Clinical and Experimental Optometry 90.4 July 2007 © 2007 The Author
Journal compilation © 2007 Optometrists Association Australia 229
CLINICAL AND EXPERIMENTAL
OPTOMETRY
Clin Exp Optom 2007; 90: 4: 229–231 DOI:10.1111/j.1444-0938.2007.00132.x
Mark G Coulthard MB BS FRACP FJFICM
Paediatric Intensive Care Unit, Royal
Children’s Hospital, Herston, Brisbane
QLD
The health-care professional-patient inter-
action is a complex process in which a
patient engages or allows the health-care
professional into a special relationship,
such that the professional’s skill, knowl-
edge and experience can effect a benefi-
cial health outcome for the patient. The
success of this interaction depends not
only on professional factors but also
on myriad intra-personal, inter-personal,
social and ethical determinants that over-
lay the basic science that underpins the
Western medical model.
Evidence-based medicine (EBM) is de-
fined precisely as the practice of medicine
that ‘requires the integration of the best re-
search evidence with our clinical expertise
and our patient’s unique values and cir-
cumstances’.
1
In other words, the clinical
state and circumstances in combination
with the patient’s preferences and personal
actions provide a framework for making
decisions, which also includes the current
best research evidence.
2
The modern
health-care professional is saturated with
the latest knowledge and therefore both ex-
perience and wisdom are required to make
the best clinical decision for the patient.
Where is the wisdom we have lost in
knowledge?
Where is the knowledge we have lost in
information?
From Choruses from the Rock by TS Eliot
3
The practice of evidence-based medi-
cine requires the application of a process
that includes five distinct steps.
1
1. asking an answerable clinical question
2. tracking down the best evidence
3. critical appraisal of the evidence
4. integration of the evidence with the
patient’s situation
5. evaluation of our effectiveness.
There is an important step that is often
missed and that is step ‘0’ or ‘permission
to ask a question’ in the first instance (Dr
Ian Scott, personal communication). An
answerable clinical question should be
framed in a specific structured format and
the most common is the PICO format. For
example, in a preschool child (popula-
tion) with acute otitis media, does treat-
ment with oral antibiotics (intervention)
compared with placebo (comparison)
reduce the duration of ear pain and fever
(outcome)?
Tracking down the best available evi-
dence requires some skills with the search
engines PubMed or Medline, access to the
Cochrane Library or the help of a librar-
ian. The ability to track down a relevant
article often depends on the precision of
the clinical question and again under-
scores the practicality of the structured
format described above so that a librarian,
who is not a clinician, can sort through the
electronic maze and pull out the most rel-
evant articles that best address the clinical
question.
Critical appraisal of the evidence
requires an assessment of the scientific
merit with which the research study was
conducted. In other words, have the inves-
tigators performed the study properly,
such that I can believe their results? This
involves some knowledge of clinical
research study design; however, a number
of critical appraisal tools are available to
assist the appraiser in asking questions
about the randomisation procedure,
patient and researcher blinding, follow-up
and intention to treat analysis.
4,5
Many of
the journals now present research papers
in a style that allows the reader to follow
this critical appraisal process in a logical
format.
6
Following our example above,
a recently updated Cochrane review sug-
gests that treating acute otitis media with
antibiotics had no effect on reduction of
pain at 24 hours but did reduce pain
between two to seven days and had no
effect on subsequent hearing problems
due to otitis media.
7
The best available research evidence
needs to be integrated with the patient’s
particular clinical situation. In a child with
full immunisation and ready access to
medical review, the decision to treat acute
otitis media may depend on patient or
rather parent viewpoints. It would be dif-
ficult to argue that a child must take anti-
biotics for acute otitis media, when serious
complications including meningitis or
mastoiditis are rare, however, the decision-
making process for a child with incom-
plete immunisations living in a remote
community would adjust the threshold for
the decision to prescribe antibiotics.
Finally, in an ideal world the effective-
ness of each health-care intervention
What is evidence-based medicine?
A personal viewpoint
GUEST EDITORIAL