Skip to contents

Cardiac Biomarkers and Heart Failure Events Data

Source: R/joint_modeling_data.R
cardiac_joint_data.Rd

A longitudinal dataset containing NT-proBNP measurements and heart failure events for 120 cardiac patients.

Usage

cardiac_joint_data

Format

A data frame with 964 observations and 8 variables:

patient_id

Character. Unique patient identifier (HF_001 to HF_120)

age

Numeric. Patient age at baseline (years)

nyha_class

Factor. NYHA functional class (Class 1, Class 2, Class 3, Class 4)

baseline_ef

Numeric. Baseline ejection fraction (%)

visit_time

Numeric. Time of NT-proBNP measurement (months from baseline)

nt_probnp

Numeric. NT-proBNP level (pg/mL)

survival_time

Numeric. Time to heart failure event or last follow-up (months)

hf_event_status

Numeric. Event indicator (0 = censored, 1 = heart failure hospitalization/death)

Source

Simulated data based on heart failure cohort studies

Details

The dataset simulates heart failure progression where:

  • NT-proBNP levels generally increase over time

  • Higher NYHA class associated with higher NT-proBNP

  • Higher NT-proBNP increases hazard of heart failure events

  • Regular monitoring every ~4 months

  • 20% event rate over 36 months follow-up

Examples

data(cardiac_joint_data)

# NT-proBNP trajectories by NYHA class
library(ggplot2)
ggplot(cardiac_joint_data, aes(x = visit_time, y = log(nt_probnp), color = nyha_class)) +
  geom_smooth(method = "loess") +
  labs(title = "NT-proBNP Trajectories by NYHA Class", 
       x = "Time (months)", y = "log(NT-proBNP)")