On the design and analysis of randomized clinical trials with multiple endpoints

This paper considers some methods for reducing the number of significance tests undertaken when analyzing and reporting results of clinical trials. Emphasis is placed on designing and analyzing clinical trials to examine a composite hypothesis concerning multiple endpoints and combining this multiple endpoint methodology with group sequential methodology. Four methods for composite hypotheses are considered: an ordinary least squares and a generalized least squares approach both due to O'Brien (1984, Biometrics 40, 1079-1087), a new modification of these, and an approximate likelihood ratio test, due to Tang, Gnecco, and Geller (1989, Biometrika 76, 577-583). These are extended for group sequential use. In particular, simulation is used to generate critical values and sequences of nominal significance levels for the approximate likelihood ratio test, which is not normally distributed. An example is given and the relative merits of the suggested approaches are discussed.

Effects of 22 months of treatment with inhaled corticosteroids and/or beta-2-agonists on lung function, airway responsiveness, and symptoms in children with asthma. The Dutch Chronic Non-specific Lung Disease Study Group

In a randomized double-blind multicenter clinical study, 116 children with asthma were randomly assigned to treatment with an inhaled beta-2-agonist (salbutamol 0.2 mg) plus an inhaled corticosteroid (budesonide 0.2 mg) three times a day (BA+CS) or to an inhaled beta-2-agonist (salbutamol 0.2 mg) plus a placebo three times a day (BA+PL). After a median follow-up time of 22 months, 26 patients receiving BA+PL (45%) had withdrawn from randomized treatment, mainly because of asthma symptoms, compared with three withdrawals in the patients receiving BA+CS (p less than 0.0001). The FEV1, expressed as a percentage of the predicted value for age, sex, and height, showed an absolute increase of 7.0% after 2 months of BA+CS compared with a decrease of 4.0% after 2 months of BA+PL. This 11% difference in percent predicted FEV1 (95% confidence interval, 7 to 15%; p less than 0.0001) was then maintained after a median follow-up period of 22 months. Postbronchodilator FEV1 showed an absolute increase of 3.7% predicted within 2 months in patients receiving BA+CS and an absolute decrease of 1.1% predicted in children receiving BA+PL (p = 0.0005). Thereafter, this difference between the two treatment groups was maintained. Average peak expiratory flow rate (PEFR) increased from baseline by 36.6 L/min in the BA+CS group compared with 3.7 L/min in the BA+PL group (p = 0.003). This difference then remained for the median follow-up time of 22 months.(ABSTRACT TRUNCATED AT 250 WORDS)

Prediction of short-term survival with an application in primary biliary cirrhosis

Many long-term follow-up studies for survival accumulate repeated measurements of prognostic factors. Survival models which include only covariate values at baseline do not use all available information, and do not relate to survival predictions for times other than at that baseline. Time-dependent covariate models (which update covariate values as measurements occur through time) might be used, though limitations of software for estimating the underlying hazard functions and difficulty in relating hazard function changes to survival prediction present serious drawbacks. By dividing each patient's follow-up into successive intervals of equal length (using a length of interest for prediction) and with measurements available at the start of each, we describe how an analysis taking person-intervals as the observation units can be undertaken using readily available software to produce short-term survival models. We show that this approach is related to both the baseline and time-dependent covariate models. The method is illustrated using data from a long-term study of patients with primary biliary cirrhosis, where interest is in short-term survival predictions to aid the decision when to undertake liver transplantation.

Repeated measures in clinical trials: analysis using mean summary statistics and its implications for design

This paper explores the use of simple summary statistics for analysing repeated measurements in randomized clinical trials with two treatments. Quite often the data for each patient may be effectively summarized by a pre-treatment mean and a post-treatment mean. Analysis of covariance is the method of choice and its superiority over analysis of post-treatment means or analysis of mean changes is quantified, as regards both reduced variance and avoidance of bias, using a simple model for the covariance structure between time points. Quantitative consideration is also given to practical issues in the design of repeated measures studies: the merits of having more than one pre-treatment measurement are demonstrated, and methods for determining sample sizes in repeated measures designs are provided. Several examples from clinical trials are presented, and broad practical recommendations are made. The examples support the value of the compound symmetry assumption as a realistic simplification in quantitative planning of repeated measures trials. The analysis using summary statistics makes no such assumption. However, allowance in design for alternative non-equal correlation structures can and should be made when necessary.

Within-subject diastolic blood pressure variability: implications for risk assessment and screening

Because of variability in diastolic blood pressure within an individual, repeated measurements increase precision in assessing an individual's underlying mean pressure and so also aid risk classification. Data from a cohort of 11,299 middle-aged men is used to model the variability in diastolic pressure between annual measurements. A simple model with pressure normally distributed about an underlying mean with standard deviation increasing with level fits the data very well. In modelling risk of cardiovascular mortality, a strong association is found with observed diastolic pressure level but not to trends in or variability between observed values. The effect of regression dilution is clear with the risk relationship appearing greater as one uses the mean of an increasing number of measurements. A method of adjusting for this regression dilution is described so giving an estimate of the relationship with underlying mean diastolic pressure. Using this survival model and the model for blood pressure variability, a method is presented for estimating both underlying mean pressure and absolute risk of cardiovascular disease given a sequence of blood pressure measurements from screening. This allows a sequential strategy for determining whether (a) antihypertensive intervention is desirable, (b) no further screening is necessary, or (c) further screening would aid the assessment, and emphasizes the need to consider blood pressure in the context of multiple risk factors.

