-
Notifications
You must be signed in to change notification settings - Fork 0
/
Copy pathMJ Asthma and Obesity
679 lines (547 loc) · 30.3 KB
/
MJ Asthma and Obesity
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
251
252
253
254
255
256
257
258
259
260
261
262
263
264
265
266
267
268
269
270
271
272
273
274
275
276
277
278
279
280
281
282
283
284
285
286
287
288
289
290
291
292
293
294
295
296
297
298
299
300
301
302
303
304
305
306
307
308
309
310
311
312
313
314
315
316
317
318
319
320
321
322
323
324
325
326
327
328
329
330
331
332
333
334
335
336
337
338
339
340
341
342
343
344
345
346
347
348
349
350
351
352
353
354
355
356
357
358
359
360
361
362
363
364
365
366
367
368
369
370
371
372
373
374
375
376
377
378
379
380
381
382
383
384
385
386
387
388
389
390
391
392
393
394
395
396
397
398
399
400
401
402
403
404
405
406
407
408
409
410
411
412
413
414
415
416
417
418
419
420
421
422
423
424
425
426
427
428
429
430
431
432
433
434
435
436
437
438
439
440
441
442
443
444
445
446
447
448
449
450
451
452
453
454
455
456
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481
482
483
484
485
486
487
488
489
490
491
492
493
494
495
496
497
498
499
500
501
502
503
504
505
506
507
508
509
510
511
512
513
514
515
516
517
518
519
520
521
522
523
524
525
526
527
528
529
530
531
532
533
534
535
536
537
538
539
540
541
542
543
544
545
546
547
548
549
550
551
552
553
554
555
556
557
558
559
560
561
562
563
564
565
566
567
568
569
570
571
572
573
574
575
576
577
578
579
580
581
582
583
584
585
586
587
588
589
590
591
592
593
594
595
596
597
598
599
600
601
602
603
604
605
606
607
608
609
610
611
612
613
614
615
616
617
618
619
620
621
622
623
624
625
626
627
628
629
630
631
632
633
634
635
636
637
638
639
640
641
642
643
644
645
646
647
648
649
650
651
652
653
654
655
656
657
658
659
660
661
662
663
664
665
666
667
668
669
670
671
672
673
674
675
676
677
678
679
library(data.table)
library(epiDisplay)
library(dplyr)
library(readxl)
library(geepack)
## order by pm
rawdata<-fread("C:\\Users\\user\\Desktop\\美兆new\\alldata_with_metabolic_syndromes.csv")
rawdata<-data.frame(rawdata[order(rawdata$serialno),])
##environment
environment<-fread("C:\\Users\\user\\Desktop\\美照處理data\\not_fill_Environment.csv")
environment<-environment[,-1]
environment<-data.frame(environment)
##combine environment energy
data<-cbind(rawdata,environment)
data<-data.frame(data)
##環境data
sum(!is.na(data$o3_patient))
length(unique(data$pid[is.na(data$o3_patient)]))
sum(!is.na(data$SO2_patient))
length(unique(data$pid[is.na(data$SO2_patient)]))
sum(!is.na(data$NO2_patient))
length(unique(data$pid[is.na(data$NO2_patient)]))
sum(!is.na(data$pm10_patient))
length(unique(data$pid[is.na(data$pm10_patient)]))
o3_NA<-data.frame(data$serialno[is.na(data$o3_patient)],data$yr[is.na(data$o3_patient)],data$longitude[is.na(data$o3_patient)],data$latitude[is.na(data$o3_patient)],data$o3_patient[is.na(data$o3_patient)])
NO2_NA<-data.frame(data$serialno[is.na(data$NO2_patient)],data$yr[is.na(data$NO2_patient)],data$longitude[is.na(data$NO2_patient)],data$latitude[is.na(data$NO2_patient)],data$NO2_patient[is.na(data$NO2_patient)])
SO2_NA<-data.frame(data$serialno[is.na(data$SO2_patient)],data$yr[is.na(data$SO2_patient)],data$longitude[is.na(data$SO2_patient)],data$latitude[is.na(data$SO2_patient)],data$SO2_patient[is.na(data$SO2_patient)])
pm10_NA<-data.frame(data$serialno[is.na(data$pm10_patient)],data$yr[is.na(data$pm10_patient)],data$longitude[is.na(data$pm10_patient)],data$latitude[is.na(data$pm10_patient)],data$pm10_patient[is.na(data$pm10_patient)])
names(o3_NA)<-c("serialno","yr","lon","lat","o3")
names(NO2_NA)<-c("serialno","yr","lon","lat","NO2")
names(SO2_NA)<-c("serialno","yr","lon","lat","SO2")
names(pm10_NA)<-c("serialno","yr","lon","lat","pm10")
write.csv(o3_NA,"C:\\Users\\user\\Desktop\\environment\\o3_NA.csv",quote=FALSE,row.names=F)
write.csv(NO2_NA,"C:\\Users\\user\\Desktop\\environment\\NO2_NA.csv",quote=FALSE,row.names=F)
