Evidence Based Medicine in the office and hospital

1
Evidence Based Medicine in the office and
hospital

• Daniel J. Van Durme, MD
• Professor and Chair
• Dept. of Family Medicine and Rural Health

2
Who is this guy and what does he know?

• Private practice in semi-rural Pasco county
  (north of Tampa) 1991-1996
• Faculty at University of South Florida College of
  Medicine 1989-1991 and 1996-2004
• Still seeing patients at Madison County Health
  Department
• Creator, Course Director, and lecturer for
  Evidence Based Medicine course at USF COM
• 40 hours of lecture and small group for Med 2s

3
I am NOT an EBM expert

• Expert from Latin
• ex has been
• spurt a drip under pressure
• I struggle with stats
• I am often overwhelmed with the volume of medical
  information and the need to critically review the
  important stuff
• BUT I can still provide high quality Evidence
  Based Care

4
Learning Objectives for today

• At the conclusion of todays session, the learner
  should be able to
• Define evidence based medicine
• Demonstrate an ability to formulate a
  patient-oriented clinical question from a
  clinical scenario
• Discuss appropriate search strategies for finding
  answer(s) to clinical questions
• Demonstrate the use of PDA and computer resources
  for finding high quality evidence based answers
• Discuss how evidence based findings would be
  applied to the care of a patient

5
EBM Original Official Definition

• The explicit, conscientious, and judicious use
  of the current best evidence in making decisions
  about the care of individual patients (and
  populations)

Evidence-Based Medicine Working Group Sackett et
al circa 1996
6
Problems with EBM definition

• EBM has been accused of being . . .
• Cookbook medicine
• It takes away the art of medicine or the
  clinical judgment
• WRONG the research results may not be
  applicable or appropriate for a given patient
• Cost-Cutting medicine
• It is all a plot by managed care companies to
  cut cost of care and increase their profit share
• WRONG - When you find what is best for a given
  patient it may cost more OR it may save money

7
Problems with EBM (cont.)

• EBM has been accused of being . . .
• Impossible or impractical
• There is no way I can spend hours looking for
  and critically reviewing medical articles for
  each of the patients that I see.
• WRONG there are many tools available at the
  point of care (PDAs and computers and texts)
  that can help you find answers in a matter of 1-2
  minutes.

8
Better EBM definition

• The integration of best research evidence with
  clinical expertise and patient values
• Sackett et al 2000

9
(No Transcript)
10
Two Fundamental Principles of true Evidence
Based Practice

• Clinical Decision Making
• Evidence is Never Enough
• a. Treatment of Pneumococcal pneumonia
  SHOULD be different for
• Terminal Cancer Patient
• Elderly, Severely Demented Patient
• Young, mother of 2 children
• b. Importance of Values/Preferences

11
Two Fundamental Principles of EBM

• A hierarchy of evidence
• a. There is a hierarchy of possible
  information
• b. Look for the highest level of evidence
  available
• BE ready to change your approach or management
  when a higher level of evidence contradicts your
  experience
• Experience Based Medicine doing the wrong
  thing with increasing confidence for an
  impressive number of years.

12
Best research evidence

• Clinically relevant not just well-done
  research
• Ideally patient-centered clinical research
• What matters to patients?
• Morbidity, mortality, quality of life
• POEM
• Patient Oriented Evidence that Matters
• Matters to my practice and my patients
• Sometimes disease-oriented evidence (DOE)
• How many irregular heartbeats per hour?
• Can be misleading (sometimes dangerously so)
• Occasionally basic science
• What is the level of C-reactive protein (CRP) in
  the serum?
• Can be VERY misleading

13
Clinical Expertise

• Use of clinical skills and past experience
• Identification of individual patients . . .
• Health status and health risks
• Personal values and expectations
• (Probable) diagnosis
• Knowledge of disease prevalence, access to
  medical or test availability, etc. in your
  community
• Did you ask the correct clinical question(s)?

