The importance of research to achieving excellence'

1
The importance of research to achieving
excellence.

• Centre for Health and Wellbeing Research, Sunley
  Management Centre, University of Northampton,
  15th December 2008.
• Professor Mike Kelly,
• The Centre for Public Health Excellence, NICE,
  London, UK

2
The importance of nineteenth century public
health in developing the evidence base

• The early pioneers John Snow, William Tennant
  Gairdner, William Duncan, Edwin Chadwick, William
  Farr.

3
John Snow
4
Gairdner
5
(No Transcript)
6
Farr
7
Chadwick
8
The classic period of public health

• The development of causal models without
  necessarily understanding the mechanisms because
  empirical observation had not yet revealed the
  underlying causes.
• Social deprivation was seen as a cause and as a
  confounder (Chadwick, Gairdner)

9
(No Transcript)
10
(No Transcript)
11
(No Transcript)
12
(No Transcript)
13
(No Transcript)
14

• The statistical patterning of mortality and
  morbidity revealed major geographical differences
  (William Farr)
• The measurement of occupation of head of
  household in the census from 1911 demonstrated
  health inequalities.
• But at the same time there has been a long run
  trend in health improvement.

15
The long run trend
Hi
Mortality
Lo
1940
1800
Time
16
Decline in mortality the middle classes
Mortality
17
Decline in mortality the less well to do
Hi
Mortality
1940
18
Implications

• Causal agents
• Patterning by time, place and person.

19
Key epidemiological studies

• Smoking and lung cancer (Doll Hill 1952)
• Exercise and heart attack (Morris et al 1953)
• Asbestos and lung cancer (Doll 1955)

20
Austen Bradford-Hill
21
Bradford-Hills principles

• Temporal sequence.
• Strength of association.
• Dose response relationship.
• Replication.
• Biological plausibility.
• Alternative explanations.
• Cessation of exposure.
• Coherence with other knowledge.
• Specificity.

22
The randomised controlled trial (RCT)

• To evaluate a specific interventions or
  treatment.
• Individuals are assigned randomly to an
  intervention or control group. This assures
  exposure is unbiased and comparability between
  groups.
• Subjects and medical staff blind to who is
  intervention and who is control

23
(No Transcript)
24
The Early Trials

• Streptomycin for Pulmonary Tuberculosis 1948.
• Whooping cough vaccine 1951

25
Enter Archie Cochrane
26
(No Transcript)
27
Archie Cochranes Principles

• The best care available to all.
• The need for a means to determine what was best
  treatment.
• The importance of rooting out harmful or useless
  practice.
• The necessity of ascertaining costs and benefits.

28
Cochranes solutions

• Health economics
• The randomised controlled trial.

29
The legacy

• The Cochrane and Campbell Collaborations.
• Health economics and the Quality Adjusted Life
  Year (QALY)
• The NHS Centre for Reviews and Dissemination
  (University of York).
• The Cochrane controlled trials register.
• Fast electronic search engines accessing large
  data bases.
• Systematic Review and metaanalysis
• NICE

30
(No Transcript)
31
Principles of building the evidence base for
excellence

• Cumulation of evidence.
• Methodological rigour by reducing bias by
  focusing on internal validity.
• Hierarchies of evidence.

32
The original hierarchy of evidence

• Systematic reviews and meta analysis.
• RCTs.
• Cohort studies.
• Cross sectional surveys.
• Case reports.
• Expert opinion.

33
Applying the evidence based approach beyond
clinical interventions.

• Evidence based public health.
• Where causal relations between the intervention
  and the outcome is distal.
• Complex interventions in complex settings.

34
Evolutionary Trends
. Source The Economist, 12 November 2003.
35

• Where X is a brief intervention in primary care
  to increase rates of exercise and where Y is
  reduced risk of heart attack

36
E
K
I
G
X
Y
D
C
B
A
L
J
H
F
37
Conceptual framework for public health guidance
Organisational vector
Life-course experience
Health Wellbeing
Environmental vector
Population- wide vector
Individual agency- life worlds-capacity for
action behaviour
Socio-cultural Society vector
38
The relationship between interventions and
outcomes

• Long causal chains.
• Importance therefore of developing logic models
  to describe the relationship between
  interventions and outcomes.
• The existing evidence base silent on large tracts
  of the logic models.
• Key points in the logic model involves evidence
  of a type that had never been near an evidence
  hierarchy.
• Two sets of causation - individual and social in
  which the causal chains are equally long

39
Practical problems of applying the existing
hierarchy to public health evidence.

• Very few RCTs to go to the top of the hierarchy.
• External validity at least as important as
  internal validity.
• Need multiple hierarchies of evidence.
• Need to deal with theoretical evidence, models
  and propositions.

40
Conclusion
41
(No Transcript)
42
(No Transcript)
43

• First come I my name is Jowett.
• Theres no knowledge but I know it.
• I am master of this college
• What I dont know isnt knowledge.
• The Masque of Balliol
• Revd. H.C. Beeching

44

• It is very important not to get stuck in a very
  narrow interpretation of what evidence based
  medicine means.
• Must not fall into the trap of assuming the
  evidence speaks for itself.

45
Because

• All evidence requires interpretation.
• Absence of evidence does not mean absence of
  effect.
• Strong evidence of effect may not relate to the
  important issue(s)
• To learn from what we have observed we must
  extrapolate beyond experience to draw out factual
  or inductive inferences from what we have
  observed to what we have not.

46
(No Transcript)
47
(No Transcript)
48
NICE

• The National Institute for Health and Clinical
  Excellence (NICE) is the independent organisation
  in the UK responsible for providing national
  guidance to the NHS and the wider public health
  community on the promotion of good health and the
  prevention and treatment of ill health.