3 4

5 6

of atenolol, nifedipine and captopril combined with bendrofluazide. Diabetic Med 1987;4:164-8. Ritchie CI, McGrath E, Hadden DR, Weaver JA, Kennedy L, Atkinson AB. Renal artery stenosis in hypertensive diabetic patients. Diabetic Med 1988;5:265-7. Chantler G, Garrett ES, Parsons V, Veall N. Glomerular filtration rate measurement in man by the single injection method using 5Cr-EDTA. Clin Sci 1969;37:169-80. Levy AS, Perrone RD, Madias NE. Serum creatinine and renal function. Annual Review of Medicine 1988;39:465-90. Gabriel R. Time to scrap creatinine clearance? BMJ 1986;293: 1119-20.

First line treatment in hypertension SIR,-Professor J D Swales's editorial' comes at a propitious time, a time when the government is endorsing health promotion clinics that include managing hypertension.' We believe, however, that many issues in the contemporary management of hypertension deserve inclusion in the recommendations for first line treatment. Firstly, non-pharmacological management options, including weight reduction, restriction of dietary salt intake, and alcohol limitation should, in our view, be the real first line treatment. This approach needs more emphasis before discussing the issue of which drug to use. All drug treatments, even those two groups of drugs favoured by Professor Swales, have been associated with well documented subjective as well as metabolic side effects.2 Secondly, the failure oftrials of thiazide diuretics and i blockers to find an effect on coronary heart disease indicates the possible importance of other risk factors and the complex interactions with antihypertensive treatment. These observations are made against a background of an alarmingly high incidence of cardiovascular disease in industrialised countries (>150 000 deaths/year in the United Kingdom' and >750000/year in the United States4). With recent data showing that over half the hypertensive population have a more atherogenic lipid profile than do normotensive subjects 6 influences on lipid and lipoprotein concentrations may be critical in determining atherosclerotic disease related to hypertension. The impact of lipid lowering on coronary heart disease is well documented. Consensus recommendations from both the United Kingdom7 and the United States' advise caution in using drugs that adversely affect lipid and lipoprotein concentrations, including thiazides and i blockers. Furthermore, thiazides and i blockers would be inappropriate in diabetic subjects, who have a high prevalence of hypertension and large vessel complications. There is good documentation that these drugs may worsen glucose tolerance and promote an atherogenic lipid profile in both diabetic and non-diabetic patients.' "' The newer drugs, by comparison, have a neutral or a beneficial effect on lipid concentrations.'0 There may, however, be a good case for using diuretics at substantially lower doses than officially recommended as effective blood pressure reduction has been observed with reduced adverse effects on other risk factors." Thirdly, an assessment of the costs and benefits, including risk factors for coronary heart disease should be favoured,' and not just a simple comparison of the cost effectiveness of the "older" and "newer" antihypertensive drugs. Of importance in cost benefit analyses are not only the effects on glucose homoeostasis and a potential atherogenic lipid profile but the reduced hypotensive effect of 0 blockers in people who smoke2 and black people with hypertension"'2; their unsuitability in the elderly people; and their wider side effects. We suggest that antihypertensive treatment should be personalised and take into account other risk factors." 4 The benefits of this approach will accrue favourably in population mortality statistics without the need to conduct expensive trials with more expensive drugs. We are falling behind our

116

transatlantic colleagues in not accepting the therapeutic advances of the past decade.'2 "We are fortunate to enter the 1990s with a wide range of drugs for the treatment of hypertension. This should allow the choice of first line drug to be widened. MICHAEL D FEHER ARIEL LANT Charing Cross and Westminster Hospital School, London SW1 2AP 1 Swales JD. First line treatment in hypertension. BMJ 1990; 301:1 172-3. (24 November.) 2 Medical Research Council Working Party. MRC trial of treatment of mild hypertension: principal results. BMJ7

1985;291:97-104. 3 Office of Health Economics. Coronary heart disease. The need for action. London: HMSO, 1990. 4 Leaf A, Ryan TJ. Prevention of coronary artery disease. N Engl

