TABLE Iv-British regional heart study. Characteristics at screening of men in different categories ofmilk intake Level of activity M1ilk intake None Teaorcoffeconly Drink Incereal Incerealplusdrink
Age (years)
Manual workers
50-8 50-4
59 67 71
50-0 49-7
46 51
50-7
Current smokers (%)
Obese*
38 50 49 29 36
Systolic blood
(%)
Serum total cholesterol (mmol/l)
27 22 22 16 16
6-36 6-32 6-31 6-25 6-30
146-0 146 9 146-4 143-9 142-7
pressure (mm Hg)
Inactive Vigorous (%) (%) 13 11 11 7 6
Recall doctors' diagnosis
Ischaemic heart disease ("/.)
Diabetes (/)
13 7 4 7 3
2-1 1-7 1.0
20 18 20 24 25
Drinking
Non-drinkers Heavy drinkers
1-5
1-8
(%)
(%)
11 5 6 6 7
16 15 16 6 5
*Body mass index ¢28-0 kg/m-. TABLE v-British regional heart study. Fat spread usually used and major ischaemic heart disease events over 9-5 years offollow up Fat spread
No of men
No (%) with ischaemic heart disease
Relative risk (95% confidence interval) of heart attack*
136 2735 1038 3808
11(8-1) 247 (9 0) 66 (6-4) 283 (7-4)
0-57(0-22to 1-46) 1 00 0-76 (055 to 1-04) 0-87 (0-79 to 1-06)
None
NMargarine only Margarine and butter Butter only
*After exclusion of men with recall of doctor's diagnosis of ischaemic heart disease at screening; cigarette smoking, and blood cholesterol concentration.
diagnosis of ischaemic heart disease than men using butter (10 1% v 4 2%). These findings strongly suggest that pre-existing illness (obesity and heart disease) is associated with choice of fat spread. After exclusion of the men with recall of a doctor's diagnosis of ischaemic heart disease and adjustment for other key risk factors no significant association was seen between butter intake and heart attack rates. There seems to be no evidence that butter eating is protective against ischaemic heart disease. Conclusions-Data from the British regional heart study show that men who had the highest milk intake at initial screening had a lower rate of heart attack than men who drank no milk. Similarly, men who used only butter as a spread at initial screening had a lower rate of heart attack than men who used margarine only or no spread at all. However, the characterstics of the men in the several milk drinking or fat spread categories are very different, and these differences must be taken into account when assessing the importance of the association in terms of possible causality. When these background characteristics are taken into account we can find no significant association between milk intake or fat spread use and the incidence of heart attack in these middle aged British men. People with disease often change their lifestyle and this must be fully considered when attempting to associate any specific behaviour with a disease end point and to interpret the findings in terms of causality. Men in the United Kingdom who do not drink milk at all or do not use butter are small groups who differ in many ways from the rest of the population. Comparisons between extreme groups may yield large relative risks, but caution is required before causality is invoked. A G SHAPER GOYA WANNAMETHEE MARY WALKER
Department of Public Health and Primary Care, Royal Free Hospital School of Medicine,
Lohdon NW3 2PF I Anonymous. MRC to review dairy product and heart disease
findings. B.J 1991;302:550. (9 March.) 2 MRC Epidemiology Unit. Epidemiological studies of cardiovascular diseases. P'enarth, South Glamorgan: Llandough Hospital, 1991. (Progress report VII.) 3 Shaper AG, Pocock SJ, Walker M, Cohen NM1, Wale CJ, Thomson AG. British regional heart studv: cardiovascular risk factors in middle aged men in 24 towns. BM7 1981;283: 179-86. 4 Walker M1, Shaper AG. Follow-up of subjects in prospective studies based in general practice. J R Coil Gen Pract 1984;34: 365-70. 5 Morris JN, Marr JW, Clayton DG. t)iet and heart: a postscript.
BAII 1977;ii:1301-68.