write.csv(SO2_NA,"C:\\Users\\user\\Desktop\\environment\\SO2_NA.csv",quote=FALSE,row.names=F)
write.csv(pm10_NA,"C:\\Users\\user\\Desktop\\environment\\pm10_NA.csv",quote=FALSE,row.names=F)
##QGIS
o3<-read.csv("C:\\Users\\user\\Desktop\\mapdata201911260954\\o3_new.csv")
No2<-read.csv("C:\\Users\\user\\Desktop\\mapdata201911260954\\NO2_new.csv")
SO2<-read.csv("C:\\Users\\user\\Desktop\\mapdata201911260954\\SO2_new.csv")
pm10<-read.csv("C:\\Users\\user\\Desktop\\mapdata201911260954\\pm10_new.csv")
###附值
data1<-data
data1[is.na(data$o3_patient),"o3_patient"]<-o3$o3
data1[is.na(data$NO2_patient),"NO2_patient"]<-No2$NO2
data1[is.na(data$SO2_patient),"SO2_patient"]<-SO2$SO2
data1[is.na(data$pm10_patient),"pm10_patient"]<-pm10$pm10
sum(is.na(data1$o3_patient))
length(unique(data1$pid[is.na(data1$o3_patient)]))
##將離島的人附值為NA
for(i in 1:length(unique(data1$pid[is.na(data1$o3_patient)]))){
data1[data1$pid==unique(data1$pid[is.na(data1$o3_patient)])[i],c("o3_patient")]<-rep(NA,length(data1[data1$pid==unique(data1$pid[is.na(data1$o3_patient)])[i],c("o3_patient")]))
}
data2<-data1
data2<-data.frame(data2)
sum(is.na(data2$o3_patient))
length(unique(data2$pid[is.na(data2$o3_patient)]))
data3<-data2[complete.cases(data2$o3_patient),]
length(unique(data3$pid))
##-------------------------------------------
##對本來就是NA的進行補植 只要自述為1 就當次病患發生過氣喘(psick14,mdrug08)
sum(is.na(data3$psick14))
sum(is.na(data3$mdrug08))
data4<-data3
data4$new_psick14<-data3$psick14
data4$new_mdrug08<-data3$mdrug08
for (i in 1:length(data3$pid[is.na(data3$psick14)])) {
aa<-data3[data3$pid==data3$pid[is.na(data3$psick14)][i],"psick14"]
if(sum(aa,na.rm = T)>=1){
aa<-rep(1,length(data3[data3$pid==data3$pid[is.na(data3$psick14)][i],"psick14"]))}
data4[data4$pid==data4$pid[is.na(data4$psick14)][i],"new_psick14"]<-aa
}
for (i in 1:length(data3$pid[is.na(data3$mdrug08)])) {
bb<-data3[data3$pid==data3$pid[is.na(data3$mdrug08)][i],"mdrug08"]
if(sum(bb,na.rm = T)>=1){
bb<-rep(1,length(data3[data3$pid==data3$pid[is.na(data3$mdrug08)][i],"mdrug08"]))
}
data4[data4$pid==data4$pid[is.na(data4$mdrug08)][i],"new_mdrug08"]<-bb
}
sum(is.na(data4$new_psick14))
sum(is.na(data4$new_mdrug08))
##設定new_psick14 NA=3 new_mdrug08 NA=3
data4$new_psick14[is.na(data4$new_psick14)]<-3
data4$new_mdrug08[is.na(data4$new_mdrug08)]<-3
table(data4$new_psick14)
data5<-data4
##只要其中一個符合就為Asthma(長期服用藥物或是自述氣喘)
data5$Asthma<-rep(0,nrow(data5))
data5[ data5$new_psick14==1 | data5$new_mdrug08==1,"Asthma"]<-1
data5[ (data5$new_psick14==3 & data5$new_mdrug08==0) |(data5$new_psick14==0 & data5$new_mdrug08==3)|(data5$new_psick14==3 & data5$new_mdrug08==3),"Asthma"]<-2
sum(is.na(data5$Asthma))
table(data5$Asthma)
data5<-data5[data5$Asthma!=2,]
length(unique(data5$pid))
##定義氣喘:只要病人有氣喘，病人其他筆資料也定義為氣喘
Asthma_id<-unique(data5[data5$Asthma==1,"pid"])
x<-data.frame(pid=Asthma_id,Asthma=rep(1,length(Asthma_id)))
new_data<-merge.data.frame(x=data5,y=x,by="pid",all.x = T)
table(new_data$Asthma.x)
table(new_data$Asthma.y)
new_data[is.na(new_data$Asthma.y),"Asthma.y"]<-0
new_data<-data.frame(new_data)
new_data<-new_data[,-which(names(new_data)=="Asthma.x")]
table(new_data$Asthma.y)
##FEV1/FVC
##將呼吸道為NA的資料刪除
sum(is.na(new_data$ratio_lung_function))
length(unique(new_data$pid[is.na(new_data$ratio_lung_function)]))
data6<-new_data[complete.cases(new_data$ratio_lung_function),]
length(unique(data6$pid))
##將呼吸道小於0.7 或大於1的資料刪除
data7<-data6[data6$ratio_lung_function>=0.7,]
length(unique(data7$pid))
data8<-data7[data7$ratio_lung_function<=1,]
length(unique(data8$pid))
##將腰圍為遺失值的刪除
sum(is.na(data8$g_wc))
data9<-data8[complete.cases(data8$g_wc),]
length(unique(data9$pid))
## BMI
#將BMI為遺失值的刪除na=88
sum(is.na(data9$g_bmi))
data10<-data9[complete.cases(data9$g_bmi),]
length(unique(data10$pid))
##age 將年齡分成10組
sum(is.na(data10$age))
summary(data10$age)
data10$age_class<-NA
data10$age_class[data10$age>=1 & data10$age<10]<-1
data10$age_class[data10$age>=10 & data10$age<20]<-2
