Matthias Egger, senior researchfelow. Correspondence to: Dr Davey Siith. BMJ 1994;308:1025-7. Commentary on the cholesterol papers. Two months ago the ...
occur. For the time being trainee surgeons will stiU be exposed to open gastrectomies, pancreatectomies, and other complex procedures, but as laparoscopic surgery becomes more and more advanced and these more complicated procedures become available to the laparoscopic approach, training in all open procedures will be a problem for the future.
Minimally invasive centres The government is setting up three centres for minimally invasive surgery: in London, Leeds, and Scotland. It is hard to see exactly what role these centres will play. They will certainly be able to teach with simulators and videos, and trainees will be able to observe surgical procedures being performed. However, they will not be able to to use porcine models. The centres may act as a focus for the organisation of hands-on courses and proctorships in the regions but are unlikely to be able to run them by themselves. The hands-on courses that have been run to date have been organised almost exclusively by industry, with obvious beneficial effects for both the surgical profession and the companies involved. It is a sad reflection on surgical teaching, however, that we have to rely on commercial companies to organise the teaching of a new technique for surgeons. This surely is the responsibility of the Royal Colleges of Surgeons and should have been organised by them. Audit Should audit be compulsory? There is little enthusiasm from general surgeons for a compulsory national audit, but this will be necessary if we are going to get
any meaningful figures as to the true incidence of mortality and complications in this exciting new field. Comprehensive audit should be introduced with training and accreditation. The Royal College of Surgeons of England has organised a voluntary confidential audit but it has only audited a fraction of the total number of laparoscopic cholecystectomies that have been performed throughout the country. As a result the true incidence of damage to the common bile duct may never be known. As newer techniques are introduced they must be audited to see if the operation is safe and has advantages over conventional surgery and to assess the long term complications. For instance, there are worrying reports of port site recurrences from colorectal laparoscopic surgery. The true incidence of this complication could be determined by a national audit. The centres for minimally invasive surgery that are being set up may have an important part to play in the development and running of such audits on a regional basis and coordinating results to produce a national picture. 1 Southern Surgeons Club. A prospective analysis of 1518 laparoscopic cholecystectomies. NEnglJMed 1991;324:10173-8. 2 Cuschieri A, Dubois F, Mouiel J, Mouret P, Becker H, Buess C, et al. The European experience with laparoscopic cholecystectomy. Am J Sug 1991; 161:385-7.
3 Dunn DC. Voluntary confidential audit of outcome of surgery. BMY 1991;303: 1272.
4 Macintryre IMC, Wilson RG. Iaparoscopic cholecystectomy. Br J Surg 1993;80:552-9. 5 Cuschieri A. Editorial.jR ColSurgEdin 1992;37:213-4. 6 Royal College of Surgeons of England. Minimnal access surgery. London: RCS, 1990. 7 Society of American Gastrointestinal Endoscopic Surgeons. Granting of privileges for laparoscopic general surgery. AmjSurg 1991;161:324-5. 8 Dent TL Training, credentialling, and granting of clinical privileges for
laparoscopic general surgesy.AmJSurg 1991;161:399-403. (Accepted 28January 1994)
Commentary on the cholesterol papers Two months ago the BMJ published three papers by Malcolm Law, Nicholas Wald, and colleagues on serum cholesterol concentration, ischaemic heart disease, and death (S February, p 363-79) This week we publish an invited commentary on these papers by George Davey Smith and Matthias Eggar and Law and Wald's response to their commentary. The debate continues on the letters pages (p 1038).
