{"id":542,"date":"2018-07-14T20:06:20","date_gmt":"2018-07-14T11:06:20","guid":{"rendered":"https:\/\/best-biostatistics.com\/toukei-er\/entry\/how-to-detect-publication-bias-by-rank-cor-and-regression-tests\/"},"modified":"2024-10-14T08:37:32","modified_gmt":"2024-10-13T23:37:32","slug":"how-to-detect-publication-bias-by-rank-cor-and-regression-tests","status":"publish","type":"post","link":"https:\/\/best-biostatistics.com\/toukei-er\/entry\/how-to-detect-publication-bias-by-rank-cor-and-regression-tests\/","title":{"rendered":"R \u3067\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u306e\u691c\u5b9a\u3092\u884c\u3046\u65b9\u6cd5"},"content":{"rendered":"\n

Begg\u691c\u5b9a\u3001Egger\u691c\u5b9a\u3001Macaskill\u691c\u5b9a\u3068\u3044\u3046\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u306e\u691c\u5b9a\u65b9\u6cd5\u306e\u89e3\u8aac\u3002<\/p>\n\n\n\n\n\n\n\n

\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u691c\u5b9a\u306e\u305f\u3081\u306e\u500b\u3005\u306e\u7814\u7a76\u30c7\u30fc\u30bf<\/h2>\n\n\n\n

\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u3092\u691c\u5b9a\u3059\u308b\u305f\u3081\u306e\u30b5\u30f3\u30d7\u30eb\u30c7\u30fc\u30bf\u306f\u4ee5\u4e0b\u306e\u901a\u308a\u3002<\/p>\n\n\n\n

a <-<\/span> c<\/span>(<\/span>3<\/span>,<\/span>7<\/span>,<\/span>5<\/span>,<\/span>102<\/span>,<\/span>28<\/span>,<\/span>4<\/span>,<\/span>98<\/span>,<\/span>60<\/span>,<\/span>25<\/span>,<\/span>138<\/span>,<\/span>64<\/span>,<\/span>45<\/span>,<\/span>9<\/span>,<\/span>57<\/span>,<\/span>25<\/span>,<\/span>65<\/span>,<\/span>17<\/span>)<\/span>\nn1 <-<\/span> c<\/span>(<\/span>38<\/span>,<\/span>114<\/span>,<\/span>69<\/span>,<\/span>1533<\/span>,<\/span>355<\/span>,<\/span>59<\/span>,<\/span>945<\/span>,<\/span>632<\/span>,<\/span>278<\/span>,<\/span>1916<\/span>,<\/span>873<\/span>,<\/span>263<\/span>,<\/span>291<\/span>,<\/span>858<\/span>,<\/span>154<\/span>,<\/span>1195<\/span>,<\/span>298<\/span>)<\/span>\nc <-<\/span> c<\/span>(<\/span>3<\/span>,<\/span>14<\/span>,<\/span>11<\/span>,<\/span>127<\/span>,<\/span>27<\/span>,<\/span>6<\/span>,<\/span>152<\/span>,<\/span>48<\/span>,<\/span>37<\/span>,<\/span>188<\/span>,<\/span>52<\/span>,<\/span>47<\/span>,<\/span>16<\/span>,<\/span>45<\/span>,<\/span>31<\/span>,<\/span>62<\/span>,<\/span>34<\/span>)<\/span>\nn0 <-<\/span> c<\/span>(<\/span>39<\/span>,<\/span>116<\/span>,<\/span>93<\/span>,<\/span>1520<\/span>,<\/span>365<\/span>,<\/span>52<\/span>,<\/span>939<\/span>,<\/span>471<\/span>,<\/span>282<\/span>,<\/span>1921<\/span>,<\/span>583<\/span>,<\/span>266<\/span>,<\/span>293<\/span>,<\/span>883<\/span>,<\/span>147<\/span>,<\/span>1200<\/span>,<\/span>309<\/span>)<\/span>\ndat <-<\/span> data.frame<\/span>(<\/span>a,<\/span>n1,<\/span>c,<\/span>n0)<\/span>\n<\/code><\/pre>\n\n\n\n
R\u306emetafor\u30d1\u30c3\u30b1\u30fc\u30b8\u3092\u4f7f\u3046\u3002<\/p>\n\n\n\n
metafor\u30d1\u30c3\u30b1\u30fc\u30b8\u306f\u521d\u56de\u3060\u3051\u30a4\u30f3\u30b9\u30c8\u30fc\u30eb\u304c\u5fc5\u8981\u3002<\/p>\n\n\n\n
install.packages<\/span>(<\/span>\"metafor\"<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
metafor\u30d1\u30c3\u30b1\u30fc\u30b8\u306eescalc()\u3067\u3001\u500b\u3005\u306e\u7814\u7a76\u306e\u63a8\u5b9a\u5024\u3068\u5206\u6563\u3092\u8a08\u7b97\u3059\u308b\u3002<\/p>\n\n\n\n
\u3053\u306e\u30c7\u30fc\u30bf\u3067\u306f\u63a8\u5b9a\u5024\u306f\u30aa\u30c3\u30ba\u6bd4\u3002<\/p>\n\n\n\n
measure=\u3067\u6307\u5b9a\u3059\u308b\u3002<\/p>\n\n\n\n
\u63a8\u5b9a\u5024yi\u3068\u5206\u6563vi\u304c\u8a08\u7b97\u3055\u308c\u308b\u3002<\/p>\n\n\n\n
\u7d71\u5408\u30aa\u30c3\u30ba\u6bd4\u306895%\u4fe1\u983c\u533a\u9593\u306f\u3001\u5236\u9650\u4ed8\u304d\u6700\u5c24\u63a8\u5b9a\u91cf\uff08REML\uff09\u3067\u8a08\u7b97\u3059\u308b\u3002<\/p>\n\n\n\n
