Competing risks- WAMS, complete withdrawal
Carolina Ferreira Atuesta
0.1 Load data from spreadsheet
data <-
read_xlsx(
"/Users/carolinaferreiraatuesta/Documents/WAMS/ASM_withdrawal_registry/WAMS_Registry.xlsx"
)
data <-
subset(
data,
data$began_wd == 1 &
szaura >= "0" &
wd_all >= "0" & wd_all_time >= "0" & time_szaura >= "0"
)
#data = data[,!(names(data) %in% drop)]
data$time_szaura <- as.numeric(data$time_szaura)
data$aura_time <- as.numeric(data$aura_time)
data$time_begin <- as.numeric(data$time_begin)
data$years_follow_up <- as.numeric(data$years_follow_up)
data$wd_all_time <- as.numeric(data$wd_all_time)
data$age_onset <- as.numeric(data$age_onset)
data$duration <- as.numeric(data$duration)0.2 Some theory
1 What are competing risks?
When subjects have multiple possible events in a time-to-event setting and that one of them precludes them from having the other.
In our case, we have to assume that seizure relapse precludes participants from completely withdrawing (but not the other way around! -> important when interpreting the results.
2 So, what’s the problem?
Conventional survival analysis (aka. KM) relies on non-informative censoring ->censoring that occurs independently of the risk for the outcome of interest. So, if our main outcome is complete withdrawal, participants who survived (did not complete withdrawal) at the end of follow up will be censored, and this censoring is assumed to be non-informative (aka, that it tells us nothing about the risk)…however, that’s not the case for our scenario: participants that are censored at the end of follow up as “did not complete withdrawal” could include those who had a seizure relapse (and therefore will never complete withdrawal). In this case, censoring IS informative and running a KM here will result in overestimated effects.
3 What to do?
In competing risk modelling, data is described using cumulative incidence functions. Incidence derived from CIFs is the probability of experiencing the primary event (wd_all) conditioned upon not experiencing either event (wd_all or relapse) until that time. The effect of covariates is determined by two different hazard functions: cause-specific and/or Fine-Gray subdistribution.
Cause-specific: Cause-specific hazard (estimate is called Aalen-Johanson) of a given event represents the rate of wd_all per unit of time among those not having had relapse.
Subdistribution hazard function Cumulative incidence (Fine Gray) represents the rate of wd_all among those that have not experienced any event PLUS those that have experienced a competing event (aka, relapse)
4 What to use:
Each of these approaches have practical considerations and give complementary information about the data.
Regression using a cause-specific hazard function will give coefficients that describes the relative effect of a covariate on the relative increase in the rate of wd_all in observations that haven’t relapse. This is ideal when examining causal relationships between risk factors and an event.
Regression using the subdistribution hazard function describes the effect of covariates on the incidence and this may be more in line with predicting the rate of occurrence of events.
Literature says its ideal to use both. But the interpretation is a bit complex.
5 Whats in here?
- Cumulative incidence curves for each predictor and each competing risk.
- Cause-specific analysis
- Subdistribution analysis
5.1 Data set-up
data$age_at_surgery_ten_cat <-
findInterval(data$age_at_surgery_ten, c(0, 2, 4), rightmost.closed = TRUE)
data$duration_ten_cat <-
findInterval(data$duration_ten, c(0, 2), rightmost.closed = TRUE)
data$age_onset_ten_cat <-
findInterval(data$age_onset_ten, c(0, 2, 4), rightmost.closed = TRUE)
data$drugs_cat <-
findInterval(data$drugs, c(0, 3), rightmost.closed = TRUE)
data$time_begin_cat <-
findInterval(data$time_begin, c(0, 2, 4), rightmost.closed = TRUE)
etime <- with(data, ifelse(szaura == 0, wd_all_time, time_szaura))
event <- with(data, ifelse(szaura == 0, 2 * wd_all, 1))
event <- factor(event, 0:2, labels = c("none", "szaura", "wdall"))
table(event)## event
## none szaura wdall
## 301 302 1955.2 Time spent in each status
list <- list(
'external',
'sex',
'febrile_sz',
'learning_disability',
'psychiatric_pre_any',
'gtcs',
'MRI_normal',
'opside',
'optemp',
'as.factor(opextent)',
'op_incomplete',
'pathology_HS',
'pathology_FCD',
'pathology_DNT',
'pathology_CAV',
'pathology_GL',
'pathology_dual',
'pathology_other',
'pathology_normal',
'acutepostszauras',
'auras',
'age_onset_ten_cat',
'age_at_surgery_ten_cat',
'duration_ten_cat',
'extent',
'extra',
'aura_relapse',
'numsz6',
'time_begin_cat',
'drugs_cat'
)
fit_cause_specific <- lapply(list, function(x)
survfit(as.formula(
paste("Surv(as.numeric(etime), event) ~", x)
), data = data))
fit_cause_specific## [[1]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## external=0, (s0) 350 0 8.263520
## external=1, (s0) 448 0 5.943924
## external=0, szaura 350 116 7.270142
## external=1, szaura 448 186 11.930215
## external=0, wdall 350 130 8.113560
## external=1, wdall 448 65 5.773083
## *restricted mean time in state (max time = 23.64722 )
##
## [[2]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## sex=0, (s0) 374 0 7.915595
## sex=1, (s0) 424 0 6.615008
## sex=0, szaura 374 133 8.975504
## sex=1, szaura 424 169 10.356667
## sex=0, wdall 374 93 6.756124
## sex=1, wdall 424 102 6.675548
## *restricted mean time in state (max time = 23.64722 )
##
## [[3]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## 65 observations deleted due to missingness
## n nevent rmean*
## febrile_sz=0, (s0) 446 0 6.684250
## febrile_sz=1, (s0) 287 0 8.065381
## febrile_sz=0, szaura 446 174 10.146971
## febrile_sz=1, szaura 287 105 8.647054
## febrile_sz=0, wdall 446 111 6.816001
## febrile_sz=1, wdall 287 83 6.934787
## *restricted mean time in state (max time = 23.64722 )
##
## [[4]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## 63 observations deleted due to missingness
## n nevent rmean*
## learning_disability=0, (s0) 673 0 7.485711
## learning_disability=1, (s0) 62 0 5.328022
## learning_disability=0, szaura 673 256 9.470245
## learning_disability=1, szaura 62 23 8.908078
## learning_disability=0, wdall 673 173 6.691267
## learning_disability=1, wdall 62 22 9.411122
## *restricted mean time in state (max time = 23.64722 )
##
## [[5]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## 63 observations deleted due to missingness
## n nevent rmean*
## psychiatric_pre_any=0, (s0) 523 0 6.899684
## psychiatric_pre_any=1, (s0) 212 0 8.305492
## psychiatric_pre_any=0, szaura 523 206 9.881985
## psychiatric_pre_any=1, szaura 212 73 8.336334
## psychiatric_pre_any=0, wdall 523 134 6.865553
## psychiatric_pre_any=1, wdall 212 61 7.005396
## *restricted mean time in state (max time = 23.64722 )
##
## [[6]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## 1 observation deleted due to missingness
## n nevent rmean*
## gtcs=0, (s0) 234 0 7.690958
## gtcs=1, (s0) 563 0 7.165633
## gtcs=0, szaura 234 69 7.705062
## gtcs=1, szaura 563 233 10.443566
## gtcs=0, wdall 234 74 8.251202
## gtcs=1, wdall 563 120 6.038023
## *restricted mean time in state (max time = 23.64722 )
##
## [[7]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## MRI_normal=0, (s0) 711 0 7.348438
## MRI_normal=1, (s0) 87 0 7.000697
## MRI_normal=0, szaura 711 266 9.382714
## MRI_normal=1, szaura 87 36 12.367119
## MRI_normal=0, wdall 711 186 6.916070
## MRI_normal=1, wdall 87 9 4.279407
## *restricted mean time in state (max time = 23.64722 )
##
## [[8]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## 1 observation deleted due to missingness
## n nevent rmean*
## opside=0, (s0) 360 0 7.135861
## opside=1, (s0) 429 0 7.510271
## opside=2, (s0) 8 0 2.687500
## opside=0, szaura 360 137 9.739226
## opside=1, szaura 429 164 9.674143
## opside=2, szaura 8 1 2.893403
## opside=0, wdall 360 92 6.772135
## opside=1, wdall 429 98 6.462808
## opside=2, wdall 8 5 18.066319
## *restricted mean time in state (max time = 23.64722 )
##
## [[9]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## optemp=0, (s0) 386 0 6.663424
## optemp=1, (s0) 412 0 7.882302
## optemp=0, szaura 386 155 11.469868
## optemp=1, szaura 412 147 8.234167
## optemp=0, wdall 386 61 5.513930
## optemp=1, wdall 412 134 7.530753
## *restricted mean time in state (max time = 23.64722 )
##
## [[10]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## 131 observations deleted due to missingness
## n nevent rmean*
## as.factor(opextent)=0, (s0) 104 0 6.749081
## as.factor(opextent)=1, (s0) 550 0 7.608952
## as.factor(opextent)=2, (s0) 13 0 7.057949
## as.factor(opextent)=0, szaura 104 33 7.920359
## as.factor(opextent)=1, szaura 550 202 9.106843
## as.factor(opextent)=2, szaura 13 2 3.330342
## as.factor(opextent)=0, wdall 104 38 8.977783
## as.factor(opextent)=1, wdall 550 149 6.931428
## as.factor(opextent)=2, wdall 13 8 13.258932
## *restricted mean time in state (max time = 23.64722 )
##
## [[11]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## 6 observations deleted due to missingness
## n nevent rmean*
## op_incomplete=0, (s0) 708 0 7.572456
## op_incomplete=1, (s0) 84 0 5.976254
## op_incomplete=0, szaura 708 259 9.269423
## op_incomplete=1, szaura 84 42 12.228017
## op_incomplete=0, wdall 708 179 6.805344
## op_incomplete=1, wdall 84 13 5.442951
## *restricted mean time in state (max time = 23.64722 )
##
## [[12]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## pathology_HS=0, (s0) 362 0 6.901347
## pathology_HS=1, (s0) 436 0 7.606000
## pathology_HS=0, szaura 362 131 9.850888
## pathology_HS=1, szaura 436 171 9.487682
## pathology_HS=0, wdall 362 80 6.894987
## pathology_HS=1, wdall 436 115 6.553541
## *restricted mean time in state (max time = 23.64722 )
##
## [[13]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## pathology_FCD=0, (s0) 730 0 7.395765
## pathology_FCD=1, (s0) 68 0 7.115867
## pathology_FCD=0, szaura 730 280 9.710629
## pathology_FCD=1, szaura 68 22 8.089548
## pathology_FCD=0, wdall 730 176 6.540828
## pathology_FCD=1, wdall 68 19 8.441807
## *restricted mean time in state (max time = 23.64722 )
##
## [[14]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## pathology_DNT=0, (s0) 749 0 7.341941
## pathology_DNT=1, (s0) 49 0 7.233879
## pathology_DNT=0, szaura 749 285 9.737981
## pathology_DNT=1, szaura 49 17 8.019535
## pathology_DNT=0, wdall 749 177 6.567300
## pathology_DNT=1, wdall 49 18 8.393809
## *restricted mean time in state (max time = 23.64722 )
##
## [[15]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## 131 observations deleted due to missingness
## n nevent rmean*
## pathology_CAV=0, (s0) 613 0 7.511449
## pathology_CAV=1, (s0) 54 0 6.868634
## pathology_CAV=0, szaura 613 225 9.036434
## pathology_CAV=1, szaura 54 12 5.740363
## pathology_CAV=0, wdall 613 172 7.099339
## pathology_CAV=1, wdall 54 23 11.038226
## *restricted mean time in state (max time = 23.64722 )
##
## [[16]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## 1 observation deleted due to