Multi-centre trial analysis revisited

Analyses of multi-centre trials must consider the effects of the individual centres and the possibility of non-constancy of treatment effect differences among centres. This usually means an ANOVA with terms for centres, treatments, and centre x treatment interactions in practice, at least in the U.S.A. Empirical and conventional Bayes methods provide attractive alternatives to conventional ANOVAs for analysing and reporting the findings from multi-centre trials and do not require more restrictive assumptions than the ANOVA approach. These approaches require regarding the centre effects as random instead of fixed, a view which often will reasonably describe outcomes of clinical trials in spite of the fact that the individual centres certainly do not comprise a random sample of all possible centres. The components of these approaches are well understood and have been employed in related applications such as meta-analysis. Combining them in a way that makes their application to routine multi-centre trial analysis relatively straightforward does not appear to have been described previously, and is what forms the topic of this paper. The empirical Bayes approach leads to useful graphical displays, including one with the data superimposed on probability contours of the joint distribution of the individual centre means and standard deviations, which provides a handy way to identify possible outliers. Covariates can be incorporated without difficulty. The Bayes approach, implemented with Gibbs sampling, provides a convenient way to construct posterior and predictive distributions for a variety of useful statistics. We compare the result of empirical and conventional Bayes analyses with the result of fixed and mixed model ANOVAs applied to data from a multi-centre trial.     

Clinical trial designs--made to order

At times, clinical trial design presents difficult problems for the consulting statistician. The nature of the disease under study or the inadequacies of standard methodology often suggest alternative strategies for the development of useful clinical trial designs. This paper examines two case of novel clinical trial designs that are "made-to-order" for the problem at hand. Both address disease-specific issues. The first case considers the use of adaptive allocation of treatments to patients in a trial in major depressive disorder. The second case reviews the design of a novel efficacy parameter in advanced pancreatic cancer.     

Admissions testing on trial.

Criticisms of test use in educational admissions decisions frequently stem from fundamental disagreements about underlying values but may still be couched in test-specific technical terms. A. Nairn's (1980) claims regarding predictive validity are discussed, and the erroneous and misleading nature of these claims is demonstrated. The alternative of relying on previous grades alone rather than in combination with tests for purposes of admissions is reviewed. Criticisms of tests based on issues of bias, coachability, and secrecy are discussed and evaluated. (63 ref) (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Overexpectation and trial massing.

Three experiments were conducted to examine the interaction of overexpectation treatment and trial massing using a Pavlovian fear conditioning procedure with rats. In first-order conditioning, Experiment 1 found the overexpectation effect (i.e., decreased conditioned responding to a cue after compound training when the elements were previously reinforced), the trial spacing effect (i.e., decreased responding to a cue when reinforced trials are massed), and a counteraction between overexpectation treatment and trial massing (i.e., an alleviation of the decrement in responding seen with overexpectation treatment or trial massing alone when the two treatments are conjointly administered). Experiment 2 replicated Experiment 1 with the critical treatments embedded within a sensory preconditioning preparation. The overexpectation effect, the trial spacing effect, and the mutual counteraction of overexpectation treatment and trial massing all proved significant. In Experiment 3, either the nontarget conditioned stimulus of overexpectation treatment or the excitatory context resulting from trial massing was extinguished. Results are best explained by the extended comparator hypothesis (J. C. Denniston, H. I. Savastano, & R. R. Miller, 2001). (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Carbamazepine trial for Lesch-Nyhan self-mutilation

Anticonvulsants may reduce the self-mutilation of acquired sensory neuropathy, and one report described sensory neuropathy in an older patient with Lesch-Nyhan syndrome. We performed nerve and muscle biopsies on four patients with Lesch-Nyhan syndrome and initiated an uncontrolled pilot trial to see if carbamazepine would reduce the self-mutilation in these patients. All of the boys had clinical features typical of Lesch-Nyhan syndrome, and the diagnosis was confirmed in each by enzyme analysis. No specific abnormalities were identified in either nerve or muscle. Nevertheless, self-mutilation and the need for constant restraint diminished in all four patients, though in one the effect was only transient. Two patients had increased self-mutilation when carbamazepine was stopped, then improved a second time when treatment was restarted. Sensory neuropathy was not confirmed, so any effect of carbamazepine is likely to be on the central nervous system.     

Designing a cost-effective clinical trial

Researchers and administrators must decide which clinical trials are worth doing and how many subjects are needed for a trial. We calculated sample size considering the costs of implementing the results of the trial and the trial costs using (1) Neyman-Pearson methods and (2) a Bayesian cost-benefit method. We illustrate these methods in a clinical trial sponsored by the National Institutes of Health that compares two levels of blood urea nitrogen clearance by haemodialysis for patients with end-stage renal disease. When applied to evaluations of research proposals, these methods may help researchers to decide whether to begin a study, and, if so, how many subjects to enrol in it. These methods should be especially useful for large studies intended to inform health policy.     

Multicenter trial of Enterococcus antibiotic susceptibility

Antibiotic susceptibility of 446 Enterococcus isolates from 9 medical centres of Moscow and St. Petersburg was tested. Among the isolates 386 belonged to E.faecalis, 48 to E.faecium and 12 to the other species. All the isolates were susceptible to vancomycin. As for E.faecalis 84 and 85 per cent of the isolates were susceptible to ampicillin and ampicillin/sulbactam respectively (no production of beta-lactamases), the frequency of high resistance to aminoglycosides amounted to 44 per cent with respect to streptomycin and to 25 per cent with respect to gentamicin, 75 per cent of the isolates was susceptible to ciprofloxacin. As for E.faecium and the rare species of Enterococcus more than 70 per cent of the isolates was resistant to ampicillin and ampicillin/sulbactam, the frequency of high resistance to aminoglycosides exceeded 60 per cent, 17 and 25 per cent of the isolates were susceptible to ciprofloxacin.     

Informed consent in phase I trial

The three-dimensional structure of a complex of cinnamycin, a 19-amino acid residue immunopotentiator peptide, and lysophosphatidylethanolamine was determined by 1H-NMR. The complex was cylindrical in shape, 11 A in diameter and 26 A in length, excluding the acyl chain of the phospholipid. The peptide had a hydrophobic pocket surrounded by residues Phe-7 through Ala(S)-14 to bind to the head group of the ligand. Fitting of the head group to the hydrophobic pocket was so good that other than a glycerophosphoethanolamine head group would be unable to fit the pocket. The goodness of the fitting is compatible with the strict specificity of ligand binding of the peptide.