Skip to main content
. Author manuscript; available in PMC: 2022 Apr 3.
Published in final edited form as: Drug Alcohol Depend. 2021 Aug 12;227:108958. doi: 10.1016/j.drugalcdep.2021.108958

Table 1.

Agreement between PRISM-5 DSM-5 and DSM-IV past-year substance use disorder diagnoses (N = 588).

Substance % DSM-5 % DSM- IV McNemar’s testa p-value Kappab (95% CI)

Alcohol 66.0 59.4 <.001 0.84 (0.79, 0.88)
Tobacco 62.1 52.2 <.001 0.80 (0.75, 0.85)
Cannabis 44.6 33.2 <.001 0.75 (0.70, 0.80)
Cannabis without withdrawal 43.7 33.2 <.001 0.76 (0.71, 0.81)
Cannabis without craving 39.5 33.2 <.001 0.78 (0.73, 0.84)
Cannabis without both withdrawal and craving 39.0 33.2 <.001 0.83 (0.78, 0.88)
Cocaine 44.7 40.5 <.001 0.91 (0.87, 0.94)
Heroin 24.1 23.6 0.083 0.99 (0.97, 1.00)
Opioids 15.6 14.6 0.014 0.96 (0.93, 0.99)
Sedatives 12.8 11.4 0.011 0.92 (0.87, 0.97)
Stimulants 2.9 2.7 0.317 0.97 (0.91, 1.00)
a

McNemar’s test evaluated whether there were significant differences between DSM-IV and DSM-5 SUD prevalences.

b

Degree of agreement, indicated by kappa values: 0–0.20, poor; 0.21–0.40, fair; 0.41–0.60, moderate; 0.61–0.80, substantial; and 0.81–1.00, excellent.