PEI Impfnebenwirkungen [Germany]

germany_vae.r

library(dplyr)
library(readxl)

url <- paste0(
  "https://www.pei.de/SharedDocs/Downloads/DE/newsroom/dossiers/rohdaten-",
  "sicherheitsberichte/download-xls-uaw-daten-2020-12-27-bis-2023-12-31",
  ".xlsx?__blob=publicationFile&v=5"
)
tempfile_path <- tempfile(fileext = ".xlsx")
download.file(url, tempfile_path, mode = "wb") # Use mode = "wb" for binary files

df <- read_xlsx(tempfile_path, skip = 1) |>
  setNames(c(
    "id", "year", "age_group", "sex", "vaxx_date", "manufacturer", "charge",
    "date_adverse", "adverse_event"
  )) |>
  mutate(
    year = as.integer(year),
    vaxx_date = as.Date(vaxx_date, tryFormats = c("%d.%m.%Y")),
    date_adverse = case_when(
      date_adverse == "--.--.----" ~ NA, # Handle missing dates explicitly
      TRUE ~ as.Date(date_adverse, format = "%d.%m.%Y") # Parse valid dates
    ),
    sex = case_when(
      sex == "männlich" ~ "m",
      sex == "weiblich" ~ "w",
      TRUE ~ "other"
    ),
    age_group = case_when(
      age_group == "18 - 59 Jahre" ~ "18-59",
      age_group == "60 Jahre und älter" ~ "60+",
      age_group == "leer" ~ "unknown",
      age_group == "7 - 17 Jahre" ~ "7-17",
      age_group == "k.A." ~ NA,
      age_group == "2 - 6 Jahre" ~ "2-6",
      age_group == "0 - 23 Monate" ~ "<2",
      age_group == "12 - 17 Jahre" ~ "12-17",
      age_group == "7 - 11 Jahre" ~ "7-11",
      TRUE ~ "other" # Catch any undefined cases
    ),
    manufacturer = case_when(
      grepl("Comirnaty", manufacturer, ignore.case = TRUE) ~ "Pfizer/BioNTech",
      grepl("Spikevax", manufacturer, ignore.case = TRUE) ~ "Moderna",
      grepl("Jcovden", manufacturer, ignore.case = TRUE) ~ "Johnson & Johnson",
      grepl("Vaxzevria", manufacturer, ignore.case = TRUE) ~ "AstraZeneca",
      grepl("Corona Impfstoff", manufacturer, ignore.case = TRUE) ~ NA,
      grepl("Nuvaxovid", manufacturer, ignore.case = TRUE) ~ "Novavax",
      grepl("Valneva", manufacturer, ignore.case = TRUE) ~ "Valneva",
      TRUE ~ NA
    )
  )

# write.csv(df, "/Users/ben/Downloads/germany_pei_vae.csv", row.names = FALSE)

# By Manufacturer
df |>
  group_by(manufacturer) |>
  summarize(n = n()) |>
  arrange(desc(n))

# By Cause
df |>
  filter(adverse_event %in% c(
    "Tod",
    "Ploetzlicher Tod",
    "Herztod",
    "Hirntod",
    "Ploetzlicher Herztod",
    "Fetaler Tod",
    "Tod eines fruehgeborenen Babys",
    "Fetaler Tod"
  )) |>
  group_by(adverse_event) |>
  summarize(n = n()) |>
  arrange(adverse_event, n)

# Min/Max By Charge
a <- df |> filter(charge == "FE6975")
min(a$vaxx_date, na.rm = TRUE)
max(a$vaxx_date, na.rm = TRUE)