860 SURGICAL TRAINING AND OVERSEAS DOCTORS SiR,—Professor Bevan (Sept 12, p 630) seems to miss a major point. There is, of course, no chance of the N...

328KB Sizes 4 Downloads 113 Views


SiR,—Professor Bevan (Sept 12, p 630) seems to miss a major point. There is, of course, no chance of the National Health Service (NHS) capsizing. Many third-world nations also have governmentrun health services that will not capsize but carry on. However, in the NHS both the quality as quantity of the service have deteriorated and private health insurance is flourishing. UK medical journals carry letters drawing attention to this deterioration and advertisements for large numbers of vacancies in orthopaedics and general surgery, for which the Royal College of Surgeons of England has started the training scheme discussed in your May 9 editorial. The Royal College of Surgeons is hatching this clever little scheme to get labour without investing to create it-to prop up the NHS. If this is not so the College should come out in favour of opening up all jobs and allow overseas doctors to compete for whatever posts they want, which is what they used to be able to do, and contribute to the running of the NHS. The Overseas Doctors Training Scheme Committee should not mislead other countries by hiding the fact that all jobs in the NHS used to be open to competition and could be again. It should not open just a small area where "hands" are. needed. The committee is being less than candid: "overseas doctors" seems to be a code word for something else.

citations was assumed to represent the importance scientists attached to the most original papers of each candidate. Data from 1974 until 1986 were obtained through the Dialog Information Retrieval Service from SCIESEARCH, an international multidisciplinary index produced by the Institute of Scientific Information, Philadephia, containing all the records published in Science Citation Index plus additional records from Current Contents. The number of publications and the impact scores were about 3 and 7 times higher, respectively, in the group of applicants who were not appointed (table). Four of the successful candidates had very few (0,1,1, and 3) citations. It would seem that the quality and quantity of the published output of candidates carried little weight in this selection committee’s deliberations. Institute of Occupational

University of Padua, 35127 Padua, Italy



Journal citation report: a bibliometnc analysis of science journals in the ISI Data Base. Science citation index annual: vol XVIII. Philadelphia: Institute for Scientific Information, 1985. 2. Gillett R, Aitkenhead M. Rank injustice in academic research. Nature 1987; 327: 381-82. 1. Garfield E. SCI:


1301 East Commonwealth Avenue, Fullerton, California 92631, USA



SIR,-In November, 1986, a committee appointed by the Italian Minister of Education published its evaluation of candidates for the position of associate professor of occupational medicine. To explore the importance of scientific productivity for an academic career in occupational medicine in Italy I examined the output of the fourteen candidates for the five posts available. The fourteen had been selected on the basis of their curricula vitae and published work. They were given an oral examination in two sessions-a discussion of publications submitted by the candidate to the committee and a lecture given by the candidate on a topic assigned to him 24 hours earlier. The committee of three professors and two associate professors of occupational medicine had been elected by their peers from a list drawn up by the Ministry of Education. The examinations were held in June, 1986. I examined the publication record of all five successful candidates and of eight of the nine not appointed. (Being myself one of the candidates who was not appointed, my records were not included, to avoid any bias in the analysis). I calculated three objective measures of quantity and quality of scientific productivitylznamely, the number of full papers, the score obtained by sum of products between each paper and the impact factor of the journal where it was published (impact score), and the number of citations received by the candidate for papers in which he was first author. The number of publications and the impact score were assumed to represent the quantity and quality of research, and the number of INDIVIDUAL AND MEAN




