Hello everyone! I am performing a meta-analysis in Medicine (Gastroenterology) on mortality rates in a liver disease (ACLF). I wish to perform a meta-regression analysis to explain heterogeneity in 30 day mortality rates based on both 1) region of the world from which study orignated and 2) mean severity of liver disease (mean MELD). For ease I've included an effect size and SE of effect size variable generated by metaprop command. Can someone help me create a result output where we can identify role of each region towards heterogeneity in mortality? (base reference for categorical variable whoregion can be taken as "4")
Code:
* Example generated by -dataex-. To install: ssc install dataex
clear
input float mortality30 long totalfor30days double(_ES _seES) byte whoregion double meanmeld
85 280 .30357142857142855 .02747829119287563 4 .
30 46 .6521739130434783 .0702237263554129 2 24.32
45 90 .5 .05270462766947299 4 .
18 29 .6206896551724138 .0901022421425571 2 20.7
164 274 .5985401459854015 .029613682945974978 2 26
33 74 .44594594594594594 .057783164016762936 6 28
64 200 .32 .03298484500494128 6 24
33 88 .375 .051607676490298154 6 26.3
38 108 .35185185185185186 .04595208114017962 4 23
12 24 .5 .10206207261596575 2 21.7
40 78 .5128205128205128 .056595237913628776 4 29.7
22 49 .4489795918367347 .0710557294866119 4 22.1
33 42 .7857142857142857 .06331466145913703 6 25
11 18 .6111111111111112 .11490438561102646 2 23.4
4 14 .2857142857142857 .1207363221040738 6 18
68 122 .5573770491803278 .04496883020924807 3 31.6
79 222 .35585585585585583 .03213306131260457 4 23.72
42 59 .711864406779661 .058961848213434656 5 36
0 103 0 .006782739974466059 4 23
38 75 .5066666666666667 .05772989468846051 4 24.6
67 252 .26587301587301587 .02783059262979156 6 26
99 99 1 .007053367989832942 4 30
171 334 .5119760479041916 .027350933725897416 3 26.8
10 24 .4166666666666667 .10063456073742666 4 29
7 15 .4666666666666667 .12881223774390613 6 24.69
15 39 .38461538461538464 .07790302823720513 4 27
17 40 .425 .07816249100431741 3 26
39 104 .375 .04747215783204687 4 21.3
17 39 .4358974358974359 .0794033618863734 4 31
11 27 .4074074074074074 .09456070200062598 6 30.52
16 59 .2711864406779661 .057878357986615475 4 19
42 77 .5454545454545454 .056744343993850845 4 26
40 71 .5633802816901409 .05886041495942303 4 27
17 38 .4473684210526316 .08066009201997748 4 23.3
49 58 .8448275862068966 .04754198431127886 4 26.5
12 50 .24 .060398675482166 4 28
7 22 .3181818181818182 .09930260198923897 4 25.5
47 92 .5108695652173914 .05211628433566957 6 18
15 57 .2631578947368421 .05832543563196217 4 27
45 125 .36 .04293250516799597 6 25.5
16 22 .7272727272727273 .09495144870310791 4 31
107 249 .42971887550200805 .03137162683122028 6 23
17 44 .38636363636363635 .07340528480781205 4 .
45 52 .8653846153846154 .04733149927744425 5 .
14 70 .2 .047809144373375745 3 .
44 58 .7586206896551724 .056188619979598856 2 .
18 40 .45 .07866066361276136 5 28
97 148 .6554054054054054 .03906415028693721 2 .
185 249 .7429718875502008 .027693435194149907 5 17.3
88 176 .5 .037688918072220454 4 .
29 89 .3258426966292135 .04968091823711921 4 22
55 153 .35947712418300654 .03879334568065554 4 26
200 346 .5780346820809249 .02655077959473571 3 27
682 1112 .6133093525179856 .014603912905486616 3 .
8 39 .20512820512820512 .064658991995499 3 .
47 102 .46078431372549017 .049354870426373425 3 .
45 91 .4945054945054945 .052411077011539936 3 29
53 100 .53 .04990991885387112 3 31
34 54 .6296296296296297 .06571489474350334 3 30.6
51 106 .4811320754716981 .04852970316111162 3 33
266 530 .5018867924528302 .021718457501703082 6 .
24 53 .4528301886792453 .06837397288386757 6 22.8
19 44 .4318181818181818 .07467372053099669 3 24
119 164 .725609756097561 .03484287360681495 6 .
12 30 .4 .08944271909999159 6 27.2
21 50 .42 .06979971346646059 3 29
4 66 .06060606060606061 .02937040512712484 6 21
447 1049 .4261201143946616 .015268232861157138 3 27
4 10 .4 .15491933384829668 6 27
58 109 .5321100917431193 .04779245478127932 6 24
28 28 1 .024171880321350914 3 36.15
0 36 0 .01898142653058624 3 26.8
115 188 .6117021276595744 .035544595504167935 6 25
10 50 .2 .0565685424949238 3 .
82 208 .3942307692307692 .03388419686555204 3 25.56
10 40 .25 .06846531968814576 6 24.6
35 446 .07847533632286996 .012733642273220071 4 19.13
41 155 .2645161290322581 .03542801393048125 6 .
1424 3009 .4732469258889997 .009101989838389698 3 28
146 689 .21190130624092887 .01556852648502191 6 23.1
28 50 .56 .07019971509913697 3 24.93
62 248 .25 .02749633406650374 6 .
23 400 .0575 .01163977555625537 6 22
109 264 .4128787878787879 .030302133538154235 2 .
55663 106634 .5220004876493426 .001529682614815464 2 .
50 67 .746268656716418 .053161503959820555 4 27
51 117 .4358974358974359 .04584355235965895 4 24
90 159 .5660377358490566 .039305209333871195 4 16.4
30 80 .375 .05412658773652741 6 26
941 1934 .48655635987590484 .0113654001627059 6 .
55 200 .275 .031573327350787724 6 24.7
15 34 .4411764705882353 .08515380416833711 6 29
18 53 .33962264150943394 .06505137344040325 6 27
195 525 .37142857142857144 .021087996484351088 6 23.5
end
label values whoregion whoregion
label def whoregion 2 "Americas", modify
label def whoregion 3 "South-East Asian", modify
label def whoregion 4 "European", modify
label def whoregion 5 "Eastern-Mediterranean Region", modify
label def whoregion 6 "Western-Pacific Region", modify
label var mortality30 "n"
label var totalfor30days "N"
label var _ES "Effect size"
label var _seES "Standard error of effect size"
label var whoregion "WHO Region"
label var meanmeld "Mean MELD"