When to stop a clinical trial

Images

The impact of stopping rules on heterogeneity of results in overviews of clinical trials

This paper explores the extent to which application of statistical stopping rules in clinical trials can create an artificial heterogeneity of treatment effects in overviews (meta-analyses) of related trials. For illustration, we concentrate on overviews of identically designed group sequential trials, using either fixed nominal or O'Brien and Fleming two-sided boundaries. Some analytic results are obtained for two-group designs and simulation studies are otherwise used, with the following overall findings. The use of stopping rules leads to biased estimates of treatment effect so that the assessment of heterogeneity of results in an overview of trials, some of which have used stopping rules, is confounded by this bias. If the true treatment effect being studied is small, as is often the case, then artificial heterogeneity is introduced, thus increasing the Type I error rate in the test of homogeneity. This could lead to erroneous use of a random effects model, producing exaggerated estimates and confidence intervals. However, if the true mean effect is large, then between-trial heterogeneity may be underestimated. When undertaking or interpreting overviews, one should ascertain whether stopping rules have been used (either formally or informally) and should consider whether their use might account for any heterogeneity found.

Can meta-analyses be trusted?

The enthusiasm for meta-analyses (or overviews) expressed by their proponents is not always shared by the broader medical community. To encourage constructive debate, we adopt a critical perspective on the conduct and interpretation of meta-analysis. We focus particularly on some of the statistical issues, especially heterogeneity between studies, and also on the extrapolation of meta-analysis findings to clinical practice. We conclude that meta-analysis is not an exact statistical science that provides definitive simple answers to complex clinical problems. It is more appropriately viewed as a valuable objective descriptive technique, which often furnishes clear qualitative conclusions about broad treatment policies, but whose quantitative results have to be interpreted cautiously.

A perspective on the role of quality-of-life assessment in clinical trials

This article summarizes some of the methodologic and practical issues relevant to health related quality-of-life (HRQL) assessment in randomized clinical trials. Particular attention is given to 1) the reasons for undertaking HRQL assessment, 2) the types of clinical trials particularly suitable for HRQL assessment, 3) the selection of appropriate HRQL measures, 4) statistical issues in the design and analysis of HRQL studies, and 5) the need to coordinate progress in HRQL studies in clinical research.

Does plasma cholesterol concentration predict mortality from coronary heart disease in elderly people? 18 year follow up in Whitehall study

OBJECTIVE

To explore the extent to which the relation between plasma cholesterol concentration and risk of death from coronary heart disease in men persists into old age.

DESIGN

18 year follow up of male Whitehall civil servants. Plasma cholesterol concentrations and other risk factors were determined at first examination in 1967-9 when they were aged 40-69. Death of men up to 31 January 1987 was recorded.

SUBJECTS

18,296 male civil servants, 4155 of whom died during follow up.

MAIN OUTCOME MEASURES

Cause and age of death. Cholesterol concentration in 1967-9 and number of years elapsed between testing and death.

RESULTS

1676 men died of coronary heart disease. The mean cholesterol concentration in these men was 0.32 mmol/l higher than that in all other men (95% confidence interval 0.26 to 0.37 mmol/l). This difference in cholesterol concentrations fell 0.15 mmol/l with every 10 years' increase in age at screening. The risk of raised cholesterol concentration fell with age at death. Compared with other men cholesterol concentration in those who died of coronary heart disease was 0.44 mmol/l higher in those who died aged less than 60 and 0.26 mmol/l higher in those aged 60-79 (p = 0.03). For a given age at death the longer the gap between cholesterol measurement and death the more predictive the cholesterol concentration, both for coronary heart disease and all cause mortality (trend test p = 0.06 and 0.03 respectively).

CONCLUSION

Reducing plasma cholesterol concentrations in middle age may influence the risk of death from coronary heart disease in old age.

Risk factors for stroke in middle aged British men

OBJECTIVE

To determine the risk factors for stroke in a cohort representative of middle aged British men.

DESIGN

Prospective study of a cohort of men followed up for eight years.

SETTING

General practices in 24 towns in England, Wales, and Scotland (the British regional heart study).

SUBJECTS

7735 men aged 40-59 at screening, selected at random from one general practice in each town.

MAIN OUTCOME MEASURE

Fatal and non-fatal strokes.

