Stroke 3 - RCT/SR Flashcards
What type of study is a randomised control trial?
Experimental
Name two important methods of avoiding selection bias in RCT
- Random assignment of intervention and control treatments
- Allocations concealment
- Efforts to minimise loss to follow-up
How to reduce information bias in RCT?
Blinding - people who collect the information and analyse the infromation
Standardised procedures for collecting/measuring outcomes
What is the difference between internal and external validity?
Internal = valid estimate of treat effect in THIS sample
External = generalisability to a wider population group
What framework is used to develop a well defined research question?
PICO
Population - e.g male over 50 with colorectal cancer
Intervention - e.g chemotherapy
Comparison - e.g radiotherapy
Outcome - e.g years survival
What are the two different types of RCT?
How are they different?
Intention to treat - population is randomised then left alone, no follow up on adherence to treatment regime -> favours real life
Per-protocol -> patients must adhere to treatment as set out to be incluced in study -> if not drop out
What type of RCT is preferred, Intention to treat or per-protocol?
WHy?
Intention to treat
Mimics real life - patient variation in compliance to treatment
Has low drop out rates
What is the purpose of CONSORT guidelines in research?
Used to create a transparent guideline for writing up RCT -> ensures all correct information is included
Results in more transparent and higher quality publications
Allows reader to critically apprais and understand reports
What is the purpose of registration of a clinical trial?
Ensures fair and unique
Identifies that trial is going on and has a purpose
What is the protocol of a RCT?
The recipe for that trial - exactly what is happening, would allow someone else to replicate the RCT exactly - in theory should get the same results
What is the purpose of the CONSORT flow chart of participants?
Creates transparency of who and why was included in a study
Enrolement - how many assessed for eligibility and number excluded with why?
Allocation = how many in each group, how many of these recieved conditions of allocation
Follow up = number discontinued or lost to follow up
Analysis = number analysied, excluded from analysis and why
What are the hierarchy of evidence?
Meta-analysis
SR
RCT
Cohort study
Case-control study
Qualitative study
Cross-sectional study
Case series/reports
What is the purpose of a forest plot?
Condense and represent the information from multiple studies/clinical trials
Shows the heterogenicity of findings and estimates the common effect.
How to interpret a forest plot?
Left column = details of included studies
Right colomn = numerical respresentation of graph
Forest plot
Vertical line = point of null effect (absolute stats at 0, relative stats at 1)
Each study = sqaure size is population, horizontal line is confidence intervals (not smaller pop tend to have larger CI)
Diamond = combined study data, size = pop, horizontal line = CI
What sort of questions should be asked in a critical appraisal?
Do I believe the results?
Does the study have any biases present?
Were the findings of the study statistically significant?
Are the findings generalizable to my patients (age, comorbidities, etc)?
Is the intervention feasible (cost, time, production)?
Are long term outcomes reported?
How does the study compare to other literature or guidance?
What guidelines are used for systematic reviews?
PRISMA guidelines
What might be included for inclusion and exclusion criteria?
Population
Intervention
Comparator
Outcome
Study design
Language
Publication date
For scoping reviews what method is used to identify is all relevant articles are found?
How to interpret it?
Funnel diagram
Expect symetrical/included within the triange = good results
Not symmetrical/within the triangle = likely missing and should search again.
What tests can be used to identify significant heterogeneity in trials in a scoping review?
I sqaured - less than 25% shows no significant heterogeneity
Chi sqaured p>0.1 no significant heterogeneity
What is a health technology assessment?
A systematic approach to evaluate the properties, effects and impacts of health technologies or interventions. For example - medical devices, health services and vaccines etc
What is the difference between cost effectiveness, cost benefit analysis and ICER?
Cost effectiveness - how well treatment works (normal medical values e.g reduction in BP) related to how much costs per NHS
Cost benefit analysis - monetary values difference between the intervention and benefits of treatment
ICER - cost of one additional unit of quality adjusted life years.
What are the three goals of NICE technology assessments?
Improving health outcomes
Providing equitable access
Ensuring cost-effectiveness and sustainabilty
