https://orcid.org/0000-0001-9054-5933
ABSTRACT
Introduction
Buprenorphine medication assisted treatment (B-MAT) adherence for opioid use disorder (OUD) is suboptimal. reSET-O, an FDA-cleared prescription digital therapeutic, delivers neurobehavioral therapy (community-reinforcement approach+fluency training+contingency management) to B-MAT-treated OUD patients.
Methods
This retrospective claims study (10/01/2018-10/31/2019) evaluated healthcare resource utilization up to 6 months before/after reSET-O initiation. Repeated-measures negative binomial models compared incidences of encounters/procedures. Net change in costs was assessed.
Results
Among 351 patients (mean age 37; 59.5% female; 82.6% Medicaid), 334 had pharmacy claims and 240 (71.9%) received buprenorphine pre-/post-index (medication possession ratio 0.73 and 0.82, respectively; P = 0.004). Facility encounters decreased, with 45 fewer inpatient (P = 0.024) and 27 fewer emergency department (ED) visits (P = 0.247). Clinical encounters with largest changes were drug testing (638 fewer; P < 0.001), psychiatry (349 fewer; P = 0.036), case management (176 additional; P = 0.588), other pathology/laboratory (166 fewer; P = 0.039), office/other outpatient (154 fewer; P = 0.302), behavioral rehabilitation (111 additional; P = 0.124), alcohol/substance rehabilitation (96 fewer; P = 0.348), other rehabilitation (66 fewer; P = 0.387), mental health rehabilitation (61 additional; P = 0.097), and surgery (60 fewer; P = 0.070). Changes in facility/clinical encounters saved $2,150/patient.
Conclusion
reSET-O initiation was associated with fewer inpatient, ED, and other clinical encounters, increased case management/rehabilitative services, and lower net costs over six months.
Expert Opinion
Real-world evidence is helpful in evaluating the effectiveness of interventions in usual-care conditions, outside of controlled research environments. Large observational studies based on health care claims are important to understand the actual pharmacoeconomic and outcomes impact of interventions at the health care system and population level.
1. Introduction
Over the last 40 years, the mortality rate for unintentional drug poisonings has increased exponentially at a rate of approximately 7% per year and has doubled approximately every 10 years [Citation1]. In recent years, opioid use disorder (OUD) has been responsible for approximately two out of every three deaths related to substance use disorder (SUD) [Citation1]. Opioid use disorder (OUD) is a chronic disease characterized by a cluster of cognitive, behavioral, and physiological symptoms indicating that an individual continues using opioid substances despite significant substance-related problems [Citation2]. OUD is defined by a problematic pattern of opioid use that has far-reaching impacts on families, communities, and society in general [Citation1,Citation3–6]. The economic burden of prescription opioid overdose, abuse, and dependence has continued to grow and was estimated to be $5040020 billion in 2015, stemming from fatalities, criminal justice and healthcare costs, and foregone earnings [Citation4,Citation7]. The cost of OUD to the US healthcare system alone is estimated at approximately $30B per year, and is largely driven by excess hospital and emergency department (ED) encounters [Citation4,Citation7]. Compounding the problem, the COVID-19 epidemic has caused significant disruption to OUD treatment programs and has contributed to patients’ loneliness, boredom, and anxiety due to social isolation, increasing the risk of OUD relapse and overdose [Citation8,Citation9]. Indeed, many U.S. states have reported a sharp increase in opioid-related mortality since the start of the COVID-19 outbreak [Citation10].
The evidence-based standard of care for OUD is medication-assisted treatment (MAT) which by definition includes pharmacotherapy (i.e. buprenorphine, methadone, or naltrexone) combined with counseling and behavioral therapies [Citation11,Citation12]. The most common form of MAT is that of buprenorphine (B-MAT) which is associated with reductions in utilization of high-cost health care services (such as inpatient/intensive care unit [ICU] stays and ED visits) as well as total cost of care [Citation7,Citation11–17]. Additionally, B-MAT is associated with reductions in opioid overdose and other outcomes such as mental health, employment, and stable housing rates with longer treatment duration [Citation11,Citation13]. However, the benefits of B-MAT are not realized by many OUD patients, particularly those who discontinue medication early [Citation18]. One of the well-documented reasons for premature discontinuation of B-MAT is lack of neurobehavioral therapy [Citation19,Citation20]. Though the relationship between the presence of neurobehavioral and persistence with B-MAT is well documented, few programs exist that effectively increase the probability of neurobehavioral therapy among B-MAT patients.
