Obesity Indices • HealthMarkers

Scope

Compute anthropometric adiposity/shape indices (BMI + category, WHR/WHtR, AVI, BAI, ABSI, BRI, Conicity Index, waist/BMI and weight/height ratios, optional WHRadjBMI and RFM) with unit conversion, NA/extreme handling, and safe-division guards.

When to use

You have weight/height/waist/hip (and optionally sex) and want a comprehensive set of body composition proxies in one call.
You need built-in unit conversion (kg/lb, cm/m), NA policy controls, and optional extreme-value scanning/capping.
You want optional BMI-adjusted WHR and Relative Fat Mass when sex coding is available.

Inputs

data: data frame/tibble with numeric columns for weight, height, waist, hip; sex needed only if include_RFM = TRUE (coded 0=male, 1=female).
Column arguments are unquoted: weight, height, waist, hip, optional sex.
Units: weight kg (or lb if weight_unit = "lb"), height m (or cm if height_unit = "cm"), waist/hip cm.
weight_unit / height_unit: convert to kg/m internally.
adjust_WHR: add BMI-adjusted WHR residuals.
include_RFM: compute Relative Fat Mass (requires sex).
na_action: keep (default) propagates NA; omit drops rows missing required inputs; error aborts when required inputs contain NA.
na_warn_prop: threshold for high-missingness debug messages (default 0.2).
check_extreme: set TRUE to scan weight_kg/height_m/waist/hip; extreme_action (warn/cap/error/ignore) controls handling. Defaults: weight_kg 20–400; height_m 1.2–2.5; waist 30–200 cm; hip 30–200 cm.
verbose: optional progress and summary logging.

Quick start

library(HealthMarkers)
library(tibble)

df <- tibble::tibble(
  wt    = c(70, 80),   # kg
  ht    = c(175, 165), # cm
  waist = c(80, 90),   # cm
  hip   = c(100, 95),  # cm
  sex   = c(0, 1)
)

obesity_indices(
  data = df,
  weight = wt,
  height = ht,
  waist = waist,
  hip = hip,
  sex = sex,
  weight_unit = "kg",
  height_unit = "cm",
  adjust_WHR = TRUE,
  include_RFM = TRUE,
  na_action = "keep",
  check_extreme = FALSE,
  verbose = FALSE
)
#> # A tibble: 2 × 15
#>   weight_kg height_m   BMI BMI_cat         WHR WHRadjBMI waist_to_height_ratio
#>       <dbl>    <dbl> <dbl> <chr>         <dbl>     <dbl>                 <dbl>
#> 1        70     1.75  22.9 Normal weight 0.8           0                  45.7
#> 2        80     1.65  29.4 Overweight    0.947         0                  54.5
#> # ℹ 8 more variables: waist_to_BMI_ratio <dbl>, weight_to_height_ratio <dbl>,
#> #   AVI <dbl>, BAI <dbl>, ABSI <dbl>, BRI <dbl>, CI <dbl>, RFM <dbl>

Extreme scan and cap

obesity_indices(
  data = df,
  weight = wt,
  height = ht,
  waist = waist,
  hip = hip,
  sex = sex,
  weight_unit = "kg",
  height_unit = "cm",
  adjust_WHR = FALSE,
  include_RFM = TRUE,
  na_action = "omit",
  check_extreme = TRUE,
  extreme_action = "cap",
  verbose = TRUE
)
#> # A tibble: 2 × 14
#>   weight_kg height_m   BMI BMI_cat         WHR waist_to_height_ratio
#>       <dbl>    <dbl> <dbl> <chr>         <dbl>                 <dbl>
#> 1        70     1.75  22.9 Normal weight 0.8                    45.7
#> 2        80     1.65  29.4 Overweight    0.947                  54.5
#> # ℹ 8 more variables: waist_to_BMI_ratio <dbl>, weight_to_height_ratio <dbl>,
#> #   AVI <dbl>, BAI <dbl>, ABSI <dbl>, BRI <dbl>, CI <dbl>, RFM <dbl>

Missing-data policy

try(
  obesity_indices(
    data = df,
    weight = wt,
    height = ht,
    waist = waist,
    hip = hip,
    na_action = "error"
  )
)
#> # A tibble: 2 × 13
#>   weight_kg height_m   BMI BMI_cat         WHR waist_to_height_ratio
#>       <dbl>    <dbl> <dbl> <chr>         <dbl>                 <dbl>
#> 1        70     1.75  22.9 Normal weight 0.8                    45.7
#> 2        80     1.65  29.4 Overweight    0.947                  54.5
#> # ℹ 7 more variables: waist_to_BMI_ratio <dbl>, weight_to_height_ratio <dbl>,
#> #   AVI <dbl>, BAI <dbl>, ABSI <dbl>, BRI <dbl>, CI <dbl>

Outputs and expectations

Adds: weight_kg, height_m, BMI, BMI_cat, WHR, waist_to_height_ratio, AVI, BAI, ABSI, BRI, CI, waist_to_BMI_ratio, weight_to_height_ratio, optional WHRadjBMI, optional RFM.
Zero denominators yield NA with warnings; WHRadjBMI requires variance in WHR and BMI; RFM requires sex coded 0/1.

Verbose diagnostics

old_opt <- options(healthmarkers.verbose = "inform")
obesity_indices(
  data = df,
  weight = wt,
  height = ht,
  waist = waist,
  hip = hip,
  sex = sex,
  weight_unit = "kg",
  height_unit = "cm",
  verbose = TRUE
)
#> obesity_indices(): preparing inputs
#> obesity_indices(): column map: weight -> 'wt', height -> 'ht', waist -> 'waist', hip -> 'hip', sex -> 'sex'
#> obesity_indices(): results: weight_kg 2/2, height_m 2/2, BMI 2/2, BMI_cat 2/2, WHR 2/2, waist_to_height_ratio 2/2, waist_to_BMI_ratio 2/2, weight_to_height_ratio 2/2, AVI 2/2, BAI 2/2, ABSI 2/2, BRI 2/2, CI 2/2
#> # A tibble: 2 × 13
#>   weight_kg height_m   BMI BMI_cat         WHR waist_to_height_ratio
#>       <dbl>    <dbl> <dbl> <chr>         <dbl>                 <dbl>
#> 1        70     1.75  22.9 Normal weight 0.8                    45.7
#> 2        80     1.65  29.4 Overweight    0.947                  54.5
#> # ℹ 7 more variables: waist_to_BMI_ratio <dbl>, weight_to_height_ratio <dbl>,
#> #   AVI <dbl>, BAI <dbl>, ABSI <dbl>, BRI <dbl>, CI <dbl>
options(old_opt)

Tips

Set weight/height units correctly; waist/hip are assumed cm.
Enable include_RFM = TRUE only with valid sex coding; leave FALSE otherwise.
WHRadjBMI needs variation in WHR and BMI; otherwise it returns NA with a warning.
Tighten extreme_rules to your cohort before using extreme_action = "cap" or "error".
Use na_action = "omit" for row-complete outputs; keep during QA to see missingness.