opioid use disorder: not enough use of replacement therapy

A recent study found that prescribing buprenorphine or methadone was confirmed to be the most effective treatments for those with opioid use disorder, but these treatments were dramatically underutilized (see opioid use disorder insuff ORT jama2020 in dropbox,  or doi:10.1001/jamanetworkopen.2019.20622.

Details:
-- 40,885 individuals with opioid use disorder (OUD) were evaluated, from the OptumLabs Data Warehouse, a large database of people with commercial or Medicare Advantage insurance, including medical, behavioral health, and pharmacy claims. From 2015-2017
-- mean age 48 (with pretty even distribution amongst the 10-year age groups), 54% male, 74% white/8% Hispanic/12% black, 58% commercial insurance/42% Medicare Advantage
-- mental health diagnoses: 45% overall (depression 24%, anxiety 26%, alcohol 10%, bipolar 8%, ADHD 4%, PTSD 4%, psychosis 4%)
-- injection drug use 14%, hepatitis C 5%, prior opioid overdose 5%
-- OUD defined as:
    -- 1 or more inpatient claims, or 2 or more outpatient claims with OUD diagnostic codes within 3 months of each other
    -- 1 or more claims for OUD plus diagnostic codes for opioid-related overdose, injection-related infection, or inpatient detoxification or residential services
    -- medication for OUD (MOUD) pharmacy claims
-- treatments received:
    -- non-intensive behavioral health: 24,258 people (59%)
    -- inpatient detoxification or residential services: 6455 (16%)
    -- Medication for OUD (MOUD) with buprenorphine or methadone: 5123 (13%)
    -- intensive behavioral health: 1970 (5%)
    -- MOUD with naltrexone: 963 (2%)
    -- no treatment: 2116 (5%)
-- mean length of stay in inpatient detox or residential services: 10 days
-- continuous commercial health insurance: 48% were disenrolled by 12 months after the index date
    -- those on buprenorphine/methadone or on naltrexone had the highest disenrollment rates (54%). No difference in those disenrolled by race/ethnicity, comorbidities, or markers of severe OUD (including history of injection-related infection, hepatitis C infection or overdose)

Results:
-- at 3 months follow-up: 707 patients (2%) experienced an overdose, and  773 (2%) had serious opioid-related acute care use

-- risk of overdose, as compared to those receiving no treatment (hazards ratios adjusted for age, sex, race/ethnicity, insurance type, baseline cost rank (i assume that reflects how much $$ they used), mental health and medical comorbidities, injection-related infections or OD):
    -- 3 months: buprenorphine or methadone: 76% reduced risk, adjusted HR 0.24 (0.14-0.41)
    -- 12 months:  buprenorphine or methadone: 59% reduced risk, aHR 0.41 (0.31-0.55)

-- risk of serious opioid-related acute care, as compared to those receiving no treatment:
    -- 3 months: buprenorphine or methadone: 32% reduced risk, aHR 0.68 (0.47-0.99); nonintensive behavioral health services; 41% reduction, aHR 0.59 (0.44-0.80)
    -- 12 months: buprenorphine or methadone: 26% reduced risk, aHR 0.74 (0.58-0.95)

-- all other treatments were statistically nonsignificant
    -- naltrexone: 41% decrease in overdoses, aHR 0.59 (0.29-1.20)

-- all treatment groups, when compared to buprenorphine/methadone, were associated with posttreatment admission to inpatient detox:
    -- those initiating treatment with inpatient detox or residential  service were  3½ times more likely to return within 3 months and 12 months
    --and, those initiating treatment with naltrexone or intensive behavioral health services having twice the risk of subsequent detox admission at 3 months and 12 months
-- treatment duration: mean duration of 74 days for naltrexone and 150 days for buprenorphine/methadone, though those who received longer duration MOUD had lower rates of overdose or serious opioid-related acute care use; at the end of 12 months:
    -- those not on MOUD: 1198 people (4%) had OD, 1204 (4%) had serious opioid-related acute care use
    -- MOUD for 1-30 days: 105 (6%) had OD, 133 (8%) had serious opioid-related acute care use
    -- MOUD for 31-180 days: 101 (3%) had OD, and 148 (5%) serious opioid-related acute care use
    -- MOUD for >180 days: 28 (1%) had OD, and 69 (3%) had serious opioid-related acute care use

Commentary:
-- as per prior blogs, OUD has had profound adverse individual and social effects, including 47,600 overdoses in 2017; associated morbidity/mortality from increasing cases of hepatitis C, HIV, and serious injection-related infections; increasing cost from hospitalizations of more than $700 million annually; and profound effects on their families and community
-- this  real-world analysis above did confirm that MOUD, especially with buprenorphine/methadone, was very effective in preventing several serious adverse outcomes: overdose, the risk of serious opioid-related acute care, and post-treatment admission for inpatient detox. And this study augments the relatively few prior studies that compare MOUD to nonpharmacologic therapies.