library<\/span>(<\/span>metafor)<\/span>\ndat.escalc <-<\/span> escalc<\/span>(<\/span>measure=<\/span>\"OR\"<\/span>,<\/span> ai=<\/span>a,<\/span> n1i=<\/span>n1,<\/span> ci=<\/span>c,<\/span> n2i=<\/span>n0,<\/span> data=<\/span>dat)<\/span>\nres.reml <-<\/span> rma.uni<\/span>(<\/span>yi,<\/span> vi,<\/span> method=<\/span>\"REML\"<\/span>,<\/span> data=<\/span>dat.escalc)<\/span>\n<\/code><\/pre>\n\n\n\n
\u30d5\u30a1\u30f3\u30cd\u30eb\u30d7\u30ed\u30c3\u30c8\u3092\u66f8\u304f\u3068\u4ee5\u4e0b\u306e\u901a\u308a\u3060\u3002<\/p>\n\n\n\n
funnel<\/span>(<\/span>res.reml)<\/span>\n<\/code><\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/span><\/p>\n\n\n\n
\u30d5\u30a1\u30f3\u30cd\u30eb\u30d7\u30ed\u30c3\u30c8\u306b\u3064\u3044\u3066\u306f\u4ee5\u4e0b\u3082\u53c2\u7167\u3002<\/p>\n\n\n
\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\u3042\u308f\u305b\u3066\u8aad\u307f\u305f\u3044<\/span>\n\t\t\t\t\t
\"\"<\/figure><\/div>\t\t\t\t\t
\n\t\t\t\t\t\tR \u3067\u30e1\u30bf\u30a2\u30ca\u30ea\u30b7\u30b9\u306e\u30d5\u30a1\u30f3\u30cd\u30eb\u30d7\u30ed\u30c3\u30c8\u3092\u66f8\u304f\u65b9\u6cd5<\/a>\n\t\t\t\t\t\t\u30d5\u30a1\u30f3\u30cd\u30eb\u30d7\u30ed\u30c3\u30c8\u3068\u306f\u4f55\u304b\uff1f \u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u3068\u306f\u4f55\u304b\uff1f \u30d5\u30a9\u30ec\u30b9\u30c8\u30d7\u30ed\u30c3\u30c8\u3068\u306e\u9055\u3044\u306f\u4f55\u304b\uff1f \u30d5\u30a1\u30f3\u30cd\u30eb\u30d7\u30ed\u30c3\u30c8\u306e\u524d\u306b\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u3068\u306f\uff1f \u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u3068\u306f\u3001\u516c\u8868\u3055\u308c…<\/span>\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t<\/div>\n\t\t<\/div>\n\n\n\n\n\n\n
yi\u3068vi\u304c\u8a08\u7b97\u3055\u308c\u308c\u3070\u3001\u30aa\u30c3\u30ba\u6bd4\u4ee5\u5916\u306e\u63a8\u5b9a\u5024\u306e\u7d71\u5408\u3082\u53ef\u80fd\u3060\u3002<\/p>\n\n\n\n
\u5e73\u5747\u5024\u306e\u5dee\u3082\u7d71\u5408\u53ef\u80fd\u3002<\/p>\n\n\n\n
\u5e73\u5747\u5024\u306e\u5dee\u306e\u30e1\u30bf\u30a2\u30ca\u30ea\u30b7\u30b9\u306f\u3053\u3061\u3089\u3092\u53c2\u7167\u3002<\/p>\n\n\n
\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\u3042\u308f\u305b\u3066\u8aad\u307f\u305f\u3044<\/span>\n\t\t\t\t\t
\"\"<\/figure><\/div>\t\t\t\t\t
\n\t\t\t\t\t\tR \u3067\u5e73\u5747\u5024\u306e\u5dee\u306e\u30e1\u30bf\u89e3\u6790\u3092\u884c\u3046\u65b9\u6cd5<\/a>\n\t\t\t\t\t\t\u5e73\u5747\u5024\u306e\u5dee\u306e\u30e1\u30bf\u89e3\u6790\u306e\u3084\u308a\u65b9\u3092\u89e3\u8aac\u3002 \u30e1\u30bf\u89e3\u6790\u306e\u3084\u308a\u65b9\u89e3\u8aac\u306e\u305f\u3081\u306e\u30b5\u30f3\u30d7\u30eb\u30c7\u30fc\u30bf \u30e1\u30bf\u89e3\u6790\u306e\u3084\u308a\u65b9\u3092\u89e3\u8aac\u3059\u308b\u305f\u3081\u306e\u30c7\u30fc\u30bf\u306f\u4ee5\u4e0b\u306e\u901a\u308a\u3002 m\u304c\u5e73\u5747\u3001s\u304c\u6a19\u6e96\u504f\u5dee\u3001n…<\/span>\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t<\/div>\n\t\t<\/div>\n\n\n\n\n\n\n
Begg\u691c\u5b9a\u3000\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u3092\u9806\u4f4d\u76f8\u95a2\u3067\u691c\u51fa\u3059\u308b\u65b9\u6cd5<\/h2>\n\n\n\n
\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u306e\u691c\u5b9a\u3068\u3057\u3066\u4e00\u3064\u76ee\u306f\u3001Begg\u306e\u9806\u4f4d\u76f8\u95a2\u3068\u3044\u3046\u65b9\u6cd5\u3092\u5b9f\u65bd\u3057\u3066\u307f\u308b\u3002<\/p>\n\n\n\n
\u5206\u6563\uff08$ s_i^2$ \u30b9\u30af\u30ea\u30d7\u30c8\u3067\u306f si2 \u3068\u3059\u308b\uff09\u306e\u9006\u6570\u3092\u91cd\u307f\u306b\u3057\u305f\u63a8\u5b9a\u5024\u306e\u91cd\u307f\u3065\u3051\u5e73\u5747\uff08$ \\hat{\\theta}$ \u30b9\u30af\u30ea\u30d7\u30c8\u4e2d\u3067\u306f theta \u3068\u3059\u308b\uff09\u3092\u8a08\u7b97\u3059\u308b\u3002<\/p>\n\n\n\n
