Abstract
This study uses simulated patient calls to a random sample of US residential addiction treatment facilities to investigate the availability of opioid agonist treatment (buprenorphine-naloxone) and nonpharmacologic therapies (eg, CBT) for opioid use disorder, and differences by for-profit vs nonprofit center status.
Residential treatment programs are a common, costly setting for treating opioid use disorder1 and are frequently promoted in policy proposals to improve access to care.2 However, concerns have been raised about the quality of care and low use of evidence-based opioid agonist treatment (OAT) in these settings.3 Most data available on programs come from self-reported surveys4 that may not reflect patient experience, so we simulated patient calls to audit treatments offered by programs nationally.
Methods
We surveyed a random sample of residential treatment programs in the United States identified from publicly available federal directories and listings of search engine advertisements. From a sampling frame of 1436 facilities, we randomly selected an equal proportion of for-profit and nonprofit (including public) programs. To reach our target number of respondents, 3 trained research assistants called 613 programs with a standardized script from June 27 to September 20, 2019. Callers posed as 27-year-old individuals who use heroin and lack insurance. They spoke with the first available person who managed prospective admissions. Callers inquired about whether they could receive OAT, specifically mentioning “Suboxone” (buprenorphine-naloxone). We systematically collected information on availability of OAT and non-OAT treatments. We also recorded whether any anti-OAT messaging (eg, “substituting one addiction for another”) was used. We stratified outcomes by facility profit status, accreditation, and state licensure. We estimated logistic regression models controlling for facility characteristics to assess independent factors associated with OAT-related outcomes. P values were estimated using 2-sided Wilcoxon rank sum tests or χ2 tests. Analyses were performed in Stata version 14 (StataCorp). The Harvard University institutional review board determined that the analysis was not human subjects research.
Results
From the 613 programs contacted, 160 were excluded as out of sample (4 most common reasons: outpatient-only program [67]; served special populations [55]; accepted only insured patients [19]; and nonworking phone [16]). We completed contact with 368 of 453 in-sample programs (81% response rate), or 26% of nonfederal programs nationally. Nonrespondents were more likely than respondents to be nonprofit (61% vs 27%, respectively).
Nationally, 107 programs (29%) offered OAT with the option to continue maintenance (Table 1). An additional 114 (31%) offered OAT only for short-term detoxification, while 143 programs (39%) did not offer OAT or were unclear about whether OAT was available. Seventy-eight programs (21%) actively discouraged callers from using OAT.
Table 1. Characteristics of Residential Treatment Programs by Reported Availability of OATa.
| All respondents, No. (%) | OAT available for maintenance treatment, No. (%)b | Adjusted difference, % (95% CI)c | OAT available for detoxification only, No. (%) | Adjusted difference, % (95% CI)c | OAT not available or unclear, No. (%) | Adjusted difference, % (95% CI)c | Anti-OAT statements made, No. (%)d | Adjusted difference, % (95% CI)c | |
|---|---|---|---|---|---|---|---|---|---|
| Total | 368 (100) | 107 (29) | 114 (31) | 143 (39) | 78 (21) | ||||
| Profit status | |||||||||
| For profit | 226 (61) | 66 (29) | 1 [Reference] | 97 (43) | 1 [Reference] | 61 (27) | 1 [Reference] | 57 (25) | 1 [Reference] |
| Not for profit | 120 (33) | 36 (31) | 4.8 (−7.3 to 16.8) | 17 (14) | −12.2 (−23.8 to −0.7) | 65 (55) | 7.2 (−4.0 to 18.4) | 20 (17) | −2.1 (−13.0 to 8.9) |