786
adjusted for age, social class,
Thrombolytic treatment for recurrent myocardial infarction SIR,-Further to Dr Harvey White's editorial regarding thrombolytic treatment for recurrent myocardial infarction,' we have recently completed a postal survey of the age related admission and thrombolysis policies of the 175 coronary care units in the United Kingdom identified from the Directory of Emergency and Special Care Units 1990.2 As part of that survey we inquired into the use of thrombolytic drugs for recurrent myocardial infarction. Of the 175 questionnaires, 134 were returned. Recombinant tissue plasminogen activator was the agent most commonly given (by 123 units) to patients who had received previous treatment for thrombolysis. Three units used anistreplase as the sole alternative to streptokinase. Eight units had no alternative thrombolytic drug to streptokinase, one of which operated a policy of giving double the standard dose of streptokinase for reinfarction. Fifty seven of the 123 units using recombinant tissue plasminogen activator indicated a time policy for when streptokinase would not be given for the treatment of reinfarction. A total of 18 different policies were operated by these units. Two units gave recombinant tissue plasminogen activator to all patients who had previously received streptokinase, no matter the time interval since the initial dose. One unit gave streptokinase up to five days after the initial treatment but from then on only recombinant tissue plasminogen activator. Of the remaining 16 policies, the two most common were to give recombinant tissue plasminogen activator if streptokinase had been administered in the previous 12 months (23 units) or previous six months (14 units). Clearly, at present there is no consensus on the most appropriate management for recurrent myocardial infarction other than to give recombinant tissue plasminogen activator if available. Some of the policies currently used are likely to provide ineffectual treatment for many patients. If streptokinase is to be used for treating recurrent infarction it may be worth considering the use of skin testing to try to identify those patients at risk of developing anaphylaxis. A previous small study using 100 IU streptokinase intradermally showed this to be a sensitive and specific indicator of raised concentrations of IgE to streptokinase.' The test takes only 15 minutes so would not cause a great delay to starting treatment. After a negative test result higher doses of streptokinase could be given with more confidence of not precipitating a major anaphylactic reaction,
although the possibility of later allergic reactions of IgG to streptokinase would still remain. N J DUDLEY E BURNS Department of Medicine for the Eldcrly, St James's University Hospital, Leeds LS9 7TF I White H. Thrombolytic treatment for recurrent myocardial infarction. BM7 1991;302:429-30. (23 February.) 2 CMA Medical Data. Directory of emergency and special care units 1990. Cambridge: CMA Medical Data, 1990. 3 Dykewicz MS, McGrath KG, Davison R, Kaplan KJ, Patterson R. Identification of patients at risk for anaphylaxis due to streptokinase. Arch Intern Med 1986;146:305-7.
Patients with chest pain in accident and emergency departments SIR,-The article by Mr S S Tachakra and colleagues' contains a number of flaws that could lead others to adopt dangerous practices in patients who present with possible myocardial infarction. Firstly, there is no record, other than stating the number referred by their general practitioners, of the presentation. There is a world of difference between patients who drop in to an accident department with chest pain that has troubled them for a few days and those who experience a severe enough pain to cause them to dial 999 for an ambulance in the belief, usually correct, that they are experiencing a heart attack. Secondly, it is not safe to assume that the patients who did not respond to the postal survey had not had further trouble. An alternative explanation is that they had all died or been admitted to another hospital with myocardial infarction. Thirdly, Mr Tachakra and colleagues should not have assumed that all the patients in whom the pain had settled down had not had a heart attack. Most patients with documented myocardial infarction experience pain for a number of hours, but many are free of pain by the next day. Fourthly, there is an implication that it was sufficient to obtain an electrocardiogram to exclude myocardial infarction (or other serious cardiovascular disease such as aortic dissection). Standard medical textbooks emphasise that a normal tracing does not exclude myocardial infarction, particularly within the first few hours, and the patient with a typical history, particularly if it is supported by the presence of one or more major risk factors, must be assumed to have had or be in the process of developing a myocardial infarct. Indeed, many doctors do not wait for the development of changes on the electrocardiogram to start
thrombolytic therapy. Fifthly, Mr Tachakra and colleagues make no reference to the use of a short stay observation ward or area, which many doctors and accident staff think is the ideal way to manage patients with chest pain in whom the diagnosis is not immediately apparent. The authors make an important point, although it is dismissed in one sentence. This concerns the training of accident and emergency staffprobably the most important single factor in avoiding unnecessary misdiagnosis. However, they give
BMJ VOLUME 302
30 MARCH 1991
no details of the training that their staff receives. The number of medicolegal cases of misdiagnosis of chest pain continues to be alarmingly high, so much so that the Medical Protection Society has highlighted it in its latest training video. The rather complacent attitude of the authors of this article could worsen rather than improve the situation. S C JORDAN Department of Cardiology, Bristol Royal Infirmary, Bristol BS2 8HW I Tachakra SS, Pausey S, Beckett Al, Potts D, Idowu A. Outcome of patients with chest pain discharged from an accident and emergency department. BMJ 1991;302:504-5. (2 March.)
