Table 1.
Characteristics of Patients in the Study Cohort a
| Characteristic | All patients (N = 27,355) | Patients with ≥1 screening potassium monitoring (N = 19,391; 71%) | Patients without screening potassium monitoring (N = 7,964; 29%) | P Value b |
|---|---|---|---|---|
| Mean age in years (SD) | 59.0 (13.3) | 60.4 (13.0) | 55.5 (13.2) | <0.001 |
| Gender: number female (%) | 13,258 (48.5) | 9,544 (49.2) | 3,714 (46.6) | <0.001 |
| Number of patients in each initial patient-drug group (%) c | <0.001 | |||
| ACEi | 25,240 (92.3) | 17,825 (91.9) | 7,415 (93.1) | |
| ARB | 1,538 (5.6) | 1,099 (5.7) | 439 (5.5) | |
| Spironolactone | 456 (1.7) | 361 (1.9) | 95 (1.2) | |
| ACEi + spironolactone | 95 (0.3) | 85 (0.4) | 10 (0.1) | |
| ACEi + ARB | 16 (0.1) | 13 (<0.1) | 3 (<0.1) | |
| ARB + spironolactone | 10 (<0.1) | 8 (<0.1) | 2 (<0.1) | |
| Duration of initial drug therapy (days) | <0.001 | |||
| Mean (SD) | 209 (143) | 238 (139) | 138 (128) | |
| Median (5th, 95th percentile) | 194 (30, 365) | 321 (30, 365) | 70 (30, 365) | |
| Duration of consecutive drug therapy (days) | <0.001 | |||
| Mean (SD) | 225 (143) | 259 (134) | 143 (130) | |
| Median (5th, 95th percentile) | 264 (30, 365) | 365 (30, 365) | 79 (30, 365) | |
| Number of chronic diseases within 6 months prior to study entry | <0.001 | |||
| Mean (SD) | 6.3 (2.3) | 6.5 (2.3) | 5.9 (2.1) | |
| Median (5th, 95th percentile) | 6 (3, 10) | 6 (3, 11) | 6 (3, 9) | |
| Hospitalization(s) and/or emergency department visits, any within 6 months prior to study entry (%) | 6,067 (22.2) | 4,549 (23.5) | 1,518 (19.1) | <0.001 |
| Outpatient visits within 6 months prior to study entry | <0.001 | |||
| Mean (SD) | 4.6 (4.7) | 4.8 (4.9) | 4.2 (4.2) | |
| Median (5th, 95th percentile) | 3 (0, 14) | 3 (0, 14) | 3 (0, 13) | |
| Heart failure diagnosis (%) d | 2,009 (7.3) | 1,727 (8.9) | 282 (3.5) | <0.001 |
| Hypertension diagnosis (%) d | 18,709 (68.4) | 14,135 (72.9) | 4,574 (57.4) | <0.001 |
| Chronic kidney disease (CKD, %) d, e | 2,176 (8.0) | 1,937 (10.0) | 239 (3.0) | <0.001 |
| CKD diagnosis code (%) | 1,812 (6.6) | 1,601 (8.3) | 211 (2.7) | <0.001 |
| CKD by creatinine clearance or glomerular filtration rate estimate (%) | 364 (1.3) | 336 (1.7) | 28 (0.4) | |
| End-stage renal disease diagnosis during study period (%) | 163 (0.6) | 132 (0.7) | 31 (0.4) | 0.004 |
| Kidney transplant (%) d | ||||
| Prior to study period | 56 (0.2) | 54 (0.3) | 2 (0.1) | <0.001 |
| During study period | 3 (<0.1) | 3 (<0.1) | 0 | |
| Deaths during study period (%) | 259 (1.0) | 207 (1.1) | 52(0.7) | <0.001 |
| Concomitant therapy f | ||||
| Digoxin | 973 (3.6) | 847 (4.4) | 126 (1.6) | <0.001 |
| Diuretic | 8,788 (32.1) | 7,200 (37.1) | 1,588 (20.0) | <0.001 |
| Potassium supplement | 3,072 (11.2) | 2,695 (13.9) | 377 (4.7) | <0.001 |
| Digoxin and potassium supplement | 411 (1.5) | 377 (1.9) | 34 (0.4) | <0.001 |
| Digoxin and diuretic | 688 (2.5) | 619 (3.2) | 69 (0.9) | <0.001 |
| Diuretic and potassium supplement | 2,821 (10.3) | 2,481 (12.8) | 340 (4.3) | <0.001 |
| Nonsteroidal anti-inflammatory agents | 5,323 (19.5) | 4,076 (21.0) | 1,247 (15.7) | <0.001 |
| Trimethoprim | 694 (2.5) | 572 (2.9) | 122 (1.5) | <0.001 |
| Beta-blocker | 8,223 (30.1) | 6,668 (34.4) | 1,555 (19.5) | <0.001 |
| Patients with potassium level measured within 30 days prior to initiating ACEi/ARB/spironolactone therapy | 244 (0.9) | 170 (0.9) | 74 (0.9) | 0.68 |
| Patients with potassium level ≥6.0 mmol/l within 180 days prior to study period | 23 (0.1) | 20 (0.1) | 3 (<0.1) | 0.09 |
| Patients with a diagnosis of hyperkalemia within 180 days prior to starting ACEi/ARB/spironolactone therapy | 140 (0.5) | 110 (0.6) | 30 (0.4) | 0.05 |
aPatients with diabetes who were new users of a RAAS inhibitor [i.e., angiotensin-converting enzyme inhibitor (ACEi), angiotensin receptor blocker (ARB) or spironolactone], with one or more dispensings during the study timeframe
bChi square, Fisher’s exact or Wilcoxon rank sum tests
cFor analysis, the ACEi, ARB and ACEi + ARB groups were considered together, and the spironolactone, ACEi + spironolactone and ARB + spironolactone patient-drug groups were considered together
dDiagnosis codes were determined from up to 180 days prior to the study period through the end of the study period
eMany of the 1,812 patients in the “diagnosis code” grouping also had creatinine clearance or GFR estimates during the study period that reflected CKD stage 3 or 4. The 364 patients in the “creatinine clearance or GFR estimate” grouping had CKD stage 3 or 4 based on at least two creatinine clearance or GFR estimates ≥90 days apart during their study period, but did not have any diagnosis code for CKD during the study timeframe
fNot mutually exclusive, that is, a patient prescribed digoxin could have also been prescribed a diuretic, etc. Diuretic concomitant therapy does not include patients receiving spironolactone