Network meta-analysis of (individual patient) time to event data alongside (aggregate) count data

BMC Medical Research Methodology

Marta O Soares²  Ling-Hsiang Chuang¹  Pedro Saramago² 

[1]Pharmerit International, Rotterdam, The Netherlands

[2]Centre for Health Economics, University of York, Heslington, York YO10 5DD, UK

关键词: Treatment-effect modifiers;    Network-meta-analysis;    Mixed treatment comparisons;    Meta-analysis;    Aggregate data;    Individual patient data;    Survival data;    Time to event data;    Evidence synthesis;   

Others  :  1091041
DOI  :  10.1186/1471-2288-14-105

 received in 2014-03-24, accepted in 2014-08-12,  发布年份 2014

【 摘 要 】

Background

Network meta-analysis methods extend the standard pair-wise framework to allow simultaneous comparison of multiple interventions in a single statistical model. Despite published work on network meta-analysis mainly focussing on the synthesis of aggregate data, methods have been developed that allow the use of individual patient-level data specifically when outcomes are dichotomous or continuous. This paper focuses on the synthesis of individual patient-level and summary time to event data, motivated by a real data example looking at the effectiveness of high compression treatments on the healing of venous leg ulcers.

Methods

This paper introduces a novel network meta-analysis modelling approach that allows individual patient-level (time to event with censoring) and summary-level data (event count for a given follow-up time) to be synthesised jointly by assuming an underlying, common, distribution of time to healing. Alternative model assumptions were tested within the motivating example. Model fit and adequacy measures were used to compare and select models.

Results

Due to the availability of individual patient-level data in our example we were able to use a Weibull distribution to describe time to healing; otherwise, we would have been limited to specifying a uniparametric distribution. Absolute effectiveness estimates were more sensitive than relative effectiveness estimates to a range of alternative specifications for the model.

Conclusions

The synthesis of time to event data considering individual patient-level data provides modelling flexibility, and can be particularly important when absolute effectiveness estimates, and not just relative effect estimates, are of interest.

【 授权许可】

   
2014 Saramago et al.; licensee BioMed Central Ltd.

【 图 表 】

Figure 1.

Figure 2.