missingness
## n nevent rmean*
## pathology_GL=0, (s0) 776 0 7.413176
## pathology_GL=1, (s0) 21 0 2.763528
## pathology_GL=0, szaura 776 295 9.680229
## pathology_GL=1, szaura 21 6 7.251629
## pathology_GL=0, wdall 776 185 6.553818
## pathology_GL=1, wdall 21 10 13.632065
## *restricted mean time in state (max time = 23.64722 )
##
## [[17]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## 1 observation deleted due to missingness
## n nevent rmean*
## pathology_dual=0, (s0) 761 0 7.341221
## pathology_dual=1, (s0) 36 0 6.164697
## pathology_dual=0, szaura 761 291 9.722327
## pathology_dual=1, szaura 36 10 8.057083
## pathology_dual=0, wdall 761 183 6.583674
## pathology_dual=1, wdall 36 12 9.425443
## *restricted mean time in state (max time = 23.64722 )
##
## [[18]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## pathology_other=0, (s0) 663 0 7.137658
## pathology_other=1, (s0) 135 0 8.948204
## pathology_other=0, szaura 663 254 9.564709
## pathology_other=1, szaura 135 48 9.741912
## pathology_other=0, wdall 663 175 6.944855
## pathology_other=1, wdall 135 20 4.957106
## *restricted mean time in state (max time = 23.64722 )
##
## [[19]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## pathology_normal=0, (s0) 778 0 7.328002
## pathology_normal=1, (s0) 20 0 8.188381
## pathology_normal=0, szaura 778 293 9.549198
## pathology_normal=1, szaura 20 9 14.313147
## pathology_normal=0, wdall 778 194 6.770022
## pathology_normal=1, wdall 20 1 1.145694
## *restricted mean time in state (max time = 23.64722 )
##
## [[20]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## acutepostszauras=0, (s0) 725 0 7.175661
## acutepostszauras=1, (s0) 73 0 8.980822
## acutepostszauras=0, szaura 725 277 9.667697
## acutepostszauras=1, szaura 73 25 9.437234
## acutepostszauras=0, wdall 725 182 6.803864
## acutepostszauras=1, wdall 73 13 5.229166
## *restricted mean time in state (max time = 23.64722 )
##
## [[21]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## 54 observations deleted due to missingness
## n nevent rmean*
## auras=0, (s0) 676 0 7.453676
## auras=1, (s0) 68 0 4.894606
## auras=0, szaura 676 240 8.872702
## auras=1, szaura 68 41 15.950184
## auras=0, wdall 676 190 7.320843
## auras=1, wdall 68 5 2.802432
## *restricted mean time in state (max time = 23.64722 )
##
## [[22]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## age_onset_ten_cat=1, (s0) 557 0 7.379274
## age_onset_ten_cat=2, (s0) 197 0 6.760903
## age_onset_ten_cat=3, (s0) 44 0 11.257718
## age_onset_ten_cat=1, szaura 557 208 9.287885
## age_onset_ten_cat=2, szaura 197 78 10.600300
## age_onset_ten_cat=3, szaura 44 16 9.237207
## age_onset_ten_cat=1, wdall 557 148 6.980063
## age_onset_ten_cat=2, wdall 197 43 6.286019
## age_onset_ten_cat=3, wdall 44 4 3.152298
## *restricted mean time in state (max time = 23.64722 )
##
## [[23]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## age_at_surgery_ten_cat=1, (s0) 59 0 5.619964
## age_at_surgery_ten_cat=2, (s0) 471 0 7.216283
## age_at_surgery_ten_cat=3, (s0) 268 0 8.073811
## age_at_surgery_ten_cat=1, szaura 59 22 10.447979
## age_at_surgery_ten_cat=2, szaura 471 174 8.949416
## age_at_surgery_ten_cat=3, szaura 268 106 10.696615
## age_at_surgery_ten_cat=1, wdall 59 13 7.579279
## age_at_surgery_ten_cat=2, wdall 471 142 7.481523
## age_at_surgery_ten_cat=3, wdall 268 40 4.876796
## *restricted mean time in state (max time = 23.64722 )
##
## [[24]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## duration_ten_cat=1, (s0) 432 0 6.763113
## duration_ten_cat=2, (s0) 366 0 7.785234
## duration_ten_cat=1, szaura 432 147 8.998724
## duration_ten_cat=2, szaura 366 155 10.533246
## duration_ten_cat=1, wdall 432 123 7.885386
## duration_ten_cat=2, wdall 366 72 5.328742
## *restricted mean time in state (max time = 23.64722 )
##
## [[25]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## 131 observations deleted due to missingness
## n nevent rmean*
## extent=0, (s0) 550 0 7.608952
## extent=1, (s0) 104 0 6.749081
## extent=2, (s0) 13 0 7.057949
## extent=0, szaura 550 202 9.106843
## extent=1, szaura 104 33 7.920359
## extent=2, szaura 13 2 3.330342
## extent=0, wdall 550 149 6.931428
## extent=1, wdall 104 38 8.977783
## extent=2, wdall 13 8 13.258932
## *restricted mean time in state (max time = 23.64722 )
##
## [[26]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## extra=0, (s0) 708 0 7.291833
## extra=1, (s0) 90 0 7.203678
## extra=0, szaura 708 275 9.847654
## extra=1, szaura 90 27 8.263164
## extra=0, wdall 708 169 6.507736
## extra=1, wdall 90 26 8.180381
## *restricted mean time in state (max time = 23.64722 )
##
## [[27]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## 2 observations deleted due to missingness
## n nevent rmean*
## aura_relapse=0, (s0) 696 0 8.3046033
## aura_relapse=1, (s0) 99 0 2.4142706
## aura_relapse=N/A, (s0) 1 0 0.6416667
## aura_relapse=0, szaura 696 202 7.4551047
## aura_relapse=1, szaura 99 99 21.2329516
## aura_relapse=N/A, szaura 1 1 23.0055556
## aura_relapse=0, wdall 696 195 7.8875143
## aura_relapse=1, wdall 99 0 0.0000000
## aura_relapse=N/A, wdall 1 0 0.0000000
## *restricted mean time in state (max time = 23.64722 )
##
## [[28]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## 1 observation deleted due to missingness
## n nevent rmean*
## numsz6=0, (s0) 16 0 4.285417
## numsz6=1, (s0) 132 0 7.400687
## numsz6=2, (s0) 411 0 7.562342
## numsz6=3, (s0) 141 0 9.074523
## numsz6=4, (s0) 97 0 5.966578
## numsz6=0, szaura 16 0 0.000000
## numsz6=1, szaura 132 43 7.717190
## numsz6=2, szaura 411 156 9.181181
## numsz6=3, szaura 141 70 12.642736
## numsz6=4, szaura 97 33 9.118449
## numsz6=0, wdall 16 7 19.361806
## numsz6=1, wdall 132 49 8.529345
## numsz6=2, wdall 411 115 6.903699
## numsz6=3, wdall 141 7 1.929963
## numsz6=4, wdall 97 16 8.562195
## *restricted mean time in state (max time = 23.64722 )
##
## [[29]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## time_begin_cat=1, (s0) 548 0 5.845545
## time_begin_cat=2, (s0) 164 0 6.142947
## time_begin_cat=3, (s0) 86 0 14.548390
## time_begin_cat=1, szaura 548 225 10.862214
## time_begin_cat=2, szaura 164 53 8.808714
## time_begin_cat=3, szaura 86 24 5.918486
## time_begin_cat=1, wdall 548 127 6.939463
## time_begin_cat=2, wdall 164 55 8.695561
## time_begin_cat=3, wdall 86 13 3.180346
## *restricted mean time in state (max time = 23.64722 )
##
## [[30]]
## Call: survfit(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data)
##
## n nevent rmean*
## drugs_cat=1, (s0) 739 0 7.395729
## drugs_cat=2, (s0) 59 0 6.288982
## drugs_cat=1, szaura 739 274 9.431099
## drugs_cat=2, szaura 59 28 12.300759
## drugs_cat=1, wdall 739 185 6.820394
## drugs_cat=2, wdall 59 10 5.057481
## *restricted mean time in state (max time = 23.64722 )5.3 Cumulative incidence p values
library(cmprsk)
cif_external <- cuminc(etime, event, group = data$external)
cif_external## Tests:
## stat pv df
## none 28.28188 1.048732e-07 1
## szaura 15.52977 8.121619e-05 1
## wdall 48.24037 3.770428e-12 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.09142857 0.1800000 0.2371429 0.2771429
## 1 none 0.30803571 0.4330357 0.4375000 NA
## 0 szaura 0.25428571 0.3114286 0.3285714 0.3314286
## 1 szaura 0.39732143 0.4129464 0.4151786 NA
## 0 wdall 0.28285714 0.3600000 0.3714286 0.3714286
## 1 wdall 0.12723214 0.1406250 0.1450893 NA
##
## $var
## 5 10 15 20
## 0 none 0.0002373621 0.0004199459 0.0005134052 0.0005700476
## 1 none 0.0004775302 0.0005522391 0.0005543608 NA
## 0 szaura 0.0005350768 0.0005959618 0.0006108939 0.0006135251
## 1 szaura 0.0005359347 0.0005437000 0.0005476395 NA
## 0 wdall 0.0005660753 0.0006339709 0.0006413191 0.0006413191
## 1 wdall 0.0002485493 0.0002705167 0.0002801222 NAcif_sex <- cuminc(etime, event, group = data$sex)
cif_sex## Tests:
## stat pv df
## none 0.9961613 0.3182412 1
## szaura 1.9119600 0.1667457 1
## wdall 0.1187791 0.7303625 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.2219251 0.3262032 0.3529412 0.3850267
## 1 none 0.2051887 0.3183962 0.3466981 0.3537736
## 0 szaura 0.3181818 0.3502674 0.3556150 0.3556150
## 1 szaura 0.3490566 0.3844340 0.3962264 0.3985849
## 0 wdall 0.2058824 0.2379679 0.2486631 0.2486631
## 1 wdall 0.1863208 0.2358491 0.2405660 0.2405660
##
## $var
## 5 10 15 20
## 0 none 0.0004621947 0.0005872023 0.0006104933 0.0006354285
## 1 none 0.0003856131 0.0005128402 0.0005354777 0.0005414510
## 0 szaura 0.0005786349 0.0006045271 0.0006085098 0.0006085098
## 1 szaura 0.0005352355 0.0005555939 0.0005617644 0.0005638426
## 0 wdall 0.0004327646 0.0004780284 0.0004921061 0.0004921061
## 1 wdall 0.0003554203 0.0004205369 0.0004264021 0.0004264021cif_febrile_sz <- cuminc(etime, event, group = data$febrile_sz)## 65 cases omitted due to missing valuescif_febrile_sz## Tests:
## stat pv df
## none 1.3007132 0.2540830 1
## szaura 0.7757101 0.3784564 1
## wdall 0.9960601 0.3182657 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.2130045 0.3251121 0.3452915 0.3542601
## 1 none 0.1324042 0.2473868 0.2926829 0.3310105
## 0 szaura 0.3408072 0.3834081 0.3901345 0.3901345
## 1 szaura 0.3205575 0.3484321 0.3623693 0.3658537
## 0 wdall 0.2062780 0.2443946 0.2488789 0.2488789
## 1 wdall 0.2195122 0.2752613 0.2891986 0.2891986
##
## $var
## 5 10 15 20
## 0 none 0.0003762470 0.0004922714 0.0005078810 0.0005147182
## 1 none 0.0004011818 0.0006485100 0.0007213410 0.0007748094
## 0 szaura 0.0005022611 0.0005263615 0.0005302039 0.0005302039
## 1 szaura 0.0007571406 0.0007858122 0.0007987566 0.0008029198
## 0 wdall 0.0003640813 0.0004092481 0.0004142601 0.0004142601
## 1 wdall 0.0005909751 0.0006830257 0.0007035401 0.0007035401cif_learning_disability <-cuminc(etime, event, group = data$learning_disability)## 63 cases omitted due to missing valuescif_learning_disability## Tests:
## stat pv df
## none 0.5033108818 0.47804895 1
## szaura 0.0002223846 0.98810194 1
## wdall 3.8118442121 0.05089139 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.1797920 0.3001486 0.3313522 0.3521545
## 1 none 0.1935484 0.2419355 0.2580645 NA
## 0 szaura 0.3283804 0.3684993 0.3789004 0.3803863
## 1 szaura 0.3709677 0.3709677 0.3709677 NA
## 0 wdall 0.2035661 0.2481426 0.2570579 0.2570579
## 1 wdall 0.3064516 0.3548387 0.3548387 NA
##
## $var
## 5 10 15 20
## 0 none 0.0002190481 0.0003112593 0.0003280103 0.0003381656
## 1 none 0.0025945126 0.0030615158 0.0032611231 NA
## 0 szaura 0.0003266025 0.0003429767 0.0003464925 0.0003470632
## 1 szaura 0.0037660605 0.0037660605 0.0037660605 NA
## 0 wdall 0.0002386103 0.0002731194 0.0002792936 0.0002792936
## 1 wdall 0.0034091299 0.0036671430 0.0036671430 NAcif_psychiatric_pre_any <-cuminc(etime, event, group = data$psychiatric_pre_any)## 63 cases omitted due to missing valuescif_psychiatric_pre_any## Tests:
## stat pv df
## none 0.03533851 0.85088809 1
## szaura 2.88299020 0.08951964 1
## wdall 0.42776432 0.51308800 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.1950287 0.2982792 0.3212237 0.3422562
## 1 none 0.1462264 0.2877358 0.3349057 0.3537736
## 0 szaura 0.3537285 0.3843212 0.3919694 0.3938815
## 1 szaura 0.2783019 0.3301887 0.3443396 0.3443396
## 0 wdall 0.2122371 0.2485660 0.2562141 0.2562141
## 1 wdall 0.2122642 0.2783019 0.2877358 0.2877358
##
## $var
## 5 10 15 20
## 0 none 0.0003003808 0.0003998162 0.0004167715 0.0004312275
## 1 none 0.0005893003 0.0009679672 0.0010549253 0.0010885175
## 0 szaura 0.0004357239 0.0004494140 0.0004525335 0.0004534516
## 1 szaura 0.0009442683 0.0010318308 0.0010545815 0.0010545815
## 0 wdall 0.0003168733 0.0003525010 0.0003595015 0.0003595015
## 1 wdall 0.0007780157 0.0009282593 0.0009471950 0.0009471950cif_gtcs <- cuminc(etime, event, group = data$gtcs)## 1 cases omitted due to missing valuescif_gtcs## Tests:
## stat pv df
## none 0.08375214 0.7722755913 1
## szaura 11.57698620 0.0006677308 1
## wdall 10.91913122 0.0009517623 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.2222222 0.3290598 0.3632479 0.3760684
## 1 none 0.2095915 0.3197158 0.3445826 0.3658970
## 0 szaura 0.2521368 0.2735043 0.2905983 0.2948718
## 1 szaura 0.3694494 0.4085258 0.4138544 0.4138544
## 0 wdall 0.2649573 0.3119658 0.3162393 0.3162393
## 1 wdall 0.1651865 0.2042629 0.2131439 0.2131439
##
## $var
## 5 10 15 20
## 0 none 0.0007420877 0.0009503445 0.0009969207 0.0010143975
## 1 none 0.0002945196 0.0003856937 0.0004005730 0.0004126406
## 0 szaura 0.0008055425 0.0008473955 0.0008802977 0.0008892158
## 1 szaura 0.0004127011 0.0004261443 0.0004277073 0.0004277073
## 0 wdall 0.0008288068 0.0009109175 0.0009180611 0.0009180611
## 1 wdall 0.0002425932 0.0002843818 0.0002932602 0.0002932602cif_MRI_normal <- cuminc(etime, event, group = data$MRI_normal)
cif_MRI_normal## Tests:
