臨床・疫学研究
RとEZRでデータ解析の実現
質的変数–数字で測れない
量的変数–数字で測れる
中心を表す量:
平均値 (mean);中央値(median);最頻値(mode)
実演: library(ShinyIntroStats) -> intro_stats_shinyapps() [3]
バラツキを表す量:
中央値と四分位範囲のペア
平均値と標準偏差
サンプルサイズ \( \Uparrow \) \( \longrightarrow \) 信頼区間の幅 \( \Downarrow \)
リスク比やオッズ比など,95%信頼区間には 1 が含まれると,統計学的に有意ではない.
実演:run shiny(confidence intervals)
データ種類 | 二値変数 | 連続変数 | 生存期間 |
---|---|---|---|
要約 | 分割表 | ヒストグラム 箱ひげ図 散布図 |
Kaplan-Meier 曲線 |
2群比較 | Fisher 正確検定 カイ二乗検定 |
t 検定 Man-Whitney U 検定 |
logrank検定 一般化 Wilcoxon 検定 |
対応のある2群比較 | McNemar 検定 | 対応のある t 検定 Wilcoxon 符号付順位和検定 |
|
3群以上の比較 | Fisher 正確検定 カイ二乗検定 |
分散分析 (ANOVA) Kruskal-Wallis 検定 |
logrank検定 一般化 Kruskal-Wallis 検定 |
対応のある3群以上の比較 | Cochran R 検定 | 反復測定分散分析 Friedman検定 |
|
(多変量) 回帰分析 | ロジスティクス回帰 | 単回帰・重回帰 | Cox比例ハザード回帰 |
daily.intake.kJ <- c(5260, 5470, 5640, 6180, 6390, 6515, 6805, 7515, 7515, 8230, 8770)
#毎日推奨エネルギー摂取量との違いを検定する
t.test(daily.intake.kJ*0.239, mu=2000)#Kj->Calの変換が必要
One Sample t-test
data: daily.intake.kJ * 0.239
t = -4.6886, df = 10, p-value = 0.0008563
alternative hypothesis: true mean is not equal to 2000
95 percent confidence interval:
1430.737 1797.501
sample estimates:
mean of x
1614.119
library(ISwR);attach(energy)#データセットをローディング
t.test(expend ~ stature, var.equal=TRUE) # "~"の符号はstatureによりグループ分けの意味
Two Sample t-test
data: expend by stature
t = -3.9456, df = 20, p-value = 0.000799
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
-3.411451 -1.051796
sample estimates:
mean in group lean mean in group obese
8.066154 10.297778
#差の95%信頼区間は0を含まれないため(p < 0.05),肥満と痩せのエネルギー消費量は統計学的に有意な違いがある.
Sample Statin..No Statin..Yes
1 Number 6058 3049
2 Glucose 9.4(5.1) 9.2(5.3)
require(lessR) #函数をローディング
tt.brief(n1 = 6058, m1 = 9.4, s1 = 5.1, n2 = 3049, m2 = 9.2, s2 = 5.3)
Compare Y across X levels Group1 and Group2
------------------------------------------------------------
Y for X Group1: n = 6058, mean = 9.4, sd = 5.1
Y for X Group2: n = 3049, mean = 9.2, sd = 5.3
---
t-cutoff: tcut = 1.960
Standard Error of Mean Difference: SE = 0.11
Hypothesis Test of 0 Mean Diff: t = 1.743, df = 9105, p-value = 0.081
Margin of Error for 95% Confidence Level: 0.22
95% Confidence Interval for Mean Difference: -0.02 to 0.42
Sample Mean Difference of Y: 0.20
Standardized Mean Difference of Y, Cohen's d: 0.04
library(ISwR);attach(intake)#データセットをローディング
post-pre #月経後のエネルギー摂取は月経前より低い
[1] -1350 -1250 -1755 -1020 -745 -1835 -1540 -1540 -725 -1330 -1435
t.test(pre, post, paired = TRUE) #自分の対応があるので,pairをTRUEに指定
Paired t-test
data: pre and post
t = 11.941, df = 10, p-value = 0.0000003059
alternative hypothesis: true difference in means is not equal to 0
95 percent confidence interval:
1074.072 1566.838
sample estimates:
mean of the differences
1320.455
#月経前後のエネルギーの差の95%信頼区間には,0が含まれていないので,統計学的に有意な違いがある.