| Public (local and state)e | 22 (6) | 5 (23) | −3.8 (−23.9 to 16.3) | 0 | NA | 17 (77) | 30.7 (9.6 to 51.8) | 1 (5) | −17.9 (−29.3 to −6.4) |
| Offering detoxification | |||||||||
| Yes | 163 (44) | 53 (33) | 1 [Reference] | 69 (43) | 1 [Reference] | 38 (24) | 1 [Reference] | 31 (19) | 1 [Reference] |
| No | 163 (44) | 38 (23) | −11.3 (−22.4 to −0.3) | 28 (17) | −11.2 (−22.4 to −0.0) | 96 (59) | 23.4 (12.4 to 34.4) | 33 (20) | 4.7 (−5.3 to 14.7) |
| Missing data | 42 (11) | 16 (38) | 7.9 (−18.8 to 34.6) | 17 (40) | 0.9 (−26.6 to 28.5) | 9 (21) | 5.2 (−17.4 to 27.7) | 14 (33) | 3.1 (−17.1 to 23.3) |
| CARF accredited | |||||||||
| Yes | 97 (26) | 27 (28) | −2.7 (−14.9 to 9.5) | 24 (25) | 10.8 (−1.9 to 23.6) | 45 (47) | −3.1 (−13.8 to 7.5) | 20 (21) | 7.5 (−4.7 to 19.7) |
| No | 271 (74) | 80 (30) | 1 [Reference] | 90 (34) | 1 [Reference] | 98 (37) | 1 [Reference] | 58 (22) | 1 [Reference] |
| Joint Commission accredited | |||||||||
| Yes | 178 (48) | 53 (30) | −2.0 (−14.4 to 10.4) | 87 (50) | 29.9 (17.0 to 42.8) | 35 (20) | −26.0 (−38.3 to −13.8) | 48 (27) | 9.9 (−1.4 to 21.3) |
| No | 190 (52) | 54 (29) | 1 [Reference] | 27 (14) | 1 [Reference] | 108 (57) | 1 [Reference] | 30 (16) | 1 [Reference] |
| State licensed | |||||||||
| Yes | 336 (91) | 99 (30) | 8.1 (−8.1 to 24.2) | 99 (30) | −9.8 (−26.5 to 7.0) | 135 (41) | 2.3 (−15.1 to 19.8) | 66 (20) | −9.9 (−27.1 to 7.3) |
| No | 32 (9) | 8 (26) | 1 [Reference] | 15 (48) | 1 [Reference] | 8 (26) | 1 [Reference] | 12 (39) | 1 [Reference] |
Abbreviations: CARF, Commission on Accreditation of Rehabilitation Facilities; NA, not applicable; OAT, opioid agonist therapy.
Facilities with missing data on OAT availability and anti-OAT statements (n = 4) were excluded from all estimates.
OAT available for maintenance treatment, OAT available for detoxification only, and OAT not available or unclear were 3 mutually exclusive outcomes for OAT availability at facilities. Percentages may not sum exactly to 100% because of rounding.
Adjusted marginal differences were estimated using logistic regression models and the “margins” function in Stata (version 14) for each of the outcomes above, controlling for all variables in the table as well as length of residential treatment offered (>30 days, ≤30 days, or both) and Census division of the facility (9 divisions: East North Central, East South Central, Middle Atlantic, Mountain, New England, Pacific, South Atlantic, West North Central, and West South Central).
“Anti-OAT statements made” was an independent outcome capturing whether facilities discouraged simulated patients from pursuing OAT when they inquired about its availability. The 3 most common anti-OAT statements were that OAT is “not clean” (n = 32), “trading one addiction for another” (n = 22), or that OAT is “addictive” (n = 16).
Public facilities (a subset of nonprofit facilities) were defined as those owned and operated by local or state governments.
The availability of OAT with maintenance treatment was not significantly different at nonprofit and for-profit programs (31% in nonprofit vs 29% in for-profit; adjusted difference, 4.8%; 95% CI, −7.3% to 16.8%), while publicly operated programs were significantly more likely to not offer or be unclear about OAT availability (77% in public vs 27% in for-profit; adjusted difference, 30.7%; 95% CI, 9.6%-51.8%). Programs’ accreditation or state licensure had no significant association with OAT availability for maintenance treatment or anti-OAT messaging.
Almost all programs (335 [92%]) offered some form of 12-step program (Table 2). Overall, 38 non-OAT treatment modes were offered by 5 or more facilities, such as cognitive behavior therapy (106 [29%]) or animal therapy (34 [9%]). Programs without clear OAT availability offered fewer non-OAT treatments (median, 2 [interquartile range, 1-4]) than those offering OAT for maintenance therapy (median, 4 [interquartile range, 2-6]; P < .001).
Table 2. Nonmedication Treatments Offered, by OAT Availabilitya.