## stat pv df
## none 3.982114 0.045985821 1
## szaura 0.744839 0.388115163 1
## wdall 7.911730 0.004911528 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.18987342 0.3037975 0.3347398 0.3544304
## 1 none 0.40229885 0.4712644 0.4712644 NA
## 0 szaura 0.32770745 0.3628692 0.3727145 0.3741210
## 1 szaura 0.39080460 0.4137931 0.4137931 NA
## 0 wdall 0.20956399 0.2531646 0.2616034 0.2616034
## 1 wdall 0.08045977 0.1034483 0.1034483 NA
##
## $var
## 5 10 15 20
## 0 none 0.0002162461 0.0002964800 0.0003119170 0.0003208549
## 1 none 0.0028501487 0.0030476120 0.0030476120 NA
## 0 szaura 0.0003086015 0.0003222186 0.0003254643 0.0003259776
## 1 szaura 0.0027911245 0.0028943304 0.0028943304 NA
## 0 wdall 0.0002302679 0.0002613018 0.0002666854 0.0002666854
## 1 wdall 0.0008754202 0.0011209221 0.0011209221 NAcif_opside <- cuminc(etime, event, group = data$opside)## 1 cases omitted due to missing valuescif_opside## Tests:
## stat pv df
## none 0.4475162 0.799508527 2
## szaura 1.4487781 0.484620548 2
## wdall 13.0179296 0.001490021 2
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.2111111 0.3083333 0.3416667 0.3611111
## 1 none 0.2121212 0.3333333 0.3566434 0.3752914
## 2 none NA NA NA NA
## 0 szaura 0.3250000 0.3694444 0.3805556 0.3805556
## 1 szaura 0.3473193 0.3729604 0.3799534 0.3822844
## 2 szaura NA NA NA NA
## 0 wdall 0.2055556 0.2500000 0.2555556 0.2555556
## 1 wdall 0.1794872 0.2191142 0.2284382 0.2284382
## 2 wdall NA NA NA NA
##
## $var
## 5 10 15 20
## 0 none 0.0004643074 0.0005948617 0.0006281004 0.0006481914
## 1 none 0.0003900141 0.0005179282 0.0005346307 0.0005473575
## 2 none NA NA NA NA
## 0 szaura 0.0006084420 0.0006432247 0.0006509509 0.0006509509
## 1 szaura 0.0005274183 0.0005426534 0.0005465445 0.0005481046
## 2 szaura NA NA NA NA
## 0 wdall 0.0004508362 0.0005152075 0.0005227880 0.0005227880
## 1 wdall 0.0003404399 0.0003939414 0.0004055859 0.0004055859
## 2 wdall NA NA NA NAcif_optemp <- cuminc(etime, event, group = data$optemp)
cif_optemp## Tests:
## stat pv df
## none 8.950958 2.773240e-03 1
## szaura 5.999819 1.430735e-02 1
## wdall 26.778831 2.281226e-07 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.3056995 0.4300518 0.4352332 0.4378238
## 1 none 0.1262136 0.2208738 0.2694175 0.3033981
## 0 szaura 0.3860104 0.3989637 0.4015544 0.4015544
## 1 szaura 0.2864078 0.3398058 0.3543689 0.3567961
## 0 wdall 0.1373057 0.1554404 0.1580311 0.1580311
## 1 wdall 0.2500000 0.3131068 0.3252427 0.3252427
##
## $var
## 5 10 15 20
## 0 none 0.0005528880 0.0006415906 0.0006445088 0.0006476948
## 1 none 0.0002671882 0.0004148288 0.0004734607 0.0005095748
## 0 szaura 0.0006160945 0.0006247504 0.0006285583 0.0006285583
## 1 szaura 0.0004915518 0.0005330540 0.0005421320 0.0005437910
## 0 wdall 0.0003081383 0.0003416034 0.0003473068 0.0003473068
## 1 wdall 0.0004460407 0.0005058336 0.0005152582 0.0005152582cif_opextent <- cuminc(etime, event, group = as.factor(data$opextent))## 131 cases omitted due to missing valuescif_opextent## Tests:
## stat pv df
## none 1.405013 0.495342221 2
## szaura 3.007815 0.222259955 2
## wdall 13.529635 0.001153658 2
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.18269231 0.26923077 0.2884615 0.3076923
## 1 none 0.17272727 0.29454545 0.3290909 0.3527273
## 2 none 0.07692308 0.07692308 0.1538462 0.1538462
## 0 szaura 0.26923077 0.31730769 0.3173077 0.3173077
## 1 szaura 0.31272727 0.35272727 0.3654545 0.3672727
## 2 szaura 0.15384615 0.15384615 0.1538462 0.1538462
## 0 wdall 0.33653846 0.36538462 0.3653846 0.3653846
## 1 wdall 0.21090909 0.26000000 0.2709091 0.2709091
## 2 wdall 0.38461538 0.61538462 0.6153846 0.6153846
##
## $var
## 5 10 15 20
## 0 none 0.0014560714 0.0019424832 0.0020448955 0.0021477504
## 1 none 0.0002598614 0.0003765867 0.0003998727 0.0004144389
## 2 none 0.0065071147 0.0065071147 0.0145294449 0.0145294449
## 0 szaura 0.0019009372 0.0020809437 0.0020809437 0.0020809437
## 1 szaura 0.0003891826 0.0004112616 0.0004172126 0.0004182172
## 2 szaura 0.0107248521 0.0107248521 0.0107248521 0.0107248521
## 0 wdall 0.0021444880 0.0022275226 0.0022275226 0.0022275226
## 1 wdall 0.0002992269 0.0003439346 0.0003527225 0.0003527225
## 2 wdall 0.0196005917 0.0208678501 0.0208678501 0.0208678501cif_op_incomplete <-cuminc(etime, event, group = data$op_incomplete)## 6 cases omitted due to missing valuescif_op_incomplete## Tests:
## stat pv df
## none 0.4925243 0.4828034308 1
## szaura 12.5094831 0.0004048916 1
## wdall 2.7310401 0.0984151729 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.2104520 0.3192090 0.3502825 0.3714689
## 1 none 0.2380952 NA NA NA
## 0 szaura 0.3177966 0.3559322 0.3644068 0.3658192
## 1 szaura 0.5000000 NA NA NA
## 0 wdall 0.2005650 0.2443503 0.2528249 0.2528249
## 1 wdall 0.1309524 NA NA NA
##
## $var
## 5 10 15 20
## 0 none 0.0002345050 0.0003059418 0.0003201672 0.0003288399
## 1 none 0.0022309985 NA NA NA
## 0 szaura 0.0003049483 0.0003207755 0.0003237927 0.0003243494
## 1 szaura 0.0030366855 NA NA NA
## 0 wdall 0.0002240710 0.0002564887 0.0002621323 0.0002621323
## 1 wdall 0.0014045631 NA NA NAcif_pathology_HS <- cuminc(etime, event, group = data$pathology_HS)
cif_pathology_HS## Tests:
## stat pv df
## none 7.0435268 0.007955223 1
## szaura 0.2123191 0.644955794 1
## wdall 1.2710312 0.259573318 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.2817680 0.3895028 0.4033149 0.4116022
## 1 none 0.1559633 0.2660550 0.3050459 0.3325688
## 0 szaura 0.3342541 0.3591160 0.3618785 0.3618785
## 1 szaura 0.3348624 0.3761468 0.3899083 0.3922018
## 0 wdall 0.1906077 0.2182320 0.2209945 0.2209945
## 1 wdall 0.1995413 0.2522936 0.2637615 0.2637615
##
## $var
## 5 10 15 20
## 0 none 0.0005608081 0.0006605853 0.0006702000 0.0006762438
## 1 none 0.0003021369 0.0004469157 0.0004848716 0.0005090455
## 0 szaura 0.0006145169 0.0006348175 0.0006376935 0.0006376935
## 1 szaura 0.0005096405 0.0005341433 0.0005407775 0.0005420956
## 0 wdall 0.0004247877 0.0004692197 0.0004737956 0.0004737956
## 1 wdall 0.0003629926 0.0004261913 0.0004384268 0.0004384268cif_pathology_FCD <- cuminc(etime, event, group = data$pathology_FCD)
cif_pathology_FCD## Tests:
## stat pv df
## none 2.2953072 0.1297655 1
## szaura 0.4901014 0.4838821 1
## wdall 1.2647929 0.2607458 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.2082192 0.3150685 0.3452055 0.3657534
## 1 none 0.2647059 NA NA NA
## 0 szaura 0.3369863 0.3726027 0.3821918 0.3835616
## 1 szaura 0.3088235 NA NA NA
## 0 wdall 0.1904110 0.2328767 0.2410959 0.2410959
## 1 wdall 0.2500000 NA NA NA
##
## $var
## 5 10 15 20
## 0 none 0.0002258182 0.0002947382 0.0003084736 0.0003169444
## 1 none 0.0029309517 NA NA NA
## 0 szaura 0.0003051094 0.0003177473 0.0003206029 0.0003210621
## 1 szaura 0.0031882835 NA NA NA
## 0 wdall 0.0002091393 0.0002409473 0.0002465205 0.0002465205
## 1 wdall 0.0027935987 NA NA NAcif_pathology_DNT <- cuminc(etime, event, group = data$pathology_DNT)
cif_pathology_DNT## Tests:
## stat pv df
## none 1.6516595 0.1987332 1
## szaura 0.6662634 0.4143574 1
## wdall 4.1079170 0.0426829 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.2202937 0.3284379 0.3551402 0.3738318
## 1 none 0.1020408 0.2244898 0.2653061 NA
## 0 szaura 0.3391188 0.3711615 0.3791722 0.3805073
## 1 szaura 0.2653061 0.3265306 0.3469388 NA
## 0 wdall 0.1882510 0.2296395 0.2363151 0.2363151
## 1 wdall 0.3061224 0.3469388 0.3673469 NA
##
## $var
## 5 10 15 20
## 0 none 0.0002291830 0.0002934866 0.0003045285 0.0003116585
## 1 none 0.0019297080 0.0037985390 0.0043864797 NA
## 0 szaura 0.0002983027 0.0003095100 0.0003119071 0.0003123726
## 1 szaura 0.0040132987 0.0044809460 0.0047292936 NA
## 0 wdall 0.0002022208 0.0002328923 0.0002374171 0.0002374171
## 1 wdall 0.0043269400 0.0046278637 0.0047802859 NAcif_pathology_CAV <-cuminc(etime, event, group = data$pathology_CAV)## 131 cases omitted due to missing valuescif_pathology_CAV## Tests:
## stat pv df
## none 0.003055019 0.95592159 1
## szaura 4.786255567 0.02868773 1
## wdall 6.147334701 0.01316103 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.1696574 0.2838499 0.3181077 0.3425775
## 1 none 0.2037037 0.3148148 0.3333333 0.3333333
## 0 szaura 0.3132137 0.3556281 0.3654160 0.3670473
## 1 szaura 0.1851852 0.2037037 0.2222222 0.2222222
## 0 wdall 0.2251223 0.2707993 0.2805873 0.2805873
## 1 wdall 0.3333333 0.4259259 0.4259259 0.4259259
##
## $var
## 5 10 15 20
## 0 none 0.0002295739 0.0003301404 0.0003520504 0.0003660921
## 1 none 0.0031004558 0.0042188532 0.0044198718 0.0044198718
## 0 szaura 0.0003489030 0.0003693205 0.0003732833 0.0003740339
## 1 szaura 0.0028548356 0.0030851795 0.0034648823 0.0034648823
## 0 wdall 0.0002806907 0.0003157924 0.0003224586 0.0003224586
## 1 wdall 0.0041990861 0.0046910702 0.0046910702 0.0046910702cif_pathology_GL <- cuminc(etime, event, group = data$pathology_GL)## 1 cases omitted due to missing valuescif_pathology_GL## Tests:
## stat pv df
## none 0.04326611 0.8352250135 1
## szaura 0.45225154 0.5012676667 1
## wdall 11.53884254 0.0006815706 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.2126289 0.3247423 0.3530928 0.3724227
## 1 none 0.2380952 NA NA NA
## 0 szaura 0.3363402 0.3698454 0.3788660 0.3801546
## 1 szaura 0.2380952 NA NA NA
## 0 wdall 0.1881443 0.2306701 0.2384021 0.2384021
## 1 wdall 0.4761905 NA NA NA
##
## $var
## 5 10 15 20
## 0 none 0.0002156048 0.0002816324 0.0002931677 0.0003003056
## 1 none 0.0098640188 NA NA NA
## 0 szaura 0.0002866107 0.0002980077 0.0003006238 0.0003010437
## 1 szaura 0.0092354057 NA NA NA
## 0 wdall 0.0001949085 0.0002251842 0.0002301770 0.0002301770
## 1 wdall 0.0135823768 NA NA NAcif_pathology_dual <-cuminc(etime, event, group = data$pathology_dual)## 1 cases omitted due to missing valuescif_pathology_dual## Tests:
## stat pv df
## none 0.2566191 0.6124525 1
## szaura 1.4535546 0.2279584 1
## wdall 1.7681102 0.1836161 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.2128778 0.3193167 0.3482260 0.3679369
## 1 none 0.2222222 0.3888889 0.3888889 NA
## 0 szaura 0.3377135 0.3731932 0.3810775 0.3823916
## 1 szaura 0.2500000 0.2500000 0.2777778 NA
## 0 wdall 0.1931669 0.2325887 0.2404731 0.2404731
## 1 wdall 0.2500000 0.3333333 0.3333333 NA
##
## $var
## 5 10 15 20
## 0 none 0.0002201747 0.0002848183 0.0002972438 0.0003049260
## 1 none 0.0050043739 0.0074009580 0.0074009580 NA
## 0 szaura 0.0002928358 0.0003048598 0.0003071122 0.0003075424
## 1 szaura 0.0053627887 0.0053627887 0.0069665721 NA
## 0 wdall 0.0002027476 0.0002309221 0.0002360784 0.0002360784
## 1 wdall 0.0053807631 0.0065486383 0.0065486383 NAcif_pathology_other <-cuminc(etime, event, group = data$pathology_other)
cif_pathology_other## Tests:
