Reitsma JB\u306e\u30aa\u30ea\u30b8\u30ca\u30eb\u306e\u8ad6\u6587\u3067\u306f\u3001\u8fd1\u4f3c\u6b63\u898f\u5206\u5e03\u3092\u7528\u3044\u308b\u305f\u3081\u3001\u75c7\u4f8b\u6570\u304c\u5c11\u306a\u304f\u3001\u611f\u5ea6\u30fb\u7279\u7570\u5ea6\u304c95%\u7a0b\u5ea6\u306b\u9ad8\u3044\u5834\u5408\u306b\u306f\u3001\u8aa4\u5dee\u304c\u5927\u304d\u304f\u306a\u308b\u3053\u3068\u3092Chu H\u3089\u304c\u6307\u6458\u3057\u3066\u304a\u308a\u3001Cochrane\u306f\u30b3\u30af\u30e9\u30f3\u306e\u30b7\u30b9\u30c6\u30de\u30c6\u30a3\u30c3\u30af\u30ec\u30d3\u30e5\u30fc\u306b\u4f7f\u7528\u3059\u308b\u3053\u3068\u306f\u63a8\u5968\u3057\u306a\u3044\u3068\u3057\u3066\u3044\u307e\u3059\u3002\u307e\u305f\u3001\u30bc\u30ed\u30a4\u30d9\u30f3\u30c8\u306e\u3042\u308b\u5834\u5408\u306b0.5\u3092\u52a0\u7b97\u3057\u3066\u88dc\u6b63\u3059\u308b\u65b9\u6cd5\u3082\u305a\u308c\u3092\u751f\u3058\u307e\u3059\u3002\u305d\u306e\u305f\u3081\u3001\u5e83\u304f\u4f7f\u7528\u3055\u308c\u3066\u3044\u308bR\u306e\u30d1\u30c3\u30b1\u30fc\u30b8\u3067\u3042\u308bmada<\/a>\u306e\u4f7f\u7528\u306f\u63a8\u5968\u3067\u304d\u306a\u3044\u3068\u8ff0\u3079\u3089\u308c\u3066\u3044\u307e\u3059\u3002\u5b9f\u969b\u306b\u3001Chu H\u3089\u306e\u8ad6\u6587\u3067\u30b7\u30df\u30e5\u30ec\u30fc\u30b7\u30e7\u30f3\u306e\u7d50\u679c\u3092\u898b\u308b\u3068\u3001\u75c7\u4f8b\u6570\u304c25\u4f8b\u4ee5\u4e0b\u3001\u611f\u5ea6\u30fb\u7279\u7570\u5ea6\u304c95%\u4ee5\u4e0a\u306b\u306a\u308b\u3068\u4e8c\u9805\u5206\u5e03\u3092\u7528\u3044\u308b\u5834\u5408\u3068\u305a\u308c\u304c\u5927\u304d\u304f\u306a\u308b\u306e\u306f\u78ba\u304b\u3067\u3059\u304c\u3001Chu H\u306eLetter\u306b\u5bfe\u3059\u308bReitsma JB\u306e\u8fd4\u7b54\u3067\u8ff0\u3079\u3089\u308c\u3066\u3044\u308b\u3088\u3046\u306b\u3001\u591a\u304f\u306e\u5834\u5408\u81e8\u5e8a\u4e0a\u306e\u554f\u984c\u3092\u751f\u3058\u308b\u307b\u3069\u3067\u306f\u306a\u3044\u3068\u8003\u3048\u3089\u308c\u307e\u3059\u3002<\/p>\n\n\n\n Cochrane Handbook for Systematic Reviews of Diagnostic Test Accuracy\u306eAppendix 14\u3067\u306f\u540c\u3058\u5bfe\u8c61\u8005\u3067CT\u3068MRI\u3092\u65bd\u884c\u3057\u3001\u8a3a\u65ad\u80fd\u3092\u6bd4\u8f03\u3057\u305f5\u3064\u306e\u7814\u7a76\u306eDTA MA\u306e\u4f8b\u304c\u8a18\u8f09\u3055\u308c\u3066\u304a\u308a\u3001R\u306elme4<\/a>\u30d1\u30c3\u30b1\u30fc\u30b8\u306eglmer()\u95a2\u6570\u3092\u7528\u3044\u3066\u3001GLMM (Generalized Linear Mixed Effects Model)\u3067\u4e8c\u9805\u5206\u5e03\u306b\u3088\u308b\u56de\u5e30\u30e2\u30c7\u30eb\u3092\u7528\u3044\u3066\u3044\u307e\u3059\u3002Appendix 