Meta analysis in stata an updated collection from the stata journal

Meta-Analysis in Stata: An Updated Collection from the Stata Journal Second Edition TOM M PALMER, collection editor Department of Mathematics and Statistics Lancaster University Lancaster, UK JONATHAN A C STERNE, collection editor School of Social and Community Medicine University of Bristol Bristol, UK H JOSEPH NEWTON, Stata Journal editor Department of Statistics Texas A&M University College Station, TX NICHOLAS J COX, Stata Journal editor Department of Geography Durham University Durham City, UK ® A Stata Press Publication StataCorp LP College Station, Texas đ Copyright â 2009, 2016 by StataCorp LP All rights reserved First edition 2009 Second edition 2016 Published by Stata Press, 4905 Lakeway Drive, College Station, Texas 77845 Typeset in LATEX Printed in the United States of America 10 Print ISBN-10: 1-59718-147-1 Print ISBN-13: 978-1-59718-147-1 ePub ISBN-10: 1-59718-221-4 ePub ISBN-13: 978-1-59718-221-8 Mobi ISBN-10: 1-59718-222-2 Mobi ISBN-13: 978-1-59718-222-5 Library of Congress Control Number: 2015950607 No part of this book may be reproduced, stored in a retrieval system, or transcribed, in any form or by any means—electronic, mechanical, photocopy, recording, or otherwise—without the prior written permission of StataCorp LP Stata, , Stata Press, Mata, StataCorp LP , and NetCourse are registered trademarks of Stata and Stata Press are registered trademarks with the World Intellectual Property Organization of the United Nations LATEX is a trademark of the American Mathematical Society Contents Introduction to the second edition References Introduction to the first edition References Install the software I Meta-analysis in Stata: metan, metaan, metacum, and metap References metan—a command for meta-analysis in Stata 1.1 Background 1.2 Data structure 1.3 Analysis of binary data using fixed-effects models 1.4 Analysis of continuous data using fixed-effects models 1.5 Test for heterogeneity 1.6 Analysis of binary or continuous data using random-effects models 1.7 Tests of overall effect 1.8 Graphical analyses 1.9 Syntax for metan 1.10 Options for metan 1.11 Saved results from metan (macros) 1.12 Syntax for funnel 1.13 Options for funnel 1.14 Syntax for labbe 1.15 Options for labbe 1.16 Example 1: Interventions in smoking cessation 1.17 Example 1.18 Formulas 1.19 Individual study responses: binary outcomes 1.20 Individual study responses: continuous outcomes 1.21 Mantel–Haenszel methods for combining trials 1.22 Inverse variance methods for combining trials 1.23 Peto’s assumption free method for combining trials 1.24 DerSimonian and Laird random-effects models 1.25 Confidence intervals 1.26 Test statistics 1.27 Acknowledgments 1.28 References metan: fixed- and random-effects meta-analysis 2.1 Introduction 2.2 Example data 2.3 Syntax 2.4 Basic use 2.5 Displaying data columns in graphs 2.6 by() processing 2.7 User-defined analyses 2.8 New analysis options 2.9 New output 2.10 More graph options 2.11 Variables and results produced by metan 2.12 References metaan: Random-effects