nri_est function

Calculation of Net Reclassification Index measures

nri_est Calculation of proportion of Reclassified persons and NRI for Cox Regression Models

nri_est(data, p0, p1, time, status, t_risk, cutoff)

Arguments

• data: Data frame with relevant predictors
• p0: risk outcome probabilities for reference model.
• p1: risk outcome probabilities for new model.
• time: Character vector. Name of time variable.
• status: Character vector. Name of status variable.
• t_risk: Follow-up value to calculate cases, controls. See details.
• cutoff: A numerical vector that defines the outcome probability cutoff values.

Returns

An object from which the following objects can be extracted:

• prop_up_case proportion of cases reclassified upwards.
• prop_down_case proportion of cases reclassified downwards.
• prop_up_ctr proportion of controls reclassified upwards.
• prop_down_ctr proportion of controls reclassified downwards.
• nri_plus proportion reclassified for events.
• nri_min proportion reclassified for nonevents.
• nri net reclassification improvement.

Details

Follow-up for which cases nd controls are determined. For censored cases before this follow-up the expected risk of being a case is calculated by using the Kaplan-Meier value to calculate the expected number of cases. These expected numbers are used to calculate the NRI proportions but are not shown by function nricens.

Examples

library(survival)
  lbpmicox1 <- subset(psfmi::lbpmicox, Impnr==1) # extract dataset
  
  fit_cox0 <- 
    coxph(Surv(Time, Status) ~ Duration + Pain, data=lbpmicox1, x=TRUE)
  fit_cox1 <- 
    coxph(Surv(Time, Status) ~ Duration + Pain + Function + Radiation, 
    data=lbpmicox1, x=TRUE)

  p0 <- risk_coxph(fit_cox0, t_risk=80)
  p1 <- risk_coxph(fit_cox1, t_risk=80)
  
  nri <- nri_est(data=lbpmicox1,
                      p0=p0,
                      p1=p1,
                      time = "Time",
                      status = "Status",
                      cutoff=0.45,
                      t_risk=80)

References

Cook NR, Ridker PM. Advances in measuring the effect of individual predictors of cardiovascular risk: the role of reclassification measures. Ann Intern Med. 2009;150(11):795-802.

Steyerberg EW, Pencina MJ. Reclassification calculations for persons with incomplete follow-up. Ann Intern Med. 2010;152(3):195-6; author reply 196-7.

Pencina MJ, D'Agostino RB Sr, Steyerberg EW. Extensions of net reclassification improvement calculations to measure usefulness of new biomarkers. Stat Med. 2011;30(1):11-21

Inoue E (2018). nricens: NRI for Risk Prediction Models with Time to Event and Binary Response Data. R package version 1.6, https://CRAN.R-project.org/package=nricens.

Author(s)

Martijn Heymans, 2023