Statistical problems George Davey Smith, Matthias Egger
Department ofPublic Health, University of Glasgow, Glasgow G12 8RZ George Davey Smith, senior lecturer in epidemiology and public health Deparment ofSocial and Preventive Medicine, University of Berne, CH-3012 Berne, Switzerland Matthias Egger, senior research felow
Correspondence to: Dr Davey Siith. BMJ 1994;308:1025-7
BMJ VOLUME 308
Recently, some authorities have called for a change in direction in health policy on blood cholesterol concentrations,' while another group considers that current efforts are correct.2 The overviews by Law and colleagues can help us make up our minds regarding which, if either, opinion is correct.-5
Underestimation of risk of ischaemic heart disease Following an old but increasingly popular tradition,6 Law and colleagues corrected the association between cholesterol concentration and risk of ischaemic heart disease for the attenuation (or regression dilution) due to measurement imprecision. Since cholesterol concentrations vary over time, prospective studies relating future risk of ischaemic heart disease to a single measurement will underestimate the true strength of the association. The adjustment is used because few epidemiological studies have obtained repeat measurements. As Spearman noted in 1904,6 however, such adjustments should be used carefully because they are not free of assumptions. They could be misleading if recent levels of a risk factor are 16APRIL1994
more important determinants of disease than earlier levels; if peaks in the level of the risk factor, rather than the usual value, are the important aetiological factor; or if changes in risk factor levels are important.7 Some studies have been able to examine the association of average cholesterol concentration with risk of ischaemic heart disease directly.89 These have indicated that although the correction factors may overestimate the strength of the association between usual cholesterol concentration and risk of ischaemic heart disease, repeated cholesterol measurements identify people at increased risk of disease better than a single measurement. Law et al have introduced an extension of the correction for attenuation, which they called correction for surrogate measurement bias. Since it is the low density lipoprotein component of total cholesterol which is considered to increase the risk of ischaemic heart disease they argue that the association of total cholesterol with ischaemic heart disease underestimates the underlying association of low density lipoprotein cholesterol with ischaemic heart disease. They determined the degree to which total cholesterol 1025
fortuitous findings in trials with favourable results, and it is obvious that reviewing and rejecting results in one direction but not the other represents a bias. In the previous analysis,'7 benefit-in terms of a favourable effect on all cause mortality-was restricted to patients at particularly high risk of death from ischaemic heart disease. This is not simply due to increased mortality in trials which recruited low risk subjects, as Law et al imply, but is a graded association of increasing benefit with increasing risk of ischaemic heart disease. Such an association would be seen if a small adverse effect is balanced by a proportionate reduction in incidence of ischaemic heart disease. This would translate into greater absolute benefit being seen for patients at higher initial risk of ischaemic heart disease. Because of their size, the three trials that Law et al consider to have produced anomalous findings contribute greatly to the analysis. Despite this the slope between mortality outcome in trials (log odds ratio for mortality between treatment and control group) and the risk of ischaemic heart disease for subjects in the trials (indexed by the mortality from ischaemic heart disease per 1000 person years in the control group of the trials) is little altered by excluding these trials, although, not surprisingly, the significance is. The slope is -0 0074 (P-0 0009) with all nonhormonal drug trials included and -0 0070 (P=0-08) when the three large trials are excluded. When these three trials are considered as a group their results show a significantly (P-0 005) less Safety ofcholesterol lowering drugs favourable effect on all cause mortality than the other Since lowering cholesterol concentration reduces the non-hormonal drug trials. This can be explained by the incidence of ischaemic heart disease and little evidence fact that they all recruited low risk participants. The exists that naturally low cholesterol concentrations apparent unfavourable outcome disappears (P=0.9) are themselves harmful""16 the safety aspects of once the level of risk of the participants is taken into interventions used to lower cholesterol become account. These could simply be trials which recruited paramount. It is important in this case to separate the subjects at low risk of ischaemic heart disease, in whom effects of cholesterol reduction from the effects of the the benefit to risk ratio for treatment was low and adverse effects could therefore become evident. treatments used to lower cholesterol."7 We have updated a previous meta-analysis to attempt. to make sense of the different interpretations that have been made of the clinical trial data.'7 As Law et al show, Implications of targeted intervention A strategy aimed at identifying people with raised increased mortality from causes other than ischaemic heart disease is seen in the trials of cholesterol lowering cholesterol concentrations would lead to large numbers drugs. Since this increase is not associated with the of people receiving dietary counselling aimed at degree of cholesterol lowering achieved and is confined reducing cholesterol concentrations. Since individual to drug trials, it seems that it is the means of bringing dietary counselling often achieves disappointing about lower cholesterol concentration that is important reductions in cholesterol many of these attempts would rather than the low cholesterol concentrations them- be unsuccessful.2'22 The prescription of cholesterol selves. Furthermore, the adverse effect is restricted to lowering drugs would probably follow. Large increases trials with long follow up, which would be expected if in prescriptions for such drugs in the United Kingdom it is related to long term ingestion of the drug. For the over the past five years suggest that such a scenario trials of non-hormonal drugs with short follow up (less could be imminent."7 In the United States, where than four years) the odds ratio for mortality from cholesterol lowering with drugs is given more priority, causes other than ischaemic heart disease is 0 79 (95% around 6% of the population aged 50 and over were confidence interval 0-55 to 1-15; P=0-2). For trials prescribed these drugs in 1988-90." Lowering cholesterol concentration produces greater with longer follow up the odds ratio is 1 23 (1 07 to benefits for those at higher risk of ischaemic heart 1 45; P-0'005). Law et al dismiss the increased mortality from causes other than ischaemic heart disease in the drug trials by Number of patients needed to be treated for one year to prevent one suggesting that there was a chance increase in three death according to risk ofdeath from ischaemic heart disease primary prevention trials: the World Health OrganisaNo of patients needed % Of deaths tion clofibrate, lipid research clinics, and Helsinki to be treated attributable to Risk of death from ischaemic heart ischaemic heart disease studies."'20 The unequivocal statement that the Trial data Best case* disease (per 1000 person years) increase is due to chance may seem odd to readers 50 102 100 80 brought up to believe that it is not possible to prove a 125 164 72 40 null hypothesis and certainly not a type I error. These 167 533 64 30 three trials contribute around two thirds of the person 250 56 20 t 500 48 10 t years of follow up and 60% of the deaths from causes 1000 5 44 t other than ischaemic heart disease in the drug trials 5000 41 1 t Law et al include. It is unjustified to dismiss these 200/o reduction in mortality from ischaemic heart disease and results as chance based on abandoning intention to *Assumingeffect on all cause mortality. adverse treat analyses and asymmetrical case review. They no tEstimates indicate overall adverse effect on mortality. did not attempt to discover what could have been Trials using hormone treatment excluded.
