| · Adverse events
· Supermarket pharmacy
· Research
· Regulation of medicines
· OTC statins
· Pricing (2)
· Pharmacy practice
· CPD
· Reciprocity
· Registration examination
· Veterinary pharmacy
· The Society
· Birdsgrove House (5)
Letters to the Editor
|
OTC statins
Worrying differences in assessment and interpretation
From Dr I. Ab I. Davies, MRPharmS
I was concerned regarding the article by Blenkinsopp et al1 entitled “The
utility and feasibility of a community pharmacist-supported protocol
assessment of eligibility for OTC statin treatment” (PDF 75K).
The data presented were somewhat disturbing.
In the introduction, the authors state that “the protocol would
exclude pre-existing conditions that indicate a high [coronary heart
disease] risk or a condition best managed under physician supervision”.
Yet in the results the authors state that 10 customers (to use the authors’ designation)
were on specific exclusion drugs, 51 were on antihypertensive drugs,
20 were on antianginal drugs and 17 were diabetic. This amounts to 98
customers, out of the 160 that were questioned. If the 34 customers that
had no risk factors are included (see Table 1 of the paper), that makes
a total of 132 leaving only 28 customers that would be appropriate for
consideration in the proposed protocol.
The primary biochemical activity of the statins is to inhibit the enzyme
3-hydroxy-3-methyl-glutaryl-coenzyme A (HMG-CoA) reductase that is responsible
for the synthesis of the key intermediate — mevalonate — in
the metabolic pathway for the synthesis of isoprenoids, sterols and steroids.
The net pharmacological effect of this activity is the reduction in the
concentration of circulating plasma cholesterol. In the report, plasma
cholesterol was neither estimated nor reported. Neither was blood pressure,
the other most prevalent contributory factor in CHD assessment, despite
the fact that 11 out of 12 pharmacies offered “either cholesterol
testing, blood pressure measurement, or both”.
In Table 1, if the number of customers (total of 126 of the 160) reporting
between one and four risk factors is multiplied in each case by the corresponding
number of factors — (68x1)+(27x2)+(18x3) +(3x4) — a value
of 208 is obtained. However, the sum of the number of subjects in Table
2 “Specific risk factors reported (n=126)”, reporting one
to four factors, would appear to be 209. These figures do not agree.
The more worrying aspect of this article is in the data presented in
Table 4 “Outcomes of pharmacists assessment for simvastatin treatment
(n=160) and concordance with the outcome measures of protocol eligibility
for treatment”. The number of subjects considered by pharmacists
to be “not at moderate risk” and to require advice only,
was 30; yet 34 subjects were reported as having none of the CHD factors
considered (Table 1). Of these 30, nine (approximately 30 per cent) were
considered by the “protocol eligibility for treatment” as
interpreted by one of the authors, to require treatment for simvastatin.
Pharmacists considered that 53 patients were suitable for treatment with
simvastatin (Table 4), but of this population 16 (approximately 30 per
cent of this population) were considered ineligible by the “protocol
eligibility for treatment” when the data were interpreted by one
of the authors of the article, and would have been advised to take simvastatin
unnecessarily.
Of the 71 customers considered by pharmacists to be in the high risk
category that pharmacists would have referred to a doctor, one author’s
interpretation of the protocol data estimated that six would have been
eligible for treatment.
Again, in Table 4, the numbers (n=160) do not add up. The numbers of
subjects listed by pharmacist amounts to 158 while the numbers considered
eligible/ ineligible by the “protocol eligibility for treatment” add
up to 154.
These discrepancies between the pharmacist’s assessment of suitability
for treatment and the author’s interpretation of the “protocol
eligibility for treatment” are worryingly large. However, the paucity
in the number of subjects, both pharmacists and “customers” should
raise serious questions regarding the validity of the statistical data
presented with regard to both the advisability of providing statins over
the counter and the suitability of the questionnaire as a means of judging
the eligibility of customers for such treatment.
The absence of cholesterol and blood pressure estimations should also
raise questions regarding the validity of the questionnaire. The questionnaire
did not consider the possibility of the subjects suffering from diabetes
(see the Joint British Societies Coronary Risk Prediction Chart reproduced
at the end of the BNF and taken from Heart (1998), based on the Framingham
(1998) data.2
The computerised version of MIMS has an interactive programme based on
the Framingham (1998) data that can be used to obtain a measure of the
susceptibility of an individual to CHD based on the sex, age, plasma
cholesterol concentration, systolic blood pressure and whether the subject
was a smoker or drank alcohol. Even so, MIMS states that “blood
cholesterol alone is a relatively poor predictor of individual CHD risk”.