J3Med 1990;330:1416-9. 5 Castelli WP, Anderson K. A population at risk. Prevalence of high cholesterol levels in hypertensive patients in the Framingham study. Amj Med 1986;80 (supply 2A):23-32. 6 Curzio JL, Kennedy FS, Elliot H, et al. Hypercholesterolaemia in treated hypertensives. J Hypertens 1989;7 (suppl 6): 254-65. 7 Shepherd J, Betteridge DJ, Durrington P, et al. Strategies for reducing coronary heart disease and desirable limits for blood lipid concentrations: guidlines of the British Hyperlipidaemia Association. BMJ 1987;295:1245-6. 8 Consensus conference. Lowering blood cholesterol to prevent heart disease. JAMA 1985;253:2080-6. 9 Feher MD, Henderson AD, Wadsworth J, et al. Alpha-blocker therapy; a possible advance in the treatment of diabetic hypertension-results of a cross-over study of doxazosin and atenolol monotherapy in hypertensive non-insulin dependent diabetic subjects. J Hum Hypertens 1990;4:571-7. 10 Weidmann P, Ferrier C, Saxenhofer H, et al. Serum Lipoproteins during treatment with antihypertensive drugs. Drugs 1988;35 (suppl 6): 118-34. 11 Carlsen JE, Kober L, Torp-Pedersen C, Johansen P. Relation between dose of bendrofluazide, antihypertensive effect and

adverse biochemical effects. BMJ 1990;300:975-8. (14 April.) 12 Weinberger MH. Racial differences in antihypertensive therapy; evidence and implications. Cardiovascular Drugs and Therapy

reserved for patients with uncomplicated hypertension. Unfortunately, over 85% of hypertensive patients have adverse lipid profiles,' which merit intervention (as recommended by the European Atherosclerosis Society and the British Hyperlipidaemia Association); up to half have left ventricular hypertrophy4; and over 10% have glucose intolerance (P McKeigue, unpublished observations). The use of diuretics and 13 blockers in the face of any of these three complications is at best suboptimal and probably contraindicated.55 NEIL POULTER SIMON THOM PETER SEVER

St Mary's Hospital, London W2 INY 1 Swales JD. First line treatment in hypertension. BM. 1990;301: 11 72-3. (24 November.) 2 Mascioli S, Svendsen K, Grimm R, et al. One year lipid changes by class of anti-hypertensive drug in the treatment of mild hypertension study (TOMHS). In: Proceedings of second international conference on preventive cardiology. Dallas: American Heart Association, 1989:86. 3 Langdon CG. Doxazosin: a study in a cohort of patients with hypertension in general practice-an interim report. Am Heartj (in press). 4 Hammond IW, Devereux RB, Alderman MH, et al. The prevalence and correlates of echocardiographic left ventricular hypertrophy among employed patients with uncomplicated hypertension. J Am Coll Cardiol 1986;7:639-50. 5 Weidman P, Vehlinger DE, Gerber A. Antihypertensive treatment and serum lipids. J Hypertens 1985;3:297-306. 6 Drayer JI, Gardin JM, Weber MA, Aronow WS. Changes in ventricular septal thickness during diuretic therapy. Clin Phar,nacol Ther 1982;32:283-8. 7 Murphy MH, Kohner E, Lewis PJ, Schumer B, Dollery CT. Glucose intolerance in hypertensive patients treated with diuretics: a fourteen year follow-up. Lancet 1982;ii: 1293-5. 8 British Heart Foundation. Fact file on treatment of hypertension. London: British Heart Foundation, 1990.

1990;4:379-82. 13 Kannel WB. Hypertension: relationship with other risk factors. Drugs 1986;31(suppl 1):1-11. 14 Moser M. Antihypertensive medications; relative effectiveness and adverse reactions. J Hypertens 1990;8 (suppl 2):S9-16.

SIR,-Professor J D Swales's recent editorial, though pragmatic, ignores several key issues. ' It is true that there are no data on long term morbidity and mortality with the newer drugs (a blockers, angiotensin converting enzyme inhibitors, and calcium antagonists), and even if a trial were to start tomorrow the results would not be available for 10 years. Unfortunately, no such data may ever be forthcoming owing to the prohibitive costs. But does this mean that we must persevere slavishly with diuretics and ,B blockers, which have repeatedly been shown to be disappointing in terms of reducing coronary heart disease? Three to four myocardial infarctions attributable to elevated blood pressure occur for each stroke; consequently preventing coronary heart disease must be the primary aim of treatment. We therefore cannot gloss over the shortfall in the effect on coronary heart disease and ignore the potential benefits of the newer drugs. We do have data suggesting that the newer drugs may be better than diuretics and 1 blockers in terms of preventing coronary heart disease. The treatment of mild hypertension study showed lasting, significant adverse effects on lipid profiles with a low dose diuretic (chlorthalidone) and a lipid friendly 13 blocker (acebutolol), whereas the newer drugs were shown to be lipid neutral or, in the case of the a blocker (doxazosin), to improve lipid profiles.2 The newer drugs do not have adverse effects on glucose intolerance and insulin resistance in contrast with diuretics and 13 blockers; nor do they produce the hyperuricaemia and hypokalaemia induced by diuretics. All these variables have an adverse impact, directly or indirectly, on coronary heart disease events. Is it not more rational to be swayed by the advantages of the newer drugs than by subgroup analyses of the Medical Research Council's trial? We agree that diuretics and 0 blockers should be

