Week 5: Efficacy of Cognitive Therapy: Systematic Review Flashcards
Systematic Review
A rigorous method for combining evidence from multiple studies.
Purpose of Systematic Review
- Collate and summarise available evidence
- Permit evidence-based recommendations about useful treatments
- Identify gaps and shortcomings in our evidence
- Generate new hypotheses that can be explored
Essential Elements of a Systematic Review
- Clear focus (one or more specific objectives/questions)
- Pre-defined eligibility criteria for studies
- Systematic literature search strategy
- Explicit, reproducible methodology
- Assessment of the quality of the study
- Systematic presentation and synthesis of studies and findings
Aim of a Systematic Review
Synthesize the evidence and arrive at answers to the starting question.
Attach more importance or weight, to the outcome of a high-quality study than a low-quality one.
Grading of Recommendations, Assessment, Development, & Evaluation (GRADE)
A commonly used system to assess quality.
Assigns evidence a rating of 4 (highest quality) to 1 (very low), and 0 (the study should be disregarded)
Upgrades / Downgrades
Essentially, the concept of multiple downgrades or upgrades reflects the iterative process of assessing evidence quality and the potential for multiple factors to influence the final rating
Possible Reasons for a Downgrade
- Lack of Blinding
- Randomization problems
- Small sample size
- Variable results
- Poor participant retention
- Missing or incomplete data
- Subjective outcomes
- Selective reporting of results
- Risk of bias: study design has significant flaws affecting the results
- Inconsistency: results from different studies are contradictory or vary widely
- Indirectness: evidence doesn’t directly address the specific question of interest
- Imprecision: confidence intervals for the effect estimates are wide
- Publication bias
Possible Reasons for an Upgrade
- Large magnitude effect
- Dose-response relationship
- Designed to minimize bias
Meta-analysis
A statistical approach that allows us to combine the findings from several different studies to draw overall conclusions about a treatment’s efficacy.
Advantage of Meta-analysis
Collate several smaller studies that, individually, may provide inconclusive evidence, but together can provide a clearer picture, and estimate the treatment effect.
Standardized Effect Size
Describe the results of the different studies in a standard way, before we combining into a meta-analysis.
Ways to Calculating Standardized Effect Size
- Continuous Data: outcomes that are assessed gradually or on a continuous scale
- Discrete Event: an occurrence that happens at a specific point in time
Cohen’s Index or Hedge’s G
Common measures of standardized effect size for continuous data. A larger effect size indicates a stronger treatment effect.
A standardized mean difference of 0.8 or above is often regarded as large, but a 0.5 moderate, and 0.2 small.
Relative Risk Ratio (RR)
A statistical measure used to compare the risk of a specific event (like recovery or relapse) between two groups.
RR>1 More likely
RR<1 Less likely
RR=1 No difference
*Can be misinterpreted if not used alongside other measures like absolute risk reduction.
Discrete Event
Typically, a one-time occurrence that is recorded as a data point. It’s a distinct event with a clear start and end point.
