Skip to main content
. Author manuscript; available in PMC: 2023 Mar 1.
Published in final edited form as: Prev Med. 2022 Feb 9;156:106993. doi: 10.1016/j.ypmed.2022.106993

Table 2:

Impact of Cannabis Dispensaries on Cannabis-Involved Pregnancy Hospitalizations per 10K Live Births, 2011–2018.

Impact of Recreational Dispensaries on Cannabis-Involved Pregnancy Hospitalizations IRR [95% CI]
Impact of recreational dispensary count 1.02* [1.00,1.04]
Impact of recreational dispensary count on counties with low number of baseline medical dispensaries 0.97** [0.96,0.99]
Impact of recreational dispensary count for counties with high number of baseline medical dispensaries 0.98* [0.96,0.99]
Time Variables
Seasonal Quarter (Quarter 1, reference)
Quarter 2 1.00 [0.95,1.06]
 Quarter 3 1.10** [1.03,1.18]
 Quarter 4 1.13** [1.04,1.23]
Year (2011, reference)
 2012 1.00 [0.88,1.13]
 2013 1.50** [1.10,2.04]
 2014 3 29*** [1.92,5.63]
 2015 5 12*** [2.74,9.56]
 2016 5.98***
[2.95,12.10]
 2017 7 78*** [3.53,17.15]
 2018 8 44*** [3.90,18.27]
Post ICD10 switch indicator (=1 in Quarter 3 2015) 1.03 [0.87,1.22]
County unemployment rate 1.19* [1.04,1.37]
Total hospital admissions 1.00+ [1.00,1.00]
N 1888
Combined IRR for the Impact of Recreational Dispensaries on Counties with Different Baseline Medical Cannabis Dispensary Densities.
No medical dispensary exposure 1.02 (p=0.042)
Low medical dispensary exposure (1–10) 0.994 (p=0.11)
High medical dispensary exposure (>10) 0.995 (p=0.000)

Notes: IRR = incidence rate ratio; CI = confidence interval. We classified counties according to their baseline level of medical dispensary exposure (no medical dispensaries; 1–9; and 10+) in 2012.

***

Indicates significance at the 0.01% level

**

significance at the 1% level

*

significance at the 5% level, and

+

indicates significance at 10% level. All models estimated using xtpoisson in STATA 16.1 with robust standard errors, using the number of live births per county per year as the exposure. The 1,888 observations reflects the number of county-quarters included in the analysis. Colorado has 64 counties, but five counties (148 county-quarter observations) drop out of the analysis because there are no pregnancy-related claims co-coded with cannabis.