M <- matrix(c(23, 7, 18, 13), 2 ,2)#データ入力
colnames(M) <- c("治癒","未治癒")#列の名前
rownames(M) <- c("薬A","薬B")#行の名前
addmargins(M)#観察値(O)
治癒 未治癒 Sum
薬A 23 18 41
薬B 7 13 20
Sum 30 31 61
addmargins(chisq.test(M)$expected)#期待値(E)
治癒 未治癒 Sum
薬A 20.163934 20.83607 41
薬B 9.836066 10.16393 20
Sum 30.000000 31.00000 61
chisq.test(M, correct = FALSE) #カイ二乗検定を行う
Pearson's Chi-squared test
data: M
X-squared = 2.394, df = 1, p-value = 0.1218
E <- chisq.test(M)$expected#期待値
O <- chisq.test(M)$observed#観察値
(O - E)^2/E #カイ二乗統計量の計算
治癒 未治癒
薬A 0.3988938 0.3860262
薬B 0.8177322 0.7913538
0.3988938 + 0.3860262 + 0.8177322 + 0.7913538
[1] 2.394006
#自由度(df) 1 のp値は0.05より大きいので,治癒率に差があると言えない.
library(RcmdrPlugin.EZR)
.Table <- matrix(c(36, 76, 15, 455), 2,2, byrow = T)
epi.tests(.Table, conf.level = 0.95)
Disease positive Disease negative Total
Test positive 36 76 112
Test negative 15 455 470
Total 51 531 582
Point estimates and 95 % CIs:
---------------------------------------------------------
Estimation Lower CI Upper CI
Apparent prevalence 0.192 0.161 0.227
True prevalence 0.088 0.066 0.114
Sensitivity 0.706 0.562 0.825
Specificity 0.857 0.824 0.886
Positive predictive value 0.321 0.236 0.416
Negative predictive value 0.968 0.948 0.982
Diagnstic accuracy 0.844 0.812 0.872
Likelihood ratio of a positive test 4.932 3.752 6.482
Likelihood ratio of a negative test 0.343 0.224 0.526
---------------------------------------------------------
検査前オッズ \( \times \) 尤度比 \( = \) 検査後オッズ
確率とオッズの変換:
検査前確率:
(陽性)尤度比の計算: \[ LR+ = \frac{\frac{a}{a+c}}{\frac{b}{b+d}} \]
AUC(Area Under the roc Curve)曲線下面積が大きいほど良い検査法.
実演:library(plotROC); shiny_plotROC()
set.seed(2529)
D.ex <- rbinom(200, size = 1, prob = .5)
M1 <- rnorm(200, mean = D.ex, sd = .65)
M2 <- rnorm(200, mean = D.ex, sd = 1.5)
test <- data.frame(D = D.ex, D.str = c("Healthy", "Ill")[D.ex + 1],
M1 = M1, M2 = M2, stringsAsFactors = FALSE)
① ②
レベル | 研究種類 |
---|---|
1a | ランダム化比較試験のメタアナリシス |
1b | 少なくとも一つのランダム化比較試験 |
2 | コホート研究(前向きが多い) |
3 | ケース・コントロール研究(後ろ向きが多い) |
4 | 処置前後の比較などの前後比較,対照群を伴わない研究 |
5 | 症例報告,ケースシリーズ |
6 | 専門家個人の意見(専門家委員会報告を含む) |
治癒 未治癒 Sum
薬 23 18 41
プラセボ 7 13 20
Sum 30 31 61
指標
リスク:割合
リスク差: Risk Difference \( = \) \( p_1 - p_2 = \) 0.2109756
治癒 未治癒 Sum
薬 23 18 41
プラセボ 7 13 20
Sum 30 31 61
治癒 未治癒 Sum
薬 23 18 41
プラセボ 7 13 20
Sum 30 31 61
prop.diff.conf(23, 41, 7, 20, 95) #リスク差の点推定値と信頼区間
[1] Difference : 0.211
[1] 95% confidence interval : -0.047 - 0.469
prop.ratio.conf(23, 41, 7, 20, 95) #リスク比の点推定値と信頼区間
[1] Ratio : 1.603
[1] 95% confidence interval : 0.832 - 3.088
治癒 未治癒 Sum
薬 23 18 41
プラセボ 7 13 20
Sum 30 31 61
Fisher's Exact Test for Count Data
data: M
p-value = 0.1737
alternative hypothesis: true odds ratio is not equal to 1