meta-analysis 3.1 Introduction 3.2 The metaan command 3.3 Methods 3.4 Example 3.5 Discussion 3.6 Acknowledgments 3.7 References Cumulative meta-analysis 4.1 Syntax 4.2 Options 4.3 Background 4.4 Example 4.5 Note 4.6 Acknowledgments 4.7 References Meta-analysis of p-values 5.1 Fisher’s method 5.2 Edgington’s methods 5.3 Syntax 5.4 Option 5.5 Example 5.6 Individual or frequency records 5.7 Saved results 5.8 References II Meta-regression: metareg References Meta-regression in Stata 6.1 Introduction 6.2 Basis of meta-regression 6.3 Relation to other Stata commands 6.4 Background to examples 6.5 New and enhanced features 6.6 Syntax, options, and saved results 6.7 Methods and formulas 6.8 Acknowledgments 6.9 References Meta-analysis regression 7.1 Background 7.2 Method-of-moments estimator 7.3 Iterative procedures 7.4 Syntax 7.5 Options 7.6 Example 7.7 Saved results 7.8 Acknowledgments 7.9 References III Investigating bias in meta-analysis: metafunnel, confunnel, metabias, metatrim, and extfunnel References Funnel plots in meta-analysis 8.1 Introduction 8.2 Funnel plots 8.3 Syntax 8.4 Description 8.5 Options 8.6 Examples 8.7 Acknowledgments 8.8 References Contour-enhanced funnel plots for meta-analysis 9.1 Introduction 9.2 Contour-enhanced funnel plots 9.3 The confunnel command 9.4 Use of confunnel 9.5 Discussion 9.6 References 10 Updated tests for small-study effects in meta-analyses 10.1 Introduction 10.2 Syntax 10.3 Options 10.4 Background 10.5 Example 10.6 Saved results 10.7 Discussion 10.8 Acknowledgment 10.9 References 11 Tests for publication bias in meta-analysis 11.1 Syntax 11.2 Description 11.3 Options 11.4 Input variables 11.5 Explanation 11.6 Begg’s test 11.7 Egger’s test 11.8 Examples 11.9 Saved results 11.10 References 12 Tests for publication bias in meta-analysis 12.1 Modification of the metabias program 12.2 References 13 Nonparametric trim and fill analysis of publication bias in metaanalysis 13.1 Syntax 13.2 Description 13.3 Options 13.4 Specifying input variables 13.5 Explanation 13.6 Estimators of the number of suppressed studies 13.7 The iterative trim and fill algorithm 13.8 Example 13.9 Remarks 13.10 Saved results 13.11 Note 13.12 References 14 Graphical augmentations to the funnel plot to assess the impact of a new study on an existing meta-analysis 14.1 Introduction 14.2 Methodology 14.3 The extfunnel command 14.4 Example uses of extfunnel 14.5 Additional feature 14.6 Discussion 14.7 Acknowledgments 14.8 References IV Multivariate meta-analysis: metandi, mvmeta References 15 metandi: Meta-analysis of diagnostic accuracy using hierarchical logistic regression 15.1 Introduction 15.2 Example: Lymphangiography for diagnosis of lymph node metastasis 15.3 Models for meta-analysis of diagnostic accuracy 15.4 metandi output 15.5 metandiplot 15.6 predict after metandi 15.7 Syntax and options for commands 15.8 Methods and formulas 15.9 Acknowledgments 15.10 References 16 Multivariate random-effects meta-analysis 16.1 Introduction 16.2 Multivariate random-effects meta-analysis with mvmeta 16.3 Details of mvmeta 16.4 A utility command to