Digital technologies have been recognized as promising ways to better assess, understand, and treat SUDs [Citation21]. In clinical use since January 2019, reSET-O® is an FDA-cleared, 84-day prescription digital therapeutic (PDT) indicated for improving retention in outpatient therapy for patients with OUD treated with buprenorphine. It delivers a form of evidence-based neurobehavioral therapy founded on the community reinforcement approach (CRA), an intensive form of cognitive behavioral therapy (CBT) validated for SUD/OUD, as a series of 67 on-demand audio and video lessons which are sequentially unlocked as patients progress through the program. Fluency training uses a simple quiz format to reinforce retention and understanding of positive adaptive behaviors immediately following each lesson. Contingency management, another proven, evidence-based neurobehavioral therapy for SUD/OUD treatment, follows each lesson/fluency training session dyad to immediately recognize and reward patients’ engagement with therapy. Patients earn rewards (merit badges or gift cards of modest value) by spinning a virtual rewards wheel for up to four lessons per week, and for each negative drug urine screen logged.
The reSET-O pivotal study showed that 82% of OUD patients who received treatment with reSET-O stayed in treatment versus 68% of those who only received treatment as usual (TAU), and the likelihood of abstinence during weeks 9–12 was 77.3% vs 62.1%, respectively [Citation22]. The objective of this study was to evaluate the real-world impact on healthcare resource utilization (HCRU) in the earliest identified cohort of patients to have been prescribed reSET-O.
2. Materials and methods
This study was a real-world, retrospective pre-post cohort study.
2.1. Data
This study used medical and pharmacy claims from the HealthVerity PrivateSource20 (PS20) database between 10/01/2018 – 10/31/2019. The database contains closed medical and pharmacy claims for approximately 70 million commercial, 60 million Medicaid, and 15 million Medicare enrollees represented across 150 payers since 2015. This study received a waiver of authorization for the use and disclosure of protected health information (PHI) and a determination of exempt status under 45 CFR § 46.104(d)(4) from Western Institutional Review Board on 2 April 2020.
2.2. Patient population
Claims were identified for adult patients who successfully activated reSET-O between 1 January 2019 and 10 April 2019. The index date was the date of reSET-O activation. Continuous enrollment in the medical (and pharmacy, as applicable) plan was required for ≥4 weeks in the 6-month pre-index and post-index periods.
2.3. Study measures
This study evaluated patient demographic characteristics included age, sex, payer type, and OUD diagnosis type. Claims in the pre-index and post-index period were identified as facility claims or clinical service claims in order to characterize patients’ HCRU. Facility encounters included all-cause inpatient stays, ICU, ED, partial hospitalizations (PH), and surgical outpatient department (SOD) identified from facility claims. Clinical service encounters included categories of all-cause Current Procedure Terminology (CPT) codes identified from clinician claims such as evaluation and management codes, medical codes (e.g. cardiovascular, psychiatry, neurology), pathology, and laboratory, and rehabilitative services. Buprenorphine adherence was assessed using the medication possession ratio (MPR) on filled prescriptions in both the pre-index and post-index periods. Costs associated with facility and clinical service encounters were evaluated.
