Seasonal Plots: Align case data for seasonal analysis"
In ggsurveillance: Tools for Outbreak Investigation/Infectious Disease Surveillance

knitr::opts_chunk$set(
  collapse = TRUE,
  comment = "#>",
  warning = FALSE,
  message = FALSE,
  fig.width = 7,
  fig.height = 4
)

library(ggplot2)
library(dplyr)
library(tidyr)
library(ggsurveillance)

The seasonal plot

This vignette is still work in progress. But the examples are hopefully already helpful and inspiring.

The seasonal plot is a commonly used plot for seasonal respiratory pathogens like Influenza and RSV. For seasons covering the turn of the year the new yearly breakpoint has to defined (e.g. week 28 instead week 1 for new year). In a second step the previous seasons have to be aligned to the current season to allow the seasonal comparison. Here we show how this is automated using ggsurveillance.

influenza_germany |>
  filter(AgeGroup == "00+") |>
  align_dates_seasonal(
    dates_from = ReportingWeek,
    date_resolution = "isoweek",
    start = 28
  ) -> df_flu_aligned

ggplot(df_flu_aligned, aes(x = date_aligned, y = Incidence)) +
  stat_summary(
    aes(linetype = "Historical Median (Min-Max)"),
    data = . %>% filter(!current_season),
    fun.data = median_hilow, geom = "ribbon", alpha = 0.3
  ) +
  stat_summary(
    aes(linetype = "Historical Median (Min-Max)"),
    data = . %>% filter(!current_season),
    fun = median, geom = "line"
  ) +
  geom_line(
    aes(linetype = "2024/25"),
    data = . %>% filter(current_season), colour = "dodgerblue4", linewidth = 2
  ) +
  labs(linetype = NULL) +
  scale_x_date(date_labels = "%b'%y") +
  theme_bw() +
  theme(legend.position = c(0.2, 0.8))

Seasonal alignment and plot

library(ggplot2)

influenza_germany |>
  align_dates_seasonal(
    dates_from = ReportingWeek, date_resolution = "isoweek", start = 28
  ) -> df_flu_aligned

ggplot(df_flu_aligned, aes(x = date_aligned, y = Incidence, color = season)) +
  geom_line() +
  facet_wrap(~AgeGroup) +
  theme_bw() +
  theme_mod_rotate_x_axis_labels_45()

influenza_germany |>
  align_dates_seasonal(dates_from = ReportingWeek) |>
  group_by(AgeGroup, season) |>
  tally(wt = Cases) |>
  pivot_wider(names_from = AgeGroup, values_from = n)

Combining everything for the seasonal plot

influenza_germany |>
  filter(AgeGroup == "00+") |>
  align_dates_seasonal(
    dates_from = ReportingWeek,
    date_resolution = "isoweek",
    start = 28
  ) -> df_flu_aligned

ggplot(df_flu_aligned, aes(x = date_aligned, y = Incidence)) +
  stat_summary(
    aes(linetype = "Historical Median (Min-Max)"),
    data = . %>% filter(!current_season),
    fun.data = median_hilow, geom = "ribbon", alpha = 0.3
  ) +
  stat_summary(
    aes(linetype = "Historical Median (Min-Max)"),
    data = . %>% filter(!current_season),
    fun = median, geom = "line"
  ) +
  geom_line(
    aes(linetype = "2024/25"),
    data = . %>% filter(current_season), colour = "dodgerblue4", linewidth = 2
  ) +
  labs(linetype = "") +
  scale_x_date(date_labels = "%b'%y") +
  theme_bw() +
  theme(legend.position = c(0.2, 0.8))

Other visualisations

influenza_germany |>
  filter(AgeGroup != "00+") |>
  align_dates_seasonal(dates_from = ReportingWeek) |>
  ggplot(aes(x = ReportingWeek, weight = Cases, fill = season)) +
  geom_vline_year(color = "grey50") +
  # Use stat = count for more efficient plotting
  geom_epicurve(color = NA, stat = "count") +
  scale_y_cases_5er() +
  theme_bw()