While Alive estimands for Recurrent Events

Klaus Holst & Thomas Scheike

2026-05-23

While Alive estimands for Recurrent Events

We consider two while-alive estimands for recurrent events data \[\begin{align*} \frac{E(N(D \wedge t))}{E(D \wedge t)} \end{align*}\] and the mean of the subject specific events per time-unit \[\begin{align*} E( \frac{N(D \wedge t)}{D \wedge t} ) \end{align*}\] for two treatment-groups in the case of an RCT. For the mean of events per time-unit it has been seen that when the sample size is small one can improve the finite sample properties by employing a transformation such as square or cube-root, and thus consider \[\begin{align*} E( (\frac{N(D \wedge t)}{D \wedge t})^.33 ) \end{align*}\]

data(hfactioncpx12)

dtable(hfactioncpx12,~status)
#> 
#> status
#>    0    1    2 
#>  617 1391  124
dd <- WA_recurrent(Event(entry,time,status)~treatment+cluster(id),hfactioncpx12,time=2,death.code=2)
summary(dd)
#> While-Alive summaries:  
#> 
#> RMST,  E(min(D,t)) 
#>            Estimate Std.Err  2.5% 97.5% P-value
#> treatment0    1.859 0.02108 1.817 1.900       0
#> treatment1    1.924 0.01502 1.894 1.953       0
#>  
#>                           Estimate Std.Err    2.5%    97.5% P-value
#> [treatment0] - [treat.... -0.06517 0.02588 -0.1159 -0.01444  0.0118
#> ────────────────────────────────────────────────────────────
#> Null Hypothesis: 
#>   [treatment0] - [treatment1] = 0 
#>  
#> chisq = 6.3405, df = 1, p-value = 0.0118
#> mean events, E(N(min(D,t))): 
#>            Estimate Std.Err  2.5% 97.5%   P-value
#> treatment0    1.572 0.09573 1.384 1.759 1.375e-60
#> treatment1    1.453 0.10315 1.251 1.656 4.376e-45
#>  
#>                           Estimate Std.Err    2.5%  97.5% P-value
#> [treatment0] - [treat....   0.1185  0.1407 -0.1574 0.3943     0.4
#> ────────────────────────────────────────────────────────────
#> Null Hypothesis: 
#>   [treatment0] - [treatment1] = 0 
#>  
#> chisq = 0.7085, df = 1, p-value = 0.4
#> _______________________________________________________ 
#> Ratio of means E(N(min(D,t)))/E(min(D,t)) 
#>    Estimate Std.Err   2.5%  97.5%   P-value
#> p1   0.8457 0.05264 0.7425 0.9488 4.411e-58
#> p2   0.7555 0.05433 0.6490 0.8619 5.963e-44
#>  
#>             Estimate Std.Err     2.5%  97.5% P-value
#> [p1] - [p2]  0.09022 0.07565 -0.05805 0.2385   0.233
#> ────────────────────────────────────────────────────────────
#> Null Hypothesis: 
#>   [p1] - [p2] = 0 
#>  
#> chisq = 1.4222, df = 1, p-value = 0.233
#> _______________________________________________________ 
#> Mean of Events per time-unit E(N(min(D,t))/min(D,t)) 
#>        Estimate Std.Err   2.5%  97.5%   P-value
#> treat0   1.0725  0.1222 0.8331 1.3119 1.645e-18
#> treat1   0.7552  0.0643 0.6291 0.8812 7.508e-32
#>  
#>                     Estimate Std.Err    2.5%  97.5% P-value
#> [treat0] - [treat1]   0.3173  0.1381 0.04675 0.5879 0.02153
#> ────────────────────────────────────────────────────────────
#> Null Hypothesis: 
#>   [treat0] - [treat1] = 0 
#>  
#> chisq = 5.2837, df = 1, p-value = 0.02153