General practice experience of patient recall SIR,-Half of the first year of the new general practitioner contract has now passed. We present an interim analysis of some of the extra work that we have been doing to meet our terms of service namely, recalls for elderly patients and those aged 16-74 not seen for three years. To spread the work evenly through the year we recalled our patients on a monthly basis by patient's month of birth. We can therefore review those patients born in the months April to September. We had 6236 patients registered on 1 April 1990: about a third were eligible for the deprivation allowance. We have an annual turnover of 20%. Elderly patients were invited to make an appointment and those aged 16-74 not seen for three years were offered a specific appointment in the afternoon, with an option to rearrange their appointment if it was unsuitable. The table shows the results; only five letters were returned "not known at this address." We have also taken over the list of a neighbouring practice. Recalling 26 of these patients to a clinic for their registration checks produced an attendance of three (with no cancellations). Attendance ofpatients recalled for routine check up

Age -_ 7 5 16-74: Men Women

No recalled

to attend

No(%) who attended

197

197

12

75 (38)

1049 961

205 154

8 5

16 (8) 30 (19)

No of patients on list

No who declined

Those general practitioners who have not begun recalling patients can fairly confidently book two elderly patients or five patients aged 16-74 for every appointment. The average general practitioner with a list of 2000 in a similar inner city practice would need to see about 80 patients

BMJ VOLUME 302

12 JANUARY 1991

annually by recall (note that we covered only half the annual workload in our table) to meet the terms of his or her service. He or she will therefore need to send about 360 invitations (that is, about 10 million letters for the whole of the United Kingdom). This does not include visits to elderly people, which must also be offered. As a practice with an interest in prevention we are disappointed, although not altogether surprised, at the attendance. It may be some consolation to note that the (unpublished) success rate from our local family health services authority smear recall system is 10-15%. We expect to improve our recall rate by providing appointments (particularly for men) in the evenings. PETER McCARTNEY GAIL COCHRANE London N I 8JQ

doctor's job. Extra fees are not paid to probation officers for writing court reports, to social workers for supporting letters to housing departments or signing mental health sections, or to school teachers for writing school reports. Why should doctors be paid an extra fee for completing a simple form? Undoubtedly, many doctors are overworked and underpaid. That is a fault of the salary structure and a quite separate issue. The question is, Why should one group of publicly employed professionals be treated so differently from others? I accept that it would be appropriate to charge a fee if a patient chooses to be treated privately. I wonder what other extra fees doctors are paid for work that should properly be regarded as part of their NHS duties.

SIR, -The use of dried blood samples from Guthrie cards to amplify DNA by the polymerase chain reaction for genetic investigation is well established. 2 This process is ofparticular importance in genetic diseases that have a lethal outcome in early infancy and in which blood storage has not previously been undertaken. My participation in a project to localise the gene for the childhood spinal muscular atrophies has led me to contact several regional neonatal screening laboratories in the United Kingdom to try to obtain a sample from a particular individual. I have been surprised by the variation in the length of time that laboratories store Guthrie cards, which in the laboratories I contacted ranged from two to 20 years. Further genetic conditions will undoubtedly be mapped to specific regions of the genome in the future. Populations served by laboratories whose financial or spatial resources limit storage times may be unable to take advantage of these genetic advances because the samples from which DNA may be generated have been destroyed. As some families will continue to depend on Guthrie cards as the only source of genetic material for antenatal testing, I wonder whether a process to save Guthrie cards from those children who die in infancy may be a feasible and less expensive method than storage of all cards. This would ensure that this valuable genetic resource is not denied to certain regions of the country.

SIR,-I am quite convinced that Dr Duncan Williams is correct when he says that the intention of fundholders is to improve patient care.' That, unfortunately, is the nub of the problem. As these centres of general practice "excellence" steadily grow on the subsidies and freedoms of fundholding they simultaneously tear the heart out of those practices judged too small to follow suit. A downward spiral is then set in motion. As more patients leave the smaller practices the service the practices are able to offer inevitably deteriorates because of-yes-lack of "funds." This, in turn, must be bad for patients. I would ask Dr Williams (and, indeed, the GMSC, which has managed to contort the position to face both ways) to think more carefully about the results of their actions before they complacently and short sightedly pull up the ladder. If they do not the current high standard of British general practice is practically guaranteed to deteriorate, and it will deteriorate because of their actions. It will be all the worse in rural areas.