95 percent confidence interval:
0.6936416 8.4948588
sample estimates:
odds ratio
2.339104
library(ISwR);library(survival);attach(stroke)#データセットをローディング
Surv.stroke <- survfit(Surv(obsmonths, dead) ~ 1)
summary(Surv.stroke)
Call: survfit(formula = Surv(obsmonths, dead) ~ 1)
time n.risk n.event survival std.err lower 95% CI upper 95% CI
0.0327 829 19 0.977 0.00520 0.967 0.987
0.0654 810 23 0.949 0.00762 0.935 0.964
0.0980 787 13 0.934 0.00864 0.917 0.951
0.1000 774 25 0.903 0.01026 0.884 0.924
0.1307 749 9 0.893 0.01075 0.872 0.914
0.1634 740 7 0.884 0.01111 0.863 0.906
0.1961 733 13 0.869 0.01174 0.846 0.892
0.2288 720 13 0.853 0.01230 0.829 0.877
0.2614 707 17 0.832 0.01297 0.807 0.858
0.2941 690 13 0.817 0.01344 0.791 0.843
0.3268 677 13 0.801 0.01387 0.774 0.829
0.3595 664 8 0.791 0.01411 0.764 0.819
0.3922 656 10 0.779 0.01440 0.752 0.808
0.4248 646 8 0.770 0.01462 0.741 0.799
0.4575 638 6 0.762 0.01478 0.734 0.792
0.4902 632 4 0.758 0.01488 0.729 0.787
0.5229 628 5 0.752 0.01501 0.723 0.782
0.5556 623 2 0.749 0.01506 0.720 0.779
0.5882 621 3 0.745 0.01513 0.716 0.776
0.6209 618 1 0.744 0.01515 0.715 0.775
0.6536 617 4 0.739 0.01524 0.710 0.770
0.6863 613 5 0.733 0.01536 0.704 0.764
0.7190 608 2 0.731 0.01540 0.701 0.762
0.7516 606 4 0.726 0.01549 0.696 0.757
0.7843 602 6 0.719 0.01561 0.689 0.750
0.8170 596 6 0.712 0.01573 0.682 0.743
0.8497 590 1 0.710 0.01575 0.680 0.742
0.8824 589 4 0.706 0.01583 0.675 0.737
0.9150 585 2 0.703 0.01587 0.673 0.735
0.9477 583 1 0.702 0.01588 0.672 0.734
0.9804 582 3 0.698 0.01594 0.668 0.730
1.0131 579 1 0.697 0.01596 0.667 0.729
1.0458 578 1 0.696 0.01598 0.665 0.728
1.0784 577 4 0.691 0.01605 0.660 0.723
1.1111 573 4 0.686 0.01611 0.656 0.719
1.1765 569 3 0.683 0.01616 0.652 0.715
1.2418 566 3 0.679 0.01621 0.648 0.712
1.2745 563 3 0.676 0.01626 0.644 0.708
1.3072 560 3 0.672 0.01631 0.641 0.705
1.3399 557 2 0.669 0.01634 0.638 0.702
1.3725 555 2 0.667 0.01637 0.636 0.700
1.4052 553 1 0.666 0.01638 0.635 0.699
1.4379 552 2 0.663 0.01641 0.632 0.696
1.5686 550 2 0.661 0.01644 0.630 0.694
1.6340 548 3 0.657 0.01648 0.626 0.691
1.6667 545 3 0.654 0.01652 0.622 0.687
1.7320 542 1 0.653 0.01654 0.621 0.686
1.7974 541 1 0.651 0.01655 0.620 0.685
1.9281 540 1 0.650 0.01656 0.619 0.683
1.9608 539 1 0.649 0.01658 0.617 0.682
2.0915 538 1 0.648 0.01659 0.616 0.681
2.1569 537 1 0.647 0.01660 0.615 0.680
2.1895 536 1 0.645 0.01662 0.614 0.679
2.2222 535 2 0.643 0.01664 0.611 0.676
2.2549 533 1 0.642 0.01665 0.610 0.675
2.3203 532 1 0.641 0.01667 0.609 0.674
2.3856 531 1 0.639 0.01668 0.607 0.673
2.4510 530 1 0.638 0.01669 0.606 0.672
2.4837 529 1 0.637 0.01670 0.605 0.671
2.5163 528 1 0.636 0.01671 0.604 0.669
