Cochrane Handbook 5.0 封面和索引

OTE/SPH OTE/SPH JWBK247-FM August 22, 2008 20:58 Printer Name: Yet to Come Contents Preface xiii Acknowledgements xv The Handbook editors xvii Major contributors xix Part 1 COCHRANE REVIEWS 1 1 Introduction 3 Sally Green, Julian PT Higgins, Philip Alderson, Mike Clarke, Cynthia D Mulrow and Andrew D Oxman 1.1 The Cochrane Collaboration 3 1.2 Systematic reviews 6 1.3 About this Handbook 7 1.4 Contributors to the Handbook 8 1.5 Chapter information 9 1.6 References 9 2 Preparing a Cochrane review 11 Edited by Sally Green and Julian PT Higgins 2.1 Rationale for protocols 11 2.2 Format of a Cochrane review 12 2.3 Logistics of doing a review 13 2.4 Publication of Cochrane reviews in print journals and books 24 2.5 Publication of previously published reviews as Cochrane reviews 26 2.6 Declaration of interest and commercial sponsorship 26 2.7 Chapter information 29 2.8 References 29 3 Maintaining reviews: updates, amendments and feedback 31 Julian PT Higgins, Sally Green and Rob JPM Scholten 3.1 Introduction 31 3.2 Some important definitions 32 CO PY RI GH TE D M AT ER IA L OTE/SPH OTE/SPH JWBK247-FM August 22, 2008 20:58 Printer Name: Yet to Come vi CONTENTS 3.3 Important dates associated with Cochrane reviews 39 3.4 Considerations when updating a Cochrane review 42 3.5 ‘What’s new’ and History tables 46 3.6 Incorporating and addressing feedback in a Cochrane review 48 3.7 Chapter information 48 3.8 References 49 4 Guide to the contents of a Cochrane protocol and review 51 Edited by Julian PT Higgins and Sally Green 4.1 Introduction 52 4.2 Title and review information (or protocol information) 52 4.3 Abstract 55 4.4 Plain language summary 55 4.5 Main text 55 4.6 Tables 70 4.7 Studies and references 72 4.8 Data and analyses 74 4.9 Figures 76 4.10 Sources of support to the review 77 4.11 Feedback 77 4.12 Appendices 78 4.13 Chapter information 78 4.14 References 78 Part 2 GENERAL METHODS FOR COCHRANE REVIEWS 81 5 Defining the review question and developing criteria for including studies 83 Edited by Denise O’Connor, Sally Green and Julian PT Higgins 5.1 Questions and eligibility criteria 84 5.2 Defining types of participants: which people and populations? 85 5.3 Defining types of interventions: which comparisons to make? 86 5.4 Defining types of outcomes: which outcome measures are most important? 87 5.5 Defining types of study 90 5.6 Defining the scope of a review question (broad versus narrow) 91 5.7 Changing review questions 93 5.8 Chapter information 93 5.9 References 94 6 Searching for studies 95 Carol Lefebvre, Eric Manheimer and Julie Glanville on behalf of the Cochrane Information Retrieval Methods Group 6.1 Introduction 96 6.2 Sources to search 98 6.3 Planning the search process 118 6.4 Designing search strategies 128 6.5 Managing references 142 OTE/SPH OTE/SPH JWBK247-FM August 22, 2008 20:58 Printer Name: Yet to Come CONTENTS vii 6.6 Documenting and reporting the search process 144 6.7 Chapter information 146 6.8 References 147 7 Selecting studies and collecting data 151 Edited by Julian PT Higgins and Jonathan J Deeks 7.1 Introduction 151 7.2 Selecting studies 152 7.3 What data to collect 156 7.4 Sources of data 163 7.5 Data collection forms 164 7.6 Extracting data from reports 167 7.7 Extracting study results and converting to the desired format 170 7.8 Managing data 182 7.9 Chapter information 183 7.10 References 183 8 Assessing risk of bias in included studies 187 Edited by Julian PT Higgins and Douglas G Altman on behalf of the Cochrane Statistical Methods Group and the Cochrane Bias Methods Group 8.1 Introduction 188 8.2 What is bias? 188 8.3 Tools for assessing quality and risk of bias 190 8.4 Introduction to sources of bias in clinical trials 193 8.5 The Cochrane Collaboration’s tool for assessing risk of bias 194 8.6 Presentation of assessments of risk of bias 202 8.7 Summary assessments of risk of bias 202 8.8 Incorporating assessments into analyses 206 