Table 3.
Associations between geographic access to MMJ dispensaries in Zip Code Tabulation Areas (ZCTA)1 and proportion of certifications for insufficient evidence conditions in Pennsylvania2
| 2019 | 2020 | 2021 | |
|---|---|---|---|
| Incident rate ratio (95% CI) | |||
| Model 1: unadjusted associations. Distance and density modeled separately3 | |||
| Distance to nearest dispensary (miles) | |||
| ≤5 | 0.67 (0.61, 0.73) | 0.62 (0.56, 0.67) | 0.62 (0.57, 0.67) |
| >5 to ≤10 | 0.73 (0.66, 0.80) | 0.67 (0.60, 0.74) | 0.70 (0.63, 0.77) |
| >10 (reference) | |||
| Density w/in 15 min | |||
| 0 (reference) | |||
| 1 | 0.83 (0.75, 0.92) | 0.76 (0.68, 0.85) | 0.78 (0.70, 0.87) |
| ≥2 | 0.67 (0.61, 0.72) | 0.68 (0.62, 0.74) | 0.66 (0.61, 0.72) |
| Model 2: adjusted for median income (quartiled). Distance and density modeled separately3 | |||
| Distance to nearest dispensary | |||
| ≤5 | 0.72 (0.66, 0.79) | 0.71 (0.64, 0.78) | 0.71 (0.65, 0.77) |
| >5 to ≤10 | 0.86 (0.78, 0.94) | 0.83 (0.75, 0.91) | 0.83 (0.75, 0.91) |
| >10 (reference) | |||
| Density w/in 15 min | |||
| 0 (reference) | |||
| 1 | 0.88 (0.80, 1.12) | 0.80 (0.73, 0.88) | 0.82 (0.75, 0.91) |
| ≥2 | 0.69 (0.63, 0.75) | 0.74 (0.68, 0.81) | 0.72 (0.68, 0.79) |
| Model 3: adjusted for race and ethnicity (quartiled). Distance and density modeled separately3,4 | |||
| Distance to nearest dispensary | |||
| ≤5 | 0.87 (0.78, 0.97) | 0.83 (0.75, 0.93) | 0.78 (0.70, 0.87) |
| >5 to ≤10 | 0.82 (0.75, 0.90) | 0.76 (0.68, 0.84) | 0.75 (0.68, 0.83) |
| >10 (reference) | |||
| Density w/in 15 min | |||
| 0 (reference) | |||
| 1 | 0.96 (0.86, 1.06) | 0.91 (0.83, 1.01) | 0.86 (0.77, 0.96) |
| ≥2 | 0.86 (0.78, 0.95) | 0.90 (0.81, 0.99) | 0.82 (0.75, 0.90) |
11608 ZCTAs with at least 100 adult residents in every year from 2018 to 2021.
2Qualifying conditions that the National Academies of Sciences, Engineering, and Medicine indicates that there is no or insufficient evidence to support or refute treatment effectiveness of MMJ: amyotrophic lateral sclerosis, epilepsy, glaucoma, Huntington’s disease, opioid use disorder, and Parkinson’s disease.
3Negative binomial models separate for each outcome and year that used count variables (i.e., number of certifications) as the outcomes and log-transformed denominators (i.e., total number of certifications) as offset terms.
4Adjusted for proportion non-White and proportion Hispanic.