Some concern has also been expressed by the medical profession regarding
the availability of statins without sufficient background knowledge of
the patient’s medical history and condition.3 Also the Framingham
(1998) report advises that blood cholesterol concentrations (fasting,
total and high-density lipoprotein) should be carried out at least three
times and on separate occasions and that the analysis “should be
made in a laboratory participating in the national quality control scheme
for cholesterol”. Likewise, blood pressure should be estimated
on at least three different occasions with both diastolic and systolic
pressures monitored. There is also the added problem of the interaction
between simvastatin and grapefruit juice.
Finally, if pharmacists are to carry out these duties — taking
patient histories (in preference to allowing the “customer” to
fill out a questionnaire) and carrying out chemical analyses — the
premises would require a sound-proofed consulting room, separate from
the shop and dispensary, to assure the customer of confidentiality, and
also a separate room to serve as an analytical laboratory.
Iolo Davies
Newtownards,
Co Down
References
1. Blenkinsopp J, Cottrell J, Mann SG. The utility and feasibility
of a community pharmacist-supported protocol assessment of eligibility
for OTC statin treatment. Pharmaceutical Journal 2004; 273:606–9
(PDF 75K)
2. Joint British recommendations on the prevention of coronary heart
disease in clinical practice. Heart 1998;80(Suppl 2):S1–S29.
3. Thompson A, Temple NJ. The case for statins: has it really been made?
Journal of the Royal Society of Medicine 2004;97:461–4.
| |
JOHN BLENKINSOPP, JEREMY COTTRELL and STEPHEN
G. MANN, authors of
the article, reply:
Thank you for the opportunity to respond to Dr Davies
concerning our study.
Dr Davies queries the numbers of subjects excluded on grounds of medical
conditions and excluding drugs, and concludes that these amounted
to 98. However none of
the non-eligible categories was exclusive, and subjects could have more than
one medical condition and be taking more than one excluding drug. As stated
in the text, 79/160 subjects had a relevant medical history that would
exclude them
from OTC simvastatin. This number of exclusions might have been of concern
if it had been selected as a typical OTC population, but in fact the
sample consisted
mainly of men who were in the pharmacies for other reasons, typically collecting
prescriptions, and with a variety of medical conditions. The objective was
to “road
test” the protocol materials rather than generate a “representative” sample — as
such the exclusions were as useful as the non-exclusions.
Dr Davies queries the absence of cholesterol and blood pressure measurements.
The pharmacy protocol provisionally approved during the licensing process
did not require a baseline cholesterol test to be performed, so this was
not required
during the study. The protocol did, however, suggest that the pharmacist
offered a blood pressure measurement if available and not recently measured,
so this
was included.
With reference to Table 2 and the n=126, this refers to the number of subjects
with one to four specific risk factors and not number of risk factors; ie,
from Table 1, 160 minus 34 equals 126.
Regarding subjects without the specific risk factors, men aged 55 years or
older without these are at moderate risk because of their age. Therefore
some of the
subjects without risk factors were in fact eligible for OTC simvastatin under
the protocol.
Table 4: Outcomes of pharmacist assessment for suitability for
simvastatin treatment (n=160) and concordance between outcomes and
protocol
|
Pharmacy outcome |
Number of subjects |
% |
Protocol eligibility
for treatment
Yes / No |
Not at moderate risk —
advice only |
30 |
18.8 |
9 / 21 |
Suitable for simvastatin therapy |
53 |
33.1 |
37 / 16 |
High risk category, excluding
medical condition or medication |
73 |
45.6 |
7 / 66 |
Missing data |
4 |
2.5 |
- / 4 |
|
Table 4 does contains some transcription errors and we apologise for
this. The Table should have read as shown above.
We do not agree that the discrepancies between the pharmacist’s assessment
and strict protocol eligibility (not “author interpretation”) were “worryingly
large”. The volunteer pharmacies were testing prototype materials that
were subsequently developed and refined as a result of the feedback. None of
the pharmacists involved had received training and simply used the prototype
materials as received. (Subsequently, during the pre-launch and early post-launch
period, there was an extensive training programme for pharmacists in the operation
of the protocol.)
Despite the lack of training and finalised materials the pharmacists operated
the protocol remarkably well; nearly 80 per cent (124/160) of the subjects
were correctly assessed according to strict protocol eligibility. Of the
remaining 20 per cent (32/160), 16 could be described as the pharmacist acting
cautiously
in not recommending a long-term treatment (especially as these were not “real” customers
requesting the product). The remaining 16 cases (pharmacy assessed as eligible
who were not strictly protocol eligible) were mostly not clear-cut “wrong” assignment
but related to interpretationally grey areas of the protocol, as discussed in
the paper. A major reason for conducting the study was to identify protocol improvement
opportunities, including these type of “grey” areas that can cause
difficulty in practice.
So far as the issue of grapefruit juice is concerned, this warning was strengthened
subsequently and is clear in the current summary of product characteristics
and protocol materials.
It is to be hoped that criticism such as this does not deter other manufacturers
from testing pre-launch materials in real pharmacies and subsequently publishing
the outcome in a transparent process. |
|
The
Pharmaceutical Journal