R\u306e\u30b9\u30af\u30ea\u30d7\u30c8\u306f\u4ee5\u4e0b\u306e\u901a\u308a\u3002<\/p>\n\n\n\n
thetai <-<\/span> dat.escalc$<\/span>yi\nsi2 <-<\/span> dat.escalc$<\/span>vi\ntheta <-<\/span> sum<\/span>(<\/span>thetai\/<\/span>si2)<\/span>\/<\/span>sum<\/span>(<\/span>1<\/span>\/<\/span>si2)<\/span>\nsi2star <-<\/span> si2 -<\/span> 1<\/span>\/<\/span>sum<\/span>(<\/span>1<\/span>\/<\/span>si2)<\/span>\nsi.star <-<\/span> sqrt<\/span>(<\/span>si2star)<\/span>\nti <-<\/span> (<\/span>thetai-<\/span>theta)<\/span>\/<\/span>si.star\n<\/code><\/pre>\n\n\n\n
\u5404\u8a66\u9a13\u306e\u63a8\u5b9a\u5024\u306e\u6a19\u6e96\u504f\u5dee\uff08\u3064\u307e\u308a\u6a19\u6e96\u8aa4\u5dee\uff09\u3092\u8a08\u7b97\u3059\u308b\u3002<\/p>\n\n\n\n
\u5206\u6563\uff08$ s_i^2$\uff09\u304b\u3089\u91cd\u307f\u306e\u5408\u8a08\u306e\u9006\u6570\u3092\u5f15\u3044\u305f\u3082\u306e\u306e\u5e73\u65b9\u6839\u304c\u5404\u8a66\u9a13\u306e\u6a19\u6e96\u8aa4\u5dee $ s_i^*$ \u3060\u3002<\/p>\n\n\n\n
$ \\hat{\\theta} $ \u3068 $ s_i^* $ \u3067\u3001\u5404\u8a66\u9a13\u306e\u63a8\u5b9a\u5024 $ \\hat{\\theta}_i $ \u3092\u6a19\u6e96\u5316\u3057\u305f\u3082\u306e\u304c\u3001$ t_i $ \u30b9\u30af\u30ea\u30d7\u30c8\u4e2d\u3067\u306f ti\u3067\u3042\u308b\u3002<\/p>\n\n\n\n
$ t_i $ \u3068 $ s_i^2 $ \u9593\u306eKendall\u306e\u9806\u4f4d\u76f8\u95a2\u304c0\u304b\u3069\u3046\u304b\u3092\u691c\u5b9a\u3059\u308b\u3002<\/p>\n\n\n\n
\u3064\u307e\u308a\u3001\u6a19\u6e96\u5316\u3057\u305f\u5404\u8a66\u9a13\u306e\u63a8\u5b9a\u5024\u3068\u3001\u5404\u8a66\u9a13\u306e\u5206\u6563\u304c\u76f8\u95a2\u3057\u3066\u3044\u308b\u306e\u304b\u3069\u3046\u304b\u3092\u691c\u5b9a\u3059\u308b\u3002<\/p>\n\n\n\n
\u5e30\u7121\u4eee\u8aac\u304c\u68c4\u5374\u3067\u304d\u305a\u3001Kendall\u306e\u9806\u4f4d\u76f8\u95a2\u304c0\u3067\u3042\u308b\u3053\u3068\u3092\u5426\u5b9a\u3067\u304d\u306a\u3051\u308c\u3070\u3001\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u306f\u306a\u3044\u3068\u5224\u65ad\u3057\u3066\u3088\u3044\u3002<\/p>\n\n\n\n
Kendall\u306e\u9806\u4f4d\u76f8\u95a2\u306f\u3053\u3061\u3089\u3082\u53c2\u7167\u3002<\/p>\n\n\n
\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\u3042\u308f\u305b\u3066\u8aad\u307f\u305f\u3044<\/span>\n\t\t\t\t\t
\"\"<\/figure><\/div>\t\t\t\t\t
\n\t\t\t\t\t\tR \u3067\u30b1\u30f3\u30c9\u30fc\u30eb\u306e\u9806\u4f4d\u76f8\u95a2\u4fc2\u6570\u3092\u8a08\u7b97\u3059\u308b\u65b9\u6cd5<\/a>\n\t\t\t\t\t\t\u30b1\u30f3\u30c9\u30fc\u30eb\u306e\u9806\u4f4d\u76f8\u95a2\u4fc2\u6570\u306f\u3069\u306e\u3088\u3046\u306b\u8a08\u7b97\u3059\u308b\u304b\u7d39\u4ecb\u3059\u308b \u30b1\u30f3\u30c9\u30fc\u30eb\u306e\u9806\u4f4d\u76f8\u95a2\u4fc2\u6570 \u30b1\u30f3\u30c9\u30fc\u30eb\uff08Kendall\uff09\u306e\u9806\u4f4d\u76f8\u95a2\u4fc2\u6570 $ \\tau $ \u306f\u3001\u9806\u4f4d\u3092\u4f7f\u308f\u306a\u3044\u76f8\u95a2\u4fc2\u6570\u3067\u3042\u308b…<\/span>\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t<\/div>\n\t\t<\/div>\n\n\n\n\n\n\n
\u7d71\u8a08\u5b66\u7684\u6709\u610f\u304b\u3069\u3046\u304b\u306f\u6709\u610f\u6c34\u6e9610%\u3092\u7528\u3044\u308b\u3002<\/p>\n\n\n\n
\u30e1\u30bf\u30a2\u30ca\u30ea\u30b7\u30b9\u306e\u5bfe\u8c61\u3068\u306a\u308b\u7814\u7a76\u6570\u304c\u5c11\u306a\u3044\u305f\u3081\u3001\u691c\u51fa\u529b\u304c\u4f4e\u3044\u304b\u3089\u3060\u3002<\/p>\n\n\n\n
\u6709\u610f\u6c34\u6e9610%\u306f\u4ed6\u306e2\u3064\u306e\u65b9\u6cd5\u3082\u540c\u69d8\u3060\u3002<\/p>\n\n\n\n
$ t_i $ \u3068$ s_i^2 $ \u306e\u6563\u5e03\u56f3\u3092\u63cf\u3044\u3066\u307f\u308b\u3002<\/p>\n\n\n\n
plot<\/span>(<\/span>ti,<\/span> si2)<\/span>\n<\/code><\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/span><\/p>\n\n\n\n