data10$age_class[data10$age>=20 & data10$age<30]<-3
data10$age_class[data10$age>=30 & data10$age<40]<-4
data10$age_class[data10$age>=40 & data10$age<50]<-5
data10$age_class[data10$age>=50 & data10$age<60]<-6
data10$age_class[data10$age>=60 & data10$age<70]<-7
data10$age_class[data10$age>=70 & data10$age<80]<-8
data10$age_class[data10$age>=80 & data10$age<90]<-9
data10$age_class[data10$age>=90 & data10$age<100]<-10
data10$age_class[ data10$age>=100]<-11
##將前面資料剔除後或是病人健檢只來一次刪除
data11<-data10
pid_data11<-table(data11$pid)
pid_data11<-data.frame(pid_data11)
pid<-as.character(pid_data11$Var1[pid_data11$Freq==1])
length(pid)
data12<-data11[-c(match(pid,data11$pid)),]
## 將資料對年月進行排序
data12<-data12[order(data12$yr,data12$mon),]
##對教育程度進行補值 (以前一次回答或是第一次回答去填補下一次回答，最後再利用眾數去填補)
aa<-list()
for (i in 1:length(unique(data12$pid))) {
aa[[i]]<-data12$education[data12$pid==unique(data12$pid)[[i]]]
##最一開始的NA用後面補值
if(length(which(is.na(aa[[i]])))==0){
next
}else if(length(which(is.na(aa[[i]])))== length(aa[[i]]) ) {
next
}else if(which(is.na(aa[[i]]))[1]==1){
aa[[i]][which(is.na(aa[[i]]))[1]]<-aa[[i]][which(is.na(aa[[i]]))[1]+1]
}
##後面NA用前一次回答補值
if(length(which(is.na(aa[[i]])))==0){
next
}else if(length(which(is.na(aa[[i]])))== length(aa[[i]]) ) {
next
}else if(which(is.na(aa[[i]]))>1){
aa[[i]][which(is.na(aa[[i]]))]<-aa[[i]][which(is.na(aa[[i]]))-1]
}}
##利用眾數進行補值
for(i in 1:length(unique(data12$pid))){
if(length(which(is.na(aa[[i]])))==0){
next
}else if(length(which(is.na(aa[[i]])))== length(aa[[i]]) ) {
next
}else{
aa[[i]][which(is.na(aa[[i]]))]<-rep(as.numeric(names(table(aa[[ i]])))[table(aa[[i]]) == max(table(aa[[i]]))],length(which(is.na(aa[[i]]))))
}
}
##附值
for(i in 1:length(unique(data12$pid))){
data12$education[data12$pid==unique(data12$pid)[[i]]]<-aa[[i]]
}
data12<-data.frame(data12)
##刪除 education missing
sum(is.na(data13$education))
data13<-data12[complete.cases(data12$education),]
##class education
data13$class_education <-NA
data13[!is.na(data13$education) & data13$education==1 | data13$education==2 ,"class_education"] <-1
data13[!is.na(data13$education) & data13$education==3| data13$education==4 ,"class_education"] <-2
data13[!is.na(data13$education) & data13$education==5| data13$education==6| data13$education==7 ,"class_education"] <-3
##smoke 將抽菸定義為 不抽 二手菸 以前抽 常抽
data13$smoke<-NA
data13[is.na(data13$smokeornot_98) & as.numeric(data13$yr) <=2002 & data13$age<18,"smokeornot_98"]<-1
data13[!is.na(data13$smokeornot_98) & data13$smokeornot_98==1 & as.numeric(data13$yr) <=2002,'smoke']<-1
data13[!is.na(data13$smokeornot_98) & data13$smokeornot_98==2 & as.numeric(data13$yr) <=2002,'smoke']<-2
data13[!is.na(data13$smokeornot_98) & data13$smokeornot_98==3 & as.numeric(data13$yr) <=2002,'smoke']<-3
data13[!is.na(data13$smokeornot_98) & data13$smokeornot_98==4 & as.numeric(data13$yr) <=2002,'smoke']<-4
data13[!is.na(data13$smokeornot_98) & data13$smokeornot_98==5 & as.numeric(data13$yr) <=2002,'smoke']<-4
data13[!is.na(data13$smokeornot_98) & data13$smokeornot_98==6 & as.numeric(data13$yr) <=2002,'smoke']<-4
data13[is.na(data13$smokeornot_03) & as.numeric(data13$yr) >2002 & data13$age<18,"smokeornot_03"]<-1
data13[!is.na(data13$smokeornot_03) & data13$smokeornot_03==1 & as.numeric(data13$yr) >2002,'smoke']<-1
data13[!is.na(data13$smokeornot_03) & data13$smokeornot_03==2 & as.numeric(data13$yr) >2002,'smoke']<-2
data13[!is.na(data13$smokeornot_03) & data13$smokeornot_03==3 & as.numeric(data13$yr) >2002,'smoke']<-3
data13[!is.na(data13$smokeornot_03) & data13$smokeornot_03==4 & as.numeric(data13$yr) >2002,'smoke']<-4
data13[!is.na(data13$smokeornot_03) & data13$smokeornot_03==5 & as.numeric(data13$yr) >2002,'smoke']<-4
data14<-data13
##smoke 補眾數
b<-list()b
sum(is.na(bb[[]]))
for (i in 1:length(unique(data14$pid))) {
bb[[i]]<-data14$smoke[data14$pid==unique(data14$pid)[[i]]]
if(length(which(is.na(bb[[i]])))==0){
next
}else if(length(which(is.na(bb[[i]])))== length(bb[[i]]) ) {
next
}else{
