Meta-analysis for the comparison of two diagnostic tests to a common gold standard: A generalized linear mixed model approach

Meta-analysis of diagnostic studies is still a rapidly developing area of biostatistical research. Especially, there is an increasing interest in methods to compare different diagnostic tests to a common gold standard. Restricting to the case of two diagnostic tests, in these meta-analyses the parameters of interest are the differences of sensitivities and specificities (with their corresponding confidence intervals) between the two diagnostic tests while accounting for the various associations across single studies and between the two tests. We propose statistical models with a quadrivariate response (where sensitivity of test 1, specificity of test 1, sensitivity of test 2, and specificity of test 2 are the four responses) as a sensible approach to this task. Using a quadrivariate generalized linear mixed model naturally generalizes the common standard bivariate model of meta-analysis for a single diagnostic test. If information on several thresholds of the tests is available, the quadrivariate model can be further generalized to yield a comparison of full receiver operating characteristic (ROC) curves. We illustrate our model by an example where two screening methods for the diagnosis of type 2 diabetes are compared.

[1]  Johannes B Reitsma,et al.  Bivariate analysis of sensitivity and specificity produces informative summary measures in diagnostic reviews. , 2005, Journal of clinical epidemiology.

[2]  Emmanuel Lesaffre,et al.  A general framework for comparative Bayesian meta-analysis of diagnostic studies , 2015, BMC Medical Research Methodology.

[3]  Theo Stijnen,et al.  Multivariate random effects meta-analysis of diagnostic tests with multiple thresholds , 2009, BMC medical research methodology.

[4]  Hitoshi Shimano,et al.  Use of high‐normal levels of haemoglobin A1C and fasting plasma glucose for diabetes screening and for prediction: a meta‐analysis , 2013, Diabetes/metabolism research and reviews.

[5]  J Menke,et al.  Bivariate Random-effects Meta-analysis of Sensitivity and Specificity with SAS PROC GLIMMIX , 2009, Methods of Information in Medicine.

[6]  Mir Said Siadaty,et al.  Repeated-measures modeling improved comparison of diagnostic tests in meta-analysis of dependent studies. , 2004, Journal of clinical epidemiology.

[7]  Oliver Kuss,et al.  Nonparametric meta‐analysis for diagnostic accuracy studies , 2015, Statistics in medicine.

[8]  Thomas A Trikalinos,et al.  Methods for the joint meta‐analysis of multiple tests , 2013, Research synthesis methods.

[9]  Maria Adam,et al.  A multivariate method for meta‐analysis and comparison of diagnostic tests , 2016, Statistics in medicine.

[10]  Pablo Martínez-Camblor,et al.  Fully non-parametric receiver operating characteristic curve estimation for random-effects meta-analysis , 2017, Statistical methods in medical research.

[11]  C Gatsonis,et al.  Meta‐analysis of Diagnostic Test Accuracy Assessment Studies with Varying Number of Thresholds , 2003, Biometrics.

[12]  Patrick Bossuyt,et al.  Systematic Reviews of Diagnostic Test Accuracy , 2008, Annals of Internal Medicine.

[13]  S. Dharmage,et al.  HbA(1c) as a screening tool for detection of Type 2 diabetes: a systematic review. , 2007, Diabetic medicine : a journal of the British Diabetic Association.

[14]  Patrick M. M. Bossuyt,et al.  Cochrane Handbook for Systematic Reviews of Diagnostic Test Accuracy , 2013 .

[15]  S. Dharmage,et al.  HbA1c as a screening tool for detection of Type 2 diabetes: a systematic review , 2007 .

[16]  Oliver Kuss,et al.  Meta‐analysis for diagnostic accuracy studies: a new statistical model using beta‐binomial distributions and bivariate copulas , 2014, Statistics in medicine.

[17]  Yemisi Takwoingi,et al.  Empirical Evidence of the Importance of Comparative Studies of Diagnostic Test Accuracy , 2013, Annals of Internal Medicine.

[18]  Richard D Riley,et al.  Meta-analysis of test accuracy studies with multiple and missing thresholds : a multivariate-normal model , 2014 .

[19]  E. Picano,et al.  The comparable diagnostic accuracies of dobutamine‐stress and dipyridamole‐stress echocardiographies: a meta‐analysis , 2000, Coronary artery disease.

[20]  Haitao Chu,et al.  Bivariate meta-analysis of sensitivity and specificity with sparse data: a generalized linear mixed model approach. , 2006, Journal of clinical epidemiology.

[21]  Haitao Chu,et al.  Meta‐analysis of diagnostic accuracy studies accounting for disease prevalence: Alternative parameterizations and model selection , 2009, Statistics in medicine.

[22]  BOULIN,et al.  [Classification and diagnosis of diabetes]. , 1953, Concours medical.

[23]  Jianfen Shu,et al.  Proportional odds ratio model for comparison of diagnostic tests in meta-analysis , 2004, BMC medical research methodology.

[24]  Constantine Gatsonis,et al.  Analysing and Presenting Results , 2010 .

[25]  David J Samson,et al.  Challenges in Systematic Reviews of Diagnostic Technologies , 2005, Annals of Internal Medicine.

[26]  O Kuss,et al.  Meta‐analysis of diagnostic tests accounting for disease prevalence: a new model using trivariate copulas , 2015, Statistics in medicine.

[27]  Jing Ning,et al.  A composite likelihood method for bivariate meta-analysis in diagnostic systematic reviews , 2017, Statistical methods in medical research.