cleanepi: Clean and standardize epidemiological data

cleanepi is an R package designed for cleaning, curating, and standardizing epidemiological data. It streamlines various data cleaning tasks that are typically expected when working with datasets in epidemiology.

Key functionalities of cleanepi include:

1. Removing irregularities: It removes duplicated and empty rows and columns, as well as columns with constant values.

2. Handling missing values: It replaces missing values with the standard NA format, ensuring consistency and ease of analysis.

3. Ensuring data integrity: It ensures the uniqueness of uniquely identified columns, thus maintaining data integrity and preventing duplicates.

4. Date conversion: It offers functionality to convert character columns to Date format under specific conditions, enhancing data uniformity and facilitating temporal analysis. It also offers conversion of numeric values written in letters into numbers.

5. Standardizing entries: It can standardize column entries into specified formats, promoting consistency across the dataset.

6. Time span calculation: It calculates the time span between two elements of type Date, providing valuable demographic insights for epidemiological analysis.

cleanepi operates on data frames or similar structures like tibbles, as well as linelist objects commonly used in epidemiological research. It returns the processed data in the same format, ensuring seamless integration into existing workflows. Additionally, it generates a comprehensive report detailing the outcomes of each cleaning task.

cleanepi is developed by the Epiverse-TRACE team at the Medical Research Council The Gambia unit at the London School of Hygiene and Tropical Medicine.

Installation

cleanepi can be installed from CRAN using

install.packages("cleanepi")

The latest development version of cleanepi can be installed from GitHub.

if (!require("pak")) install.packages("pak")
pak::pak("epiverse-trace/cleanepi")
library(cleanepi)

Quick start

The main function in cleanepi is clean_data(), which internally makes call of almost all standard data cleaning functions, such as removal of empty and duplicated rows and columns, replacement of missing values, etc. However, each function can also be called independently to perform a specific task. This mechanism is explained in details in the vignette. Below is typical example of how to use the clean_data() function.

# READING IN THE TEST DATASET
test_data <- readRDS(
  system.file("extdata", "test_df.RDS", package = "cleanepi")
)

# READING IN THE DATA DICTIONARY
test_dictionary <- readRDS(
  system.file("extdata", "test_dictionary.RDS", package = "cleanepi")
)

# SCAN THROUGH THE DATA
scan_res <- cleanepi::scan_data(test_data)

# DEFINING THE CLEANING PARAMETERS
replace_missing_values <- list(target_columns = NULL, na_strings = "-99")
remove_duplicates <- list(target_columns = NULL)
standardize_dates <- list(
  target_columns = NULL,
  error_tolerance = 0.4,
  format = NULL,
  timeframe = as.Date(c("1973-05-29", "2023-05-29")),
  orders = list(
    world_named_months = c("Ybd", "dby"),
    world_digit_months = c("dmy", "Ymd"),
    US_formats = c("Omdy", "YOmd")
  )
)
standardize_subject_ids <- list(
  target_columns = "study_id",
  prefix = "PS",
  suffix = "P2",
  range = c(1, 100),
  nchar = 7
)
remove_constants <- list(cutoff = 1)
standardize_column_names <- list(
  keep = "date.of.admission",
  rename = c(DOB = "dateOfBirth")
)
to_numeric <- list(target_columns = "sex", lang = "en")

# PERFORMING THE DATA CLEANING
cleaned_data <- clean_data(
  data = test_data,
  standardize_column_names = standardize_column_names,
  remove_constants = remove_constants,
  replace_missing_values = replace_missing_values,
  remove_duplicates = remove_duplicates,
  standardize_dates = standardize_dates,
  standardize_subject_ids = standardize_subject_ids,
  to_numeric = to_numeric,
  dictionary = test_dictionary,
  check_date_sequence = NULL
)
#> ℹ Cleaning column names
#> ℹ Replacing missing values with NA
#> ℹ Removing constant columns and empty rows
#> ℹ Removing duplicated rows
#> ℹ No duplicates were found.
#> ℹ Standardizing Date columns
#> ! Detected 8 values that comply with multiple formats and no values that are
#>   outside of the specified time frame.
#> ℹ Enter `print_report(data = dat, "date_standardization")` to access them,
#>   where "dat" is the object used to store the output from this operation.
#> ℹ Checking subject IDs format
#> 
#> ! Detected 0 missing, 0 duplicated, and 3 incorrect subject IDs.
#> ℹ Enter `print_report(data = dat, "incorrect_subject_id")` to access them,
#>   where "dat" is the object used to store the output from this operation.
#> ℹ You can use the `correct_subject_ids()` function to correct them.
#> ℹ Converting the following  column into numeric: sex
#> 
#> ℹ Performing dictionary-based cleaning

# ADD THE DATA SCANNING RESULT TO THE REPORT
cleaned_data <- cleanepi::add_to_report(
  x = cleaned_data,
  key = "scanning_result",
  value = scan_res
)

# DISPLAY THE DATA CLEANING REPORT
print_report(cleaned_data, print = TRUE)

Vignette

browseVignettes("cleanepi")

Lifecycle

This package is currently an experimental, as defined by the RECON software lifecycle. This means that it is functional, but interfaces and functionalities may change over time, testing and documentation may be lacking.

Contributions

Contributions are welcome via pull requests.

Code of Conduct

Please note that the cleanepi project is released with a Contributor Code of Conduct. By contributing to this project, you agree to abide by its terms.

Citing this package

citation("cleanepi")
#> To cite package 'cleanepi' in publications use:
#> 
#>   Mané K, Degoot A, Ahadzie B, Mohammed N, Bah B (2025). _cleanepi:
#>   Clean and Standardize Epidemiological Data_.
#>   doi:10.5281/zenodo.11473985
#>   <https://doi.org/10.5281/zenodo.11473985>,
#>   <https://epiverse-trace.github.io/cleanepi/>.
#> 
#> A BibTeX entry for LaTeX users is
#> 
#>   @Manual{,
#>     title = {cleanepi: Clean and Standardize Epidemiological Data},
#>     author = {Karim Mané and Abdoelnaser Degoot and Bankolé Ahadzie and Nuredin Mohammed and Bubacarr Bah},
#>     year = {2025},
#>     doi = {10.5281/zenodo.11473985},
#>     url = {https://epiverse-trace.github.io/cleanepi/},
#>   }