Table 1 Summary of key parameter estimates from fitting within-host and between-host models to data.

From: Modeling mitigation of influenza epidemics by baloxavir

Parameter

Median

95% CI lower

95% CI upper

Antiviral efficacy \(\epsilon\) for baloxavir

0.9997

0.9996

0.9999

Antiviral efficacy \(\epsilon\) for oseltamivir

0.89

0.88

0.90

Initial sensitive viral load V0 (TCID50/ml)

258.2

3.3a

2268.9a

Basic reproduction number R0 in 2016–2017 season

1.09

1.06

1.11

Basic reproduction number R0 in 2017–2018 season

1.15

1.12

1.17

Basic reproduction number R0 in 2018–2019 season

1.10

1.08

1.13

Baseline distribution of treatment initiation time, G0–48 (hours after symptom onset, truncated at 48 h)

G(4.0, 6.3)

Accelerated distribution of treatment initiation times, G0–24 (hours after symptom onset, truncated at 24 h)

G(4.0, 6.3)/2

Delayed distribution of treatment initiation times, G24–48 (hours after symptom onset, compressed to 24–48 h window)

G(4.0, 6.3)/2 + 24

Distribution of time lag between infection and symptom onset, L (hours)

24aL(0.37,0.41)

  1. aFor estimates derived by simulated annealing, we provide 95 percentile range rather than confidence intervals.
  2. Viral replication and antiviral efficacy are estimated via simulated annealing46 and approximate Bayesian computation38,39,47 fitting of deterministic within-host model to clinical trial data;8 season-specific transmission rates are estimated via approximate Bayesian computation38,39 fitting of stochastic population-level influenza transmission model (Supplementary Section 1) to US seasonal influenza incidence data12. Parameters for distributions of time between infection and symptom onset (lognormal) and from symptom onset to treatment (gamma) were estimated by the interior-point algorithm fitting of clinical trial data8. The key parameter estimates of within-host model and between-host model are summarized here, whereas others are in Supplementary Tables 2 and 3.
Back to article page