AUTHORS' REPLY,-The purpose of our study was to determine whether or not the junior doctors working in an accident and emergency department could, with adequate training, safely determine which patients are likely to have had a myocardial infarction and require admission and which do not. It is by no means uncommon for walking patients with apparently trivial chest pains to have been found to have suffered a recent myocardial infarct, and conversely some patients arriving by ambulance are found to have trivial conditions. In our study 33 of 122 (27%) walking patients were admitted and 47 of 133 (35%) who arrived by ambulance were discharged. In response to Dr Jordan's second point, we should have made it clear that if a patient did not respond to three letters the general practitioner was contacted. Some patients who were discharged may have had small infarcts and subsequently settled without further symptoms. These patients could be detected only by a study that included serial electrocardiography and measurement of cardiac enzymes in patients discharged with a diagnosis of non-cardiac pain; this was beyond the scope of our study. Such a project, though interesting, would be difficult to perform. We agree that every casualty officer must be aware that it is possible to have a normal electrocardiogram in a patient suffering a myocardial infarct. In most cases, however, the tracing will show some abnormality and in cases of doubt, a second electrocardiogram after one hour is often very helpful. We believe that it is a wise precaution to obtain an electrocardiogram in all patients over 50 years presenting with chest pain as a routine. Patients may trivialise their symptoms and try to persuade the doctor that the pain is due to a muscular injury or other minor conditions. An electrocardiogram, even if not clearly diagnostic, is a useful reminder to the busy casualty officer to consider cardiac disease. We agree with Dr Jordan on the overriding importance of training and supervision of casualty officers. We believe that compulsory attendance at a formal weekend induction course is beneficial. Such a course for newly appointed casualty officers has been run at Central Middlesex Hospital since 1984, and we have been fortunate in being able to attract high quality speakers from a large area. This introduction needs to be reinforced by continuous tuition in the department by motivated middle and senior grade staff. S STACHAKRA S PAWSEY
M BECKETT D POTTS A IDOWU
Accident and Emergency Department, Central Middlesex Hospital, London NW10 7NS
302
30
1 Rawlins MD. Extending the role of the community pharmacist. BAJ 1991;302:427-8. (23 February.)
Child health computing SIR,-Mr D J Hewitt's response' to the editorial by Drs Euan Ross and Norman Begg on child health computing2 suggests that his interest in health information, like that of most NHS management, is limited to that necessary for "purchasing" or invoicing. Most of the present impetus for better information systems in the NHS is driven by these requirements. This is not much comfort to doctors in public health medicine or community paediatrics, who need good population data. The child health system is much more than a call and recall system. It holds a record of the child's development from birth to school leaving age, details of immunisation state, and other information such as special needs-for example, handicapping conditions. This constitutes a huge databank on the child population that is just starting to be exploited by paediatric epidemiologists. To suggest that all that community paediatricians need is historical data from general practitioners carrying out child health surveillance shows a lack of understanding of the information needs of community paediatricians (and an unrealistic expectation of general practitioners). A ABRA
SIR,-The Royal Pharmaceutical Society of Great Britain is pleased that Professor Michael D Rawlins generally supports the development of the role of the community pharmacist, particularly in VOLUME
JOHN FERGUSON Secretary and registrar, Royal Pharmaceutical Society of Great Britain, London SEl 7JN
Department of Child Health, St Richard's Hospital, Chichester, West Sussex
Community pharmacist
BMJ