RESULTS

110 of the men had at least one stroke; there were four times as many non-fatal as fatal strokes. The relative risk of stroke was 12.1 in men who had high blood pressure (systolic blood pressure greater than or equal to 160 mm Hg) and were current smokers compared with normotensive, non-smoking men. Diastolic blood pressure yielded no additional information, and former cigarette smokers had the same risk as men who had never smoked. Heavy alcohol intake was associated with a relative risk of stroke of 3.8 in men without previously diagnosed cardiovascular disease. Men with pre-existing ischaemic heart disease had an increased risk of stroke, but only when left ventricular hypertrophy on electrocardiography was also present.

CONCLUSIONS

Systolic blood pressure, cigarette smoking, and left ventricular hypertrophy on electrocardiography in men with pre-existing ischaemic heart disease were found to be the major risk factors for stroke in middle aged British men. Heavy alcohol intake seemed to increase the risk of stroke in men without previously diagnosed cardiovascular disease. A large proportion of strokes should be preventable by controlling blood pressure and stopping smoking.

Prognostic scores for detecting a high risk group: estimating the sensitivity when applied to new data

The sensitivity of a prognostic scoring system will tend to be exaggerated if the scoring system is both derived and validated on the same data. This paper provides, by analogy to regression with error in an explanatory variable, an intuitive basis for the methodological results of Copas which seek to estimate the degree of such exaggeration. There was good agreement between Copas' results and those achieved in a series of cross-validation exercises where logistic regression models predicting the risk of ischaemic heart disease were derived using data from the prospective British Regional Heart Study. When truly important variables were included, the exaggeration of the sensitivity increased as the number of cases of disease available decreased. It is concluded that Copas' method, which is easy to implement in practice, may be helpful in realistically anticipating the extent of such exaggeration, and that it can be usefully employed before pursuing a scoring system on newly collected data.

Estimation issues in clinical trials and overviews

There is a general move towards greater emphasis on point and interval estimates of treatment effect in reporting of clinical trials, so that significance testing plays a lesser role. In this article we examine a number of issues which affect the use and interpretation of conventional estimation methods. Should we accept or avoid the stereotypes of 95 per cent confidence? Should the abstract of a trial report include confidence intervals for major endpoints? Are frequentist confidence intervals being interpreted correctly, and should Bayesian probability intervals be more widely used in trial reports? Does the timing of publication, such as early stopping because of a large observed treatment difference, lead to exaggerated point and interval estimates? How can we produce realistic estimates from subgroup analyses? Is publication bias seriously affecting our ability to obtain unbiased estimates? Is the emphasis on estimation methods a powerful tool for encouraging larger sample sizes? Can we resolve the controversy concerning fixed or random effects models for estimation in overviews of related trials? Our arguments are illustrated by results from recent trials in cardiovascular disease.

The variability of serum cholesterol measurements: implications for screening and monitoring

The reliability of screening for high serum total cholesterol is adversely affected by the variability of cholesterol levels over time. This problem is investigated using data on repeated cholesterol measurements for 14,600 men and women in the MRC Mild Hypertension Trial. For measurements 1 year apart, the within-person coefficient of variation (CV) is 7%, which is substantial compared with the between-person CV of 15%. In a screening programme, this within-person variability may lead to the misclassification of individuals and inappropriate intervention. For example, 28% of middle-aged British men with a single cholesterol measurement above 6.9 mmol/l have a long-term average cholesterol below that value even without intervention. Using averages of several cholesterol measurements reduces, but does not eliminate, these problems. Furthermore, monitoring the effect of interventions in individuals by sequential cholesterol measurement may be unhelpful or even misleading. These problems cast serious doubt on the value of general population screening for high cholesterol levels.

Practical problems in interim analyses, with particular regard to estimation

This article considers some of the practical problems inherent in interim analyses and stopping rules for randomized clinical trials. Topics covered include group sequential designs, trials with unplanned interim analyses, estimation problems in clinical trials with planned interim analyses, and the balance between individual and collective ethics. Particular attention is paid to the fact that clinical trials that stop early are prone to exaggerate the magnitude of treatment effect. Accordingly, a Bayesian "shrinkage" method of analysis is proposed to help quantify the extent to which surprisingly large point and interval estimates of treatment difference in clinical trials that stop early should be moderated.

Concentrations of high density lipoprotein cholesterol, triglycerides, and total cholesterol in ischaemic heart disease

OBJECTIVE

To assess the roles of serum concentrations of total cholesterol, high density lipoprotein cholesterol, and triglycerides in predicting major ischaemic heart disease.

DESIGN

Men recruited for the British regional heart study followed up for a mean of 7.5 years.

SETTING

General practices in 24 British towns.

PATIENTS

7735 Middle aged men.

END POINT

Predictive value of serum concentrations of lipids.

