Confidence intervals and p-values for Williams' and other step-down multiple comparison tests against control.

When single-step multiple comparison tests against control, such as Dunnett's test, are used, p-values and confidence intervals can be reported. However, Williams' test and other step-down multiple comparison tests only provide results in terms of statistical significance. In this paper, approximate confidence intervals are proposed for some step-down multiple comparison tests and their coverage properties are assessed by simulation; p-values are readily calculated. The proposed simultaneous confidence intervals associated with Williams' test, Dunnett's step-down test, and the closed t test are all found to have good coverage, typically between 94% and 96% for a nominal value of 95%. Thus practicing statisticians can now quote p-values for these tests and use simple confidence intervals to aid interpretation of test results.

Minimization in crossover trials with non-prognostic strata: theory and practical application.

To obtain meaningful results in any clinical trial, patients need to be allocated to treatments in such a way that valid analysis can be carried out. Balancing treatment groups before analysis is carried out is more desirable than trying to compensate for incomparability at a later date. Therefore, the development of allocation procedures to produce comparable groups in which prognostic factors are equally represented is important. Minimization, a deterministic allocation method, aims to ensure balance on such factors, particularly in small trials when traditional randomization methods are likely to fail. However, views on the use of conventional analysis following minimization are divided. The use of minimization in two randomised crossover trials is described where, in addition to the comparisons between randomised treatments, it was desired to have balance between groups based on differential trial procedures. Theoretical concerns about the use of minimization are not applicable in this setting, and therefore minimization is shown to be a useful technique for obtaining balance.

Long-term evaluation of treatment for chronic heart failure: a 1 year comparative trial of flosequinan and captopril.

Two hundred and nine patients with moderate to severe chronic heart failure, all of whom remained symptomatic despite at least 80 mg of frusemide daily, were randomized to 12 months treatment with flosequinan or captopril. The patients were stratified into two groups, a treadmill group and a corridor walk test group, depending upon their exercise capability. Sixty-five out of 102 patients randomized to flosequinan and 43 out of 107 randomized to captopril (p < 0.001) did not complete the study. There was no difference between the groups in mortality: 19 patients died while taking flosequinan and 15 while taking captopril. Both drugs had similar effects on treadmill exercise tolerance; the mean increase at week 52 was 117 seconds in the flosequinan group and 156 seconds (p = 0.57) for the captopril group. For those patients stratified to the corridor walk test only, there was also very little difference in the improvement at 52 weeks; the mean increase for patients randomized to flosequinan was 61 meters and captopril was 75 meters (p = 0.65). However, when the walk tests from all patients are examined, captopril produced a significant improvement compared with flosequinan at week 52 (p = 0.015). Flosequinan has similar long-term efficacy to captopril but is associated with a higher incidence of adverse events.

Placebo-controlled trial of flosequinan in moderate heart failure. The possible importance of aetiology and method of analysis in the interpretation of the results of heart failure trials.

One hundred-thirty five patients with moderate heart failure, recruited from 18 centres, were included in a double blind, placebo controlled study to evaluate the effects of flosequinan on symptom limited tread-mill exercise tolerance. Fifteen patients in the placebo group were withdrawn from the study compared with 14 from the group given flosequinan. New York Heart Association classification was improved at week 16 in the flosequinan group relative to those randomised to placebo (P < 0.01). Depending how the other results are analysed flosequinan either appeared to have no effect on symptom limited exercise tolerance in those who completed the study; a suggestion of superiority if an analysis at endpoint is used (P = 0.09), or, if a covariate analysis at endpoint is used, then a significant improvement can be demonstrated (P = 0.04). Subset analysis suggests that the aetiology of the heart failure and the dose of diuretics used might have a major effect on the response to treatment. The best way of analysing clinical trials in heart failure is not clear as the results can be profoundly influenced by the way data from withdrawn patients are handled. The aetiology and diuretic requirement of patients may influence their response to treatment.

A preliminary study of the effects of flosequinan on psychomotor function in healthy volunteers.

The central nervous system effects of flosequinan (100 mg), a chemically novel quinolone vasodilator, were assessed by a double-blind crossover comparison with placebo and diazepam (10 mg) in 12 healthy volunteers. After five practice sessions on a battery of automated psychomotor tests, assessments of psychomotor function and mood ratings were made on each volunteer at baseline and 1, 3, 6 and 24 h after dosing. Compared with placebo, diazepam (10 mg), the verum control, significantly (p less than 0.05) reduced subjective alertness, impaired critical flicker fusion threshold at 1 and 3 h, digit symbol substitution at 6 h, overall total choice reaction time and overall rate on two of five finger tapping tests. Flosequinan (100 mg), however, was indistinguishable from placebo in all tests with two contrasting exceptions: improved alternate right and left finger tapping (mean 5.1/s) compared to either diazepam (4.7/s) or placebo (4.8/s) (p less than 0.05), and impaired digit symbol substitution at 6 h (45.7/min) in comparison with placebo (50.7/min) (p less than 0.01). Ten volunteers reported 12 adverse effects after flosequinan treatment (10 of which were headaches), two reported drowsiness after diazepam and one reported headache after placebo. It was concluded that flosequinan has no central nervous system depressant effects despite the occurrence of headache in 10 volunteers.