Home > Research > Publications & Outputs > Quantitative assessment of inconsistency in met...

Electronic data

  • Sousa-Pinto et al 2025

    Accepted author manuscript, 1.56 MB, PDF document

    Available under license: CC BY: Creative Commons Attribution 4.0 International License

Links

Text available via DOI:

View graph of relations

Quantitative assessment of inconsistency in meta-analysis using decision thresholds with two new indices

Research output: Contribution to Journal/MagazineJournal articlepeer-review

E-pub ahead of print

Standard

Quantitative assessment of inconsistency in meta-analysis using decision thresholds with two new indices. / Sousa-Pinto, Bernardo; Neumann, Ignacio; Vieira, Rafael José et al.
In: Journal of clinical epidemiology, Vol. 181, 111725, 31.05.2025.

Research output: Contribution to Journal/MagazineJournal articlepeer-review

Harvard

Sousa-Pinto, B, Neumann, I, Vieira, RJ, Bognanni, A, Marques-Cruz, M, Gil-Mata, S, Mordue, S, Nevill, C, Baio, G, Whaley, P, Schwarzer, G, Steele, J, Stewart, G, Schünemann, HJ & Azevedo, LF 2025, 'Quantitative assessment of inconsistency in meta-analysis using decision thresholds with two new indices', Journal of clinical epidemiology, vol. 181, 111725. https://doi.org/10.1016/j.jclinepi.2025.111725

APA

Sousa-Pinto, B., Neumann, I., Vieira, R. J., Bognanni, A., Marques-Cruz, M., Gil-Mata, S., Mordue, S., Nevill, C., Baio, G., Whaley, P., Schwarzer, G., Steele, J., Stewart, G., Schünemann, H. J., & Azevedo, L. F. (2025). Quantitative assessment of inconsistency in meta-analysis using decision thresholds with two new indices. Journal of clinical epidemiology, 181, Article 111725. Advance online publication. https://doi.org/10.1016/j.jclinepi.2025.111725

Vancouver

Sousa-Pinto B, Neumann I, Vieira RJ, Bognanni A, Marques-Cruz M, Gil-Mata S et al. Quantitative assessment of inconsistency in meta-analysis using decision thresholds with two new indices. Journal of clinical epidemiology. 2025 May 31;181:111725. Epub 2025 Mar 11. doi: 10.1016/j.jclinepi.2025.111725

Author

Sousa-Pinto, Bernardo ; Neumann, Ignacio ; Vieira, Rafael José et al. / Quantitative assessment of inconsistency in meta-analysis using decision thresholds with two new indices. In: Journal of clinical epidemiology. 2025 ; Vol. 181.

Bibtex

@article{b2b60054f03e469ca9a533372420f555,
title = "Quantitative assessment of inconsistency in meta-analysis using decision thresholds with two new indices",
abstract = "OBJECTIVES: In evidence synthesis, inconsistency is typically assessed visually and with the I 2 and the Q statistics. However, these measures have important limitations (i) if there are few primary studies of small sample sizes or (ii) if there are multiple studies with precise estimates. In addition, with the increasing use of decision thresholds (DT), for example in Grading of Recommendations Assessment, Development and Evaluation evidence to decision (EtD) frameworks, inconsistency judgments can be anchored around DTs. In this article, we developed quantitative measures to assess inconsistency based on DTs. STUDY DESIGN AND SETTING: We developed two measures to quantify inconsistency based on DTs - the decision inconsistency (DI) and the across-studies inconsistency (ASI) indices. The DI and the ASI are based on the distribution of the posterior samples studies' effect sizes (ES) across interpretation categories defined by DTs. We developed these indices for the Bayesian context, followed by a frequentist extension.RESULTS: The DI informs on the overall inconsistency of ESs across interpretation categories, while the ASI quantifies how different studies are compared to each other (in relation to interpretation categories) based on absolute effects. A DI ≥ 50% and an ASI ≥ 25% are suggestive of important inconsistency. We provide an R package (metainc) and a web tool (https://metainc.med.up.pt/) to support the computation of the DI and ASI, including in the context of sensitivity analyses assessing the impact of potential uncertainty in inconsistency.CONCLUSION: The DI and the ASI can contribute to quantitatively assess inconsistency, particularly as DTs are gaining recognition in evidence synthesis and health decision-making.",
keywords = "Systematic review, GRADE, Inconsistency, Heterogeneity, Meta-analysis",
author = "Bernardo Sousa-Pinto and Ignacio Neumann and Vieira, {Rafael Jos{\'e}} and Antonio Bognanni and Manuel Marques-Cruz and Sara Gil-Mata and Simone Mordue and Clareece Nevill and Gianluca Baio and Paul Whaley and Guido Schwarzer and James Steele and Gavin Stewart and Sch{\"u}nemann, {Holger J} and Azevedo, {Lu{\'i}s Filipe}",
year = "2025",
month = mar,
day = "11",
doi = "10.1016/j.jclinepi.2025.111725",
language = "English",
volume = "181",
journal = "Journal of clinical epidemiology",
issn = "0895-4356",
publisher = "Elsevier USA",