bb[[i]][which(is.na(bb[[i]]))]<-rep(as.numeric(names(table(bb[[i]])))[table(bb[[i]]) == max(table(bb[[i]]))],length(which(is.na(bb[[i]]))))
}}
for (i in 1:length(unique(data14$pid))) {
data14$smoke[data14$pid==unique(data14$pid)[[i]]]<-bb[[i]]
}
data14<-data.frame(data14)
data15<-data14[complete.cases(data14$smoke),]
library(data.table)
data15<-fread("C:\\Users\\user\\Desktop\\美兆整理\\data16.csv")
##--MET-----------------
data15$vlightsport<-data15$vmeansport<-data15$vheavysport<-data15$vvioltsport<-data15$vsportornot_98<-data15$vfirstsport<-data15$vfirstsportfrequ<-data15$vfirstsporttime<-NA
data15[(!is.na(data15$lightsport) & data15$lightsport==1),c("vlightsport")]<-2.5
data15[(!is.na(data15$lightsport) & data15$lightsport==0),c("vlightsport")]<-0
data15[is.na(data15$lightsport),c("vlightsport")]<-0
data15[(!is.na(data15$meansport) & data15$meansport==1),c("vmeansport")]<-4.5
data15[(!is.na(data15$meansport) & data15$meansport==0),c("vmeansport")]<-0
data15[is.na(data15$meansport),c("vmeansport")]<-0
data15[(!is.na(data15$heavysport) & data15$heavysport==1),c("vheavysport")]<-6.5
data15[(!is.na(data15$heavysport) & data15$heavysport==0),c("vheavysport")]<-0
data15[is.na(data15$heavysport),c("vheavysport")]<-0
data15[(!is.na(data15$violtsport) & data15$violtsport==1),c("vvioltsport")]<-8.5
data15[(!is.na(data15$violtsport) & data15$violtsport==0),c("vvioltsport")]<-0
data15[is.na(data15$violtsport),c("vvioltsport")]<-0
data15[(!is.na(data15$sportornot_98) & data15$sportornot_98==1),c("vsportornot_98")]<-0.5
data15[(!is.na(data15$sportornot_98) & data15$sportornot_98==2),c("vsportornot_98")]<-1.5
data15[(!is.na(data15$sportornot_98) & data15$sportornot_98==3),c("vsportornot_98")]<-3.5
data15[(!is.na(data15$sportornot_98) & data15$sportornot_98==4),c("vsportornot_98")]<-5.5
data15[(!is.na(data15$sportornot_98) & data15$sportornot_98==5),c("vsportornot_98")]<-7.5
data15[is.na(data15$sportornot_98),c("vsportornot_98")]<-0
data15[(!is.na(data15$firstsport) & data15$firstsport==1),c("vfirstsport")]<-2.5
data15[(!is.na(data15$firstsport) & data15$firstsport==2),c("vfirstsport")]<-4.5
data15[(!is.na(data15$firstsport) & data15$firstsport==3),c("vfirstsport")]<-6.5
data15[(!is.na(data15$firstsport) & data15$firstsport==4),c("vfirstsport")]<-8.5
data15[(!is.na(data15$firstsport) & data15$firstsport==5),c("vfirstsport")]<-0
data15[is.na(data15$firstsport),c("vfirstsport")]<-0
data15[(!is.na(data15$firstsportfrequ) & data15$firstsportfrequ==1),c("vfirstsportfrequ")]<-17.5
data15[(!is.na(data15$firstsportfrequ) & data15$firstsportfrequ==2),c("vfirstsportfrequ")]<-7
data15[(!is.na(data15$firstsportfrequ) & data15$firstsportfrequ==3),c("vfirstsportfrequ")]<-2.8
data15[(!is.na(data15$firstsportfrequ) & data15$firstsportfrequ==4),c("vfirstsportfrequ")]<-1
data15[(!is.na(data15$firstsportfrequ) & data15$firstsportfrequ==5),c("vfirstsportfrequ")]<-0.5
data15[is.na(data15$firstsportfrequ),c("vfirstsportfrequ")]<-0
data15[(!is.na(data15$firstsporttime) & data15$firstsporttime==1),c("vfirstsporttime")]<-0.25
data15[(!is.na(data15$firstsporttime) & data15$firstsporttime==2),c("vfirstsporttime")]<-0.75
data15[(!is.na(data15$firstsporttime) & data15$firstsporttime==3),c("vfirstsporttime")]<-1.5
data15[(!is.na(data15$firstsporttime) & data15$firstsporttime==4),c("vfirstsporttime")]<-2.5
data15[is.na(data15$firstsporttime),c("vfirstsporttime")]<-0
data15$met_hr <-data15$vfirstsport * data15$vfirstsportfrequ * data15$vfirstsporttime + data15$vlightsport*data15$vsportornot_98 + data15$vmeansport*data15$vsportornot_98 + data15$vheavysport*data15$vsportornot_98 + data15$vvioltsport*data15$vsportornot_98
##met group 7.5< 7.5-21 >21
data15$metgroup<-NA
data15$metgroup[data15$met_hr<7.5]<-1
data15$metgroup[data15$met_hr>=7.5 & data15$met_hr<21]<-2
data15$metgroup[data15$met_hr>=21]<-3
table(data15$metgroup)
###class_g_bmi class 1=normal 2=underweight 3=overweight 4=obesity
