Skip to contents

Wald test for treatment effect using negative binomial model (Mütze et al.)

Source: R/mutze_test.R
mutze_test.Rd

Fits a negative binomial (or Poisson) log-rate model to the aggregated subject-level data produced by cut_data_by_date(). The method matches the Wald test described by Mütze et al. (2019) for comparing treatment arms with recurrent event outcomes.

Usage

mutze_test(data, method = c("nb", "poisson"), conf_level = 0.95)

Arguments

data

A data frame with at least the columns treatment, events, and tte (follow-up time). Typically output from cut_data_by_date().

method

Type of model to fit: "nb" (default) uses a negative binomial GLM via MASS::glm.nb(), "poisson" fits a Poisson GLM.

conf_level

Confidence level for the rate ratio interval. Default 0.95.

Value

A list containing the fitted model summary with elements:

  • estimate: log rate ratio (experimental vs control).

  • se: standard error for the log rate ratio.

  • z: Wald statistic.

  • p_value: two-sided p-value.

  • rate_ratio: estimated rate ratio and its confidence interval.

  • dispersion: estimated dispersion (theta) when method = "nb".

  • group_summary: observed subjects/events/exposure per treatment.

Examples

enroll_rate <- data.frame(rate = 20 / (5 / 12), duration = 5 / 12)
fail_rate <- data.frame(treatment = c("Control", "Experimental"), rate = c(0.5, 0.3))
dropout_rate <- data.frame(
  treatment = c("Control", "Experimental"),
  rate = c(0.1, 0.05), duration = c(100, 100)
)
sim <- nb_sim(enroll_rate, fail_rate, dropout_rate, max_followup = 2, n = 40)
cut <- cut_data_by_date(sim, cut_date = 1.5)
mutze_test(cut)
#> $method
#> [1] "Negative binomial Wald"
#> 
#> $estimate
#> [1] -0.4016237
#> 
#> $se
#> [1] 0.4928251
#> 
#> $z
#> [1] -0.8149418
#> 
#> $p_value
#> [1] 0.4151056
#> 
#> $rate_ratio
#> [1] 0.6692325
#> 
#> $conf_int
#> [1] 0.254732 1.758209
#> 
#> $conf_level
#> [1] 0.95
#> 
#> $dispersion
#> [1] 5580.973
#> 
#> $model
#> 
#> Call:  MASS::glm.nb(formula = events ~ treatment + offset(log(tte)), 
#>     data = df, init.theta = 5580.973267, link = log)
#> 
#> Coefficients:
#>           (Intercept)  treatmentExperimental  
#>               -0.7111                -0.4016  
#> 
#> Degrees of Freedom: 39 Total (i.e. Null);  38 Residual
#> Null Deviance:	    35.59 
#> Residual Deviance: 34.92 	AIC: 72.15
#> 
#> $group_summary
#>      treatment subjects events exposure
#> 1      Control       20     10 20.36196
#> 2 Experimental       20      7 21.29804
#>