MEASUREMENTS AND MAIN RESULTS

At initial screening serum concentrations of total cholesterol, high density lipoprotein cholesterol, and triglycerides were determined from non-fasting blood samples. Altogether 443 major ischaemic heart disease events (fatal and non-fatal) occurred during the study. Men in the highest fifth of the distribution of total cholesterol concentration (greater than or equal to 7.2 mmol/l) had 3.5 times the risk of ischaemic heart disease than did men in the lowest fifth (less than 5.5 mmol/l) after adjustment for high density lipoprotein cholesterol concentration and other risk factors. Men in the lowest fifth of high density lipoprotein cholesterol concentration (less than 0.93 mmol/l) had 2.0 times the risk of men in the highest fifth (greater than or equal to 1.33 mmol/l) after adjustment for total cholesterol concentration and other risk factors. Men in the highest fifth of triglyceride concentration (greater than or equal to 2.8 mmol/l) had only 1.3 times the risk of those in the lowest fifth (less than 1.08 mmol/l) after adjustment for total cholesterol concentration and other risk factors; additional adjustment for high density lipoprotein cholesterol concentration made the association with ischaemic heart disease disappear.

CONCLUSIONS

Serum concentration of total cholesterol is the most important single blood lipid risk factor for ischaemic heart disease in men. High density lipoprotein cholesterol concentration is less important, and triglyceride concentrations do not have predictive importance once other risk factors have been taken into account.

Diurnal variations in serum biochemical and haematological measurements

Twenty five biochemical and haematological measurements were determined on nonfasting blood and serum samples collected between 9 am and 7 pm from a representative group of 7685 British middle-aged men. Most measurements showed significant diurnal variations, but only for bilirubin, phosphate, and triglyceride did time of day account for more than 5% of the between subject variance. Serum bilirubin concentrations showed a pronounced downward trend in the afternoon, the mean value after 6 pm being 30% lower than the mean value in the morning. Mean serum triglyceride and phosphate concentrations increased steadily through the day. Mean concentrations of potassium, haemoglobin, and haematocrit and red cell count were higher in the morning, while urea and creatinine concentrations and white cell count had higher means in the afternoon. Glucose showed a pattern consistent with short term response to meals. The effects of these diurnal trends on routine use of biochemical tests needs careful consideration, and a greater understanding of their biological mechanisms is required.

Sample size requirements for prospective studies, with examples for coronary heart disease

Methods of determining the required number of disease cases for estimation of relative odds in prospective studies are evaluated, with examples from coronary heart disease. Data from a British prospective study of coronary heart disease are used in simulation exercises to assess the reliability of estimation formulae for both continuous and categorical risk factors. For continuous risk factors, a univariate formula based on estimation of the standardized relative odds (Whittemore A. S. JAMA 1981; 76: 27-32 [1]), gives reliable estimation of the required number of disease cases, provided the risk factor has a near normal distribution. An extension of the formula to adjustment for other risk factors, was less satisfactory, perhaps because of departures from multivariate normality. For categorical risk factors, an adaption of a univariate method for case control studies (Smith PG, Day NE. Int J Epidemiol 1984; 13: 356-365 [2]), gives reliable estimates of the number of cases required. However, this depends on approximate prior knowledge of the relative odds. In general, prospective studies of coronary heart disease risk factors should aim for at least 400 cases to enable sufficient accuracy of estimation.

Stopping rules and estimation problems in clinical trials

Stopping rules in clinical trials can lead to bias in point estimation of the magnitude of treatment difference. A simulation exercise, based on estimation of the risk ratio in a typical post-myocardial infarction trial, examines the nature of this exaggeration of treatment effect under various group sequential plans and also under continuous naive monitoring for statistical significance. For a fixed treatment effect the median bias in group sequential design is small, but it is greatest for effects that the trial has reasonable power to detect. Bias is evidently greater in trials that stop early and is dramatic under naive monitoring for significance. Group sequential plans lead to a multimodal sampling distribution of treatment effect, which poses problems for incorporating their estimates into meta-analyses. By simulating a population of trials with treatment effects modelled by an underlying distribution of true risk ratios, a Bayesian method is proposed for assessing the plausible range of true treatment effect for any trial based on interim results. This approach is particularly useful for producing shrinkage of the unexpectedly large and imprecise observed treatment effects that arise in clinical trials that stop early. Its implications for trial design are discussed.

The role of risk factors in heart attacks occurring in men with pre-existing ischaemic heart disease

The importance of three risk factors--serum total cholesterol, systolic blood pressure, and cigarette smoking--on the risk of new major ischaemic heart disease events in men who already have evidence of ischaemic heart disease was assessed. Data from the initial examination in a large prospective study of cardiovascular disease in middle aged men (the British Regional Heart Study) were used to separate 7710 men into three groups on the basis of a resting electrocardiogram, a standardised chest pain questionnaire, and recall of a doctor's diagnosis of angina or of a previous heart attack: (group 1) no evidence of ischaemic heart disease (75%), (group 2) evidence of ischaemic heart disease short of a definite myocardial infarction (20%), (group 3) definite myocardial infarction (6%). In the average follow up period of 7.5 years, 443 men suffered a new major event caused by ischaemic heart disease (fatal or non-fatal myocardial infarction or sudden cardiac death). Age standardised event rates were determined for each of the three groups for varying levels of the established risk factors. Cigarette smoking is strongly associated with the event rate in group 1 but in men with existing heart disease, especially group 3, differences in risk between the smoking categories were smaller. The strong relation between systolic blood pressure and event rate persisted in groups 1 and 2 but not in group 3. The positive association between serum concentration of total cholesterol and the event rate was strongest in group 1 and weaker in groups 2 and 3, though it remained highly significant. These observations, taken together with the results of previous prospective studies and intervention trials, suggest that the important association between serum total cholesterol and the risk of heart attack persists in men with pre-existing ischaemic heart disease, including myocardial infarction. Therefore, in these men the reduction of serum total cholesterol concentration may be at least as important as it is in men without evidence of ischaemic heart disease.

