Roy Poses correctly points out the flip side to the problem I mentioned of
the "knee-jerk demand for RCT proof" on the part of payers. Namely that
there are all too many situations where, in spite of a lack of evidence or
even evidence against, payers have found it expedient to cover an
intervention, usually to avoid consumer and political pressure, bad
publicity from a poorly informed media, or to avoid law suits. When the
bottom line is the only concern, paying for bad treatments may be cheaper
than fighting lawsuits. A recent example of this has been coverage of
highdose chemotherapy with stem cell transplantation for breast cancer.
This was widely covered before there was adequate evidence. As the evidence
has come in, it has been unconvincing. Roy's example of CABG is a little
more complicated. This is not a yes or no coverage decision, but involves
development and adherence to guidelines for who is appropriate for the
procedure and who isn't (HDC/SCT for breast cancer may also evolve into this
type of problem). Payers have been even less on top of these types of
problems than the yes or no situations. Once an intervention such as CABG
is covered for an appropriate situation, it may well begin to be used and
covered for other situations without adequate evidence. There is also the
problem of many "grandfathered in" long standing interventions that are
covered in spite of poor evidence support. So Roy is correct that there is
much unjustified coverage going on.
In spite of the above types of exceptions with unjustified coverage, my
impression is that consumers and many providers are correct in their belief
that the default position for payers concerning new interventions is no
coverage without RCTs. When RCTs are impractical or unethical or for some
other reason not forthcoming, certain hot-button interventions might get
around this obstacle, for better or worse; but other interventions with less
support from consumers, politicians and providers may languish in limbo.
What I am interested in is how we can extricate from limbo the worthwhile
interventions which may never get adequate RCTs. This is a major problem
for the health technology assessment community, because these types of
problems are a major proportion of the interventions for which policy makers
turn to us for help. If the intervention has a nice set of RCTs, our work
is easy. Which is why we get few of those topics - policy makers may think
they can handle those situations without us. But when there are no RCTs,
and possibly never will be, do we throw up our hands and by ommission lend
support to possibly economically driven decisions for no coverage?
There will always be a need for methods and guidelines for analysis and
decision making involving non-RCT evidence. Let's take a look at
informative examples where this was done successfully at the same time we
are compiling examples of RCT successes.
David L. Doggett, Ph.D.
Senior Medical Research Analyst
Technology Assessment Group
ECRI, a non-profit health services research organization
5200 Butler Pike
Plymouth Meeting, PA 19462-1298, USA
Phone: +1 (610) 825-6000 ext.5509
Fax: +1(610) 834-1275
E-mail: [log in to unmask]
%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%%
|