2.5817 527 2 0.633 0.01674 0.601 0.667
2.7451 525 1 0.632 0.01675 0.600 0.666
2.7778 524 2 0.630 0.01677 0.598 0.663
2.8105 522 1 0.628 0.01678 0.596 0.662
2.8431 521 1 0.627 0.01679 0.595 0.661
2.8758 520 1 0.626 0.01680 0.594 0.660
2.9085 519 2 0.624 0.01683 0.592 0.658
3.0719 517 1 0.622 0.01684 0.590 0.656
3.1046 516 1 0.621 0.01685 0.589 0.655
3.3007 515 1 0.620 0.01686 0.588 0.654
3.3333 514 2 0.618 0.01688 0.585 0.652
3.4641 512 1 0.616 0.01689 0.584 0.650
3.6275 511 1 0.615 0.01690 0.583 0.649
3.7582 510 1 0.614 0.01691 0.582 0.648
3.8562 509 1 0.613 0.01692 0.581 0.647
3.9216 508 2 0.610 0.01694 0.578 0.644
3.9542 506 2 0.608 0.01696 0.576 0.642
4.0196 504 1 0.607 0.01697 0.574 0.641
4.2157 503 1 0.606 0.01697 0.573 0.640
4.2810 502 1 0.604 0.01698 0.572 0.639
4.3464 501 1 0.603 0.01699 0.571 0.637
4.3791 500 1 0.602 0.01700 0.570 0.636
4.6078 499 2 0.600 0.01702 0.567 0.634
4.6405 497 2 0.597 0.01704 0.565 0.631
4.6732 495 1 0.596 0.01704 0.563 0.630
4.7059 494 2 0.593 0.01706 0.561 0.628
4.7386 492 1 0.592 0.01707 0.560 0.627
4.8039 491 1 0.591 0.01708 0.559 0.626
4.9346 490 1 0.590 0.01708 0.557 0.624
5.0654 489 1 0.589 0.01709 0.556 0.623
5.5229 488 1 0.587 0.01710 0.555 0.622
5.5556 487 1 0.586 0.01711 0.554 0.621
5.5882 486 1 0.585 0.01711 0.552 0.620
5.6209 485 1 0.584 0.01712 0.551 0.618
5.6536 484 1 0.583 0.01713 0.550 0.617
6.0131 483 1 0.581 0.01713 0.549 0.616
6.0458 482 1 0.580 0.01714 0.548 0.615
6.1765 481 1 0.579 0.01715 0.546 0.614
6.4379 480 1 0.578 0.01715 0.545 0.612
6.4706 479 1 0.577 0.01716 0.544 0.611
6.6013 478 1 0.575 0.01717 0.543 0.610
6.6340 477 1 0.574 0.01717 0.541 0.609
7.0915 476 1 0.573 0.01718 0.540 0.608
7.2222 475 3 0.569 0.01720 0.537 0.604
7.2549 472 1 0.568 0.01720 0.535 0.603
7.3203 471 2 0.566 0.01721 0.533 0.601
7.4837 469 1 0.565 0.01722 0.532 0.599
7.5490 468 1 0.563 0.01723 0.531 0.598
7.6144 467 2 0.561 0.01724 0.528 0.596
7.7451 465 1 0.560 0.01724 0.527 0.595
8.0719 464 1 0.559 0.01725 0.526 0.593
8.4967 463 1 0.557 0.01725 0.524 0.592
8.5294 462 1 0.556 0.01726 0.523 0.591
9.0196 461 1 0.555 0.01726 0.522 0.590
9.3464 460 1 0.554 0.01727 0.521 0.589
9.5098 459 2 0.551 0.01727 0.518 0.586
9.9346 457 1 0.550 0.01728 0.517 0.585
10.0980 456 1 0.549 0.01728 0.516 0.584
10.1961 455 1 0.548 0.01729 0.515 0.583
10.2614 454 1 0.546 0.01729 0.514 0.581
10.2941 453 1 0.545 0.01729 0.512 0.580
10.4575 452 1 0.544 0.01730 0.511 0.579
10.8497 451 1 0.543 0.01730 0.510 0.578
10.8824 450 1 0.542 0.01731 0.509 0.577
11.2418 449 1 0.540 0.01731 0.508 0.575
11.3725 448 1 0.539 0.01731 0.506 0.574
11.8301 447 1 0.538 0.01732 0.505 0.573
12.8105 446 1 0.537 0.01732 0.504 0.572