overweight_boy<-c(14.8,18.3,17.4,17.0,16.7,16.7,16.9,17.9,19.0,19.5,20.0,20.7,21.3,21.9,22.5,22.9,23.3,25.5)
underweight_boy<-c(11.5,14.8,14.2,13.7,13.4,13.3,13.5,13.8,14.1,14.3,14.5,14.8,15.2,15.7,16.3,16.9,17.4,17.8)
obesity_boy<-c(15.8,19.2,18.3,17.8,17.6,17.7,18.5,20.3,21.6,22.3,22.7,23.2,23.9,24.5,25.0,25.4,25.6,25.6)
overweight_girl<-c(14.7,17.9,17.2,16.9,16.8,17.0,17.2,17.7,18.4,19.1,19.7,20.5,21.3,21.9,22.5,22.7,22.7,22.7)
underweight_girl<-c(11.5,14.2,13.7,13.5,13.2,13.1,13.1,13.4,13.8,14.0,14.3,14.7,15.2,15.7,16.3,16.7,17.1,17.3)
obesity_girl<-c(15.5,19.0,18.1,17.8,17.9,18.1,18.8,19.6,20.7,21.3,22.0,22.7,23.5,24.3,24.9,25.2,25.3,25.3)
x<-data.frame(age=c(0:17),underweight_boy,overweight_boy,obesity_boy,underweight_girl,overweight_girl,obesity_girl)
littleboy<-data15[data15$gender==1 & data15$age<18,c("age","g_bmi")]
littlegirl<-data15[data15$gender==2 & data15$age<18,c("age","g_bmi")]
##計算0-17歲男性體重分組
littleboy$class_g_bmi<-rep(1,nrow(littleboy))
littleboy$class_g_bmi[littleboy$g_bmi<x[littleboy$age,"underweight_boy"]]<-2
littleboy$class_g_bmi[littleboy$g_bmi>=x[littleboy$age,"overweight_boy"] & littleboy$g_bmi<x[littleboy$age,"obesity_boy"]]<-3
littleboy$class_g_bmi[littleboy$g_bmi>=x[(littleboy$age),"obesity_boy"] ]<-4
table(littleboy$class_g_bmi)
##計算0-17歲女性體重分組
littlegirl$class_g_bmi<-rep(1,nrow(littlegirl))
littlegirl$class_g_bmi[littlegirl$g_bmi<x[littlegirl$age,"underweight_girl"]]<-2
littlegirl$class_g_bmi[littlegirl$g_bmi>=x[littlegirl$age,"overweight_girl"] & littlegirl$g_bmi<x[littlegirl$age,"obesity_girl"]]<-3
littlegirl$class_g_bmi[littlegirl$g_bmi>=x[littlegirl$age,"obesity_girl"] ]<-4
data15$class_g_bmi<-rep(1,nrow(data15))
data15[data15$gender==1 & data15$age<18,'class_g_bmi']<-littleboy$class_g_bmi
data15[data15$gender==2 & data15$age<18,'class_g_bmi']<-littlegirl$class_g_bmi
data15[data15$g_bmi<18.5 & data15$age>=18,'class_g_bmi']<-2
data15[data15$g_bmi>=24 & data15$g_bmi <27 & data15$age>=18,'class_g_bmi']<-3
data15[data15$g_bmi>=27 & data15$age>=18,'class_g_bmi']<-4
table(data15$class_g_bmi)
##wc分組男1 女2
data15$class_wc<-0
data15$class_wc[data15$gender==1 & data15$g_wc>=90]<-1
data15$class_wc[data15$gender==2 & data15$g_wc>=80]<-1
##定義名稱
names(data15)[187:190]<-c("o3","NO2","SO2","pm10")
data16<-data15
names(data16)[191:193]<-c("new_asthma","new_asthma_drug","asthma_finally")
names(data16)[c(29,30,31,32,34,41,42,44,45)]<-c("HTN_drug","DM_drug","hyperlipidemic_drug",
"asthma_drug","Psychiatric_drug","HTN","DM","CVD","asthma")
##疾病填補根據之前或之後回答來判斷
HTN<-list()
for(i in 1:length(data16$pid[is.na(data16$HTN)])){
HTN[[i]]<-data16[data16$pid==unique(data16$pid[is.na(data16$HTN)])[i],c("HTN")]
}
sum(is.na(data16$HTN))
DM<-list()
for(i in 1:length(data16$pid[is.na(data16$DM)])){
DM[[i]]<-data16[data16$pid==unique(data16$pid[is.na(data16$DM)])[i],c("DM")]
}
sum(is.na(data16$DM))
CVD<-list()
for(i in 1:length(data16$pid[is.na(data16$CVD)])){
CVD[[i]]<-data16[data16$pid==unique(data16$pid[is.na(data16$CVD)])[i],c("CVD")]
}
sum(is.na(data16$CVD))
HTN_drug<-list()
for(i in 1:length(data16$pid[is.na(data16$HTN_drug)])){
HTN_drug[[i]]<-data16[data16$pid==unique(data16$pid[is.na(data16$HTN_drug)])[i],c("HTN_drug")]
}
DM_drug<-list()
for(i in 1:length(data16$pid[is.na(data16$DM_drug)])){
DM_drug[[i]]<-data16[data16$pid==unique(data16$pid[is.na(data16$DM_drug)])[i],c("DM_drug")]
}
hyperlipidemic_drug<-list()
for(i in 1:length(data16$pid[is.na(data16$hyperlipidemic_drug)])){
hyperlipidemic_drug[[i]]<-data16[data16$pid==unique(data16$pid[is.na(data16$hyperlipidemic_drug)])[i],c("hyperlipidemic_drug")]
}
Psychiatric_drug<-list()
for(i in 1:length(data16$pid[is.na(data16$Psychiatric_drug)])){
Psychiatric_drug[[i]]<-data16[data16$pid==unique(data16$pid[is.na(data16$Psychiatric_drug)])[i],c("Psychiatric_drug")]
}
data16$HTN[is.na(data16$HTN)]<-rep(0,sum(is.na(data16$HTN)))
data16$DM[is.na(data16$DM)]<-rep(0,sum(is.na(data16$DM)))
data16$CVD[is.na(data16$CVD)]<-rep(0,sum(is.na(data16$CVD)))