the provision of more advice on health care and of diagnostic services to a specified standard in the community pharmacy setting.' Professor Rawlins can be assured that the society has already taken action to ensure that the education and training of pharmacists meet the requirements for these developing roles. Therapeutics was added to the pharmacy degree course some years ago, and response to symptoms now forms an integral part of core tuition in every course. The structured preregistration experience programme is being revised to make it competency based. Social and behavioural sciences will feature prominently in both the degree and the programme, but the essential basis of pharmaceutical science in the undergraduate course will be maintained, as it must be if pharmacists are to continue to be expert in the ways in which the formulation of medicinal products affects treatment. In recent years there has also been a substantial increase in funding from the Department of Health for continuing education for pharmacists working in the NHS, and the council of the society is tackling the thorny question of the assessment of competence in practice. Training courses for support staff on medicines counters have been developed in recent years. Well trained assistants recognise when the pharmacists needs to be personally involved, and the pharmacist in turn recognises when reference to a medical practitioner is essential. The full potential of community pharmacists, not only in advising on the treatment of symptoms of self limiting ailments but also in activities designed to promote good health, must be tapped for the future benefit of the NHS. Doctors, pharmacists, and other members of the health care team need to coordinate their activities and ensure that the advice given by various sectors of the health service is consistent.
MARCH
1991
1 Hewitt DJ. Child health computing. BMJ 1991;302:409-10. (16 February.) 2 Ross E, Begg N. Child health computing. BMJ7 1991;302:5-6. (5 January.)
Childhood immunisation SIR,-Dr Peter T Rudd emphasises the importance of the new general practitioner contract in achieving better protection of children against infectious diseases.' The contract, introduced on 1 April 1990, includes a target payment system in an attempt to boost vaccine uptake. Unfortunately this new system, which replaces the old item of service payments, includes two anomalies that unnecessarily threaten the success of the programme; they would require only minor contractual modifications to be completely rectified. General practitioners can no longer be paid for giving pertussis immunisation to a child after the third birthday or for giving measles, mumps, and rubella immunisation to children between the third and sixth birthdays. Before this new contract they were paid item of service fees for giving these immunisations at these ages. Nothing has changed in the interim to make them any less important; indeed there is increasing recognition of the need for high uptake of immunisation at all ages.2 There is no medical reason why these anomalies should persist. It is our belief that they exist because of administrative error rather than medical intent, yet requests for modification have thus far been rejected. If measles, mumps, and rubella immunisation can be justified in a 6 year old why not in a 4 year old? If pertussis immunisation can be justified in a one year old why not in a 3 year old? Although whooping cough is a less serious infection in older children, complications do occur in this group, and these children can act as a source of infection for younger siblings. There are two possible solutions to this problem. An amendment to the contract could allow directors of public health to authorise payment of general practitioners by health authorities on an item of service basis for immunising children outside the currently recognised age bands. Alternatively, on a national basis, item of service payments could be extended for these vaccines to any age up to 18 years. The first would be the simplest to bring into effect, but the second would offer a national solution that avoids the obvious drawbacks of regional variations in immunisation
practice. The Joint Committee on Vaccination and Immunisation and the Department of Health would do well to examine this situation and rectify the anomalies as soon as possible. Efforts to eradicate these infections should not be hampered by unnecessary obstacles. DREW WALKER DONALD COID I G JONES
Fife Health Board, Glenrothes, Fife KY7 5PB 1 Rudd PT. Childhood immunisation in the new decade. BM7 1991;302:481-2. (2 March.) 2 Department of Health, Hong Kong. Expanded programme on unmunisation-measles outbreak. Weekly Epidemiological Record 1990;49:379-81.
SIR,-Dr Peter Rudd's editorial rightly emphasises that immunisation rates in the United Kingdom are presently at record levels,' but we question the validity of attributing the reasons given-introduction of immunisation targets into the general practitioner contract, the publication of a guide to immunisation by the Department of Health,2 and the publication of the British Paediatric Association's manual on infection and immunisation.3 These are all fairly recent events, whereas immunisation rates in the United Kingdom have been steadily increasing over the past 10 years. The figure, based on statistics from the Department of Health and the Scottish Office, shows immunisation uptake rates in 1979-89 for children who achieved their second birthday.
