Formal analyses of the effect of major policy changes were conducted using mixed-effects linear models with a random intercept; counties were treated as the clustering variable. The outcome was analyzed was the proportion of eligible adults registered with one of four organ donor organizations. Four successive models were created: a model only analyzing a linear effect of time; a model including a term for a spline corresponding with a 2012 policy requiring that DMV drivers license application force people to decide whether or not they will join the registry; a model including a term for a spline corresponding with a 2016 policy that added organ donor registry to the New York state health insurance marketplace application; a model including a term for a spline corresponding with a 2017 policy that lowered the age from eligible enrollees from 18 to 16 years old; and a model that included all splines. All models were adjusted for the specific organization used in each county; models were compared using likelihood ratio tests.

The above plot shows the increasing trend of organ donor enrollment over time. The dashed lines represent the beginning of the 3 policies we analyzed in this project.

Term	\(\beta\)	Standard error	Lower bound	Upper bound
Time (year)	3.088	0.022	3.045	3.131
2012 spline	-0.982	0.041	-1.063	-0.902
2016 spline	0.333	0.150	0.039	0.627
2017 spline	2.465	0.213	2.048	2.883
Note:
Splines correspond with dates in which 3 major policy changes were inacted. Estimates are adjusted for organ donor organization

The final model analyzed accounted for all three separate policy changes. Main effects were found for all three splines. This model found that after the 2012 policy was enacted, the estimated rate of increase in population enrollment decreased by .98 percentage points (95% CI: -1.06, -0.90). However, after the 2016 policy went into effect the rate of enrollment increased again by 0.33 percentage points (95% CI: 0.039, 0.627). Lastly, after the 2017 policy went into effect the rate of enrollment increased again by 2.47 percentage points (95% CI: 2.05, 2.88). Ultimately, after all three policies came into effect the estimated increase in organ donor enrollment is 4.9 percentage points per year (an estimated net effect of 1.82 percentage points).

 

---

A P8105 Final Project by Nick , Jack , SiJia , Alina and Shu