Time Series Models for Disease Surveillance

Fits time trend models for routine disease surveillance tasks and returns probability distributions for a variety of quantities of interest, including age-standardized rates, period and cumulative percent change, and measures of health inequality. The models are appropriate for count data such as disease incidence and mortality data, employing a Poisson or binomial likelihood and the first-difference (random-walk) prior for unknown risk. Optionally add a covariance matrix for multiple, correlated time series models. Inference is completed using Markov chain Monte Carlo via the Stan modeling language. References: Donegan, Hughes, and Lee (2022) ; Stan Development Team (2021) < https://mc-stan.org>; Theil (1972, ISBN:0-444-10378-3).

surveil: Public health surveillance

The surveil R package provides time series models for routine public health surveillance tasks: model time trends in mortality or disease incidence rates to make inferences about levels of risk, cumulative and period percent change, age-standardized rates, and health inequalities.

surveil is an interface to Stan, a state-of-the-art platform for Bayesian inference. For analysis of spatial health data see the geostan R package.

Installation

surveil is available on CRAN; install from R using:

install.packages("surveil")

Vignettes

Review the package vignettes to get started:

• vignette("demonstration"): An introduction to health modeling with surveil
• vignette("age-standardization"): How to age-standardize surveil model results and compare age-stratified populations

Also see the online documentation.

Usage

Model time series data of mortality or disease incidence by loading the surveil package into R together with disease surveillance data. Tables exported from CDC WONDER are automatically in the correct format.

library(surveil)
library(knitr)
data(cancer)

kable(head(cancer), 
      booktabs = TRUE,
      caption = "Table 1. A glimpse of cancer surveillance data")

Model trends in risk and easily view functions of risk estimates, such as cumulative percent change:

fit <- stan_rw(data = cancer,
               time = Year, 
               group = Age,
           cores = 4 # multi-core processing for speed
           )

fit_apc <- apc(fit)
plot(fit_apc, cumulative = TRUE)