Meta-analysis for orthodontists: Part I – How to choose effect measure and statistical model
The DOI matches its Crossref record; 5 of the 5 references it cites were found in a registry or an index.
Abstract
Systematic reviews ideally provide a comprehensive and unbiased summary of existing evidence from clinical studies, whilst meta-analysis combines the results of these studies to produce an overall estimate. Collectively, this makes them invaluable for clinical decision-making. Although the number of published systematic reviews and meta-analyses in orthodontics has increased, questions are often raised about their methodological soundness. In this primer, the first steps of meta-analysis are discussed, namely the choice of an effect measure to express the results of included studies, and the choice of a statistical model for the meta-analysis. Clinical orthodontic examples are given to explain the various options available, the thought process behind the choice between them and their interpretation.
Overview
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Integrity signals
Registry
Cites
Statistics
Verification
Claims the paper makes
The claims and statistics of this paper haven’t been checked yet.
Free: it reads the open-access text and checks the first 5 claims that cite a source we can check. Create a free account to check every claim.
Self-check
Percentages, sums and reported differences re-computed from the paper’s own numbers.
The self-check needs the paper’s full text.
Citations
Cites
Beta blockade during and after myocardial infarction: An overview of the randomized trials
Yusuf, S., Peto, R., Lewis, J., Collins, R., & Sleight, P. (1985) · Progress in Cardiovascular Diseases · doi:10.1016/s0033-0620(85)80003-7 · Cited by 3,205 · cited 1×
“For ORs, there is also the Peto method (Yusuf et al., 1985), which is a good method for studies with few events, small effects (OR close to 1) and similar numbers in the experimental and control group.”
MethodA Re-Evaluation of Random-Effects Meta-Analysis
Higgins, J. P. T., Thompson, S. G., & Spiegelhalter, D. J. (2009) · Journal of the Royal Statistical Society Series A: Statistics in Society · doi:10.1111/j.1467-985x.2008.00552.x · Cited by 2,569 · cited 1×
“Additionally, 95% predictive intervals should always be calculated for random-effects meta-analyses (Higgins et al., 2009).”
BackgroundDeeks, J. J. (2002) · Statistics in Medicine · doi:10.1002/sim.1188 · Cited by 660 · cited 1×
“According to existing empirical studies (Engels et al., 2000; Deeks, 2002; Furukawa et al., 2002), the following recommendations can be made about the choice of effect measure for binary outcomes: (i) the use of RD should be avoided, (ii) the RR for harm (i.e. incidence of incisal trauma) is…”
Furukawa, T. A., Guyatt, G. H., & Griffith, L. E. (2002) · International Journal of Epidemiology · doi:10.1093/ije/31.1.72 · Cited by 318 · cited 3×
“N Therefore, Furukawa et al. (2002) proposed the calculation of an individualised NNT based on the Patient’s Expected Event Rate (PEER) and either the RR or the OR: NNT~ 1 PEER|(1{RR) or NNT~ 1{PEERzOR|PEER PEER|(1{OR)|(1{PEER) Therefore, suppose we want to calculate the NNT for two patient…”
MethodInfluentialDetecting and describing heterogeneity in meta-analysis
Hardy, R. J., & Thompson, S. G. (1998) · Statistics in Medicine · doi:10.1002/(sici)1097-0258(19980430)17:8<841::aid-sim781>3.0.co;2-d · cited 1×
“The biggest limitation of the fixed-effect model is that existing means for the identification of heterogeneity are very imprecise (Hardy and Thompson, 1998), meaning that heterogeneity is very likely to exist, whether we find it or not.”
Background
Cited by
Zymperdikas, V. F., Koretsi, V., Papageorgiou, S. N., & Papadopoulos, M. A. (2016) · The European Journal of Orthodontics · doi:10.1093/ejo/cjv034 · Cited by 297
Self-citationInflammatory bowel disease and oral health: systematic review and a meta‐analysis
Papageorgiou, S. N., Hagner, M., Nogueira, A. V. B., Franke, A., Jäger, A., & Deschner, J. (2017) · Journal of Clinical Periodontology · doi:10.1111/jcpe.12698 · Cited by 195
“Heterogeneity could not be explained in some cases due to the limited number of studies, while publication bias diagnostics and sensitivity analyses could not be performed, which might pose a threat to the results’ validity (Papageorgiou 2014b, Papageorgiou et al. 2015c).”
MethodSelf-citationBasic study design influences the results of orthodontic clinical investigations
Papageorgiou, S. N., Xavier, G. M., & Cobourne, M. T. (2015) · Journal of Clinical Epidemiology · doi:10.1016/j.jclinepi.2015.03.008 · Cited by 108
“Finally, when incorporating non-RCTs in systematic reviews, it will always make sense to explore potential sources of heterogeneity and adopt a random-effects approach to acknowledge the unexplained heterogeneity [56,57].”
MethodSelf-citationPapageorgiou, S. N., Antonoglou, G. N., Michelogiannakis, D., Kakali, L., Eliades, T., & Madianos, P. (2022) · Journal of Clinical Periodontology · doi:10.1111/jcpe.13529 · Cited by 75
Self-citationHou, H., Wang, C., Sun, F., Zhao, L., Dun, A., & Sun, Z. (2015) · Inflammation Research · doi:10.1007/s00011-015-0850-9 · Cited by 55
“A fixed effect model was used to calculate pooled OR when I(2)\ 50 %; otherwise, a random effect model was adopted to combine the data [44].”
Method
Metadata
- Authors
- Papageorgiou, S. N.
- Venue
- Journal of Orthodontics
- Publisher
- Informa UK Limited
- Published
- 2014-12
- Volume
- 41
- Issue
- 4
- Pages
- 317-326
- ISSN
- 1465-3125 (ISSN-L)
- 1465-3133
- OpenAlex
- W2001541413
- Cited by
- 82
- References
- 5
- FWCI
- 7.37
- Type
- journal article
- Language
- en
- Open access
- closed
About these checks
Registry, retraction, Cites and statistics results are computed, not judged, with one exception: a printed reference that no index lists is read once by a model before it counts as a citation error. Claims are extracted by a model and kept only when their quote appears verbatim in the text; each cited source is then checked, with a model’s help, for whether it says what the claim relies on. Those verdicts can be wrong — read the claim’s quote and the cited source before relying on one.
Statistics and arithmetic are recomputed, not judged.
Confidence bands are shown only for fields calibrated on our annotated test set. None is calibrated yet, so no band is shown.
What counts as a claim ›How we measure confidence ›How we check this ›
Are you an author, or do you see a problem with this page? Request a review
Data sources
Bibliographic data: Semantic Scholar, OpenAlex, Crossref.
Contains information from the Semantic Scholar Academic Graph Datasets, made available under the ODC Attribution License (ODC-BY 1.0).
Semantic Scholar’s data is described in: Kinney, R., et al. (2023). The Semantic Scholar Open Data Platform. arXiv:2301.10140.