data16$HTN_drug[is.na(data16$HTN_drug)]<-rep(0,sum(is.na(data16$HTN_drug)))
data16$DM_drug[is.na(data16$DM_drug)]<-rep(0,sum(is.na(data16$DM_drug)))
data16$hyperlipidemic_drug[is.na(data16$hyperlipidemic_drug)]<-rep(0,sum(is.na(data16$hyperlipidemic_drug)))
data16$Psychiatric_drug[is.na(data16$Psychiatric_drug)]<-rep(0,sum(is.na(data16$Psychiatric_drug)))
##將年齡分組 <20 20-59 >60
data16$agegroup[data16$age<20]<-1
data16$agegroup[data16$age>=20 & data16$age<60]<-2
data16$agegroup[data16$age>=60 ]<-3
##呼吸道
data16$FEV1_FVC<-data16$ratio_lung_function*100
##分成氣喘以及非氣喘
all<-data16[,c("pid","yr","mon","n","nn","gender","age","agegroup","education","HTN_drug","DM_drug","hyperlipidemic_drug",
"Psychiatric_drug","HTN","DM","CVD","g_wc","g_bmi","ratio_lung_function",
"FEV1_FVC","o3","NO2","SO2","pm10","class_g_bmi","class_wc","age_class","class_education","smoke","met_hr",
"metgroup","new_asthma","new_asthma_drug","asthma_finally")]
all<-data.frame(all)
all$pid<-as.factor(all$pid)
all$agegroup<-as.factor(all$agegroup)
all$gender<-as.factor(all$gender)
all$class_g_bmi<-as.factor(all$class_g_bmi)
all$class_wc<-as.factor(all$class_wc)
all$class_education<-as.factor(all$class_education)
all$HTN<-as.factor(all$HTN)
all$DM<-as.factor(all$DM)
all$CVD<-as.factor(all$CVD)
all$HTN_drug<-as.factor(all$HTN_drug)
all$DM_drug<-as.factor(all$DM_drug)
all$DM_drug<-as.factor(all$DM_drug)
all$hyperlipidemic_drug<-as.factor(all$hyperlipidemic_drug)
all$Psychiatric_drug<-as.factor(all$Psychiatric_drug)
all$smoke<-as.factor(all$smoke)
all$metgroup<-as.factor(all$metgroup)
all<-all[order(all$pid),]
##將資料分為2組
asthma<-all[all$asthma_finally==1,]
health<-all[all$asthma_finally==0,]
##對於氣喘與非氣喘組的環境濃度變數進行分組
##asthma
asthma$class_pm10[asthma$pm10<as.numeric(summary(asthma$pm10)[2])]<-1
asthma$class_pm10[asthma$pm10>=as.numeric(summary(asthma$pm10)[2]) & asthma$pm10<as.numeric(summary(asthma$pm10)[3]) ]<-2
asthma$class_pm10[asthma$pm10>=as.numeric(summary(asthma$pm10)[3]) & asthma$pm10<as.numeric(summary(asthma$pm10)[5]) ]<-3
asthma$class_pm10[asthma$pm10>=as.numeric(summary(asthma$pm10)[5])]<-4
asthma$class_o3[asthma$o3<as.numeric(summary(asthma$o3)[2])]<-1
asthma$class_o3[asthma$o3>=as.numeric(summary(asthma$o3)[2]) & asthma$o3<as.numeric(summary(asthma$o3)[3]) ]<-2
asthma$class_o3[asthma$o3>=as.numeric(summary(asthma$o3)[3])& asthma$o3<as.numeric(summary(asthma$o3)[5]) ]<-3
asthma$class_o3[asthma$o3>=as.numeric(summary(asthma$o3)[5])]<-4
asthma$class_no2[asthma$NO2<as.numeric(summary(asthma$NO2)[2])]<-1
asthma$class_no2[asthma$NO2>=as.numeric(summary(asthma$NO2)[2]) & asthma$NO2<as.numeric(summary(asthma$NO2)[3]) ]<-2
asthma$class_no2[asthma$NO2>=as.numeric(summary(asthma$NO2)[3]) & asthma$NO2<as.numeric(summary(asthma$NO2)[5]) ]<-3
asthma$class_no2[asthma$NO2>=as.numeric(summary(asthma$NO2)[5])]<-4
asthma$class_so2[asthma$SO2<as.numeric(summary(asthma$SO2)[2])]<-1
asthma$class_so2[asthma$SO2>=as.numeric(summary(asthma$SO2)[2]) & asthma$SO2<as.numeric(summary(asthma$SO2)[3]) ]<-2
asthma$class_so2[asthma$SO2>=as.numeric(summary(asthma$SO2)[3]) & asthma$SO2<as.numeric(summary(asthma$SO2)[5]) ]<-3
asthma$class_so2[asthma$SO2>=as.numeric(summary(asthma$SO2)[5])]<-4
asthma$class_pm10<-as.factor(asthma$class_pm10)
asthma$class_o3<-as.factor(asthma$class_o3)
asthma$class_no2<-as.factor(asthma$class_no2)
asthma$class_so2<-as.factor(asthma$class_so2)
## Q1=77 Q2=83 Q3=89 男性
summary(asthma$g_wc[asthma$gender==1])
## Q1=67 Q2=71 Q3=78 女性
summary(asthma$g_wc[asthma$gender==2])
##wc 按照性別不同 分成Q1 Q2 Q3
asthma$new_wc[asthma$gender==1 & asthma$g_wc<as.numeric(summary(asthma$g_wc[asthma$gender==1])[2])]<-1
asthma$new_wc[asthma$gender==2 & asthma$g_wc<as.numeric(summary(asthma$g_wc[asthma$gender==2])[2]) ]<-1
asthma$new_wc[asthma$gender==1 & asthma$g_wc>=as.numeric(summary(asthma$g_wc[asthma$gender==1])[2]) & asthma$g_wc <as.numeric(summary(asthma$g_wc[asthma$gender==1])[3])]<-2
asthma$new_wc[asthma$gender==2 & asthma$g_wc>=as.numeric(summary(asthma$g_wc[asthma$gender==2])[2]) & asthma$g_wc <as.numeric(summary(asthma$g_wc[asthma$gender==2])[3])]<-2
