Reporting, handling and assessing the risk of bias associated with missing participant data in systematic reviews: a methodological survey

被引：22

作者：

Akl, Elie A. ^{[1
,2
]}

Carrasco-Labra, Alonso ^{[2
,3
]}

Brignardello-Petersen, Romina ^{[2
,3
,4
]}

Neumann, Ignacio ^{[5
]}

Johnston, Bradley C. ^{[4
,6
]}

Sun, Xin ^{[2
,7
]}

Briel, Matthias ^{[2
,8
]}

Busse, Jason W. ^{[2
,9
,10
]}

Ebrahim, Shanil ^{[2
,9
,11
]}

Granados, Carlos E. ^{[12
]}

Iorio, Alfonso ^{[2
,13
]}

Irfan, Affan ^{[14
]}

Martinez Garcia, Laura ^{[15
]}

Mustafa, Reem A. ^{[2
,16
,17
,18
]}

Ramirez-Morera, Anggie ^{[19
]}

Selva, Anna ^{[15
]}

Sola, Ivan ^{[15
]}

Sanabria, Andrea Juliana ^{[15
]}

Tikkinen, Kari A. O. ^{[2
,20
,21
,22
]}

Vandvik, Per O. ^{[23
,24
]}

Vernooij, Robin W. M.

Zazueta, Oscar E. ^{[15
]}

Zhou, Qi ^{[2
]}

Guyatt, Gordon H. ^{[2
]}

Alonso-Coello, Pablo ^{[15
]}

机构：

[1] Amer Univ Beirut, Dept Internal Med, Beirut, Lebanon

[2] McMaster Univ, Dept Clin Epidemiol & Biostat, Hamilton, ON, Canada

[3] Univ Chile, Fac Dent, Evidence Based Dent Unit, Santiago, Chile

[4] Univ Toronto, Inst Hlth Policy Management & Evaluat, Toronto, ON, Canada

[5] Pontificia Univ Catolica Chile, Dept Med, Santiago, Chile

[6] Hosp Sick Children, Dept Anesthesia & Pain Med, Toronto, ON M5G 1X8, Canada

[7] Sichuan Univ, West China Hosp, Chinese Evidence Based Med Ctr, Clin Res & Evaluat Unit, Chengdu 610064, Peoples R China

[8] Univ Basel Hosp, Basel Inst Clin Epidemiol & Biostat, Dept Clin Res, CH-4031 Basel, Switzerland

[9] McMaster Univ, Dept Anesthesia, Hamilton, ON, Canada

[10] McMaster Univ, Michael G DeGroote Inst Pain Res & Care, Hamilton, ON, Canada

[11] Stanford Univ, Dept Med, Stanford Prevent Res Ctr, Palo Alto, CA USA

[12] Univ La Sabana, Fac Med, Area Invest, Chia, Colombia

[13] McMaster Univ, Dept Med, Hamilton, ON, Canada

[14] Univ Louisville, Dept Cardiol, Louisville, KY 40292 USA

[15] Biomed Res Inst St Pau CIBER Epidemiol & Publ Hlt, Iberoamer Cochrane Ctr, Barcelona, Spain

[16] Univ Missouri, Dept Med, Kansas City, MO 64110 USA

[17] Univ Missouri, Dept Nephrol, Kansas City, MO 64110 USA

[18] Univ Missouri, Dept Biomed & Hlth Informat, Kansas City, MO 64110 USA

[19] IHCAI Fdn & Cent Amer Cochrane, Hlth Care Dev Div, San Jose, Costa Rica

[20] Univ Helsinki, Cent Hosp, Dept Urol, Helsinki, Finland

[21] Univ Helsinki, Cent Hosp, Dept Publ Hlth, Helsinki, Finland

[22] Univ Helsinki, Helsinki, Finland

[23] Norwegian Knowledge Ctr Hlth Serv, Oslo, Norway

[24] Innlandet Hosp Trust, Dept Med, Gjovik, Norway

来源：

BMJ OPEN | 2015年 / 5卷 / 09期

基金：

加拿大健康研究院;

关键词：

ADDRESSING CONTINUOUS DATA; TRIAL ANALYSIS; METAANALYSIS; UNCERTAINTY; GUIDE;

D O I：

10.1136/bmjopen-2015-009368

中图分类号：

R5 [内科学];

学科分类号：

1002 ; 100201 ;