## stat pv df
## none 13.9605175 0.0001866903 1
## szaura 0.1114865 0.7384580609 1
## wdall 7.1476742 0.0075061264 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.1885370 0.2941176 0.3242836 0.3438914
## 1 none 0.3333333 0.4592593 0.4740741 0.4888889
## 0 szaura 0.3333333 0.3710407 0.3815988 0.3831071
## 1 szaura 0.3407407 0.3555556 0.3555556 0.3555556
## 0 wdall 0.2111614 0.2549020 0.2639517 0.2639517
## 1 wdall 0.1185185 0.1481481 0.1481481 0.1481481
##
## $var
## 5 10 15 20
## 0 none 0.0002307407 0.0003123318 0.0003295007 0.0003397440
## 1 none 0.0016681768 0.0018790293 0.0018923067 0.0019136952
## 0 szaura 0.0003337980 0.0003486264 0.0003521057 0.0003526652
## 1 szaura 0.0016801782 0.0017211740 0.0017211740 0.0017211740
## 0 wdall 0.0002482330 0.0002813832 0.0002875764 0.0002875764
## 1 wdall 0.0007826620 0.0009519109 0.0009519109 0.0009519109cif_pathology_normal <- cuminc(etime, event, group = data$pathology_normal)
cif_pathology_normal## Tests:
## stat pv df
## none 8.438626 0.003673337 1
## szaura 1.106452 0.292854399 1
## wdall 2.458371 0.116899966 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.2069409 0.3174807 0.3457584 0.3650386
## 1 none 0.4500000 NA NA NA
## 0 szaura 0.3329049 0.3663239 0.3753213 0.3766067
## 1 szaura 0.4000000 NA NA NA
## 0 wdall 0.1992288 0.2416452 0.2493573 0.2493573
## 1 wdall 0.0500000 NA NA NA
##
## $var
## 5 10 15 20
## 0 none 0.0002108256 0.0002775972 0.0002895995 0.0002970653
## 1 none 0.0144619842 NA NA NA
## 0 szaura 0.0002843015 0.0002958446 0.0002985019 0.0002989276
## 1 szaura 0.0130234748 NA NA NA
## 0 wdall 0.0002028756 0.0002317028 0.0002364363 0.0002364363
## 1 wdall 0.0025569236 NA NA NAcif_acutepostszauras <-cuminc(etime, event, group = data$acutepostszauras)
cif_acutepostszauras## Tests:
## stat pv df
## none 4.9301203 0.02639248 1
## szaura 0.1624343 0.68692516 1
## wdall 1.8144919 0.17797032 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.2041379 0.3117241 0.3393103 0.3572414
## 1 none 0.3013699 0.4246575 0.4520548 NA
## 0 szaura 0.3379310 0.3724138 0.3806897 0.3820690
## 1 szaura 0.3013699 0.3287671 0.3424658 NA
## 0 wdall 0.2000000 0.2427586 0.2510345 0.2510345
## 1 wdall 0.1506849 0.1780822 0.1780822 NA
##
## $var
## 5 10 15 20
## 0 none 0.0002239875 0.0002949061 0.0003080217 0.0003159068
## 1 none 0.0029589925 0.0034865694 0.0035823061 NA
## 0 szaura 0.0003073566 0.0003195492 0.0003220563 0.0003225327
## 1 szaura 0.0029353229 0.0031020766 0.0031945838 NA
## 0 wdall 0.0002183388 0.0002493120 0.0002547480 0.0002547480
## 1 wdall 0.0017876906 0.0020669300 0.0020669300 NAcif_auras <- cuminc(etime, event, group = data$auras)## 54 cases omitted due to missing valuescif_auras## Tests:
## stat pv df
## none 0.0614198 8.042658e-01 1
## szaura 18.4924177 1.705816e-05 1
## wdall 12.7401823 3.578828e-04 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.18786982 0.30029586 0.33284024 0.35502959
## 1 none 0.20588235 0.30882353 0.30882353 0.30882353
## 0 szaura 0.30769231 0.34319527 0.35355030 0.35502959
## 1 szaura 0.55882353 0.60294118 0.60294118 0.60294118
## 0 wdall 0.22633136 0.27218935 0.28106509 0.28106509
## 1 wdall 0.04411765 0.07352941 0.07352941 0.07352941
##
## $var
## 5 10 15 20
## 0 none 0.0002254866 0.0003094891 0.0003268088 0.0003374884
## 1 none 0.0025070818 0.0033449868 0.0033449868 0.0033449868
## 0 szaura 0.0003138441 0.0003304676 0.0003346804 0.0003353509
## 1 szaura 0.0037051172 0.0037103535 0.0037103535 0.0037103535
## 0 wdall 0.0002561823 0.0002881831 0.0002936522 0.0002936522
## 1 wdall 0.0006343149 0.0011646965 0.0011646965 0.0011646965cif_age_onset_ten_cat <-cuminc(etime, event, group = data$age_onset_ten_cat)
cif_age_onset_ten_cat## Tests:
## stat pv df
## none 14.930471 0.0005726503 2
## szaura 0.820040 0.6636369742 2
## wdall 6.188287 0.0453138138 2
## Estimates and Variances:
## $est
## 5 10 15 20
## 1 none 0.19569120 0.2962298 0.3249551 0.3500898
## 2 none 0.22842640 0.3451777 0.3756345 0.3807107
## 3 none 0.36363636 NA NA NA
## 1 szaura 0.32854578 0.3608618 0.3716338 0.3734291
## 2 szaura 0.34517766 0.3908629 0.3959391 0.3959391
## 3 szaura 0.36363636 NA NA NA
## 1 wdall 0.20825853 0.2567325 0.2657092 0.2657092
## 2 wdall 0.18274112 0.2131980 0.2182741 0.2182741
## 3 wdall 0.09090909 NA NA NA
##
## $var
## 5 10 15 20
## 1 none 0.0002825292 0.0003730617 0.0003921985 0.0004074828
## 2 none 0.0009048382 0.0011659881 0.0012156831 0.0012310384
## 3 none 0.0055536959 NA NA NA
## 1 szaura 0.0003940972 0.0004098287 0.0004144087 0.0004152770
## 2 szaura 0.0011524801 0.0012164226 0.0012302635 0.0012302635
## 3 szaura 0.0054635284 NA NA NA
## 1 wdall 0.0002925049 0.0003358969 0.0003430267 0.0003430267
## 2 wdall 0.0007607285 0.0008554457 0.0008732645 0.0008732645
## 3 wdall 0.0020080087 NA NA NAcif_age_at_surgery_ten_cat <- cuminc(etime, event, group = data$age_at_surgery_ten_cat)
cif_age_at_surgery_ten_cat## Tests:
## stat pv df
## none 13.7853937 0.0010151724 2
## szaura 0.7227131 0.6967305391 2
## wdall 21.3510527 0.0000231035 2
## Estimates and Variances:
## $est
## 5 10 15 20
## 1 none 0.3220339 0.3898305 0.4067797 NA
## 2 none 0.1613588 0.2547771 0.2887473 0.3163482
## 3 none 0.2798507 0.4253731 0.4440299 0.4514925
## 1 szaura 0.3559322 0.3559322 0.3559322 NA
## 2 szaura 0.3184713 0.3566879 0.3694268 0.3694268
## 3 szaura 0.3582090 0.3917910 0.3955224 0.3955224
## 1 wdall 0.1694915 0.2203390 0.2203390 NA
## 2 wdall 0.2377919 0.2929936 0.3014862 0.3014862
## 3 wdall 0.1268657 0.1417910 0.1492537 0.1492537
##
## $var
## 5 10 15 20
## 1 none 0.0038372661 0.0043303714 0.0044797717 NA
## 2 none 0.0002873240 0.0004024634 0.0004350556 0.0004593263
## 3 none 0.0007562875 0.0009194410 0.0009314573 0.0009388803
## 1 szaura 0.0039734241 0.0039734241 0.0039734241 NA
## 2 szaura 0.0004590411 0.0004826365 0.0004894206 0.0004894206
## 3 szaura 0.0008598815 0.0008914254 0.0008967679 0.0008967679
## 1 wdall 0.0024549319 0.0030747428 0.0030747428 NA
## 2 wdall 0.0003807056 0.0004323768 0.0004392462 0.0004392462
## 3 wdall 0.0004136289 0.0004543684 0.0004763145 0.0004763145cif_duration_ten_cat <- cuminc(etime, event, group = data$duration_ten_cat)
cif_duration_ten_cat## Tests:
## stat pv df
## none 0.004334304 0.947508847 1
## szaura 4.287475003 0.038394145 1
## wdall 8.970386939 0.002743905 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 1 none 0.2199074 0.3333333 0.3611111 0.3680556
## 2 none 0.2049180 0.3087432 0.3360656 0.3688525
## 1 szaura 0.3078704 0.3310185 0.3379630 0.3402778
## 2 szaura 0.3661202 0.4125683 0.4234973 0.4234973
## 1 wdall 0.2384259 0.2824074 0.2847222 0.2847222
## 2 wdall 0.1448087 0.1830601 0.1967213 0.1967213
##
## $var
## 5 10 15 20
## 1 none 0.0003976487 0.0005152492 0.0005357483 0.0005406381
## 2 none 0.0004462833 0.0005832150 0.0006095689 0.0006383776
## 1 szaura 0.0004924437 0.0005108098 0.0005166460 0.0005191995
## 2 szaura 0.0006338097 0.0006588101 0.0006632498 0.0006632498
## 1 wdall 0.0004165558 0.0004635786 0.0004659554 0.0004659554
## 2 wdall 0.0003368576 0.0004046275 0.0004270968 0.0004270968cif_extent <- cuminc(etime, event, group = data$extent)## 131 cases omitted due to missing valuescif_extent## Tests:
## stat pv df
## none 1.405013 0.495342221 2
## szaura 3.007815 0.222259955 2
## wdall 13.529635 0.001153658 2
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.17272727 0.29454545 0.3290909 0.3527273
## 1 none 0.18269231 0.26923077 0.2884615 0.3076923
## 2 none 0.07692308 0.07692308 0.1538462 0.1538462
## 0 szaura 0.31272727 0.35272727 0.3654545 0.3672727
## 1 szaura 0.26923077 0.31730769 0.3173077 0.3173077
## 2 szaura 0.15384615 0.15384615 0.1538462 0.1538462
## 0 wdall 0.21090909 0.26000000 0.2709091 0.2709091
## 1 wdall 0.33653846 0.36538462 0.3653846 0.3653846
## 2 wdall 0.38461538 0.61538462 0.6153846 0.6153846
##
## $var
## 5 10 15 20
## 0 none 0.0002598614 0.0003765867 0.0003998727 0.0004144389
## 1 none 0.0014560714 0.0019424832 0.0020448955 0.0021477504
## 2 none 0.0065071147 0.0065071147 0.0145294449 0.0145294449
## 0 szaura 0.0003891826 0.0004112616 0.0004172126 0.0004182172
## 1 szaura 0.0019009372 0.0020809437 0.0020809437 0.0020809437
## 2 szaura 0.0107248521 0.0107248521 0.0107248521 0.0107248521
## 0 wdall 0.0002992269 0.0003439346 0.0003527225 0.0003527225
## 1 wdall 0.0021444880 0.0022275226 0.0022275226 0.0022275226
## 2 wdall 0.0196005917 0.0208678501 0.0208678501 0.0208678501cif_extra <- cuminc(etime, event, group = data$extra)
cif_extra## Tests:
## stat pv df
## none 0.6500572 0.4200922 1
## szaura 3.1551830 0.0756862 1
## wdall 1.5415363 0.2143893 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.2019774 0.3163842 0.3446328 0.3658192
## 1 none 0.3000000 0.3666667 0.3888889 0.3888889
## 0 szaura 0.3432203 0.3771186 0.3870056 0.3884181
## 1 szaura 0.2666667 0.3000000 0.3000000 0.3000000
## 0 wdall 0.1892655 0.2302260 0.2387006 0.2387006
## 1 wdall 0.2444444 0.2888889 0.2888889 0.2888889
##
## $var
## 5 10 15 20
## 0 none 0.0002277453 0.0003049288 0.0003182230 0.0003273294
## 1 none 0.0023539336 0.0026282772 0.0027233463 0.0027233463
## 0 szaura 0.0003174142 0.0003295026 0.0003324847 0.0003329849
## 1 szaura 0.0021847138 0.0023488995 0.0023488995 0.0023488995
## 0 wdall 0.0002147314 0.0002468316 0.0002528800 0.0002528800
## 1 wdall 0.0020640771 0.0022970574 0.0022970574 0.0022970574cif_aura_relapse <- cuminc(etime, event, group = data$aura_relapse)## 2 cases omitted due to missing valuescif_aura_relapse## Tests:
## stat pv df
## none 29.71844 3.521460e-07 2
## szaura 266.15749 0.000000e+00 2
## wdall 30.77503 2.076284e-07 2
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.2428161 0.3663793 0.3979885 0.4195402