14
Patient values

• Patient preferences and concerns
• Cultural influences
• Religious/spiritual influences
• Psychosocial issues
• May include . . .
• Reimbursement or insurance status
• Access to care
• Societal factors
• Other influences

15
Why do we need EBM?

• Stay up to date
• Medical information changes constantly
• Unlike bread our knowledge does not become
  visibly moldy or stale we just keep using it

16
Why do we need EBM?

• Save LIVES!
• Encainide and flecainide for ventricular
  arrhythmia
• Well proven to decrease the number of premature
  ventricular beats became widely used 1980s
• BUT
• Further studies showed significant INCREASE in
  MORTALITY died from other cardiac complications
  and dysrhythmias ( a dangerous DOE)
• Thrombolytics for acute MI
• CLEAR evidence of benefit in the 1970s
• Not widely recommended until 1988 almost 13 yrs
  later
• How many thousands of people died unnecessarily
  in the years in between?

17
Why do we need EBM?

• We want to do the right thing what is best
  for our patients
• Practice variations that do not make sense . . .
• Not to doctors
• Not to patients
• Not to payors
• Not to policy makers

18
Assessment of Radical Prostatectomy Time Trends,
Geographic Variation, and Outcomes
Lu-Yao JAMA, Volume 269(20). May 26, 1993.
19
So why not get info from textbooks and review
articles?

• Texts and review articles?
• Dated perhaps by several years
• Often heavily biased
• Author chooses article that he/she agrees with
  (or has written)
• May help more with background knowledge (help me
  learn about disease) not foreground (help me
  answer the specific clinical question for this
  patient)

20
Foreground questions
Background questions
Experienced clinician
Medical student
21
But how does EBM REALLY work?

• Step 1 Translate clinical scenarios into an
  answerable clinical questions
• TRUE STORY
• My 54 yr old patient was just diagnosed with
  prostate cancer
• I received pathology report and he is coming in
  to see me tomorrow

22
What are my questions?

• What do I know about prostate cancer?
• How common is it?
• Is it usually aggressive and rapidly fatal?
• How can it be treated surgery, chemotherapy,
  radiation?
• What about family history what should I tell
  him about his sons risk?
• Etc.
• These are called background questions

23
Foreground questions apply to that specific
patient (or population)

• After meeting with patient and spouse we find
  that he has seen the urologist who recommended
  surgery but the patient is reluctant
• 54 year old male patient was diagnosed with
  intermediate grade prostate cancer and wants to
  know whether to get a radical prostatectomy or
  radiation treatment. He is concerned about death
  from prostate CA and also risks of impotence and
  incontinence.

24
Question?

• Population
• For middle aged males with intermediate stage
  prostate cancer,
• Intervention
• Treated with radical prostatectomy
• Comparison
• Compared to radiation treatment
• Outcome
• What are the rates of incontinence, impotence and
  cancer-related mortality?

25

• Developing the question requires
• Some background knowledge of the condition
• Understanding of the patient and what are the
  outcomes that matter in this patient
• Death?
• Disability?
• Quality of life? Anxiety, Impotence, etc.
• Cost?

26
Hands on Part 1

• Think in your practice THIS week what was a
  clinical question you had?
• Think of a foreground question (not just a drug
  dose or drug interaction)
• What diagnostic test would have been best for
  that pt with abdominal pain?
• What treatment would have been best for the pt
  with Parkinsons?
• What about the patient who was asking about
  acupuncture for osteoarthritis?

27
How does EBM REALLY work?

• Step 2 Translate question into effective
  searches for the best evidence
• Requires knowledge of medical informatics
• How to search what terms to use, what types of
  studies, etc.
• Where to search utility of varied sources of
  information
• Evidence based sources, Texts, Medline,

28
Purpose-specific resources

• CDC Travel
• Drug information resources
• Patient Education handouts
• Medical Search engines
• Textbooks
• Journals

29
EBM sources

• EBM sources Cochrane, USPSTF, Clinical Evidence
• Ideally best information source hard to argue
  with, will explicitly state the level of evidence
  (weak to strong)
• - There may not be any good evidence

30
How does EBM REALLY work?

• Step 3 Critically appraise the evidence
• Validity of the evidence
• Internal study design, blinding, randomized,
  sample size, appropriate statistics, etc.
• Relevance of the evidence
• Did they measure something pts care about?
• Is population similar (enough) to mine?
• Is the intervention feasible?
• Importance of the evidence
• Magnitude of effect or clinical significance?
• P values, confidence intervals, relative risk or
  absolute risk reduction

