Table 2.
Estimated changes in early and inadequate prenatal care before and after the policy across birthing person education and race/ethnicity
| Outcome | Effect heterogeneity | Average % change in pre vs. post policy (95% CIs) |
% Change in pre-policy trend vs. post-policy trend (95% CIs) |
||
|---|---|---|---|---|---|
| Q1 County prison admissions increased the most |
Q4 County Prison admissions decreased the most |
Q1 County prison admissions increased the most |
Q4 County Prison admissions decreased the most |
||
| First trimester prenatal care | Education | ||||
|
> High school High school < High school |
2.71 (1.62, 3.81) 3.99 (2.29, 5.72) 4.58 (0.73, 8.59) |
0.80 (0.09, 1.52) 4.12 (3.02, 5.23) 8.13 (6.35, 9.95) |
0.00% (-0.10, 0.11) 0.10% (-0.06, 0.27) 0.06% (-0.31, 0.44) |
-0.03% (-0.08, 0.02) 0.23% (0.14, 0.32) 0.22% (0.06, 0.37) |
|
| Race/ethnicity | |||||
|
Non-Hispanic White Non-Hispanic Black |
2.97 (2.01, 3.93) 6.14 (-1.97, 14.91) |
1.55 (0.83, 2.29) 4.68 (3.22, 6.16) |
0.04% (-0.05, 0.13) 0.25% (-0.53, 1.03) |
-0.01% (-0.06, 0.04) 0.23% (0.11, 0.35) |
|
| Inadequate prenatal care | Education | ||||
|
> High school High school < High school |
-6.64 (-13.04, 0.22) 9.20 (1.09, 17.96) 5.25 (-0.02, 10.80) |
-17.92 (-20.56, -15.20) -7.06 (-10.11, -3.91) -3.74 (-6.89, -0.48) |
0.06% (-0.62, 0.76) 0.19% (-0.55, 0.95) -0.15% (-0.65, 0.35) |
0.21% (-0.05, 0.48) -0.42% (-0.68, -0.15) -0.28% (-0.56, -0.00) |
|
| Race/ethnicity | |||||
|
Non-Hispanic White Non-Hispanic Black |
3.49 (-0.73, 7.88) -3.17 (-19.13, 15.93) |
-14.53 (-17.27, -11.71) -4.36 (-7.68, -0.93) |
-0.02% (-0.42, 0.38) 0.23% (-1.47, 1.95) |
0.23% (-0.03, 0.51) -0.37% (-0.62, -0.12) |
|
Average percent change in each outcome was estimated from Poisson models with robust error variance that interacted a post policy variable with birthing person education and, separately, race/ethnicity. The change in pre- versus post-policy trends in each outcome were estimated by interacting the post-policy variable with a linear monthly time trend and birthing person education and, separately, race/ethnicity. All models were stratified across quartiles of post-policy changes in county prison admissions, and adjust for age, marital status, insurance type, crime rate, and included county-level fixed effects. In Q1 counties, rates of prison admissions increased by 1.34–10.6 per 100,000, whereas in Q4 counties prison admissions decreased by 1.10 - 4.28 per 100,000