produce data in the correct format: mvmeta_make 16.5 Example 1: Telomerase data 16.6 Example 2: Fibrinogen Studies Collaboration data 16.7 Perfect prediction 16.8 Discussion 16.9 Acknowledgments 16.10 References 17 Multivariate random-effects meta-regression: Updates to mvmeta 17.1 Introduction 17.2 mvmeta: Multivariate random-effects meta-regression 17.3 Details 17.4 Example 17.5 Difficulties and limitations 17.6 Acknowledgments 17.7 References V Individual patient data meta-analysis: ipdforest and ipdmetan References 18 A short guide and a forest plot command (ipdforest) for one-stage meta-analysis 18.1 Introduction 18.2 Individual patient data meta-analysis 18.3 The ipdforest command 18.4 Discussion 18.5 Acknowledgments 18.6 References 19 Two-stage individual participant data meta-analysis and generalized forest plots 19.1 Introduction 19.2 Two-stage IPD meta-analysis 19.3 The ipdmetan command 10 Talwar, M., 22.2 Tansella, M., 22.3.7 Tatsioni, A., 15.1 Taylor, J E., 6.5.5 Taylor, S J., Teo, K K., 9.2 Terpos, E., 20.3 Terrin, N., , 8.2.1 , 8.2.3 Tervonen, T., 21.5 Tetzlaff, J., Thatcher, N., 21.3.7 The ALSPAC Study Team, 6.5.5 Thiel, E., 20.3 Thijs, L., 19.1 , 19.4.4 Thijs, V., 22.2 Thomas, A., 2.7.2 , 22.3.1 Thomas, J., 21.3.2 Thompson, D., 20.4 Thompson, J R., 14 , 14.1 , 14.2.4 , 14.3.2 , 15.8 , 16.1 , 16.3.2 , 16.5 , 16.5.1 , 16.5.2 , 17.1 , 17.2.2 , 17.3.3 , 17.3.5 , 22 , 25.4.1 , 27 , 27.1 , 27.2.1 , 27.2.2 , 27.7 Thompson, R., 25.2.2 Thompson, S G., 2.4.2 , 2.6 , 2.7.2 , 2.9.1 , 2.9.2 , 3.1 , 3.3 , , 6.1 , 6.3 , 6.4 , 6.5 , 6.5.1 , 6.5.2 , 6.5.3 , 6.5.5 , 6.5.6 , 6.6.2 , 6.6.3 , 6.7.1 , 7.6 , 8.2.5 , 16.1 , 16.2.2 , 16.8.2 , 17.1 , 17.2.3 , 17.3.4 , 17.3.6 , 18.2 , 18.3.4 , 19.1 , 19.2.1 , 19.2.2 , 19.2.3 , 21.3.5 , 22.3.6 , 25.2.1 , 25.2.2 , 25.3.1 Thorlund, K., 20.4 , 21.1 , 26.1 , 26.2 , 26.5.1 Tierney, J F., 19.1 , 19.2.3 , 19.4 , 26.1 Tobias, A., , 27.8 Tramer, M R., 8.2.3 Tsionos, K., 20.3 Tsuruta, H., 22.1 Tudur Smith, C., 19.1 Turesson, I., 20.3 Turner, E H., 14.1 Turner, R M., 18.2 , 25.2.1 Tweedie, R L., , 9.4.1 , 9.4.2 , 13.2 , 19.2.2 U Ungerleider, R S., 14.2.4 647 V Valentine, J C., 22.3.6 van der Windt, D A W M., 15.1 van Houwelingen, H C., 16.1 , 16.3.2 , 16.8.1 , 17.1 van Houwelingen, J C., 16.8.1 van Valkenhoef, G., 21.5 Vangsted, A., 20.3 Vasiliadis, H S., 20.1 , 22.1 , 22.3.4 , 22.3.5 Verbeke, G., 17.2.3 Veroniki, A A., 22.1 , 22.3.4 Verweij, P J M., 16.8.1 Viechtbauer, W., 19.2.2 , 22.3.4 , 25.2.2 , 25.6 Vignon, E., 20.3 Viniou, N., 20.3 Virkkunen, P., 20.3 Viwatwongkasem, C., 3.3 W Waage, A., 20.3 Wacholder, S, 6.5.5 Wald, N J., 9.4.2 Walsh, T., 20.4 Walter, S D., 10.4.2 , 15.3 , 20.1 , 22.1 , 22.3.4 Wang, J., 19.1 , 19.4.4 Warn, D E., 2.7.2 Watanabe, N., 22.3.7 Weintraub, M., 1.17 Wells, G A., 20.2 Welton, N J., 20.1 , 21.3.6 , 21.4.1 , 21.5 , 22.1 , 22.3.3 , 22.3.5 Westfall, P H., 6.5.5 Westin, J., 20.3 Wetterslev, J., 26.1 , 26.2 , 26.5.1 Wetterwald, M., 20.3 Wheatley, K., 20.3 White, I R., 3.1 , , 14 , 16.2.2 , 17.1 , 17.2.1 , 17.2.4 , 17.3.3 , 17.3.4 , 17.4.1 , 19 , 19.4.6 , 20.1 , 21.1 , 21.2 , 21.3.1 , 21.3.2 , 