serves as a proxy measure of low density lipoprotein and used this to inflate the relation between total cholesterol concentration and ischaemic heart disease. This is perhaps an unhappy addition to the statistical armoury, particularly since there are now many studies which have directly examined the association between low density lipoprotein cholesterol and risk of ischaemic heart disease.Y1'3 The results of these direct comparisons undermine the argument for applying the correction for surrogate measurement bias.'2 13 Furthermore, it is unclear why we should stop the correction for surrogate measurement bias at low density lipoprotein. The atherogenic component of low density lipoprotein cholesterol is thought to be oxidised low density lipoprotein,'4 and measurements of low density lipoprotein will serve only as proxies for this. Thus a further magnification of the risk associations should be introduced to correct this surrogate measurement bias. Once this adjustment has been made we must consider the fact that the presence of oxidised low density lipoprotein is only a surrogate marker of the degree to which unregulated scavenger receptors take up oxidised low density lipoprotein.'4 After a few more corrections for surrogate measurement bias the association becomes implausibly large, making the problem of this procedure clear. There is, unfortunately, no statistical substitute for actually measuring what you want to study, especially when trying to quantify risk precisely.
1026
BMJ VOLUME 308
16 APRIL 1994
disease, an effect which is even more pronounced when absolute risk rather than relative risk is examined.24 The table shows the number of patients who would need to be treated with drugs for a year to prevent one death, estimated from the data from trials that have been done or from a best case assumption of a 20% reduction in risk of death from ischaemic heart disease and no counterbalancing effect on death from other causes. The real data show an adverse effect for patients who are not at high risk of death from ischaemic heart disease. However, even in the best case large numbers of patients with a risk of death from ischaemic heart disease of 1-5/1000 person years would need to be treated to prevent one death. This range of risk includes many groups of patients whom current guidelines identify as candidates for drug treatment. Clearly the costs-in terms of patient time, the psychological effects of taking treatment to prevent a serious disease, the side effects of drug treatment, and direct expenses for medical care and the drugs-will be great compared with the potential benefits.
Conclusions Does the current evidence regarding cholesterol, cholesterol lowering, and mortality suggest that we should change direction' or carry on with current practice? There is no good evidence that naturally having a low blood cholesterol concentration is harmful. Nor does reducing cholesterol concentration without using drugs seem to be harmful. Reduction of cholesterol concentration will have a small benefit for most individuals, but in whole populations many deaths could be delayed. People and communities should be given this information and the means to adapt to it as they see fit. On the other hand, strategies to identify individuals with high cholesterol concentrations will be associated with the usual material and psychological costs of screening. Many will fail to respond adequately to dietary intervention and become candidates for lifelong drug treatment. This will be accompanied by high medical costs and the usual array of minor side effects. Except for patients at greatly increased risk of ischaemic heart disease, current evidence suggests such drug treatment will be associated with, at best, little benefit, if not an adverse overall effect on mortality. Here we need a change of direction-to turn away from the identification and drug treatment
of asymptomatic people with isolated mildly or moderately raised cholesterol concentrations. ME is supported by the Swiss National Science Foundation. 1 Hulley SB, Walsh JMB, Newman TB. Health policy on blood cholesterol: time to change directions. Circulasion 1992;86:1026-9. 2 Stamler J, Stamler R, Brown WV, Gotto AM, Greenland P, Grundy S, et al. Serum cholesterol: doing the right thing. Circsuaion 1993;88:1954-60. 3 Law MR, Wald NJ, Wu T, Hackshaw A, Bailey A. Systematic underestimation of association between serum cholesterol concentration and ischaemic heart disease in observational studies: data from the BUPA study. BM3 1993;308:363-6. 4 Law MR, Wald NJ, Thompson SG. By how much and how quickly does reduction in serum cholesterol concentration lower risk of ischaemic heart disease. BMJ 1993;308:367-72. 5 Law MR, Thomson SG, Wald NJ. Assessing public hazards of reducing serum cholesterol. BM3 1993;308:373-9. 