Critical Appraisal Flashcards
What is critical appraisal? What are the THREE factors that are assessed?
The process of carefully and systematically examining research to judge its trustworthiness, and its value and relevance in a particular context
Validity, results and clinical relevance are assessed
What is PICO when talking about critical appraisal?
Population: Who are they? What are their characteristics?
Intervention: What was done? What drug was given at what dose?
Comparator: New intervention investigated, what was it compared to as stdnadard care?
Outcome: What was measured and what was found?
What are the two questions that have to be asked before doing a critical appraisal?
- Where is the trial published? –> High-quality peer-reviewed journals
- Is the trial sponsored? –> Any conflicts of interest or bias that may affect how results are presented
What are the four types of BIAS in clinical trials?
- Selection bias
- Performance bias
- Detection bias
- Attrition bias
How to minimise bias?
- Effective randomisation
- Matching of participants
- Other treatments, measurements
- Blinding of participants and investigators
Define the following abbreviations for the study methods of an critical appraisal;
A) R (population)
B) A
C) M
D) M,b,o (outcome)
A)
- Recruit a large enough sample PLUS Recruit subjects randomly
- OR Recruit consecutive patients
B)
- Allocate randomly to groups (and conceal allocation) OR Adjust for confounding
C)
- Manage groups equally
- Follow up all subjects
D)
- Measure outcomes with:
- blinded subjects and assessors and/or
- objective measures –> using numbers as much as possible
What are the examples of descriptive studies used in critical appraisals?
- Mean ± SD for continuous variables (or median [range or IQR])
- Number (%) for categorical variables
What are the examples of comparative statistics used in critical appraisals?
- t-tests to compare means; ANOVA for 3 or more means
- x2 analysis for categorical variables
Explain the meaning of the following results
A) Relative risk (RR). RR = risk in the intervention group/ risk in the control group. What does a value of 1, <1 and > 1 mean?
B) Odds ratio (OR). OR = odds of an event occurring/ odds of it not occurring. What does a value of 1, <1 and > 1 mean?
C) Absolute risk reduction (ARR). ARR = risk of an event in the control group - risk in the intervention group. What does a value of 0, +ve and -ve mean?
D) Relative risk reduction.(RRR) RRR = 1 – ARR
E) Number needed to treat (NNT) NNT = 1/ARR
A)
- RR = 1. No difference
- RR < 1. Intervention reduces risk
- RR > 1. Intervention increases risk
B)
- OR = 1. No difference
- OR < 1. Exposure associated with reduced risk
- OR > 1. Exposure associated with increased risk
C)
- ARR = 0. No difference
- ARR +ve. Intervention beneficial
- ARR –ve. Intervention harmful
D)
- Reflects reduction in the rate of the event in intervention vs control group
E)
- Number of patients we need to treat to prevent 1 negative outcome; a small number is desirable (also consider NNH)
What do checklists for critical appraisals assess?
- Safety
- Tolerability
- Efficacy
- Price