dd <- WA_recurrent(Event(entry,time,status)~treatment+cluster(id),hfactioncpx12,time=2,
           death.code=2,trans=.333)
summary(dd,type="log")
#> While-Alive summaries, log-scale:  
#> 
#> RMST,  E(min(D,t)) 
#>            Estimate  Std.Err   2.5%  97.5% P-value
#> treatment0   0.6199 0.011340 0.5977 0.6421       0
#> treatment1   0.6543 0.007807 0.6390 0.6696       0
#>  
#>                           Estimate Std.Err     2.5%     97.5% P-value
#> [treatment0] - [treat.... -0.03446 0.01377 -0.06145 -0.007478 0.01231
#> ────────────────────────────────────────────────────────────
#> Null Hypothesis: 
#>   [treatment0] - [treatment1] = 0 
#>  
#> chisq = 6.2656, df = 1, p-value = 0.01231
#> mean events, E(N(min(D,t))): 
#>            Estimate Std.Err   2.5%  97.5%   P-value
#> treatment0   0.4523 0.06090 0.3329 0.5716 1.119e-13
#> treatment1   0.3739 0.07097 0.2348 0.5130 1.376e-07
#>  
#>                           Estimate Std.Err    2.5%  97.5% P-value
#> [treatment0] - [treat....  0.07835 0.09352 -0.1049 0.2616  0.4022
#> ────────────────────────────────────────────────────────────
#> Null Hypothesis: 
#>   [treatment0] - [treatment1] = 0 
#>  
#> chisq = 0.7018, df = 1, p-value = 0.4022
#> _______________________________________________________ 
#> Ratio of means E(N(min(D,t)))/E(min(D,t)) 
#>    Estimate Std.Err    2.5%    97.5%   P-value
#> p1  -0.1676 0.06224 -0.2896 -0.04563 7.081e-03
#> p2  -0.2804 0.07192 -0.4214 -0.13947 9.651e-05
#>  
#>             Estimate Std.Err     2.5%  97.5% P-value
#> [p1] - [p2]   0.1128 0.09511 -0.07361 0.2992  0.2356
#> ────────────────────────────────────────────────────────────
#> Null Hypothesis: 
#>   [p1] - [p2] = 0 
#>  
#> chisq = 1.4067, df = 1, p-value = 0.2356
#> _______________________________________________________ 
#> Mean of Events per time-unit E(N(min(D,t))/min(D,t)) 
#>        Estimate Std.Err    2.5%   97.5%   P-value
#> treat0  -0.3833 0.04939 -0.4801 -0.2865 8.487e-15
#> treat1  -0.5380 0.05666 -0.6491 -0.4270 2.191e-21
#>  
#>                     Estimate Std.Err     2.5%  97.5% P-value
#> [treat0] - [treat1]   0.1548 0.07517 0.007459 0.3021 0.03948
#> ────────────────────────────────────────────────────────────
#> Null Hypothesis: 
#>   [treat0] - [treat1] = 0 
#>  
#> chisq = 4.2403, df = 1, p-value = 0.03948

We see that the ratio of means is not very different between groups, but that the subject-specific mean of events per time-unit shows that those on the active treatment have fewer events per time-unit on average.

We can also fit a regression model for the mean of the subject-specific events per time-unit, here using the exponential link.

hfactioncpx12$age <- rnorm(741)[hfactioncpx12$id]
hfactioncpx12$sex <- rbinom(741,1,0.5)[hfactioncpx12$id]

dd <- WA_reg(Event(entry,time,status)~treatment+age+sex+cluster(id),hfactioncpx12,time=2,death.code=2)
summary(dd)
#>    n events
#>  741     86
#> 
#>  741 clusters
#> coeffients:
#>               Estimate    Std.Err       2.5%      97.5% P-value
#> (Intercept)  0.0059349  0.1018293 -0.1936468  0.2055166  0.9535
#> treatment1  -0.3475423  0.1399946 -0.6219268 -0.0731578  0.0130
#> age          0.0447745  0.0619333 -0.0766125  0.1661615  0.4697
#> sex          0.1252971  0.1557374 -0.1799426  0.4305367  0.4211
#> 
#> exp(coeffients):
#>             Estimate    2.5%  97.5%
#> (Intercept)  1.00595 0.82395 1.2282
#> treatment1   0.70642 0.53691 0.9295
#> age          1.04579 0.92625 1.1808
#> sex          1.13349 0.83532 1.5381

Composite outcomes involving death and marks

The event count can be generalised in various ways by using outcomes other than \(N(D \wedge t)\), for example, \[\begin{align*} \tilde N(D \wedge t) = \int_0^t I(D \geq s) M(s) dN(s) + \sum_j M_j I(D \leq t,\epsilon=j) ) \end{align*}\] where \(M(s)\) are the marks associated with \(N(s)\) and \(M_j\) are marks associated with the different causes of the terminal event. This provides an extension of the weighted composite outcomes measure of Mao & Lin (2022).