8.9 Sequence generation 210 8.10 Allocation sequence concealment 214 8.11 Blinding of participants, personnel and outcome assessors 217 8.12 Incomplete outcome data 219 8.13 Selective outcome reporting 226 8.14 Other potential threats to validity 230 8.15 Chapter information 234 8.16 References 235 9 Analysing data and undertaking meta-analyses 243 Edited by Jonathan J Deeks, Julian PT Higgins and Douglas G Altman on behalf of the Cochrane Statistical Methods Group 9.1 Introduction 244 9.2 Types of data and effect measures 249 9.3 Study designs and identifying the unit of analysis 260 9.4 Summarizing effects across studies 263 9.5 Heterogeneity 276 9.6 Investigating heterogeneity 282 9.7 Sensitivity analyses 289 9.8 Chapter information 292 9.9 References 293 OTE/SPH OTE/SPH JWBK247-FM August 22, 2008 20:58 Printer Name: Yet to Come viii CONTENTS 10 Addressing reporting biases 297 Edited by Jonathan AC Sterne, Matthias Egger and David Moher on behalf of the Cochrane Bias Methods Group 10.1 Introduction 298 10.2 Types of reporting biases and the supporting evidence 299 10.3 Avoiding reporting biases 308 10.4 Detecting reporting biases 310 10.5 Chapter information 324 10.6 References 325 11 Presenting results and ‘Summary of findings’ tables 335 Holger J Schu¨nemann, Andrew D Oxman, Julian PT Higgins, Gunn E Vist, Paul Glasziou and Gordon H Guyatt on behalf of the Cochrane Applicability and Recommendations Methods Group and the Cochrane Statistical Methods Group 11.1 Introduction 335 11.2 ‘Characteristics of included studies’ tables 336 11.3 Data and analyses 337 11.4 Figures 341 11.5 ‘Summary of findings’ tables 342 11.6 Additional tables 350 11.7 Presenting results in the text 351 11.8 Writing an abstract 352 11.9 Writing a plain language summary 355 11.10 Chapter information 356 11.11 References 357 12 Interpreting results and drawing conclusions 359 Holger J Schu¨nemann, Andrew D Oxman, Gunn E Vist, Julian PT Higgins, Jonathan J Deeks, Paul Glasziou and Gordon H Guyatt on behalf of the Cochrane Applicability and Recommendations Methods Group 12.1 Introduction 360 12.2 Assessing the quality of a body of evidence 361 12.3 Issues in applicability 367 12.4 Interpreting results of statistical analyses 369 12.5 Interpreting results from dichotomous outcomes (including numbers needed to treat) 372 12.6 Interpreting results from continuous outcomes (including standardized mean differences) 377 12.7 Drawing conclusions 380 12.8 Chapter information 382 12.9 References 383 Part 3 SPECIAL TOPICS 389 13 Including non-randomized studies 391 Barnaby C Reeves, Jonathan J Deeks, Julian PT Higgins and George A Wells on behalf of the Cochrane Non-Randomised Studies Methods Group 13.1 Introduction 392 13.2 Developing criteria for including non-randomized studies 396 OTE/SPH OTE/SPH JWBK247-FM August 22, 2008 20:58 Printer Name: Yet to Come CONTENTS ix 13.3 Searching for non-randomized studies 404 13.4 Selecting studies and collecting data 407 13.5 Assessing risk of bias in non-randomized studies 412 13.6 Synthesis of data from non-randomized studies 419 13.7 Interpretation and discussion 424 13.8 Chapter information 428 13.9 References 429 14 Adverse effects 433 Yoon K Loke, Deirdre Price and Andrew Herxheimer on behalf of the Cochrane Adverse Effects Methods Group 14.1 Introduction 433 14.2 Scope of a review addressing adverse effects 434 14.3 Choosing which adverse effects to include 437 14.4 Types of studies 438 14.5 Search methods for adverse effects 439 14.6 Assessing risk of bias for adverse effects 442 14.7 Chapter information 445 14.8 References 446 15 Incorporating economics evidence 449 Ian Shemilt, Miranda Mugford, Sarah Byford, Michael Drummond, Eric Eisenstein, Martin Knapp, Jacqueline Mallender, David McDaid, Luke Vale and Damian Walker on behalf of the Campbell and Cochrane Economics Methods Group 15.1 The role and relevance of economics evidence in Cochrane reviews 449 15.2 Planning the economics