Kendall\u306e\u9806\u4f4d\u76f8\u95a2\u3092\u8a08\u7b97\u3059\u308b\u3002<\/p>\n\n\n\n
cor.test<\/span>(<\/span>ti,<\/span> si2,<\/span> method=<\/span>\"kendall\"<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
\u76f8\u95a2\u4fc2\u6570\u306f-0.059\u3067p\u5024\u306f0.777\u3067\u3001\u76f8\u95a2\u4fc2\u6570\u304c0\u3067\u3042\u308b\u3068\u3044\u3046\u5e30\u7121\u4eee\u8aac\u3092\u68c4\u5374\u3067\u304d\u306a\u3044\u3002<\/p>\n\n\n\n
\u3064\u307e\u308a\u76f8\u95a2\u4fc2\u6570\u304c\u30bc\u30ed\u3067\u3042\u308b\u3053\u3068\u3092\u5426\u5b9a\u3067\u304d\u306a\u3044\u3002<\/p>\n\n\n\n
\u6a19\u6e96\u5316\u3055\u308c\u305f\u63a8\u5b9a\u5024\u3068\u5404\u7814\u7a76\u306e\u91cd\u307f\u306e\u9006\u6570\uff08\u5206\u6563\uff09\u306e\u9593\u306b\u306f\u76f8\u95a2\u304c\u306a\u304f\u3001\u7814\u7a76\u7d50\u679c\u3084\u7cbe\u5ea6\u306b\u3088\u308b\u30d0\u30a4\u30a2\u30b9\u306f\u691c\u51fa\u3055\u308c\u306a\u304b\u3063\u305f\u3002<\/p>\n\n\n\n
><\/span> cor.test<\/span>(<\/span>ti,<\/span> si2,<\/span> method=<\/span>\"kendall\"<\/span>)<\/span>\nKendall's rank correlation tau<\/span>\ndata:  ti and si2<\/span>\nT = 64, p-value = 0.7765<\/span>\nalternative hypothesis: true tau is not equal to 0<\/span>\nsample estimates:<\/span>\n        tau <\/span>\n-0.05882353 <\/span>\n<\/code><\/pre>\n\n\n\n
metafor\u30d1\u30c3\u30b1\u30fc\u30b8\u306eranktest()\u3092\u4f7f\u3046\u3068\u305a\u3063\u3068\u7c21\u5358\u3002<\/p>\n\n\n\n
ranktest<\/span>(<\/span>res.reml)<\/span>\n<\/code><\/pre>\n\n\n\n
\u30e1\u30bf\u30a2\u30ca\u30ea\u30b7\u30b9\u306e\u8a08\u7b97\u7d50\u679c\u304c\u683c\u7d0d\u3055\u308c\u3066\u3044\u308b\u30aa\u30d6\u30b8\u30a7\u30af\u30c8\uff08\u3053\u3053\u3067\u306fres.reml\uff09\u304c\u3042\u308c\u3070\u3001\u63a8\u5b9a\u5024\u306e\u91cd\u307f\u3065\u3051\u5e73\u5747\u3084\u3001\u5404\u7814\u7a76\u306e\u6a19\u6e96\u504f\u5dee\u3001\u63a8\u5b9a\u5024\u306e\u6a19\u6e96\u5316\u306e\u8a08\u7b97\u306f\u5168\u304f\u4e0d\u8981\u3067\u3001\u4e00\u77ac\u3067\u7d50\u679c\u304c\u5f97\u3089\u308c\u308b\u3002<\/p>\n\n\n\n
><\/span> ranktest<\/span>(<\/span>res.reml)<\/span>\nRank Correlation Test for<\/span> Funnel Plot Asymmetry\nKendall's tau = -0.0588, p = 0.7765<\/span>\n<\/code><\/pre>\n\n\n\n
\uff1e\uff1e\u3082\u3046\u7d71\u8a08\u3067\u60a9\u3080\u306e\u306f\u7d42\u308f\u308a\u306b\u3057\u307e\u305b\u3093\u304b\uff1f\u00a0<\/a><\/strong><\/span><\/p>\r\n\"\"<\/a>\r\n
\u21911\u4e07\u4eba\u4ee5\u4e0a\u306e\u533b\u7642\u5f93\u4e8b\u8005\u304c\u8cfc\u8aad\u4e2d<\/span><\/strong><\/span><\/p><\/div>
Egger\u691c\u5b9a\u3000\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u3092\u56de\u5e30\u5206\u6790\u3067\u691c\u51fa\u3059\u308b\u65b9\u6cd5<\/h2>\n\n\n\n
Egger\u306e\u65b9\u6cd5\u306f\u3001\u4ee5\u4e0b\u306e\u56de\u5e30\u30e2\u30c7\u30eb\u3092\u7528\u3044\u308b\u3002<\/p>\n\n\n\n
$$ \\left( \\frac{\\hat{\\theta}_i}{s_i} \\right) = \\alpha + \\beta \\left( \\frac{1}{s_i} \\right) + \u8aa4\u5dee $$<\/p>\n\n\n\n\n\n\n\n
\u5e30\u7121\u4eee\u8aac\u306e\u5143\u3067\u306f\u3001\u30d5\u30a1\u30f3\u30cd\u30eb\u30d7\u30ed\u30c3\u30c8\u304c\u5de6\u53f3\u5bfe\u79f0\u5f62\u306a\u3089\u3070\u3001\u3053\u306e\u56de\u5e30\u5f0f\u306f\u539f\u70b9\u3092\u901a\u308a\uff08\u03b1= 0\uff09\u3001\u50be\u304d\u306f\u7d71\u5408\u63a8\u5b9a\u5024\u306b\u7b49\u3057\u304f\uff08$ \\hat{\\beta} = \\hat{\\theta} $\uff09\u306a\u308b\u306e\u3067\u3001\u03b1 = 0 \u3092\u691c\u5b9a\u3059\u308b\u3002<\/p>\n\n\n\n
\u524d\u7bc0\u3067\u4f7f\u3063\u305f\u5909\u6570\u3092\u5f15\u304d\u7d9a\u304d\u4f7f\u3046\u3002<\/p>\n\n\n\n
si <-<\/span> sqrt<\/span>(<\/span>si2)<\/span>\ntheta.si <-<\/span> thetai\/<\/span>si\ninv.si <-<\/span> 1<\/span>\/<\/span>si\n<\/code><\/pre>\n\n\n\n
