Identifying items to assess methodological quality in physical therapy trials: a factor analysis.

Area of Science:

• Clinical Research Methodology
• Physical Therapy Evidence Synthesis

Background:

• Numerous tools exist to assess randomized controlled trial (RCT) methodological quality.
• Varying use of assessment items across health areas indicates a lack of consensus on relevance to trial quality or bias.
• A need exists for standardized, validated items specific to physical therapy research.

Purpose of the Study:

• To identify the underlying component structure of items used for assessing RCT quality.
• To determine relevant items for evaluating the quality and risk of bias in physical therapy trials.
• To utilize exploratory factor analysis (EFA) for item reduction and structure identification.

Main Methods:

• Methodological research design employing exploratory factor analysis (EFA).
• RCTs were randomly selected from the Cochrane Database of Systematic Reviews.
• Two reviewers assessed RCTs using 45 items from seven quality tools, followed by principal axis factoring (PAF) and varimax rotation.

Main Results:

• Exploratory factor analysis (EFA) identified 34 items loading onto nine common factors.
• These factors encompass selection bias, performance/detection bias, intervention/outcome details, outcome measurement properties, contamination/adherence, attrition bias, data analysis, sample size, and control adequacy.
• The identified factors provide a potential framework for evaluating physical therapy trial quality.

Conclusions:

• This study presents the first factor analysis exploring items for evaluating methodological quality or risk of bias in physical therapy RCTs.
• The identified items and factors serve as a foundation for future quality assessment in this field.
• Confirmatory factor analysis and empirical evidence linking these items to treatment effects are required for validation.