Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2020 Jun;183(3):1189-1210.
doi: 10.1111/rssa.12579. Epub 2020 Jun 7.

Multilevel network meta-regression for population-adjusted treatment comparisons

David M Phillippo  1 , Sofia Dias  2 , A E Ades  1 , Mark Belger  3 , Alan Brnabic  4 , Alexander Schacht  5 , Daniel Saure  5 , Zbigniew Kadziola  6 , Nicky J Welton  1
Affiliations

Multilevel network meta-regression for population-adjusted treatment comparisons

David M Phillippo et al. J R Stat Soc Ser A Stat Soc. 2020 Jun.

Abstract

Standard network meta-analysis (NMA) and indirect comparisons combine aggregate data from multiple studies on treatments of interest, assuming that any effect modifiers are balanced across populations. Population adjustment methods relax this assumption using individual patient data from one or more studies. However, current matching-adjusted indirect comparison and simulated treatment comparison methods are limited to pairwise indirect comparisons and cannot predict into a specified target population. Existing meta-regression approaches incur aggregation bias. We propose a new method extending the standard NMA framework. An individual level regression model is defined, and aggregate data are fitted by integrating over the covariate distribution to form the likelihood. Motivated by the complexity of the closed form integration, we propose a general numerical approach using quasi-Monte-Carlo integration. Covariate correlation structures are accounted for by using copulas. Crucially for decision making, comparisons may be provided in any target population with a given covariate distribution. We illustrate the method with a network of plaque psoriasis treatments. Estimated population-average treatment effects are similar across study populations, as differences in the distributions of effect modifiers are small. A better fit is achieved than a random effects NMA, uncertainty is substantially reduced by explaining within- and between-study variation, and estimates are more interpretable.

Keywords: Effect modification; Indirect comparison; Individual patient data; Network meta‐analysis.

PubMed Disclaimer

Figures

Figure 1
Figure 1
Simple scenario with one 1 versus 2 and one 1 versus 3 study: individual patient data are available for the 1 versus 2 study (formula image); only aggregate data are available for the 1 versus 3 study (formula image); an indirect comparison compares treatments 2 and 3 via the common 1 arm
Figure 2
Figure 2
The UNCOVER (Griffiths et al., 2015; Gordon et al., 2016) and FIXTURE (Langley et al., 2014) trials form a network of six treatments: shading indicates comparisons made in a the UNCOVER‐2 and UNCOVER‐3 trials, b the UNCOVER‐1 trial and c the FIXTURE trial (formula image, availability of IPD on a comparison; formula image, availability of aggregate data on a comparison; PBO, placebo; IXE, ixekizumab; SEC, secukinumab; ETN, etanercept); IXE and SEC were each investigated with two different dosing regimens; the MAIC analysis included only information from the IXE Q2W, SEC 300 and ETN arms, whereas ML‐NMR makes use of all available information
Figure 3
Figure 3
Estimated proportion of individuals achieving PASI 75 on each treatment, in each study population (the MAIC estimate is produced in the FIXTURE study population, and the corresponding interval is a 95% confidence interval as MAIC is a frequentist method) (formula image, ML‐NMR; formula image, MAIC; formula image, observed): (a) FIXTURE; (b) UNCOVER‐1; (c) UNCOVER‐2; (d) UNCOVER‐3
See this image and copyright information in PMC

Similar articles

See all similar articles

Cited by

See all "Cited by" articles

References

    1. Ades, A. E. (2003) A chain of evidence with mixed comparisons: models for multi‐parameter synthesis and consistency of evidence. Statist. Med., 22, 2995–3016. - PubMed
    1. Berlin, J. A. , Santanna, J. , Schmid, C. H. , Szczech, L. A. and Feldman, H. I. (2002) Individual patient‐ versus group‐level data meta‐regressions for the investigation of treatment effect modifiers: ecological bias rears its ugly head. Statist. Med., 21, 371–387. - PubMed
    1. Bucher, H. C. , Guyatt, G. H. , Griffith, L. E. and Walter, S. D. (1997) The results of direct and indirect treatment comparisons in meta‐analysis of randomized controlled trials. J. Clin. Epidem., 50, 683–691. - PubMed
    1. Caflisch, R. E. (1998) Monte Carlo and quasi‐Monte Carlo methods. Acta Numer., 7, 1–49.
    1. Caldwell, D. M. , Ades, A. E. and Higgins, J. P. T. (2005) Simultaneous comparison of multiple treatments: combining direct and indirect evidence. Br. Med. J., 331, 897–900. - PMC - PubMed

LinkOut - more resources