787
Age (years)
Manual workers
50-8 50-4
59 67 71
50-0 49-7
46 51
50-7
Current smokers (%)
Obese*
38 50 49 29 36
Systolic blood
(%)
Serum total cholesterol (mmol/l)
27 22 22 16 16
6-36 6-32 6-31 6-25 6-30
146-0 146 9 146-4 143-9 142-7
pressure (mm Hg)
Inactive Vigorous (%) (%) 13 11 11 7 6
Recall doctors' diagnosis
Ischaemic heart disease ("/.)
Diabetes (/)
13 7 4 7 3
2-1 1-7 1.0
20 18 20 24 25
Drinking
Non-drinkers Heavy drinkers
1-5
1-8
(%)
(%)
11 5 6 6 7
16 15 16 6 5
*Body mass index ¢28-0 kg/m-. TABLE v-British regional heart study. Fat spread usually used and major ischaemic heart disease events over 9-5 years offollow up Fat spread
No of men
No (%) with ischaemic heart disease
Relative risk (95% confidence interval) of heart attack*
136 2735 1038 3808
11(8-1) 247 (9 0) 66 (6-4) 283 (7-4)
0-57(0-22to 1-46) 1 00 0-76 (055 to 1-04) 0-87 (0-79 to 1-06)
None
NMargarine only Margarine and butter Butter only
*After exclusion of men with recall of doctor's diagnosis of ischaemic heart disease at screening; cigarette smoking, and blood cholesterol concentration.
diagnosis of ischaemic heart disease than men using butter (10 1% v 4 2%). These findings strongly suggest that pre-existing illness (obesity and heart disease) is associated with choice of fat spread. After exclusion of the men with recall of a doctor's diagnosis of ischaemic heart disease and adjustment for other key risk factors no significant association was seen between butter intake and heart attack rates. There seems to be no evidence that butter eating is protective against ischaemic heart disease. Conclusions-Data from the British regional heart study show that men who had the highest milk intake at initial screening had a lower rate of heart attack than men who drank no milk. Similarly, men who used only butter as a spread at initial screening had a lower rate of heart attack than men who used margarine only or no spread at all. However, the characterstics of the men in the several milk drinking or fat spread categories are very different, and these differences must be taken into account when assessing the importance of the association in terms of possible causality. When these background characteristics are taken into account we can find no significant association between milk intake or fat spread use and the incidence of heart attack in these middle aged British men. People with disease often change their lifestyle and this must be fully considered when attempting to associate any specific behaviour with a disease end point and to interpret the findings in terms of causality. Men in the United Kingdom who do not drink milk at all or do not use butter are small groups who differ in many ways from the rest of the population. Comparisons between extreme groups may yield large relative risks, but caution is required before causality is invoked. A G SHAPER GOYA WANNAMETHEE MARY WALKER
Department of Public Health and Primary Care, Royal Free Hospital School of Medicine,
Lohdon NW3 2PF I Anonymous. MRC to review dairy product and heart disease
findings. B.J 1991;302:550. (9 March.) 2 MRC Epidemiology Unit. Epidemiological studies of cardiovascular diseases. P'enarth, South Glamorgan: Llandough Hospital, 1991. (Progress report VII.) 3 Shaper AG, Pocock SJ, Walker M, Cohen NM1, Wale CJ, Thomson AG. British regional heart studv: cardiovascular risk factors in middle aged men in 24 towns. BM7 1981;283: 179-86. 4 Walker M1, Shaper AG. Follow-up of subjects in prospective studies based in general practice. J R Coil Gen Pract 1984;34: 365-70. 5 Morris JN, Marr JW, Clayton DG. t)iet and heart: a postscript.
BAII 1977;ii:1301-68.