31

• Step 3 Critically appraise the evidence (cont.)
• Requires some knowledge of basic epidemiology and
  biostatistics
• Sensitivity, specificity, prevalence, likelihood
  ratios
• Absolute risk reduction, relative risk reduction,
  odds ratios, number needed to treat
• Requires knowledge of study types
• ASSUMING THAT IT IS A WELL DESIGNED STUDY
• Appropriate sample size, randomization, stats,
  treatment allocation, etc., etc.
• Meta-analysis of RCTs gt RCT gt Cohort gt Case
  Control gt Case Series gt Case Report

32
Hierarchy of studies
33
Step 3 Critical appraisal of medical literature

• This is often confused with EBM
• they are not the same thing
• This is often the toughest part of EBM
• Skipped by many doctors suffering from
  photonumerophobia
• The fear that ones fear of numbers and
  statistics will come to light
• This is where most attempts come to a halt
• Not enough time and expertise

34

• EBM Databases
• Systematic Literature Searches
• Cochrane Library (OVID)
• Clinical Evidence
• Systematic Literature Surveillance
• ACP Journal Club (OVID)
• DARE
• DynaMed
• Medical InfoRetriever
• Journal of Family Practice POEMS
• EMB Search Engine
• TRIP Database

35
Drilling for the Best Information
36
Cochrane Library

• The current resource with the highest
  methodological standards
• For each clinical question, all of the English
  literature meticulously searched for randomized
  trials
• Large systematic reviews with valid methods
  collaborative effort
• Conclusions are based on all the evidence from
  valid randomized trials

37

• Cochrane Library
• Included in OVID subscription
• Limitations
• limited to English
• only addresses questions amenable to randomized
  trials
• most of medicine has not been studied enough to
  allow for conclusions
• 235/year or abstracts only

38
InfoRetriever

• 104 journals surveyed for Evidence-Based Practice
  Newsletter
• Over 1300 article synopses/ POEMS
• Cochrane abstracts
• Selected evidence-based guidelines (USPSTF, CDC,
  others)
• Basic drug info
• ICD-9 codes
• Clinical calculators/prediction rules

39
InfoRetriever Symbols
40
InfoRetriever

• Comes in web, desktop and PDA versions
• Explicitly states Levels of Evidence
• Limitations
• individual article summaries may not account for
  the big picture
• may have to read multiple items
• 249/year
• Optimized for use with Internet Explorer 5.x or
  Netscape 6.x

41
Hands on with InfoRetriever

• 1. Look up migraine
• 5 min clinical consult level 5 evidence
• Background info
• Overview practice guidelines
• ACEP guidelines for ED
• Tx Drug treatment anticonvulsants?
• Note symbols for Cochrane database or InfoPOEM
• Info available on CAM, screening, Pt ed, etc

42
Levels of Evidence

• Level 1 Randomized Clinical Trials
• Level 2 Head to Head Trial or
  Systematic Review of Cohort Studies
• Level 3 Case-Control Studies
• Level 4 Case-series
• Level 5 Expert Opinion

43
Guidelines

• What is a guideline?
• Guidelines may be
• Explicitly evidence-based
• Evidence-based
• Research-based (highly referenced)
• Opinion-based
• expert consensus

44
Guidelines

• National Guideline Clearinghouse
• Primary Care Clinical Practice Guidelines
• Agency/Association sites
• AAFP
• AAP
• ACS

45
Clinical Evidence

• BMJ
• Summaries of Evidence
• Specific clinical questions treatment
• Makes specific recommendations
• States when there is a lack of evidence
• Free from United Health Foundation

46
Clinical Evidence
47
Hands On with Clinical Evidence

• Look up Stroke Prevention in Clinical Evidence
• Beneficial control BP and cholesterol and give
  aspirin
• Unknown other antiplatelet agents showed no
  benefit over aspirin
• Ineffective or harmful anticoagulant for those
  in sinus rhythm carotid endarterectomy for
  those with lt30 symptomatic stenosis

48
Hands on (POSSIBLE example or use your own!)

• Patient wants to know if Gingko biloba will help
  her moms Alzheimers
• See InfoRetriever dementia
• Treatment Complementary and alternative
  medicine
• Mixed results in InfoPOEMS some say maybe yes,
  some say no
• Cochrane says it seems safe, but studies are
  weak, we really do not know more study is
  needed
• See Clinical Evidence - dementia

49
ACP Journal Club

• About 100 journals systematically surveyed
• Highest-validity articles abstracted
• Structured abstracts to guide critical appraisal
• Clinical commentary
• Included in our OVID subscription

50
ACP Journal Club

• Limitations
• individual article summaries may not account for
  the big picture
• may have to read multiple items
• No control over what is covered
• 78/year ?

