EVIDENCE-BASED MEDICINE Dr. Nigel Hart Clinical Research Registrar Dept. of General Practice QUB.

EVIDENCE-BASED MEDICINE

Dr. Nigel HartClinical Research RegistrarDept. of General Practice

QUB

OBJECTIVES

Define the term EBM

Understand the components involved in

EBM

Describe how to develop an ‘answerable

question’

List several methods of how to evaluate

your performance

List several resources of evidence

How do we make clinical decisions?

Dogma

Policy

Experiential

Whimsical

Nihilism

This is the best way to do it

This is the way we do it around here

This way worked the last few times

This way might work

It doesn’t really matter what we do

Rule of least chagrin

Expert deferential

Patient deferential

Do what you will regret the least

What would you do?

How would you like us to proceed?

Pet owners How much can you afford?

Schools Of Thought……..

What is EBM?

‘Evidence-based medicine is the conscientious, explicit and

judicious use of current best evidence in making decisions about the care of individual

patients.’

Sackett et al. BMJ 1996

Traditional Approaches/New Paradigms

• Traditional Approach

– The amount of (quantity) clinical experience is a good measure of the clinical acumen (quality) exhibited by a practitioner

– Knowledge of pathophysiology is essential for making effective decisions

– Didactic medical training and clinical intuition are sufficient

New Paradigm

• EBM Approach

– Clinicians should strive to apply results from systematic, robust,

reproducible, randomized and controlled research to augment and

improve their certainty toward decisions about diagnostic tests,

treatments and assessment of outcomes.

– This approach will allow practitioners of all stages (students to

experts) to come to the same conclusions because they use the same

rigorous approach to obtain answers. It ‘evens the playing field’.

– Understanding that you have not in the past, do not now, or will not

in the future have all the information you need to answer every

question. It’s about identifying your information needs and knowing

how to find and formulate it in the proper context.

– Apply this knowledge consistently and objectively

Why Bother?

• New evidence is being produced every day which could make major changes in the way that we care for our patients (NB – HRT).

• Clinical governance / Revalidation.• Without keeping up-to-date our clinical

performance may deteriorate.• EBM has been shown ultimately improve

health care.• GMC regulations.• General public have greater access to health

care information eg the web, media etc.

How many minutes a week did you spend last week reading around your patients?

• Medical students

• PRHOs

• SHOs

• SpRs

• Consultants

How many minutes a week did you spend last week reading around your patients?

• Medical students

120mins

• PRHOs

• SHOs

• SpRs

• Consultants

• Medical students

120mins

• PRHOs

10mins

• SHOs

• SpRs

• Consultants

How many minutes a week did you spend last week reading around your patients?

• Medical students

120mins

• PRHOs

10mins

• SHOs 30mins

• SpRs

• Consultants

How many minutes a week did you spend last week reading around your patients?

• Medical students

120mins

• PRHOs

10mins

• SHOs 30mins

• SpRs

45mins

• Consultants

How many minutes a week did you spend last week reading around your patients?

How many minutes a week did you spend last week reading around your patients?

• Medical students

120mins

• PRHOs

10mins

• SHOs 30mins

• SpRs

45mins

• Consultants

60mins

The important elements of EBM?

‘EBM begins and ends with patients’

EBM – The search algorithm

1)Convert patient health care needs into

answerable questions.

2) Track down the best evidence.

3) Critically appraise evidence.

4) Apply the results into clinical practice.

5) Evaluate your performance.

The important elements of EBM?

‘EBM begins and ends with patients’

EBM – The search algorithm

1)Convert patient health care needs into

answerable questions.

Asking clinical questions that you can answer….

‘Answerable questions are the backbone of practising EBM’

In practice, good questions usually include:- ‘PICO’

• Patient’s clinical needs

• Intervention or exposure

• Comparison intervention (if appropriate)

• Clinical Outcome(s) [diagnosis/screening, prognosis, therapy, event, harm, or prevention]

The important elements of EBM?

‘EBM begins and ends with patients’

EBM – The search algorithm

1)Convert patient health care needs into

answerable questions.

2) Track down the best evidence.

Tracking down best evidence…..

General search strategy:-

• Define searchable question

• Select evidence resource

• Design search strategy

Tracking down best evidence…..

Hierarchy of evidence:

• Systematic review• Meta-analysis• RCT (DB/SB)• Cohort study• Case controlled

study• Cross sectional study• Case series

The important elements of EBM?

‘EBM begins and ends with patients’

EBM – The search algorithm

1)Convert patient health care needs into

answerable questions.

2) Track down the best evidence.

3) Critically appraise evidence.

Critically appraising the evidence….