786
adjusted for age, social class,
Thrombolytic treatment for recurrent myocardial infarction SIR,-Further to Dr Harvey White's editorial regarding thrombolytic treatment for recurrent myocardial infarction,' we have recently completed a postal survey of the age related admission and thrombolysis policies of the 175 coronary care units in the United Kingdom identified from the Directory of Emergency and Special Care Units 1990.2 As part of that survey we inquired into the use of thrombolytic drugs for recurrent myocardial infarction. Of the 175 questionnaires, 134 were returned. Recombinant tissue plasminogen activator was the agent most commonly given (by 123 units) to patients who had received previous treatment for thrombolysis. Three units used anistreplase as the sole alternative to streptokinase. Eight units had no alternative thrombolytic drug to streptokinase, one of which operated a policy of giving double the standard dose of streptokinase for reinfarction. Fifty seven of the 123 units using recombinant tissue plasminogen activator indicated a time policy for when streptokinase would not be given for the treatment of reinfarction. A total of 18 different policies were operated by these units. Two units gave recombinant tissue plasminogen activator to all patients who had previously received streptokinase, no matter the time interval since the initial dose. One unit gave streptokinase up to five days after the initial treatment but from then on only recombinant tissue plasminogen activator. Of the remaining 16 policies, the two most common were to give recombinant tissue plasminogen activator if streptokinase had been administered in the previous 12 months (23 units) or previous six months (14 units). Clearly, at present there is no consensus on the most appropriate management for recurrent myocardial infarction other than to give recombinant tissue plasminogen activator if available. Some of the policies currently used are likely to provide ineffectual treatment for many patients. If streptokinase is to be used for treating recurrent infarction it may be worth considering the use of skin testing to try to identify those patients at risk of developing anaphylaxis. A previous small study using 100 IU streptokinase intradermally showed this to be a sensitive and specific indicator of raised concentrations of IgE to streptokinase.' The test takes only 15 minutes so would not cause a great delay to starting treatment. After a negative test result higher doses of streptokinase could be given with more confidence of not precipitating a major anaphylactic reaction,
although the possibility of later allergic reactions of IgG to streptokinase would still remain. N J DUDLEY E BURNS Department of Medicine for the Eldcrly, St James's University Hospital, Leeds LS9 7TF I White H. Thrombolytic treatment for recurrent myocardial infarction. BM7 1991;302:429-30. (23 February.) 2 CMA Medical Data. Directory of emergency and special care units 1990. Cambridge: CMA Medical Data, 1990. 3 Dykewicz MS, McGrath KG, Davison R, Kaplan KJ, Patterson R. Identification of patients at risk for anaphylaxis due to streptokinase. Arch Intern Med 1986;146:305-7.
Patients with chest pain in accident and emergency departments SIR,-The article by Mr S S Tachakra and colleagues' contains a number of flaws that could lead others to adopt dangerous practices in patients who present with possible myocardial infarction. Firstly, there is no record, other than stating the number referred by their general practitioners, of the presentation. There is a world of difference between patients who drop in to an accident department with chest pain that has troubled them for a few days and those who experience a severe enough pain to cause them to dial 999 for an ambulance in the belief, usually correct, that they are experiencing a heart attack. Secondly, it is not safe to assume that the patients who did not respond to the postal survey had not had further trouble. An alternative explanation is that they had all died or been admitted to another hospital with myocardial infarction. Thirdly, Mr Tachakra and colleagues should not have assumed that all the patients in whom the pain had settled down had not had a heart attack. Most patients with documented myocardial infarction experience pain for a number of hours, but many are free of pain by the next day. Fourthly, there is an implication that it was sufficient to obtain an electrocardiogram to exclude myocardial infarction (or other serious cardiovascular disease such as aortic dissection). Standard medical textbooks emphasise that a normal tracing does not exclude myocardial infarction, particularly within the first few hours, and the patient with a typical history, particularly if it is supported by the presence of one or more major risk factors, must be assumed to have had or be in the process of developing a myocardial infarct. Indeed, many doctors do not wait for the development of changes on the electrocardiogram to start
thrombolytic therapy. Fifthly, Mr Tachakra and colleagues make no reference to the use of a short stay observation ward or area, which many doctors and accident staff think is the ideal way to manage patients with chest pain in whom the diagnosis is not immediately apparent. The authors make an important point, although it is dismissed in one sentence. This concerns the training of accident and emergency staffprobably the most important single factor in avoiding unnecessary misdiagnosis. However, they give
BMJ VOLUME 302
30 MARCH 1991
no details of the training that their staff receives. The number of medicolegal cases of misdiagnosis of chest pain continues to be alarmingly high, so much so that the Medical Protection Society has highlighted it in its latest training video. The rather complacent attitude of the authors of this article could worsen rather than improve the situation. S C JORDAN Department of Cardiology, Bristol Royal Infirmary, Bristol BS2 8HW I Tachakra SS, Pausey S, Beckett Al, Potts D, Idowu A. Outcome of patients with chest pain discharged from an accident and emergency department. BMJ 1991;302:504-5. (2 March.)
