%0 Journal Article %J Archives of Public Health %D 2024 %T Methods for meta-analysis and meta-regression of binomial data: concepts and tutorial with Stata command metapreg %A Victoria N Nyaga %A M. Arbyn %K Binomal %K logistic regression %K meta-analyses %K Meta-regressions %K Network Meta-Analysis %K Stata %X

Background

Despite the widespread interest in meta-analysis of proportions, its rationale, certain theoretical and methodological concepts are poorly understood. The generalized linear models framework is well-established and provides a natural and optimal model for meta-analysis, network meta-analysis, and meta-regression of proportions. Nonetheless, generic methods for meta-analysis of proportions based on the approximation to the normal distribution continue to dominate.

Methods

We developed metapreg, a tool with advanced statistical procedures to perform a meta-analysis, network meta-analysis, and meta-regression of binomial proportions in Stata using binomial, logistic and logistic-normal models. First, we explain the rationale and concepts essential in understanding statistical methods for meta-analysis of binomial proportions and describe the models implemented in metapreg. We then describe and demonstrate the models in metapreg using data from seven published meta-analyses. We also conducted a simulation study to compare the performance of metapreg estimators with the existing estimators of the population-averaged proportion in metaprop and metan under a broad range of conditions including, high over-dispersion and small meta-analysis.

Conclusion

metapreg is a flexible, robust and user-friendly tool employing a rigorous approach to evidence synthesis of binomial data that makes the most efficient use of all available data and does not require ad-hoc continuity correction or data imputation. We expect its use to yield higher-quality meta-analysis of binomial proportions.

%B Archives of Public Health %V 82 %8 29/01/2024 %G eng %R https://link.springer.com/article/10.1186/s13690-023-01215-y %0 Journal Article %J Br J Cancer %D 2023 %T Accuracy of HPV E6/E7 oncoprotein tests to detect high-grade cervical lesions: a systematic literature review and meta-analysis. %A Laura Downham %A Iman Jaafar %A Mary Luz Rol %A Victoria N Nyaga %A Joan Valls %A Armando Baena %A Zhang, Li %A Marc J Gunter %A M. Arbyn %A Maribel Almonte %K Cervical cardiogenesis %K Cervical lesions %K HPV %X

BACKGROUND:
Cervical carcinogenesis is mediated by the HPV-E6 and E7 oncoproteins, considered as biomarkers usable in managing screen-positive women.

METHODS:
We conducted a systematic review and meta-analysis assessing the accuracy of HPV-E6/E7-oncoprotein tests to detect underlying cervical-precancer and cancer. We included studies reporting data on oncoprotein test accuracy detecting cervical intraepithelial neoplasia grade 3 or worse. Random effects logistic regression models were applied for pooling absolute and relative accuracy.

RESULTS:
Twenty-two studies were included. Sensitivity and specificity estimates ranged from 54.2% (95%CI: 45.2-63.0) to 69.5% (95%CI:60.8-76.9) and from 82.8% (95%CI: 50.4-95.8) to 99.1 (95%CI: 98.8-99.3), respectively in the population irrespective of HPV status. Higher sensitivity estimates ranging from 60.8% (95%CI: 49.6-70.9) to 75.5% (95%CI: 71.7-78.9) but lower specificity estimates ranging from 83.7% (95%CI: 76.1-89.3) to 92.1% (95%CI: 88.5-94.6) were observed in studies enrolling high-risk-HPV-positive women. Studies recruiting only HIV-positive women showed a pooled sensitivity of 46.9% (95%CI: 30.6-63.9) with a specificity of 98.0% (95%CI: 96.8-98.7).

CONCLUSIONS:
The high specificity of oncoprotein tests supports its use for triaging HPV-positive women. However, oncoprotein-negative women would not be recommended to undertake routine screening, requiring further follow-up. Large-scale and longitudinal studies are needed to further investigate the role of E6/E7-oncoprotein detection in predicting the risk of developing cervical pre-cancer and cancer.

%B Br J Cancer %8 2023 Nov 16 %G eng %R 10.1038/s41416-023-02490-w %0 Journal Article %J Res Synth Methods %D 2023 %T Comparison and validation of metadta for meta-analysis of diagnostic test accuracy studies. %A Victoria N Nyaga %A M. Arbyn %K Diagnostic Tests, Routine %K Network Meta-Analysis %K Research Design %K SOFTWARE %X

We developed metadta, a flexible, robust, and user-friendly statistical procedure that fuses established and innovative statistical methods for meta-analysis, meta-regression, and network meta-analysis of diagnostic test accuracy studies in Stata. Using data from published meta-analyses, we validate metadta by comparing and contrasting its features and output to popular procedures dedicated to the meta-analysis of diagnostic test accuracy studies; (midas [Stata], metandi [Stata], metaDTA [web application], mada [R], and MetaDAS [SAS]). We also demonstrate how to perform network meta-analysis with metadta, for which no alternative procedure is dedicated to network meta-analysis of diagnostic test accuracy data in the frequentist framework. metadta generated consistent estimates in simple and complex diagnostic test accuracy data sets. We expect its availability to stimulate better statistical practice in the evidence synthesis of diagnostic test accuracy studies.