## 1 none 0.0000000 0.0000000 0.0000000 NA
## N/A none NA NA NA NA
## 0 szaura 0.2586207 0.2859195 0.2902299 0.2902299
## 1 szaura 0.8686869 0.9494949 0.9898990 NA
## N/A szaura NA NA NA NA
## 0 wdall 0.2241379 0.2715517 0.2801724 0.2801724
## 1 wdall 0.0000000 0.0000000 0.0000000 NA
## N/A wdall NA NA NA NA
##
## $var
## 5 10 15 20
## 0 none 0.0002640172 0.0003324211 0.0003428708 0.0003491022
## 1 none 0.0000000000 0.0000000000 0.0000000000 NA
## N/A none NA NA NA NA
## 0 szaura 0.0002747781 0.0002915346 0.0002940291 0.0002940291
## 1 szaura 0.0012016215 0.0005364062 0.0001667079 NA
## N/A szaura NA NA NA NA
## 0 wdall 0.0002475771 0.0002800800 0.0002852853 0.0002852853
## 1 wdall 0.0000000000 0.0000000000 0.0000000000 NA
## N/A wdall NA NA NA NAcif_numsz6 <- cuminc(etime, event, group = data$numsz6)## 1 cases omitted due to missing valuescif_numsz6## Tests:
## stat pv df
## none 39.17272 6.416981e-08 4
## szaura 28.46667 1.003080e-05 4
## wdall 43.07731 9.972606e-09 4
## Estimates and Variances:
## $est
## 5 10 15 20
## 0 none 0.56250000 NA NA NA
## 1 none 0.12878788 0.22727273 0.2727273 NA
## 2 none 0.16058394 0.26034063 0.2968370 0.3236010
## 3 none 0.30496454 0.44680851 NA NA
## 4 none 0.36082474 0.49484536 NA NA
## 0 szaura 0.00000000 NA NA NA
## 1 szaura 0.26515152 0.31818182 0.3181818 NA
## 2 szaura 0.32116788 0.36253041 0.3795620 0.3795620
## 3 szaura 0.48936170 0.49645390 NA NA
## 4 szaura 0.31958763 0.34020619 NA NA
## 0 wdall 0.37500000 NA NA NA
## 1 wdall 0.28787879 0.35606061 0.3712121 NA
## 2 wdall 0.22141119 0.27250608 0.2798054 0.2798054
## 3 wdall 0.04255319 0.04964539 NA NA
## 4 wdall 0.14432990 0.15463918 NA NA
##
## $var
## 5 10 15 20
## 0 none 0.0184217228 NA NA NA
## 1 none 0.0008558277 0.0013337443 0.0015131006 NA
## 2 none 0.0003281352 0.0004676260 0.0005068051 0.0005325854
## 3 none 0.0015322450 0.0018165307 NA NA
## 4 none 0.0024412038 0.0027097985 NA NA
## 0 szaura 0.0000000000 NA NA NA
## 1 szaura 0.0014667261 0.0016108379 0.0016108379 NA
## 2 szaura 0.0005275174 0.0005560050 0.0005658017 0.0005658017
## 3 szaura 0.0017939660 0.0018006937 NA NA
## 4 szaura 0.0022765488 0.0023636398 NA NA
## 0 wdall 0.0183515439 NA NA NA
## 1 wdall 0.0015364468 0.0016924880 0.0017241319 NA
## 2 wdall 0.0004134747 0.0004726656 0.0004802205 0.0004802205
## 3 wdall 0.0002965845 0.0003450226 NA NA
## 4 wdall 0.0013023571 0.0013860251 NA NAcif_time_begin_cat <- cuminc(etime, event, group = data$time_begin_cat)
cif_time_begin_cat## Tests:
## stat pv df
## none 0.9735549 0.614603796 2
## szaura 11.4806376 0.003213744 2
## wdall 11.6027216 0.003023438 2
## Estimates and Variances:
## $est
## 5 10 15 20
## 1 none 0.24087591 0.3321168 0.3521898 NA
## 2 none 0.19512195 0.3109756 0.3414634 NA
## 3 none 0.06976744 0.2790698 0.3488372 0.4883721
## 1 szaura 0.38321168 0.4087591 0.4105839 NA
## 2 szaura 0.26219512 0.3109756 0.3170732 NA
## 3 szaura 0.16279070 0.2209302 0.2790698 0.2790698
## 1 wdall 0.20437956 0.2262774 0.2317518 NA
## 2 wdall 0.26829268 0.3292683 0.3353659 NA
## 3 wdall 0.00000000 0.1279070 0.1511628 0.1511628
##
## $var
## 5 10 15 20
## 1 none 0.0003337620 0.0004045552 0.0004170454 NA
## 2 none 0.0009632553 0.0013149311 0.0013917662 NA
## 3 none 0.0007655603 0.0023799439 0.0026957653 0.003000155
## 1 szaura 0.0004290111 0.0004370058 0.0004383205 NA
## 2 szaura 0.0011827779 0.0013033537 0.0013288159 NA
## 3 szaura 0.0016050225 0.0020309352 0.0023837253 0.002383725
## 1 wdall 0.0002930642 0.0003143360 0.0003197223 NA
## 2 wdall 0.0011914861 0.0013315565 0.0013437890 NA
## 3 wdall 0.0000000000 0.0013222154 0.0015235488 0.001523549cif_drugs_cat <- cuminc(etime, event, group = data$drugs_cat)
cif_drugs_cat## Tests:
## stat pv df
## none 0.05302614 0.81787882 1
## szaura 3.04447643 0.08101223 1
## wdall 1.70934016 0.19107100 1
## Estimates and Variances:
## $est
## 5 10 15 20
## 1 none 0.2083897 0.3234100 0.3491204 0.3694181
## 2 none 0.2711864 0.3050847 NA NA
## 1 szaura 0.3261164 0.3599459 0.3694181 0.3707713
## 2 szaura 0.4406780 0.4745763 NA NA
## 1 wdall 0.1975643 0.2422192 0.2503383 0.2503383
## 2 wdall 0.1694915 0.1694915 NA NA
##
## $var
## 5 10 15 20
## 1 none 0.0002230811 0.0002950021 0.0003061562 0.0003141779
## 2 none 0.0035166783 0.0038474078 NA NA
## 1 szaura 0.0002963427 0.0003093162 0.0003124374 0.0003129322
## 2 szaura 0.0042753571 0.0043749685 NA NA
## 1 wdall 0.0002124771 0.0002445530 0.0002497874 0.0002497874
## 2 wdall 0.0024605738 0.0024605738 NA NA5.4 Cumulative incidence curves
plot(cif_time_begin_cat, col = 1:9)
mfit2 <-
survfit(Surv(as.numeric(etime), event) ~ as.factor(time_begin_cat), data =
data)
print(mfit2)## Call: survfit(formula = Surv(as.numeric(etime), event) ~ as.factor(time_begin_cat),
## data = data)
##
## n nevent rmean*
## as.factor(time_begin_cat)=1, (s0) 548 0 5.845545
## as.factor(time_begin_cat)=2, (s0) 164 0 6.142947
## as.factor(time_begin_cat)=3, (s0) 86 0 14.548390
## as.factor(time_begin_cat)=1, szaura 548 225 10.862214
## as.factor(time_begin_cat)=2, szaura 164 53 8.808714
## as.factor(time_begin_cat)=3, szaura 86 24 5.918486
## as.factor(time_begin_cat)=1, wdall 548 127 6.939463
## as.factor(time_begin_cat)=2, wdall 164 55 8.695561
## as.factor(time_begin_cat)=3, wdall 86 13 3.180346
## *restricted mean time in state (max time = 23.64722 )plot(
mfit2,
ylim = c(0, 1),
col = c(1, 2, 3, 4, 1, 2, 3, 4),
lty = c(1, 1, 1, 2, 2, 2, 3, 3, 3),
mark.time = FALSE,
lwd = 2,
xscale = 12,
noplot = NULL,
xlab = "Years post wd",
ylab = "Probability in State"
)
legend(
18,
.9,
c(
"non:0-2",
"non:2-4",
"non:4-",
"sz:0-2",
"sz:2-4",
"sz:4-",
"wd:0-2",
"wd:2-4",
"wd:4-"
),
col = c(1, 2, 3, 4, 1, 2, 3, 4),
lty = c(1, 1, 1, 2, 2, 2, 3, 3, 3),
lwd = 2,
bty = 'n'
)
mfit3 <- survfit(Surv(as.numeric(etime), event) ~ sex, data = data)
print(mfit3)## Call: survfit(formula = Surv(as.numeric(etime), event) ~ sex, data = data)
##
## n nevent rmean*
## sex=0, (s0) 374 0 7.915595
## sex=1, (s0) 424 0 6.615008
## sex=0, szaura 374 133 8.975504
## sex=1, szaura 424 169 10.356667
## sex=0, wdall 374 93 6.756124
## sex=1, wdall 424 102 6.675548
## *restricted mean time in state (max time = 23.64722 )plot(
mfit3,
ylim = c(0, 1),
col = c(1, 2, 3, 1, 2, 3),
lty = c(1, 1, 2, 2, 3, 3),
mark.time = FALSE,
lwd = 2,
xscale = 12,
noplot = NULL,
xlab = "Years post wd",
ylab = "Probability in State"
)
legend(
18,
1.0,
c("none:m", "none:f", "sz:m", "sz:f", "wd:m", "wd:f"),
col = c(1, 2, 3, 1, 2, 3),
lty = c(1, 1, 2, 2, 3, 3),
lwd = 2,
bty = 'n'
)
mfit4 <- survfit(Surv(as.numeric(etime), event) ~ auras, data = data)
print(mfit4)## Call: survfit(formula = Surv(as.numeric(etime), event) ~ auras, data = data)
##
## 54 observations deleted due to missingness
## n nevent rmean*
## auras=0, (s0) 676 0 7.453676
## auras=1, (s0) 68 0 4.894606
## auras=0, szaura 676 240 8.872702
## auras=1, szaura 68 41 15.950184
## auras=0, wdall 676 190 7.320843
## auras=1, wdall 68 5 2.802432
## *restricted mean time in state (max time = 23.64722 )plot(
mfit4,
ylim = c(0, 1),
col = c(1, 2, 3, 1, 2, 3),
lty = c(1, 1, 2, 2, 3, 3),
mark.time = FALSE,
lwd = 2,
xscale = 12,
noplot = NULL,
xlab = "Years post wd",
ylab = "Probability in State"
)
legend(
18,
1.0,
c("none:noaura", "none:aura", "sz:noaura", "sz:auras", "wd:noaura", "wd:auras"),
col = c(1, 2, 3, 1, 2, 3),
lty = c(1, 1, 2, 2, 3, 3),
lwd = 2,
bty = 'n'
)
mfit5 <- survfit(Surv(as.numeric(etime), event) ~ gtcs, data = data)
print(mfit5)## Call: survfit(formula = Surv(as.numeric(etime), event) ~ gtcs, data = data)
##
## 1 observation deleted due to missingness
## n nevent rmean*
## gtcs=0, (s0) 234 0 7.690958
## gtcs=1, (s0) 563 0 7.165633
## gtcs=0, szaura 234 69 7.705062
## gtcs=1, szaura 563 233 10.443566
## gtcs=0, wdall 234 74 8.251202
## gtcs=1, wdall 563 120 6.038023
## *restricted mean time in state (max time = 23.64722 )plot(
mfit5,
ylim = c(0, 1),
col = c(1, 2, 3, 1, 2, 3),
lty = c(1, 1, 2, 2, 3, 3),
mark.time = FALSE,
lwd = 2,
xscale = 12,
noplot = NULL,
xlab = "Years post wd",
ylab = "Probability in State"
)
legend(
18,
1.0,
c("none:nogtcs", "none:gtcs", "sz:nogtcs", "sz:gtcs", "wd:nogtcs", "wd:gtcs"),
col = c(1, 2, 3, 1, 2, 3),
lty = c(1, 1, 2, 2, 3, 3),
lwd = 2,
bty = 'n'
)
mfit6 <- survfit(Surv(as.numeric(etime), event) ~ extra, data = data)
print(mfit6)## Call: survfit(formula = Surv(as.numeric(etime), event) ~ extra, data = data)
##
## n nevent rmean*
## extra=0, (s0) 708 0 7.291833
## extra=1, (s0) 90 0 7.203678
## extra=0, szaura 708 275 9.847654
## extra=1, szaura 90 27 8.263164
## extra=0, wdall 708 169 6.507736
## extra=1, wdall 90 26 8.180381
## *restricted mean time in state (max time = 23.64722 )plot(
mfit6,
ylim = c(0, 1),
col = c(1, 2, 3, 1, 2, 3),
lty = c(1, 1, 2, 2, 3, 3),
mark.time = FALSE,
lwd = 2,
xscale = 12,
noplot = NULL,
xlab = "Years post wd",
ylab = "Probability in State"
)
legend(
18,
1.0,
c("none:tempsurgery", "none:extratemp", "sz:tempsurgery", "sz:extratemp", "wd:tempsurgery", "wd:extratemp"),
col = c(1, 2, 3, 1, 2, 3),
lty = c(1, 1, 2, 2, 3, 3),
lwd = 2,
bty = 'n'
)
5.5 Cause specific hazard regression - estimates and p values for each predictor and competing outcome (1:2 -> relapse, 1:3-> wdall)
library(prodlim)
library(riskRegression)## riskRegression version 2022.03.09cs <- lapply(list, function(x)
coxph(as.formula(
paste("Surv(as.numeric(etime), event) ~", x)
), data = data, id = id_wams))## Warning in coxph.fit(X, Y, istrat, offset, init, control, weights = weights, :
## Loglik converged before variable 3,4 ; coefficient may be infinite.cs## [[1]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## external 0.7782 2.1775 0.1229 0.1151 6.758 1.4e-11
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## external -0.1588 0.8532 0.1551 0.1530 -1.038 0.299
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=42.88 on 2 df, p=4.889e-10
## n= 798, number of events= 497
##
## [[2]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## sex 0.1459 1.1571 0.1161 0.1159 1.259 0.208
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## sex -0.01386 0.98624 0.14360 0.14382 -0.096 0.923
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=1.6 on 2 df, p=0.4499
## n= 798, number of events= 497
##
## [[3]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## febrile_sz -0.2511 0.7780 0.1243 0.1242 -2.022 0.0432
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## febrile_sz -0.1437 0.8661 0.1459 0.1444 -0.995 0.32
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=5.13 on 2 df, p=0.0769
## n= 733, number of events= 473
## (65 observations deleted due to missingness)
##
## [[4]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## learning_disability 0.2213 1.2477 0.2184 0.2217 0.998 0.318
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## learning_disability 0.8118 2.2520 0.2276 0.2400 3.383 0.000716
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=11.35 on 2 df, p=0.003431
## n= 735, number of events= 474
## (63 observations deleted due to missingness)
##
## [[5]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## psychiatric_pre_any -0.3369 0.7140 0.1366 0.1329 -2.536 0.0112
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## psychiatric_pre_any -0.2047 0.8149 0.1548 0.1512 -1.354 0.176
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=8.19 on 2 df, p=0.01666
