Model selection for component network meta-analysis in connected and disconnected networks: a simulation study

BMC Medical Research Methodology

Research

Stephanie Weibel¹  Peter Kranke¹  Gerta Rücker²  Maria Petropoulou²  Guido Schwarzer² 

[1] Department of Anaesthesiology, Intensive Care, Emergency and Pain Medicine, University Hospital Würzburg, Oberdürrbacher Straße 6, 97080, Würzburg, Germany;Institute of Medical Biometry and Statistics, Faculty of Medicine and Medical Center – University of Freiburg, Stefan-Meier-Straße 26, 79104, Freiburg, Germany;

关键词: Component network meta-analysis;    Disconnected networks;    Model selection;    Multicomponent interventions;    Simulation;   

DOI  :  10.1186/s12874-023-01959-9

 received in 2022-10-11, accepted in 2023-05-29,  发布年份 2023

来源: Springer

【 摘 要 】

BackgroundNetwork meta-analysis (NMA) allows estimating and ranking the effects of several interventions for a clinical condition. Component network meta-analysis (CNMA) is an extension of NMA which considers the individual components of multicomponent interventions. CNMA allows to “reconnect” a disconnected network with common components in subnetworks. An additive CNMA assumes that component effects are additive. This assumption can be relaxed by including interaction terms in the CNMA.MethodsWe evaluate a forward model selection strategy for component network meta-analysis to relax the additivity assumption that can be used in connected or disconnected networks. In addition, we describe a procedure to create disconnected networks in order to evaluate the properties of the model selection in connected and disconnected networks. We apply the methods to simulated data and a Cochrane review on interventions for postoperative nausea and vomiting in adults after general anaesthesia. Model performance is compared using average mean squared errors and coverage probabilities.ResultsCNMA models provide good performance for connected networks and can be an alternative to standard NMA if additivity holds. For disconnected networks, we recommend to use additive CNMA only if strong clinical arguments for additivity exist.ConclusionsCNMA methods are feasible for connected networks but questionable for disconnected networks.