asthma$new_wc[asthma$gender==1 & asthma$g_wc>=as.numeric(summary(asthma$g_wc[asthma$gender==1])[3]) & asthma$g_wc <as.numeric(summary(asthma$g_wc[asthma$gender==1])[5])]<-3
asthma$new_wc[asthma$gender==2 & asthma$g_wc>=as.numeric(summary(asthma$g_wc[asthma$gender==2])[3]) & asthma$g_wc <as.numeric(summary(asthma$g_wc[asthma$gender==2])[5])]<-3
asthma$new_wc[asthma$gender==1 & asthma$g_wc>=as.numeric(summary(asthma$g_wc[asthma$gender==1])[5]) ]<-4
asthma$new_wc[asthma$gender==2 & asthma$g_wc>=as.numeric(summary(asthma$g_wc[asthma$gender==2])[5]) ]<-4
asthma$new_wc<-as.factor(asthma$new_wc)
table(asthma$new_wc)
##Health
##asthma
health$class_pm10[health$pm10<as.numeric(summary(health$pm10)[2])]<-1
health$class_pm10[health$pm10>=as.numeric(summary(health$pm10)[2]) & health$pm10<as.numeric(summary(health$pm10)[3]) ]<-2
health$class_pm10[health$pm10>=as.numeric(summary(health$pm10)[3]) & health$pm10<as.numeric(summary(health$pm10)[5]) ]<-3
health$class_pm10[health$pm10>=as.numeric(summary(health$pm10)[5])]<-4
health$class_o3[health$o3<as.numeric(summary(health$o3)[2])]<-1
health$class_o3[health$o3>=as.numeric(summary(health$o3)[2]) & health$o3<as.numeric(summary(health$o3)[3]) ]<-2
health$class_o3[health$o3>=as.numeric(summary(health$o3)[3])& health$o3<as.numeric(summary(health$o3)[5]) ]<-3
health$class_o3[health$o3>=as.numeric(summary(health$o3)[5])]<-4
health$class_no2[health$NO2<as.numeric(summary(health$NO2)[2])]<-1
health$class_no2[health$NO2>=as.numeric(summary(health$NO2)[2]) & health$NO2<as.numeric(summary(health$NO2)[3]) ]<-2
health$class_no2[health$NO2>=as.numeric(summary(health$NO2)[3]) & health$NO2<as.numeric(summary(health$NO2)[5]) ]<-3
health$class_no2[health$NO2>=as.numeric(summary(health$NO2)[5])]<-4
health$class_so2[health$SO2<as.numeric(summary(health$SO2)[2])]<-1
health$class_so2[health$SO2>=as.numeric(summary(health$SO2)[2]) & health$SO2<as.numeric(summary(health$SO2)[3]) ]<-2
health$class_so2[health$SO2>=as.numeric(summary(health$SO2)[3]) & health$SO2<as.numeric(summary(health$SO2)[5]) ]<-3
health$class_so2[health$SO2>=as.numeric(summary(health$SO2)[5])]<-4
health$class_pm10<-as.factor(health$class_pm10)
health$class_o3<-as.factor(health$class_o3)
health$class_no2<-as.factor(health$class_no2)
health$class_so2<-as.factor(health$class_so2)
## Q1=77 Q2=83 Q3=89 男性
summary(health$g_wc[health$gender==1])
## Q1=66 Q2=70 Q3=76 女性
summary(health$g_wc[health$gender==2])
##wc 按照性別不同 分成Q1 Q2 Q3
health$new_wc[health$gender==1 & health$g_wc<as.numeric(summary(health$g_wc[health$gender==1])[2])]<-1
health$new_wc[health$gender==2 & health$g_wc<as.numeric(summary(health$g_wc[health$gender==2])[2]) ]<-1
health$new_wc[health$gender==1 & health$g_wc>=as.numeric(summary(health$g_wc[health$gender==1])[2]) & health$g_wc <as.numeric(summary(health$g_wc[health$gender==1])[3])]<-2
health$new_wc[health$gender==2 & health$g_wc>=as.numeric(summary(health$g_wc[health$gender==2])[2]) & health$g_wc <as.numeric(summary(health$g_wc[health$gender==2])[3])]<-2
health$new_wc[health$gender==1 & health$g_wc>=as.numeric(summary(health$g_wc[health$gender==1])[3]) & health$g_wc <as.numeric(summary(health$g_wc[health$gender==1])[5])]<-3
health$new_wc[health$gender==2 & health$g_wc>=as.numeric(summary(health$g_wc[health$gender==2])[3]) & health$g_wc <as.numeric(summary(health$g_wc[health$gender==2])[5])]<-3
health$new_wc[health$gender==1 & health$g_wc>=as.numeric(summary(health$g_wc[health$gender==1])[5]) ]<-4
health$new_wc[health$gender==2 & health$g_wc>=as.numeric(summary(health$g_wc[health$gender==2])[5]) ]<-4
health$new_wc<-as.factor(health$new_wc)
table(health$new_wc)
##analysis-----------------------
var<-c("class_g_bmi","class_wc","gender","agegroup","class_education","class_pm10","class_so2",
"class_no2","HTN","DM","CVD","HTN_drug","DM_drug","hyperlipidemic_drug","Psychiatric_drug","smoke","metgroup")
"new_wc"