\u7dda\u5f62\u30e2\u30c7\u30eblm()\u3067\u3001\u4e0a\u8a18\u56de\u5e30\u30e2\u30c7\u30eb\u3092\u63a8\u5b9a\u3059\u308b\u3002<\/p>\n\n\n\n
res.theta.si <-<\/span> lm<\/span>(<\/span>theta.si ~<\/span> inv.si)<\/span>\nsummary<\/span>(<\/span>res.theta.si)<\/span>\n<\/code><\/pre>\n\n\n\n
\u7d50\u679c\u306f\u4ee5\u4e0b\u306e\u901a\u308a\u3002<\/p>\n\n\n\n
(Intercept)\u306e\u884c\u304c \u03b1 = 0 \u306e\u691c\u5b9a\u7d50\u679c\u3002<\/p>\n\n\n\n
p\u5024\u304c0.817\u3067\u3001\u30bc\u30ed\u3092\u5426\u5b9a\u3059\u308b\u3053\u3068\u306f\u3067\u304d\u306a\u3044\u3002<\/p>\n\n\n\n
\u3086\u3048\u306b\u30d5\u30a1\u30f3\u30cd\u30eb\u30d7\u30ed\u30c3\u30c8\u306f\u5de6\u53f3\u5bfe\u79f0\u5f62\u3067\u3042\u308b\u5e30\u7121\u4eee\u8aac\u306f\u68c4\u5374\u3055\u308c\u305a\u3001\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u304c\u3042\u308b\u53ef\u80fd\u6027\u306f\u4f4e\u3044\u3068\u8a00\u3048\u308b\u3002<\/p>\n\n\n\n
><\/span> summary<\/span>(<\/span>res.theta.si)<\/span>\nCall:<\/span>\nlm<\/span>(<\/span>formula =<\/span> theta.si ~<\/span> inv.si)<\/span>\nResiduals:<\/span>\nMin      1Q  Median      3Q     Max\n-<\/span>1.9909<\/span> -<\/span>0.7373<\/span> -<\/span>0.2046<\/span>  0.8214<\/span>  2.5698<\/span>\nCoefficients:<\/span>\nEstimate Std. Error t value Pr<\/span>(<\/span>>|<\/span>t|<\/span>)<\/span>\n(<\/span>Intercept)<\/span>  -<\/span>0.1518<\/span>     0.6463<\/span>  -<\/span>0.235<\/span>    0.817<\/span>\ninv.si       -<\/span>0.2152<\/span>     0.1395<\/span>  -<\/span>1.543<\/span>    0.144<\/span>\nResidual standard error:<\/span> 1.194<\/span> on 15<\/span> degrees of freedom\nMultiple R-<\/span>squared:<\/span>  0.1369<\/span>,<\/span>    Adjusted R-<\/span>squared:<\/span>  0.07937<\/span>\nF<\/span>-<\/span>statistic:<\/span> 2.379<\/span> on 1<\/span> and 15<\/span> DF,<\/span>  p-<\/span>value:<\/span> 0.1438<\/span>\n<\/code><\/pre>\n\n\n\n
\u5404\u7814\u7a76\u3092\u30d7\u30ed\u30c3\u30c8\u3057\u3066\u3001\u56de\u5e30\u76f4\u7dda\u3092\u5f15\u3044\u3066\u307f\u308b\u3002<\/p>\n\n\n\n
\u56de\u5e30\u76f4\u7dda\uff08\u7834\u7dda\uff09\u306f\u307b\u307c\u539f\u70b9\u3092\u901a\u3063\u3066\u3044\u308b\u3053\u3068\u304c\u308f\u304b\u308b\u3002<\/p>\n\n\n\n
plot<\/span>(<\/span>theta.si ~<\/span> inv.si,<\/span> xlab=<\/span>\"1\/SE\"<\/span>,<\/span> ylab=<\/span>\"log(OR)\/SE\"<\/span>,<\/span> xlim=<\/span>c<\/span>(<\/span>0<\/span>,<\/span>10<\/span>))<\/span>\nabline<\/span>(<\/span>h=<\/span>0<\/span>,<\/span> v=<\/span>0<\/span>,<\/span> lty=<\/span>1<\/span>)<\/span>\nabline<\/span>(<\/span>res.theta.si,<\/span> lty=<\/span>2<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/span><\/p>\n\n\n\n
metafor\u30d1\u30c3\u30b1\u30fc\u30b8\u306eregtest()\u3092\u4f7f\u3046\u3068\u7c21\u5358\u306b\u8a08\u7b97\u3067\u304d\u308b\u3002<\/p>\n\n\n\n
regtest<\/span>(<\/span>res.reml,<\/span> model=<\/span>\"lm\"<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
\u7d50\u679c\u306f\u4ee5\u4e0b\u306e\u901a\u308a\u3002<\/p>\n\n\n\n
t\u5024\u3068p\u5024\u304c\u3001regtest()\u3092\u4f7f\u308f\u306a\u3044\u7d50\u679c\u3068\u4e00\u81f4\u3057\u3066\u3044\u308b\u306e\u304c\u308f\u304b\u308b\u3002<\/p>\n\n\n\n
><\/span> regtest<\/span>(<\/span>res.reml,<\/span> model=<\/span>\"lm\"<\/span>)<\/span>\nRegression Test for<\/span> Funnel Plot Asymmetry\nModel:<\/span>     weighted regression with multiplicative dispersion\nPredictor:<\/span> standard error\nTest for<\/span> Funnel Plot Asymmetry:<\/span> t =<\/span> -<\/span>0.2349<\/span>,<\/span> df =<\/span> 15<\/span>,<\/span> p =<\/span> 0.8175<\/span>\nLimit Estimate <\/span>(<\/span>as sei -><\/span> 0<\/span>):<\/span>   b =<\/span> -<\/span>0.2152 <\/span>(<\/span>CI:<\/span> -<\/span>0.5126<\/span>,<\/span> 0.0822<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