21.3.4 , 21.3.5 , 21.3.9 , 21.5 , 22 , 22.1 , 22.3.1 , 22.3.3 , 22.3.6 , 22.4 , 24.1 , 24.2.2 , 24.2.4 , 24.3.1 , 24.4.2 , 24.5 , 25.1 , 25.2.2 , 25.5 , 25.6 Whitehead, A., 10.4.2 , 18.2 , 21.2 Whitehead, J., 10.4.2 648 Whiting, P., 14 , 15.1 , 15.2 , 15.3 , 15.8 , 15.8.2 , 15.8.3 , 17.5 , 25.4.1 , 25.6 , 27.2.2 , 27.4.2 Wieland, G D., 8.2.3 Wilde, M., 25.2.2 , 25.4 , 25.6 Willan, A., 16.1 , 16.3.2 Willett, W C., 23.1 , 23.4.2 Williamson, P R., 17.5 , 19.1 Wilson, D B., 6.5.3 Wilson, J., 20.3 Wilson, K., 20.3 Wilson, M E., 2.2 , 7.6 Wisløff, F., 20.3 Wolk, A., 23.1 , 23.4.2 , 23.4.3 , 23.4.4 Wood, A M., 24.1 , 24.2.2 , 24.2.4 , 24.3.1 , 24.4.2 , 24.5 Woods, B S., 21.3.7 Woolf, B., 10.4.1 Wright, J., 14.4.1 Y Yaju, Y., 22.1 Yang, J., 14.4.1 , 14.4.2 Yang, M., 18.2 , 25.2.1 Yataganas, X., 20.3 Young, S S., 6.5.5 Yu, K K., 15.2 Yusuf, S., 1.3 , 9.2 , 19.1 , 19.4.5 , 26.1 , 26.2 Z Zarin, D A., 15.1 Zee, B., 20.3 Zhang, D., 14.4.1 , 14.4.2 Zhang, Y., 14.4.1 , 14.4.2 Zhou, D., 14.4.1 , 14.4.2 Zhou, M., 14.4.1 , 14.4.2 Zwinderman, A H., 14 , 15.1 , 15.2 , 15.3 , 15.3.2 , 15.4 , 15.8 , 15.8.1 , 15.8.3 , 25.4.1 , 25.4.2 649 Subject index C clusterank command, 22 , 22.5 confunnel command, , 9.5 E extfunnel command, 14 , 14.7 F funnel command, 1.12 , 1.13 , 1.16 G glst command, 23 , 23.6 I ifplot command, 22 , 22.5 indirect command, 20 , 20.4 intervalplot command, 22 , 22.5 ipdforest command, 18 , 18.5 ipdmetan command, 19 , 19.6 L labbe command, 1.13 , 1.15 , 1.16 M mdsrank command, 22 , 22.5 meta command, 7.6 , 7.6 meta_lr command, 27.8 metaan command, , 3.6 metabias command, 9.5 , 10.8 , 11 , 11.9 , 11.9 , 12.1 metacum command, , 4.6 metacumbounds command, 26 , 26.7 metacurve command, 27.8 metaeff command, 27.8 metafunnel command, , 8.7 , 9.2 , 9.2 , 10.5 metamiss command, 24 , 24.5 metan command, 1.1 , 1.27 , , 2.11.2 , 4.4 , 8.6 , 10.5 650 , 10.5 metandi command, 15 , 15.9 predict after, 15.5 , 15.6 , 15.7.2 , 15.7.3 metandiplot command, 15.4 , 15.5 , 15.7.1 , 15.7.2 metaninf command, 27.8 metannt command, 27.8 , 27.8 metap command, 4.6 , 5.7 metaparm command, 27.8 , 27.8 metapow command, 27 , 27.8 metapowplot command, 27 , 27.8 metareg command, , 6.8 , , 7.8 metasim command, 27 , 27.8 metatrim command, 13 , 13.11 midas command, 27.8 mvmeta command, 16 , 16.9 , 17 , 17.6 mvmeta_make command, 16.3.4 , 16.4.2 , 16.6 N netfunnel command, 22 , 22.5 netleague command, 22 , 22.5 netweight command, 22 , 22.5 network components command, 20.4 , 21.6 network convert command, 20.4 , 21.6 network forest command, 20.4 , 21.6 network import command, 20.4 , 21.6 network map command, 20.4 , 21.6 network meta command, 20.4 , 21.6 network pattern command, 20.4 , 21.6 network query command, 20.4 , 21.6 network rank command, 20.4 , 21.6 network setup command, 20.4 , 21.6 network sidesplit command, 20.4 , 21.6 network table command, 20.4 , 21.6 network unset command, 20.4 , 21.6 networkplot command, 22 , 22.5 S sucra command, 22 , 22.5 651 目录 Introduction to the second edition References 14 16 Introduction to the first edition References 19 21 Install the software I Meta-analysis