Skip to main content
. Author manuscript; available in PMC: 2022 May 1.
Published in final edited form as: Drug Alcohol Depend. 2021 Mar 18;222:108674. doi: 10.1016/j.drugalcdep.2021.108674

Table 4.

Multivariable logistic regression analyses examining factors associated with needing behavioral healthcare in the past 12 months but not having received it among transgender adults who use substances in Massachusetts and Rhode Island (N = 326).

Block 1: Demographics Block 2: Demographics, Victimization, Mental Health, Social Support & Substance Misuse
Multivariable Multivariable
aOR 95% CI P-Value aOR 95% CI P-Value
DEMOGRAPHICS - CURRENT
Young Adult
 No (30+) 1.00 ---- ---- 1.00 ---- ----
 Yes (18–29) 2.67 1.34–5.33 0.01 2.79 1.30–5.99 0.01
Race/ethnicity
 White (non-Hispanic) 1.00 ---- ---- 1.00 ---- ----
 Racial/ethnic minority 1.95 0.99–3.87 0.05 1.95 0.88–4.30 0.10
Gender Identity
 Trans woman 0.82 0.37–1.83 0.62 0.71 0.29–1.75 0.46
 Trans man 0.65 0.33–1.30 0.22 0.65 0.30–1.40 0.27
 Non-binary 1.00 ---- ---- 1.00 ---- ----
Health Insurance Status
 No insurance 4.28 0.98–18.74 0.05 4.12 0.83–20.46 0.08
 Public insurance 1.44 0.73–2.84 0.29 1.49 0.68–3.26 0.32
 Private insurance 1.00 ---- ---- 1.00 ---- ----
VICTIMIZATION - PAST 3 YEARS
Experiencing Stigma in a Healthcare Setting
 No 1.00 ---- ----
 Yes 2.45 1.14–5.26 0.02
MENTAL HEALTH & SELF-HARM
Post-Traumatic Stress Disorder - Current
 No 1.00 ---- ----
 Yes 0.41 0.19–0.87 0.02
Attempted Suicide - Lifetime
 No 1.00 ---- ----
 Yes 0.58 0.28–1.16 0.12
SOCIAL SUPPORT - CURRENT
Family Support of Gender Identity/Expression
 Not supportive ---- ---- ----
 Somewhat supportive 0.31 0.12–0.80 0.02
 Supportive 0.45 0.14–1.39 0.17
SUBSTANCE MISUSE - PAST 12 MONTHS
Number of Other Substances
 Continuous 0.78 0.57–1.07 0.12
Opioid Pain Medication
 No 1.00 ---- ----
 Yes 3.14 1.03–9.63 0.04

OR=Odds Ratio; aOR= Adjusted Odds Ratio; NSSI = non-suicidal self-injury; bolded values = significant at p<0.05.

*

Not inclusive of opioid pain medications

Note. Outcome: Needing behavioral healthcare in the past 12 months but did not receive it (ref=needed behavioral healthcare and received it). Blocked multivariable regression modeling was used. Demographics were entered into the first block. Variables significant at p<0.25 were entered into the final multivariable model.