Table 2. Number of Events, Crude, and Weighted Rate of Thromboembolic Events in Patients Not Receiving Anticoagulation Therapy According to Presence of New-Onset AF.
| Strata | 1-y Follow-up | 3-y Follow-up | ||||
|---|---|---|---|---|---|---|
| Events, No. | Rate, events/100 person-years (95% CI) | Events, No. | Rate, events/100 person-years (95% CI) | |||
| Crude | Weighteda | Crude | Weighteda | |||
| Pneumonia without AF | 2059 | 0.93 (0.89-0.97) | NA | 4982 | 0.83 (0.81-0.86) | NA |
| Pneumonia with new-onset AF | 137b | 2.80 (2.37-3.31) | 2.82 (2.38-3.37) | 272b | 2.21 (1.97-2.49) | 2.29 (1.99-2.65) |
| Baseline stroke riskc | ||||||
| Low | <5b | 0.30 (0.04-2.14) | NAd | <5b | 0.31 (0.10-0.97) | NAd |
| Intermediate | 38 | 1.91 (1.39-2.63) | 1.88 (1.36-2.68) | 89 | 1.71 (1.39-2.11) | 1.75 (1.41-2.20) |
| High | 99 | 3.80 (3.12-4.63) | 3.87 (3.17-4.76) | 183 | 2.92 (2.53-3.38) | 3.07 (2.56-3.70) |
Abbreviations: AF, atrial fibrillation; NA, not applicable.
Patients with new-onset AF were censored when they started anticoagulation therapy and were weighted by the inverse of the probability of being censored.
As required by Danish data protection law, percentages and counts were suppressed for observations with fewer than 5 incidents to prevent disclosure of potentially identifiable information. Accordingly, the number of events among patients at low risk is not included in the overall number of events.
Among patients with new-onset AF, baseline stroke risk is determined according to CHA2DS2-VASc score (congestive heart failure or left ventricular ejection fraction ≤40%; hypertension; age ≥75 years; diabetes; stroke, transient ischemic attack, or thromboembolism history; vascular disease; age 65-74 years; and female sex) categorized by assigned points: low risk (0, no risk factors when disregarding female sex as a lone risk factor), intermediate risk (1-2 risk factors), and high risk (≥3 risk factors) where points for female sex were not taken into account.
Because of low sample size and few events (<5 events), weighted rates were not estimated for patients at low baseline risk of stroke (CHA2DS2-VASc score of 0 to 1 when disregarding female sex as a lone risk factor).