component of a Cochrane review 454 15.3 Locating studies 459 15.4 Selecting studies and collecting data 462 15.5 Addressing risk of bias 463 15.6 Analysing and presenting results 468 15.7 Addressing reporting biases 472 15.8 Interpreting results 474 15.9 Conclusions 474 15.10 Chapter information 476 15.11 References 476 16 Special topics in statistics 481 Edited by Julian PT Higgins, Jonathan J Deeks and Douglas G Altman on behalf of the Cochrane Statistical Methods Group 16.1 Missing data 482 16.2 Intention-to-treat issues 488 16.3 Cluster-randomized trials 493 16.4 Cross-over trials 498 16.5 Studies with more than two intervention groups 508 16.6 Indirect comparisons and multiple-treatments meta-analysis 513 16.7 Multiplicity and the play of chance 516 16.8 Bayesian and hierarchical approaches to meta-analysis 518 16.9 Rare events (including zero frequencies) 520 16.10 Chapter information 524 16.11 References 524 OTE/SPH OTE/SPH JWBK247-FM August 22, 2008 20:58 Printer Name: Yet to Come x CONTENTS 17 Patient-reported outcomes 531 Donald L Patrick, Gordon H Guyatt and Catherine Acquadro on behalf of the Cochrane Patient Reported Outcomes Methods Group 17.1 What are patient-reported outcomes? 532 17.2 Patient-reported outcomes and Cochrane reviews 533 17.3 Health status and quality of life as PRO outcomes 534 17.4 Issues in the measurement of patient-reported outcomes 537 17.5 Locating and selecting studies with patient-reported outcomes 538 17.6 Assessing and describing patient-reported outcomes 539 17.7 Comparability of different patient-reported outcome measures 540 17.8 Interpreting results 541 17.9 Chapter information 543 17.10 References 544 18 Reviews of individual patient data 547 Lesley A Stewart, Jayne F Tierney and Mike Clarke on behalf of the Cochrane Individual Patient Data Meta-analysis Methods Group 18.1 Introduction 548 18.2 The collaborative nature of IPD meta-analyses 550 18.3 Dealing with data 551 18.4 Analysis 553 18.5 Limitations and caveats 555 18.6 Chapter information 556 18.7 References 557 19 Prospective meta-analysis 559 Davina Ghersi, Jesse Berlin and Lisa Askie on behalf of the Cochrane Prospective Meta-analysis Methods Group 19.1 Introduction 559 19.2 The collaborative nature of prospective meta-analyses 562 19.3 The prospective meta-analysis protocol 563 19.4 Data collection in prospective meta-analysis 566 19.5 Analysis issues in prospective meta-analysis 567 19.6 Chapter information 569 19.7 References 569 20 Qualitative research and Cochrane reviews 571 Jane Noyes, Jennie Popay, Alan Pearson, Karin Hannes and Andrew Booth on behalf of the Cochrane Qualitative Research Methods Group 20.1 Introduction 572 20.2 Incorporating evidence from qualitative research in Cochrane Intervention reviews: concepts and issues 572 20.3 Qualitative evidence synthesis 576 20.4 Chapter information 583 20.5 References 584 20.6 Further selected reading 587 OTE/SPH OTE/SPH JWBK247-FM August 22, 2008 20:58 Printer Name: Yet to Come CONTENTS xi 21 Reviews in public health and health promotion 593 Edited by Rebecca Armstrong, Elizabeth Waters and Jodie Doyle 21.1 Introduction 593 21.2 Study designs to include 594 21.3 Searching 594 21.4 Assessment of study quality and risk of bias 595 21.5 Ethics and inequalities 597 21.6 Context 599 21.7 Sustainability 600 21.8 Applicability and transferability 601 21.9 Chapter information 603 21.10 References 603 22 Overviews of reviews 607 Lorne A Becker and Andrew D Oxman 22.1 Introduction 607 22.2 Preparing a Cochrane Overview of reviews 608 22.3 Format of a Cochrane Overview 613 22.4 Chapter information 631 22.5 References 631 Index 633 OTE/SPH OTE/SPH JWBK247-FM August 22, 2008 20:58 Printer Name: Yet to Come OTE/SPH OTE/SPH JWBK247-ind August 25, 2008 19:58 Printer Name: Yet to Come Index absolute risk reduction (ARR) see risk difference abstracts 55, 352–5 adverse effects 441–2 non-randomized studies 406 Overviews of reviews 615–16 searching 