12.9085 445 1 0.536 0.01732 0.503 0.571
13.1046 444 1 0.534 0.01732 0.501 0.569
13.2353 443 1 0.533 0.01733 0.500 0.568
13.3660 442 1 0.532 0.01733 0.499 0.567
13.5621 441 1 0.531 0.01733 0.498 0.566
13.6275 440 1 0.530 0.01734 0.497 0.565
13.8889 439 1 0.528 0.01734 0.495 0.563
13.9542 438 1 0.527 0.01734 0.494 0.562
13.9869 437 1 0.526 0.01734 0.493 0.561
14.9673 436 1 0.525 0.01734 0.492 0.560
15.6863 435 1 0.524 0.01735 0.491 0.559
15.7190 434 1 0.522 0.01735 0.489 0.557
15.9150 433 2 0.520 0.01735 0.487 0.555
15.9477 431 1 0.519 0.01735 0.486 0.554
15.9804 430 1 0.517 0.01736 0.485 0.553
16.0131 429 1 0.516 0.01736 0.483 0.551
16.2745 428 1 0.515 0.01736 0.482 0.550
16.3725 427 1 0.514 0.01736 0.481 0.549
17.0261 426 2 0.511 0.01736 0.479 0.547
18.0392 424 1 0.510 0.01736 0.477 0.545
18.2353 423 1 0.509 0.01736 0.476 0.544
18.5621 422 1 0.508 0.01736 0.475 0.543
18.6275 421 1 0.507 0.01736 0.474 0.542
18.9542 420 1 0.505 0.01736 0.473 0.541
19.6732 419 1 0.504 0.01737 0.471 0.539
19.7059 418 1 0.503 0.01737 0.470 0.538
19.9020 417 1 0.502 0.01737 0.469 0.537
20.0000 416 1 0.501 0.01737 0.468 0.536
20.0327 415 1 0.499 0.01737 0.466 0.535
20.3268 414 1 0.498 0.01737 0.465 0.533
20.3922 413 1 0.497 0.01737 0.464 0.532
20.5556 412 1 0.496 0.01737 0.463 0.531
20.8497 411 1 0.495 0.01736 0.462 0.530
21.4379 410 1 0.493 0.01736 0.460 0.529
21.6667 409 1 0.492 0.01736 0.459 0.527
21.6993 408 1 0.491 0.01736 0.458 0.526
21.9608 407 1 0.490 0.01736 0.457 0.525
22.1569 406 1 0.489 0.01736 0.456 0.524
22.4510 405 1 0.487 0.01736 0.454 0.523
22.4837 404 1 0.486 0.01736 0.453 0.521
23.5948 403 1 0.485 0.01736 0.452 0.520
23.6275 402 1 0.484 0.01736 0.451 0.519
23.6601 401 1 0.483 0.01736 0.450 0.518
24.1830 396 1 0.481 0.01735 0.448 0.517
24.2157 395 1 0.480 0.01735 0.447 0.515
24.3137 391 1 0.479 0.01735 0.446 0.514
24.3464 390 1 0.478 0.01735 0.445 0.513
24.4444 388 1 0.476 0.01735 0.444 0.512
24.5098 385 1 0.475 0.01735 0.442 0.510
24.6405 382 1 0.474 0.01735 0.441 0.509
24.6732 381 1 0.473 0.01735 0.440 0.508
24.9020 379 1 0.471 0.01735 0.439 0.507
25.0327 376 1 0.470 0.01734 0.437 0.505
25.1961 373 1 0.469 0.01734 0.436 0.504
25.7843 366 1 0.468 0.01734 0.435 0.503
26.4379 355 1 0.466 0.01735 0.434 0.502
26.5359 352 1 0.465 0.01735 0.432 0.500
26.6013 351 1 0.464 0.01735 0.431 0.499
27.5163 340 1 0.462 0.01735 0.430 0.498
27.8758 337 1 0.461 0.01735 0.428 0.496
27.9085 333 1 0.460 0.01736 0.427 0.495
28.1046 331 1 0.458 0.01736 0.425 0.493
28.1373 330 1 0.457 0.01736 0.424 0.492
29.1503 312 1 0.455 0.01737 0.422 0.491
29.3137 309 1 0.454 0.01737 0.421 0.489
29.7059 303 1 0.452 0.01738 0.420 0.488
29.7712 301 1 0.451 0.01739 0.418 0.486
30.4248 295 1 0.449 0.01740 0.416 0.485
