Skip to main content
. 2021 Sep 14;16:66. doi: 10.1186/s13011-021-00398-3

Table 3.

Determine and compare the validity of self-reported substance use in Rafsanjan Youth Cohort Study and primary health care setting (N = 1007)

Rafsanjan Youth Cohort Study (research setting) primary health care setting compare the validity of self-reported substance use in two settings
Drug/gender Sensitivity 95% CI Specificitya 95% CI Conditional κ 95% CI Sensitivity 95% CI Specificitya 95% CI Conditional κ 95% CI p-valuec
opioid /total
past 72 h 39.4 (22.9–57.9) 100 (99.4–100) 0.381 (0.376–0.385) 50 (32.4–67.6) 100 (99–100) 0.478 (0.470–0.486) 0.383
past 12 months 57.6 (39.2–74.5) 0.523 (0.522–0.524) 64.7 (46.5–80.3) 0.614 (0.608–0.620) -d
opioid /men
past 72 h 37.5 (21.1–56.3) 100 (99.2–100) 0.360 (0.353–0.366) 53.3 (34.3–71.7) 100 (98.2–100) 0.498 (0.482–0.515) 0.211
past 12 months 56.3 (37.7–73.6) 0.501 (0.499–0.503) 70 (50.6–85.3) 0.651 (0.640–0.660) -d
opioid /women
past 72 h 100 (2.5–100) 100 (96.8–100) 1 1 25 (.63–80.6) 100 (97.8–100) 0.246 (0.242–0.250) 0.171
past 12 months 100 (2.5–100) 1 1 25 (.63–80.6) -d
Cannabis/men
past 72 h 10 (.253–44.5) 100 (99.4–100) 0.10 (0.096–0.101) 12.5 (1.55–38.3) 100 (98.3–100) 0.117 (0.105–0.130) 0.846
past 12 months 20 (2.52–55.6) 0.187 (0185–0.189) 12.5 (1.55–38.3) 0.105 (0.094–0.117) -d
Methadone/total
past 72 h 20 (5.73–43.7) 100 (99.4–100) 0.195 (0.192–0.198) 20.7 (7.99–39.7) 100 (99–100) 0.199 (0.190–0.208) 0.953
past 12 months 40 (19.1–63.9) 0.384 (0.381–.0386) 31 (15.3–50.8) 0.291 (0.283–0.299) -d
Methadone/men
past 72 h 21.1 (6.05–45.6) 100 (99.2–100) 0.210 (0.199–0.208) 24 (9.36–45.1) 100 (98.2–100) 0.220 (0.202–0.237) 0.817
past 12 months 42.1 (20.3–66.5) 0.402 (0.399–0.405) 36 (18–57.5) 0.327706 (0.312–0.343) -d
Methadone/women
past 72 h 0.000 -b 100 0.00 0.00 100 0.00
past 12 months 0.000 -b 0.00 0.00 0.00
amphetamine and Methamphetamine/men
past 72 h 0.000 -b 100 0.00 0.00 100 0.00
past 12 months 0.000 -b 0.00 0.00 0.00

a Specificity was not calculated for past-12-month use because of the potential for inflated false negatives

b Confidence interval could not be calculated because sensitivity was zero

cChi-Square test: Comparison of self-reported substance use in two settings among urinalysis-confirmed users

dp-value was not calculated to compare of self-reported substance use over the past year, as some people may have correctly self-report of substance use over the past year, but a urine test could not determine substance use over the past year