## n= 735, number of events= 474
## (63 observations deleted due to missingness)
##
## [[6]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## gtcs 0.4200 1.5220 0.1371 0.1334 3.148 0.00165
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## gtcs -0.3050 0.7371 0.1479 0.1506 -2.026 0.0428
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=14.22 on 2 df, p=0.0008157
## n= 797, number of events= 496
## (1 observation deleted due to missingness)
##
## [[7]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## MRI_normal1 0.3210 1.3785 0.1787 0.1757 1.827 0.0676
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## MRI_normal1 -0.4685 0.6259 0.3426 0.3367 -1.391 0.164
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=5.14 on 2 df, p=0.0766
## n= 798, number of events= 497
##
## [[8]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## opside -0.007784 0.992246 0.111922 0.110078 -0.071 0.944
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## opside 0.03371 1.03428 0.14095 0.14825 0.227 0.82
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=0.06 on 2 df, p=0.9694
## n= 797, number of events= 497
## (1 observation deleted due to missingness)
##
## [[9]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## optemp -0.5367 0.5846 0.1179 0.1150 -4.668 3.04e-06
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## optemp 0.0845 1.0882 0.1569 0.1562 0.541 0.588
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=20.98 on 2 df, p=2.787e-05
## n= 798, number of events= 497
##
## [[10]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## as.factor(opextent)1 0.03236 1.03289 0.18811 0.18823 0.172 0.864
## as.factor(opextent)2 -0.88497 0.41273 0.72866 0.74644 -1.186 0.236
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## as.factor(opextent)1 -0.5256 0.5912 0.1823 0.1947 -2.699 0.00695
## as.factor(opextent)2 0.2615 1.2988 0.3899 0.4384 0.596 0.55092
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=12.85 on 4 df, p=0.01202
## n= 667, number of events= 432
## (131 observations deleted due to missingness)
##
## [[11]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## op_incomplete 0.7489 2.1147 0.1675 0.1860 4.026 5.66e-05
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## op_incomplete 0.2123 1.2365 0.2886 0.2730 0.778 0.437
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=17.27 on 2 df, p=0.0001777
## n= 792, number of events= 493
## (6 observations deleted due to missingness)
##
## [[12]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## pathology_HS -0.1283 0.8796 0.1171 0.1174 -1.093 0.274
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## pathology_HS -0.2058 0.8140 0.1470 0.1468 -1.402 0.161
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=3.13 on 2 df, p=0.2093
## n= 798, number of events= 497
##
## [[13]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## pathology_FCD -0.02456 0.97574 0.22229 0.21985 -0.112 0.911
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## pathology_FCD 0.5084 1.6627 0.2438 0.2459 2.068 0.0387
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=3.83 on 2 df, p=0.1472
## n= 798, number of events= 497
##
## [[14]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## pathology_DNT -0.2665 0.7661 0.2498 0.2314 -1.152 0.249
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## pathology_DNT 0.1901 1.2094 0.2476 0.2454 0.775 0.438
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=1.79 on 2 df, p=0.4081
## n= 798, number of events= 497
##
## [[15]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## pathology_CAV -0.5083 0.6015 0.2963 0.2854 -1.781 0.0749
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## pathology_CAV 0.4281 1.5344 0.2221 0.2287 1.872 0.0611
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=6.77 on 2 df, p=0.03386
## n= 667, number of events= 432
## (131 observations deleted due to missingness)
##
## [[16]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## pathology_GL -0.02723 0.97313 0.41314 0.38714 -0.07 0.944
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## pathology_GL 1.4389 4.2162 0.3288 0.3282 4.384 1.17e-05
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=13.07 on 2 df, p=0.001451
## n= 797, number of events= 496
## (1 observation deleted due to missingness)
##
## [[17]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## pathology_dual -0.2981 0.7423 0.3217 0.3145 -0.948 0.343
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## pathology_dual 0.4091 1.5055 0.2982 0.3015 1.357 0.175
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=2.62 on 2 df, p=0.2702
## n= 797, number of events= 496
## (1 observation deleted due to missingness)
##
## [[18]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## pathology_other 0.07141 1.07402 0.15774 0.16032 0.445 0.656
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## pathology_other -0.3200 0.7262 0.2364 0.2414 -1.325 0.185
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=2.19 on 2 df, p=0.3337
## n= 798, number of events= 497
##
## [[19]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## pathology_normal 0.6281 1.8741 0.3395 0.3352 1.874 0.0609
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## pathology_normal -0.7863 0.4555 1.0031 1.0693 -0.735 0.462
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=3.65 on 2 df, p=0.1613
## n= 798, number of events= 497
##
## [[20]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## acutepostszauras -0.07058 0.93186 0.20886 0.21739 -0.325 0.745
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## acutepostszauras -0.3060 0.7364 0.2871 0.2915 -1.05 0.294
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=1.36 on 2 df, p=0.507
## n= 798, number of events= 497
##
## [[21]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## auras 0.7562 2.1302 0.1697 0.1708 4.428 9.51e-06
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## auras -1.0561 0.3478 0.4535 0.4399 -2.401 0.0164
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=24.36 on 2 df, p=5.13e-06
## n= 744, number of events= 476
## (54 observations deleted due to missingness)
##
## [[22]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## age_onset_ten_cat 0.15101 1.16301 0.09875 0.10083 1.498 0.134
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## age_onset_ten_cat -0.1313 0.8770 0.1432 0.1400 -0.938 0.348
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=3.12 on 2 df, p=0.2101
## n= 798, number of events= 497
##
## [[23]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## age_at_surgery_ten_cat 0.08954 1.09367 0.10225 0.10442 0.858 0.391
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## age_at_surgery_ten_cat -0.3589 0.6985 0.1318 0.1215 -2.954 0.00314
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=8.22 on 2 df, p=0.0164
## n= 798, number of events= 497
##
## [[24]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## duration_ten_cat 0.1365 1.1463 0.1154 0.1149 1.188 0.235
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## duration_ten_cat -0.5329 0.5869 0.1488 0.1477 -3.607 0.00031
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=14.69 on 2 df, p=0.0006469
## n= 798, number of events= 497
##
## [[25]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## extent -0.1517 0.8592 0.1610 0.1581 -0.96 0.337
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## extent 0.4562 1.5780 0.1322 0.1453 3.139 0.00169
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=11.19 on 2 df, p=0.003719
## n= 667, number of events= 432
## (131 observations deleted due to missingness)
##
## [[26]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## extra -0.2347 0.7908 0.2020 0.1943 -1.208 0.227
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## extra 0.3320 1.3938 0.2114 0.2186 1.519 0.129
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=3.72 on 2 df, p=0.156
## n= 798, number of events= 497
##
## [[27]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## aura_relapse1 1.7152 5.5580 0.1237 0.1153 14.88 <2e-16
## aura_relapseN/A 2.3900 10.9138 1.0062 0.1106 21.61 <2e-16
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## aura_relapse1 -1.661e+01 6.109e-08 1.044e+03 1.967e-01 -84.44 <2e-16
## aura_relapseN/A -1.665e+01 5.856e-08 3.963e+04 1.069e+00 -15.58 <2e-16
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=180.2 on 4 df, p=< 2.2e-16
## n= 796, number of events= 497
## (2 observations deleted due to missingness)
##
## [[28]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## numsz6 0.26309 1.30094 0.06248 0.05923 4.442 8.91e-06
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## numsz6 -0.25295 0.77651 0.09241 0.09921 -2.55 0.0108
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=24.95 on 2 df, p=3.816e-06
## n= 797, number of events= 496
## (1 observation deleted due to missingness)
##
## [[29]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## time_begin_cat -0.56080 0.57075 0.09531 0.09292 -6.035 1.59e-09
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## time_begin_cat -0.55934 0.57158 0.10620 0.08968 -6.237 4.45e-10
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=72.63 on 2 df, p=< 2.2e-16
## n= 798, number of events= 497
##
## [[30]]
## Call:
## coxph(formula = as.formula(paste("Surv(as.numeric(etime), event) ~",
## x)), data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z p
## drugs_cat 0.4448 1.5601 0.1990 0.1964 2.265 0.0235
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## drugs_cat 0.002012 1.002014 0.325525 0.327931 0.006 0.995
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=4.44 on 2 df, p=0.1088
## n= 798, number of events= 497cscon <- lapply(cs, confint)
cscon## [[1]]