51
Need to read the key

• Levels of Evidence
• Level 1 Highest
• Level 2
• Level 3
• Level 4
• Level 5 Lowestbut still evidence

52
Read the key

• Levels of Recommendation (USPSTF)
• A Highest Strongly recommended (PAP smears)
• B Recommended (Mammograms age 40)
• C no recommendation for or against (too close a
  balance between harm/benefit) (osteoporosis
  screening below age 60)
• D Recommend AGAINST (ovarian cancer)
• I insufficient evidence to make any
  recommendation for or against (Prostate cancer
  screening)

53
Other guidelines

• A good evidence
• B fair evidence
• C based on expert opinion and/or consensus
• X evidence of harm

54
Essential principle

• Be ready to surrender to a higher level of
  evidence when it becomes available
• Do not become entrenched in what has been done
  for years
• A bad idea done by a LOT of people for a LONG
  time, is still a bad idea

55
Evidence based information, recommendations,
reviews

• Not all that claims to be evidence based, is
  really EBM
• Should include explicit statements about search
  methods, findings, appraisal and level of
  evidence (or strength of recommendation)
• High quality sources
• Cochrane, AHRQ, USPSTF, ACP Journal Club,
  Clinical Evidence, InfoRetriever
• Questionable sources
• Developed by BOGSAT methodology
• Bunch Of Guys Sitting Around a Table
• Sometimes called consensus, argument may be won
  based on volume and stamina

56
How does EBM REALLY work?

• Step 4 Implement information into practice
• Integrate information with patients values and
  preferences
• Patient-centered care

57
How does EBM REALLY work?

• Step 4 Implement information into practice
• Integrate information with patients values and
  preferences
• Patient-centered care
• Demographics, age, socioeconomics, fear, etc.
• Evidence may point to surgery as better treatment
  but patient refuses
• This does NOT mean EBM is out the window
• Your job is to understand the magnitude of
  benefit and the level of evidence
• Then translate into useable information for the
  patient

58

• As patient participates in care decisions, you
  are practicing TRUE evidence based medicine

59
OK I am convinced, how can I start to practice
evidence based medicine?

• Step 1 Ask the questions
• Use your clinical experiences to find 1-2 case
  scenarios every day that translate into clinical
  questions
• Ask your student to help he/she may REALLY
  appreciate that you explicitly tried to help find
  a REAL answer that would help an actual patient
• PICO population intervention comparison -
  outcome

60

• Use your growing clinical skills but do not be
  swayed by YUCKs
• Your Unsubstantiated Clinical Knowledge (and
  experience)
• Regularly seek to find the best available
  evidence to guide you
• Especially review common topics, you may be
  getting stale without realizing it

61

• Step 2 Search for the evidence
• When searching for background information
• Critically appraise your texts for known
  problems/biases
• Date of publication, references, source
• Try to use systematic review articles
• Explicit statements of how and where they
  searched, and statements of strength of
  recommendation of level of evidence

62

• Step 2 searching (continued)
• Locate and regularly use YODAs
• Your Own Data Analyzer
• Let others do hard work for you
• It is their full-time job, do you really have the
  time and expertise to do better?
• Try InfoRetriever and Clinical Evidence a few
  times a week
• Save your questions on a card and find answers
  over lunch or end of the day
• Look at the Cochrane reports first
• BUT even those may be dated!

63

• Step 3 Critical Appraisal
• Do not fall for three common myths
• Newer article, by bigger name, and a famous
  journal, does NOT mean it is better
• Use three quick tips
• Is it relevant first?, dont get overwhelmed by
  the stats, was it from YODA?
• PRACTICE CRITICAL APPRAISAL of original research
  if you do not use it, you will lose it
• I often let headlines drive this then I need to
  know NOW

64

• Step 4 Integrate into patient care
• Take your findings back to your patients
• Sometimes this may be 2 minutes later or 2 weeks
  later
• Discuss how to integrate this into care of your
  patient
• Tell the patient that you have been looking up
  the latest information and they will appreciate
  it!

65

• Step 5 Self evaluation how did you do?
• Learn to improve your . . .
• Framing of the question
• Search terms
• Search locations
• Critical appraisal skills
• Patient understanding
• Patient centered approach

66
Questions?