Jie Fistulogram v1
Dihua Xu
January 21, 2016
1. How many patients have repeat fistulogram in the follow up period?
count any patient who had multiple rows in the dataset
## [1] 0.508count only those who had repeated fistulogram in the same lesion
## [1] 0.2182. How many patients developed new lesions in the subsequent fistulogram?
## [1] 0.2583. The distribution of lesions in each type of fistula (locationoffistula)
##
## locationofthefistula l2 l4 l5 l6 l7 l8 l9 l10 l11 l12 l13 l14
## 1 21 6 0 14 10 0 4 0 0 4 0 0
## 2 28 10 64 25 9 2 19 2 3 0 0 56
## 3 18 12 0 17 5 2 13 0 1 0 0 234. Angioplasty survival for each lesion:
## Call: survfit(formula = Surv(surv.days, as.numeric(event)) ~ lesion.group,
## data = subset(dat.surv.lesion, !is.na(lesion.group)))
##
## lesion.group=l12l13l14
## time n.risk n.event survival std.err lower 95% CI upper 95% CI
## 5 39 1 0.974 0.0253 0.926 1.000
## 41 37 2 0.922 0.0434 0.840 1.000
## 54 35 1 0.895 0.0495 0.803 0.998
## 66 34 1 0.869 0.0546 0.768 0.983
## 75 33 1 0.843 0.0590 0.735 0.967
## 84 32 1 0.816 0.0627 0.702 0.949
## 89 31 2 0.764 0.0689 0.640 0.911
## 91 29 1 0.737 0.0713 0.610 0.891
## 98 28 1 0.711 0.0735 0.581 0.871
## 100 27 1 0.685 0.0753 0.552 0.849
## 105 26 1 0.658 0.0769 0.524 0.828
## 111 25 1 0.632 0.0782 0.496 0.806
## 112 24 2 0.579 0.0801 0.442 0.760
## 138 22 1 0.553 0.0806 0.416 0.736
## 182 21 1 0.527 0.0810 0.390 0.712
## 187 20 1 0.500 0.0811 0.364 0.687
## 195 19 1 0.474 0.0810 0.339 0.663
## 205 18 1 0.448 0.0807 0.314 0.637
## 215 16 1 0.420 0.0803 0.288 0.611
## 235 13 1 0.387 0.0804 0.258 0.582
## 272 11 1 0.352 0.0804 0.225 0.551
## 436 6 1 0.293 0.0858 0.165 0.521
##
## lesion.group=l2l4
## time n.risk n.event survival std.err lower 95% CI upper 95% CI
## 6 61 1 0.984 0.0163 0.952 1.000
## 8 60 1 0.967 0.0228 0.924 1.000
## 15 59 1 0.951 0.0277 0.898 1.000
## 72 58 2 0.918 0.0351 0.852 0.990
## 75 56 1 0.902 0.0381 0.830 0.980
## 91 55 1 0.885 0.0408 0.809 0.969
## 93 54 1 0.869 0.0432 0.788 0.958
## 94 53 1 0.852 0.0454 0.768 0.946
## 98 51 1 0.836 0.0475 0.748 0.934
## 99 50 1 0.819 0.0494 0.728 0.922
## 105 49 1 0.802 0.0511 0.708 0.909
## 112 48 1 0.786 0.0527 0.689 0.896
## 154 47 1 0.769 0.0542 0.670 0.883
## 178 44 1 0.751 0.0557 0.650 0.869
## 182 43 1 0.734 0.0571 0.630 0.855
## 213 38 1 0.715 0.0588 0.608 0.840
## 255 35 1 0.694 0.0605 0.585 0.824
##
## lesion.group=l5
## time n.risk n.event survival std.err lower 95% CI upper 95% CI
## 2 38 1 0.974 0.0260 0.924 1.000
## 28 37 1 0.947 0.0362 0.879 1.000
## 36 36 1 0.921 0.0437 0.839 1.000
## 41 35 2 0.868 0.0548 0.767 0.983
## 54 33 1 0.842 0.0592 0.734 0.966
## 70 32 1 0.816 0.0629 0.701 0.949
## 75 31 1 0.789 0.0661 0.670 0.930
## 76 30 1 0.763 0.0690 0.639 0.911
## 89 29 1 0.737 0.0714 0.609 0.891
## 94 28 1 0.711 0.0736 0.580 0.870
## 111 27 1 0.684 0.0754 0.551 0.849
## 112 26 1 0.658 0.0770 0.523 0.827
## 168 24 1 0.630 0.0785 0.494 0.805
## 216 21 1 0.600 0.0803 0.462 0.780
## 267 18 1 0.567 0.0825 0.426 0.754
## 322 17 1 0.534 0.0841 0.392 0.727
##
## lesion.group=l6l7l8
## time n.risk n.event survival std.err lower 95% CI upper 95% CI
## 15 35 1 0.971 0.0282 0.918 1.000
## 43 34 2 0.914 0.0473 0.826 1.000
## 54 32 1 0.886 0.0538 0.786 0.998
## 70 31 1 0.857 0.0591 0.749 0.981
## 85 30 1 0.829 0.0637 0.713 0.963
## 98 29 2 0.771 0.0710 0.644 0.924
## 100 27 1 0.743 0.0739 0.611 0.903
## 105 26 1 0.714 0.0764 0.579 0.881
## 114 25 1 0.686 0.0785 0.548 0.858
## 123 24 1 0.657 0.0802 0.517 0.835
## 170 21 1 0.626 0.0823 0.484 0.810
## 192 20 1 0.595 0.0839 0.451 0.784
## 197 19 1 0.563 0.0851 0.419 0.757
## 201 18 1 0.532 0.0860 0.388 0.730
## 252 13 2 0.450 0.0901 0.304 0.666
## Call:
## coxph(formula = Surv(surv.days, as.numeric(event)) ~ lesion.group,
## data = subset(dat.surv.lesion, !is.na(lesion.group)))
##
## n= 173, number of events= 78
##
## coef exp(coef) se(coef) z Pr(>|z|)
## lesion.groupl2l4 -0.9994 0.3681 0.3098 -3.226 0.00126 **
## lesion.groupl5 -0.4522 0.6363 0.3150 -1.436 0.15112
## lesion.groupl6l7l8 -0.3023 0.7391 0.3093 -0.977 0.32838
## ---
## Signif. codes: 0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1
##
## exp(coef) exp(-coef) lower .95 upper .95
## lesion.groupl2l4 0.3681 2.717 0.2006 0.6756
## lesion.groupl5 0.6363 1.572 0.3432 1.1796
## lesion.groupl6l7l8 0.7391 1.353 0.4031 1.3551
##
## Concordance= 0.592 (se = 0.033 )
## Rsquare= 0.063 (max possible= 0.987 )
## Likelihood ratio test= 11.24 on 3 df, p=0.0105
## Wald test = 10.63 on 3 df, p=0.01391
## Score (logrank) test = 11.25 on 3 df, p=0.010465. The distribution of number of lesions in each fistulogram
## rowSums(dat[, 15:26] == "Y")
## 0 1 2 3 4
## 8 161 77 15 2