Systematic reviews of diagnostic tests: a guide to methods and application

Area of Science:

• Medical research methodology
• Evidence-based medicine
• Diagnostic test evaluation

Background:

• Systematic reviews are crucial for synthesizing evidence on diagnostic test accuracy.
• Evaluating test accuracy requires a structured approach to minimize bias.
• Previous reviews may lack standardized methodologies, impacting reliability.

Purpose of the Study:

• To delineate the fundamental steps involved in conducting systematic reviews for test accuracy.
• To identify and discuss potential threats to the validity of systematic reviews at each stage.
• To provide a practical framework illustrated with a real-world example.

Main Methods:

• Defining review questions using a structured PICO (Population, Intervention, Comparison, Outcome) format.
• Conducting comprehensive literature searches across multiple databases.
• Assessing the methodological quality and risk of bias in included studies.
• Synthesizing study characteristics and results, including meta-analysis where appropriate.
• Interpreting findings for clinical relevance and generating recommendations.

Main Results:

• The process involves five core steps: question formulation, literature search, quality assessment, data synthesis, and inference generation.
• Each step presents specific threats to validity that must be addressed.
• A review on cervicovaginal fetal fibronectin test accuracy serves as a practical illustration.

Conclusions:

• Adherence to systematic review methodology enhances the reliability of test accuracy evaluations.
• Careful attention to each step, from question design to inference, is vital for valid conclusions.
• This framework supports evidence-based decision-making in clinical practice.