It appears that the initial resistance/scepticism
towards evidence-based medicine (EBM) has declined and the current
challenge for professionals is "how to" rather than "why to" practice EBM.
The major hurdles include limitations related to: (i) availability
(of high quality evidence on problems/interventions relevant to our
setting), (ii) accessibility (to current evidence at the
point-of-care), (iii) appraisal (of available evidence to judge
reliability/validity), and (iv) applicability (of evidence
developed for/from another setting into the local setting). The sterling
contribution of the Cochrane Collaboration worldwide and the South Asian
Cochrane Network (now Centre) in India have greatly facilitated progress
towards overcoming these barriers.
Beneath EBM
The foundation of evidence-based medicine is evident
(pun not intended) from its definition, which is the "integration of best
research evidence, with clinical expertise and patient values"(2,3). Three
important points should be noted to avoid misperception. (i) ‘Best’
research evidence implies systematic identification, critical appraisal
(of methodology) and synthesis of scientific literature; not merely
searching for (and citing) randomized controlled trials or systematic
reviews to support a given action/decision/guideline/recommendation. The
latter avoidable error unjustifiably tarnishes the principle and process
of EBM. (ii) Clinical expertise is integral (rather central) to
effective health-care; EBM reinforces expertise with best research
evidence. Therefore ‘evidence-based’ is not distinct from ‘experience or
expertise based’, but complementary. (iii) ‘Patient values’ include
the unique circumstances (such as health-care setting, personal/social
issues, etc) of individual patients for/by whom health-care decisions are
made. This concept also facilitates shared (patient-professional)
decision-making which is a goal of health care.
Despite tremendous progress in the production,
appraisal, and access to the best research evidence; understanding of
patient values has lagged in developing countries where health-care is
often ‘provider-driven’ with a ‘take-it-or-leave-it’ attitude. Attempting
to apply best research evidence, bypassing/ignoring patient values can do
more harm than good, a fact that needs to be recognised when ‘best
evidence’ from other settings is directly extrapolated to the local
health-care setting. Paradoxically, a shared decision not to apply best
research evidence on account of patient values, can also be regarded as
sound evidence-based medicine.
Behind EBM
Despite its numerous strengths, a potential limitation
of EBM is that ‘health-care interventions’ (therapeutic/diagnostic) rather
than ‘health problems’ are the usual starting point for generating
research evidence, whether primary (comparative trials) or secondary
(systematic reviews). In other words, while EBM tries to answer "Does X
intervention work for Y clinical problem?", it does not directly address
the more important issue of "What’s the best approach for Y clinical
problem?" For example, EBM can answer "Is hepatitis B vaccination
efficacious?" but not "What’s the best approach to control hepatitis B in
India? (choosing from one or more of - universal vaccination, selective
vaccination, health education strategies, treatment of cases, screening,
etc)." The latter question is more complex, but necessary to make
appropriate decisions. One of the reasons that research revolves around
interventions is that industry produces and markets a variety of products
that need to be evaluated.
Besides EBM
Evidence-based practice necessitates that ‘best
research evidence’ itself be critically appraised to judge validity,
clinical significance, and applicability. There are currently a host of
tools for appraisal of validity based on methodological quality of
research and evaluation of sources of bias (systematic error). However,
judging clinical significance of research findings requires expertise and
experience. Assessment of applicability is perhaps the toughest component
and requires judgment of several factors described below.
Often (primary and secondary) research does not provide
conclusive evidence on all the outcomes relevant to multiple stakeholders.
Taking the example of hepatitis B vaccination, the decision to initiate
vaccination does not depend on evidence of efficacy (does it work?)
alone. Other issues like safety (short and long term), cost,
cost-effectiveness, feasibility, comparison with other possible
interventions, etc need to be factored-in, to make an appropriate
decision. Consideration of these factors together adds up to evidence of
effectiveness (will it work in this setting?)(3). Further, the
outcomes determining ‘effectiveness’ could vary among different
stakeholders. For hepatitis B, the main outcome of interest for
health-care professionals could be prevention of hepatitis B and its
complications; for policy-makers could be cost-effectiveness, feasibility
and prioritization against other health-care needs; for ‘consumers’, the
guarantee of individual protection, freedom from side effects and
convenience. ‘Best’ research evidence usually does not address all these
complex but important issues. The additional problem is that research
often presents secondary/surrogate outcomes, that are expected to
correlate with the main outcomes of interest (for example hepatitis B
surface antigen evaluated in most trials is a surrogate marker for
hepatitis B infection diagnosed by histopathology). The extent to which
various secondary outcomes actually reflect the primary outcome of
interest, necessitates critical judgment. Based on these facts, it is easy
to appreciate that highly objective ‘evidence of efficacy’ should be
superseded by ‘evidence of effectiveness’, which could have an additional
subjective component. Therefore there has been a gradual shift from
‘evidence-based medicine’ towards a more practical concept of
‘evidence-informed health-care/decision making’; which is based on more
than systematic reviews of efficacy/safety.
Beyond EBM
The various components that together facilitate
informed decision-making comprise the discipline of Health Technology
Assessment. Although it sounds like a misnomer, "health technology" is a
loose term covering all methods used to promote health, prevent or treat
disease and improve rehabilitation or long-term care(4). My own definition
includes eight Ps viz the Products, Practices, Procedures,
Processes, Programs and Principles that Promote health or Prevent disease.
Health Technology Assessment (HTA) is the scientific process of examining
the medical (efficacy, safety), economic (cost, cost-effectiveness),
social, logistic and ethical aspects pertaining to the application of a
given health technology. It has been described as a bridge between
health-care research and real-world decision-making(5). The final product
of HTA is a document that ‘informs’ various stakeholders. Most developed
countries have well-established HTA units/organizations/institutions that
guide individual /community/national policy in their setting. This is
deficient in developing countries, where competing priorities for limited
resources demands that health-care decisions be based on robust scientific
principles.
The Way Forward
The leadership position of the Indian Academy of
Pediatrics and the prestige of Indian Pediatrics makes them natural
vehicles to foster a culture of evidence-informed healthcare in India.
Some of the practical ways this can be achieved are:
1. Mathew JL, Singh M. Evidence based child health: fly
but with feet on the ground! Indian Pediatr 2008; 45: 95-98.
2. Straus SE, Sackett DL. Using research findings in
clinical practice. BMJ 1998; 317: 339-342.
3. Sehon SR, Stanley DE. A philosophical analysis of
the evidence-based medicine debate. BMC Health Serv Res 2003; 3: 14.
4. Mathew JL. Evidence, EURECA and Evidence-Based Child
Health. Indian Pediatr 2008; 45: 518.
5. No authors listed. Health Technology Assessment.
From: http://inahta.episerverhotell.net/HTA/. Accessed on 28 January,
2010.
6. Battista RN. Expanding the scientific basis of
health technology assessment: A research agenda for the next decade. Int J
Tech Assess Health Care 2006; 22: 275-280.