find_column_preening

Internal helper function to find a column matching a list of patterns

internal

Facilitates probabilistic record linkage between infectious disease
surveillance datasets (notifiable disease registers, outbreak line-lists),
vaccination registries, and hospitalization records using methods based on
Fellegi and Sunter (1969) <doi:10.1080/01621459.1969.10501049> and
Sayers et al. (2016) <doi:10.1093/ije/dyv322>. The package provides
core functions for data preparation, linkage, and analysis: clean_the_nest()
standardizes variable names and formats across heterogeneous datasets;
murmuration() performs machine learning-based record linkage using blocking
variables and similarity metrics; molting() deidentifies datasets for secure
sharing; homing() re-identifies previously deidentified datasets; plumage()
identifies and categorizes comorbidities; and preening() creates analysis-ready
variables including age categories and temporal groupings. Designed for
epidemiological research linking acute and post-acute disease outcomes to
vaccination status and healthcare utilization. Supports multiple linkage
scenarios including case-to-vaccination, case-to-hospitalization, and
event-based vaccination status determination (e.g., outbreak attendees, flight
passengers, exposure site visitors).

Nicolas Smoll

starling

Link Infectious Disease Cases to Vaccination and Hospitalization
Records

find_column_preening function

<dl><dt>data</dt>
<dd>A data frame</dd>
<dt>patterns</dt>
<dd>Character vector of column name patterns to search for</dd></dl>

Arguments

Find Column by Pattern Matching — find_column_preening

<dl>

<dt>data</dt>
<dd>A data frame</dd>


<dt>patterns</dt>
<dd>Character vector of column name patterns to search for</dd>

</dl>

find_column_preening: Find Column by Pattern Matching

Description

Usage

Value

Arguments