Skip to main content
. 2021 Jun 14;2(6):e210771. doi: 10.1001/jamahealthforum.2021.0771

Table 1. Sample Characteristics, Optimal Bandwidth, and Total Sample, 2014 and 2015a.

Characteristic No. (%)
Optimal bandwidthb Total sample
2014 2015 2014 2015
FPL, % 120-156 75-400
Open enrollment 32 091 55 481 142 176 213 561
Age, y
19-25 4242 (13.22) 7727 (13.93) 18 075 (12.71) 28 637 (13.41)
26-34 8936 (27.85) 15 832 (28.54) 38 273 (26.92) 59 199 (27.72)
35-44 7568 (23.58) 13 506 (24.34) 33 099 (23.28) 51 337 (24.04)
45-54 5889 (18.35) 9848 (17.75) 25 723 (18.09) 37 842 (17.72)
55-64 5456 (17.00) 8568 (15.44) 27 006 (18.99) 36 546 (17.11)
Sex
Female 19 017 (59.26) 32 843 (59.20) 86 124 (60.58) 128 721 (60.27)
Male 13 063 (40.71) 22 612 (40.76) 55 999 (39.39) 84 767 (39.69)
Chronic condition
Any 8131 (25.34) 13 257 (23.89) 38 090 (26.79) 52 408 (24.54)
None 23 960 (74.66) 42 224 (76.11) 104 086 (73.21) 161 153 (75.46)
Residence
Urban 22 578 (70.36) 40 544 (73.08) 98 594 (69.35) 154 929 (72.55)
Rural 3295 (10.27) 4941 (8.91) 14 894 (10.48) 19 924 (9.33)

Abbreviation: FPL, federal poverty level.

a

Source: Analysis of Colorado all-payer claims database from 2014-2015.

b

The optimal bandwidth, here defined as 120% to 156% FPL, was set using the STATA SE 14 MSE−optimal bandwidth selector for the RD treatment effect estimator e(h_mserd), defined by the command rdbwselect using the 2014 enrollment outcome.