The relationship between blood lead, blood pressure, stroke, and heart attacks in middle-aged British men

The relationship between blood lead concentration and blood pressure is examined in a survey of 7371 men aged 40 to 59 from 24 British towns. After allowance for relevant confounding variables, including town of residence and alcohol consumption, there exists a very weak but statistically significant positive association between blood lead and both systolic and diastolic blood pressure. These cross-sectional data indicate that an estimated mean increase of 1.45 mm Hg in systolic blood pressure occurs for every doubling of blood lead concentration with a 95% confidence interval of 0.47 to 2.43 mm Hg. After 6 years of follow-up, 316 of these men had major ischemic heart disease, and 66 had a stroke. After allowance for the confounding effects of cigarette smoking and town of residence there is no evidence that blood lead is a risk factor for these cardiovascular events. However, as the blood lead-blood pressure association is so weak, it is unlikely that any consequent association between lead and cardiovascular disease could be demonstrated from prospective epidemiological studies. An overview of data from this and other large epidemiological surveys provides reasonably consistent evidence on lead and blood pressure. While NHANES II data on 2254 U.S. men indicate a slightly stronger association between blood lead and systolic blood pressure, data from two Welsh studies on over 2000 men did not show a statistically significant association. However, the overlapping confidence limits for all these studies suggest that there may be a weak positive statistical association whereby systolic blood pressure is increased by about 1 mm Hg for every doubling of blood lead concentration.(ABSTRACT TRUNCATED AT 250 WORDS)

The relationship between blood lead, blood pressure, stroke, and heart attacks in middle-aged British men

The relationship between blood lead concentration and blood pressure is examined in a survey of 7371 men aged 40 to 59 from 24 British towns. After allowance for relevant confounding variables, including town of residence and alcohol consumption, there exists a very weak but statistically significant positive association between blood lead and both systolic and diastolic blood pressure. These cross-sectional data indicate that an estimated mean increase of 1.45 mm Hg in systolic blood pressure occurs for every doubling of blood lead concentration with a 95% confidence interval of 0.47 to 2.43 mm Hg. After 6 years of follow-up, 316 of these men had major ischemic heart disease, and 66 had a stroke. After allowance for the confounding effects of cigarette smoking and town of residence there is no evidence that blood lead is a risk factor for these cardiovascular events. However, as the blood lead-blood pressure association is so weak, it is unlikely that any consequent association between lead and cardiovascular disease could be demonstrated from prospective epidemiological studies. An overview of data from this and other large epidemiological surveys provides reasonably consistent evidence on lead and blood pressure. While NHANES II data on 2254 U.S. men indicate a slightly stronger association between blood lead and systolic blood pressure, data from two Welsh studies on over 2000 men did not show a statistically significant association. However, the overlapping confidence limits for all these studies suggest that there may be a weak positive statistical association whereby systolic blood pressure is increased by about 1 mm Hg for every doubling of blood lead concentration.(ABSTRACT TRUNCATED AT 250 WORDS)

Prediction of men at high risk of heart attack and its relevance to pilots

In this paper we have extrapolated from data on the general population of middle-aged men and made suggestions for risk prediction and prevention policy in aircrew. As suggested previously, it would be helpful in future if data on the risk of heart attack in pilots could be generated from a central recording system covering both current pilots and those who retire for whatever reason. Indefinite follow-up information on ischaemic events would also be particularly helpful.

Blood pressure and hypertension in middle-aged British men

Blood pressure measurements in 7735 middle-aged men from general practices in 24 towns in England, Wales and Scotland provide information on the prevalence of hypertension and its management in Great Britain. Despite a substantial correlation (r = 0.70) between systolic and diastolic blood pressures, individuals can show considerable discrepancies between these two measurements; they are not interchangeable. This observation has important implications for the choice of criteria used to define hypertension. However defined, the prevalence of hypertension increases markedly with age, increasing body mass index and with heavy alcohol consumption. It is not related to smoking and only to a small extent to social class. Diastolic hypertension (greater than or equal to 90 mmHg) was present in 26% and systolic hypertension (greater than or equal to 160 mmHg) in 22% of these men. In both systolic and diastolic hypertension, only one quarter of affected men could recall having been diagnosed as hypertensive by a doctor, and only one third of these were on regular antihypertensive treatment. There is a threefold variation in the prevalence of measured hypertension in the 24 towns with a trend towards higher rates in Northern England and Scotland. No relationship was seen between the prevalence rates of measured hypertension in the towns and the rates of doctor diagnosis of hypertension. Cardiovascular mortality rates in the towns were correlated with the measured prevalence rates for systolic and diastolic hypertension (r = 0.70 and r = 0.57, respectively). The geographic variations in blood pressure and hypertension in Great Britain provide a major opportunity for research into the causes of 'essential' hypertension.