12\u3067\u306f\u540c\u3058\u5bfe\u8c61\u8005\u3067\u4e8c\u3064\u306e\u8a3a\u65ad\u6cd5\u3092\u5b9f\u65bd\u3057\u305f\u306e\u3067\u306f\u306a\u304f\u3001\u5225\u306e\u5bfe\u8c61\u8005\u3067\u305d\u308c\u305e\u308c\u306e\u8a3a\u65ad\u6cd5\u306e\u611f\u5ea6\u30fb\u7279\u7570\u5ea6\u3092\u6e2c\u5b9a\u3057\u305f\u7814\u7a76\u3092\u3082\u3068\u306b\u3001\u4e8c\u3064\u306e\u8a3a\u65ad\u6cd5\u306e\u8a3a\u65ad\u80fd\u3092\u6bd4\u8f03\u3059\u308b\u305f\u3081\u306e\u89e3\u6790\u6cd5\u304c\u8a18\u8f09\u3055\u308c\u3066\u3044\u307e\u3059\u3002\u3053\u3061\u3089\u306f\u3001\u9593\u63a5\u7684\u306a\u6bd4\u8f03\u3068\u3044\u3046\u8868\u73fe\u304c\u4f7f\u308f\u308c\u3066\u304a\u308a\u3001\u540c\u3058\u5bfe\u8c61\u8005\u3067\u76f4\u63a5\u6bd4\u8f03\u3057\u305f\u7814\u7a76\u3082\u542b\u3081\u3066\u89e3\u6790\u3067\u304d\u308b\u65b9\u6cd5\u304c\u793a\u3055\u308c\u3066\u3044\u307e\u3059\u3002<\/p>\n\n\n\n \u611f\u5ea6\u306e\u30ed\u30b8\u30c3\u30c8\u3068\u507d\u967d\u6027\u7387\u306e\u30ed\u30b8\u30c3\u30c8\u3092\u5f93\u5c5e\u5909\u6570\u3068\u3057\u3001\u5404\u7814\u7a76\u306e\u611f\u5ea6\u306e\u30ed\u30b8\u30c3\u30c8\u3068\u507d\u967d\u6027\u7387\u30ed\u30b8\u30c3\u30c8\u3067\u56de\u5e30\u30e2\u30c7\u30eb\u3092\u4f5c\u3063\u3066\u3044\u307e\u3059\u3002\u30ed\u30b8\u30c3\u30c8\u306f\u30aa\u30c3\u30ba\u306e\u81ea\u7136\u5bfe\u6570\u3001\u3064\u307e\u308alogit(se)=ln[se\/(1-se)], logit(sp)=ln[sp\/(1-sp)], logit(fpr)=-logit(sp)\u3067\u3059\u3002<\/p>\n\n\n\n \u5177\u4f53\u7684\u306b\u306f\u56f31\u306b\u793a\u3059\u3088\u3046\u306a\u30c7\u30fc\u30bf\u306b\u5bfe\u3057\u3066\u3001\u56de\u5e30\u5206\u6790\u3092\u884c\u3044\u3001\u30ed\u30b8\u30c3\u30c8\u306eExponential\u3067\u30aa\u30c3\u30ba\u306b\u5909\u63db\u3057\u3001\u3055\u3089\u306b\u30aa\u30c3\u30ba\/\uff08\uff11\uff0b\u30aa\u30c3\u30ba\uff09\u3067\u5272\u5408\u306b\u5909\u63db\u3057\u3066\u611f\u5ea6\u30fb\u7279\u7570\u5ea6\u306e\u5024\u3092\u6c42\u3081\u3066\u3044\u307e\u3059\u3002\u56f31\u306e\u4f8b\u306f\u540c\u4e00\u75c7\u4f8b\u3067CT\u3068MRI\u3092\u65bd\u884c\u3057\u3066\u304a\u308a\u3001\u3053\u308c\u3089\u4e8c\u3064\u306e\u691c\u67fb\u6cd5\u306e\u611f\u5ea6\u30fb\u7279\u7570\u5ea6\u306e\u7d71\u8a08\u5b66\u7684\u306a\u6bd4\u8f03\u304c\u884c\u308f\u308c\u3066\u3044\u307e\u3059\u3002<\/p>\n\n\n\n \u901a\u5e38\u3001\u5143\u306b\u306a\u308b\u30c7\u30fc\u30bf\u306f\u5404\u7814\u7a76ID\u3068TP, FP, FN, TN \uff08True Positive, False Positive, False Negative, True Negative)\u306e\u4eba\u6570\u306e\u30c7\u30fc\u30bf\u3068\u691c\u67fb\u6cd5\u306e\u30c7\u30fc\u30bf\u3067\u3059\u3002\u305d\u308c\u3089\u304b\u3089\u3001\u56f31\u306e\u5f62\u5f0f\u306e\u30c7\u30fc\u30bf\u30d5\u30ec\u30fc\u30e0\u3092\u4f5c\u6210\u3057\u3001\u4ee5\u4e0b\u306e\u30b9\u30af\u30ea\u30d7\u30c8\u3067\u89e3\u6790\u3092\u884c\u3044\u307e\u3059\u3002<\/p>\n\n\n\n ###Comparison of sensitivity and specificity between two tests done in the same subjects### \u5b9f\u969b\u306b\u306f\u5404\u7814\u7a76ID\u3068TP, FP, FN, TN \u306e\u30c7\u30fc\u30bf\u304b\u3089\u56f31\u306e\u5f62\u5f0f\u306e\u30c7\u30fc\u30bf\u30d5\u30ec\u30fc\u30e0\u3092\u4f5c\u6210\u3059\u308b\u90e8\u5206\u306e\u30b9\u30af\u30ea\u30d7\u30c8\u3082\u5fc5\u8981\u3067\u3059\u304c\u3001\u3053\u3053\u3067\u306flmer()\u95a2\u6570\u306e\u56de\u5e30\u30e2\u30c7\u30eb\u3068\u56f31\u306e\u30c7\u30fc\u30bf\u30e9\u30d9\u30eb\u306e\u95a2\u4fc2\u3092\u8003\u3048\u3066\u3001\u56de\u5e30\u5206\u6790\u3068\u30e1\u30bf\u30a2\u30ca\u30ea\u30b7\u30b9\u306e\u95a2\u4fc2\u3092\u8003\u3048\u308b\u304d\u3063\u304b\u3051\u306b\u306a\u308c\u3070\u3068\u601d\u3044\u307e\u3059\u3002<\/p>\n\n\n\n \u306a\u304a\u3001\u3053\u3053\u3067\u306f\u89e6\u308c\u307e\u305b\u3093\u3067\u3057\u305f\u304c\u3001\u611f\u5ea6\u30fb\u7279\u7570\u5ea6\u306e\u7d71\u5408\u5024\u306895%\u4fe1\u983c\u533a\u9593\u306e\u8a08\u7b97\u306fCT\u3068MRI\u306b\u3064\u3044\u3066\u5225\u3005\u306b\u56de\u5e30\u5206\u6790\u3092\u884c\u3063\u305f\u7d50\u679c\u304b\u3089\u7b97\u51fa\u3057\u307e\u3059\u3002<\/p>\n\n\n\n \u6587\u732e\uff1a Harbord RM, Deeks JJ, Egger M, Whiting P, Sterne JA: A unification of models for meta-analysis of diagnostic accuracy studies. Biostatistics 2007;8:239-51. doi: 10.1093\/biostatistics\/kxl004 PMID: 16698768<\/a> Diagnostic Test Accuracy (DTA) \u8a3a\u65ad\u7cbe\u5ea6\u7814\u7a76\u306eMeta-analysis (MA)\u30e1\u30bf\u30a2\u30ca\u30ea\u30b7\u30b9\u306f\u4ecb\u5165\u306e\u52b9\u679c\u306e\u30e1\u30bf\u30a2\u30ca\u30ea\u30b7\u30b9\u3068\u306f\u7570\u306a\u308a\u307e\u3059\u3002\u624b\u6cd5\u304c\u7570\u306a\u308b\u3060\u3051\u3067\u306a\u304f\u3001\u7d50\u679c\u306e\u89e3\u91c8\u3084\u6d3b\u7528\u6cd5\u3082\u7570\u306a\u308a … “DTA\u306e\u30e1\u30bf\u30a2\u30ca\u30ea\u30b7\u30b9” \u306e<\/span>\u7d9a\u304d\u3092\u8aad\u3080<\/a><\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[32,2],"tags":[],"class_list":["post-1341","post","type-post","status-publish","format-standard","hentry","category-statistics","category-sr"],"_links":{"self":[{"href":"https:\/\/info.zanet.biz\/index.php?rest_route=\/wp\/v2\/posts\/1341","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/info.zanet.biz\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/info.zanet.biz\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/info.zanet.biz\/index.php?rest_route=\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/info.zanet.biz\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