%B Res Synth Methods %V 14 %8 2023 May %G eng %N 3 %R 10.1002/jrsm.1634 %0 Journal Article %J Archives of Public Health %D 2019 %T The EMPCAN study: protocol of a population-based cohort study on the evolution of the socio-economic position of workers with cancer %A Regine Kiasuwa %A Victoria N Nyaga %A Otter, Renée %A De Brouwer, Christophe %A Bouland, Catherine %K cancer %K Determining factors %K Life course %K Return-to-work %X

Background: The improvements in cancer control led to an increase in the number of cancer survivors, notably, in
the working age population (16–64 years). There is a strong need to assess and understand their reintegration on
the labour market, which underlines and ensures their social integration and quality of life. The objectives of the
EMPCAN study is therefore to measure the scale of return-to-work after cancer and to identify the determining
factors, allowing for the implementation of an adequate socio-professional support.
Methods: We requested data from the Belgian Cancer Registry and the Crossroad Bank for Social Security. We
included all socially insured Belgian workers diagnosed between 2004 and 2011 with colorectal, breast, head &
neck, prostate, testis, lung and corpus uteri cancer. The end of (administrative) follow-up was 31st December 2012.
We include demographic, health-related and work-related factors in the analysis and observed how these factors
interplay to determine the working status. After having solved legal, ethical and technical issues for the coupling,
we will perform survival analysis with competing risks using the Fine and Gray model; we will also perform a
multistate model using transitions probabilities; and finally, a group-based modeling for longitudinal data using the
‘proc traj’ package in SAS.
Discussion: The results of the EMPCAN study will allow the provision of an evidence-based support to professional
reintegration policies. It will also bring some key features for the prediction of the cancer-related social security
needs. Besides the raise of awareness among health professionals and policy makers, this study could lead to a
better planning and organization of vocational rehabilitation programs.
 

%B Archives of Public Health %V 77 %8 Jan-12-2019 %G eng %N 1 %R 10.1186/s13690-019-0337-1 %0 Journal Article %J Stat Methods Med Res %D 2016 %T ANOVA model for network meta-analysis of diagnostic test accuracy data. %A Victoria N Nyaga %A Aerts, Marc %A M. Arbyn %X

Procedures combining and summarising direct and indirect evidence from independent studies assessing the diagnostic accuracy of different tests for the same disease are referred to network meta-analysis. Network meta-analysis provides a unified inference framework and uses the data more efficiently. Nonetheless, handling the inherent correlation between sensitivity and specificity continues to be a statistical challenge. We developed an arm-based hierarchical model which expresses the logit transformed sensitivity and specificity as the sum of fixed effects for test, correlated study-effects to model the inherent correlation between sensitivity and specificity and a random error associated with various tests evaluated in a given study. We present the accuracy of 11 tests used to triage women with minor cervical lesions to detect cervical precancer. Finally, we compare the results with those from a contrast-based model which expresses the linear predictor as a contrast to a comparator test. The proposed arm-based model is more appealing than the contrast-based model since the former permits more straightforward interpretation of the parameters, makes use of all available data yielding shorter credible intervals, and models more natural variance-covariance matrix structures.

%B Stat Methods Med Res %8 2016 Sep 20 %G eng %1 http://www.ncbi.nlm.nih.gov/pubmed/27655805?dopt=Abstract %R 10.1177/0962280216669182 %0 Journal Article %J Arch Public Health %D 2014 %T Metaprop: a Stata command to perform meta-analysis of binomial data. %A Victoria N Nyaga %A M. Arbyn %A Aerts, Marc %X

BACKGROUND: Meta-analyses have become an essential tool in synthesizing evidence on clinical and epidemiological questions derived from a multitude of similar studies assessing the particular issue. Appropriate and accessible statistical software is needed to produce the summary statistic of interest.

METHODS: Metaprop is a statistical program implemented to perform meta-analyses of proportions in Stata. It builds further on the existing Stata procedure metan which is typically used to pool effects (risk ratios, odds ratios, differences of risks or means) but which is also used to pool proportions. Metaprop implements procedures which are specific to binomial data and allows computation of exact binomial and score test-based confidence intervals. It provides appropriate methods for dealing with proportions close to or at the margins where the normal approximation procedures often break down, by use of the binomial distribution to model the within-study variability or by allowing Freeman-Tukey double arcsine transformation to stabilize the variances. Metaprop was applied on two published meta-analyses: 1) prevalence of HPV-infection in women with a Pap smear showing ASC-US; 2) cure rate after treatment for cervical precancer using cold coagulation.

RESULTS: The first meta-analysis showed a pooled HPV-prevalence of 43% (95% CI: 38%-48%). In the second meta-analysis, the pooled percentage of cured women was 94% (95% CI: 86%-97%).

CONCLUSION: By using metaprop, no studies with 0% or 100% proportions were excluded from the meta-analysis. Furthermore, study specific and pooled confidence intervals always were within admissible values, contrary to the original publication, where metan was used.

%B Arch Public Health %V 72 %P 39 %8 2014 %G eng %N 1 %1 http://www.ncbi.nlm.nih.gov/pubmed/25810908?dopt=Abstract %R 10.1186/2049-3258-72-39