2.4. Analyses
HCRU was evaluated with incidence rates for each type of facility and clinical service encounter using a repeated-measures negative binomial model adjusted for the observed number of days in the 6-month pre-index and post-index periods ( presents the study design). The adjusted number of facility and clinical service encounters were derived from the incidence rates by multiplying the incidence rate by the sample size (i.e. 351), and the difference in encounters between the pre-index and post-index periods was then calculated. No adjustments for multiplicity were undertaken. Adjusted mean MPR was calculated as the total days’ supply of buprenorphine within the post-index period divided by the total number of days in the post-index period [Citation23]. A scenario analysis of the cost impact of changes in facility and clinical service encounters was conducted using published facility costs for OUD patients ($11,731 for inpatient stays, $20,500 for ICU stays, and $504 for ED visits) [Citation24] and 2020 Medicare reimbursement rates for remaining facility and clinician services [Citation25,Citation26].
Figure 1. Study design
3. Results
3.1. Patient selection and attrition
Of 1,265 patients who activated reSET-O between 1 January 019 and 10 April 2019, 399 (31.5%) had medical enrollment data in PS20. Among those, 351 patients had ≥4 weeks of continuous pre-index and post-index medical enrollment and 334 also had ≥4 weeks of continuous pre-index and post-index medical and pharmacy enrollment ().
Figure 2. Patient enrollment
3.2. Demographic and clinical characteristics
The mean (SD) patient age was 37 (8.6) years with 79.5% of patients between the ages of 25 and 44 years. Over half (59.5%) of patients were female; most were covered by Medicaid (82.6%), and most had 10th revision International Classification of Disease (ICD-10) diagnosis codes indicative of uncomplicated OUD (90.2%) (). Among those with pharmacy claims 71.9% had a buprenorphine pharmacy claim before and after reSET-O initiation, and overall 76.7% of patients had a buprenorphine pharmacy claim before initiating reSET-O, compared to 72.8% following reSET-O initiation. Adherence to buprenorphine was high among the 240 patients with buprenorphine use in both the pre-index period (MPR = 0.73; 75.4% with MPR≥ 0.6) and the post-index period (MPR = 0.82; 78.8% with MPR≥ 0.6) ().
Table 1. Demographic characteristics of patients who initiated re-SET-O
Table 2. Buprenorphine adherence among patients with buprenorphine prescriptions pre- and post- re-SET-O initiation
3.3. Facility encounters comparison
The facility encounters with the largest decrease were inpatient stays and ED visits. There were 45 fewer adjusted inpatient stays (62% reduction; P = 0.024), and 27 fewer adjusted ED visits (20% reduction; P = 0.247). Among the inpatient stays, four included ICU stays, all of which occurred in the pre-index period. There were four additional partial hospitalizations (16 in the pre-index period, 20 in the post-index period; P = 0.813), and four fewer surgical outpatient department visits (5 in the pre-index period, none in the post-index period; P = NA) ().
Table 3. Incidence of facility and clinical service encounters with the largest difference 6 months pre- and post- re-SET-O initiation (ranked by magnitude of the difference)
3.4. Clinical service encounters comparison
Clinical service encounters experienced large adjusted pre-post changes in utilization (). The clinical encounters categories with the largest adjusted change were: drug testing (638 fewer; P < 0.001), psychiatry (349 fewer; P = 0.036), case management (176 additional; P = 0.588), other pathology and lab (166 fewer; P = 0.039), office/other outpatient (154 fewer; P = 0.302), behavioral health rehabilitative services (111 additional; P = 0.124), alcohol and substance rehabilitative services (96 fewer; P = 0.348), other rehabilitative services (69 fewer; P = 0.387), mental health rehabilitative services (61 additional; P = 0.097), and surgery (60 fewer; P = 0.070). Full results are shown in Supplementary Table S1.
3.5. Scenario analysis of costs
Facility costs were estimated to have experienced a net decrease of $603,965, including reductions of $522,933 for inpatient stays, $81,309 for ICU stays, and $13,799 for emergency department visits. Clinical encounters were estimated to have experienced a net decrease of $150,692. Combined, these represent cost savings of $754,657, or $2,150 per patient ().
