Prescribing of lipid regulating drugs and admissions for myocardial infarction in England
BMJ 2004; 329 doi: https://doi.org/10.1136/bmj.329.7467.645 (Published 16 September 2004) Cite this as: BMJ 2004;329:645All rapid responses
Rapid responses are electronic comments to the editor. They enable our users to debate issues raised in articles published on bmj.com. A rapid response is first posted online. If you need the URL (web address) of an individual response, simply click on the response headline and copy the URL from the browser window. A proportion of responses will, after editing, be published online and in the print journal as letters, which are indexed in PubMed. Rapid responses are not indexed in PubMed and they are not journal articles. The BMJ reserves the right to remove responses which are being wilfully misrepresented as published articles or when it is brought to our attention that a response spreads misinformation.
From March 2022, the word limit for rapid responses will be 600 words not including references and author details. We will no longer post responses that exceed this limit.
The word limit for letters selected from posted responses remains 300 words.
Dear Editor,
Majeed et al.'s exploration of data on statin prescribing and
admission rates for myocardial infarction raises the question of what the
exact purpose of the Dr Fosters case notes initiative is. The authors note
some of the many reasons why the relationship they describe could exist,
to which one could add several others. Thus we have an intervention
(increased prescription of statins) taking place set against a background
of a 50% reduction in deaths from myocardial infarction that has taken
place since the late 1970s, multiple other concurrent interventions
(ranging from changing diets to better treatment of hypertension and
earlier interventional treatment for atheroma), and changing case
definitions (introduction of troponin assays). When one adds the
probability of a time-lag effect, with any increase in prescribing likely
to precede a reduction in mortality, incomplete ascertainment (e.g. those
suffering myocardial infarctions who die before reaching hospital are
excluded), and the well known problems of data quality with Hospital
Episode Statistics, one has almost every problem one could imagine in
trying to make sense of an observed association. It is very likely that
the benefits of statins have been exaggerated, especially when
administered at low doses to people at low initial risk [1], but it is not
clear that this paper adds anything to what was already known.
1. Thompson A, Temple NJ. The case for statins: has it really been
made? J R Soc Med 2004; 97: 461-464.
Competing interests:
CHKS provides commercial benchmarking and comparative information services to the NHS
Competing interests: No competing interests
Majeed et al. demonstrate convincingly that ‘admissions for
myocardial infarction (MI)’ have not fallen nearly as much as the potential 30%
reduction anticipated from increasing Statin prescription. I believe that changing diagnostic criteria
can account for all of the demonstrated dissonance, and that the data remain
fully compatible with the trial evidence.
The number of hospital admissions is but a rough measure of MI incidence. The new criteria for diagnosis1 of MI may lead to a more
than 50% increase in cases - swamping
even the 30% maximum hoped-for relative risk reduction benefits of Statin,
which in absolute terms should have resulted at best in a 5% reduction in MI
incidence.
What is the most we should expect, if the Statin trial data
were indeed generalisable ?
“During the study period, statins were recommendedfor high risk patients, but many myocardial infarctions occur in people at low risk who would not
have been recommended fortreatment with statins. This would limit
the population impactof the increase in the prescribing of
statins.”
Certainly it is true that only treated patients stand to
gain, but the whole point of the recommended ‘treat those at high-risk’
strategy is that it will have the maximum population impactfor any
given spend. Using my own general practice database, I calculated the
percentages in each calculated Framingham Risk , and derived the following
crude UK population estimates in each risk category:-
FRAMINGHAM RISK % |
0_ |
10_ |
20_ |
30 + |
|
TOTAL |
Persons |
40.8m |
12.8m |
4.5m |
1.9m |
|
60million |
Population % |
68% |
21% |
7% |
4% |
|
100% |
Expected MI’s annually |
123434 |
176867 |
109876 |
67575 |
|
|
Maximum MIs preventable at a 30% RRR |
37030 |
53060 |
32963 |
20272 |
|
|
Estimated target Spending to achieve the above today Simvastatin 28x40mg = |
£ 10289m |
£ 3237m |
£ 1126m |
£ 467m |
|
|
“prescriptions of lipid regulating drugs (largelystatins) increased from 3.1m to 17.6m items during 1996-2002”
Given that ‘prescription items’ last on average for 1 month or more, then this might represent an
increase from 0.26m (at most) to 1.63m
persons (at most) on long-term treatment.
