buprenorphine was cost-effective compared with extended-release naltrexone and that this finding was robust to a wide range of sensitivity analyses (Murphy et al. 2019). A noted limitation of this modeling study was that it did not consider out-of-pocket costs to patients. While some studies have investigated MOUD prescription out-of-pocket costs (Strahan et al. 2023), to our knowledge, no study has reported overall out-of-pocket spending associated with the MOUD initiation process. Thus, while our study does not evaluate long-term outcomes or cost-effectiveness, it highlights differences in short-term cost burdens for patients initiating different MOUD types. In combination with patient preferences and existing evidence regarding clinical- and cost-effectiveness of different MOUD options, this could be important for providers and patients to consider when choosing MOUD.

A strength of our study is that it uses real-world administrative claims data from inpatient, outpatient, and pharmacy services to descriptively compare out-of-pocket spending and healthcare utilization associated with the MOUD initiation process among individuals with OUD initiating different MOUD types. The Patient-Centered Outcomes Research Institute (PCORI) recommends that researchers assess “the full range of clinical and patient-centered outcomes,” including, “the potential burdens and economic impacts” (Patient-Centered Outcomes Research Institute 2022). Understanding out-of-pocket spending associated with different treatments is integral to understanding patient burden (Patient-Centered Outcomes Research Institute 2022). Evidence from a randomized controlled clinical trial shows that initiating extended-release naltrexone may be more challenging than initiating buprenorphine due to the need to be abstinent from opioids for as much as one week prior to naltrexone initiation (Lee et al. 2018). Our results are consistent with this evidence, as we find that a higher proportion of people initiating naltrexone have detoxification episodes in the month prior to MOUD initiation. While the detoxification process has been described as a clinical barrier to naltrexone initiation, we find that higher detoxification-related economic costs may be associated with naltrexone induction, which could pose an additional burden for patients. Specifically among people with OUD, high out-of-pocket costs have been associated with lower initiation, retention and adherence to MOUDs for commercially-insured people with OUD (Morgan et al. 2022; Dunphy et al. 2021; Strahan et al. 2023). Thus, our findings may have implications for clinical decision-making about which MOUD type to initiate.

Our study has several limitations. First, it is cross-sectional and lacks detailed demographic or clinical information about underlying characteristics, such as disease severity and patient behavior, that may differ across individuals who initiate different MOUD types. Therefore, our study does not establish a causal relationship between type of MOUD initiated and out-of-pocket costs. Given this limitation, it is useful to consider the mechanisms that could be driving the observed association. Because detoxification is required for naltrexone initiation, it is likely that many individuals would go through the costly detoxification process with the intent to initiate naltrexone. This may drive observed differences between buprenorphine and both naltrexone types (as oral and extended-release naltrexone both require full detoxification). Alternatively, individuals with certain underlying characteristics (including medical complexity [e.g., psychiatric co-morbidities], OUD severity, or insurance plan characteristics) may be more likely to seek care in an inpatient setting and/or be prescribed naltrexone in such a setting, potentially due to setting-specific standard practices or provider preferences favoring oral or extended-release naltrexone. Differences in underlying patient and setting/provider characteristics could be driving observed differences between buprenorphine and naltrexone, and/or differences between oral and extended-release naltrexone. For example, prior research has found that oral naltrexone is disproportionately prescribed by psychiatrists, and individuals engaged in psychiatric care may have higher utilization of services compared to those who initiate the extended-release formulation (Morgan et al. 2018). Future research should further investigate these possible pathways to naltrexone initiation, including differences between oral and extended-release formulations. Irrespective of the causal pathways driving the observed association, characterizing the short-term out-of-pocket costs incurred with detoxification and other intensive services prior to naltrexone initiation is important for patients and providers as they engage in shared decision-making about choice of MOUD. A second limitation is that administrative claims are subject to data entry errors including incorrect cost-sharing values, ICD, NDC, or revenue codes. Third, as our study utilized data from a commercially-insured cohort, findings may not be generalizable to a publicly-insured population, for example, one study found different MOUD prescription out-of-pocket costs by insurance type, where Medicaid beneficiaries incurred the lowest costs (Strahan et al. 2023). Individuals with low incomes and/or with disabilities using Medicaid, and veterans using Veteran’s Health Administration-sponsored insurance, should be included in future research, as these populations may be disproportionately impacted by OUD. Last, we were unable to capture out-of-pocket costs associated with initiation of methadone, an effective MOUD; as methadone was not covered by commercial insurance at all until 2017, and only rarely thereafter (Barocas et al. 2021).