AUTHORS' REPLY,-The purpose of our study was to determine whether or not the junior doctors working in an accident and emergency department could, with adequate training, safely determine which patients are likely to have had a myocardial infarction and require admission and which do not. It is by no means uncommon for walking patients with apparently trivial chest pains to have been found to have suffered a recent myocardial infarct, and conversely some patients arriving by ambulance are found to have trivial conditions. In our study 33 of 122 (27%) walking patients were admitted and 47 of 133 (35%) who arrived by ambulance were discharged. In response to Dr Jordan's second point, we should have made it clear that if a patient did not respond to three letters the general practitioner was contacted. Some patients who were discharged may have had small infarcts and subsequently settled without further symptoms. These patients could be detected only by a study that included serial electrocardiography and measurement of cardiac enzymes in patients discharged with a diagnosis of non-cardiac pain; this was beyond the scope of our study. Such a project, though interesting, would be difficult to perform. We agree that every casualty officer must be aware that it is possible to have a normal electrocardiogram in a patient suffering a myocardial infarct. In most cases, however, the tracing will show some abnormality and in cases of doubt, a second electrocardiogram after one hour is often very helpful. We believe that it is a wise precaution to obtain an electrocardiogram in all patients over 50 years presenting with chest pain as a routine. Patients may trivialise their symptoms and try to persuade the doctor that the pain is due to a muscular injury or other minor conditions. An electrocardiogram, even if not clearly diagnostic, is a useful reminder to the busy casualty officer to consider cardiac disease. We agree with Dr Jordan on the overriding importance of training and supervision of casualty officers. We believe that compulsory attendance at a formal weekend induction course is beneficial. Such a course for newly appointed casualty officers has been run at Central Middlesex Hospital since 1984, and we have been fortunate in being able to attract high quality speakers from a large area. This introduction needs to be reinforced by continuous tuition in the department by motivated middle and senior grade staff. S STACHAKRA S PAWSEY
M BECKETT D POTTS A IDOWU
Accident and Emergency Department, Central Middlesex Hospital, London NW10 7NS
302
30
1 Rawlins MD. Extending the role of the community pharmacist. BAJ 1991;302:427-8. (23 February.)
Child health computing SIR,-Mr D J Hewitt's response' to the editorial by Drs Euan Ross and Norman Begg on child health computing2 suggests that his interest in health information, like that of most NHS management, is limited to that necessary for "purchasing" or invoicing. Most of the present impetus for better information systems in the NHS is driven by these requirements. This is not much comfort to doctors in public health medicine or community paediatrics, who need good population data. The child health system is much more than a call and recall system. It holds a record of the child's development from birth to school leaving age, details of immunisation state, and other information such as special needs-for example, handicapping conditions. This constitutes a huge databank on the child population that is just starting to be exploited by paediatric epidemiologists. To suggest that all that community paediatricians need is historical data from general practitioners carrying out child health surveillance shows a lack of understanding of the information needs of community paediatricians (and an unrealistic expectation of general practitioners). A ABRA
SIR,-The Royal Pharmaceutical Society of Great Britain is pleased that Professor Michael D Rawlins generally supports the development of the role of the community pharmacist, particularly in VOLUME
JOHN FERGUSON Secretary and registrar, Royal Pharmaceutical Society of Great Britain, London SEl 7JN
Department of Child Health, St Richard's Hospital, Chichester, West Sussex
Community pharmacist
BMJ
the provision of more advice on health care and of diagnostic services to a specified standard in the community pharmacy setting.' Professor Rawlins can be assured that the society has already taken action to ensure that the education and training of pharmacists meet the requirements for these developing roles. Therapeutics was added to the pharmacy degree course some years ago, and response to symptoms now forms an integral part of core tuition in every course. The structured preregistration experience programme is being revised to make it competency based. Social and behavioural sciences will feature prominently in both the degree and the programme, but the essential basis of pharmaceutical science in the undergraduate course will be maintained, as it must be if pharmacists are to continue to be expert in the ways in which the formulation of medicinal products affects treatment. In recent years there has also been a substantial increase in funding from the Department of Health for continuing education for pharmacists working in the NHS, and the council of the society is tackling the thorny question of the assessment of competence in practice. Training courses for support staff on medicines counters have been developed in recent years. Well trained assistants recognise when the pharmacists needs to be personally involved, and the pharmacist in turn recognises when reference to a medical practitioner is essential. The full potential of community pharmacists, not only in advising on the treatment of symptoms of self limiting ailments but also in activities designed to promote good health, must be tapped for the future benefit of the NHS. Doctors, pharmacists, and other members of the health care team need to coordinate their activities and ensure that the advice given by various sectors of the health service is consistent.