Table 4. Scenario analysis of costs related to facility and clinical encounters resource utilization
4. Discussion
This real-world claims data analysis of an early cohort of patients initiating reSET-O showed a significant decrease in facility encounters and clinical encounters at 6 months post-reSET-O initiation compared to the preceding 6 months. A total of 75 unique facility encounters were avoided, including 45 inpatient hospital stays and 27 ED visits, and among the top 10 clinical encounters there were close to 1181 fewer claims post-index. These findings are especially noteworthy given the high buprenorphine treatment and adherence rates observed in this cohort both pre- and post-index and support an incremental benefit of reSET-O in real-world, usual care conditions.
Notably, there was a significant number of pre- and post-index hospital-related visits (228 pre-index and 153 post-index) in this highly buprenorphine-adherent population. These findings underscore the challenges of OUD treatment and highlight the value of adding behavioral treatment delivered by a PDT which has been shown to improve retention and abstinence in a randomized trial [Citation22]. Patients retained in treatment are in a better position to avoid exposure to illicit opioids, greatly reducing their risk of an overdose or acute care event. The clinically meaningful post-index decreases in hospital and emergency department-related visits observed in this study (72 per 351 treated patients, or approximately one event avoided per 5 treated patients) suggests that overall total costs may be reduced upon reSET-O initiation. However, this will need to be confirmed in future real-world evidence studies. Nevertheless, a calculation of the potential reduction of costs associated with hospital and emergency department encounters suggests that more than $750,000 in savings would have been realized through the treatment of this cohort of 351 patients with reSET-O, representing a savings of $2,150 per patient. Given that the cost assumptions used in this study are for the most part conservative, a more comprehensive cost analysis incorporating additional cost scenarios and changes in efficiency of care (i.e. impact of reduced, per-patient face-to-face clinician time through reSET-O-enabled CBT delivery) should be conducted to determine additional near-term cost impact scenarios.
It is also noteworthy that drug testing was the encounter category with the largest change in this study, and its reduction could be a cause for concern given the importance of objectively measuring abstinence from illicit opioids and other substances. On the other hand, a reduction in the volume of drug testing could be indicative of a reduction in treatment intensity as patients are able to cope better and as cravings stabilize [Citation27], an observation which is supported by the high buprenorphine adherence rate post index, the observed reduction in inpatient, emergency department, and the majority of clinical encounters, and the concomitant increase in utilization of case management and behavioral rehabilitative services (176 and 111 additional visits post-index, respectively). Although a causal effect on these reductions in all-cause facility and clinical encounters cannot be attributed to reSET-O, the findings are strongly suggestive.
Furthermore, the findings are consistent with other published claims data analyses in patients with OUD. Studies by Ruetsch and colleagues and Lynch and colleagues have also shown decreased emergency department and inpatient visits in patients with more comprehensive therapy [Citation16] and with adherence to MAT [Citation15]. Similar to Lynch and colleagues, the present study showed fewer mental health and primary care visits [Citation16], a finding which the authors hypothesized could be related to increased coordination and consolidation of care [Citation16]. Lastly, similar to Ruetsch and colleagues and other studies, the present study also observed increases in select clinician encounters that may be indicative of greater patient engagement with treatment, such mental and behavioral rehabilitative services [Citation15–17].
As with any healthcare claims-based analysis, there are potential limitations and mitigation factors that should be noted. First, the majority of the patients that were identified in the claims data source were Medicaid patients; thus, the results may be less generalizable to patient with other types of insurance. Additionally, mortality cannot be assessed using claims data and it is possible that some patients may have died in the early (<4 weeks) or late (>6 months) post-index period. Given the excess risk of accidental overdose with opioids, this important outcome will need to be assessed in future analyses.
The pre-post study design may be prone to bias given the strong potential for increased health care resource utilization in the pre-index period, prior to the intensification of OUD treatment with reSET-O. However, this approach avoids the pitfalls of comparing to dissimilar populations while allowing for an examination of the overall impact of the intervention with each patient acting as their own control. Using this approach, reductions in hospitalizations and relapse indicators were observed despite high adherence levels to buprenorphine in both the pre- and post-index periods. In this study, adherence to buprenorphine was defined as an MPR >0.6 which is the adherence level at which Ruetsch and colleagues observed a dramatic improvement in health care resource utilization.