| Total | OAT availability | |||||
|---|---|---|---|---|---|---|
| Maintenance | Detoxification only | P value, detoxification vs maintenanceb | Not available or unclear | P value, not available or unclear vs maintenanceb | ||
| No. of programs | 368 | 107 | 114 | 143 | ||
| No. of non-OAT treatments offered, median (IQR) | 3 (1-5) | 4 (2-6) | 5 (3-6) | .22 | 2 (1-4) | <.001 |
| Top 10 non-OAT treatments offered, No. (%)c | ||||||
| 12-Step program | 335 (92) | 95 (90) | 109 (96) | .13 | 129 (91) | .92 |
| Cognitive behavior therapy | 106 (29) | 26 (24) | 48 (42) | .005 | 32 (22) | .72 |
| Yoga | 65 (18) | 27 (25) | 28 (25) | .91 | 10 (7) | <.001 |
| Dialectical behavior therapy | 59 (16) | 23 (21) | 23 (20) | .81 | 13 (9) | .006 |
| Relapse prevention | 53 (14) | 19 (18) | 13 (11) | .18 | 21 (15) | .51 |
| Meditation or mindfulness | 51 (14) | 16 (15) | 24 (21) | .24 | 11 (8) | .07 |
| Group therapy | 37 (10) | 17 (16) | 5 (4) | .004 | 15 (10) | .21 |
| Art therapy | 35 (10) | 10 (9) | 17 (15) | .22 | 8 (6) | .26 |
| Animal therapy (eg, dogs, horses) | 34 (9) | 13 (12) | 19 (17) | .34 | 2 (1) | <.001 |
| Exercise program | 32 (9) | 9 (8) | 15 (13) | .26 | 8 (6) | .38 |
Abbreviation: OAT, opioid agonist therapy.
Facilities with missing data on OAT availability (n = 3), 12-step program availability (n = 3), or both (n = 1) were excluded from all estimates.
P values were estimated using a Wilcoxon rank sum test (No. of non-OAT treatments) or a χ2 test (individual non-OAT treatments).
Multiple answers per program were possible. There were 38 total treatments offered by 5 or more residential programs. The 28 other treatments offered were eye movement desensitization and reprocessing (n=31 programs), family programming (n=30), outdoor/adventure programming (n=30), SMART Recovery (n=28), trauma-based programming (n=27), acupuncture (n=26), music therapy n=(24), massage (n=22), faith-based programming (n=21), anger management (n=14), coping skills (n=14), general education (n=13), general individual therapy (n=12), nutrition (n=11), refuge recovery (n=10), life skills (n=10), biofeedback (n=10), process group (n=8), chiropractor (n=7), seeking safety (n=7), breathing exercises (n=6), rational emotive behavioral therapy (n=6), emotion management (n=6), experiential therapy (n=6), psychoeducation (n=6), brain mapping (n=5), health and wellness programming (n=5), and trigger awareness (n=5).
Discussion
In this national survey of residential programs for opioid use disorder, 29% of programs offered OAT as maintenance therapy, the standard of care for opioid use disorder, while many actively discouraged use of OAT to callers. Thirty-one percent of programs offered OAT only for detoxification, which has worse outcomes compared with OAT maintenance.5
The presence of licensure or accreditation did not ensure availability of OAT or low use of anti-OAT language. Therefore, these “seals of approval” do not appear to guide consumers to programs consistently offering the most effective treatment for opioid use disorder or direct them away from those discouraging its use.
This study has limitations. Differences between nonrespondents and respondents could bias results. Because callers spoke only with the first available person, facilities’ true breadth of treatments may not have been captured. Also, callers had a fixed script representing an uninsured, cash-paying individual, so results may not generalize to other populations.
Overall, these findings raise concerns about the quality of care offered by residential programs.
Section Editor: Jody W. Zylke, MD, Deputy Editor.
References
- 1.Substance Abuse and Mental Health Services Administration Behavioral Health Spending & Use Accounts, 2006-2015 Published February 2019. Accessed October 30, 2019. https://store.samhsa.gov/product/Behavioral-Health-Spending-and-Use-Accounts-2006-2015/SMA19-5095
- 2.Christie C, Baker C, Cooper R, Kennedy PJ, Madras B, Bondi P. The President’s Commission on Combating Drug Addiction and the Opioid Crisis: Final Report Published November 1, 2017. Accessed July 9, 2020. https://www.whitehouse.gov/sites/whitehouse.gov/files/images/Final_Report_Draft_11-15-2017.pdf
- 3.Wakeman SE, Larochelle MR, Ameli O, et al. . Comparative effectiveness of different treatment pathways for opioid use disorder. JAMA Netw Open. 2020;3(2):e1920622. doi: 10.1001/jamanetworkopen.2019.20622 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Substance Abuse and Mental Health Services Administration National Survey of Substance Abuse Treatment Services (N-SSATS): 2018 Published September 2019. Accessed October 30, 2019. https://www.samhsa.gov/data/sites/default/files/cbhsq-reports/NSSATS-2018.pdf
- 5.Fiellin DA, Schottenfeld RS, Cutter CJ, Moore BA, Barry DT, O’Connor PG. Primary care-based buprenorphine taper vs maintenance therapy for prescription opioid dependence: a randomized clinical trial. JAMA Intern Med. 2014;174(12):1947-1954. doi: 10.1001/jamainternmed.2014.5302 [DOI] [PMC free article] [PubMed] [Google Scholar]