MARCH
1991
1 Hewitt DJ. Child health computing. BMJ 1991;302:409-10. (16 February.) 2 Ross E, Begg N. Child health computing. BMJ7 1991;302:5-6. (5 January.)
Childhood immunisation SIR,-Dr Peter T Rudd emphasises the importance of the new general practitioner contract in achieving better protection of children against infectious diseases.' The contract, introduced on 1 April 1990, includes a target payment system in an attempt to boost vaccine uptake. Unfortunately this new system, which replaces the old item of service payments, includes two anomalies that unnecessarily threaten the success of the programme; they would require only minor contractual modifications to be completely rectified. General practitioners can no longer be paid for giving pertussis immunisation to a child after the third birthday or for giving measles, mumps, and rubella immunisation to children between the third and sixth birthdays. Before this new contract they were paid item of service fees for giving these immunisations at these ages. Nothing has changed in the interim to make them any less important; indeed there is increasing recognition of the need for high uptake of immunisation at all ages.2 There is no medical reason why these anomalies should persist. It is our belief that they exist because of administrative error rather than medical intent, yet requests for modification have thus far been rejected. If measles, mumps, and rubella immunisation can be justified in a 6 year old why not in a 4 year old? If pertussis immunisation can be justified in a one year old why not in a 3 year old? Although whooping cough is a less serious infection in older children, complications do occur in this group, and these children can act as a source of infection for younger siblings. There are two possible solutions to this problem. An amendment to the contract could allow directors of public health to authorise payment of general practitioners by health authorities on an item of service basis for immunising children outside the currently recognised age bands. Alternatively, on a national basis, item of service payments could be extended for these vaccines to any age up to 18 years. The first would be the simplest to bring into effect, but the second would offer a national solution that avoids the obvious drawbacks of regional variations in immunisation
practice. The Joint Committee on Vaccination and Immunisation and the Department of Health would do well to examine this situation and rectify the anomalies as soon as possible. Efforts to eradicate these infections should not be hampered by unnecessary obstacles. DREW WALKER DONALD COID I G JONES
Fife Health Board, Glenrothes, Fife KY7 5PB 1 Rudd PT. Childhood immunisation in the new decade. BM7 1991;302:481-2. (2 March.) 2 Department of Health, Hong Kong. Expanded programme on unmunisation-measles outbreak. Weekly Epidemiological Record 1990;49:379-81.
SIR,-Dr Peter Rudd's editorial rightly emphasises that immunisation rates in the United Kingdom are presently at record levels,' but we question the validity of attributing the reasons given-introduction of immunisation targets into the general practitioner contract, the publication of a guide to immunisation by the Department of Health,2 and the publication of the British Paediatric Association's manual on infection and immunisation.3 These are all fairly recent events, whereas immunisation rates in the United Kingdom have been steadily increasing over the past 10 years. The figure, based on statistics from the Department of Health and the Scottish Office, shows immunisation uptake rates in 1979-89 for children who achieved their second birthday.
787