Given NICE prescribing guidelines targeting those at 30% risk or more (
including all those already diagnosed with CHD ), there is clearly still some way to go before we cover the 1.85m
estimated to be at high-risk, even if targeting was 100% accurate. The table above shows that IF the 1.85m persons at high risk all received regular
Simvastatin 40mg, then £467m would be spent, and IF the trial evidence of a 30%
relative reduction was achieved across the real population, then 20272 MIs
would be prevented each year. Thus we
could not have prevented more than 20272 MIs for an optimally targeted Statin spend
of 467m, representing a 4.2% reduction in the Framingham estimated
incidence. We actually spend £571m, somewhat less well targeted, and see a 5%
reduction in MI incidence, even with the changing diagnostic criteria..
Interestingly, the new diagnostic criteria are revealing a
progressive change in the proportion of STEMI to NSTEMI events. Whatever the true annual incidence of
MI, the annual death rates should be a
reliable criterion by which we should judge the success or failure of MI prevention strategies.
BMJ 2003;326:134-135 ( 18 January ) Impact of changing diagnostic criteria
on incidence,
management, and outcome of acute myocardial infarction: retrospective
cohort study. Pell et al.
Competing interests:
None declared
Competing interests: Majeed et al. demonstrate convincingly that ‘admissions formyocardial infarction (MI)’ have not fallen nearly as much as the potential 30%reduction anticipated from increasing Statin prescription.
Majeed and Jarman's contribution to the Dr Foster casenote series is
seriously misleading. During the period of their review that compares
statin prescribing with AMI admission rate, there have been major changes
in diagnostic criteria (eg troponins used as markers) and such variable
interpretation of the criteria that the codings used under Hospital
Episode Statistics have failed to keep pace. The epidemiology of acute
myocardial infarction (AMI) cases admitted to hospital has therefore
changed beyond recognition including many more cases than would have been
the case in the late 1990s.
The national audit of AMI suggests that the number of ST elevation AMI
(STEMIs) may be falling quite rapidly while the number of non STEMIs is
rising. This may be a direct and desirable result of statin prescribing
and a major contribution in reducing CHD fatality rates particularly in
the under 65s of both sexes.
While Capewell and colleagues rightly suggest that life style changes
provide the larger contribution to reducing CHD deaths, treatment effects
are also important. Both areas require attention and resource and that is
exactly what the National Service Framework programme has delivered and
will do in the future. The literature showing benefit from statin therapy
is to my mind overwhelming and also subject to an appraisal by NICE in the
near future.
Competing interests:
None declared
Competing interests: No competing interests
Sir
DR FOSTER’S CASE NOTES. Prescribing of lipid regulating drugs and
admissions for myocardial infarction in England (BMJ 2004: 329; 645).
This interesting article concludes “Large increases in the cost and
volume of prescribing of lipid regulating drugs have been associated with
only a modest decline in standardised admission ratios for acute
myocardial infarction”. This raises a number of issues.
Firstly, the problem about using admissions for myocardial infarction
as a proxy for the burden of coronary heart disease (CHD). Although
myocardial admissions have hardly fallen, (for several reasons, as
discussed in the article), age-adjusted CHD event rates and mortality
rates have halved in the last two decades (1). There were 68,230 fewer CHD
deaths in 2000 than in 1981 (1, 2). CHD prevalence has also fallen (3).
Secondly, how much of the CHD mortality fall between 1981 and 2000
can be explained by medical treatments?
The answer, (as Douglas Adams readers will know), is approximately 42%.
Much of the remaining 58% was explained by decreases in major risk factors
in the population, particularly smoking (2).
Total cholesterol levels in the general population fell by only about
4% during this time, accounting for less than 10% of the overall CHD
mortality fall (approximately 7,900 fewer CHD deaths). Statins accounted
for less than 2,000 fewer deaths, barely 3% of the overall fall (2,4).
Meanwhile, countries such as Finland and Mauritius have achieved
dramatic total cholesterol reductions of 1mmol/l or more, mostly through
effective national dietary policies rather than statin prescribing. In
Finland, this contributed to a massive 70% reduction in CHD mortality
rates (5)(6).
In conclusion, statins are visible and expensive, but have a
surprisingly small public health impact. In future, the biggest national
gains will probably come from comprehensive primary prevention programmes
such as Heart of Mersey (5)(6).
Yours Sincerely
Simon Capewell,
Belgin Unal,
Julia Critchley,
Robin Ireland
REFERENCES
1. British Heart Foundation Statistics Database. Coronary Heart
Disease Statistics. 2004. Available at:
http://www.heartstats.org/homepage.asp (Accessed September 27, 2004.)