The important elements of EBM?

‘EBM begins and ends with patients’

EBM – The search algorithm

1)Convert patient health care needs into

answerable questions.

2) Track down the best evidence.

3) Critically appraise evidence.

4) Apply the results into clinical practice.

Applying the results to clinical practice….

• John, a retired 78yr old

gentleman is your last patient

of the day. He has a long

history of AF and is on aspirin.

He says to you ‘my friend has

atrial fibrillation and he is on

warfarin - should I be on

warfarin?’

Element Example

Patient In patients with AF and who are >65….

Intervention ….would adding

warfarin….

Comparison ….when compared to

aspirin ….

Outcome ….lead to lower

mortality or morbidity

Is this enough to be worth the increased

risk of bleeding?

Example: Atrial Fibrillation

Stroke Risk with AF

Example: Atrial Fibrillation

Stroke Risk with AF on WarfarinStroke Risk with AF on Aspirin

Can you apply this evidence into clinical practice?

• Can the results be extrapolated to your

patient?

• Availability of tests/treatment

• Affordability of tests/treatment (NB NICE

etc)

• Are there adverse risks?

• Are there alternatives?

• What are the patient preferences?

Example: Atrial Fibrillation

• Major bleeding risk

– Population prevalence: 10/1000

– Aspirin: Not significantly different to placebo

– Warfarin: 15/1000

• Are you going to give him warfarin?

• Your patient is very unsteady on his feet and is prone to falls

• Are you still going to give him warfarin?

The important elements of EBM?

‘EBM begins and ends with patients’

EBM – The search algorithm

1)Convert patient health care needs into

answerable questions.

2) Track down the best evidence.

3) Critically appraise evidence.

4) Apply the results into clinical practice.

5) Evaluate your performance.

Evaluate Performance

Audit

Clinical efficacy

Cost analysis

Patient surveys

Prescribing / referral rates

Mortality / morbidity rates

The important elements of EBM?

‘EBM begins and ends with patients’

EBM – The search algorithm

1)Convert patient health care needs into

answerable questions.

2) Track down the best evidence.

3) Critically appraise evidence.

4) Apply the results into clinical practice.

5) Evaluate your performance.

The evidence resources

1. Systematic review databases:

• The Cochrane Library:

– Best source for structured, systematic reviews

– Explicit search & quality criteria (free abstracts)

• Database of Abstracts of Reviews of Effectiveness (DARE/CRD)

– Structured abstracts of systematic reviews meeting highest quality

standards

– A (free) part of the Cochrane Library

• Evidence-Based Medicine Reviews (EBMR) integrates the above with

secondary journal content:

– ACP Journal Club / Best Evidence

– Medline searching and some full-text journals.

The evidence resources…….contd

2. Secondary Journals:

• Evidence-Based “secondary journals” provide structured abstracts with balanced commentary

• Selected from high quality publications

• Can be best resource to start with when investigating rare clinical conditions (saves time).

• Example

– Evidence-Based Medicine, ACP Journal Club

The evidence resources…….contd

3. EBM guidelines and textbooks:

• Clinical practice guidelines

– The best sources rate the strength of evidence

– Important one in the UK – SIGN guidelines

• SIGN

– Always consider external validity to your patient

• Evidence-based textbooks

– Least detail, but most efficient source for simple clinical queries

– Clinical Evidence is the best (explicit protocols)

– Good exceptions: UpToDate, Scientific American Medicine

The evidence resources…….contd

4. Medline/Electronic databases:

• If the Cochrane Library has not addressed our question, next try

Medline…

• The largest biomedical literature database, but:

– Misses some journal articles, misclassifies others

– Can be overwhelming if not searching selectively

• PubMed is a free Medline service with Boolean and other search

capabilities.

• Clinical query filter facilitates clinical searches

– Short Medline strategy and a link to 12 tips

– Sample clinical query here

The evidence resources…..contd

5. The least reliable:

• Colleagues, experts, & “throw-away” journals

– Convenient and fast

– Often invalid , incomplete, and biased information

• Textbooks (usually)

– Generally not systematically researched

– Usually based on “expert opinion”

– Most are out of date - check for recent citations

Some final thoughts……..

Evidence Based Medicine

• The Goal: – To provide exemplary care for our patients using the current

best evidence• The Issues:

– Minutes per ‘patient encounter’ to seek out best evidence– Minutes to hours per week of reading to ‘stay current’ – Research data is accruing faster than our ability to read!

• The Challenge: How do we best make use of our limited time for…– Information retrieval and ‘reading time’ – Information management– Evidence-based medical practice

Thanks for your attention……..

Any Questions…………….