Confidence intervals for adjusted incidence estimates were estimated using the 95% confidence intervals of the proportion of RSV positive tests assuming a binomial distribution.
Descriptive statistics and bivariate prevalence estimates were estimated using the design-based F test.
Standard errors of lifetime risk estimates were estimated using the jack-knife repeated replication method implemented in a SAS macro [ 17]. *Age of Onset.
Uncertainty ranges for HALE estimates were estimated using Monte Carlo simulation techniques to quantify the uncertainty in life expectancy projections, in GBD estimates for prevalence and disability severity, and in the survey-based prevalence estimates [ 23].
Meta-analytic estimates were estimated using fixed-effects models in the absence of heterogeneity between cohorts (p-value of heterogeneity > 0.1), and using random-effects models when heterogeneity between cohorts was present.
Heterogeneity of the cohort-specific estimates was estimated using Cochrane's Q test.
Standard errors of sensitivity, false discovery rate, and genomic distance estimates are estimated using 1,000 bootstrap samples.
The extent of optimism in the r and AUC estimates was estimated using bootstrap resampling (n = 1000 bootstrap samples).
Standard errors for parameter estimates in equation (1) were estimated using the Huber-White sandwich estimator.
Models were estimated using generalized estimating equations with a first-order autoregressive correlation structure (Hilbe [2011]).