Parental death from heart disease and the risk of heart attack

In the British Regional Heart Study, 7735 men aged 40-59 years were asked at initial screening whether their father or mother was alive or dead and to what cause any death had been attributed. They were followed up for the occurrence of major ischaemic heart disease events. At screening the men reported that 23% of the fathers and 43% of the mothers were alive and that 19% of the fathers and 11% of the mothers had died of heart trouble. In a sample of parental deaths, the death certificate was obtained and it was found that a son's report of a parental death from 'heart trouble' was a reliable indication that death had been certified to ischaemic heart disease. However, about half of the parental deaths certified as ischaemic heart disease were not reported as 'heart trouble' by their sons. A major ischaemic heart disease event was experienced by 336 men over an average follow-up of 6.2 years. Men who said their father had died from 'heart trouble' were at twice the risk of a major ischaemic heart disease event compared with men whose fathers were still alive and 1.5 times the risk of men who reported their father to be dead from another or unknown cause, even after accounting for age, systolic blood pressure, serum total cholesterol, cigarette smoking and serum HDL-cholesterol. For men with mother dead from heart trouble, the corresponding figures were 1.3 and 1.0.(ABSTRACT TRUNCATED AT 250 WORDS)

Blood cadmium concentrations in the general population of British middle-aged men

Blood cadmium concentrations were determined for 6919 men aged 40-59 randomly selected from general practice registers in 24 British towns. The mean and median blood cadmium were 1.9 and 1.4 micrograms/1 respectively and the distribution was highly skewed. The mean levels in non-smokers was 1.0 micrograms/l and current smokers showed a marked gradient with the daily amount smoked, with a mean of 3.9 micrograms/l in men smoking 40 or more cigarettes per day. Whereas 95% of men who never smoked had blood cadmium less than 2.0 micrograms/l, 80% of men smoking 20 or more cigarettes a day exceeded this figure. 1% of the men had blood cadmium concentrations greater than or equal to 7 micrograms/l virtually all of whom currently smoked cigarettes. Blood cadmium levels in ex-smokers were much lower than in current smokers even for those who had stopped within the past year. However, the mean levels in ex-smokers remained higher than the 'never smoked' for several years after stopping. There was little evidence that age, social class, or alcohol consumption were associated with blood cadmium levels after allowance for cigarette smoking. There is substantial geographic variation in mean blood cadmium for middle-aged men which could not be completely accounted for by smoking differences. Towns in the south and east of England all had mean levels under 2.0 micrograms/l whereas the majority of towns in other parts of Britain had mean levels greater than 2.0 micrograms/l. Possible reasons for this geographic pattern (e.g. geochemistry, industrial exposure, dietary differences) need further exploration.

State of the art in comparative cancer clinical trials

In order to investigate the current state of the art in clinical cancer research, a survey of comparative cancer clinical trials was conducted using a MEDLINE literature search of the 6-month interval from July to December of 1985. Data were obtained primarily from the published abstracts. The major observations from the study are as follows. First, trials in cancer research are published in an extraordinarily wide range of journals covering most of the major medical specialties in addition to the many cancer specialty journals. Second, randomization is now used quite extensively. It is the predominant method of control in chemotherapy trials and in trials reported in cancer journals. However, its use is much less common in other specialties, especially surgery, so efforts to popularize randomization in these specialties would be beneficial. Third, sample sizes are highly variable, and the median sample size (96 patients) is too small to reliably detect and evaluate moderate treatment advances. Finally, our survey reveals that published trials show an inordinately large proportion of breakthroughs in treatment compared with the generally accepted view that only slow progress is being made in developing effective cancer treatments. Our view is that this result reflects the strong tendency to publish only studies with positive results. This hypothesis is supported by the fact that in nonrandom trials there tend to be more positive conclusions and an absence of a strong association between conclusions and sample size.

The analysis of multiple endpoints in clinical trials

Treatment comparisons in randomized clinical trials usually involve several endpoints such that conventional significance testing can seriously inflate the overall Type I error rate. One option is to select a single primary endpoint for formal statistical inference, but this is not always feasible. Another approach is to apply Bonferroni correction (i.e., multiply each P-value by the total number of endpoints). Its conservatism for correlated endpoints is examined for multivariate normal data. A third approach is to derive an appropriate global test statistic and this paper explores one such test applicable to any set of asymptotically normal test statistics. Quantitative, binary, and survival endpoints are all considered within this general framework. Two examples are presented and the relative merits of the proposed strategies are discussed.