1 McCabe ERB, Huang S-Z, Seltzer W-K, Law ML. DNA microextraction from dried blood spots on filter paper blotters: potential applications to newborn screening. Hum Genet 1987;75:213-6. 2 Jinks DC, Minter M, Tarver DA, Vanderford M, Heimancik JF, McCabe ERB. Molecular genetic diagnosis of sickle cell disease using dried blood specimens on blotters used for newborn screening. Hum Genet 1989;81:363-6.

Fees for completing attendance allowance reports SIR, -Some of my colleagues and I were surprised and disappointed to discover recently that doctors receive £13.90 for completing the medical statement for patients claiming attendance allowance under the new "special rules." The report form is very simple, and it cannot possibly take a doctor more than I1/2 minutes to complete. In fact it will probably take longer to complete the claim part of the form to say where and how the fee should be paid. Surely, completing forms for social security purposes for NHS patients should be part of a BMJ

VOLUME 302

12 JANUARY 1991

I Drummond M, Crump B, Hawkes R, Marchment M. General practice fundholding. BMJ 1990;301:1288-9. (8 December.)

Future of maternity services

Clarke's legacy

NEIL H THOMAS

W F G TUCKER P V S SCOTT

Alexandra Hospital, Redditch B98 7UB

CHRISTOPHER LAW

St Bartholomew's Hospital, London ECIA 7BE

Storage times for Guthrie cards

Royal Postgraduate Medical School, London W12 ONN

If they cannot achieve priority why should patients join a fundholding practice? If they do achieve priority what effect will that have on the ethic of the NHS: equal access to health provision for all? As Professor Drummond and colleagues said, "the biggest risk is that fundholding will increase the inequalities in health care provision if some practices can deliver a better standard of service than others."

DAVID ROBERTS

Dispensing Doctors' Association, Welford, Northampton NN6 7HG

SIR,-John Warden draws the profession's attention to the NHS Management Executive's instruction to health authorities to review maternity and neonatal services. ' Once again Bradford's poor perinatal figures are quoted, but they are only a part of the district's problems; all Bradford's standardised mortality ratios are above the national averages. This and the Jarman indices indicate that the city has a high risk population. Such a population requires an appropriately funded high quality service. Our plans to centralise maternity and neonatal services by 1994 have been set back by reductions in capital funding. Those working locally in health care are striving valiantly under great pressure-but until the district is funded adequately, taking into account the demography, we will continue to be judged unfairly. There seems to be a political habit to criticise without considering the population's needs. It is time to decide priorities and to target health care on those who need it and not on those who shout the loudest. MICHAEL L SMITH J G CRAIG Paediatric Unit, St Luke's Hospiial, Bradford BD5 ONA 1 Warden J. A friend at the top. (5 December.)

BMJ7

1990;301:1297.

1 Williams D. Clarke's legacy. BMJ 1990;301:1336. (8 December.)

Military secrecy and medical preparedness General practice fundholding SIR,-The editorial by Professor Michael Drummond and colleagues' is proving to be of great help to us as we try to negotiate hospital contracts with fundholding family doctors. We accept that there are many ways in which we can improve the system that provides hospital services for patients, and welcome the prospect of a fundholding general practitioner (or any general practitioner) insisting that we should improve it. We are as aware of the shortcomings as any general practitioner, and we are equally demoralised. Many improvements that could be made relate to hospital administration and in particular to the paucity of secretarial help for consultants. These are problems we can look at; whether our bankrupt hospitals can solve them is a moot point. Professor Drummond and colleagues write that "the other main benefits [of general practice fundholding] are likely to come ... from the greater access to hospital services that fundholding practices may negotiate." What does greater access mean? Does it mean speedier access? If so patients referred to hospital by fundholding doctors will expect to receive priority in outpatient appointments and inpatient treatment over patients from non-fundholding doctors: a two tier system, in fact.

SIR,-Should hostilities break out in the Middle East there will be many military and civilian casualties. With the rapidity of modern warfare and sophisticated evacuation and first aid, many patients will be returned to the United Kingdom in less time than it would have taken to reach a field dressing station during the battle of the Somme in 1915. It is then expected that much of the medical workload will fall on the NHS. There is widespread speculation that chemical, bacteriological, and tactical nuclear weapons may be used, but the nature of the expected chemical wounds in particular is veiled in secrecy. This information has not, at the time of writing, been circulated among plastic and burns surgeons. Much more precise information needs to be made available on the potential chemical agents and how to treat them, on the correct procedures for detecting and handling radioactive contamination, and on the risks of biological weapons. Military secrecy must not leave us unprepared. In all previous wars there have never been enough bandages. D A McGROUTHER N PARKHOUSE

Department of Surgery, University College and Middlesex School of Medicine, London WC1E 6JJ

117