Opioid addiction has been reported to have a seasonal variation, which was not evaluated in this study. Given the study period, there were not enough patients in each season to examine potential association on HCRU incidence rates but will be examined as more patients initiate reSET-O therapy in the future.
Lastly, the lack of follow-up beyond 6 months is a limitation. Nonetheless, the analyses showed early reductions in hospital-based care in a relatively large real-world sample of mostly Medicaid-covered patients. The avoidance of hospital stays may drive significant cost reductions over the near-term and warrants further study.
5. Conclusions
In an early cohort of patients with opioid use disorder, hospital-based HCRU and the majority of clinical encounters decreased, while clinical encounters associated with recovery and rehabilitation increased following initiation of a novel prescription digital therapeutic delivering cognitive behavioral therapy, fluency training, and contingency management. The findings from this study indicate a strong potential for quality improvement and overall near-term total cost reduction in this patient population through a new low-barrier approach to the delivery of neurobehavioral therapy.
Author contribution statement
All authors were involved with conception, study design, interpretation of the data, writing the manuscript, and the final approval of the version to be published.
Declaration of interest
FFV is an employee of Pear Therapeutics, Inc. SC, LK, CR, and KA provided consulting services to Pear Therapeutics, Inc. for the conduct of the study. No author received an honorarium related to the development of this manuscript. The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.
Reviewers disclosure
Peer reviewers on this manuscript have no relevant financial relationships or otherwise to disclose.
Supplemental Material
Download MS Excel (17.3 KB)Acknowledgments
The authors would like to thank Dr. Sean Murphy for his helpful insights during critical review of the results, Dr. Heather Shapiro for her assistance with data integrity and manuscript review, Natalie Edwards for assistance with writing support and critical review of the manuscript, and most of all to the patients for their courage in the face of all the challenges on the road to recovery.
Data availability statement
The data that support the findings of this study are available from the authors upon reasonable request and with permission from HealthVerity.
Supplementary material
Supplemental data for this article can be accessed here.
Additional information
Funding
References
- Jalal H, Buchanich JM, Roberts MS, et al. Changing dynamics of the drug overdose epidemic in the United States from 1979 through 2016. Science. 2018 Sept 21;361(6408).
- American Psychiatric Association (APA). Diagnostic and statistical manual of mental disorders. 5thed. (DSM-5). Washington (DC): (APA) APA, editor; 2013.
- Hagemeier NE. Introduction to the opioid epidemic: the economic burden on the healthcare system and impact on quality of life. Am J Manag Care. 2018 May;24(10 Suppl):S200–S206.
- Florence CS, Zhou C, Luo F, et al. The economic burden of prescription opioid overdose, abuse, and dependence in the United States, 2013. Med Care. 2016;54(10):901–906.
- Ray GT, Mertens JR, Weisner C. The excess medical cost and health problems of family members of persons diagnosed with alcohol or drug problems. Med Care. 2007 Feb;45(2):116–122.
- Kirson NY, Scarpati LM, Enloe CJ, et al. The economic burden of opioid abuse: updated findings. J Manag Care Spec Pharm. 2017 Apr;23(4):427–445.
- The Council of Economic Advisers. The underestimated cost of the opioid crisis. 2017 [cited 2020 July 22]. Available from: https://www.whitehouse.gov/briefings-statements/cea-report-underestimated-cost-opioid-crisis/
- Becker WC, Fiellin DA. When epidemics collide: coronavirus disease 2019 (COVID-19) and the opioid crisis. Ann Intern Med. 2020;173(1):59–60.
- The Opioid Response Network. The opioid epidemic and COVID-19: how social determinants of health create differential impacts. 2020 [cited 2020 July 22]. Available from: https://opioidresponsenetwork.org/documents/SDH_OUD-COVID-19.pdf
- American Medical Association (AMA). Issue brief: reports of increases in opioidrelated overdose and other concerns during COVID pandemic. 2020 [cited 2020 July 22]. Available from: https://www.ama-assn.org/system/files/2020-07/issue-brief-increases-in-opioid-related-overdose.pdf
- Wakeman SE, Larochelle MR, Ameli O, et al. Comparative effectiveness of different treatment pathways for opioid use disorder. JAMA Network Open. 2020;3(2):e1920622–e1920622.
