Table 3.
Selected baseline characteristics of cohorts to assess hospitalization for acute liver injury, full sample before propensity score trimming
| CPRD | HIRD | Medicare | ||||
|---|---|---|---|---|---|---|
| Dapagliflozin (n = 10,466) | Comparator GLD (n = 39,173) | Dapagliflozin (n = 17,187) | Comparator GLD (n = 195,393) | Dapagliflozin (n = 13,280) | Comparator GLD (n = 199,193) | |
| Age, mean (SD),a years | 57.3 (10.6) | 59.2 (11.1) | 51.7 (8.6) | 51.7 (8.7) | 69.9 (4.5) | 69.9 (4.6) |
| Female sex, n (%) | 4224 (40.4) | 15,862 (40.5) | 7441 (43.3) | 86,669 (44.4) | 6217 (46.8) | 93,255 (46.8) |
| Race/ethnicity,b n (%) | ||||||
| Asian | NA | NA | NA | NA | 552 (4.2) | 7976 (4.0) |
| Black | NA | NA | NA | NA | 914 (6.9) | 18,465 (9.3) |
| Hispanic | NA | NA | NA | NA | 519 (3.9) | 7613 (3.8) |
| White | NA | NA | NA | NA | 10,595 (79.8) | 153,313 (77.0) |
| Otherc | NA | NA | NA | NA | 326 (2.5) | 6292 (3.2) |
| Unknown | NA | NA | NA | NA | 374 (2.8) | 5534 (2.8) |
| Insulin use at the index date, n (%) | 1402 (13.4) | 2,210 (5.6) | 2,424 (14.1) | 21,516 (11.0) | 2399 (18.1) | 26,360 (13.2) |
| 1 or more drugs with a known association with liver injury,d n (%) | 9607 (91.8) | 35,152 (89.7) | 15,039 (87.5) | 169,028 (86.5) | 10,820 (81.5) | 166,556 (83.6) |
| Indicators of diabetes severity, n (%) | ||||||
| Diabetic nephropathy or renal insufficiency | 113 (1.1) | 380 (1.0) | 306 (1.8) | 4,208 (2.2) | 910 (6.9) | 17,651 (8.9) |
| Retinopathy | 3080 (29.4) | 9753 (24.9) | 4056 (23.6) | 41,067 (21.0) | 4665 (35.1) | 60,494 (30.4) |
| Peripheral neuropathy | 336 (3.2) | 1056 (2.7) | 297 (1.7) | 3007 (1.5) | 557 (4.2) | 8551 (4.3) |
| Peripheral vascular diseasee | 333 (3.2) | 1468 (3.7) | 3771 (21.9) | 39,351 (20.1) | 4289 (32.3) | 58,901 (29.6) |
| Coronary heart disease | 1194 (11.4) | 4989 (12.7) | 1249 (7.3) | 14,712 (7.5) | 3003 (22.6) | 44,048 (22.1) |
| Cerebrovascular disease | 452 (4.3) | 2316 (5.9) | 228 (1.3) | 3509 (1.8) | 1187 (8.9) | 18,769 (9.4) |
| Amputation | 79 (0.8) | 367 (0.9) | 48 (0.3) | 844 (0.4) | 54 (0.4) | 1453 (0.7) |
| Body mass index (kg/m2),f n (%) | ||||||
| < 20 (underweight) | 21 (0.2) | 197 (0.5) | NA | NA | NA | NA |
| 20 to < 25 (normal) | 357 (3.4) | 3,182 (8.1) | NA | NA | NA | NA |
| 25 to < 30 (overweight) | 2138 (20.4) | 10,484 (26.8) | NA | NA | NA | NA |
| 30 to < 40 (obese) | 5747 (54.9) | 18,473 (47.2) | NA | NA | NA | NA |
| ≥ 40 (severely obese) | 2031 (19.4) | 5418 (13.8) | NA | NA | NA | NA |
| Unknown | 172 (1.6) | 1419 (3.6) | NA | NA | NA | NA |
| Healthcare utilization in the 180 days before the index date | ||||||
| No. of outpatient visits,g n (%) | ||||||
| 0 | 487 (4.7) | 1629 (4.2) | 389 (2.3) | 8797 (4.5) | 767 (5.8) | 14,155 (7.1) |
| 1 | 905 (8.6) | 3267 (8.3) | 693 (4.0) | 9477 (4.9) | 919 (6.9) | 16,082 (8.1) |
| 2 or more | 9074 (86.7) | 34,277 (87.5) | 16,105 (93.7) | 177,119 (90.6) | 11,594 (87.3) | 168,956 (84.8) |