\u3061\u306a\u307f\u306b\u3001\u7814\u7a76\u9593\u306b\u7121\u8996\u3067\u304d\u306a\u3044\u7570\u8cea\u6027\u304c\u691c\u51fa\u3055\u308c\u305f\u5834\u5408\u3001\u5909\u91cf\u52b9\u679c\u30e2\u30c7\u30eb\u3067\u7d71\u5408\u3059\u308b\u3053\u3068\u306b\u306a\u308b\u304c\u3001\u305d\u306e\u3068\u304d\u306fEgger\u306e\u65b9\u6cd5\u3082\u5909\u91cf\u52b9\u679c\u30e2\u30c7\u30eb\u3067\u5b9f\u65bd\u3059\u308b\u3068\u3088\u3044\u3002<\/p>\n\n\n\n
\u4eca\u56de\u306e\u4f8b\u984c\u306f\u3059\u3067\u306bREML\u3092\u7528\u3044\u3066\u5909\u91cf\u52b9\u679c\u30e2\u30c7\u30eb\u3067\u7d71\u5408\u3057\u3066\u3044\u308b\u306e\u3067\u3001\u89e3\u6790\u7d50\u679c\u306e\u30aa\u30d6\u30b8\u30a7\u30af\u30c8\u306f\u305d\u306e\u307e\u307e\u3067OK\u3002<\/p>\n\n\n
\t\t\t
\n\t\t\t\t
\n\t\t\t\t\t\u3042\u308f\u305b\u3066\u8aad\u307f\u305f\u3044<\/span>\n\t\t\t\t\t
\"\"<\/figure><\/div>\t\t\t\t\t
\n\t\t\t\t\t\tR \u3067\u5236\u9650\u4ed8\u304d\u6700\u5c24\u63a8\u5b9a\u91cf REML \u3067\u30aa\u30c3\u30ba\u6bd4\u3092\u7d71\u5408\u3059\u308b\u30e1\u30bf\u30a2\u30ca\u30ea\u30b7\u30b9<\/a>\n\t\t\t\t\t\t\u30aa\u30c3\u30ba\u6bd4\u3092\u5236\u9650\u4ed8\u304d\u6700\u5c24\u63a8\u5b9a\u91cf REML \u3067\u7d71\u5408\u3059\u308b\u65b9\u6cd5\u306e\u89e3\u8aac\u3002 R \u3067\u5b9f\u884c\u3059\u308b\u3002 \u5236\u9650\u4ed8\u304d\u6700\u5c24\u63a8\u5b9a\u91cf\u3067\u30aa\u30c3\u30ba\u6bd4\u3092\u7d71\u5408\u3059\u308b\u610f\u5473 \u30aa\u30c3\u30ba\u6bd4\u306e\u30e1\u30bf\u30a2\u30ca\u30ea\u30b7\u30b9\u3092\u3059\u308b\u3068\u304d\u306b\u3001\u5bfe\u6570…<\/span>\t\t\t\t\t<\/div>\n\t\t\t\t<\/div>\n\t\t\t<\/div>\n\t\t<\/div>\n\n\n\n\n\n\n
\u5148\u307b\u3069 model = “lm” \u3068\u3044\u3046\u30aa\u30d7\u30b7\u30e7\u30f3\u3092\u5165\u308c\u305f\u304c\u3001\u305d\u306e\u30aa\u30d7\u30b7\u30e7\u30f3\u3092\u5165\u308c\u306a\u3044\u3068\u3001\u5909\u91cf\u52b9\u679c\u30e2\u30c7\u30eb\u306e\u7d50\u679c\u306b\u306a\u308b\u3002<\/p>\n\n\n\n
><\/span> regtest<\/span>(<\/span>res.reml)<\/span>\nRegression Test for<\/span> Funnel Plot Asymmetry\nModel:<\/span>     mixed-<\/span>effects meta-<\/span>regression model\nPredictor:<\/span> standard error\nTest for<\/span> Funnel Plot Asymmetry:<\/span> z =<\/span> -<\/span>0.7411<\/span>,<\/span> p =<\/span> 0.4586<\/span>\nLimit Estimate <\/span>(<\/span>as sei -><\/span> 0<\/span>):<\/span>   b =<\/span> -<\/span>0.1311 <\/span>(<\/span>CI:<\/span> -<\/span>0.4408<\/span>,<\/span> 0.1787<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
\u6709\u610f\u78ba\u7387\u306f p=0.4586 \u3068\u3084\u3084\u4f4e\u4e0b\u3057\u305f\u304c\u3001\u6709\u610f\u6c34\u6e9610\uff05\u3068\u3057\u3066\u3001\u7d71\u8a08\u5b66\u7684\u6709\u610f\u3067\u306f\u306a\u304f\u3001\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u304c\u3042\u308b\u3068\u306f\u8a00\u3048\u306a\u3044\u3068\u3044\u3046\u7d50\u679c\u3068\u306a\u308b\u3002<\/p>\n\n\n\n
Macaskill\u691c\u5b9a\u3000\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u3092\u56de\u5e30\u5206\u6790\u3067\u691c\u51fa\u3059\u308b\u65b9\u6cd5<\/h2>\n\n\n\n
Macaskill\u306e\u65b9\u6cd5\u306f\u3001\u4ee5\u4e0b\u306e\u56de\u5e30\u30e2\u30c7\u30eb\u3092\u7528\u3044\u305f\u91cd\u307f\u3065\u3051\u56de\u5e30\u5206\u6790\u3060\u3002<\/p>\n\n\n\n
$$ \\hat{\\theta}_i = \\alpha + \\beta n_i + \u8aa4\u5dee $$<\/p>\n\n\n\n\n\n\n\n
\u63a8\u5b9a\u5024 $ \\hat{\\theta}_i $ \u3092\u30b5\u30f3\u30d7\u30eb\u30b5\u30a4\u30ba $ n_i $ \u3067\u4e88\u6e2c\u3059\u308b\u56de\u5e30\u5f0f\u3060\u3002<\/p>\n\n\n\n
\u91cd\u307f\u306f\u5206\u6563\u306e\u9006\u6570\u3092\u7528\u3044\u308b\u3002<\/p>\n\n\n\n
