Evidence Based Medicine

1
Evidence Based Medicine Semmelweis
University, Budapest, April 2006 Ofer Allan
Avital M.D., CEO, Founder, Evidence Matters
2
Evidence Based Medicine (EBM) is the
conscious, explicit, and judicious use of the
best availalbe evidence to inform and improve
clinical decisions for patient care. Sackett
- 1996
3

• EBM is a process, which integrates
• the clinicians experience and judgement
• facts from the research evidence
• patient values and preferences.
• Without the first point, medicine becomes
  cookbook practice, and is tyrannized by
  external research, which may not apply in a local
  setting or for a particular patient
• Without the second point, there is a risk of
  expertise deteriorating over time, to the
  detriment of the patient
• Without the third point, the patient will be
  unsatisfied or will not co-operate

4
The Type of Question Determines Type of Evidence
If the Question is Relevant Research
Types Diagnostic Transverse Prognos
tic Cohorts Therapy (85) 10 levels
of evidence Systematic reviews
5
Why do we need Evidence-Based Medicine?

• Half of all medical knowledge is found to be
  false every 5 years (Sackett 1997). i.e.
  Half-life 5 years
• Sackett found in a Canadian study that the choice
  of antihypertensive medication was determined not
  by patient factors (e.g. Level of diastolic blood
  pressure, end-organ damage) but by the year of
  graduation of the doctor!
• To use new therapies, clinicians must keep up
  with the scientific literature in a critical
  manner

6
Why is it such a challenge to keep up?

• There are over 10,000 scientific journals
  publishing over 2 millon medically-related
  articles a year.
• For a doctor to keep up in just his/her
  specialty, the doctor would have to read 19
  articles a day, 365 days a year! (Sackett 1997)
• Not all the articles are scientifically valid,
  or clinically relevant.

7
EBM offers a system for continuous
self-education, focused on the needs of your
patients 1. Convert the clinical information
need into an answerable question (P.I.C.O)
2. Retrieve a range of evidence in a systematic
manner 3. Critically evaluate the evidence 4.
Apply the results in clinical practice 5.
Evaluate the results for your patients!
8
Levels of Evidence for Research on Therapies
9
Critically Evaluating Evidence

• Chance How likely are the results due to
  chance? What is the P value p-value
  setting? What are the confidence intervals?
• Systematic errors We can reduce these by
• a) Random assignment of patients to groups
• b) Blinding patients and clinicians to therapy
  given.
• Confounding Variables Try to control these
  with
• a) Randomized therapy assignment,
• b) Precise definitions of variables,
• c) Stratification analysis
• d) Multivariate analysis
• Many more...

10
Basic Bibliography
Daniel Friedland, in his book "Evidence-Based
Medicine. A Framework for Clinical
Practice" David L. Sackett "Evidence-based
Medicine. How to practice and teach EBM"