Table 2.

Association between stigmatizing attitudes and opioid use disorder (OUD) treatment practices among U.S. primary care physicians^1,2 (N=336)

Dependent variables	Average marginal effect of stigma³ (95% CI)
Provision of OUD treatment
Prescribe either buprenorphine or injectable extended-release naltrexone for OUD	−0.1%^** (0.16, −0.07)
Prescribe buprenorphine (also called Suboxone or Subutex) for OUD	−0.08^*** (0.12, −0.03)
Prescribe injectable extended-release naltrexone (also called Vivitrol) for OUD	−0.07^*** (0.11, −0.03)
Provide OUD counseling themselves	−0.03 (−0.11, 0.05)
Referral to OUD treatment⁴
Refer patients to another clinician who prescries buprenorphine, injectable extended-release naltrexone, or methadone for OUD	−0.20^*** (0.29, −0.11)
Refer patients to another clinician who prescribes buprenorphine for OUD	−0 17^*** (0.26, −0.08)
Refer patients to another clinician who prescribes injectable extended-release naltrexone for OUD	−0.07^* (0.14, 0.00)
Refer patients to an OUD methadone treatment program	−0.04 (−0.10, 0.03)
Refer patients to OUD counseling	−0.02 (−0.10, 0.06)
Refer patients to OUd counseling outside their practice	0.03 (−0.07, 0.12)
Refer patients to OUD counseling inside their practice	−0.10^* (0.17, −0.03)

p<0.05

^**

p<0.01

^***

p<0.001

All models control for sex, age, region, highest degree (MD/DO), year of graduation from medical school, specialty, practice type, and personal experience (self, close friend, or family member) with OUD and are weighted for nonresponse.

Independent variable is a continuous summary stigma measure created using the mean of an individuals’ responses to the 12 stigma items (range: 1–5) with 1 indicating the lowest degree of stigma and 5 indicating the highest degree of stigma.

The average marginal effect indicates the percentage point change in the predicted probability of the outcome associated with a one-unit change in the stigma scale (range 1–5).

⁴

Among primary care physician respondents who do not provide treatment themselves.