5. Conclusions

In conclusion, our findings suggest that individuals who initiate naltrexone have higher utilization of intensive health services in the period prior to MOUD initiation, resulting in significantly higher out-of-pocket spending in this short-term period compared with individuals who initiate buprenorphine. This may reflect the need for medically-managed opioid withdrawal before naltrexone initiation, which may not be required prior to buprenorphine initiation. Current evidence shows medically-managed withdrawal can be a clinical barrier for initiating naltrexone. However, few studies have investigated economic costs associated with the MOUD initiation process. We find that unanticipated high out-of-pocket costs, not directly attributable to the cost of the MOUD drug itself, may be associated with initiating naltrexone. These out-of-pocket costs may contribute to the induction challenge of naltrexone, and while we do not investigate long-term costs or cost-effectiveness, these pre-treatment costs may present a short-term economic shock to patients and increase patient burden. These results are relevant for policymakers and payers to understand the full “true cost” associated with MOUD choice, and how costs are passed on to patients. We believe that pre-treatment cost is a noteworthy aspect of MOUD treatment that should be considered in combination with patient preferences and evolving clinical effectiveness and cost-effectiveness evidence, as part of a patient-centered shared decision-making process between patients and their providers when deciding on treatment for OUD.

Supplementary Material

1

Highlights.

  • Initiating naltrexone was associated with higher out-of-pocket spending than buprenorphine

  • Spending differences were driven by healthcare utilization before initiation

  • Results should be considered for provider-patient MOUD treatment decision-making

  • Payers should consider non-drug costs associated with MOUD initiation

Funding:

This work was supported by the National Institute on Drug Abuse (grant numbers R01DA046527 [Morgan] and P30DA040500 [Morgan] and 1T32DA041898-01A1 [McCann]). Funding sources had no role in the research.