【 授权许可】

CC BY   
© The Author(s) 2023

【 图 表 】

Fig. 1

Fig. 8

Fig. 4

Fig. 2

Fig. 3

Fig. 1

Fig. 3

Fig. 2

Fig. 3

Fig. 6

Fig. 1

Fig. 5

Fig. 7

Fig. 6

Fig. 6

Fig. 4

13690_2023_1130_Article_IEq62.gif

Fig. 1

Fig. 2

13690_2023_1130_Article_IEq40.gif

13690_2023_1130_Article_IEq39.gif

Fig. 2

13690_2023_1130_Article_IEq36.gif

13690_2023_1130_Article_IEq35.gif

13690_2023_1130_Article_IEq34.gif

13690_2023_1130_Article_IEq33.gif

13690_2023_1130_Article_IEq32.gif

Fig. 1

Fig. 1

13690_2023_1130_Article_IEq29.gif

13690_2023_1130_Article_IEq28.gif

13690_2023_1130_Article_IEq27.gif

13690_2023_1130_Article_IEq26.gif

13690_2023_1130_Article_IEq22.gif

Fig. 2

13690_2023_1130_Article_IEq20.gif

13690_2023_1130_Article_IEq17.gif

13690_2023_1130_Article_IEq16.gif

13690_2023_1130_Article_IEq15.gif

13690_2023_1130_Article_IEq14.gif

13690_2023_1130_Article_IEq13.gif

13690_2023_1130_Article_IEq12.gif

13690_2023_1130_Article_IEq11.gif

13690_2023_1130_Article_IEq1.gif

Fig. 2

Fig. 1

40517_2023_252_Article_IEq139.gif

40517_2023_252_Article_IEq138.gif

40517_2023_252_Article_IEq136.gif

12938_2023_1116_Article_IEq21.gif

40517_2023_252_Article_IEq132.gif

Fig. 6

12938_2023_1116_Article_IEq11.gif

Fig. 2

Fig. 5

Fig. 4

40517_2023_252_Article_IEq129.gif

Fig. 3

Fig. 1

Fig. 1

Fig. 3

40517_2023_252_Article_IEq123.gif

40517_2023_252_Article_IEq122.gif

40517_2023_252_Article_IEq121.gif

40517_2023_252_Article_IEq120.gif

Fig. 2

40517_2023_252_Article_IEq117.gif

40517_2023_252_Article_IEq116.gif

40517_2023_252_Article_IEq115.gif

40517_2023_252_Article_IEq114.gif

Fig. 4

40517_2023_252_Article_IEq112.gif

40517_2023_252_Article_IEq111.gif

40517_2023_252_Article_IEq109.gif

40517_2023_252_Article_IEq107.gif

Fig. 2

Fig. 2

Fig. 2

40517_2023_252_Article_IEq103.gif

Fig. 1

Fig. 1

Fig. 2

Fig. 4

Fig. 1

40517_2023_252_Article_IEq99.gif

40517_2023_252_Article_IEq98.gif

40517_2023_252_Article_IEq97.gif

40517_2023_252_Article_IEq95.gif

Fig. 4

Fig. 1

40517_2023_252_Article_IEq92.gif

40517_2023_252_Article_IEq91.gif

40517_2023_252_Article_IEq90.gif

40517_2023_252_Article_IEq89.gif

40517_2023_252_Article_IEq88.gif

40517_2023_252_Article_IEq86.gif

Fig. 1

40517_2023_252_Article_IEq84.gif

40517_2023_252_Article_IEq83.gif

40517_2023_252_Article_IEq82.gif

40517_2023_252_Article_IEq81.gif

40517_2023_252_Article_IEq80.gif

40517_2023_252_Article_IEq79.gif

40517_2023_252_Article_IEq78.gif

40517_2023_252_Article_IEq77.gif

40517_2023_252_Article_IEq76.gif

Fig. 3

40517_2023_252_Article_IEq74.gif

40517_2023_252_Article_IEq73.gif

40517_2023_252_Article_IEq69.gif

Fig. 2

Fig. 4

Fig. 1

Fig. 3

Fig. 1

40517_2023_252_Article_IEq55.gif

40517_2023_252_Article_IEq53.gif

Fig. 3

40517_2023_252_Article_IEq51.gif

40517_2023_252_Article_IEq50.gif

40517_2023_252_Article_IEq49.gif

40517_2023_252_Article_IEq48.gif

40517_2023_252_Article_IEq47.gif

40517_2023_252_Article_IEq46.gif

40517_2023_252_Article_IEq45.gif

40517_2023_252_Article_IEq44.gif

40517_2023_252_Article_IEq43.gif

40517_2023_252_Article_IEq42.gif

40517_2023_252_Article_IEq41.gif

40517_2023_252_Article_IEq39.gif

40517_2023_252_Article_IEq37.gif

40517_2023_252_Article_IEq35.gif

Fig. 2

40517_2023_252_Article_IEq33.gif

Fig. 1

40517_2023_252_Article_IEq31.gif

40517_2023_252_Article_IEq30.gif

40517_2023_252_Article_IEq29.gif

Fig. 3

40517_2023_252_Article_IEq26.gif

Fig. 3

Fig. 4

Fig. 19

Fig. 18

Fig. 12

Fig. 2

Fig. 4

Fig. 2

Fig. 16

Fig. 1

Fig. 3

Fig. 14

Fig. 3

Fig. 13

Fig. 12

Fig. 1

Fig. 11

42004_2023_911_Article_IEq62.gif

Fig. 10

42004_2023_911_Article_IEq60.gif

Fig. 9

Fig. 7

Fig. 5

Fig. 9

Fig. 1

Fig. 1

42004_2023_911_Article_IEq48.gif

Fig. 5

Fig. 4

Fig. 3

Fig. 2

Fig. 2

Fig. 1

Fig. 5

42004_2023_911_Article_IEq37.gif

42004_2023_911_Article_IEq45.gif

Fig. 2

Fig. 1

Fig. 5

42004_2023_911_Article_IEq031.gif

42004_2023_911_Article_IEq30.gif

Fig. 5

42004_2023_911_Article_IEq28.gif

Fig. 8

Fig. 1

42004_2023_911_Article_IEq25.gif

42004_2023_911_Article_IEq10.gif

42004_2023_911_Article_IEq9.gif

42004_2023_911_Article_IEq8.gif

Fig. 7

Fig. 2

Fig. 1

Fig. 10

Fig. 1

12302_2023_750_Article_IEq5.gif

Fig. 2

Fig. 1

12864_2023_9424_Article_IEq11.gif

Fig. 6

Fig. 4

Fig. 3

Fig. 2

Fig. 3

Fig. 4

Fig. 6

Fig. 4

Fig. 1

Fig. 4

Fig. 3

Fig. 5