2. B Unal, J A Critchley & S Capewell. Explaining the decline in
coronary heart disease mortality in England & Wales between 1981 and
2000. Circulation 2004, 109(9); 1101-7.
3. Lampe FC. Morris RW. Whincup PH. Walker M. Ebrahim S. Shaper AG.
Is the prevalence of coronary heart disease falling in British men?
Heart 2001; 86(5):499-505.
4. B Unal, JA Critchley, S Capewell. Reducing CHD deaths in England
and Wales: has primary prevention been overlooked? J Epid Comm Health
2004 Supplement 1; A19.
5. WHO/EURO Programme. CINDI (Countrywide Integrated Noncommunicable
Disease Intervention). WHO Collaborating Centre for Community Programmes
in Chronic Disease Prevention and Health Promotion National Public Health
Institute, Helsinki, Finland. http://www.ktl.fi/eteo/cindi/ (accessed 27
09 04)
6. J A Critchley, S Capewell. Substantial potential for reductions
in coronary heart disease mortality in the UK through changes in risk
factor levels J Epid Comm Health 2003, i57: 243-9.
Competing interests:
None declared
Competing interests: No competing interests
I find it quite appalling that the statin bill for 2002 has risen to
some £571M. Total NHS spending on dietetics of any kind is at most £90M.
Perhaps if £571M were spent we might have some chance of helping patients
to make informed choices about food and health rather than dosing them up
with one medication after another.
Competing interests:
None declared
Competing interests: No competing interests
Dear Dr. Majeed,
I agree with your sentiments - statins are largely
ineffective, a sober reviewing of Heart Protection Study, PROSPER
and others proves this point.
However, please tell me how you can be sure that the modest
decline in admission rates for MI is indeed related to statin
prescribing and not due to increased blood pressure control/
diabestes control/ increased life style awareness? Sureley, the
prescribing for antihypertensive drugs soared too over the
last years - giving a similar curve.
I am looking foreward to hearing from you.
Philipp Conradi
Competing interests:
None declared
Competing interests: No competing interests
Sir,
I read with great interest the article by Azeem Majeed et.al. It seems
that the main message is that despite an immense increase in expenditure
on lipid lowering therapy, there is only a modest decrease in admissions
for acute coronary syndrome (ACS) and myocardial infarctions (MI).
However, the most important piece of information seems to be missing from
the review. As many surveys have demonstrated, lipid lowering therapy, in
most cases, does not reach recommended target levels. So, it seems a bit
premature and inaccurate to claim that despite inceasing costs there's
only a modest reduction in admissions for ACS.
Maybe, if less people were treated and a higher proportion reached
recommended target values, the picure would have been different. Efficacy
is not measured by expenditure, but, by reaching target levels of
cholesterol and it's components.
Competing interests:
None declared
Competing interests: No competing interests
Dr Martin still believes in the fairy tale that statins will prevent
disease.
Given the abundancy of available information plus the availability, I
fail to see why anyone should be polite any longer.
Statins make lots of money. Statins cause more illness than they
could ever prevent.
Statins are possibly an attempt to implement population control.
If this sounds strange to you, perhaps a look at the budgets of
governments would convince you.
The business of disease management will, in the end, be self-defeating.
No one will be able to afford the drugs.
But - and here comes the knell that will restore some semblance of
reality - health care will have to return to
the the individual, and this will be a giant step toward restoration of
traditional values.
No one needs the drugs? Well? Wouldn't that be something.
Increasing the dosage means that the brain of the prescriber has gone on
holiday,
yet there is an even better chance that malicious intent by is involved.
I, personally, am waiting for the Yanks to alert the lawyers.
Competing interests:
None declared
Competing interests: No competing interests
I wish to comment on the reasons given by Majeed et al in Dr Foster's
Case Notes as to why the increased rates of statin prescribing have not
resulted in a more significant reduction in MIs. I believe Majeed et al
have missed the 2 main reasons:
1. Statins have been prescribed to many patients for primary
prevention who are at low risk. An audit conducted in a Bury practice in
2002 revealed that two thirds of patients receiving statins for primary
prevention had a 10 year risk score of 0-15%.
2. The average dose of simvastatin in the 4S study (1) was 27mg.
However, prescribing data show the majority of simvastatin being
prescribed is of the 10 and 20mg strengths (2). Therefore, even when used
in secondary prevention, it is likely that patients are receiving
inadequate doses of statins.
Ref 1: Scandinavian Simvastatin Survival Study Group. Randomised
trial of cholesterol lowering in 4444 patients with coronary heart
disease: the Scandinavian Simvastatin Survival Study (4S). Lancet
1994:344:1383–9.
Ref 2: Prescription Cost Analysis 2002. Government Statistical
Service.
