Table 1.
Patient characteristics before and after propensity score matching
| Optum database | Before matching |
After matching |
||||
|---|---|---|---|---|---|---|
| ICM (N = 695) | Non-ICM (N = 15,099) | P value | ICM (N = 691) | Non-ICM (N = 2073) | P value | |
| Age (years), mean | 64.6 | 64.4 | .4481 | 64.6 | 65.0 | .1203 |
| Female, % | 37.2% | 33.2% | .0003 | 37.2% | 37.8% | .6544 |
| Medicare Advantage | 49.8% | 51.2% | .2098 | 49.8% | 50.8% | .3348 |
| Region, n (%) | .9001 | .9429 | ||||
| Midwest | 182 (26.3%) | 3937 (26.2%) | 181 (26.2%) | 540 (26.1%) | ||
| Northeast | 65 (9.4%) | 1322 (8.8%) | 65 (9.4%) | 194 (9.4%) | ||
| South | 287 (41.5%) | 6179 (41.1%) | 287 (41.5%) | 885 (42.7%) | ||
| West | 158 (22.8%) | 3586 (23.9%) | 158 (22.9%) | 454 (21.9%) | ||
| CHAD2S2-VASc, mean (SD)† | 2.29 (1.53) | 2.17 (1.51) | .0374 | 2.29 (1.53) | 2.29 (1.53) | 1.0000 |
| CHAD2S2-VASc, median (range) | 2 (1–3) | 2 (1–3) | 2 (1–3) | 2 (1–3) | ||
| Charlson score | 1.39 | 1.40 | .7748 | 1.39 | 1.41 | .6738 |
| Total healthcare costs in baseline (excluding ICM cost) | $29,602 | $27,492 | .0101 | $29,263 | $29,661 | .7413 |
| Patient history, % | ||||||
| Heart failure | 20.0% | 21.3% | .1624 | 20.0% | 21.4% | .2369 |
| Hypertension | 79.0% | 75.9% | .0012 | 79.0% | 79.4% | .7466 |
| Ischemic heart disease | 41.1% | 39.6% | .1842 | 41.1% | 42.4% | .3935 |
| Diabetes | 23.2% | 24.2% | .3189 | 23.2% | 24.2% | .4084 |
| Sleep apnea | 34.2% | 26.6% | <.0001 | 34.2% | 32.9% | .3148 |
| Oral anticoagulant use | 73.6% | 70.1% | .0006 | 73.7% | 73.8% | .9151 |
| Antiarrhythmic use | 74.0% | 70.9% | .0031 | 74.0% | 73.6% | .7749 |
| Year of index ablation, % | ||||||
| 2012 | 2.8% | 13.2% | <.0001 | 2.8% | 2.6% | .5863 |
| 2013 | 4.2% | 10.6% | <.0001 | 4.2% | 3.9% | .4309 |
| 2014 | 11.7% | 11.4% | .6845 | 11.7% | 12.0% | .7829 |
| 2015 | 17.7% | 13.6% | <.0001 | 17.7% | 18.6% | .4201 |
| 2016 | 24.2% | 14.4% | <.0001 | 24.2% | 22.6% | .1681 |
| 2017 | 28.4% | 19.5% | <.0001 | 28.4% | 29.1% | .5708 |
| 2018 | 8.7% | 6.3% | .0002 | 8.7% | 8.9% | .7829 |
| Medicare 5% database | ICM (N = 311) | Non-ICM (N = 5951) | P value | ICM (N = 309) | Non-ICM (N = 925) | P value‡ |
|---|---|---|---|---|---|---|
| Age (years), mean | 70.4 | 70.4 | .8396 | 70.4 | 70.7 | .2901 |
| Female, % | 45.0% | 41.8% | .0623 | 45.0% | 48.0% | .2004 |
| Medicare FFS | 100.0% | 100.0% | 1.000 | 100.0% | 100.0% | 1.000 |
| Region, n (%) | .0388 | .9176 | ||||
| Midwest | 45 (14.5%) | 1243 (20.9%) | 45 (14.6%) | 131 (14.2%) | ||
| Northeast | 46 (14.8%) | 910 (15.3%) | 46 (14.9%) | 133 (14.4%) | ||
| South | 149 (48.1%) | 2653 (44.6%) | 149 (48.2%) | 467 (50.5%) | ||
| West | 70 (22.6%) | 1143 (19.2%) | 69 (22.3%) | 194 (21.0%) | ||
| CHA2DS2-VASc, mean (SD)† | 3.17 (1.36) | 3.07 (1.39) | .2283 | 3.17 (1.36) | 3.17 (1.36) | .9967 |
| CHA2DS2-VASc, median (range) | 3 (2–4) | 3 (2–4) | 3 (2–4) | 3 (2–4) | ||
| Charlson score | 1.71 | 1.90 | .0028 | 1.71 | 1.70 | .9096 |
| Total healthcare costs in baseline (excluding ICM cost) | $14,972 | $14,637 | .6140 | $14,975 | $14,467 | .5158 |
| Patient history, % | ||||||
| Heart failure | 20.4% | 28.4% | .0000 | 20.4% | 22.0% | .4358 |
| Hypertension | 88.0% | 86.5% | .1915 | 88.0% | 90.2% | .1476 |
| Ischemic heart disease | 50.7% | 52.3% | .3418 | 50.7% | 50.3% | .6330 |
| Diabetes | 31.3% | 32.3% | .5395 | 31.4% | 36.0% | .0324∗ |
| Sleep apnea | 30.4% | 24.7% | .0005 | 30.3% | 28.0% | .3038 |
| Oral anticoagulant use | 9.7% | 5.1% | <.0001 | 9.6% | 9.5% | .5873 |
| Antiarrhythmic use | 17.0% | 8.6% | <.0001 | 16.7% | 16.5% | .755 |
| Year of index ablation, % | ||||||
| 2012 | 2.6% | 13.1% | <.0001 | 2.6% | 2.2% | .2254 |
| 2013 | 2.6% | 12.4% | <.0001 | 2.6% | 3.1% | .2755 |
| 2014 | 13.9% | 13.3% | .6129 | 14.0% | 14.4% | .9394 |
| 2015 | 22.3% | 15.8% | <.0001 | 22.3% | 22.8% | .6422 |
| 2016 | 29.5% | 15.6% | <.0001 | 29.4% | 27.1% | .2220 |
| 2017 | 25.9% | 18.4% | <.0001 | 26.0% | 27.6% | .4071 |
| 2018 | - | - | - | - | - | - |
ICM = insertable cardiac monitor.
†
Patients were matched on exact CHA2DS2-VASc score.
‡
Statistically significant difference between the 2 cohorts is indicated by an asterisk.