Table 2 Association between timing of nirmatrelvir/ritonavir initiation and 28-day all-cause mortality or hospitalization

From: Optimal timing of nirmatrelvir/ritonavir treatment after COVID-19 symptom onset or diagnosis: target trial emulation

All-cause mortality or hospitalization

Cumulative incidence

Absolute risk reductiona

95% CI

Relative riska

95% CI

Early

Late

    

N

RiskΔ

N

RiskΔ

    

Primary analysis

Target trial emulation with IPW

43,625

5.09%

43,445

6.59%

1.50%

(1.17%, 1.80%)

0.77

(0.73, 0.82)

Sensitivity analyses

Extended the follow-up to 42 days

43,625

5.20%

43,445

6.70%

1.50%

(1.14%, 1.81%)

0.78

(0.74, 0.82)

Days 0–2 vs days >2

64,690

5.24%

22,380

7.50%

2.25%

(1.40%, 3.02%)

0.70

(0.63, 0.79)

Days 0–3 vs days >3

76,935

5.54%

10,135

8.36%

2.82%

(0.65%, 5.40%)

0.66

(0.50, 0.90)

Exclude those who initiated nirmatrelvir/ritonavir beyond 5 days after diagnosis or symptom onset

43,625

4.71%

41,686

6.81%

2.10%

(1.77%, 2.43%)

0.69

(0.66, 0.73)

Subgroup analyses

Male

19,218

5.75%

18,486

7.55%

1.80%

(1.32%, 2.21%)

0.76

(0.72, 0.82)

Female

24,407

4.61%

24,959

5.82%

1.21%

(0.78%, 1.55%)

0.79

(0.74, 0.86)

March–June 2022‡

2900

13.13%

2,753

11.63%

−1.50%

(−2.83%, 0.48%)

1.13

(0.96, 1.25)

July–October 2022‡

16,639

6.19%

17,119

8.45%

2.26%

(1.78%, 2.73%)

0.73

(0.69, 0.78)

November 2022–January 2023‡

24,086

3.52%

23,573

4.89%

1.37%

(0.82%, 1.82%)

0.72

(0.65, 0.81)

Fully vaccinated or boosted

38,552

4.02%

37,784

5.55%

1.53%

(1.23%, 1.91%)

0.72

(0.67, 0.77)

Not fully vaccinated

5073

12.20%

5661

13.67%

1.47%

(0.53%, 2.39%)

0.89

(0.83, 0.96)

Charlson’s index 0–6

43,405

4.62%

43,199

6.04%

1.42%

(1.08%, 1.76%)

0.77

(0.72, 0.82)

Charlson’s index >6

220

97.51%

246

97.97%

0.46%

(−2.93%, 2.81%)

1.00

(0.97, 1.03)

Concomitant corticosteroid use

425

95.51%

499

98.48%

2.97%

(1.07%, 4.92%)

0.97

(0.95, 0.99)

No concomitant corticosteroid use

43,200

4.11%

42,946

5.50%

1.38%

(1.09%, 1.65%)

0.75

(0.71, 0.79)

Immunocompromised

603

74.74%

720

83.84%

9.10%

(4.55%, 14.89%)

0.89

(0.82, 0.95)

Not immunocompromised

43,022

3.98%

42,725

5.20%

1.22%

(0.91%, 1.57%)

0.77

(0.71, 0.82)

With at least one Ct value measurement

2398

70.20%

2871

85.36%

15.16%

(13.07%, 17.22%)

0.82

(0.80, 0.85)

Without any Ct value measurements

41,227

0.64%

40,574

0.72%

0.09%

(−0.08%, 0.25%)

0.88

(0.72, 1.14)

Documented symptom onset date as index date

29,159

4.70%

39,615

6.11%

1.42%

(1.12%, 1.71%)

0.77

(0.72, 0.81)

Date of COVID-19 diagnosis as index date

14,534

4.15%

3870

5.45%

1.31%

(0.76%, 1.87%)

0.76

(0.69, 0.85)

  1. IPW inverse probability weighting, CI confidence interval.
  2. ΔRisk represents the incidence of 28-day all-cause mortality or hospitalization after inverse probability weighting (IPW) between the two treatment groups against baseline covariates.
  3. aAbsolute risk reduction >0 (or <0) and relative risk <1 (or >1) indicate early initiators (day 0–1) had a lower (higher) risk of the designated outcome compared to late initiators (days ≥2).
  4. ‡Omicron subvariant BA.2 was dominant from March to June 2022, BA.2 and BA.4/BA.5 from July to October 2022 and BA.5, BA.2.75, and BQ.1 from November 2022 to January 2023.
Back to article page