The marks (or weights) can be stochastic, for example when counting hospital expenses, and are stored as a vector in the data frame. The marks for the event times (defined through the causes) are then used.

Here we weight death by 2 and otherwise count recurrent events as before (with weight 1):

hfactioncpx12$marks <- runif(nrow(hfactioncpx12))

##ddmg <- WA_recurrent(Event(entry,time,status)~treatment+cluster(id),hfactioncpx12,time=2,
##cause=1:2,death.code=2,marks=hfactioncpx12$marks)
##summary(ddmg)

ddm <- WA_recurrent(Event(entry,time,status)~treatment+cluster(id),hfactioncpx12,time=2,
cause=1:2,death.code=2,marks=hfactioncpx12$status)

SessionInfo

sessionInfo()
#> R version 4.6.0 (2026-04-24)
#> Platform: x86_64-pc-linux-gnu
#> Running under: Ubuntu 24.04.4 LTS
#> 
#> Matrix products: default
#> BLAS:   /home/kkzh/.asdf/installs/r/4.6.0/lib/R/lib/libRblas.so 
#> LAPACK: /usr/lib/x86_64-linux-gnu/lapack/liblapack.so.3.12.0  LAPACK version 3.12.0
#> 
#> locale:
#>  [1] LC_CTYPE=en_US.UTF-8       LC_NUMERIC=C              
#>  [3] LC_TIME=en_US.UTF-8        LC_COLLATE=en_US.UTF-8    
#>  [5] LC_MONETARY=en_US.UTF-8    LC_MESSAGES=en_US.UTF-8   
#>  [7] LC_PAPER=en_US.UTF-8       LC_NAME=C                 
#>  [9] LC_ADDRESS=C               LC_TELEPHONE=C            
#> [11] LC_MEASUREMENT=en_US.UTF-8 LC_IDENTIFICATION=C       
#> 
#> time zone: Europe/Copenhagen
#> tzcode source: system (glibc)
#> 
#> attached base packages:
#> [1] splines   stats     graphics  grDevices utils     datasets  methods  
#> [8] base     
#> 
#> other attached packages:
#> [1] prodlim_2026.03.11 timereg_2.0.7      survival_3.8-6     mets_1.3.10       
#> 
#> loaded via a namespace (and not attached):
#>  [1] Matrix_1.7-5           future.apply_1.20.2    jsonlite_2.0.0        
#>  [4] ucminf_1.2.3           compiler_4.6.0         Rcpp_1.1.1-1.1        
#>  [7] parallel_4.6.0         jquerylib_0.1.4        globals_0.19.1        
#> [10] yaml_2.3.12            fastmap_1.2.0          lattice_0.22-9        
#> [13] R6_2.6.1               knitr_1.51             future_1.70.0         
#> [16] bslib_0.11.0           rlang_1.2.0            cachem_1.1.0          
#> [19] xfun_0.57              sass_0.4.10            otel_0.2.0            
#> [22] cli_3.6.6              digest_0.6.39          grid_4.6.0            
#> [25] mvtnorm_1.3-7          lifecycle_1.0.5        lava_1.9.1            
#> [28] RcppArmadillo_15.2.6-1 KernSmooth_2.23-26     evaluate_1.0.5        
#> [31] data.table_1.18.4      numDeriv_2016.8-1.1    listenv_0.10.1        
#> [34] codetools_0.2-20       stats4_4.6.0           parallelly_1.47.0     
#> [37] rmarkdown_2.31         tools_4.6.0            htmltools_0.5.9