Clinical Decision Curve Analysis (DCA) — decision_curve

Evaluates whether using a predictive model to guide clinical interventions yields higher Net Benefit than default strategies ("treat all" or "treat none") across decision thresholds (Vickers et al., 2006).

Usage

decision_curve_analysis(
  y_true,
  y_prob,
  threshold_range = c(0.05, 0.35),
  step = 0.01,
  harm = 0
)

Arguments

y_true: Binary vector of true outcome values (0 or 1).
y_prob: Numeric vector of predicted event probabilities (between 0 and 1).
threshold_range: Numeric vector of length 2 c(min, max) thresholds (default c(0.05, 0.35)).
step: Numeric step increment (default 0.01).
harm: Numeric test/intervention harm (default 0.0).

Value

A decision_curve_analysis S3 object containing $net_benefit and $summary.