Statistical problems in the reporting of clinical trials. A survey of three medical journals

Reports of clinical trials often contain a wealth of data comparing treatments. This can lead to problems in interpretation, particularly when significance testing is used extensively. We examined 45 reports of comparative trials published in the British Medical Journal, the Lancet, or the New England Journal of Medicine to illustrate these statistical problems. The issues we considered included the analysis of multiple end points, the analysis of repeated measurements over time, subgroup analyses, trials of multiple treatments, and the overall number of significance tests in a trial report. Interpretation of large amounts of data is complicated by the common failure to specify in advance the intended size of a trial or statistical stopping rules for interim analyses. In addition, summaries or abstracts of trials tend to emphasize the more statistically significant end points. Overall, the reporting of clinical trials appears to be biased toward an exaggeration of treatment differences. Trials should have a clearer predefined policy for data analysis and reporting. In particular, a limited number of primary treatment comparisons should be specified in advance. The overuse of arbitrary significance levels (for example, P less than 0.05) is detrimental to good scientific reporting, and more emphasis should be given to the magnitude of treatment differences and to estimation methods such as confidence intervals.

Social class differences in ischaemic heart disease in British men

To examine why ischaemic heart disease (IHD) mortality rates in Britain are higher in manual than in non-manual workers 7735 middle-aged men in the British Regional Heart Study were followed up for 6 years, during which time 336 men experienced a major IHD event (fatal or non-fatal myocardial infarction or sudden cardiac death). The prevalence rates of IHD at screening, were higher in manual workers. Also, the attack rate of major IHD events during follow-up was 44% higher in manual workers. Marked differences in cigarette smoking contributed substantially to the increased risk of IHD in manual workers, who also had higher levels of blood pressure, were more obese, and took much less physical activity in leisure time. Adjustment for differences in these risk factors narrowed the gap between manual and non-manual workers in attack rates of IHD. Since the risk of IHD in Great Britain is high in all social classes, there would seem to be little justification for any overall policy for prevention of IHD to focus on social class. However, anti-smoking strategies might well take into account the social class differences described.

A scoring system to identify men at high risk of a heart attack

A scoring system for identification of men at high risk of a heart attack within 5 years of screening is presented. The full scoring system includes an electrocardiogram and blood cholesterol measurement and the top fifth of the distribution of this full score yields 59% of the major ischaemic heart disease events occurring in the 5 years after screening. An intermediate scoring system, without an electrocardiogram but retaining blood cholesterol, yields 58% of cases from the top fifth of the score distribution. A basic (GP) score, without electrocardiogram or blood cholesterol measurement, yields 54% of cases and is recommended for use in opportunistic screening in general practice. This high risk strategy would increase public awareness of the size of the problem, help to prevent premature death and provide a useful complement to the population strategies of health education and government policy.

Alcohol and ischaemic heart disease in middle aged British men

The relation between alcohol intake and ischaemic heart disease was examined in a large scale prospective study of middle aged men drawn from general practices in 24 British towns. After an average follow up of 6.2 years 335 of the 7729 men had experienced a myocardial infarction (fatal or non-fatal) or sudden cardiac death. No significant relation was found between reported alcohol intake and the incidence of such events. Though the group of light daily drinkers had the lowest incidence of ischaemic heart disease events, it also contained the lowest proportion of current smokers, had the lowest mean blood pressure, had the lowest mean body mass index, and contained the lowest proportion of manual workers. These characteristics are more likely to account for the apparent protective effect of alcohol against ischaemic heart disease than a direct effect of alcohol. Compared with the effects of established risk factors alcohol seems to be quite unimportant in the development of ischaemic heart disease.

Interim analyses in randomized clinical trials: ramifications and guidelines for practitioners

Recent developments in group sequential methods have had a great impact on the design and analysis of randomized clinical trials. The consequences for both planned and unplanned interim analyses are discussed using several real trials as illustrations. Guidelines for the conduct of interim analysis are given, including tables of nominal significance levels and required sample sizes for several group sequential plans. Areas in need of further theoretical advance include multiple endpoints, estimation of treatment differences, stratification, and design of multiple-armed trials.

Lead exposure and children's intellectual performance

The Institute of Child Health/Southampton study is the largest cross-sectional survey of lead exposure and children's intelligence. 402 six year olds in London with tooth lead concentration in three pre-defined ranges were selected for neuropsychological testing. This paper presents new findings on the relationship between child IQ and tooth lead levels which build on previous findings in four respects: Rather than simply classifying children into high, medium and low lead groups the actual concentrations of lead in each child's tooth have been used to provide a more powerful assessment of the association between IQ and body lead burden. The influence of parental and social factors on child IQ is explored in detail in order to see if any residual lead-IQ association exists after allowance for such confounders. The methods of multiple regression, including an 'optimal' statistical policy, are more fully described. The possibility of interactions between lead and confounders is explored. Findings are that parental IQ is the most important influence on child IQ, though several other factors (eg: family size, social class and quality of marital relationships) were also significantly related. There was no overall evidence that tooth lead concentrations were related to child IQ once these other factors were taken into account. However, a significant interaction between tooth lead and sex of child indicates that the lead-IQ association appears much more pronounced in boys. This unexpected finding needs cautious interpretation and further exploration in other studies.