6 Spearman C. The proof and measurement of association between two things. Am3'Psychol 1904;15:72-101. 7 Phillips AN, Davey Smith G. The design of prospective epidemiological studies: when smalter is better. Clin Epidemiol 1993;46:1203-1 1. 8 Davis CE, Rifitnd BM, Brenner H, Gordon DJ. A single cholesterol measurement underestimates the risk of coronary heart disease. An empirical example from the lipid research clinics mortality foliow-up study. JAMA 1990;264:3044-6. 9 National Health Screening Service. Predicton based on one or two measurements of a variable. Oslo: National Health Screening Service, 1991. 10 Manolio TA, Pearson TA, Wenger NK, Barrett-Connor E, Payne GH, Harlan WR. Cholesterol and health disease in older persons and women: review of an NHLBI workshop. Annals of Epidemiology 1992;2:161-76. 11 Assmann G, Schulte H. Relation of high-density lipoprotein cholesterol and triglycerides to incidence of atherosclerotic coronary artery disease (the PROCAM experience). Am e Cardiol 1992;70:733-7. 12 Shestov DB, Deev AD, Klimov AN, Davis CE, Tyroler HA. Increased risk of coronary heart disease in men with low total and low-density lipoprotein cholesterol in the Russian lipid research dinics prevalence follow-up study. Circulation 1993;88:846-53. 13 Bass KM, Newschaffer CJ, Klag MJ, Bush TL Plasma lipoprotein levels as predictors of cardiovascular death in women. Arch Intern Med 1993;153: 2209-16. 14 Witztum .I, Steinberg D. Role of oxidized LDL in atherogenesis. I Clin Invest 1991;U:1785-92. 15 Davey Smith G, Shipley MJ, Marmot MG, Rose G. Plasma cholesterol concentration and mortality: the Whitehall study. JAMA 1992;267:70-6. 16 Davey Smith G. Blood cholesterol and non-coronary mortality. Coronary Artery Disease 1993;4:860-6. 17 Davey Smith G, Song F, Sheldon TA. Cholesterol lowering and mortality: the importance of considering initial level of risk. BMJ 1993;306:1367-73. 18 Committee of Principal Investigators. A co-operative trial in the primary prevention of ischaemic heart disease using clofibrate. Br Heart a 1978;40: 1069-118. 19 Lipid Research Clinics Program. The lipid research clinics coronary primary prevention trial results. 1. Reduction in incidence of coronary heart
disease.gAMA 1984;251:351-64.
20 Frick MH, Elo 0, Haapa K, Heinsaimi P, Helo P, Huttunen JK, et al. Helsinki heart study: primary-prevention trial with gemfibrozil in middle aged men with dyslipidemia. NEnglyMed 1987;317:1237-45. 21 Ramsay LE, Yeo WW, Jackson PR. Dietary reduction of serum cholesterol concentration: time to think again. BMJ 1991;303:953-7. 22 Hunninghake DB, Stein EA, Dujovne CA, Harris WS, Feldman EB, Miller VT, et al. The effect of intensive dietary therapy alone or combined with lovastatin in outpatients with hypercholesterolaemia. N Engl g Med
1993;328:1213-9. 23 Giles WH, Anda RF, Jones DH, Serdula MK, Merritt RK, DeStefano F. Recent trends in the identification and treatment of high blood cholesterol by physicians: progress and missed opportunities. JAMA 1993;269:1133-8. 24 Davey Smith G, Egger M. Who benefits from medical interventions? BMJ 1994;308:72-4.
Disagreements are not substantial M R Law, N J Wald
Department of Environmental and Preventive Medicine, Wolfson Institute of Preventive Medicine, St Bartholomew's Hospital Medical College, London ECiM 6BQ M R Law, senior lecturer N J Wald, professor
Correspondence to: Professor Wald. BM_ 1994;308:1027-9
BMJ VOLUME 308
George Davey Smith and Matthias Egger agree with the conclusions and recommendations in our cholesterol papers1-3 and also with our recommendation against mass cholesterol testing.4 They disagree with us on two aspects but these are both minor. One relates to the methods and the other to interpretation. Neither affects the importance of serum cholesterol concentration in the aetiology and prevention of ischaemic heart disease, but both raise issues of general epidemiological application.
Correction for underestimation in observational studies It is well recognised that cohort studies under16APRL1994
estimate the dose-response relation between an imprecisely measured risk factor (here serum cholesterol) and its effect (here mortality from ischaemic heart disease).5 The statistical procedure to adjust for this regression dilution bias is simple, accurate, and corroborated by direct measurement. Davey Smith and Egger do not give the correct reason for taking account of the second source of underestimation, the surrogate dilution effect. It is not that low density lipoprotein cholesterol measures an effect closer to the target site than total cholesterol. It is because it permits the quantitative reconciliation of data from observational studies with data from randomised trials. In the trials the reduction in total serum cholesterol concentration is almost entirely due 1027
