A Comparison of Bayesian and Maximum Likelihood Methods to Determine the Performance of a Point of Care Test for Helicobacter pylori in the Office Setting

Objective. Evaluations of point of care tests (PCT) are often hampered by a lack of appropriate gold standards. This study aimed to compare the results of a Bayesian statistical analysis and a maximum likelihood method to evaluate the performance of a PCT for Helicobacter pylori in primary care. Methods. The Helisal Rapid Blood Test (Cortecs Diagnostics) was performed in 311 patients from 6 primary care centers, and a concurrent venous sample was taken for 2 enzyme linked immunosorbent assays (ELISA) performed at the laboratory, blind to the PCT result. The Bayesian analysis was conducted using Markov Chain Monte Carlo methods (WinBUGS). The performance characteristics of the PCT and the 2 ELISA tests were estimated together with 95% credible intervals (95% CIs). Results. The estimate of prevalence of H. pylori in this population was 64% (95% CI, 59% to 70%), the sensitivity and specificity of the PCT were 89% (84% to 94%) and 84% (77% to 91%), respectively (likelihood ratios positive 5.6, negative 0.13). The equivalent maximum likelihood results were prevalence, 65%; sensitivity, 90%; and specificity, 83%. Conclusions. The Helisal Rapid Blood Test performed as well as laboratory-based ELISA tests in this cohort of patients. The Bayesian analysis and the maximum likelihood method gave similar results, the Bayesian method also simultaneously estimating 95% CIs.

[1]  Gordon H. Guyatt,et al.  Users' Guides to the Medical Literature: III. How to Use an Article About a Diagnostic Test A. Are the Results of the Study Valid? , 1994 .

[2]  R J Lilford,et al.  For Debate: The statistical basis of public policy: a paradigm shift is overdue , 1996, BMJ.

[3]  A. Feinstein,et al.  Spectrum Bias in the Evaluation of Diagnostic Tests: Lessons from the Rapid Dipstick Test for Urinary Tract Infection , 1992, Annals of Internal Medicine.

[4]  R. Logan,et al.  Accuracy of near-patient blood tests for Helicobacter pylori , 1996, The Lancet.

[5]  Bradley P. Carlin,et al.  Markov Chain Monte Carlo conver-gence diagnostics: a comparative review , 1996 .

[6]  D. Sanders,et al.  Comparison of serum, salivary, and rapid whole blood diagnostic tests for Helicobacter pylori and their validation against endoscopy based tests. , 1997, Gut.

[7]  A. Blum,et al.  Intragastric acidity as a predictor of the success of Helicobacter pylori eradication: a study in peptic ulcer patients with omeprazole and amoxicillin. , 1995, Gut.

[8]  Walter R. Gilks,et al.  A Language and Program for Complex Bayesian Modelling , 1994 .

[9]  R. Clemen,et al.  Correlations and Copulas for Decision and Risk Analysis , 1999 .

[10]  L. Joseph,et al.  Bayesian estimation of disease prevalence and the parameters of diagnostic tests in the absence of a gold standard. , 1995, American journal of epidemiology.

[11]  R. Logan,et al.  Testing for Helicobacter pylori infection: validation and diagnostic yield of a near patient test in primary care , 1999, BMJ.

[12]  Frederick Mosteller,et al.  Guidelines for Meta-analyses Evaluating Diagnostic Tests , 1994, Annals of Internal Medicine.

[13]  P. Malfertheiner,et al.  Performance of a Rapid Whole Blood Test for Helicobacter pylori in Primary Care: A German Multicenter Study , 1998, Helicobacter.

[14]  T. Borody,et al.  Evaluation of whole blood antibody kit to detect active Helicobacter pylori infection. , 1996, The American journal of gastroenterology.

[15]  Douglas G. Altman,et al.  Practical statistics for medical research , 1990 .

[16]  A R Feinstein,et al.  Use of methodological standards in diagnostic test research. Getting better but still not good. , 1995, JAMA.

[17]  M. Farthing,et al.  Ornithine decarboxylase as a marker for premalignancy in the stomach. , 1995, Gut.

[18]  S D Walter,et al.  Estimation of test error rates, disease prevalence and relative risk from misclassified data: a review. , 1988, Journal of clinical epidemiology.

[19]  G. Guyatt,et al.  Users' Guides to the Medical Literature: III. How to Use an Article About a Diagnostic Test: B. What Are the Results and Will They Help Me In Caring for My Patients? , 1994 .

[20]  Brendan Delaney,et al.  Near-Patient Tests in Primary Care: Setting the Standards for Evaluation , 2000, Journal of health services research & policy.

[21]  P N Valenstein,et al.  Evaluating diagnostic tests with imperfect standards. , 1990, American journal of clinical pathology.

[22]  D. Fitzmaurice,et al.  Systematic review of near patient test evaluations in primary care , 1999, BMJ.

[23]  A. Axon,et al.  Validation of a rapid whole blood test for diagnosing Helicobacter pylori infection , 1997, BMJ.

[24]  J. Hippisley-Cox,et al.  A pilot study of a randomized controlled trial of pragmatic eradication of Helicobactor pylori in primary care. , 1997, The British journal of general practice : the journal of the Royal College of General Practitioners.

[25]  R. Logan,et al.  Accuracy of near-patient blood tests for Helicobacter pylori , 1996, The Lancet.

[26]  N. Talley,et al.  Challenges in managing dyspepsia in general practice , 1997, BMJ.

[27]  Susan R. Wilson,et al.  A systematic review of near patient testing in primary care , 1997 .

[28]  L. Joseph,et al.  Inferences for Likelihood Ratios in the Absence of a "Gold Standard" , 1996, Medical decision making : an international journal of the Society for Medical Decision Making.

[29]  R. Jones,et al.  An evaluation of near‐patient testing for Helicobacter pylori in general practice , 1997, Alimentary pharmacology & therapeutics.