SIR,-Dr Gore and Dr Gilks (Aug 29, p 507) claim a potential saving to the National Health Service of at least cl653 000 if all kidneys were "beneficially" matched. The concept of beneficial matching derives from a retrospective analysis of a large number of UK Transplant Service (UKTS) first transplants from 1979 to 1984.1 The data have not been independently confirmed. While many would share the UKTS ideal of achieving the best use of a limited number of cadaver transplants, this cannot be viewed only in terms of increased long-term survival for those fortunate enough to receive them. Waiting for a beneficially matched cadaver kidney will increase the waiting times for transplant in the short term, which then places increased demands on stretched dialysis services. In Edinburgh, in common with other renal units, we offer about 50% of our locally obtained cadaver kidneys to UKTS and retain the other 50 % for transplantation locally, usually to a young hospital haemodialysis patient. This creates a space so that we can treat another patient with chronic renal failure. We are forced to continue this pragmatic policy while places for hospital haemodialysis are limited. Offering all our kidneys to UKTS for beneficial matching would, in the short term, severely curtail our transplant programme. Most beneficially matched kidneys will go to the largest transplant centres, some of which put fewer kidneys into UKTS than they take out. Although UKTS intends to address this imbalance, it cannot easily do that and pursue the concept of beneficial matching. In the longterm it may be that the number of kidneys we send to UKTS will be balanced once more by an equal number of kidneys received, 60% of which would be beneficially matched. In the meantime, at least half of our patients will have to wait longer for a transplanted kidney and the increased hospital and home dialysis costs would easily wipe out any economic benefit of the extra two years of graft function achieved with the beneficial match. Medical Renal Unit,

Royal Infirmary, Edinburgh EH3 9YW


1. Gilks WR, Bradley BA, Gore SM, Klouda PT. Substantial benefits of tissue matching in renal transplantation. Transplantation 1987; 43: 669-74.


*p < 0 02; tp < 0-01; tNS, all by Student’s t test.

SIR,-A full reply to Dr Skrabanek’s letter (Aug 29, p 510) would require a comparison of all of his statements with the published data, leading to a longer response than seems justified. Here, however, is one example of how he operates. He claims that by providing figures for a wider age group (20-75) we "falsely inflate the effect of screening since most of the older women were not


screened and yet they had a sharper fall in mortality rates (between 1967-70 and 1975-78) than the younger age group". In the paper referred to’ we find the following: (1) Age-specific mortality rates per year per 105 women for the age group 20-59 averaged9-8,16-6,20 6,17-9,10 2,and6 6,fbrthe six 4-year periods 1955-58 to 1975-78, showing a 66 % fall between 1963-70 and 1975-78. The corresponding fall for the age group 20-74 is 60 %. Thus our error in giving figures for age 20-74 in our July 25 letter cannot have resulted in false inflation since the percentage fall in mortality rate was less for the age group 20-74 than for the 20-59 age band on which Skrabanek concentrates. (2) In 1969 screening was extended to all women in Iceland aged 25-69. Fig 1 in the paper shows that among all women over 60, some 40% had been screened by the end of 1974 and some 70% by the end of 1977. Table I and fig 1 taken together indicate that among women aged 60-74, some 55% had been screened by the end of 1974 and over 85 % by the end of 1977. So much for Skrabanek’s remark that most women aged 60-74 were not screened. (3) Table n shows that the fall in the average age-specific mortality rate between 1967-70 and 1975-78 was 63% for the women aged 20-59 and 60 % for the women aged 60-74; if one takes the drop in rates between 1963-1970 and 1975-78, to cover more completely the period immediately before screening might be effective, the corresponding figures are 66% and 53%. Yet Skrabanek says the older age group had a sharper fall in mortality. Skrabanek clearly hopes that Lancet readers will not examine the articles that he cites. MRC Biostatistics Unit,

Shaftesbury Road, Cambridge CB2 2BW 5


1 Johannesson G, Gensson G, Day N, Tulinius H Screening for cancer of the uterine cervix in Iceland, 1965-1978. Acta Obstet Gynecol Scand 1982; 61: 199-203


SIR,-Dr Evans (Sept 5, p 574) highlights a very important problem. Many haemophiliac boys in the UK under the age of 17 are HIV antibody positive and are approaching the years when nature has programmed them for sexual experimentation. They must be told of their positivity and counselled patiently, expertly, kindly, and repeatedly. Teenagers pay little attention to contraception. Many will also disregard advice about smoking, drinking, and drug taking. Will they accept advice that will inhibit the spontaneity of their sexual activity? Perhaps not, but because of the consequences of unrestricted sexual activity in this group we dare not give up. My experience leads me to believe that the quality and intensity of counselling before the adolescent is told of his seropositivity will influence the acceptance of subsequent advice. An education programme about HIV disease is likely to precipitate inquiries from young people about their own HIV status. If the positive aspects of the disease have been emphasised a truthful answer can then be given more easily and the child should not be devastated. From then on counselling must gradually emphasise responsibility towards sexual partners. The logistics of such a programme may be daunting but this matter is serious and those of us who care for haemophiliac children should be addressing it urgently. Treloar

