Skip to main content
. 2020 Aug;110(8):1214–1220. doi: 10.2105/AJPH.2020.305666

TABLE 2—

Difference-in-Differences Estimates of the Association of DelCAN With the Primary Contraceptive Method Type Among Adult Family Planning Clients at Risk for Unintended Pregnancy: Delaware, 2008 to 2017

% Using Method in Delaware During the Preimplementation Period DD Estimates (SE) P
Contraceptive method categories
 Permanent methods 5.3 −1.7 (0.90) .1
 LARCs 8.0 3.2 (1.58) .012
 Moderate methods 44.1 1.9 (3.35) .34
 Less effective methods 25.9 −1.2 (2.99) .46
 No methods 16.7 −2.2 (3.10) .22

Note. DD = difference-in-differences; DelCAN = Delaware Contraceptive Access Now; LARCs = long-acting reversible contraceptives. Each estimate comes from a separate regression. All estimates are weighted by the number of female family planning clients by age group, state, and year. Covariates are described in the text. SEs account for imputation variance and state clustering. P values were obtained from a bootstrap that accounted for state clustering and a single treated cluster. The unweighted sample size was 1470 age group by state by year cells. On average, there were 13 265 adult female clients per year in Delaware and 3.1 million per year in the comparison states from 2008 to 2017. The comparison states were the District of Columbia and all states except Colorado and Texas.

Source. Restricted use versions of the 2008–2017 Title X Family Planning Annual Report; 2008–2017 American Community Survey; Kaiser Family Foundation26; and the Guttmacher Institute.