Skip to main content
. Author manuscript; available in PMC: 2022 Oct 1.
Published in final edited form as: Pharmacoepidemiol Drug Saf. 2021 Jun 14;30(10):1420–1427. doi: 10.1002/pds.5303

Table 3:

Adjusted sensitivity and secondary analyses 1

No. of
patients 2
Metoprolol
Events (IR)
Carvedilol
Events (IR)
HR (95% CI)

Primary Analysis
24,239 8,150 (33.5) 8,576 (33.4) 0.97 (0.94, 1.01)
Subgroup analysis 3
 Male 9,408 3,279 (37.2) 3,472 (36.1) 0.98 (0.94, 1.03)
 Female 14,739 4,965 (32.1) 5,077 (31.8) 0.99 (0.95, 1.03)
 Age > 80 years 8,901 2,945 (36.6) 3,051 (37.9) 0.96 (0.91, 1.01)
 Digoxin or spironolactone use 4,286 1,426 (33.5) 1,415 (31.8) 1.03 (0.96, 1.11)
 Recent hospitalization (30 days) 14,436 4,654 (32.1) 5,022 (32.8) 0.95 (0.91, 0.99)
Sensitivity analysis
 Intention to treat analysis 24,239 9,683 (26.9) 9,716 (28.1) 0.98 (0.95, 1.01)
 Broad HF definition 3 56,060 19,878 (31.6) 20,920 (31.7) 0.97 (0.95, 0.99)
Secondary Outcomes
 Insulin initiation 3,4 11,840 2,676 (16.3) 2,845 (16.3) 0.98 (0.93, 1.03)
 Second-line therapy initiation 3,5 3,754 1,549 (41.2) 1,641 (44.4) 0.94 (0.87, 1.00)
 Hyperglycemic hospitalization 24,239 3,396 (9.8) 3,658 (9.8) 0.99 (0.94, 1.03)
 Emergency room visit 24,239 868 (2.3) 962 (2.3) 0.97 (0.88, 1.06)
 Severe hyperglycemic event 6 24,239 3,822 (11.2) 4,151 (11.3) 0.98 (0.93, 1.02)

Abbreviations: CI: Confidence Interval; IR: Incidence Rates per 100 person-years of follow up; HF Heart Failure

1

Analyses were adjusted using propensity score matching. See text for details.

2

In each group (represents the number of matched pairs).

3

Propensity score was re-estimated and patients were re-matched

4

Comprised of insulin naïve patients

5

Initiation of second-line antidiabetic therapy was assessed in patients who were on metformin monotherapy at baseline.

6

Composite of inpatient or emergency department visits relating to hyperglycemia; see Appendix Table 1 for codes.