Competing interests:
None declared
Competing interests: No competing interests
The Italian public investment on statins seems to yield a lower than expected return.
In the 2000-2004 five years period, the Italian National Healthcare
System invested cumulatively over 3.3. billion euro to fund the
prescribing of statins in primary care. In 2004, the annual public cost of
statins exceeded 1 billion euro, showing a 3 fold increase compared to
2000. [1]
Based on the Framingham coronary prediction algorithm, the expected
return on this considerable investment would be a significant reduction of
the Italian population’s risk for Coronary Heart Disease (CHD). The
observed trend of hospital admissions for myocardial infarction over the
same five years period seems to confirm the concerns raised by the Majeed
et al. about the prescribing of lipid regulating drugs in England. [2]
We examined the utilisation of statins and the number of hospital
admissions for acute myocardial infarction in Italy from 2000 to 2004. The
relatively short period of observation (5 years) was determined by the
availability of comparable data from public sources [1,3]. Statin’s
utilisation was measured in DDDs (Defined Daily Doses), while hospital
admissions referred to Disease Related Group (DRG) 121, 122 and 123 to
avoid double counting of diagnoses. Both units of analysis were
standardised over 1,000 residents. To account for ageing population, the
2000-2005 Italian resident population obtained by the National Institute
of Statistic (ISTAT) was weighted using the ASSET age/sex weightings. [4]
In the 2000-2005 five years period, the utilisation of statins showed
a threefold increase, from 16.5 to 52.3 DDDs/1,000 weighted residents,
while the admissions for myocardial infarction decreased only marginally
from 1.40 to 1.30 admissions/1,000 weighted residents.
Several caveats prevent the inference of any definitive conclusion
from this analysis:
a. While the Framingham algorithm is based on a 10 years risk assessment,
the observation period of analysis was significantly shorter;
b. While we adjusted for ageing population, the impact of other important
cardiovascular risk factors, such as cigarette smoking habit, blood
pressure and diabetes, was ignored.
c. We used hospital admissions for acute myocardial infarction as a proxy
of Coronary Heart Disease: angina pectoris and coronary disease admissions
were ignored to avoid double counting.
With all the possible cautions in mind, we believe that it would
still be appropriate for healthcare administrators and policy makers to
question the value of current utilisation of statins in general practice.
More specifically, a limited patient compliance would definitely reduce
the contribution of lipid lowering agents to the reduction of the economic
burden of Coronary Heart Disease. Italian data on compliance to statins’
treatment are controversial. A recent study based on the average annual
exposure to statins calculated from a single Local Healthcare Authority
(Pavia), concluded that over 70% of patients stayed on treatment for
longer than 180 days per year. [5] According to a similar study, performed
in an entire Regional Healthcare Authority (Umbria), the median
persistence on statin treatment was just 5.3 months, while only 49.6% of
patients renewed their prescription for consecutive years. [6]
A focused emphasis on individual compliance to treatment would
significantly increase the expected outcomes and, consequently, the
economic value of statins prescribing in general practice.
[1] Agenzia Italiana del Farmaco AIFA. L’uso dei farmaci in Italia.
OSMED reports 2000-2004. Available online at:
http://www.agenziafarmaco.it/aifa/servlet/section.ktml?target=&area_tema...§ion_code=AIFA_PUB_RAP_OSMED&cache_session=true
[2] Majeed A, Aylin P, Williams S, Bottle A, Jarman B (2004). Prescribing
of lipid regulating drugs and admissions for myocardial infarction in
England. BMJ, 2004 329:645, doi: 10.1136/bmj.329.7467.645
[3] Ministero della Salute. Rapporti annuali sui ricoveri ospedalieri,
2000-2004. Available online at:
http://www.ministerosalute.it/programmazione/sdo/sezDocumenti.jsp?label=osp.
[4] Favato G, Mariani P, Mills RW, Capone A, Pelagatti M, et al. (2007)
ASSET (Age/Sex Standardised Estimates of Treatment): A Research Model to
Improve the Governance of Prescribing Funds in Italy. PLoS ONE 2(7): e592.
doi:10.1371/journal.pone.0000592
[5] Lucioni C, Mazzi S, Cerra C et al (2006), Uno studio di Drug
Utilisation delle statine nella recente prassi terapeutica italiana.
PharmacoEconomics-IRA. 8(1): 3-17
[6] Abraha I, Montedori A, et al (2003), Statin compliance in the Umbrian
population. European Journal of Clinical Pharmacology. Vol. 59 Numbers 8-
9.
Competing interests:
None declared
Competing interests: No competing interests