This discussion has highlighted some of the major problems that hinder the
widespread implementation of EBP. There are several reasons why high quality
evidence-based decisions may come up with different answers for populations
as opposed to individuals.
Firstly there is the issue of perspective. Quite rightly NICE and other
policy makers take a societal perspective and such decisions must maximise
cost-effectiveness. However, individuals in most health-care systems are not
interested in cost-effectiveness, but in clinical effectiveness and using a
different outcome measure (e.g. cost per QALY gained vs QALE) may provide a
different answer to the same question. This is a difficult dilemma for
clinicians who may be torn between their responsibilities to the individual
and to society and is probably at the root of many of the issues recently
aired relating to politicians and the media. In terms of newspaper coverage a
good personal anecdote always beats a responsible decision based on careful
statistical analysis and critical appraisal.
However there are other good reasons why individual decisions may differ from
the advice based on a high quality systematic review.
Patients are individuals and even the best RCT evidence is based upon average
results from a selected patient cohort. Our ability to predict the risks and
benefits of specific treatments in individuals with known characteristics may
allow a tailor made decision through formal or informal techniques with an
accuracy that exceeds a decision based upon subgroup analyses from RCTs.
Another important issue is that decisions may hinge upon the weighting or
utilities attached to specific outcomes. These vary from patient to patient.
I would certainly prefer a clinical decision that affects me to reflect my
own views about the utility of possible outcomes rather than an averaged
societal utility - particularly one based upon such questionable techniques
as the EoroQol.
The underlying problem is that clear guidance tends to be over-simplified as
any degree of clinical freedom leads to a lack of equity. There is no
theoretical reason why decision aids could not take into account patient
characteristics, risk factors, local outcome data and differing patient
preferences. Perhaps the EBH community should be taking more interest in the
way that techniques such as decision and utility analysis can be combined
with the results of RCTs and other forms of evidence to make individualised
decisions that remain evidence-based.
Jonathan Michaels
Jonathan Michaels, MChir. FRCS.
Consultant Vascular Surgeon,
Northern General Hospital,
Herries Road
Sheffield S5 7AU
UK
Tel: +44 114 271 4968
Fax: +44 114 271 4747
email: [log in to unmask]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|