Intended for healthcare professionals

Rapid response to:

Analysis

Getting better value from the NHS drug budget

BMJ 2010; 341 doi: https://doi.org/10.1136/bmj.c6449 (Published 17 December 2010) Cite this as: BMJ 2010;341:c6449

Rapid Response:

Changing prescribing may not be so simple

Moon et al make a convincing case that a relatively small number of
changes in prescribing could save considerable money, but delivering such
savings may not be so simple.

First, there is not always consensus that different drugs are
therapeutically equivalent. They estimate 84 million pounds of savings
from switching patients from tiotropium to ipratropium, but this is
contrary to the 2010 NICE COPD guideline update
(http://guidance.nice.org.uk/CG101/Guidance/pdf/English -recommendation
U4). The evidence underlying this recommendation is open to
interpretation, but cost driven switching is inevitably less persuasive in
such situations.

Second, they give the example of switching patients from modified
release nifedipine to amlodipine. For some years, my old practice was
always near the top of our local league table for cost-effective calcium
channel blocker use because we routinely used the recommended cheaper drug
- modified release nifedipine. When amlodipine came off patent and it
became formulary first choice, we went from hero to zero overnight. We
decided that we would immediately switch to using amlodipine for new
patients, but that we would not routinely change existing patients'
treatment. This was primarily because changing drugs causes significant
side effects in a proportion of patients. These side effects are a
mixture of placebo (or more accurately nocebo) effects and genuine
differences in tolerance. But their cause is irrelevant, as is blinded
trial evidence that side effect rates are similar with two drugs. The key
issue is that a significant proportion of patients experience symptoms as
a result of medication change, and this needs to be managed even if they
are 'just' nocebo. Managing the process of reviewing 1.3 million patients
on losartan is therefore no small task, and is unlikely to be achieved
quickly unless it is done without discussion with patients. As David
Phizackerley highlights, where does that leave "no decision about me
without me"?

Third, we had the reverse experience of going from zero to hero with
proton pump inhibitors. While both were still on patent, we had a higher
proportion of patients on omeprazole (more expensive) compared to
lansoprazole (cheaper and local guideline recommended as a result). When
omeprazole came off patent and became formulary first choice, our PPI
prescribing cost indicator 'performance' miraculously improved. Such
sudden changes in 'quality' risk breeding cynicism in those being
measured, but highlights a more general difficulty with a generic
anticipation policy. The drug due to come off patent first may be the
class leader and the most expensive while on patent. That implies that
drug switching for 'generic anticipation' has to be achieved in very short
time-frames if it isn't to adversely affect budgets now, irrespective of
likely savings in the future.

None of this means that the NHS shouldn't seek to minimise
prescribing costs as far as possible, but systematically and rapidly
delivering the changes they identify will require upfront investment that
seems unlikely in the current financial climate.

Bruce Guthrie
b.guthrie@chs.dundee.ac.uk

Competing interests: No competing interests

24 December 2010
Bruce Guthrie
Professor of Primary Care Medicine
University of Dundee