98, 110, 143, 145, 406, 441–2 selecting studies 153 accumulation of changes 38 acknowledgements 68, 624 ACR see assumed control risks Additional tables 72, 234, 350–1, 468, 539 adherence 160, 369 adverse effects 433–48 data collection 161–2 definitions 434 drop-outs 438 eligibility criteria 434–8 narrow/broad focus 437–8 non-randomized studies 397, 400, 427, 438, 443–5 review questions 90 risk of bias/quality 442–5 searching 439–42 separate reviews 436–7 sources of information 439–40 strategies for addressing 434–7 Summary of findings tables 348–9 types of studies 438, 442–5 Adverse Effects Methods Group (AEMG) 446 advisory groups 19–20 AEMG see Adverse Effects Methods Group aetiological research questions 399 agreement measures 155–6, 169 allied health-specific databases 103–4 allocation concealment 193, 211, 214–17 results sections 65 Risk of bias tool 196, 198–9, 203–4 amendments 33, 34–7 analysing data see meta-analysis analysis of covariance (ANCOVA) 270 analysis plans 58, 60–3, 244–6, 564, 565 prospective meta-analyses 564, 565 ANCOVA see analysis of covariance appendices 78 search strategies 144, 146 applicability discussion sections 66 interpreting results 367–9 Overviews of reviews 622–3 public health and health promotion 601–3 Applicability and Recommendations Methods Group (ARMG) 383 ARR (absolute risk reduction) see risk difference as-treated analyses 225 assessing a review as up to date 40, 43–4, 54, 625–6 assumed control risks (ACR) 347, 349, 374, 376–7 attrition bias 194–6, 200, 219 authors’ conclusions see conclusions authorship 36, 38, 39, 52–3 available case analysis 489, 490, 492 background sections 56–7, 617–18 baseline imbalance 231–2 baseline risk see control group risk Bayesian methods 210, 518–19 before-and-after studies 393, 596–7 bias see publication bias; reporting bias; risk of bias Bias Methods Group (BMG) 324 bibliographic databases 98–107 CO PY RI GH TE D M AT ER IA L OTE/SPH OTE/SPH JWBK247-ind August 25, 2008 19:58 Printer Name: Yet to Come 634 INDEX binary outcome data see dichotomous outcome data biologic variation 368 biology-specific databases 103 blinding 193–4, 217–9 results sections 65 review authors 168, 192, 382 Risk of bias tool 195–196, 199, 203–4, 217–19 blocked randomization 194, 212, 232 BMG see Bias Methods Group Boolean operators 132–3, 141 broad questions 91–3 Campbell and Cochrane Economics Methods Group (CCEMG) 454, 463, 468, 475, 543 Campbell Collaboration 147 Campbell Coordinating Groups 584 Campbell Equity Methods Group 597–8 carry-over effects 499–503 case reports 438, 444–5 case series 393, 438 case-control studies 393, 399, 415–16, 438, 443–4 CBA see cost-benefit analysis CBT see cognitive behavioural therapy CCEMG see Campbell and Cochrane Economics Methods Group CCTs see controlled clinical trials CDSR see Cochrane Database of Systematic Reviews CEA see cost-effectiveness analysis CENTRAL see Cochrane Central Register of Controlled Trials central randomization 216 Centre for Reviews and Dissemination (CRD) 111, 460–2, 463 CGR see control group risk change scores 172, 269–70 standard deviations 485–8 changing review questions 93 Characteristics of studies tables 63–4, 70–2 economics evidence 468 excluded studies 64, 71, 154 included studies 63–4, 70, 336–7, 468 ongoing studies 64, 72 Overviews of reviews 620, 625–6 studies awaiting classification 64, 71 checklists 191 chi-squared tests 277–8, 281, 284, 338–9 CINAHL database 103, 579 citation bias 298, 306 indexes 105 versions 33–6, 37–9, 55 citing protocols 58 reviews 25–6 clinical event pathways 455–6 clinical trials see randomized trials cluster-allocated studies 409–11 see also cluster-randomized trials clustering 498 cluster-randomized trials 260, 493–8 effective sample sizes 495–7 inflating standard errors 497 meta-analysis 495–6, 497 methods of analysis 495 public health and health promotion 594, 595, 599 risk of bias 494–5 sensitivity analyses 291 CMR see Cochrane Methodology