dsr (version 0.2.2)

dsrrec: Compute Directly Standardized Rates for Recurrent Events

Description

Computes directly standardized rates for recurrent events by subgroup with confidence intervals.

Usage

dsrrec(data, event, fu, subgroup, ..., refdata, sig = 0.95, mp, decimals)

Arguments

data

A data frame with counts and unit-times summarized by the standardization variables.

event

A variable within the input data that corresponds to the event counts.

fu

A variable within the input data that corresponds to the unit-time.

subgroup

A variable within the input data frame for which rates are calculated by.

...

Variables(s) within the input data that for which rates are to be standardized by. The input data and ref data should both be summarized by these.

refdata

A data frame with population unit-times summarized by the standardization variables. The unit-time variable name must named pop.

sig

The desired level of confidence in computing confidence intervals. The default is 0.95 for 95 percent CIs.

mp

A constant to multiply rates by (e.g. mp=1000 for rates per 1000).

decimals

Round estimates to a desired decimal place.

References

Stukel, T. A., Glynn, R. J., Fisher, E. S., Sharp, S. M., Lu-Yao, G and Wennberg, J. E. (1994). Standardized rates of recurrent outcomes. Statistics in Medicine, 13, 1781-1791.

Fay, M.P., & Feuer, E.J. (1997). Confidence intervals for directly standardized rates: a method based on the gamma distribution. Statistics in Medicine, 16, 791-801.

Examples

# NOT RUN {
#An example of directly standardized rates for recurrent events

library(frailtypack)
library(dplyr)
library(dsr)
data(readmission)

#Make an individual level dataset with total event counts and total observation times
treadm <- as.data.frame(readmission %>%
                         group_by(id) %>%
                         filter(max(enum)==enum ) %>%
                         mutate(events=enum-1, time=t.stop) %>%
                         select(id, events, time, sex, dukes))

#Make the standard pop
tref <- as.data.frame(treadm %>%
                     group_by(sex) %>%
                     mutate(pop=sum(time)) %>%
                     select(sex, pop) %>%
                     distinct(sex, pop))

#Get directly standardized rates (age-adjusted) for readmissions by Dukes' tumor grade.
analysis <- dsrrec(data=treadm,
                  event=events,
                  fu=time,
                  refdata=tref,
                  subgroup=dukes,
                  sex,
                  mp=1000,
                  decimals=3)
# }