【 参考文献 】

• [1]Bucher H, Guyatt G, Griffith L, Walter S: The results of direct and indirect treatment comparisons in meta-analysis of randomized controlled trials. J Clin Epidemiol 1997, 50(6):683-691.
• [2]Sutton A, Ades A, Cooper N, Abrams K: Use of indirect and mixed treatment comparisons for technology assessment. Pharmacoeconomics 2008, 26(9):753-767.
• [3]Dias S, Welton NJ, Sutton AJ, Ades AE: Evidence synthesis for decision making 1: introduction. Med Decis Making 2013, 33(5):597-606.
• [4]Saramago P, Manca A, Sutton AJ: Deriving input parameters for cost-effectiveness modelling: taxonomy of data types and approaches to their statistical synthesis. Value Health 2012, 15(5):639-649.
• [5]Dias S, Sutton AJ, Ades AE, Welton NJ: Evidence synthesis for decision making 2: a generalized linear modeling framework for pairwise and network meta-analysis of randomized controlled trials. Med Decis Making 2013, 33(5):607-617.
• [6]Saramago P, Sutton AJ, Cooper NJ, Manca A: Mixed treatment comparisons using aggregate- and individual-participant level data. Stat Med 2012, 10; 31(28):3516-3536.
• [7]Jansen J, Cope S: Network meta-analysis of individual and aggregate level data. Value Health 2012, 15(4):A159.
• [8]Donegan S, Williamson P, D’Alessandro U, Garner P, Smith C: Combining individual patient data and aggregate data in mixed treatment comparison meta-analysis: individual patient data may be beneficial if only for a subset of trials. Stat Med 2013, 32(6):914-930.
• [9]Berlin J, Santanna J, Schmid C, Szczech L, Feldman H: Individual patient- versus group-level data meta-regressions for the investigation of treatment effect modifiers: ecological bias rears its ugly head. Stat Med 2002, 21(3):371-387.
• [10]Riley R, Lambert P, Staessen J, Wang J, Gueyffier F, Thijs L, Boutitie F: Meta-analysis of continuous outcomes combining individual patient data and aggregate data. Stat Med 2008, 27(11):1870-1893.
• [11]Sutton AJ: Methods for Meta-Analysis in Medical Research. Chichester ; New York: J. Wiley; 2000.
• [12]Welton NJ, Willis SR, Ades AE: Synthesis of survival and disease progression outcomes for health technology assessment of cancer therapies. Res Synth Method 2010, 1:239-257. doi: 10.1002/jrsm.21
• [13]Smith C, Williamson P, Marson A: An overview of methods and empirical comparison of aggregate data and individual patient data results for investigating heterogeneity in meta-analysis of time-to-event outcomes. J Eval Clin Pract 2005, 11(5):468-478.
• [14]Woods B, Hawkins N, Scott D: Network meta-analysis on the log-hazard scale, combining count and hazard ratio statistics accounting for multi-arm trials: a tutorial. BMC Med Res Methodol 2010, 10:54.
• [15]Soares MO, Dumville JC, Ades AE, Welton NJ: Treatment comparisons for decision making: facing the problems of sparse and few data. J R Stat Soc A Stat Soc 2014, 177:259-279. doi: 10.1111/rssa.12010
• [16]Sutton A, Kendrick D, Coupland C: Meta-analysis of individual- and aggregate-level data. Stat Med 2008, 27(5):651-669.
• [17]Ashby RL, Gabe R, Ali S, Saramago P, Chuang LH, Adderley U, Bland JM, Cullum NA, Dumville JC, Iglesias CP, Soares MO, Stubbs NC, Torgerson DJ: VenUS IV (venous leg ulcer study IV): a randomised controlled trial of compression hosiery versus compression bandaging in the treatment of venous leg ulcers. Health Technology Assessment (forthcoming) 2014.
• [18]Duby T, Hoffman D, Cameron J, Doblhoffbrown D, Cherry G, Ryan T: A randomized trial in the treatment of venous Leg ulcers comparing short stretch bandages, 4 layer bandage system, and a long stretch paste bandage system. Wounds-a Compendium Clin Res Pract 1993, 5(6):276-279.
• [19]Scriven J, Taylor L, Wood Rgn A, Bell P, Naylor A, London N: A prospective randomised trial of four-layer versus short stretch compression bandages for the treatment of venous leg ulcers. Ann R Coll Surg Engl 1998, 80(3):215-220.
• [20]Partsch H, Damstra R, Tazelaar D, Schuller-Petrovic S, Velders A, De Rooij M, Tjon Lim Sang R, Quinlan D: Multicentre, randomised controlled trial of four-layer bandaging versus short-stretch bandaging in the treatment of venous leg ulcers. Vasa 2001, 30(2):108-113.
• [21]Ukat A, Konig M, Vanscheidt W, Munter K: Short-stretch versus multilayer compression for venous leg ulcers: a comparison of healing rates. J Wound Care 2003, 12(4):139-143.
• [22]Franks P, Moody M, Moffatt C, Martin R, Blewett R, Seymour E, Hildreth A, Hourican C, Collins J, Heron A: Randomized trial of cohesive short-stretch versus four-layer bandaging in the management of venous ulceration. Wound Repair Regen 2004, 12(2):157-162.