in Stata: metan, metaan, metacum, and metap References 24 26 27 metan—a command for meta-analysis in Stata 1.1 Background 1.2 Data structure 1.3 Analysis of binary data using fixed-effects models 1.4 Analysis of continuous data using fixed-effects models 1.5 Test for heterogeneity 1.6 Analysis of binary or continuous data using random-effects models 1.7 Tests of overall effect 1.8 Graphical analyses 1.9 Syntax for metan 1.10 Options for metan 1.11 Saved results from metan (macros) 1.12 Syntax for funnel 1.13 Options for funnel 1.14 Syntax for labbe 1.15 Options for labbe 1.16 Example 1: Interventions in smoking cessation 1.17 Example 1.18 Formulas 1.19 Individual study responses: binary outcomes 1.20 Individual study responses: continuous outcomes 1.21 Mantel–Haenszel methods for combining trials 652 28 28 29 30 31 32 32 33 33 33 34 41 43 43 44 44 45 49 51 51 52 53 1.22 Inverse variance methods for combining trials 1.23 Peto’s assumption free method for combining trials 1.24 DerSimonian and Laird random-effects models 1.25 Confidence intervals 1.26 Test statistics 1.27 Acknowledgments 1.28 References 54 55 55 56 56 57 57 metan: fixed- and random-effects meta-analysis 60 2.1 Introduction 2.2 Example data 2.3 Syntax 2.4 Basic use 2.5 Displaying data columns in graphs 2.6 by() processing 2.7 User-defined analyses 2.8 New analysis options 2.9 New output 2.10 More graph options 2.11 Variables and results produced by metan 2.12 References 61 62 64 65 70 72 76 78 79 82 85 86 metaan: Random-effects meta-analysis 89 3.1 Introduction 3.2 The metaan command 3.3 Methods 3.4 Example 3.5 Discussion 3.6 Acknowledgments 3.7 References 89 91 95 98 101 101 101 Cumulative meta-analysis 104 4.1 Syntax 4.2 Options 4.3 Background 4.4 Example 4.5 Note 104 105 110 110 115 653 4.6 Acknowledgments 4.7 References 115 115 Meta-analysis of p-values 117 5.1 Fisher’s method 5.2 Edgington’s methods 5.3 Syntax 5.4 Option 5.5 Example 5.6 Individual or frequency records 5.7 Saved results 5.8 References II Meta-regression: metareg 117 118 118 118 119 120 120 121 123 References 123 Meta-regression in Stata 125 6.1 Introduction 6.2 Basis of meta-regression 6.3 Relation to other Stata commands 6.4 Background to examples 6.5 New and enhanced features 6.6 Syntax, options, and saved results 6.7 Methods and formulas 6.8 Acknowledgments 6.9 References Meta-analysis regression 125 126 129 130 130 146 152 156 156 159 7.1 Background 7.2 Method-of-moments estimator 7.3 Iterative procedures 7.4 Syntax 7.5 Options 7.6 Example 7.7 Saved results 7.8 Acknowledgments 7.9 References 654 159 160 161 161 162 162 168 168 168 III Investigating bias in meta-analysis: metafunnel, confunnel, metabias, metatrim, and extfunnel References 170 171 Funnel plots in meta-analysis 8.1 Introduction 8.2 Funnel plots 8.3 Syntax 8.4 Description 8.5 Options 8.6 Examples 8.7 Acknowledgments 8.8 References 173 173 174 181 181 182 183 188 188 Contour-enhanced funnel plots for meta-analysis 9.1 Introduction 9.2 Contour-enhanced funnel plots 9.3 The confunnel command 9.4 Use of confunnel 9.5 Discussion 9.6 References 192 193 