A hybrid model for combining case–control and cohort studies in systematic reviews of diagnostic tests

Systematic reviews of diagnostic tests often involve a mixture of case-control and cohort studies. The standard methods for evaluating diagnostic accuracy only focus on sensitivity and specificity and ignore the information on disease prevalence contained in cohort studies. Consequently, such methods cannot provide estimates of measures related to disease prevalence, such as population averaged or overall positive and negative predictive values, which reflect the clinical utility of a diagnostic test. In this paper, we propose a hybrid approach that jointly models the disease prevalence along with the diagnostic test sensitivity and specificity in cohort studies, and the sensitivity and specificity in case-control studies. In order to overcome the potential computational difficulties in the standard full likelihood inference of the proposed hybrid model, we propose an alternative inference procedure based on the composite likelihood. Such composite likelihood based inference does not suffer computational problems and maintains high relative efficiency. In addition, it is more robust to model mis-specifications compared to the standard full likelihood inference. We apply our approach to a review of the performance of contemporary diagnostic imaging modalities for detecting metastases in patients with melanoma.

[1]  John Weiner,et al.  Letter to the Editor , 1992, SIGIR Forum.

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

[3]  S. Zeger,et al.  Longitudinal data analysis using generalized linear models , 1986 .

[4]  Richard D Riley,et al.  Rejoinder to commentaries on ‘Multivariate meta‐analysis: Potential and promise’ , 2011 .

[5]  J. Kent.  Robust properties of likelihood ratio tests , 1982 .

[6]  Richard E. Chandler,et al.  Inference for clustered data using the independence loglikelihood , 2007 .

[7]  C. Harris,et al.  Developmental Toxicology , 2012, Methods in Molecular Biology.

[8]  Henrik Holmberg,et al.  Generalized linear models with clustered data: Fixed and random effects models , 2011, Comput. Stat. Data Anal..

[9]  Xiao-Hua Zhou,et al.  Statistical Methods in Diagnostic Medicine , 2002 .

[10]  H. Honest,et al.  Reporting of measures of accuracy in systematic reviews of diagnostic literature , 2002, BMC health services research.

[11]  Carlo Gaetan,et al.  Composite likelihood methods for space-time data , 2006 .

[12]  Benjamin Djulbegovic,et al.  Quality and methods of developing practice guidelines , 2002, BMC Health Services Research.

[13]  Dan Jackson,et al.  Multivariate meta-analysis: Potential and promise , 2011, Statistics in medicine.

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

[15]  Haitao Chu,et al.  Statistical methods for multivariate meta-analysis of diagnostic tests: An overview and tutorial , 2016, Statistical methods in medical research.

[16]  P Glasziou,et al.  Meta-analytic methods for diagnostic test accuracy. , 1995, Journal of clinical epidemiology.

[17]  Johannes B Reitsma,et al.  Meta-Analysis of Diagnostic Studies: A Comparison of Random Intercept, Normal-Normal, and Binomial-Normal Bivariate Summary ROC Approaches , 2008, Medical decision making : an international journal of the Society for Medical Decision Making.

[18]  Barry McDonald,et al.  Estimating Logistic Regression Parameters for Bivariate Binary Data , 1993 .

[19]  N. Jewell,et al.  Hypothesis testing of regression parameters in semiparametric generalized linear models for cluster correlated data , 1990 .

[20]  G. Molenberghs,et al.  Models for Discrete Longitudinal Data , 2005 .

[21]  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.

[22]  Roger M Harbord,et al.  A unification of models for meta-analysis of diagnostic accuracy studies. , 2007, Biostatistics.

[23]  H C Van Houwelingen,et al.  A bivariate approach to meta-analysis. , 1993, Statistics in medicine.

[24]  Satterthwaite Fe.  An approximate distribution of estimates of variance components. , 1946 .

[25]  D. Bates,et al.  Approximations to the Log-Likelihood Function in the Nonlinear Mixed-Effects Model , 1995 .

[26]  Sarah J E Barry,et al.  Linear mixed models for longitudinal shape data with applications to facial modeling. , 2008, Biostatistics.

[27]  J. Douglas,et al.  Confidence Regions for Parameter Pairs , 1993 .

[28]  J. C. Houwelingen,et al.  Bivariate Random Effects Meta-Analysis of ROC Curves , 2008, Medical decision making : an international journal of the Society for Medical Decision Making.

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

[30]  Luigi Pace,et al.  ADJUSTING COMPOSITE LIKELIHOOD RATIO STATISTICS , 2009 .

[31]  P. Song,et al.  Composite Likelihood Bayesian Information Criteria for Model Selection in High-Dimensional Data , 2010 .

[32]  Geert Verbeke,et al.  Pairwise Fitting of Mixed Models for the Joint Modeling of Multivariate Longitudinal Profiles , 2006, Biometrics.

[33]  C. Varin,et al.  A note on composite likelihood inference and model selection , 2005 .

[34]  Linear Mixed Model Approach in Penalized Spline Regression , 2013 .

[35]  Bruce G. Lindsay,et al.  Moment-Based Approximations of Distributions Using Mixtures: Theory and Applications , 2000 .

[36]  Adrian Pagan,et al.  Estimation, Inference and Specification Analysis. , 1996 .

[37]  C A Gatsonis,et al.  Regression methods for meta-analysis of diagnostic test data. , 1995, Academic radiology.

[38]  Yan Xing,et al.  Contemporary diagnostic imaging modalities for the staging and surveillance of melanoma patients: a meta-analysis. , 2011, Journal of the National Cancer Institute.

[39]  G. Molenberghs,et al.  Pseudolikelihood Modeling of Multivariate Outcomes in Developmental Toxicology , 1999 .

[40]  Theo Stijnen,et al.  Advanced methods in meta‐analysis: multivariate approach and meta‐regression , 2002, Statistics in medicine.

[41]  M. Bartlett.  Properties of Sufficiency and Statistical Tests , 1992 .

[42]  F. E. Satterthwaite.  An approximate distribution of estimates of variance components. , 1946, Biometrics.

[43]  S. Walter,et al.  Properties of the summary receiver operating characteristic (SROC) curve for diagnostic test data , 2002, Statistics in medicine.

[44]  S. Shapiro,et al.  An Analysis of Variance Test for Normality (Complete Samples) , 1965 .

[45]  P. Bossuyt,et al.  BMC Medical Research Methodology , 2002 .

[46]  L E Moses,et al.  Combining independent studies of a diagnostic test into a summary ROC curve: data-analytic approaches and some additional considerations. , 1993, Statistics in medicine.

[47]  C M Rutter,et al.  A hierarchical regression approach to meta‐analysis of diagnostic test accuracy evaluations , 2001, Statistics in medicine.

[48]  R. Henderson,et al.  A serially correlated gamma frailty model for longitudinal count data , 2003 .

[49]  Haitao Chu,et al.  A unification of models for meta-analysis of diagnostic accuracy studies. , 2009, Biostatistics.

[50]  Kalyan Das,et al.  Miscellanea. On the efficiency of regression estimators in generalised linear models for longitudinal data , 1999 .

[51]  Peter Trovitch,et al.  Early detection and treatment of skin cancer , 2002 .

[52]  L E Moses,et al.  Estimating Diagnostic Accuracy from Multiple Conflicting Reports , 1993, Medical decision making : an international journal of the Society for Medical Decision Making.

[53]  Haitao Chu,et al.  Bivariate Random Effects Meta-Analysis of Diagnostic Studies Using Generalized Linear Mixed Models , 2010 .