Interpreting results

1

• Interpreting results

Behnam Shakiba
2
Starting at the end .

• We know that many people who read a review begin
  at the end, by looking at the conclusions and
  discussions, often, the brief conclusions in the
  abstract.

3

• The purpose of Discussion and Conclusion section
  of your review is for you to help the reader
  interpret your results and make a decision.

4
Three main things you need to cover

• Any limitations of your review
• The strength of the evidence
• The applicability of the evidence

5
Strength of the evidence

• internal factors
• external factors

6
internal factors

• Methodological issue
• adequate allocation concealment
• careful control for confounding
• little missing data

7
2.The number of studies in your review and the
number of participants in the studies

• Study A 3 trials with total of 256 participants
• Study B 21 trials with total of 2343
  participants
• The strength of evidence in study B is
  greater.

8
3. The size of the treatment effect

• Study A OR0.31 0.24,0.42
• Study B OR0.52 0.29,0.93
• The strength of evidence in study A is greater.

9
4. The precision of the treatment effect

• Study A CI 0.29,0.38
• Study B CI 0.25,0.73
• The strength of evidence in study A is greater.

10
5.The consistency of the outcomes

• study A
• The strength of evidence in
• study B is greater.
• 
  
  
• 
  studyB

11
6.The consistency of results

• Study A
  Study B
• The strength of evidence in study A is greater.

12
7.Apparent dose responserelationship

• Multiple doses appear to have a greater effect
  than single doses, which may strengthen the
  evidence.

13
External factors

• Biological plausibility
• Other evidence
• related reviews

14
Applicability of the evidence

• the results of your review may be used in many
  different populations and settings around the
  world.

15
Important variations

• Biological and cultural variation
• Variation in compliance
• Variation in baseline risk

16

• The Discussion section
• interpreted strength of clarification
• the results evidence trade-offs
• making a decision

17

• Try to draw some discussions

18
Identifying all the possible outcomes

• positive effects
• adverse effects
• cost of treatment

.
19
Drawing up a balance sheet

• beneficial harms
• effects
  effects

20
The example

• Treatment of a sore throat with Antibiotics

21
The potential positive effects of treatment
include

• Reduction in severity of illness
• Reduction in pain
• Reduction in the duration of illness
• Reduction in infections following on (such as
  sinusitis, and acute otitis media)
• Reduction in subsequent non infective
  complications (such as rheumatic fever and acute
  glomerulonephritis)

22
The potential negative effects of treatment
include

• Adverse effects of antibiotics (such as
  diarrhea).
• The cost to society in terms of antibiotic
  prescription and overuse

23
What if there are no data available on some
outcomes?

• This should be clearly stated in your review.
• Use other types of data
• Recommendations for future research

24
What do users of reviews need to know about these
outcomes?

• Absolute measures is often more meaningful to
  users than relative measures.
• The natural history of a disease may be
  important.
• How common the outcome is in their particular
  setting .

25
Try to avoid value judgments

• Value judgments are any statements where you make
  assumptions about the value placed on particular
  outcomes.
• Would everyone agree with this statement?

26
The results of studies included within a review
may vary

• age
• gender
• presence of some biochemical marker
• If these differences might be clinically
  relevant they would be worth highlighting when
  discussing the evidence in your review.

27
Other useful information

• Size of a particular health care problem
• Current clinical practice
• Epidemiological data
• Cost

28
It would be good if we could provide users of a
review with all the information they require to
make a decision , this is rarely possible.
29
A final check .

• Have all the main outcomes been considered?
• Have I considered any factors that might limit
  the application of these results in different
  situations?
• Are these results consistent across the
  included studies or do they vary for some reason?
• Have I avoided making value judgments about how
  to interpret my findings?

30

• Try to draw some conclusions

31
Whose conclusions?

• Users in different settings will often interpret
  the same evidence from the same review quite
  differently.

32
Usually there are three aspects to drawing
conclusions

• Does the intervention work at all?
• What have we learned from this review that can
  be applied to clinical practice?
• What have we learned from this review about the
  need for further evaluation and research?

33
What are the implications for clinical practice

• beneficial harms
• effects
  effects
• trade-offs
• reflect this
• in conclusion

34
What are the implications for further research

• While there may sometimes be a need for more
  research in order to gather sufficient evidence.
• The further research needs to be targeted to
  specific issues that have arisen out of the
  review.

35
(No Transcript)
36
common mistakes frequently made by reviewers

• Evidence of no effect versus no evidence of
  effect
• Evidence-conclusion mismatches
• Further research is needed

37

• Any Questions ?