194 197 199 205 206 10 Updated tests for small-study effects in meta-analyses 209 10.1 Introduction 10.2 Syntax 10.3 Options 10.4 Background 10.5 Example 10.6 Saved results 10.7 Discussion 10.8 Acknowledgment 10.9 References 210 210 211 211 216 221 221 222 222 11 Tests for publication bias in meta-analysis 11.1 Syntax 11.2 Description 11.3 Options 225 225 225 226 655 11.4 Input variables 11.5 Explanation 11.6 Begg’s test 11.7 Egger’s test 11.8 Examples 11.9 Saved results 11.10 References 227 228 230 231 233 237 238 12 Tests for publication bias in meta-analysis 12.1 Modification of the metabias program 12.2 References 13 Nonparametric trim and fill analysis of publication bias in meta-analysis 13.1 Syntax 13.2 Description 13.3 Options 13.4 Specifying input variables 13.5 Explanation 13.6 Estimators of the number of suppressed studies 13.7 The iterative trim and fill algorithm 13.8 Example 13.9 Remarks 13.10 Saved results 13.11 Note 13.12 References 14 Graphical augmentations to the funnel plot to assess the impact of a new study on an existing meta-analysis 14.1 Introduction 14.2 Methodology 14.3 The extfunnel command 14.4 Example uses of extfunnel 14.5 Additional feature 14.6 Discussion 14.7 Acknowledgments 14.8 References 239 239 241 243 243 243 244 246 247 248 251 252 255 257 257 257 259 260 262 264 268 275 277 277 277 656 IV Multivariate meta-analysis: metandi, mvmeta References 279 279 15 metandi: Meta-analysis of diagnostic accuracy using hierarchical logistic regression 15.1 Introduction 15.2 Example: Lymphangiography for diagnosis of lymph node metastasis 15.3 Models for meta-analysis of diagnostic accuracy 15.4 metandi output 15.5 metandiplot 15.6 predict after metandi 15.7 Syntax and options for commands 15.8 Methods and formulas 15.9 Acknowledgments 15.10 References 16 Multivariate random-effects meta-analysis 16.1 Introduction 16.2 Multivariate random-effects meta-analysis with mvmeta 16.3 Details of mvmeta 16.4 A utility command to produce data in the correct format: mvmeta_make 16.5 Example 1: Telomerase data 16.6 Example 2: Fibrinogen Studies Collaboration data 16.7 Perfect prediction 16.8 Discussion 16.9 Acknowledgments 16.10 References 281 281 283 284 286 288 290 293 296 300 300 304 304 305 307 309 311 313 317 320 322 322 17 Multivariate random-effects meta-regression: Updates 325 to mvmeta 17.1 Introduction 17.2 mvmeta: Multivariate random-effects meta-regression 17.3 Details 17.4 Example 657 325 326 329 333 17.5 Difficulties and limitations 17.6 Acknowledgments 17.7 References 341 341 341 V Individual patient data meta-analysis: ipdforest and ipdmetan References 344 344 18 A short guide and a forest plot command (ipdforest) for one-stage meta-analysis 18.1 Introduction 18.2 Individual patient data meta-analysis 18.3 The ipdforest command 18.4 Discussion 18.5 Acknowledgments 18.6 References 19 Two-stage individual participant data meta-analysis and generalized forest plots 19.1 Introduction 19.2 Two-stage IPD meta-analysis 19.3 The ipdmetan command 19.4 Example 19.5 Discussion 19.6 Acknowledgments 19.7 References VI Network