\u30d5\u30a1\u30f3\u30cd\u30eb\u30d7\u30ed\u30c3\u30c8\u56de\u5e30\u6cd5\u3068\u3082\u547c\u3070\u308c\u308b\u3002<\/p>\n\n\n\n
\u3082\u3057\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u304c\u3042\u308b\u306a\u3089\u3001\u30b5\u30f3\u30d7\u30eb\u30b5\u30a4\u30ba\u304c\u5c0f\u3055\u3044\u7814\u7a76\u307b\u3069\u3001\u7814\u7a76\u7d50\u679c\u306e\u63a8\u5b9a\u5024\u304c\u5927\u304d\u3044\u307b\u3046\u306b\u504f\u308b\u306f\u305a\u3002<\/p>\n\n\n\n
\u3053\u308c\u306f\u3001\u771f\u306e\u63a8\u5b9a\u5024\u304c\u6b63\u306e\u5834\u5408\u3067\u3042\u308b\u3002<\/p>\n\n\n\n
\u8ca0\u306e\u5834\u5408\u306f\u3001\u5c0f\u3055\u3044\u307b\u3046\u306b\u504f\u308b\u3002<\/p>\n\n\n\n
\u30b5\u30f3\u30d7\u30eb\u30b5\u30a4\u30ba\u304c\u5c0f\u3055\u3044\u3068\u7d71\u8a08\u5b66\u7684\u6709\u610f\u306b\u306a\u308a\u306b\u304f\u304f\u3001\u63a8\u5b9a\u5024\u304c\u5927\u304d\u3044\u304c\u305f\u3081\u306b\u7d71\u8a08\u5b66\u7684\u6709\u610f\u306b\u306a\u3063\u305f\u3068\u3044\u3046\u504f\u3063\u305f\u7814\u7a76\u306e\u307f\u304c\u63a1\u629e\u3055\u308c\u3066\u3044\u308b\u53ef\u80fd\u6027\u304c\u3042\u308b\u3002<\/p>\n\n\n\n
\u3053\u306e\u504f\u308a\u3092\u691c\u51fa\u3059\u308b\u308f\u3051\u3060\u3002<\/p>\n\n\n\n
\u30d5\u30a1\u30f3\u30cd\u30eb\u30d7\u30ed\u30c3\u30c8\u304c\u5de6\u53f3\u5bfe\u79f0\u5f62\u306a\u3089\u3001\u56de\u5e30\u5f0f\u306e\u50be\u304d\u304c\u30bc\u30ed\u306b\u306a\u308b\u3002<\/p>\n\n\n\n
\u3059\u306a\u308f\u3061\u5e30\u7121\u4eee\u8aac\u306f \u03b2 = 0 \u3060\u3002<\/p>\n\n\n\n
\u5404\u7814\u7a76\u306e\u30b5\u30f3\u30d7\u30eb\u30b5\u30a4\u30ba $ n_i $ \u3092\u8a08\u7b97\u3057\u3066\u3001\u5148\u307b\u3069\u306e\u56de\u5e30\u30e2\u30c7\u30eb\u3067 \u03b2 \u3092\u63a8\u5b9a\u3059\u308b\u3002<\/p>\n\n\n\n
ni\u3000<-<\/span> dat$<\/span>n1+<\/span>dat$<\/span>n0\nres.theta.n <-<\/span> lm<\/span>(<\/span>thetai ~<\/span> ni,<\/span> weights=<\/span>1<\/span>\/<\/span>dat.escalc$<\/span>vi)<\/span>\nsummary<\/span>(<\/span>res.theta.n)<\/span>\n<\/code><\/pre>\n\n\n\n
\u7d50\u679c\u306f\u4ee5\u4e0b\u306e\u901a\u308a\u3002<\/p>\n\n\n\n
><\/span> summary<\/span>(<\/span>res.theta.n)<\/span>\nCall:<\/span>\nlm<\/span>(<\/span>formula =<\/span> thetai ~<\/span> ni,<\/span> weights =<\/span> 1<\/span>\/<\/span>dat.escalc$<\/span>vi)<\/span>\nWeighted Residuals:<\/span>\nMin      1Q  Median      3Q     Max\n-<\/span>1.9432<\/span> -<\/span>0.7026<\/span> -<\/span>0.3085<\/span>  0.7728<\/span>  2.5467<\/span>\nCoefficients:<\/span>\nEstimate Std. Error t value Pr<\/span>(<\/span>>|<\/span>t|<\/span>)<\/span>\n(<\/span>Intercept)<\/span> -<\/span>2.269e-01<\/span>  1.250e-01<\/span>  -<\/span>1.816<\/span>   0.0895<\/span> .\nni          -<\/span>8.469e-06<\/span>  5.209e-05<\/span>  -<\/span>0.163<\/span>   0.8730<\/span>\n---<\/span>\nSignif. codes:<\/span>  0<\/span> \u2018***<\/span>\u2019 0.001<\/span> \u2018**<\/span>\u2019 0.01<\/span> \u2018*<\/span>\u2019 0.05<\/span> \u2018.\u2019 0.1<\/span> \u2018 \u2019 1<\/span>\nResidual standard error:<\/span> 1.196<\/span> on 15<\/span> degrees of freedom\nMultiple R-<\/span>squared:<\/span>  0.001759<\/span>,<\/span>  Adjusted R-<\/span>squared:<\/span>  -<\/span>0.06479<\/span>\nF<\/span>-<\/span>statistic:<\/span> 0.02643<\/span> on 1<\/span> and 15<\/span> DF,<\/span>  p-<\/span>value:<\/span> 0.873<\/span>\n<\/code><\/pre>\n\n\n\n
ni\u306eCoefficient\u306eEstimate\u3092\u898b\u308b\u3068\u307b\u307c\u30bc\u30ed\u3002<\/p>\n\n\n\n
p\u5024\uff08Pr(>|t|)\uff09\u3092\u898b\u308b\u30680.873\u3067\u3001\u03b2 = 0 \u306e\u5e30\u7121\u4eee\u8aac\u3092\u68c4\u5374\u3067\u304d\u306a\u3044\u3002<\/p>\n\n\n\n
\u3064\u307e\u308a\u3001\u03b2 = 0 \u304c\u5426\u5b9a\u3067\u304d\u306a\u3044\u3002<\/p>\n\n\n\n