Haemophilia Centre, Holybourne, Alton, Hampshire



SIR,-I read with great interest the report of Dr Sinclair and

colleagues (July 4, p 38) on the prognosis of meningococcal septicaemia. Why did they select those seven factors to score and how did they allocate 1, 2, or 3 points? They proposed to apply this scoring system to all patients, in shock or not. However, patients not in shock usually survive, and no scoring system is necessary to predict a good outcome. How many patients without hypotension and

or a skin-rectum temperature difference of more than 30 °C (which are signs of shock) died? Would the score established by the

French Club of Paediatric Intensive Care for patients with shockl have allowed prediction of outcome in the patients studied by Sinclair et al? I agree that prognostic scoring systems are very important in the evaluation of new forms of treatment. Infant Resuscitation Unit,

Hôpital Calmette,


59037 Lille, France 1. Leclerc



Care Med




factor of severe infectious purpura


children. Intensive

11: 140-43.

SiR,—On Aug 10-25, 1987, we experienced a large epidemic of meningococcal meningitis in 700 patients, 70 of them children. We used the scoring system devised by Dr Sinclair and colleagues to rationalise intensive care; the number of the patients overstretched the services available. 10 patients presented with fulminant septicaemic and encephalitic meningococcaemia, of whom 8 died, confirming the high case fatality rate for these two manifestations. 12 All patients received "conventional medication". The outcome was clearly less favourable when conventional medication was given, in contrast to the claim by Dr Hampton’s group (Aug 15, p 395). Our experience confirms the validity of Sinclair’s scale in quickly predicting which children were likely to die despite intensive care. Immunological detection of meningococcal capsular antigen, to determine the severity, is not readily available in all centres.3 Combined plasmapheresis and leucapheresis or blood exchange4 are the only options left open, for the time being, for the desperately ill patients who can readily be identified by use of Sinclair’s scale. King Fahd Hospital, POB 7990 Jeddah, Saudi Arabia


1. Banks HS, Meningococcosis a protean disease. Lancet 1986; ii 635-40. 2. Banks HS. Meningococcosis a protean disease. Lancet 1984; ii: 677-81 3. Edwards EA. Immunologic investigation of meningococci disease I: Group-specific Neisseria meningitidis antigens present in the serum of patients with fulminant meningococcaemia. JImmunol 1971; 106: 314-17. 4. Bjorvatan B, Bjertnaes L, Fadnes HO, et al. Meningococcal septicaemia treated with combined plasmapheresis and leucapheresis or with blood exchange. Br Med J

1984, 288: 439-41.

SIR,-Dr Sinclair and colleagues pick out six bedside findings and one laboratory factor for evaluating prognosis in meningococcal septicaemia. Their selection agrees fairly well with our unpublished meta-analysis, a review of our own cases and published material. Sinclair’s method of factor weighting was not given. Our material1,2 showed that the most interesting factors were correlated so that weightings were not of decisive importance. For routine use, however, the composition of the score is very important. When choosing a standard score for clinical use, we think that a pure bedside clinical score is generally much more applicable in the important early stages before and on admission in hospital. Such a choice may help to improve understanding and communications between health-care staff. Multivariate analysis of our first material’ prompted us to choose factors to record on admission. Analysis of that material then led to our scoring system, with the following elements: systolic blood pressure (age adjusted) less than 100 mm Hg; cyanosis; ecchymosis (skin haemorrhages of at least 5 mm diameter); diarrhoea, before or on admission; cold extremities (significant skin/rectal temperature difference); no nuchal or back ridigity; rectal temperature of at least 4O.0oC. We score the percentage of these features found present. Sinclair et al stress the need for a good coma scale. We too found the degree of consciousness a feature worth elaborating on, especially for a late sequelae severity score. This score differs from scores aimed at predicting a fatal outcome,3’ and in animal experiments, clinical practice, and epidemiological research these two objectives must be differentiated. Dr Hampton and her colleagues (Aug 15, p 395) seem to presume that the main aim of prognostication is to pick out those who will inevitably die. Such an approach, to avoid stressful and expensive, but useless, intensive care, is seldom a practical option for meningococcal disease management in western countries today. However, with a high cut-off and a score validated as almost ideal, this aim is theoretically attainable. Hampton et al, among others,