Evidence Based Practice for Beginners

1
Evidence Based Practice for Beginners

• An Introduction to EBP
• Joanne Oakes, Clinical Effectiveness Support
  Officer
• North Tees PCT

2
Workshop Structure

• Background research and EBP
• What is the PICO structure and how is it useful?
• Making sense of research papers how to appraise
  a paper and work out what the numbers really
  mean.
• Interactive and Informal (and at times Intense so
  be warned!)

3

• Lets start with a quiz!
• (You may think this is a bit cruel at first but
  youll see the reasoning behind it later...)

4
Early research James Lind and Scurvy

• In 1747 James Lind performed an experiment to
  determine what might be effective as a cure for
  scurvy.
• Lind divided his 12 men into 6 groups of two. All
  12 shared a common diet for breakfast, lunch and
  dinner but each group received a different
  supplement ranging from half a pint of sea water
  daily to 2 oranges and 1 lemon daily.

5
Randomised Controlled Trials (RCTs)

• Early randomised controlled trials were carried
  out by farmers in the former Yugoslavia
• Seeds were randomised to different conditions
  (e.g. receiving a particular fertiliser, more or
  less water, a control group etc) in a controlled
  environment.

6
Different types of Research Methods
Systematic Review
Meta-Analysis
Randomised Controlled Trial
Cohort Study
Quantitative
Case Control Study
Descriptive Studies
Qualitative
7
What is Evidence Based Practice?
8
What is Evidence Based Practice?

• Evidence-based practice (EBP) is the integration
  of best research evidence with clinical expertise
  and patient values.

9
Implementing Research Findings in Practice (EBP
Cycle)
Need for a decision
Implement findings
Information need
Formulate answerable question
Appraise for validity and usefulness
Find evidence from research
10
Example of an information need.

• Mrs Itchy has three children Bobby, 6, Billy, 9
  and Belinda, 13. Bobby keeps coming home from
  school infected with head lice and passing them
  on to his siblings and parents. Mrs Itchy is
  finding that its costing her a small fortune for
  expensive treatments from the chemist. She
  wonders whether buying a nit comb would be a
  better alternative.

11
PICO

• P Patient/ Population/ Problem
• I Intervention
• C Comparison/ Control
• O Outcome
• Exercise create a PICO structure for the head
  lice example.

12
Finding the information

• Electronic databases
• Journal Articles (electronic and paper)
• Internet
• Courses are available for people whod like to
  further develop their searching skills and
  theyre FREE!

13
So weve found 3 papers which is best?

• Exercise
• We have carried out a search and tracked down
  three papers on head lice that look interesting.
  Look at their PICO structures and see if any of
  them warrant further investigation.

14
Appraising the paper

• Validity
• Were patients randomised into the different
  treatment groups?
• Were all of the patients who started the trial
  accounted for at the end of the trial? (Was
  anyone not followed up? Were patients analysed
  using the intention to treat principle?)
• Were patients, health workers and study personnel
  blind to treatment?
• Were the groups similar at the start of the
  trial?
• Were the groups treated equally apart from the
  intervention?

15
What do all of the numbers mean?
16
Control Event Rate and Experimental Event Rate

• CER No of events in the control group (e.g
  deaths)
• No of patients in the control group
• EER No of events in the experiment
  group
• No of patients in the experiment group
• Exercise work out the CER and EER of being
  cured from head lice from the head lice paper
• CER 31/40 0.775 78 EER 12/32 0.375
  38

17
Absolute Benefit Increase

• ABI Absolute difference between the CER and the
  EER
• (i.e. CER - EER)
• (Alternatively called the Absolute Risk Reduction
  (ARR) when looking at a negative outcome)
• Exercise work out the ABI from the head lice
  paper
• 78 - 38 40. This means that if 100 people
  used malathion instead of bug busting, 40 would
  better off.

18
Number Needed to Treat

• We know from the ABI that treating 100 people
  with malathion instead of bug-busting will see 40
  of them better off.
• So how many do we need to treat to see 1 person
  better off?
• We use the calculation 100/ABI NNT
• (If you remember algebra from school its just a
  case of dividing both sides by the ABI i.e if 100
  ABI then 100/ABI ABI/ABI 1)
• Exercise work out the NNT for the head lice
  paper
• 100/40 2.5. Always round the NNT up 3. That
  means that we need to treat 3 people with
  malathion to see one extra person head lice free
  seven days after the end of the treatment.

19
Absolute and Relative
75p
1
Has only gone up by 25p (absolute increase) but
25p is a third of the original price so you could
say that the price has gone up 33 (relative
increase).
20
Relative Benefit (RB) and Relative Benefit
Increase (RBI)

• The relative benefit (or relative risk if youre
  looking at a negative outcome) looks at your
  chance of having the outcome in the experimental
  group compared with your chance of having the
  outcome in the control group.
• RB EER/CER
• RBI 1-RB
• Exercise work out the RB and the RBI for the
  head lice paper
• RB 38/78 0.487 49 RBI 1-0.489 51
• This means that you are increasing the relative
  benefit by 51

21
It sometimes helps to think of it in terms of a
chart.
Relative Benefit Increase
22
Last bit of stats!

• Confidence Intervals and Statistical
  Significance.
• A 95 confidence interval means that you can
  expect the true result to lie within the given
  limits 95 times out of 100.
• If the 95 confidence intervals crosses 1 then
  the result is NOT statistically significant (this
  is because a value of 1 means that there is no
  difference between the interventions).
• e.g 3.2 (0.6 to 7.1).

23
Transferability and Patient Values

• Can the results be applied to the local
  population?
• Were all clinically important outcomes
  considered?
• Does the regimen suit the individual patient?

24
Implementation something to think about

• Even when you do have a good research base it
  often takes an age to implement the findings!
• Going back to the James Lind example, although
  his research was carried out in 1747, it wasnt
  until 1795 that lemon juice was issued to sailors
  which banished scurvy from the fleet.
• And even in the modern day we havent got all
  that much quicker! Consistent research on
  Thrombolisis has been around since at least 1970
  but weve only really starting implementing it
  over the past few years.

25
Quiz Time Again!

• Now lets work out your ABSOLUTE improvement and
  your RELATIVE improvement from the two different
  test results.

26
How do I find out more?

• Visit www.teesebp.net or try reading one of the
  following

Evidence Based Medicine Toolkit Douglas Badenoch,
Carl HeneghanBMJ Books November 2001
How to Read a Paper The basics of evidence based
medicine Trisha GreenhalghBMJ Books November
2000
Evidence Based Medicine, Second Edition How to
Practice and Teach EBM David L. Sackett
(Editor)Churchill Livingstone November 1999
27