Register Cochrane, Archie 4, 450 Cochrane Central Register of Controlled Trials (CENTRAL) 5, 98–100, 108, 118–29, 133, 139–42, 406 Cochrane Collaboration 3–6 maintenance of Cochrane reviews 32 preparation of Cochrane reviews 11, 18, 24–8 principles 4, 8 publication of Cochrane reviews 5–6 structure 4–5 Cochrane Database of Systematic Reviews (CDSR) 5–6 citation versions 33–6 citing 25–6 dates 40, 54–5 Overviews of reviews 608 prospective meta-analyses 566 protocols 12–13, 17 searching see Cochrane Library withdrawn reviews 46 Cochrane Health Equity Field 597–8 Cochrane Library 5 Cochrane Review Groups 22–4 economics evidence 460–1 feedback 48 Overviews of reviews 608 searching 99–100, 119, 121, 124–6, 133–4 Cochrane Methodology Register (CMR) 5 OTE/SPH OTE/SPH JWBK247-ind August 25, 2008 19:58 Printer Name: Yet to Come INDEX 635 Cochrane Review Groups (CRGs) 4–5, 7 economics evidence 453–4 guidance and support 17, 20–4, 27, 59, 68, 91 maintenance of Cochrane reviews 32, 35, 37–8, 41, 45–6, 48 non-randomized studies 398, 399–400, 403, 406–7 Overviews of reviews 607–8, 613 plain language summaries 355–6 prospective meta-analyses 566 public health and health promotion 595 qualitative research 576, 583–4 searching 59–60, 96–8, 102, 108, 118–19, 123–5, 130, 144 statisticians 292 Cochrane reviews Diagnostic test accuracy 7, 20 Intervention 6–7, 12–16 Methodology 20 Overviews of reviews 20, 607–31 coding schemes 167 cognitive behavioural therapy (CBT) 536 cohort studies 393, 399, 415, 438, 443–4 collaborations individual patient data 549, 550–1 prospective meta-analyses 562–3 see also Cochrane Collaboration combining studies see meta-analysis combining subgroups/intervention groups 177 comments and criticisms see feedback commercial interests 26–8, 303–4, 624 community healthcare-specific databases 104 comparisons 74–5, 86–7, 245, 247–8 completeness of evidence 66 concealment of allocation see allocation concealment conclusions 67, 380–2 abstract 353, 355 changed 34, 37–8 economics evidence 474 Overviews of reviews 616, 623–4 qualitative research 578 conditional outcomes 492–3 conference abstracts/proceedings 110–11 confidence intervals 369–71 continuous outcome data 172–5, 370 data extraction 173–5, 181–2 dichotomous outcome data 171, 370 forest plots 338 rare events 522 Summary of findings tables 349–50 confidentiality 550–1, 563 conflicts of interest see declarations of interest confounding GRADE/quality of evidence 362, 366 heterogeneity 287 indirect comparisons 514 non-randomized studies 395, 397, 412, 416–17, 420–2, 425–6 public health and health promotion 594 consensus 169 consistency see heterogeneity consumer involvement 18–19 contact persons 53–4 contents of Cochrane reviews 12–16, 51–79 abstracts 55 appendices 78 data and analyses 74–5 feedback 77 figures 76–7 plain language summaries 55 protocol information 52–5 review information 52–5 studies and references 72–4 tables 70–2 text 55–70 titles 52–3 context 368–9, 599–600 continuous outcome data 250, 255–7 confidence intervals 172–5, 370 counts 179 data extraction 171–7, 179 forest plots 338–9 funnel plots 317–18 individual patient data 554, 555 intention to treat 491–2 interpreting results 377–9 meta-analysis 268–72, 290 non-randomized studies 421 Overviews of reviews 629 results sections 338–9, 349, 370, 377–9 risk of bias 200, 223–4, 227 Summary of findings tables 349 see also patient-reported outcomes contour-enhanced funnel plots 313–15 contributions of authors 68–9 control group risk (CGR) effect measures 251–3 interpreting results 346–7, 349, 374, 377 meta-regression 288–9 see also assumed control risk OTE/SPH OTE/SPH JWBK247-ind August 25, 2008 19:58 Printer Name: Yet to Come 636 INDEX controlled before-and-after studies 393 vocabulary 130–1, 440, 538–9 controlled clinical trial (CCT) 119–23, 126 copyright 97–8 correlation coefficients 486–8, 496–8, 505–7 correspondence with investigators 163, 166, 193 cost data 470–1 cost-benefit a