## 2.5 % 97.5 %
## external_1:2 0.5524964 1.0038589
## external_1:3 -0.4585830 0.1409801
##
## [[2]]
## 2.5 % 97.5 %
## sex_1:2 -0.08122638 0.3730525
## sex_1:3 -0.29574251 0.2680274
##
## [[3]]
## 2.5 % 97.5 %
## febrile_sz_1:2 -0.4944574 -0.007700644
## febrile_sz_1:3 -0.4268168 0.139376854
##
## [[4]]
## 2.5 % 97.5 %
## learning_disability_1:2 -0.2131687 0.6557171
## learning_disability_1:3 0.3415355 1.2821454
##
## [[5]]
## 2.5 % 97.5 %
## psychiatric_pre_any_1:2 -0.5973449 -0.07652076
## psychiatric_pre_any_1:3 -0.5009500 0.09158856
##
## [[6]]
## 2.5 % 97.5 %
## gtcs_1:2 0.1584818 0.681582763
## gtcs_1:3 -0.6001780 -0.009892498
##
## [[7]]
## 2.5 % 97.5 %
## MRI_normal1_1:2 -0.02328974 0.6653536
## MRI_normal1_1:3 -1.12846655 0.1914627
##
## [[8]]
## 2.5 % 97.5 %
## opside_1:2 -0.2235336 0.2079647
## opside_1:3 -0.2568484 0.3242615
##
## [[9]]
## 2.5 % 97.5 %
## optemp_1:2 -0.7621170 -0.3113696
## optemp_1:3 -0.2215387 0.3905463
##
## [[10]]
## 2.5 % 97.5 %
## as.factor(opextent)1_1:2 -0.3365668 0.4012817
## as.factor(opextent)2_1:2 -2.3479772 0.5780311
## as.factor(opextent)1_1:3 -0.9072951 -0.1439401
## as.factor(opextent)2_1:3 -0.5978349 1.1207854
##
## [[11]]
## 2.5 % 97.5 %
## op_incomplete_1:2 0.3843578 1.1134646
## op_incomplete_1:3 -0.3227937 0.7473329
##
## [[12]]
## 2.5 % 97.5 %
## pathology_HS_1:2 -0.3584495 0.10177842
## pathology_HS_1:3 -0.4934457 0.08185818
##
## [[13]]
## 2.5 % 97.5 %
## pathology_FCD_1:2 -0.45544621 0.4063326
## pathology_FCD_1:3 0.02648087 0.9904104
##
## [[14]]
## 2.5 % 97.5 %
## pathology_DNT_1:2 -0.7200616 0.1870574
## pathology_DNT_1:3 -0.2907811 0.6710027
##
## [[15]]
## 2.5 % 97.5 %
## pathology_CAV_1:2 -1.0675765 0.05100578
## pathology_CAV_1:3 -0.0200127 0.87628542
##
## [[16]]
## 2.5 % 97.5 %
## pathology_GL_1:2 -0.7860167 0.7315494
## pathology_GL_1:3 0.7955775 2.0822766
##
## [[17]]
## 2.5 % 97.5 %
## pathology_dual_1:2 -0.9144669 0.3183497
## pathology_dual_1:3 -0.1817686 0.9999799
##
## [[18]]
## 2.5 % 97.5 %
## pathology_other_1:2 -0.2428083 0.3856291
## pathology_other_1:3 -0.7931626 0.1531924
##
## [[19]]
## 2.5 % 97.5 %
## pathology_normal_1:2 -0.02882741 1.285077
## pathology_normal_1:3 -2.88215621 1.309609
##
## [[20]]
## 2.5 % 97.5 %
## acutepostszauras_1:2 -0.4966514 0.3555014
## acutepostszauras_1:3 -0.8773390 0.2652798
##
## [[21]]
## 2.5 % 97.5 %
## auras_1:2 0.4215021 1.090946
## auras_1:3 -1.9182121 -0.193935
##
## [[22]]
## 2.5 % 97.5 %
## age_onset_ten_cat_1:2 -0.04660564 0.3486315
## age_onset_ten_cat_1:3 -0.40568859 0.1431153
##
## [[23]]
## 2.5 % 97.5 %
## age_at_surgery_ten_cat_1:2 -0.1151190 0.2942050
## age_at_surgery_ten_cat_1:3 -0.5969657 -0.1207513
##
## [[24]]
## 2.5 % 97.5 %
## duration_ten_cat_1:2 -0.08871316 0.3617106
## duration_ten_cat_1:3 -0.82244559 -0.2432796
##
## [[25]]
## 2.5 % 97.5 %
## extent_1:2 -0.4615574 0.1580967
## extent_1:3 0.1713457 0.7409881
##
## [[26]]
## 2.5 % 97.5 %
## extra_1:2 -0.61555716 0.1460990
## extra_1:3 -0.09650161 0.7605504
##
## [[27]]
## 2.5 % 97.5 %
## aura_relapse1_1:2 1.489336 1.941133
## aura_relapseN/A_1:2 2.173273 2.606785
## aura_relapse1_1:3 -16.996465 -16.225364
## aura_relapseN/A_1:3 -18.748431 -14.558154
##
## [[28]]
## 2.5 % 97.5 %
## numsz6_1:2 0.1470045 0.37917521
## numsz6_1:3 -0.4473949 -0.05849533
##
## [[29]]
## 2.5 % 97.5 %
## time_begin_cat_1:2 -0.7429243 -0.3786707
## time_begin_cat_1:3 -0.7351055 -0.3835822
##
## [[30]]
## 2.5 % 97.5 %
## drugs_cat_1:2 0.05983095 0.8297178
## drugs_cat_1:3 -0.64072234 0.6447454csmulti <- coxph(Surv(as.numeric(etime), event) ~ duration_ten_cat + age_at_surgery_ten_cat + gtcs + numsz6 + learning_disability + factor(extent==1) +pathology_FCD + pathology_GL + time_begin + auras, data = data, id = id_wams)
csmulti## Call:
## coxph(formula = Surv(as.numeric(etime), event) ~ duration_ten_cat +
## age_at_surgery_ten_cat + gtcs + numsz6 + learning_disability +
## factor(extent == 1) + pathology_FCD + pathology_GL + time_begin +
## auras, data = data, id = id_wams)
##
##
## 1:2 coef exp(coef) se(coef) robust se z
## duration_ten_cat 0.239586 1.270723 0.144753 0.151751 1.579
## age_at_surgery_ten_cat 0.050448 1.051742 0.131286 0.133406 0.378
## gtcs 0.415458 1.515064 0.151014 0.148588 2.796
## numsz6 0.174450 1.190591 0.074404 0.074925 2.328
## learning_disability 0.355510 1.426909 0.241497 0.248127 1.433
## factor(extent == 1)TRUE -0.006701 0.993321 0.196140 0.202043 -0.033
## pathology_FCD 0.153598 1.166022 0.240244 0.243437 0.631
## pathology_GL 0.389328 1.475989 0.425123 0.452459 0.860
## time_begin -0.136250 0.872625 0.028888 0.036120 -3.772
## auras 1.084932 2.959239 0.194950 0.203330 5.336
##
## 1:2 p
## duration_ten_cat 0.114380
## age_at_surgery_ten_cat 0.705317
## gtcs 0.005173
## numsz6 0.019896
## learning_disability 0.151922
## factor(extent == 1)TRUE 0.973541
## pathology_FCD 0.528068
## pathology_GL 0.389529
## time_begin 0.000162
## auras 9.51e-08
##
##
## 1:3 coef exp(coef) se(coef) robust se z p
## duration_ten_cat -0.25846 0.77224 0.16343 0.16875 -1.532 0.125612
## age_at_surgery_ten_cat -0.33857 0.71279 0.14048 0.13673 -2.476 0.013282
## gtcs -0.17140 0.84248 0.15019 0.15669 -1.094 0.274006
## numsz6 -0.19687 0.82130 0.09501 0.10386 -1.896 0.058022
## learning_disability 0.77709 2.17512 0.23233 0.25580 3.038 0.002383
## factor(extent == 1)TRUE 0.17503 1.19128 0.19841 0.20716 0.845 0.398153
## pathology_FCD 0.32809 1.38831 0.25336 0.26539 1.236 0.216363
## pathology_GL 1.31824 3.73684 0.36726 0.33927 3.885 0.000102
## time_begin -0.22030 0.80228 0.04011 0.03478 -6.334 2.39e-10
## auras -0.52512 0.59148 0.46107 0.41633 -1.261 0.207199
##
## States: 1= (s0), 2= szaura, 3= wdall
##
## Likelihood ratio test=168.1 on 20 df, p=< 2.2e-16
## n= 665, number of events= 430
## (133 observations deleted due to missingness)exp(confint(csmulti))## 2.5 % 97.5 %
## duration_ten_cat_1:2 0.9438000 1.7108890
## age_at_surgery_ten_cat_1:2 0.8097557 1.3660430
## gtcs_1:2 1.1322761 2.0272606
## numsz6_1:2 1.0279829 1.3789198
## learning_disability_1:2 0.8773848 2.3206101
## factor(extent == 1)TRUE_1:2 0.6685134 1.4759421
## pathology_FCD_1:2 0.7235912 1.8789724
## pathology_GL_1:2 0.6080646 3.5827502
## time_begin_1:2 0.8129851 0.9366397
## auras_1:2 1.9865757 4.4081372
## duration_ten_cat_1:3 0.5547764 1.0749544
## age_at_surgery_ten_cat_1:3 0.5452232 0.9318579
## gtcs_1:3 0.6197075 1.1453454
## numsz6_1:3 0.6700270 1.0067147
## learning_disability_1:3 1.3174787 3.5910753
## factor(extent == 1)TRUE_1:3 0.7937487 1.7879196
## pathology_FCD_1:3 0.8252553 2.3355203
## pathology_GL_1:3 1.9218250 7.2659893
## time_begin_1:3 0.7494066 0.8588741
## auras_1:3 0.2615543 1.3375945csh <- lapply(list, function(x)
CSC(as.formula(
paste("Hist(as.numeric(etime), event) ~", x)
), data = data))## Warning in coxph.fit(X, Y, istrat, offset, init, control, weights = weights, :
## Loglik converged before variable 1,2 ; coefficient may be infinite.## Warning in coxph.fit(X, Y, istrat, offset, init, control, weights = weights, :
## Loglik converged before variable 1,2 ; coefficient may be infinite.5.6 Subdistribution model data set up
data <-
read_xlsx(
"/Users/carolinaferreiraatuesta/Documents/WAMS/ASM_withdrawal_registry/WAMS_Registry.xlsx"
)
data <-
subset(
data,
data$began_wd == 1 &
wd_all_time >= "0" &
szaura >= "0" &
wd_all >= "0" & wd_all_time >= "0" & time_szaura >= "0"
)
data$age_at_surgery_ten_cat <-
findInterval(data$age_at_surgery_ten, c(0, 2, 4), rightmost.closed = TRUE)
data$duration_ten_cat <-
findInterval(data$duration_ten, c(0, 2), rightmost.closed = TRUE)
data$age_onset_ten_cat <-
findInterval(data$age_onset_ten, c(0, 2, 4), rightmost.closed = TRUE)
data$drugs_cat <-
findInterval(data$drugs, c(0, 3), rightmost.closed = TRUE)
data$time_begin_cat <-
findInterval(data$time_begin, c(0, 2, 4), rightmost.closed = TRUE)
etime <- with(data, ifelse(szaura == 0, wd_all_time, time_szaura))
event <- with(data, ifelse(szaura == 0, 2 * wd_all, 1))
event <- factor(event, 0:2, labels = c("none", "szaura", "wdall"))
table(event)## event
## none szaura wdall
## 301 302 195szdat <-
finegray(
Surv(as.numeric(etime), event) ~ .,
data = data,
etype = "szaura",
id = id_wams
)
wddat <-
finegray(
Surv(as.numeric(etime), event) ~ .,
data = data,
etype = "wdall",
id = id_wams
)5.7 Subdistribution model - only time_begin as example but i should run this for all predictors
pfitsz <-
survfit(
Surv(fgstart, fgstop, fgstatus) ~ as.factor(time_begin_cat),
data = szdat,
weight = fgwt
)
dfitwdall <-
survfit(
Surv(fgstart, fgstop, fgstatus) ~ as.factor(time_begin_cat),
data = wddat,
weight = fgwt
)
fgfitsz <-
coxph(
Surv(fgstart, fgstop, fgstatus) ~ as.factor(time_begin_cat),
data = szdat,
weight = fgwt
)
#summary(fgfitsz)
fgfitwdall <-
coxph(
Surv(fgstart, fgstop, fgstatus) ~ as.factor(time_begin_cat),
data = wddat,
weight = fgwt
)
#summary(fgfitwdall)
finegray <- lapply(list, function(x)
coxph(as.formula(
paste("Surv(fgstart, fgstop, fgstatus) ~", x)
), data = wddat,
weight = fgwt))## Warning in agreg.fit(X, Y, istrat, offset, init, control, weights = weights, :
## Loglik converged before variable 1,2 ; beta may be infinite.finegray## [[1]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## external -0.7062 0.4935 0.1524 0.1416 -4.987 6.13e-07
##
## Likelihood ratio test=22.79 on 1 df, p=1.804e-06
## n= 10056, number of events= 195
##
## [[2]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## sex -0.06488 0.93718 0.14340 0.13308 -0.487 0.626
##
## Likelihood ratio test=0.2 on 1 df, p=0.6511
## n= 10056, number of events= 195
##
## [[3]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## febrile_sz 0.02618 1.02652 0.14521 0.13321 0.197 0.844
##
## Likelihood ratio test=0.03 on 1 df, p=0.857
## n= 9090, number of events= 194
## (966 observations deleted due to missingness)
##
## [[4]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## learning_disability 0.5406 1.7171 0.2265 0.2069 2.613 0.00898
##
## Likelihood ratio test=4.96 on 1 df, p=0.02593
## n= 9092, number of events= 195
## (964 observations deleted due to missingness)
##
## [[5]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## psychiatric_pre_any 0.003427 1.003432 0.154510 0.141265 0.024 0.981
##
## Likelihood ratio test=0 on 1 df, p=0.9823
## n= 9092, number of events= 195
## (964 observations deleted due to missingness)
##
## [[6]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## gtcs -0.4801 0.6187 0.1478 0.1382 -3.473 0.000514
##
## Likelihood ratio test=10.07 on 1 df, p=0.001507
## n= 10055, number of events= 194
## (1 observation deleted due to missingness)
##
## [[7]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## MRI_normal1 -0.7884 0.4546 0.3414 0.3275 -2.407 0.0161
##
## Likelihood ratio test=6.82 on 1 df, p=0.008997
## n= 10056, number of events= 195
##
## [[8]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## opside 0.008784 1.008823 0.140904 0.139916 0.063 0.95
##
## Likelihood ratio test=0 on 1 df, p=0.9503
## n= 10055, number of events= 195
## (1 observation deleted due to missingness)
##
## [[9]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## optemp 0.5174 1.6777 0.1548 0.1442 3.587 0.000334
##
## Likelihood ratio test=11.8 on 1 df, p=0.0005929
## n= 10056, number of events= 195
##
## [[10]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## as.factor(opextent)1 -0.4656 0.6277 0.1819 0.1728 -2.694 0.00705
## as.factor(opextent)2 0.5332 1.7043 0.3891 0.3777 1.411 0.15811
##
## Likelihood ratio test=10.94 on 2 df, p=0.004217
## n= 7450, number of events= 195
## (2606 observations deleted due to missingness)
##
## [[11]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## op_incomplete -0.4055 0.6666 0.2873 0.2657 -1.526 0.127
##
## Likelihood ratio test=2.24 on 1 df, p=0.1343
## n= 10050, number of events= 192
## (6 observations deleted due to missingness)
##
## [[12]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## pathology_HS -0.01747 0.98268 0.14586 0.13459 -0.13 0.897
##
## Likelihood ratio test=0.01 on 1 df, p=0.9047
## n= 10056, number of events= 195
##
## [[13]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## pathology_FCD 0.3450 1.4120 0.2419 0.2251 1.533 0.125
##
## Likelihood ratio test=1.86 on 1 df, p=0.173
## n= 10056, number of events= 195
##
## [[14]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## pathology_DNT 0.3777 1.4589 0.2474 0.2244 1.683 0.0924
##
## Likelihood ratio test=2.1 on 1 df, p=0.1468
## n= 10056, number of events= 195
##