Footnotes

Declarations of Interest: None

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

References

  1. Barocas Joshua A., Morgan Jake R., Wang Jianing, Dylan McLoone Alysse Wurcel, and Stein Michael D.. 2021. “Outcomes Associated With Medications for Opioid Use Disorder Among Persons Hospitalized for Infective Endocarditis.” Clinical Infectious Diseases: An Official Publication of the Infectious Diseases Society of America 72 (3): 472–78. 10.1093/cid/ciaa062. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bisaga Adam, Mannelli Paolo, Sullivan Maria A., Vosburg Suzanne K., Compton Peggy, Woody George E., and Kosten Thomas R.. 2018. “Antagonists in the Medical Management of Opioid Use Disorders: Historical and Existing Treatment Strategies.” The American Journal on Addictions 27 (3): 177–87. 10.1111/ajad.12711. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Blanco Carlos, and Volkow Nora D.. 2019. “Management of Opioid Use Disorder in the USA: Present Status and Future Directions.” Lancet (London, England) 393 (10182): 1760–72. 10.1016/S0140-6736(18)33078-2. [DOI] [PubMed] [Google Scholar]
  4. Cochran Bryan N., Flentje Annesa, Heck Nicholas C., Bos Jill Van Den, Perlman Dan, Torres Jorge, Valuck Robert, and Carter Jean. 2014. “Factors Predicting Development of Opioid Use Disorders among Individuals Who Receive an Initial Opioid Prescription: Mathematical Modeling Using a Database of Commercially-Insured Individuals.” Drug and Alcohol Dependence 138 (May): 202–8. 10.1016/j.drugalcdep.2014.02.701. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Dunphy Christopher, Peterson Cora, Zhang Kun, and Jones Christopher M.. 2021. “Do Out-of-Pocket Costs Influence Retention and Adherence to Medications for Opioid Use Disorder?” Drug and Alcohol Dependence 225 (August): 108784. 10.1016/j.drugalcdep.2021.108784. [DOI] [PMC free article] [PubMed] [Google Scholar]
  6. Lee Joshua D., Nunes Edward V., Novo Patricia, Bachrach Ken, Bailey Genie L., Bhatt Snehal, Farkas Sarah, et al. 2018. “Comparative Effectiveness of Extended-Release Naltrexone versus Buprenorphine-Naloxone for Opioid Relapse Prevention (X:BOT): A Multicentre, Open-Label, Randomised Controlled Trial.” Lancet (London, England) 391 (10118): 309–18. 10.1016/S0140-6736(17)32812-X. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Merative. 2022. “Merative MarketScan Research Databases.” 2022. https://www.merative.com/documents/brief/marketscan-explainer-general. [Google Scholar]
  8. Morgan Jake R., Quinn Emily K., Chaisson Christine E., Ciemins Elizabeth, Stempniewicz Nikita, White Laura F., Linas Benjamin P., Walley Alexander Y., and Marc R. LaRochelle. 2022. “Variation in Initiation, Engagement, and Retention on Medications for Opioid Use Disorder Based on Health Insurance Plan Design.” Medical Care 60 (3): 256–63. 10.1097/MLR.0000000000001689. [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Morgan Jake R., Schackman Bruce R., Leff Jared A., Linas Benjamin P., and Walley Alexander Y.. 2018. “Injectable Naltrexone, Oral Naltrexone, and Buprenorphine Utilization and Discontinuation among Individuals Treated for Opioid Use Disorder in a United States Commercially Insured Population.” Journal of Substance Abuse Treatment 85 (February): 90–96. 10.1016/j.jsat.2017.07.001. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Morgan Jake R., Schackman Bruce R., Weinstein Zoe M., Walley Alexander Y., and Linas Benjamin P.. 2019. “Overdose Following Initiation of Naltrexone and Buprenorphine Medication Treatment for Opioid Use Disorder in a United States Commercially Insured Cohort.” Drug and Alcohol Dependence 200 (July): 34–39. 10.1016/j.drugalcdep.2019.02.031. [DOI] [PMC free article] [PubMed] [Google Scholar]
  11. Murphy Sean M., McCollister Kathryn E., Leff Jared A., Yang Xuan, Jeng Philip J., Lee Joshua D., Nunes Edward V., Novo Patricia, Rotrosen John, and Schackman Bruce R.. 2019. “Cost-Effectiveness of Buprenorphine-Naloxone Versus Extended-Release Naltrexone to Prevent Opioid Relapse.” Annals of Internal Medicine 170 (2): 90–98. 10.7326/M18-0227. [DOI] [PMC free article] [PubMed] [Google Scholar]
  12. National Institute on Drug Abuse. 2021. “Medications to Treat Opioid Use Disorder Research Report: Overview.” National Institutes of Health. December 2021. https://nida.nih.gov/publications/research-reports/medications-to-treat-opioid-addiction/overview#:~:text=Disorder%20Research%20Report-,Overview,a%20prescription%20opioid%20use%20disorder. [Google Scholar]
  13. Oesterle Tyler S., Thusius Nuria J., Rummans Teresa A., and Gold Mark S.. 2019. “Medication-Assisted Treatment for Opioid-Use Disorder.” Mayo Clinic Proceedings 94 (10): 2072–86. 10.1016/j.mayocp.2019.03.029. [DOI] [PubMed] [Google Scholar]
  14. Patient-Centered Outcomes Research Institute. 2022. “Patient-Centered Economic Outcomes.” October 5, 2022. https://www.pcori.org/research/about-our-research/patient-centered-economic-outcomes. [Google Scholar]
  15. Phillips Mark R., Wykoff Charles C., Thabane Lehana, Bhandari Mohit, Chaudhary Varun, and Retina Evidence Trials InterNational Alliance (R.E.T.I.N.A.) Study Group. 2022. “The Clinician’s Guide to p Values, Confidence Intervals, and Magnitude of Effects.” Eye (London, England) 36 (2): 341–42. 10.1038/s41433-021-01863-w. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Prel Jean-Baptist du, Hommel Gerhard, Röhrig Bernd, and Blettner Maria. 2009. “Confidence Interval or P-Value?: Part 4 of a Series on Evaluation of Scientific Publications.” Deutsches Arzteblatt International 106 (19): 335–39. 10.3238/arztebl.2009.0335. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Shah Ankit, Duncan Margaret, Atreja Nipun, Tai Kei Sing, and Gore Mugdha. 2018. “Healthcare Utilization and Costs Associated with Treatment for Opioid Dependence.” Journal of Medical Economics 21 (4): 406–15. 10.1080/13696998.2018.1427101. [DOI] [PubMed] [Google Scholar]
  18. Strahan Andrea E., Desai Shaina, Zhang Kun, and Guy Gery P.. 2023. “Trends in Out-of-Pocket Costs for and Characteristics of Pharmacy-Dispensed Buprenorphine Medications for Opioid Use Disorder Treatment by Type of Payer, 2015 to 2020.” JAMA Network Open 6 (2): e2254590. 10.1001/jamanetworkopen.2022.54590. [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. U.S. Department of Health & Human Services. 2021. “Overdose Prevention Strategy.” HHS.Gov. October 27, 2021. https://www.hhs.gov/overdose-prevention/.

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

1
NIHMS1954143-supplement-1.docx (64KB, docx)

RESOURCES