Skip to contents

Systematic COronary Risk Evaluation (SCORE) model function including chronic kidney disease (CKD) measures

Source: R/13_SCORE2_CKD_func.R
SCORE2-slash-OP-CKD.Rd

This function implements the SCORE2 and SCORE2 older population (OP) score calculation as a vector plus the ckd add-ons as suggested on the https://doi.org/10.1093/eurjpc/zwac216 publication

*Diabetes mellitus was included in the modelling since this was necessary for the recalibration approach, which relies data from the whole population, including those with diabetes. However, SCORE2 is not intended for use in individuals with diabetes and has not been validated in this population. For risk prediction in the target population of individuals without diabetes this risk factor will always be 0, meaning the coefficient can effectively be ignored.

formula in SCORE2 Updated Supplementary Material page 9. paper: "SCORE2 risk prediction algorithms: new models to estimate 10-year risk of cardiovascular disease in Europe"

Age 10-year risk of fatal and non-fatal cardiovascular disease

| Low risk | Moderate risk | High risk |
| :———— | ————- | ————— | ———-: |
| < 50 years | <2.5 | 50 - 69 years | <5 | => 70 years | <7.5

above classifications referred from https://www.inanutshell.ch/en/digital-doctors-bag/score2-and-score2-op/#:~:text=SCORE2

The underlying model was developed and validated using eligible data from 3,054,840 individuals from 34 different cohorts and 5,997,719 individuals from 34 independent cohorts, respectively, from the Chronic Kidney Disease Prognosis Consortium. https://doi.org/10.1093/eurjpc/zwac176

Usage

SCORE2_CKD(
  Risk.region,
  Age = Age,
  Gender = Gender,
  smoker = smoker,
  systolic.bp = systolic.bp,
  diabetes = diabetes,
  total.chol = total.chol,
  total.hdl = total.hdl,
  eGFR = eGFR,
  ACR = NA,
  trace = NA,
  addon = "ACR",
  classify
)

Arguments

Risk.region

a character value to set the desired risk region calculations. Categories should include "Low", "Moderate", "High", or "Very high"

Age

a numeric vector of age values, in years

Gender

a binary character vector of Gender values. Categories should include only 'male' or 'female'

smoker

a binary numeric vector, 1 = yes and 0 = no

systolic.bp

a numeric vector of systolic blood pressure continuous values

diabetes

a binary numeric vector, 1 = yes and 0 = no

total.chol

a numeric vector of total cholesterol values, in mmol/L

total.hdl

a numeric vector of total high density lipoprotein total.hdl values, in mmol/L

eGFR

a numeric vector of total estimated glomerular rate (eGFR) values, in mL/min/1.73m2

ACR

a numeric vector of total urine albumin to creatine ratio (ACR) values, in mg/g. Default set to NA

trace

a character vector of urine protein dipstick categories. Categories should include 'negative', 'trace', '1+', '2+', '3+', '4+. Default set to NA

addon

set the add-on you wish to calculate. Categories should include only 'ACR' or 'trace'. Default set to 'ACR'

classify

set TRUE if wish to add a column with the scores' categories

Value

A vector with SCORE2/OP score calculations with CKD add-ons and/or a vector of their classifications if indicated

Examples


# Create a data frame or list with the necessary variables
# Set the number of rows
num_rows <- 100

# Create a larger dataset with 100 rows
cohort_xx <- data.frame(
  typical_symptoms.num = as.numeric(sample(0:6, num_rows, replace = TRUE)),
  ecg.normal = as.numeric(sample(c(0, 1), num_rows, replace = TRUE)),
  abn.repolarisation = as.numeric(sample(c(0, 1), num_rows, replace = TRUE)),
  ecg.st.depression = as.numeric(sample(c(0, 1), num_rows, replace = TRUE)),
  Age = as.numeric(sample(40:85, num_rows, replace = TRUE)),
  diabetes = sample(c(1, 0), num_rows, replace = TRUE),
  smoker = sample(c(1, 0), num_rows, replace = TRUE),
  hypertension = sample(c(1, 0), num_rows, replace = TRUE),
  hyperlipidaemia = sample(c(1, 0), num_rows, replace = TRUE),
  family.history = sample(c(1, 0), num_rows, replace = TRUE),
  atherosclerotic.disease = sample(c(1, 0), num_rows, replace = TRUE),
  presentation_hstni = as.numeric(sample(10:100, num_rows, replace = TRUE)),
  Gender = sample(c("male", "female"), num_rows, replace = TRUE),
  sweating = as.numeric(sample(c(0, 1), num_rows, replace = TRUE)),
  pain.radiation = as.numeric(sample(c(0, 1), num_rows, replace = TRUE)),
  pleuritic = as.numeric(sample(c(0, 1), num_rows, replace = TRUE)),
  palpation = as.numeric(sample(c(0, 1), num_rows, replace = TRUE)),
  ecg.twi = as.numeric(sample(c(0, 1), num_rows, replace = TRUE)),
  second_hstni = as.numeric(sample(1:200, num_rows, replace = TRUE)),
  killip.class = as.numeric(sample(1:4, num_rows, replace = TRUE)),
  systolic.bp = as.numeric(sample(90:180, num_rows, replace = TRUE)),
  heart.rate = as.numeric(sample(0:300, num_rows, replace = TRUE)),
  creat = as.numeric(sample(0:4, num_rows, replace = TRUE)),
  cardiac.arrest = as.numeric(sample(c(0, 1), num_rows, replace = TRUE)),
  previous.pci = as.numeric(sample(c(0, 1), num_rows, replace = TRUE)),
  previous.cabg = as.numeric(sample(c(0, 1), num_rows, replace = TRUE)),
  aspirin = as.numeric(sample(c(0, 1), num_rows, replace = TRUE)),
  number.of.episodes.24h = as.numeric(sample(0:20, num_rows, replace = TRUE)),
  total.chol = as.numeric(round(runif(num_rows, 3.9, 7.2), 1)),
  total.hdl = as.numeric(round(runif(num_rows, .8, 2.1), 1)),
  Ethnicity = sample(c("white", "black", "asian", "other"), num_rows, replace = TRUE),
  eGFR = as.numeric(sample(15:120, num_rows, replace = TRUE)),
  ACR = as.numeric(sample(5:1500, num_rows, replace = TRUE)),
  trace = sample(c("trace", "1+", "2+", "3+", "4+"), num_rows, replace = TRUE)
)

# Call the function with the cohort_xx

  results <- cohort_xx %>% rowwise() %>%
  mutate(SCORE2OP_CKD_score = SCORE2_CKD(Risk.region = "Low", Age, Gender,
  smoker, systolic.bp, diabetes, total.chol, total.hdl, eGFR, ACR, trace,
  classify = FALSE))