Skip to main content
. 2021 Nov 1;17(11):2205–2214. doi: 10.5664/jcsm.9418

Table 3.

Multivariate logistic regression analysis determining associations between insufficient sleep and both lifetime and current prescription opioid misuse.

Variables Lifetime Prescription Opioid Misuse Current Prescription Opioid Misuse
aOR 95% CI P aOR 95% CI P
Sleep
 Sleeps 8–9 h per night Ref Ref
 Sleeps < 8 h per night 1.4 (1.1–1.2) .006 1.0 (0.7–1.4) .850
Age 1.0 (0.9–1) .36 0.9 (0.8–1.1) .238
Sex
 Male Ref Ref
 Female 1.0 (0.8–1.3) .971 1.0 (0.8–1.2) .878
Race and ethnicity
 White, non-Hispanic Ref Ref
 Black, non-Hispanic 1.4 (0.9–2.0) .116 1.8 (1–3.2) .035
 Hispanic 1.0 (0.7–1.6) .856 1.9 (1.2–3.1) .013
 Asian 1.0 (0.6–1.5) .928 1.3 (0.8–2.3) .322
 American Indian/Alaska Native 1.5 (0.4–5.8) .58 0.6 (0.2–2.4) .495
 Native Hawaiian/other Pacific Islander 0.6 (0.1–3.4) .597 1.7 (0.3–9.9) .539
 Multiple race 1.3 (0.9–1.9) .103 1.8 (1.2–2.8) .006
Current tobacco use
 No Ref Ref
 Yes 1.2 (1–1.6) .066 1.6 (1.2–2.1) .002
Current marijuana use
 No Ref Ref
 Yes 1.9 (1.5–2.4) <.0001 1.3 (1–1.7) .059
Current alcohol use
 No Ref Ref
 Yes 1.7 (1.4–2.1) <.0001 2.3 (1.7–3.1) <.0001
Felt sad or hopeless
 No Ref Ref
 Yes 2.3 (2–2.7) <.0001 2.4 (1.8–3.2) <.0001
Sexual assault history
 No Ref Ref
 Yes 2.0 (1.4–2.9) <.0001 2.0 (1.5–2.8) <.0001
Threatened or injured with a weapon on school property
 No Ref Ref
 Yes 2.0 (1.5–2.7) < .0001 1.9 (1.3–2.9) .002

aOR = adjusted odds ratio, CI = confidence interval, Ref = reference.