Table 3.
Association between cognitive function (CF) and cumulative lifetime exposure to marijuana in ‘marijuana-years’ among those without recent use.a
| Standardized difference in each CF measure (95% CI)c | ||||
|---|---|---|---|---|
| Cognitive Function Measure |
Unadjusted model |
Adjusted for age, race, sex, education, study center, and with IPCW f |
Additionally adjusted for substance use, depression and cardiovascular risk factors g |
Additionally adjusted for mirror star tracing at the Year 2 visit h |
| - Cumulative lifetime exposure in marijuana-years b | ||||
|
Rey Auditory Verbal Learning Test (RAVLT) |
||||
| - Never used marijuana | 0 (Ref.) | 0 (Ref.) | 0 (Ref.) | 0 (Ref.) |
| - 1 day to <0.5 marijuana-years | 0.06 (−0.04 to 0.16) | −0.01 (−0.11 to 0.08) | −0.02 (−0.12 to 0.08) | −0.03 (−0.13 to 0.08) |
| - 0.5 to <2 marijuana-years | −0.17 (−0.28 to −0.06) | −0.07 (−0.18 to 0.04) | −0.07 (−0.21 to 0.06) | −0.08 (−0.22 to 0.06) |
| - 2 to <5 marijuana-years | −0.33 (−0.51 to −0.15) | −0.11 (−0.28 to 0.06) | −0.09 (−0.28 to 0.09) | −0.08 (−0.27 to 0.11) |
| >5 marijuana-years | −0.52 (−0.75 to −0.29) | −0.27 (−0.49 to −0.05) | −0.31 (−0.54 to −0.07) | −0.25 (−0.50 to −0.01) |
| p-value for trend | <0.001 | 0.007 | 0.01 | 0.04 |
| For every 5 marijuana-years | −0.34 (−0.45 to −0.24) | −0.15 (−0.24 to −0.05) | −0.15 (−0.25 to −0.04) | −0.13 (−0.24 to −0.02) |
| p-value | <0.001 | 0.002 | 0.005 | 0.02 |
|
Digit symbol substitution test (DSST) |
||||
| - Never used marijuana | 0 (Ref.) | 0 (Ref.) | 0 (Ref.) | 0 (Ref.) |
| - 1 day to <0.5 marijuana-years | 0.17 (0.07 to 0.27) | 0.03 (−0.06 to 0.12) | 0.06 (−0.04 to 0.16) | 0.06 (−0.04 to 0.16) |
| - 0.5 to <2 marijuana-years | −0.08 (−0.19 to 0.03) | −0.03 (−0.13 to 0.07) | 0.07 (−0.06 to 0.19) | 0.05 (−0.08 to 0.18) |
| - 2 to <5 marijuana-years | −0.33 (−0.51 to −0.16) | −0.12 (−0.28 to 0.04) | −0.03 (−0.21 to 0.15) | −0.02 (−0.20 to 0.17) |
| >5 marijuana-years | −0.25 (−0.48 to −0.02) | −0.04 (−0.24 to 0.15) | 0.12 (−0.08 to 0.33) | 0.13 (−0.09 to 0.34) |
| p-value for trend | <0.001 | 0.26 | 0.5 | 0.5 |
| For every 5 marijuana-years | −0.31 (−0.41 to −0.20) | −0.08 (−0.17 to 0.01) | −0.01 (−0.10 to 0.08) | −0.03 (−0.12 to 0.07) |
| p-value | <0.001 | 0.08 | 0.8 | 0.6 |
| Stroop interference test d | ||||
| - Never used marijuana | 0 (Ref.) | 0 (Ref.) | 0 (Ref.) | 0 (Ref.) |
| - 1 day to <0.5 marijuana-years | 0.12 (0.02 to 0.22) | 0.06 (−0.05 to 0.17) | 0.05 (−0.06 to 0.17) | 0.05 (−0.07 to 0.17) |
| - 0.5 to <2 marijuana-years | 0.09 (−0.02 to 0.20) | 0.10 (−0.02 to 0.23) | 0.13 (−0.04 to 0.29) | 0.11 (−0.06 to 0.27) |
| - 2 to <5 marijuana-years | −0.03 (−0.21 to 0.15) | 0.10 (−0.09 to 0.29) | 0.08 (−0.13 to 0.29) | 0.10 (−0.11 to 0.31) |
| >5 marijuana-years | −0.12 (−0.36 to 0.11) | −0.08 (−0.32 to 0.17) | −0.02 (−0.30 to 0.24) | −0.09 (−0.37 to 0.20) |
| p-value for trend | 0.12 | 0.7 | 0.9 | 0.7 |
| For every 5 marijuana-years | −0.09 (−0.20 to 0.01) | −0.02 (−0.12 to 0.09) | −0.01 (−0.13 to 0.10) | −0.04 (−0.16 to 0.08) |
| p-value | 0.08 | 0.8 | 0.8 | 0.5 |
Abbreviations: CF: Cognitive function; IPCW: Inverse probability of censoring weighting; Ref.: Reference
Cumulative exposure to marijuana expressed in ‘marijuana-years’, with 1 marijuana-year of exposure equivalent to 365 days of marijuana use (see Methods). Current marijuana users within the 30 days prior of the Year 25 visit excluded (N=392).
Years of marijuana exposure was modeled first as a 5-level categorical predictor, and then as a continuous linear predictor, per 5 marijuana-years (separate models).
Linear regression models used to determine the association between CF scores and cumulative exposure to marijuana use. Negative standardized scores indicate worse CF.
The inverse of the Stroop score used in the present analyses to allow interpretation of worse CF with negative standardized scores for all three CF tests.
Adjusted for age, race/ethnicity, sex, study site and years of education. Analyses weighted by the inverse probability of censoring (IPCW) to address potential bias by informative censoring (eMethods).
Model described in f additionally adjusted for cumulative and current exposure to licit and illicit substances and other covariates (see Methods).
Model described in g additionally adjusted for total number of stars completed and errors made drawing the stars. Participants with missing data on mirror star tracing excluded (N=280).