cusum.sprt: Creats a risk-adjusted sequential probability ratio test...

In meyera/rcusum: Tools for the easy creation cusum analysis charts

Description

Implementation of a risk-adjusted sequential probability ratio test (SPRT) chart with control limits as described in Rogers et al. (2004)

Usage

  cusum.sprt(failure_indicator, p0, OR, alpha = 0.01,
    beta = 0.01, by = NULL)

Arguments

failure_indicator	a numeric indicator variable consiting of only c(0,1), where 0 is no failure and 1 is failure for each procedure
p0	a numeric vector representing the acceptable risk score/acceptable failure rate for each single individual. I.e. STS Score values, or emperically modeled risks
OR	the increase in relative risk to the modeled acceptable risk, where An odds ratio of 2, for example, would equate approximately to a doubling of patientspecific risk of failure, an odds ratio of 1.5 to a 50 percent increase in failure risk, and so on.
alpha	Type I error (the probability of concluding that the failure rate has increased, when in fact it has not)
beta	Type II error (the probability of concluding that the failure rate has not increased, when in fact it has)
by	a factor vector consisting of the stratification variable.

Details

For the unadjusted chart, increase in risk is defined in terms of the unacceptable failure rate. However, when risk for each patient varies, it does not make sense to have a common unacceptable rate applied across all operations. This variable unacceptable rate is achieved by defining the increase in terms of a relative risk (ie. odds ratio), rather than a specific rate.

Value

an object of the class ggplot

Author(s)

Alexander Meyer

References

Rogers, C. A., Reeves, B. C., Caputo, M., Ganesh, J. S., Bonser, R. S., & Angelini, G. D. (2004). Control chart methods for monitoring cardiac surgical performance and their interpretation Chris. The Journal of Thoracic and Cardiovascular Surgery, 128(6), 811–819. doi:10.1016/j.jtcvs.2004.03.011

See Also

Other cusum: cusum, cusum.obs_minus_exp

Examples

set.seed(16)
df = data.frame(
 is_failure = c(rbinom(50,1,0.10),rbinom(50,1,0.08),rbinom(50,1,0.05),
                 rbinom(50,1,0.10),rbinom(50,1,0.13),rbinom(50,1,0.14),
                 rbinom(50,1,0.14),rbinom(50,1,0.09),rbinom(50,1,0.25)
 ),
 p0 = c(rnorm(50, 0.10, 0.03),rnorm(50, 0.10, 0.03),rnorm(50, 0.10, 0.03),
         rnorm(50, 0.10, 0.03),rnorm(50, 0.10, 0.03),rnorm(50, 0.10, 0.03),
         rnorm(50, 0.10, 0.03),rnorm(50, 0.15, 0.03),rnorm(50, 0.20, 0.03)
 ),
 by=rep(factor(c("Surgeon A", "Surgeon B", "Surgeon C")), times=c(150,150,150))
)

sprt1= cusum.sprt(rbinom(200,1,0.10), rnorm(200, 0.10, 0.03), 1.5)
print(sprt1)
sprt2= cusum.sprt(df$is_failure, df$p0, 1.5, by=df$by)
print(sprt2)

meyera/rcusum documentation built on May 22, 2019, 7:54 p.m.