| No. of hospitalizations, n (%) | ||||||
| 0 | 9586 (91.6) | 34,656 (88.5) | 16,748 (97.5) | 187,475 (95.9) | 12,817 (96.5) | 187,364 (94.1) |
| 1 | 658 (6.3) | 2888 (7.4) | 397 (2.3) | 7174 (3.7) | 370 (2.8) | 8452 (4.2) |
| 2 or more | 222 (2.1) | 1629 (4.1) | 42 (0.2) | 744 (0.4) | 93 (0.7) | 3377 (1.7) |
| No. of GLD classes h used within 12 monthsi before the index date, n (%) | ||||||
| 0 | 114 (1.1) | 1705 (4.4) | 2392 (13.9) | 37,568 (19.2) | 782 (5.9) | 16,973 (8.5) |
| 1–2 | 6326 (60.4) | 32,478 (82.9) | 11,320 (65.9) | 140,761 (72.0) | 7616 (57.3) | 147,140 (73.9) |
| 3–4 | 3976 (38.0) | 4936 (12.6) | 3425 (19.9) | 16,951 (8.7) | 4721 (35.5) | 34,271 (17.2) |
| 5–8 | 50 (0.5) | 54 (0.1) | 50 (0.3) | 113 (0.1) | 161 (1.2) | 809 (0.4) |
| Type of index therapy, j n (%) | ||||||
| Index monotherapy with no prior treatment | 222 (2.1) | 1102 (2.8) | 1380 (8.0) | 18,408 (9.4) | 1227 (9.2) | 14,139 (7.1) |
| Combined index therapy with no prior treatment | 150 (1.4) | 1831 (4.7) | 1283 (7.5) | 22,039 (11.3) | 650 (4.9) | 20,208 (10.1) |
| Add-on index therapy | 6185 (59.1) | 26,106 (66.6) | 11,612 (67.6) | 119,280 (61.0) | 7519 (56.6) | 100,722 (50.6) |
| Switched-to index therapy | 388 (3.7) | 2972 (7.6) | 357 (2.1) | 3536 (1.8) | 1047 (7.9) | 25,043 (12.6) |
| Add-on and switched-to index therapy | 3180 (30.4) | 5727 (14.6) | 1390 (8.1) | 10,678 (5.5) | 2114 (15.9) | 29,115 (14.6) |
| Non-evaluablek | 341 (3.3) | 1435 (3.7) | 1165 (6.8) | 21,452 (11.0) | 723 (5.4) | 9,966 (5.0) |
CPRD Clinical Practice Research Datalink, GLD glucose-lowering drug, HIRD HealthCore Integrated Research Database, NA not applicable, SD standard deviation
aPatients were aged 18 years or older in CPRD, 18–64 years in the HIRD, and 65 years or older in Medicare
bData on race/ethnicity were available only in Medicare
cIncludes patients categorized as Other or as North American Native in Medicare
dDrugs with a known association with liver injury are listed in Table S2 of the electronic supplementary material
eIncludes peripheral artery disease
fData on body mass index were available only in CPRD
gOutpatient visits included general practitioner and outpatient hospital visits
hGlucose-lowering drug classes that were considered were insulin, sulfonylureas, thiazolidinediones, dipeptidyl peptidase-4 (DPP-4) inhibitors, glucagon-like peptide-1 (GLP-1) receptor agonists, biguanides (metformin), alpha glucosidase inhibitors, and meglitinides
iThose with at least 180 days of available lookback data before the index date were eligible for inclusion in the study, and therefore some patients had less than 12 months of available lookback data
jDetailed definitions for the index therapy type categories are provided in Online Resource 1
kPatients who did not have sufficient follow-up time to assess the 90-day add-on/switch requirement