• [23]Junger M, Wollina U, Kohnen R, Rabe E: Efficacy and tolerability of an ulcer compression stocking for therapy of chronic venous ulcer compared with a below-knee compression bandage: results from a prospective, randomized, multicentre trial. Curr Med Res Opin 2004, 20(10):1613-1623.
• [24]Kralj B, Kosicek M: Randomised Comparative Trial of Single-Layer and Multi-Layer Bandages in the Treatment of Venous leg Ulcer. Harrogate, UK: 6th European Conference on Advances in Wound Management: 1995; 1995:158-160.
• [25]Polignano R, Bonadeo P, Gasbarro S, Allegra C: A randomised controlled study of four-layer compression versus Unna’s boot for venous ulcers. J Wound Care 2004, 13(1):21-24.
• [26]Wilkinson E, Buttfield S, Cooper S, Young E: Trial of two bandaging systems for chronic venous leg ulcers. J Wound Care 1997, 6(7):339-340.
• [27]Colgan MP, Teevan M, McBride C, O’Sullivan L, Moore D, Shanik G: Cost Comparisons in the Management of Venous Ulceration. Harrogate, UK: 5th European Conference on Advances in Wound Management: 21–24 November 1995; 1995.
• [28]Blecken S, Villavicencio J, Kao T: Comparison of elastic versus nonelastic compression in bilateral venous ulcers: a randomized trial. J Vasc Surg 2005, 42(6):1150-1155.
• [29]Moffatt C, Edwards L, Collier M, Treadwell T, Miller M, Shafer L, Sibbald G, Brassard A, McIntosh A, Reyzelman A, Price P, Kraus SM, Walters SA, Harding K: A randomised controlled 8-week crossover clinical evaluation of the 3M coban 2 layer compression system versus profore to evaluate the product performance in patients with venous leg ulcers. Int Wound J 2008, 5(2):267-279.
• [30]Szewczyk M, Jawie A, Cierzniakowska K, Cwajda-Bialasik J, Moscicka P: Comparison of the effectiveness of compression stockings and layer compression systems in venous ulceration treatment. Archives of Medical Science 2010, 6(5):793-799.
• [31]Wong I, Andriessen A, Charles H, Thompson D, Lee D, So W, Abel M: Randomized controlled trial comparing treatment outcome of two compression bandaging systems and standard care without compression in patients with venous leg ulcers. J Eur Acad Dermatol Venereol 2012, 26(1):102-110.
• [32]Iglesias C, Nelson E, Cullum N, Torgerson D: VenUS I: a randomised controlled trial of two types of bandage for treating venous leg ulcers. Health Technol Assess 2004, 8(29):iii, 1-iii, 105.
• [33]Collett D: Modelling Survival Data in Medical Research. 2nd edition. Boca Raton, Fla: Chapman & Hall/CRC; 2003.
• [34]Cooper N, Sutton A, Morris D, Ades A, Welton N: Addressing between-study heterogeneity and inconsistency in mixed treatment comparisons: application to stroke prevention treatments in individuals with non-rheumatic atrial fibrillation. Stat Med 2009, 28(14):1861-1881.
• [35]Lunn DJ, Thomas A, Best N, Spiegelhalter D: WinBUGS - a Bayesian modelling framework: concepts, structure, and extensibility. Statistics and Computing 2000, 10(4):325-337.
• [36]Spiegelhalter DJ, Best NG, Carlin BR, van der Linde A: Bayesian measures of model complexity and fit. J R Stat Soc B Stat Soc 2002, 64:583-616.
• [37]Ades A, Sculpher M, Sutton A, Abrams K, Cooper N, Welton N, Lu G: Bayesian methods for evidence synthesis in cost-effectiveness analysis. Pharmacoeconomics 2006, 24(1):1-19.
• [38]StataCorp A: Stata Statistical Software: Release 12. College Station, TX: StataCorp LP; 2011.
• [39]Akaike H: Information Theory and the Maximum Likelihood Principle. 2nd International Symposium in Information Theory: 1973; 1973:267-281.
• [40]O’Meara S, Tierney J, Cullum N, Bland J, Franks P, Mole T, Scriven M: Four layer bandage compared with short stretch bandage for venous leg ulcers: systematic review and meta-analysis of randomised controlled trials with data from individual patients. BMJ (Clinical research ed) 2009, 338(7702):1054-1057.
• [41]Guyot P, Ades AE, Ouwens MJ, Welton NJ: Enhanced secondary analysis of survival data: reconstructing the data from published Kaplan-Meier survival curves. BMC Med Res Methodol 2012, 12:9.
• [42]Espinoza M: Heterogeneity in cost-effectiveness analysis: methods to explore the value of subgroups and individualised care in a collectively funded health system. University of York; 2012.
• [43]Medical Research Council Data Sharing http://www.mrc.ac.uk/research/research-policy-ethics/data-sharing/ webcite (accessed in September 2014)
• [44]National Institute for Health Research - Health Technology Assessment Data Sharing http://www.journalslibrary.nihr.ac.uk/information-for-authors/data-sharing webcite (accessed in September 2014)