\u3059\u306a\u308f\u3061\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u304c\u3042\u308b\u53ef\u80fd\u6027\u306f\u4f4e\u3044\u3068\u8003\u3048\u3089\u308c\u308b\u3002<\/p>\n\n\n\n
\u5404\u7814\u7a76\u3092\u30d7\u30ed\u30c3\u30c8\u3057\u3066\u3001\u56de\u5e30\u76f4\u7dda\u3092\u5f15\u3044\u3066\u307f\u308b\u3002<\/p>\n\n\n\n
\u50be\u304d\u304c\u307b\u307c\u30bc\u30ed\u3067\u3042\u308b\u3053\u3068\u304c\u78ba\u8a8d\u3067\u304d\u308b\u3002<\/p>\n\n\n\n
\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u304c\u3042\u308b\u3068\u306f\u65ad\u5b9a\u3067\u304d\u306a\u3044\u3002<\/p>\n\n\n\n
plot<\/span>(<\/span>thetai ~<\/span> ni,<\/span> xlab=<\/span>\"Sample Size\"<\/span>,<\/span> ylab=<\/span>\"log(OR)\"<\/span>,<\/span> xlim=<\/span>c<\/span>(<\/span>0<\/span>,<\/span>4000<\/span>))<\/span>\nabline<\/span>(<\/span>h=<\/span>0<\/span>,<\/span> v=<\/span>0<\/span>,<\/span> lty=<\/span>1<\/span>)<\/span>\nabline<\/span>(<\/span>res.theta.n,<\/span> lty=<\/span>2<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
\"\"<\/figure>\n\n\n\n
<\/span><\/p>\n\n\n\n
metafor\u30d1\u30c3\u30b1\u30fc\u30b8\u306eregtest()\u3092\u4f7f\u3046\u3068\u3053\u3061\u3089\u3082\u7c21\u5358\u3060\u3002<\/p>\n\n\n\n
regtest<\/span>(<\/span>res.reml,<\/span> model=<\/span>\"lm\"<\/span>,<\/span> predictor=<\/span>\"ni\"<\/span>)<\/span>\n<\/code><\/pre>\n\n\n\n
\u7d50\u679c\u306f\u4ee5\u4e0b\u306e\u901a\u308a\u3067\u3001<\/p>\n\n\n\n
t\u5024\u3068p\u5024\u304c\u3001regtest()\u3092\u4f7f\u308f\u306a\u3044\u65b9\u6cd5\u3068\u4e00\u81f4\u3057\u3066\u3044\u308b\u3002<\/p>\n\n\n\n
><\/span> regtest<\/span>(<\/span>res.reml,<\/span> model=<\/span>\"lm\"<\/span>,<\/span> predictor=<\/span>\"ni\"<\/span>)<\/span>\nRegression Test for<\/span> Funnel Plot Asymmetry\nmodel:<\/span>     weighted regression with multiplicative dispersion\npredictor:<\/span> sample size\ntest for<\/span> funnel plot asymmetry:<\/span> t =<\/span> -<\/span>0.1626<\/span>,<\/span> df =<\/span> 15<\/span>,<\/span> p =<\/span> 0.8730<\/span>\n<\/code><\/pre>\n\n\n\n
\u307e\u3068\u3081<\/h2>\n\n\n\n
\u51fa\u7248\u30d0\u30a4\u30a2\u30b9\u306e\u6709\u7121\u3092\u691c\u8a0e\u3059\u308b\u65b9\u6cd5\u3068\u3057\u3066\u3001Begg\u306e\u65b9\u6cd5\u3001Egger\u306e\u65b9\u6cd5\u3001Macaskill\u306e\u65b9\u6cd5\u3092\u5b9f\u65bd\u3057\u3066\u307f\u305f\u3002<\/p>\n\n\n\n
\u30d5\u30a1\u30f3\u30cd\u30eb\u30d7\u30ed\u30c3\u30c8\u3068\u6bd4\u3079\u3066\u6570\u5024\u3068\u7d71\u8a08\u5b66\u7684\u691c\u5b9a\u3067\u5224\u65ad\u3067\u304d\u308b\u306e\u3067\u3001\u767d\u9ed2\u3064\u3051\u3084\u3059\u304f\u308f\u304b\u308a\u3084\u3059\u3044\u304c\u3001\u30d5\u30a1\u30f3\u30cd\u30eb\u30d7\u30ed\u30c3\u30c8\u3068\u5408\u308f\u305b\u3066\u7dcf\u5408\u7684\u306b\u5224\u65ad\u3059\u308b\u5fc5\u8981\u304c\u3042\u308b\u3002<\/p>\n\n\n\n
R\u306emetafor\u30d1\u30c3\u30b1\u30fc\u30b8\u306e\u95a2\u6570\u3092\u4f7f\u3046\u3068\u7c21\u5358\u306b\u8a08\u7b97\u3067\u304d\u308b\u3002<\/p>\n\n\n\n
\u4f55\u3089\u304b\u53c2\u8003\u306b\u306a\u308c\u3070\u5e78\u3044\u3002<\/p>\n\n\n\n
\u53c2\u8003\u66f8\u7c4d<\/h2>\n\n\n\n
\u4e39\u5f8c\u4fca\u90ce\u8457\u3000\u30e1\u30bf\u30fb\u30a2\u30ca\u30ea\u30b7\u30b9\u5165\u9580\u3000\u671d\u5009\u66f8\u5e97
6. Publication bias\u3078\u306e\u6311\u6226
6.1 \u516c\u8868\u30d0\u30a4\u30a2\u30b9\u306e\u691c\u51fa<\/p>\n\n\n\n
\"\u30e1\u30bf\u30fb\u30a2\u30ca\u30ea\u30b7\u30b9\u5165\u9580\u2015\u30a8\u30d3\u30c7\u30f3\u30b9\u306e\u7d71\u5408\u3092\u3081\u3056\u3059\u7d71\u8a08\u624b\u6cd5<\/a><\/figure>\n\n\n\n
<\/div>\n\n\n\n
\n
\u30e1\u30bf\u30fb\u30a2\u30ca\u30ea\u30b7\u30b9\u5165\u9580\u2015\u30a8\u30d3\u30c7\u30f3\u30b9\u306e\u7d71\u5408\u3092\u3081\u3056\u3059\u7d71\u8a08\u624b\u6cd5 (\u533b\u5b66\u7d71\u8a08\u5b66\u30b7\u30ea\u30fc\u30ba)<\/a><\/p>\n