Intended for healthcare professionals

Rapid response to:

Editorials

Benefits and harms of drug treatments

BMJ 2004; 329 doi: https://doi.org/10.1136/bmj.329.7456.2 (Published 01 July 2004) Cite this as: BMJ 2004;329:2

Rapid Response:

"Bangs for backfires" - a useful measure of benefits and harms? E.g. Rosuvastatin.

In his editorial, Jan Vandenbroucke discusses the importance of
observational data in relation to assessing drug safety.1 On 26th June,
The Lancet published a letter from Sidney Wolfe of Public Citizen calling
for the withdrawal of Rosuvastatin2 that made UK national news.3 This
seems premature and based on incomplete consideration of risk and benefit,
but represents a contemporary (if flawed) attempt to incorporate later
observational data into a view based on existing trial data. A simple and
intelligible measure allowing meaningful comparisons between “me-too”
drugs is clearly needed.

Using the rhabdomyolysis and script volume data Wolfe reports for the
US market,2 I estimate a rate of between 8 - 20 cases per million scripts
in the first full year for the 10mg dose, the low end of the range being
most likely (unless reports increase). Since rhabdomyolysis is a dose-
related side-effect, comparison of statin strengths equipotent in reducing
LDL-C would be logical, whereas the crude rates of the other statins
reported by Wolfe are very likely skewed downwards by their lower
efficacies. Not to be forgotten is an established death rate from statin-
induced rhabdomyolysis of 1 per 10 million scripts.4

In these consumer days of “bangs for bucks”, we surely need to be
comparing “bangs for backfires” as well; serious vascular events avoided
for each serious adverse event incurred through statin treatment. A 10y
CHD risk of 30% is a reasonable indication for a statin. If 1mM
decrease in LDL-C reduces CHD events by at least 33% from the 3rd year of
treatment5 and Rosuvastatin 10mg daily reduces LDL-C by 2.1mM,5 then 10y
CHD risk would be reduced to around 10%. So if 10,000 patients were
treated for 10 - 12 years, current data would imply around 2,000 avoided
CHD events (including deaths) at a cost of 8 – 20 cases of rhabdomyolysis
– better than 100:1 “bangs for backfires”, but in need of consideration by
both patient and clinician. Some patients will want higher benefit at
higher risk, or will require so to reach treatment targets.

So does the market need another statin? Well, yes, if it can better
address the needs of higher risk patients at an acceptable level of extra
risk. In the UK, only Atorvastatin and Rosuvastatin currently achieve LDL
-C reductions greater than 50%. That sector of the market (our highest
risk patients) is probably not best served by a single product. Their UK
pricing structures clearly reveal the value of competition to the NHS.

More robust and comparable safety data are required, together with
meaningful ways of communicating treatment risk. (Above estimations are
presented in good faith for purpose of example and argument only.)

1 Vandenbroucke JP. Benefits and harms of drug treatments. BMJ
2004; 329: 2-3.

2 Wolfe SM. Dangers of rosuvastatin identified before and after FDA
approval. Lancet 2004; 363: 2189-90.

3 http://news.bbc.co.uk/1/hi/health/3838915.stm

4 Staffa JA, Chang J, Green L. Cerivastatin and reports of fatal
rhabdomyolysis. N Engl J Med 2002; 346: 539-40.

5 Law MR, Wald NJ, Rudnicka AR. Quantifying effect of statins on low
density lipoprotein cholesterol, ischaemic heart disease, and stroke:
systematic review and meta-analysis. BMJ 2003; 326: 1423-30.

Competing interests:
None - unless as a result of a future prescribing incentive scheme

Competing interests: No competing interests

05 July 2004
David R Purdy
GP prescribing lead, Northamptonshire Heartlands PCT
Rushden Medical Centre, Adnitt Rd, Rushden. NN10 9TR.