30.7516 289 1 0.448 0.01740 0.415 0.483
31.8627 275 2 0.444 0.01743 0.412 0.480
32.8105 265 1 0.443 0.01744 0.410 0.478
32.9739 264 1 0.441 0.01746 0.408 0.477
33.2026 261 1 0.439 0.01747 0.406 0.475
33.3007 259 1 0.438 0.01749 0.405 0.473
33.3333 258 1 0.436 0.01750 0.403 0.472
33.6601 255 1 0.434 0.01752 0.401 0.470
34.3791 248 1 0.433 0.01753 0.400 0.468
34.5098 245 1 0.431 0.01755 0.398 0.467
34.7059 240 1 0.429 0.01757 0.396 0.465
35.1634 232 1 0.427 0.01759 0.394 0.463
35.9150 222 1 0.425 0.01762 0.392 0.461
35.9804 221 1 0.423 0.01764 0.390 0.459
36.4379 214 1 0.421 0.01767 0.388 0.457
37.4510 201 1 0.419 0.01771 0.386 0.455
38.4641 193 1 0.417 0.01775 0.384 0.453
38.7908 188 1 0.415 0.01779 0.381 0.451
39.8039 180 1 0.413 0.01784 0.379 0.449
39.8693 178 1 0.410 0.01789 0.377 0.447
41.3399 163 1 0.408 0.01796 0.374 0.444
42.5490 154 1 0.405 0.01803 0.371 0.442
42.8431 152 1 0.402 0.01811 0.368 0.440
43.6275 135 1 0.399 0.01822 0.365 0.437
47.1242 104 1 0.396 0.01845 0.361 0.433
49.2810 86 1 0.391 0.01880 0.356 0.430
51.1765 68 1 0.385 0.01938 0.349 0.425
51.2092 67 1 0.379 0.01992 0.342 0.421
51.8954 61 1 0.373 0.02055 0.335 0.416
53.6601 47 1 0.365 0.02159 0.325 0.410
library(survival)
survdiff(Surv(obsmonths, dead) ~ sex) #logrank検定の函数
Call:
survdiff(formula = Surv(obsmonths, dead) ~ sex)
N Observed Expected (O-E)^2/E (O-E)^2/V
sex=Female 510 321 280 6.11 14.6
sex=Male 319 164 205 8.32 14.6
Chisq= 14.6 on 1 degrees of freedom, p= 0.000132
系統的レビュー(systematic review)の一部:
チフスに対する新しいワクチンを開発し,いくつかの異なる集団において,同じワクチンの有効性を検査して,メタ解析を行う.
OR 95%-CI %W(fixed) %W(random)
HospitalSA 2.8571 [0.6011; 13.5804] 0.9 2.1
GarrisonLadysmith 0.9572 [0.4308; 2.1269] 4.9 7.4
SpecialRegimenSA 2.4098 [1.0233; 5.6753] 2.9 6.5
SpecialHospitalSA 1.5285 [1.2059; 1.9374] 50.1 39.1
MilitaryHospitalSA 1.9814 [1.5058; 2.6072] 35.2 34.3
IndianArmy 2.6684 [1.4016; 5.0803] 6.0 10.7
Number of studies combined: k = 6
OR 95%-CI z p-value
Fixed effect model 1.7659 [1.4979; 2.0820] 6.77 < 0.0001
Random effects model 1.7877 [1.4212; 2.2486] 4.96 < 0.0001
Quantifying heterogeneity:
tau^2 = 0.0204; H = 1.17 [1.00; 1.81]; I^2 = 26.6% [0.0%; 69.5%]
Test of heterogeneity:
Q d.f. p-value
6.81 5 0.2350
Details on meta-analytical method:
- Mantel-Haenszel method
- DerSimonian-Laird estimator for tau^2
metabias(res, k.min = 5)
Linear regression test of funnel plot asymmetry (efficient score)
data: res
t = 0.51543, df = 4, p-value = 0.6334
alternative hypothesis: asymmetry in funnel plot
sample estimates:
bias se.bias slope
0.4428032 0.8590997 0.4306045
アドレスはこちら:http://winterwang.github.io/Epi_exercise/slides.html#/
アドレスはこちら:https://github.com/winterwang/Epi_exercise