Characteristics of repeat and first-occurrence adverse drug events (n = 1296)
| Characteristic | Repeat adverse drug events n = 421 | First-occurrence adverse drug events n = 875 | ||
|---|---|---|---|---|
| No. | % (95% CI*) | No. | % (95% CI*) | |
| Adverse drug event type | ||||
| Adverse drug reaction | 132 | 31.4 (26.8–35.9) | 330 | 37.7 (34.4–41.0) |
| Nonadherence | 85 | 20.2 (16.3–24.0) | 163 | 18.6 (15.9–21.3) |
| Needs additional drug/untreated indication | 63 | 15.0 (11.6–18.4) | 88 | 10.1 (8.0–12.1) |
| Low dose | 57 | 13.5 (10.2–16.8) | 78 | 8.9 (7.0–10.8) |
| High dose | 54 | 12.8 (9.5–16.1) | 100 | 11.4 (9.2–13.6) |
| Ineffective drug | 22 | 5.2 (3.1–7.4) | 57 | 6.5 (4.9–8.2) |
| Drug interaction | 7 | 1.7 (0.4–2.9) | 48 | 5.5 (3.9–7.0) |
| Drug withdrawal | 1 | 0.2 (0.0–0.7) | 6 | 0.7 (0.1–1.2) |
| Transcribing/dispensing/administration error | 0 | – | 3 | 0.3 (0.0–0.7) |
| Other | 0 | – | 2 | 0.2 (0.0–0.5) |
| Adverse drug event severity | ||||
| Mild | 20 | 4.8 (2.7–6.8) | 19 | 2.2 (1.2–3.2) |
| Moderate | 280 | 66.5 (62.0–71.0) | 566 | 64.7 (61.5–67.9) |
| Severe | 121 | 28.7 (24.4–33.1) | 290 | 33.1 (30.0–36.3) |
| Fatal | 0 | – | 0 | – |
| Intervention(s) required | ||||
| Repeat clinical assessment | 267 | 63.4 (58.7–68.2) | 533 | 60.9 (57.7–64.1) |
| Add a medication | 214 | 50.8 (46.0–55.6) | 457 | 52.2 (48.8–55.7) |
| Follow-up laboratory tests | 149 | 35.4 (30.8–40.0) | 296 | 33.8 (30.6–37.1) |
| Hospital admission | 129 | 30.6 (26.2–35.1) | 302 | 34.5 (31.3–37.7) |
| Stop in medication | 107 | 25.4 (21.1–29.7) | 307 | 35.1 (31.8–38.3) |
| Change in medication dose | 98 | 23.3 (19.1–27.5) | 166 | 19.0 (16.3–21.7) |
| Vital sign monitoring | 96 | 22.8 (18.6–27.0) | 226 | 25.8 (22.8–28.8) |
| Other | 24 | 5.7 (3.5–7.9) | 34 | 3.9 (2.6–5.2) |
| Outcomes | ||||
| No harm | 71 | 16.9 (13.2–20.5) | 135 | 15.4 (13.1–17.8) |
| Temporary harm | 343 | 81.5 (77.7–85.2) | 714 | 81.6 (79.0–84.2) |
| Permanent harm/death | 7 | 1.7 (0.4–2.9) | 26 | 3.0 (1.8–4.1) |
Note: CI = confidence interval.
↵* 95% confidence intervals were adjusted for clustering of adverse drug event characteristics in patients with multiple events.