Giving up smoking and the risk of heart attacks. A report from The British Regional Heart Study

In a prospective study of 7735 middle-aged men, both current and ex-cigarette-smokers had a risk of a major IHD event, within an average 6.2 years of screening, more than twice that in men who had never smoked cigarettes; men who had given up smoking more than 20 years ago still had an increased risk. This excess risk among ex-smokers is only to a small extent explained by their higher blood pressure, serum total cholesterol, and body-mass index. An increased prevalence of IHD in men who had recently given up smoking also made a small contribution to excess risk. In both current and former cigarette smokers, the number of years a man had smoked cigarettes ("smoking-years") was the clearest indicator of IHD risk due to cigarettes. The major benefit of giving up smoking may lie in halting the accumulation of smoking years.

Identifying men at high risk of heart attacks: strategy for use in general practice

A strategy was devised for identifying men at high risk of acute myocardial infarction or sudden ischaemic death. A risk score was devised using cigarette smoking, mean blood pressure, recall of ischaemic heart disease or diabetes mellitus diagnosed by a doctor, history of parental death from "heart trouble," and the presence of angina reported on a questionnaire. The top fifth of the score distribution identified 53% of ischaemic heart disease cases--that is, men who subsequently experienced major ischaemic heart disease over the next five years. The addition of serum total cholesterol concentration and electrocardiographic evidence only slightly improved prediction (to 59%) and would have considerably increased the cost and effort of screening. Using this risk score on an opportunistic basis could be particularly valuable in general practice. Management of this high risk group is regarded as appropriate medical care and is complementary to the population approach to preventing ischaemic heart disease. Such a strategy for reducing the incidence of and mortality from ischaemic heart disease in men at high risk would also increase professional and public awareness of the need for preventive action.

High density lipoprotein cholesterol is not a major risk factor for ischaemic heart disease in British men

The concentration of high density lipoprotein cholesterol (HDL cholesterol) in serum was measured at initial examination in a large prospective study of men aged 40-59 drawn from general practices in 24 British towns. After an average follow up of 4.2 years 193 cases of major ischaemic heart disease had been registered in 7415 men in whom both HDL cholesterol and total cholesterol values had been measured. The mean HDL cholesterol concentration was lower in the men with ischaemic heart disease ("cases") compared with other men, but the difference became small and non-significant after adjustment for age, body mass index, blood pressure, cigarette smoking, and concentration of non-HDL cholesterol. The higher mean concentration of non-HDL cholesterol in "cases" remained highly significant after adjustment for other factors. Men in the highest fifth of non-HDL cholesterol values had over three times the risk of major ischaemic heart disease compared with men in the lowest fifth. Multivariate analysis showed that non-HDL cholesterol was a more powerful predictor of risk than the HDL to total cholesterol ratio. These British findings were compared with six other prospective studies. All the larger studies showed similar results, suggesting that HDL cholesterol is not a major risk factor in the aetiology of ischaemic heart disease.

Risk factors for ischaemic heart disease: the prospective phase of the British Regional Heart Study

Risk factors for major ischaemic heart disease (acute myocardial infarction or sudden death) have been investigated in a prospective study of 7735 men aged 40-59 years drawn from general practices in 24 British towns. After a mean follow-up of 4.2 years, there have been 202 cases of major ischaemic heart disease. Univariate estimates of the risk of ischaemic heart disease show that serum total cholesterol, HDL-cholesterol and triglyceride concentrations, systolic and diastolic blood pressures, cigarette smoking, and body mass index are all associated with increased risk of ischaemic heart disease. Evidence of ischaemic heart disease at initial examination is also strongly associated with increased risk of subsequent ischaemic heart disease. All these factors were then considered simultaneously using multiple logistic models. Definite myocardial infarction on electrocardiogram and recall of a doctor diagnosis of ischaemic heart disease remained predictive of subsequent major ischaemic heart disease, after allowance for all other risk factors. Serum total cholesterol, blood pressure, and cigarette smoking each remained as highly significant independent risk factors whereas overweight, above average levels of HDL-cholesterol and serum triglyceride were not predictive of risk after allowance for the above factors. Men with and without pre-existing ischaemic heart disease were examined separately in the same way (using multiple logistic models). The strength of association between the principal risk factors and subsequent major ischaemic heart disease was reduced in the men with pre-existing ischaemic heart disease, only age and serum total cholesterol remaining highly significant. Overall the levels of the major risk factors commonly encountered in British men have a marked effect on the risk of ischaemic heart disease. Modification of these risk factors in the general population constitutes an important national priority.