-- of note, increases in opioid-related deaths have occurred in the setting of some decreases in prescription opioid supply in the community, likely related to decreased prescriptions by clinicians (68% of OD deaths in 2017 were still from prescription opioids). This all emphasizes the importance of also decreasing the social and economic issues creating the demand for OUD (eg, see http://gmodestmedblogs.blogspot.com/2020/01/opioid-od-mortality-increased-with-auto.html)

-- obstacles to buprenorphine or methadone access include:
    -- lack of access to buprenorphine prescribers in many parts of the country. The concern here is that there is a special training required to be a buprenorphine prescriber, yet buprenorphine is by far the safest opioid we have; in contrast pretty much anyone with a license to practice medicine or dentistry can prescribe OxyContin, fentanyl, and other potentially highly toxic opiates
    -- insurance obstacles such as requiring prior approvals for medications, which clearly limit their being prescribed regularly
    -- inability to prescribe methadone for OUD in the community: allowing community-based clinicians to prescribe methadone maintenance, especially in cases where patients are intolerant of buprenorphine, have insufficient effect from buprenorphine, and are not interested in going to methadone clinics on a very regular basis to get their methadone (e.g., the clinics are far away and the patient needs to work, patients have concerns about privacy/confidentiality, the methadone clinics are not helpful/patient-centered/friendly/and some are now run by private corporations, they prefer getting medications from their well-established primary-care site and clinician, etc.) in terms of privatization of methadone clinics, see https://www.npr.org/sections/health-shots/2016/06/10/480663056/investors-see-big-opportunities-in-opioid-addiction-treatment

-- a few major concerns with detox programs:
    -- perhaps the biggest is that they don’t work very well. Repeated studies since the 1960s have confirmed their low rate of success. Yet they are pushed, often from a moralistic perspective similar to what gave rise to Prohibition in the late 1910s
    -- and, detoxification may lead to increased sensitivity to opiates (e.g. up-regulation of mu-receptors) and potentially to overdoses related to people unwittingly resuming their old opiate doses after release from the detox program, similar to the increased overdoses found in people released from jail
    -- few addiction treatment programs offer MOUD; my limited experience with patients going through detox programs is that even when we strongly suggested that they call us/come by immediately on release from the detox programs to be put on buprenorphine, this happens relatively infrequently
-- in a Massachusetts study of 17,568 adults who survived in opioid overdose between 2012-2014, those on methadone maintenance therapy for a median of 5 months, buprenorphine for median of 4 months, or naltrexone for median of one month had less than half the all-cause and opioid-related mortality (see opiate after OD bup methadone dec mortality MA AIM2018 in dropbox, or doi:10.7326/M17-3107)
-- Hard to comment much about the role of naltrexone. Relatively few patients were on it and they were on it an average of only 2½ months, likely  playing a part in the lack of statistical significance, especially since the trend to benefit was quite strong

-- Limitations of study:
    -- this study was limited to patients with commercial insurance or Medicare advantage programs and not those either without insurance or on Medicaid.
    -- they could not distinguish between disenrollment attributable to death vs changes in health insurance
    -- the treatment duration for MOUD was only 2½ months for naltrexone and 4 months for buprenorphine/methadone (though those who received longer duration MOUD had lower rates of overdose or serious opioid related acute care use).  ?was the short duration of treatments related to changes in insurance/death of patients. Also those who continue with buprenorphine/methadone are the select group who are most likely to have benefit [of note, the 12-month benefit rates were less than the 3-month ones, likely related to the fact that the mean duration of buprenorphine/methadone was only 5 months]
    -- data were not available for those who access public sector treatments, especially for methadone, which is not covered by Medicare and may well have significant copayments for those with commercial insurance plans
    -- there are limited granular data: unclear what the exact content of the various strategies were.  ?If those on MOUD had counseling, support groups, were seen in rapid turnover clinics vs integrated programs vs in the context of primary care, ?types and intensity of behavioral health interventions, ?how aggressive follow-up was for those who missed appointments
    -- this was an observational study. Patients assigned to these different treatment groups may be very different from each other. So, as with observational studies in general, it is hard to attribute outcomes with certainty to the different treatment groups (e.g. confounding by unknown parameters)
    -- also, it is hard to get high-level, consistent data in the US healthcare system, where patients are in and out of insurance, being forced to change insurance plans with different systems of care and coverage at the discretion of employers, in and out of access to prescription meds, and there are no all-inclusive high-quality national  databases.

So,
-- the study does confirm that treatment with buprenorphine or methadone is the most effective therapy for OUD, with major decreases in several major OUD problems, including ODs, need for acute inpatient services, and subsequent admission for inpatient detox. there were no mortality data in this study, though several others have found decreased overdose and all-cause mortality
--BUT only 1 in 8 patients in this study were actually put on these buprenorphine/methadone, despite the fact that studies the last  50 to 60 years have consistently found that methadone (in the older studies) was dramatically more effective when given as maintenance, perhaps with very slowly decreasing doses, vs detox. Newer studies also confirm that buprenorphine is best given as maintenance therapy (eg, see opioid bup maint vs detox jamaintmed2014 in dropbox, or doi:10.1001/jamainternmed.2014.5302)
    -- most patients in the new study above initiated treatment with psychosocial services alone or inpatient detox
-- and, results are much better with longer-term involvement in buprenorphine/methadone programs, suggesting:
    -- there should be very easy, immediate access to these medications, with essentially no barriers to care
    -- the importance of having systems to support patients with OUD to help them remain in treatment. The system should have both clinical and public health initiatives, including aggressive public campaigns and education, outreach into high-risk areas, individualized treatment plans and outreach as needed, etc.
    -- and, we as a country need to focus on the social basis for opiate (and other harmful substance) use: the increasing social inequities. one recurrent example we see is patients getting out of jail, and because of their incarceration are not able to get a job/get housing/have a satisfying life. and end up using and/or selling drugs again just to deal with their circumstances/survive

geoff​

If you would like to be on the regular email list for upcoming blogs, please contact me at gmodest@uphams.org

to get access to all of the blogs:

1. go to http://gmodestmedblogs.blogspot.com/ to see them in reverse chronological order

2. click on 3 parallel lines top left, if you want to see blogs by category, then click on "labels" and choose a category​

3. or you can just click on the magnifying glass on top right, then  type in a name in the search box and get all the blogs with that name in them

please feel free to circulate this to others. also, if you send me their emails, i can add them to the list