=1341"}],"version-history":[{"count":5,"href":"https:\/\/info.zanet.biz\/index.php?rest_route=\/wp\/v2\/posts\/1341\/revisions"}],"predecessor-version":[{"id":1359,"href":"https:\/\/info.zanet.biz\/index.php?rest_route=\/wp\/v2\/posts\/1341\/revisions\/1359"}],"wp:attachment":[{"href":"https:\/\/info.zanet.biz\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=1341"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/info.zanet.biz\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=1341"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/info.zanet.biz\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=1341"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}![\"\"](\"https:\/\/info.zanet.biz\/wp\/wp-content\/uploads\/2023\/02\/dta_pair_ct_mri_data-1024x401.png\")
###Random-effects meta-analysis with bivariate model using binomial distribution###
library(lme4)
library(lmtest)
###Y is a data frame as shown in Fig. 1###
(B = glmer(formula=cbind(true, n – true) ~ 0 + seCT + seMRI + spCT + spMRI + (0+sens + spec|Study_ID), data=Y, family=binomial))
(C = glmer(formula=cbind(true, n – true) ~ 0 + sens + spCT + spMRI + (0+sens + spec|Study_ID), data=Y, family=binomial))
###Is there a statistically significant difference in sensitivity between CT and MRI?
lrtest(B,C)<\/p>\n\n\n\n![\"\"](\"https:\/\/info.zanet.biz\/wp\/wp-content\/uploads\/2023\/02\/dta_B-C_se_comparison.png\")
<\/figure>\n\n\n\n
###Is there a statistically significant difference in specificity between CT and MRI?
lrtest(B,D)<\/p>\n\n\n\n![\"\"](\"https:\/\/info.zanet.biz\/wp\/wp-content\/uploads\/2023\/02\/dta_B-D_sp_comparison.png\")
<\/figure>\n\n\n\n
Chu H, Cole SR: Bivariate meta-analysis of sensitivity and specificity with sparse data: a generalized linear mixed model approach. J Clin Epidemiol 2006;59:1331-2 author reply 1332-3. doi: 10.1016\/j.jclinepi.2006.06.011 PMID: 17098577<\/a>
Reitsma JB, Glas AS, Rutjes AW, Scholten RJ, Bossuyt PM, Zwinderman AH: Bivariate analysis of sensitivity and specificity produces informative summary measures in diagnostic reviews. J Clin Epidemiol 2005;58:982-90. doi: 10.1016\/j.jclinepi.2005.02.022 PMID: 16168343<\/a>
Rutter CM, Gatsonis CA: A hierarchical regression approach to meta-analysis of diagnostic test accuracy evaluations. Stat Med 2001;20:2865-84. PMID: 11568945<\/a><\/p>\n\n\n\n
Arends LR, Hamza TH, van Houwelingen JC, Heijenbrok-Kal MH, Hunink MG, Stijnen T: Bivariate random effects meta-analysis of ROC curves. Med Decis Making 2008;28:621-38. doi: 10.1177\/0272989X08319957 PMID: 18591542<\/a><\/p>\n","protected":false},"excerpt":{"rendered":"