## [[15]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## pathology_CAV 0.5690 1.7666 0.2221 0.2078 2.738 0.00618
##
## Likelihood ratio test=5.68 on 1 df, p=0.01714
## n= 7450, number of events= 195
## (2606 observations deleted due to missingness)
##
## [[16]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## pathology_GL 1.1590 3.1866 0.3255 0.2773 4.179 2.93e-05
##
## Likelihood ratio test=9.18 on 1 df, p=0.002443
## n= 10014, number of events= 195
## (42 observations deleted due to missingness)
##
## [[17]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## pathology_dual 0.4568 1.5790 0.2982 0.2726 1.676 0.0938
##
## Likelihood ratio test=2.06 on 1 df, p=0.1512
## n= 10014, number of events= 195
## (42 observations deleted due to missingness)
##
## [[18]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## pathology_other -0.4542 0.6349 0.2361 0.2274 -1.997 0.0458
##
## Likelihood ratio test=4.17 on 1 df, p=0.04107
## n= 10056, number of events= 195
##
## [[19]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## pathology_normal -1.3620 0.2562 1.0026 1.0107 -1.347 0.178
##
## Likelihood ratio test=3.04 on 1 df, p=0.08122
## n= 10056, number of events= 195
##
## [[20]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## acutepostszauras -0.3017 0.7396 0.2871 0.2735 -1.103 0.27
##
## Likelihood ratio test=1.21 on 1 df, p=0.2722
## n= 10056, number of events= 195
##
## [[21]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## auras -1.4482 0.2350 0.4531 0.4370 -3.314 0.00092
##
## Likelihood ratio test=16.78 on 1 df, p=4.194e-05
## n= 9197, number of events= 195
## (859 observations deleted due to missingness)
##
## [[22]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## age_onset_ten_cat -0.2644 0.7676 0.1423 0.1293 -2.046 0.0408
##
## Likelihood ratio test=3.73 on 1 df, p=0.05336
## n= 10056, number of events= 195
##
## [[23]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## age_at_surgery_ten_cat -0.3866 0.6794 0.1253 0.1026 -3.769 0.000164
##
## Likelihood ratio test=9.51 on 1 df, p=0.002043
## n= 10056, number of events= 195
##
## [[24]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## duration_ten_cat -0.5065 0.6026 0.1486 0.1375 -3.683 0.00023
##
## Likelihood ratio test=12.02 on 1 df, p=0.0005256
## n= 10056, number of events= 195
##
## [[25]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## extent 0.4823 1.6199 0.1355 0.1293 3.73 0.000192
##
## Likelihood ratio test=10.92 on 1 df, p=0.0009524
## n= 7450, number of events= 195
## (2606 observations deleted due to missingness)
##
## [[26]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## extra 0.3548 1.4259 0.2109 0.1987 1.786 0.0742
##
## Likelihood ratio test=2.6 on 1 df, p=0.1071
## n= 10056, number of events= 195
##
## [[27]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## aura_relapse1 -1.830e+01 1.127e-08 1.521e+03 8.638e-02 -211.9 <2e-16
## aura_relapseN/A -1.830e+01 1.132e-08 1.589e+04 3.096e-01 -59.1 <2e-16
##
## Likelihood ratio test=70.27 on 2 df, p=5.5e-16
## n= 10054, number of events= 195
## (2 observations deleted due to missingness)
##
## [[28]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## numsz6 -0.45045 0.63734 0.09138 0.09392 -4.796 1.62e-06
##
## Likelihood ratio test=25.91 on 1 df, p=3.575e-07
## n= 10055, number of events= 194
## (1 observation deleted due to missingness)
##
## [[29]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## time_begin_cat -0.22495 0.79856 0.10457 0.08381 -2.684 0.00727
##
## Likelihood ratio test=4.95 on 1 df, p=0.02612
## n= 10056, number of events= 195
##
## [[30]]
## Call:
## coxph(formula = as.formula(paste("Surv(fgstart, fgstop, fgstatus) ~",
## x)), data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## drugs_cat -0.3342 0.7159 0.3247 0.3068 -1.089 0.276
##
## Likelihood ratio test=1.17 on 1 df, p=0.2789
## n= 10056, number of events= 195lapply(finegray, confint)## [[1]]
## 2.5 % 97.5 %
## external -0.9837258 -0.428645
##
## [[2]]
## 2.5 % 97.5 %
## sex -0.3257187 0.1959624
##
## [[3]]
## 2.5 % 97.5 %
## febrile_sz -0.2349121 0.2872699
##
## [[4]]
## 2.5 % 97.5 %
## learning_disability 0.1351086 0.9461367
##
## [[5]]
## 2.5 % 97.5 %
## psychiatric_pre_any -0.2734473 0.2803004
##
## [[6]]
## 2.5 % 97.5 %
## gtcs -0.7510282 -0.209181
##
## [[7]]
## 2.5 % 97.5 %
## MRI_normal1 -1.430333 -0.1465465
##
## [[8]]
## 2.5 % 97.5 %
## opside -0.2654456 0.2830136
##
## [[9]]
## 2.5 % 97.5 %
## optemp 0.2347191 0.8001095
##
## [[10]]
## 2.5 % 97.5 %
## as.factor(opextent)1 -0.8043671 -0.1269149
## as.factor(opextent)2 -0.2071859 1.2735199
##
## [[11]]
## 2.5 % 97.5 %
## op_incomplete -0.926365 0.1153525
##
## [[12]]
## 2.5 % 97.5 %
## pathology_HS -0.2812539 0.2463125
##
## [[13]]
## 2.5 % 97.5 %
## pathology_FCD -0.09617643 0.7862293
##
## [[14]]
## 2.5 % 97.5 %
## pathology_DNT -0.06215348 0.8175818
##
## [[15]]
## 2.5 % 97.5 %
## pathology_CAV 0.1617062 0.9763597
##
## [[16]]
## 2.5 % 97.5 %
## pathology_GL 0.6153883 1.702543
##
## [[17]]
## 2.5 % 97.5 %
## pathology_dual -0.07746397 0.9910597
##
## [[18]]
## 2.5 % 97.5 %
## pathology_other -0.9000279 -0.00844968
##
## [[19]]
## 2.5 % 97.5 %
## pathology_normal -3.342982 0.6190811
##
## [[20]]
## 2.5 % 97.5 %
## acutepostszauras -0.8377918 0.2344132
##
## [[21]]
## 2.5 % 97.5 %
## auras -2.30475 -0.5916612
##
## [[22]]
## 2.5 % 97.5 %
## age_onset_ten_cat -0.5177751 -0.01109838
##
## [[23]]
## 2.5 % 97.5 %
## age_at_surgery_ten_cat -0.5876512 -0.1855767
##
## [[24]]
## 2.5 % 97.5 %
## duration_ten_cat -0.7760693 -0.2369708
##
## [[25]]
## 2.5 % 97.5 %
## extent 0.2288857 0.7358142
##
## [[26]]
## 2.5 % 97.5 %
## extra -0.03464227 0.7442755
##
## [[27]]
## 2.5 % 97.5 %
## aura_relapse1 -18.46999 -18.13138
## aura_relapseN/A -18.90348 -17.69001
##
## [[28]]
## 2.5 % 97.5 %
## numsz6 -0.6345276 -0.2663807
##
## [[29]]
## 2.5 % 97.5 %
## time_begin_cat -0.3892092 -0.06068195
##
## [[30]]
## 2.5 % 97.5 %
## drugs_cat -0.9355648 0.267141Plot wdall using both competing risks models and time to begin
#rawfit<- survfit(Surv(time_szaura, szaura) ~ time_begin_cat, data=data)
mfit2 <-
survfit(Surv(as.numeric(etime), event) ~ time_begin_cat, data = data) #reprise the AJ
plot(
mfit2[, 'wdall'],
col = 1:3,
ylim = c(0, 1),
lwd = 2,
xscale = 12,
conf.times = c(5, 15, 25) * 12,
xlab = "Years post wd",
ylab = "Fraction with wdall"
)
ndata <- data.frame(time_begin_cat = c("1", "2", "3"))
fgsurv <- survfit(fgfitwdall, ndata)
lines(
fgsurv,
fun = "event",
lty = 2,
lwd = 2,
col = 1:3
)
#lines(rawfit, fun="event", lty=2, lwd=2, col=1:2)
legend(
"topleft",
c(
"0-2, Aalen-Johansen",
"2-4, Aalen-Johansen",
"4, Aalen-Johansen",
"0-2, Fine-Gray",
"2-4, Fine-Gray",
"4-, Fine-Gray"
),
col = 1:3,
lty = c(1, 1, 1, 2, 2, 2),
bty = 'n'
)
# rate models with only
cfitr <-
coxph(Surv(as.numeric(etime), event) ~ time_begin_cat, data, id = id_wams)
#rcurve <- survfit(cfitr, newdata=ndata)
#lines(rcurve[, 'pcm'], col=6:7) # makes the plot too crowsdedPlot szaura relapse model using both competing risks models and time to begin
mfit2 <-
survfit(Surv(as.numeric(etime), event) ~ time_begin_cat, data = data)
plot(
mfit2[, 'szaura'],
col = 1:3,
ylim = c(0, 1),
lwd = 2,
xscale = 12,
conf.times = c(5, 15, 25) * 12,
xlab = "Years post wd",
ylab = "Fraction with szaura"
)
ndata <- data.frame(time_begin_cat = c("1", "2", "3"))
fgsurv <- survfit(fgfitsz, ndata)
lines(
fgsurv,
fun = "event",
lty = 2,
lwd = 2,
col = 1:3
)
legend(
"topleft",
c(
"0-2, Aalen-Johansen",
"2-4, Aalen-Johansen",
"4, Aalen-Johansen",
"0-2, Fine-Gray",
"2-4, Fine-Gray",
"4-, Fine-Gray"
),
col = 1:3,
lty = c(1, 1, 1, 2, 2, 2),
bty = 'n'
)
cfitr <-
coxph(Surv(as.numeric(etime), event) ~ time_begin_cat, data, id = id_wams)
#rcurve <- survfit(cfitr, newdata=ndata)finegraymulti <-
coxph(Surv(fgstart, fgstop, fgstatus) ~ age_onset_ten_cat + duration_ten_cat + age_at_surgery_ten_cat + gtcs + numsz6 + learning_disability +MRI_normal + factor(extent==1) +pathology_CAV +pathology_GL +pathology_other+ time_begin_cat + auras, data = wddat, weight = fgwt)
finegraymulti## Call:
## coxph(formula = Surv(fgstart, fgstop, fgstatus) ~ age_onset_ten_cat +
## duration_ten_cat + age_at_surgery_ten_cat + gtcs + numsz6 +
## learning_disability + MRI_normal + factor(extent == 1) +
## pathology_CAV + pathology_GL + pathology_other + time_begin_cat +
## auras, data = wddat, weights = fgwt)
##
## coef exp(coef) se(coef) robust se z p
## age_onset_ten_cat -0.42775 0.65197 0.18301 0.16877 -2.535 0.01126
## duration_ten_cat -0.51499 0.59751 0.18466 0.16529 -3.116 0.00184
## age_at_surgery_ten_cat -0.17681 0.83794 0.16646 0.14821 -1.193 0.23288
## gtcs -0.38477 0.68061 0.15115 0.14264 -2.697 0.00699
## numsz6 -0.26618 0.76630 0.09534 0.10050 -2.649 0.00808
## learning_disability 0.47907 1.61457 0.24014 0.22692 2.111 0.03476
## MRI_normal1 -0.48389 0.61638 0.35139 0.33409 -1.448 0.14751
## factor(extent == 1)TRUE 0.03361 1.03418 0.20670 0.20011 0.168 0.86662
## pathology_CAV 0.61842 1.85599 0.26520 0.25808 2.396 0.01657
## pathology_GL 1.12872 3.09170 0.35948 0.27759 4.066 4.78e-05
## pathology_other -0.11464 0.89169 0.25364 0.23381 -0.490 0.62393
## time_begin_cat -0.38621 0.67962 0.10989 0.09263 -4.170 3.05e-05
## auras -1.09114 0.33583 0.45749 0.43468 -2.510 0.01207
##
## Likelihood ratio test=82.77 on 13 df, p=3.32e-12
## n= 7407, number of events= 194
## (2649 observations deleted due to missingness)exp(confint(finegraymulti))## 2.5 % 97.5 %
## age_onset_ten_cat 0.4683565 0.9075766
## duration_ten_cat 0.4321590 0.8261151
## age_at_surgery_ten_cat 0.6266968 1.1203874
## gtcs 0.5146120 0.9001534
## numsz6 0.6292865 0.9331393
## learning_disability 1.0349036 2.5189044
## MRI_normal1 0.3202371 1.1863865
## factor(extent == 1)TRUE 0.6986591 1.5308306
## pathology_CAV 1.1191729 3.0779073
## pathology_GL 1.7943827 5.3269496
## pathology_other 0.5638865 1.4100560
## time_begin_cat 0.5667947 0.8149160
## auras 0.1432581 0.7872779convencoxmulti <-
coxph(Surv(as.numeric(wd_all_time), wd_all) ~ febrile_sz + learning_disability +extra + pathology_HS +pathology_FCD +as.numeric(time_begin), data = data)
convencoxmulti## Call:
## coxph(formula = Surv(as.numeric(wd_all_time), wd_all) ~ febrile_sz +
## learning_disability + extra + pathology_HS + pathology_FCD +
## as.numeric(time_begin), data = data)
##
## coef exp(coef) se(coef) z p
## febrile_sz 0.10568 1.11147 0.13490 0.783 0.4334
## learning_disability 0.34284 1.40895 0.20104 1.705 0.0881
## extra 0.41437 1.51341 0.21252 1.950 0.0512
## pathology_HS 0.32244 1.38049 0.15229 2.117 0.0342
## pathology_FCD -0.05184 0.94948 0.23572 -0.220 0.8259
## as.numeric(time_begin) -0.15929 0.85275 0.02887 -5.518 3.42e-08
##
## Likelihood ratio test=51.8 on 6 df, p=2.046e-09
## n= 733, number of events= 283
## (65 observations deleted due to missingness)exp(confint(convencoxmulti))## 2.5 % 97.5 %
## febrile_sz 0.8532432 1.4478405
## learning_disability 0.9501029 2.0893874
## extra 0.9978305 2.2954013
## pathology_HS 1.0242438 1.8606547
## pathology_FCD 0.5981933 1.5070597
## as.numeric(time_begin) 0.8058454 0.9023872