}

RIS

TY - JOUR

T1 - Quantitative assessment of inconsistency in meta-analysis using decision thresholds with two new indices

AU - Sousa-Pinto, Bernardo

AU - Neumann, Ignacio

AU - Vieira, Rafael José

AU - Bognanni, Antonio

AU - Marques-Cruz, Manuel

AU - Gil-Mata, Sara

AU - Mordue, Simone

AU - Nevill, Clareece

AU - Baio, Gianluca

AU - Whaley, Paul

AU - Schwarzer, Guido

AU - Steele, James

AU - Stewart, Gavin

AU - Schünemann, Holger J

AU - Azevedo, Luís Filipe

PY - 2025/3/11

Y1 - 2025/3/11

N2 - OBJECTIVES: In evidence synthesis, inconsistency is typically assessed visually and with the I 2 and the Q statistics. However, these measures have important limitations (i) if there are few primary studies of small sample sizes or (ii) if there are multiple studies with precise estimates. In addition, with the increasing use of decision thresholds (DT), for example in Grading of Recommendations Assessment, Development and Evaluation evidence to decision (EtD) frameworks, inconsistency judgments can be anchored around DTs. In this article, we developed quantitative measures to assess inconsistency based on DTs. STUDY DESIGN AND SETTING: We developed two measures to quantify inconsistency based on DTs - the decision inconsistency (DI) and the across-studies inconsistency (ASI) indices. The DI and the ASI are based on the distribution of the posterior samples studies' effect sizes (ES) across interpretation categories defined by DTs. We developed these indices for the Bayesian context, followed by a frequentist extension.RESULTS: The DI informs on the overall inconsistency of ESs across interpretation categories, while the ASI quantifies how different studies are compared to each other (in relation to interpretation categories) based on absolute effects. A DI ≥ 50% and an ASI ≥ 25% are suggestive of important inconsistency. We provide an R package (metainc) and a web tool (https://metainc.med.up.pt/) to support the computation of the DI and ASI, including in the context of sensitivity analyses assessing the impact of potential uncertainty in inconsistency.CONCLUSION: The DI and the ASI can contribute to quantitatively assess inconsistency, particularly as DTs are gaining recognition in evidence synthesis and health decision-making.

AB - OBJECTIVES: In evidence synthesis, inconsistency is typically assessed visually and with the I 2 and the Q statistics. However, these measures have important limitations (i) if there are few primary studies of small sample sizes or (ii) if there are multiple studies with precise estimates. In addition, with the increasing use of decision thresholds (DT), for example in Grading of Recommendations Assessment, Development and Evaluation evidence to decision (EtD) frameworks, inconsistency judgments can be anchored around DTs. In this article, we developed quantitative measures to assess inconsistency based on DTs. STUDY DESIGN AND SETTING: We developed two measures to quantify inconsistency based on DTs - the decision inconsistency (DI) and the across-studies inconsistency (ASI) indices. The DI and the ASI are based on the distribution of the posterior samples studies' effect sizes (ES) across interpretation categories defined by DTs. We developed these indices for the Bayesian context, followed by a frequentist extension.RESULTS: The DI informs on the overall inconsistency of ESs across interpretation categories, while the ASI quantifies how different studies are compared to each other (in relation to interpretation categories) based on absolute effects. A DI ≥ 50% and an ASI ≥ 25% are suggestive of important inconsistency. We provide an R package (metainc) and a web tool (https://metainc.med.up.pt/) to support the computation of the DI and ASI, including in the context of sensitivity analyses assessing the impact of potential uncertainty in inconsistency.CONCLUSION: The DI and the ASI can contribute to quantitatively assess inconsistency, particularly as DTs are gaining recognition in evidence synthesis and health decision-making.

KW - Systematic review

KW - GRADE

KW - Inconsistency

KW - Heterogeneity

KW - Meta-analysis

U2 - 10.1016/j.jclinepi.2025.111725

DO - 10.1016/j.jclinepi.2025.111725

M3 - Journal article

C2 - 39955079

VL - 181

JO - Journal of clinical epidemiology

JF - Journal of clinical epidemiology

SN - 0895-4356

M1 - 111725

ER -