Guidelines Flashcards
What is the definition of a clinical guideline, and what are sources of evidence to make clinical guidelines?
A systematically developed statement
Assists practitioners and patients in making decisions regarding healthcare in clinical circumstances
Sources of evidence = RCT’s/on-going clinical trials -> meta-analyses and SR’s are carried out on these, to give higher value results than clinical opinion alone
What are the requirements of a guideline?
Valid Reliable Flexible Applies to wide population Reviewed regularly (-> patients are put in danger if these requirements are not met)
Describe the hierarchy of evidence used to create systematic reviews and meta-analyses
Meta-analyses/systematic review RCT Uncontrolled RCT Case / cohort studies Expert opinion
What parameters are used to assess evidence in a systematic review?
BLEADFIPP/CARLOP
- bias
- literature review completed?
- exclusion criteria?
- analysis
- duplication of results (have other studies agreed?)
- funding
- inclusion criteria
- priori design used (authors refer to published protocol/research objectives)
- PICO used
- confounding factors?
- analysis appropriate?
- representative?
- loss to follow up?
- outcomes adequately measured?
- prognostic factors adequately measured?
What are the strengths of a meta-analysis?
- good summary
- large sample size in total = increases power and reliability of results
- saves money and time compared to carrying out a new trial
- comes to conclusions about best treatment
- often multi-centred and multi-national
What are the weaknesses of a meta-analysis?
- can be subject to lots of different types of bias
- results of poorer trials may dilute effect seen from results of positive trials
What are features of a well conducted RCT?
large multinational control group placebo/active control good randomisation well balanced group characteristics blinding appropriate methodology appropriate inclusion/exclusion criteria ITT analysis documenting appropriately/deals with drop outs
What are the 7 types of bias?
CLOP To Some Music
Citation bias
- highly cited, high impact studies are more readily available and included more in studies
Language bias
- studies with high impact and which are highly cited tend to be published in english -> this means important studies in tother languages may not be included in meta analyses
Observation bias
- analysers of a study measuring height may not have accurate measurement tool/may round UP fractionate answers, introducing bias
Publication bias
- studies with positive results tend to be published more than those with negative/neutral results -> means these studies may be omitted from meta-analyses
Time lag bias
- trials with positive outcomes are published sooner -> meaning they dominate the literature until negative resulting trials come out, influences result of meta-analysis
Selection bias
- only individuals who are able to say come to a clinic can take part in a trial (omits those with an exposure but who can’t drive etc.)
Multiple publication bias
- positive trials are often published more than once = they can be included multiple times in a meta-analysis
Give an example of a study which utilised rapid evidence review and its pros/cons:
Burton et al 2017 (published in Lancet) -> rapid evidence review of alcohol-related harm control policies in England.
Pros -> quicker and cheaper then SR
Cons -> more restricted search terms, different inclusion periods for each policy area
Give an example of a study where meta-analysis resulted in changing of current practice:
Bouri et al 2014 -> carried out a meta-analysis for the use of B-blockers started peri-operatively for patients who were high/intermediate risk undergoing surgery/vascular surgery.
Previously the DECREASE set of studies had contained fictitious data and showed B-blockers were beneficial, but this meta-analysis showed that they actually resulted in a 27% increased risk of 30-day mortality -> therefore, guidelines immediately revoked and changed