for(i in 1:length(var)){
formula<-formula(sprintf("FEV1_FVC~%2s",var[i]))
model<-geeglm(formula=formula,data=asthma,id=pid,corstr = "exchangeable")
print(summary(model)$coefficient)
}
##
new<-model.matrix(~class_g_bmi+gender+agegroup+class_education+HTN+DM+CVD+HTN_drug+hyperlipidemic_drug+
Psychiatric_drug+class_pm10+class_so2+class_no2+smoke+metgroup+class_g_bmi*gender,data=asthma)
new<-new[,-1]
new<-data.frame(new)
new$pid<-asthma$pid
new$FEV1_FVC<-asthma$FEV1_FVC
a<-(names(new)[1:28])
##找出本來會放入的變數
fit<-geeglm(FEV1_FVC~class_g_bmi2+class_g_bmi3+class_g_bmi4+gender2+agegroup2+agegroup3+class_education2+
class_education3+HTN1+DM1+HTN_drug1+hyperlipidemic_drug1+Psychiatric_drug1+class_pm102+
class_pm103+class_pm104+class_so22+class_so23+class_so24+class_no22+class_no23+class_no24+
smoke2+smoke3+smoke4+metgroup2+metgroup3,data=new,id=pid,corstr ="exchangeable")
model<-geeglm(FEV1_FVC~class_g_bmi+gender+agegroup+class_education+HTN+DM+CVD+HTN_drug+hyperlipidemic_drug+
Psychiatric_drug+class_pm10+class_so2+class_no2+smoke+metgroup,data=asthma,id=pid,corstr ="exchangeable")
summary(model)
gee_stepper(model,formula(model))
##判斷VIF
m<-lm(FEV1_FVC~class_g_bmi+gender+agegroup+class_education+class_pm10+class_no2+metgroup+hyperlipidemic_drug+HTN,data=asthma)
library(car)
vif(m)
##進行單變量交互項檢查
##EV1_FVC~class_g_bmi+gender+agegroup+class_education+class_pm10+class_no2+metgroup+hyperlipidemic_drug+HTN
##BMI CHECK
var<-c("gender","agegroup","class_education","class_pm10","class_no2","HTN_drug","hyperlipidemic_drug","smoke","metgroup")
for (i in 1:length(var)) {
formula<-formula(sprintf("FEV1_FVC~class_g_bmi*%2s",var[i]))
model<-geeglm(formula=formula,data=asthma,id=pid,corstr = "exchangeable")
print(summary(model))
}
# agegroup class_education class_no2 HTN_drug hyperlipidemic_drug
#gender check
var1<-c("agegroup","class_education","class_pm10","class_no2","HTN_drug","hyperlipidemic_drug","smoke","metgroup")
for (i in 1:length(var)) {
formula<-formula(sprintf("FEV1_FVC~gender*%2s",var[i]))
model<-geeglm(formula=formula,data=asthma,id=pid,corstr = "exchangeable")
print(summary(model))
}
## agegroup class_education smoke
##agegroup check
var2<-c("class_education","class_pm10","class_no2","HTN_drug","hyperlipidemic_drug","smoke","metgroup")
for (i in 1:length(var2)) {
formula<-formula(sprintf("FEV1_FVC~agegroup*%2s",var2[i]))
model<-geeglm(formula=formula,data=asthma,id=pid,corstr = "exchangeable")
print(summary(model))
}
##class_no2 HTN_drug hyperlipidemic_drug
##class_education check
var3<-c("class_pm10","class_no2","HTN_drug","hyperlipidemic_drug","smoke","metgroup")
for (i in 1:length(var3)) {
formula<-formula(sprintf("FEV1_FVC~class_education*%2s",var3[i]))
model<-geeglm(formula=formula,data=asthma,id=pid,corstr = "exchangeable")
print(summary(model))
}
## class_pm10 HTN_drug hyperlipidemic_drug
## class_pm10 check
var4<-c("class_no2","HTN_drug","hyperlipidemic_drug","smoke","metgroup")
for (i in 1:length(var4)) {
formula<-formula(sprintf("FEV1_FVC~class_pm10*%2s",var4[i]))
model<-geeglm(formula=formula,data=asthma,id=pid,corstr = "exchangeable")
print(summary(model))
}
##class_no2 smoke
## class_no2 check
var5<-c("HTN_drug","hyperlipidemic_drug","smoke","metgroup")
for (i in 1:length(var5)) {
formula<-formula(sprintf("FEV1_FVC~class_no2*%2s",var5[i]))
model<-geeglm(formula=formula,data=asthma,id=pid,corstr = "exchangeable")
print(summary(model))
}
##HTN_drug
##HTN_drug check
var6<-c("hyperlipidemic_drug","smoke","metgroup")
for (i in 1:length(var6)) {
formula<-formula(sprintf("FEV1_FVC~HTN_drug*%2s",var6[i]))
model<-geeglm(formula=formula,data=asthma,id=pid,corstr = "exchangeable")
print(summary(model))
}
## hyperlipidemic_drug
## hyperlipidemic_drug check
model<-geeglm(FEV1_FVC~hyperlipidemic_drug*metgroup,data=asthma,id=pid,corstr = "exchangeable")
print(summary(model))
##smoke check
model<-geeglm(FEV1_FVC~smoke*metgroup,data=asthma,id=pid,corstr = "exchangeable")
print(summary(model))