Skip to contents

Derive Length of Stay from Date Columns

Source: R/prep_derivation.R
prep_derive_los_from_dates.Rd

Computes LOS when missing (or always when overwrite = TRUE). Adds los_method and los_confidence audit columns. For AIIMS data, uses unit_admission_date + unit_duration_days as a fallback when primary date columns are absent.

Usage

prep_derive_los_from_dates(
  data,
  admission_col = "admission_date",
  outcome_col = "outcome_date",
  los_col = "los_days",
  unit_admission_col = "unit_admission_date",
  unit_duration_col = "unit_duration_days",
  overwrite = FALSE
)

Arguments

data

Data frame.

admission_col

Character. Admission date column. Default "admission_date".

outcome_col

Character. Outcome/discharge date column. Default "outcome_date".

los_col

Character. Target LOS column to populate. Default "los_days".

unit_admission_col

Character. AIIMS unit admission date fallback. Default "unit_admission_date".

unit_duration_col

Character. AIIMS unit duration fallback. Default "unit_duration_days".

overwrite

Logical. Recalculate even when los_col is already present. Default FALSE.

Value

Data frame with los_col, los_method, and los_confidence populated where possible.

Details

Priority order:

  1. outcome_col - admission_col

  2. unit_admission_date + unit_duration_days (AIIMS fallback)

  3. Already-present los_col value retained as-is