摘要：

Objectives: To describe how systematic reviewers are reporting missing data for dichotomous outcomes, handling them in the analysis and assessing the risk of associated bias. Methods: We searched MEDLINE and the Cochrane Database of Systematic Reviews for systematic reviews of randomised trials published in 2010, and reporting a meta-analysis of a dichotomous outcome. We randomly selected 98 Cochrane and 104 non-Cochrane systematic reviews. Teams of 2 reviewers selected eligible studies and abstracted data independently and in duplicate using standardised, piloted forms with accompanying instructions. We conducted regression analyses to explore factors associated with using complete case analysis and with judging the risk of bias associated with missing participant data. Results: Of Cochrane and non-Cochrane reviews, 47% and 7% (p<0.0001), respectively, reported on the number of participants with missing data, and 41% and 9% reported a plan for handling missing categorical data. The 2 most reported approaches for handling missing data were complete case analysis (8.5%, out of the 202 reviews) and assuming no participants with missing data had the event (4%). The use of complete case analysis was associated only with Cochrane reviews (relative to non-Cochrane: OR=7.25; 95% CI 1.58 to 33.3, p=0.01). 65% of reviews assessed risk of bias associated with missing data; this was associated with Cochrane reviews (relative to non-Cochrane: OR=6.63; 95% CI 2.50 to 17.57, p=0.0001), and the use of the Grading of Recommendations Assessment, Development and Evaluation (GRADE) methodology (OR=5.02; 95% CI 1.02 to 24.75, p=0.047). Conclusions: Though Cochrane reviews are somewhat less problematic, most Cochrane and non-Cochrane systematic reviews fail to adequately report and handle missing data, potentially resulting in misleading judgements regarding risk of bias.

引用

页数：8

共 21 条

[11] GRADE guidelines: 4. Rating the quality of evidence-study limitations (risk of bias) [J].

Guyatt, Gordon H. ;

Oxman, Andrew D. ;

Vist, Gunn ;

Kunz, Regina ;

Brozek, Jan ;

Alonso-Coello, Pablo ;

Montori, Victor ;

Akl, Elie A. ;

Djulbegovic, Ben ;

Falck-Ytter, Yngve ;

Norris, Susan L. ;

Williams, John W., Jr. ;

Atkins, David ;

Meerpohl, Joerg ;

Schuenemann, Holger J. .

JOURNAL OF CLINICAL EPIDEMIOLOGY, 2011, 64 (04) :407-415

[12] Imputation methods for missing outcome data in meta-analysis of clinical trials [J].

Higgins, Julian P. T. ;

White, Ian R. ;

Wood, Angela M. .

CLINICAL TRIALS, 2008, 5 (03) :225-239

[13] Incorporation of assessments of risk of bias of primary studies in systematic reviews of randomised trials: a cross-sectional study [J].

Hopewell, Sally ;

Boutron, Isabelle ;

Altman, Douglas G. ;

Ravaud, Philippe .

BMJ OPEN, 2013, 3 (08)

[14]

Moher D, 2009, ANN INTERN MED, V151, P264, DOI [10.7326/0003-4819-151-4-200908180-00135, 10.1136/bmj.b2700, 10.1371/journal.pmed.1000097, 10.1136/bmj.i4086, 10.1016/j.ijsu.2010.02.007, 10.1016/j.ijsu.2010.07.299, 10.1136/bmj.b2535, 10.1186/2046-4053-4-1]

[15] Cochrane reviews used more rigorous methods than non-Cochrane reviews: survey of systematic reviews in physiotherapy [J].

Moseley, Anne M. ;

Elkins, Mark R. ;

Herbert, Robert D. ;

Maher, Christopher G. ;

Sherrington, Catherine .

JOURNAL OF CLINICAL EPIDEMIOLOGY, 2009, 62 (10) :1021-1030

[16]

Schulz KF, 2010, J CLIN EPIDEMIOL, V63, P834, DOI [10.1016/j.jclinepi.2010.02.005, 10.1136/bmj.c332, 10.1016/j.ijsu.2011.09.004, 10.4103/0976-500X.72352, 10.1136/bmj.c869, 10.1186/1741-7015-8-18]

[17] AMSTAR is a reliable and valid measurement tool to assess the methodological quality of systematic reviews [J].

Shea, Beverley J. ;

Hamel, Candyce ;

Wells, George A. ;

Bouter, Lex M. ;

Kristjansson, Elizabeth ;

Grimshaw, Jeremy ;

Henry, David A. ;

Boers, Maarten .

JOURNAL OF CLINICAL EPIDEMIOLOGY, 2009, 62 (10) :1013-1020

[18]

The Cochrane Collarobation, 2011, COCHR HDB SYST REV I

[19] Allowing for uncertainty due to missing data in meta-analysis - Part 1: Two-stage methods [J].

White, Ian R. ;

Higgins, Julian P. T. ;

Wood, Angela M. .

STATISTICS IN MEDICINE, 2008, 27 (05) :711-727

[20] Allowing for uncertainty due to missing data in meta-analysis - Part 2: Hierarchical models [J].

White, Ian R. ;

Welton, Nicky J. ;

Wood, Angela M. ;

Ades, A. E. ;

Higgins, Julian P. T. .

STATISTICS IN MEDICINE, 2008, 27 (05) :728-745

← 1 2 3 →