1 Identifying Barriers to Uptake of Medication Assisted Therapy Among Patients Diagnosed with Opioid Use Disorder June 4, 2019 Ezra Fishman, Geetanjoli Banerjee, John Barron, and Gosia Sylwestrzak HealthCore, Inc., Wilmington, DE efishman@healthcore.com
Disclosures 2 All authors are employees of HealthCore, Inc., a wholly-owned, independently-operated research subsidiary of Anthem, Inc. Anthem Inc. had no role in the conduct of the analysis or the decision to submit the abstract. Results do not necessarily reflect the opinions or policies of Anthem Inc.
Motivation: MAT is effective but under-utilized 3 • Among patients with opioid use disorder (OUD), medication- assisted treatment (MAT), in conjunction with psychosocial treatment, reduces non-medical opioid use and all-cause and overdose mortality (Sordo 2017 BMJ ). • However, MAT is under-utilized (McCarty 2017 Ann Rev Pub Health , Saloner 2015 JAMA ). Reasons include: • Persistent expectations of abstinence as the proper treatment course • Lack of prescribers licensed to dispense MAT – https://www.samhsa.gov/medication-assisted-treatment/training-materials-resources/buprenorphine-waiver
MAT uptake among patients diagnosed with OUD 4 OBJECTIVE 1 OBJECTIVE 2 Quantify MAT Identify barriers to uptake among MAT uptake patients newly among such patients diagnosed with OUD
Study population 5 • N=70,437 patients below age 65 and newly diagnosed with OUD between 01 January 2013 and 30 April 2018. • All patients had ≥12 months’ continuous enrollment in commercial health plan prior to first OUD diagnosis date (index date). • We followed patients from the first OUD diagnosis date until the date of first MAT dispensing/administration (the outcome of interest), death, disenrollment from the health plan, or end of study period. • Data were queried in August 2018 from our commercial claims database, the HealthCore Integrated Research Database (HIRD SM ).
OUD is not only a disorder for young adults 6 30 25 23.3% Percent of Cohort (%) 23.1% 20 18.1% 16.6% 16.3% 15 10 5 2.5% 0 0-17 18-24 25-34 35-44 45-54 55-64 Age at Index (years)
OUD diagnosis became more common over time 7 30 25.0% 25 23.5% Percent of Cohort (%) 20.9% 20 16.8% 13.9% 15 10 5 0 2013 2014 2015 2016 2017 Index Year Note: Patients diagnosed in 2018 were excluded from this figure. They represented 3.6% of the cohort because data were only available from the first quarter of 2018.
Other baseline characteristics of study cohort 8 47% 67% Rx 24% FEMALE documented Rx fill for opioid in baseline period saw a “special specialist” on date of OUD diagnosis 89% OUD diagnosis characterized as moderate to severe (ICD-10 dx code F11.20 or equivalent) saw a specialist in psychiatry, addiction 37% medicine, or pain medicine (a “special specialist”) in the baseline period 29% Rx fill for benzodiazepine in mean Elixhauser 2.2 baseline period Comorbidity Index (range: 0-19)
Distribution of follow-up time 9 26 24 22 Percent of Cohort (%) 20 Mean: Median: Max: 18 16.4% 13.4 mo 9.3 mo 64 mo 16 14 12.6% 12.3% 12.2% 12 10.6% 10.6% 10.1% 10 8.1% 7.1% 8 6 4 2 0 <1 1-2 3-4 5-7 8-11 12-17 18-23 24-29 30+ Months at risk
Outcome summary 10 16,577 patients (=24%) initiated MAT in follow-up period 21 patients initiating MAT per 100 person-years of follow-up
Analytic method 11 • Cox proportional hazards model • The following predictors of MAT uptake were included: • Patient age • Sex • Elixhauser comorbidity index (ECI) • Year of diagnosis of OUD • Geographic region on date of diagnosis • Encounter with “special specialist” on index date • Any benzo Rx in baseline period • Any opioid Rx fill in baseline period • Index OUD diagnosis classified as moderate/severe • Outcome = first MAT dispensing, death, or disenrollment from plan, whichever comes first Special specialist = Physician with specialty in psychiatry, addiction medicine, or pain medicine