- Boyett B. The individualized treatment of opioid use disorder. J Manag Care Spec Pharm. 2019 June;25(6):634–636.
- U.S. Department of Health & Human Services. Facing addiction in America - the surgeon general’s spotlight on opioids. In: Services USDoHH, editor. Washington, DC; 2018.
- Ling W, Nadipelli VR, Solem CT, et al. Effects of monthly buprenorphine extended-release injections on patient-centered outcomes: a long-term study. J Subst Abuse Treat. 2020 Mar;110:1–8.
- Ruetsch C, Tkacz J, Nadipelli VR, et al. Heterogeneity of nonadherent buprenorphine patients: subgroup characteristics and outcomes. Am J Manag Care. 2017 June 1;23(6):e172–e179.
- Lynch FL, McCarty D, Mertens J, et al. Costs of care for persons with opioid dependence in commercial integrated health systems. Addict Sci Clin Pract. 2014 Aug 14;9(1):16.
- Ronquest NA, Willson TM, Montejano LB, et al. Relationship between buprenorphine adherence and relapse, health care utilization and costs in privately and publicly insured patients with opioid use disorder. Subst Abuse Rehabil. 2018;9:59–78.
- Madras B, Ahmad NJ, Wen J, et al. and The Prevention, Treatment, and Recovery Working Group of the Action Collaborative on Countering the U.S. Opioid Epidemic. Improving Access to evidence-based medical treatment for opioid use disorder: strategies to address key barriers within the treatment system. 2020 [cited 2020 July 22]. Available from: https://nam.edu/improving-access-to-evidence-based-medical-treatment-for-opioid-use-disorder-strategies-to-address-key-barriers-within-the-treatment-system/
- Bentzley BS, Barth KS, Back SE, et al. Discontinuation of buprenorphine maintenance therapy: perspectives and outcomes. J Subst Abuse Treat. 2015;52:48–57.
- Dutra L, Stathopoulou G, Basden SL, et al. A meta-analytic review of psychosocial interventions for substance use disorders. Am J Psychiatry. 2008 Feb;165(2):179–187.
- Marsch LA, Campbell A, Campbell C, et al. The application of digital health to the assessment and treatment of substance use disorders: the past, current, and future role of the national drug abuse treatment clinical trials network. J Subst Abuse Treat. 2020 Mar;112S:4–11.
- Christensen DR, Landes RD, Jackson L, et al. Adding an Internet-delivered treatment to an efficacious treatment package for opioid dependence. J Consult Clin Psychol. 2014;82(6):964–972.
- Cramer JA, Roy A, Burrell A, et al. Medication compliance and persistence: terminology and definitions. Value Health. 2008 Jan-Feb;11(1):44–47.
- Premier. Opioid overdoses costing U.S. hospitals an estimated $11 billion annually. 2019 [cited 2020 Jul 28]. Available from: https://www.premierinc.com/newsroom/press-releases/opioid-overdoses-costing-u-s-hospitals-an-estimated-11-billion-annually
- Centers for Medicare and Medicaid Services (CMS). Medicare hospital outpatient prospective payment system addendum A, April 2020 file. 2020 [cited 2020 Jul 28]. Available from: https://www.cms.gov/Medicare/Medicare-Fee-For-Service-Payment/HospitalOutpatientPPS/Addendum-A-and-Addendum-B-Updates
- Centers for Medicare and Medicaid Services (CMS). Physician fee schedule relative value file, April 2020 release (RVU20B). 2020 [cited 2020 July 28]. Available from: https://www.cms.gov/medicaremedicare-fee-service-paymentphysicianfeeschedpfs-relative-value-files/rvu20b
- Barbour Jordan S, Jarvis Margaret A, Withers David J. How geisinger dramatically reduced deaths from opioid use disorder. NEJM Catal. 2020 Mar-Apr;1(2):1–20.