Evidence-Based Practice 奇美醫學中心 林宏榮
What evidence-based medicine is: “Evidence-based medicine is the integration of best research evidence with clinical expertise and patient values” - Sackett, et al 2001
Clinical Expertise Best Evidence What evidence-based medicine is Patient Values
Rule 31 – Review the World Literature Fortnightly* *"Kill as Few Patients as Possible" - Oscar London 5,000? per day 1,400 per day 55 per day
The Airline industry Boeing 777 manuals 24 binders 10 feet shelf space Conversion to CD Reduced search by 60% The Health Industry Memorize “the manuals” Exams, audits, etc to check Managing Information
Systematic review of bed rest after medical procedures • 10 trials of bed rest after spinal puncture • no change in headache with bed rest • Increase in back pain • Protocols in UK neurology units - 80% still recommend bed rest after LPSerpell M, BMJ 1998;316:1709–10 • …evidence of harm available for 17 years preceding... Allen, Glasziou, Del Mar. Lancet, 1999
Getting Evidence in to PracticeHow do you “do” EBP? • What EBP do you do/help with? • What other EBP do you know of? • Compare with you neighbour Teaching Tip: Special background for activities.
Managing Information“Push” and “Pull”methods • “Push” - alerts us to new information • “Just in Case” learning • Use ONLY for important, new, valid research • “Pull” – access information when needed • “Just in Time” learning • Use whenever questions arise • EBM Steps: Question; search; appraise; apply
Bimonthly “just in case” journalValid, Relevant & (almost) No Effort! • 80 journals scanned • Is it valid? • Intervention: RCT • Prognosis: inception cohort • Etc • Is it relevant? • GPs & specialists ask:Will this change your practice? www.evidence-basedmedicine.com
“Just in Time” learning:Doctor’s information needs • Setting: 64 residentsat 2 New Haven hospitals • Method: Interviewed after 401 consultations • Questions • Asked 280 questions (2 per 3 patients) • Pursued an answer for 80 questions (29%) • Not pursued because • Lack of time • Forgot the question • Sources of answers • Textbooks (31%), articles (21%), consultants (17%) Green, Am J Med 2000
Doctor’s information needs • Most of our questions are NEVER answered • When answered, the information is likely to be neither the best nor up-to-date
Your Clinical Questions • Write down one recent patient problem • What was the critical question? • Did you answer it? If so, how?
Good questions • Important to your practice • Important to your patients • Specific • Answerable!
Good Questions • Which patients is this question about? • What is the main intervention? • Is there an alternative intervention? • What can I hope to accomplish?
“Hunting” questions - “PICO”: • “P” - patient or problem • “I” - intervention (e.g., diagnostic test, treatment, cause, prognostic factor) • “C” - comparison intervention (if necessary) • “O” - outcome
Examples of good questions • In patients with insulin-dependent diabetes mellitus • receiving current standard insulin therapy • will an intensive insulin regime • reduce the risk of developing microvascular complications
Examples of good questions • Among women in premature labour expected to deliver before thirty weeks of gestation • does an intensive corticosteroid regime • compared with the standard regime • reduce the risk of RDS in their babies?
Information “pull”Steps in EBM process • Formulate an answerable question • Track down the best evidence • Critically appraise the evidence • Integrate with clinical expertise and patient values
An example: “the first sign of hyperkalaemia is death” • An anxious laboratory technician phoned about a potassium of 7.3 mmol/l (Ref Range 3.5-5.0) found on a routine blood test of a 50 year old woman. • I arranged an urgent repeat of the electrolytes (to rule out a spurious elevation) and an ECG. • The latter was reassuringly normal, but left me asking: Does a normal ECG rule out a serious elevation of potassium?
1. The question • Does a normal ECG rule out a serious elevation of potassium? • Population - In suspected hyperkalemia • Indicator - does a normal ECG • Comparator - • Outcome - rule out hyperkalemia?
1. The question • Does a normal ECG rule out a serious elevation of potassium? • Population – hyperkal* • Indicator – ECG OR EKG • Comparator - • Outcome – hyperkal* • Underline keywords; think of synonyms
Step #2Efficiently track down the best evidence to answer clinical questions
Useful data sources MEDLINE Cochrane Library Clinical Evidence
searchable through Medline
Using the tools • NLM (who make Medline) index thousands of medical journals • Each article is given keywords - • Major MESH terms • Minor MESH terms • The article title and abstract are also searchable - as Textwords
Using the tools • Search engines will sometimes match your entry to the nearest MESH term. • Sometimes they don’t • Experiment!
Filters • A filter is a sequence of Medline search instructions intended to locate specific types of study design • Filters exist for • clinical trials • studies of prognosis • studies of adverse effects • and many others….
Filters • Some search engines provide prepackaged filters • PubMed for example • Most don’t
PubMed via Google Diagnosis button “OR” synonyms * Means any letters
Limit to EBM Reviews
Most Recent Update
The “best” evidence depends on the type of question • What are the phenomena/problems? • Observation (e.g., qualitative research) • What is frequency of the problem? (FREQUENCY) • Random (or consecutive) sample • Does this person have the problem? (DIAGNOSIS) • Random (or consecutive) sample with Gold Standard • Who will get the problem? (PROGNOSIS) • Follow-up of inception cohort • How can we alleviate the problem? (INTERVENTION/THERAPY) • Randomised controlled trial
Treating hyperkalemia • She refused to go to hospital • Resonium A, but it is around $100 (RPBS but not PBS) which she could not afford. • My search had mentioned albuterol as a treatment.
Dave Sackett “Just in Time” learningThe EBM Alternative Approach • Shift focus to current patient problems(“just in time” education) • Relevant to YOUR practice • Memorable • Up to date • Learn to obtain best current answers
Advanced threshing • Read the abstract • Read the author list • Read references cited in several other papers • Consider levels of evidence • (as far as you can from abstracts)
Internal validity • Is the study credible? • Was it done welll? • Was it done right? • Do you believe the authors? • Is the study good enough to consider making decisions based on its results?
Levels of evidence • Randomised controlled trials • Cohort studies • Case-control studies • Routine data hunting • Case series • Case reports • Allow for serendipity
Quality of evidence • Use Sackett’s guidelines for the various different types of study • Gain experience • Quality assessment is quite subjective, no matter how experienced you are • Allow for serendipity