Factors associated with MAT uptake 12 Male Special Specialist Index Dt Benzo Rx Fill Baseline Opioid Rx Fill Baseline OUD Diagnosis Moderate/Severe 0.9 1 1.1 1.2 1.3 1.4 1.5 1.6 1.7 1.8 Adjusted hazard ratio, 95% CI Higher hazard ratio = more likely to initiate MAT
Factors associated with MAT uptake 13 +1 yr age +1 ECI NE (vs. MW) S (vs. MW) W (vs. MW) 0.7 0.8 0.9 1 1.1 1.2 1.3 Adjusted Hazard Ratio, 95% CI Higher hazard ratio = more likely to initiate MAT
After adjusting for covariates, each year of diagnosis 14 after 2013 was associated with lower MAT uptake 2014 2015 2016 2017 2018 0.5 0.6 0.7 0.8 0.9 1 1.1 Adjusted hazard Ratio, 95% CI Higher hazard ratio = more likely to initiate MAT
Summary 15 • Only ¼ of newly diagnosed OUD patients initiated MAT. • Estimate from National Survey of Drug Use and Health: 21.5%* • Diagnosis by a “special specialist” was associated with higher MAT uptake. • Males diagnosed with OUD uptake MAT at higher rates than females. Why? *Saloner B, Karthikeyan S. Changes in substance abuse treatment use among individuals with opioid use disorders in the United States, 2004-2013. JAMA . 2015 Oct 13;314(14):1515-7. Special specialist = Physician with specialty in psychiatry, addiction medicine, or pain medicine.
Summary 16 • More-recent years associated with lower MAT uptake. Possible explanations: • Increasing provider tendency to apply OUD diagnosis (so less-severe patients get diagnosed) without concurrent increase in tendency to prescribe MAT • Increasing tendency for OUD to be diagnosed by physicians not authorized to dispense MAT • “Depletion of the susceptibles” - Those most receptive to treatment were already getting it early on • Crowding of behavioral health services
Strengths and limitations of the study 17 • Large sample size from all regions of the U.S. • Control for secular increase in OUD diagnosis when assessing Strengths other barriers to MAT uptake • Connect OUD diagnosis to MAT dispensing • Unmeasured confounding • Population “at risk” of MAT is defined imprecisely Limitations • Methadone dispensing at clinic is likely unobserved • Opioids obtained without insurance involvement are unobserved • Illicit opioid use is unobserved
Implications 18 Future efforts to increase MAT uptake may require increased involvement of, and patient access to, specialists in addiction medicine, pain management, and psychiatry. Greater efforts to train primary care physicians and other specialists in implementation of MAT to treat OUD may also reduce barriers to OUD treatment.
Thank you 19 Ezra Fishman, PhD Contact: efishman@healthcore.com Geetanjoli Banerjee, PhD John Barron, PharmD Gosia Sylwestrzak, MA
20
Appendix material 21
Full cohort inclusion/exclusion criteria 22 # Criterion Count 1 Starting Population: Any diagnosis code for opioid use disorder at any time 349,760 2 Index OUD Diagnosis in intake period 210,396 4 Exclude those with faulty Death Dates 210,305 5 Continuously Enrolled 12 Months Baseline 111,493 6 No Baseline OUD Diagnosis 102,544 7 Age LT 65 91,524 8 No Cancer 87,013 9 No Hospice 86,504 10 No Baseline MAT 73,406 11 Any Encounter in Baseline Period 70,438 12 Exclude those with faulty birth date 70,437
Regional distribution of study cohort 23 Percent of patients by region on index date Unknown 6% MW 22% W 22% NE 14% S 35% MW = Midwest, NE = Northeast, S = South, W = West.
Age pattern of OUD diagnosis in our cohort resembles 24 age pattern in opioid overdose hospitalization rates Opioid overdose hospitalization rate, 2015, United States Nonfatal overdose hospitalizations 42.4 45 40 36.3 35 per 100,000 persons 29.5 28.7 28 30 24.2 25 20 15 9.1 10 5 0.8 0 0-14 15-19 20-24 25-34 35-44 45-54 55-64 65+ Age Source: Centers for Disease Control and Prevention (CDC). 2018. Highlights from the 2018 Annual Surveillance Report of Drug-Related Risks and Outcomes - United States. Slide 31. https://www.cdc.gov/drugoverdose/pdf/pubs/CDC_2018_Surveillance-Report_DataSummary_presentation.pdf Weighted national estimates from Healthcare Cost and Utilization Project Nationwide Inpatient Sample, 2015.
Recommend
More recommend