meta-analysis: indirect, network package, network_graphs package References 345 345 347 350 359 360 360 362 362 364 372 380 390 390 391 395 396 20 Indirect treatment comparison 20.1 Introduction 20.2 Adjusted indirect treatment comparison 20.3 Example: Zoledronate versus Pamidronate in multiple myeloma 20.4 Conclusion 20.5 References 658 398 399 399 402 405 405 21 Network meta-analysis 410 21.1 Introduction 21.2 Model for network meta-analysis 21.3 The network commands 21.4 Examples 21.5 Discussion 21.6 Acknowledgments 21.7 References 22 Visualizing assumptions and results in network metaanalysis: The network graphs package 22.1 Introduction 22.2 Example datasets 22.3 The network graphs package 22.4 Discussion 22.5 Acknowledgments 22.6 References VII Advanced methods: glst, metamiss, sem, gsem, metacumbounds, metasim, metapow, and metapowplot References 410 411 413 430 446 447 447 451 451 453 453 499 499 500 505 506 23 Generalized least squares for trend estimation of summarized dose–response data 508 23.1 Introduction 23.2 Method 23.3 The glst command 23.4 Examples 23.5 Empirical comparison of the WLS and GLS estimates 23.6 Conclusion 23.7 References 509 511 519 521 526 527 528 24 Meta-analysis with missing data 24.1 Introduction 24.2 metamiss command 24.3 Examples 24.4 Details 530 530 532 537 542 659 24.5 Discussion 24.6 References 543 543 25 Fitting fixed- and random-effects meta-analysis models using structural equation modeling with the sem and gsem commands 25.1 Introduction 25.2 Univariate outcome meta-analysis models 25.3 Univariate outcome meta-regression models 25.4 Multivariate outcome meta-analysis with zero within-study covariances 25.5 Multivariate outcome meta-analysis with nonzero within-study covariances 25.6 Conclusion 25.7 Acknowledgments 25.8 References 26 Trial sequential boundaries for cumulative metaanalyses 26.1 Introduction 26.2 Methods 26.3 R statistical software 26.4 The metacumbounds command 26.5 Examples 26.6 Discussion 26.7 Acknowledgment 26.8 References 545 545 546 559 565 570 575 577 577 582 583 583 586 587 590 597 598 598 27 Simulation-based sample-size calculation for designing new clinical trials and diagnostic test accuracy studies to 600 update an existing meta-analysis 27.1 Introduction 27.2 Methods 27.3 The metasim command 27.4 The metapow command 27.5 The metapowplot command 27.6 Other uses 601 602 605 610 613 621 660 27.7 Discussion 27.8 Acknowledgments 27.9 References 623 623 623 Appendix 626 27.10 References 628 661 ... reflects the continuing innovations in meta-analysis software made by the Stata community since the publication of the first edition in 2009 This new collection of articles about meta-analysis from the. .. (Reprinted in this collection in chapter 16.) White, I R., and J P T Higgins 2009 Meta-analysis with missing data Stata Journal 9: 57–69 (Reprinted in this collection in chapter 24.) 23 Install the. .. rather than at or nulloff removes the null hypothesis line from the graph favours(string # string) applies a label saying something about the treatment effect to either side of the graph (strings