Sample size determination in complex clinical trials comparing more than two groups for survival endpoints

This paper presents a sample size formula for testing the equality of kappa (> or = 2) survival distributions using the Tarone-Ware class of test statistics in the presence of non-proportional hazards, time dependent losses, non-compliance and drop-in. This method extends the derivation by Lakatos of a sample size formula for comparing two survival distributions. A sample size formula is also presented for the stratified logrank test. We describe how one can utilize these generalized formulae in calculating sample sizes and assessing power in complex multi-arm clinical trials.     

Efficient allocation of patients to treatment cells in clinical trials with more than two treatment conditions

OBJECTIVE: Clinical trials generally allocate patients to equal-sized treatment groups. The authors propose that it may be more efficient to allocate unequal proportions of the total sample size to treatments when more than two treatments are being compared. METHOD: This proposal is illustrated with two examples. One involved a comparison of three treatments and used a dichotomous categorical outcome. The other involved comparison of three treatments and used a continuous measure. RESULTS: In both examples, a considerable increase in efficiency was realized by reducing the number of patients assigned to the placebo cell. CONCLUSIONS: Unequal allocation of patients to treatments should be considered when more than two groups are compared.     

Statistical significance testing and clinical trials.

The efficacy of treatments is better expressed for clinical purposes in terms of these treatments' outcome distributions and their overlapping rather than in terms of the statistical significance of these distributions' mean differences, because clinical practice is primarily concerned with the outcome of each individual client rather than with the mean of the variety of outcomes in any group of clients. Reports of the obtained outcome distributions for the comparison groups of all competently designed and executed randomized clinical trials should be publicly available no matter what the statistical significance of the mean differences among these groups, because all of these studies' outcome distributions provide clinically useful information about the efficacy of the treatments compared. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

Statistical applications for in vitro diagnostic tests and other medical device clinical trials

Some statistical methods applied to in vitro diagnostic tests for the three primary indications (screening, diagnosis, and monitoring) are discussed. Various examples with practical statistical applications are presented, including test for k by k ordered categorical matched-pair data for screening of cervical cancer, receiver operating characteristic (ROC) curve for diagnosis or screening, and the Cox time-to-event model to estimate relative risk of first cancer progression by monitoring carcinoembryonic (CEA) levels for stage IV breast cancer patients.     

On the use of the ratio or the odds ratio of cure rates in therapeutic equivalence clinical trials with binary endpoints

Acutely ill patients who are immunocompromised but not neutropenic most commonly are: (1) diabetic; (2) on chronic high-dose steroid therapy; (3) have lupus; or (4) have impaired or absent splenic function. These patients often present in the CCU because of the severity of their infection. Differential diagnosis may be approached by first considering the patient's underlying disease, i.e., SLE. The next step in the diagnostic process is to appreciate the immune defect associated with these disorders. The nature of the immune defect determines which clinical pathogens are related to the immune defect. Pathogens are associated with a sterotyped pattern of organ involvement. The object of the diagnostic analysis is to determine the most likely organism affecting a particular organ system, given the defect in host defenses associated with the patient's underlying illness. In this way, a useful clinical diagnosis can be made rapidly, and appropriate clinical specimens obtained for diagnostic testing. Often empiric therapy must be started pending the results of diagnostic testing. In such situations, empiric therapy ordinarily is directed against the bacterial pathogens most likely to cause disease relevant to the patient's impaired defenses. Specific therapy for unusual or exotic pathogens should not be empiric and should be based on demonstration of a pathogenic role by the microorganism. In the case of miliary tuberculosis or invasive fungal disease, a case may be made for early empiric therapy to cover these organisms if there is sufficient clinical suspicion based on the presenting signs and symptoms as well as the pattern of organ involvement. As with all infections, but particularly in immunocompromised patients, the early initiation of appropriate antimicrobial therapy is essential and often life-saving.     

Long-term clinical angiographic follow-up undergoing PTCA more than 10 years

Previously, it has been shown that human immunodeficiency virus (HIV)-1 envelope proteins gp160 and gp41 bind to Candida albicans. Whether this interaction affects candidal virulence in vitro was investigated. HIV-1 gp160 or gp120 treatment of C. albicans significantly altered neither growth nor phospholipase activity of the fungus. However, treatment of C. albicans with gp160, but not with gp120, led to an elevation of free and cell-bound aspartate proteinase. In addition, culture supernatants obtained from C. albicans treated with gp160 or gp41, but not with gp120, showed a strong increase in proteinase activity. Finally, C. albicans viable yeast cells treated with gp160 or gp41 and serum were phagocytosed by polymorphonuclear leukocytes to a lesser extent than was C. albicans treated with gp120 and serum or serum alone. These findings suggest that the interaction between HIV-1 gp160 and C. albicans may promote the virulence of C. albicans in HIV-1-positive patients.     

A comparative study of patients with cancer of the ovary, who have survived more or less than 10 years

With the purpose of elucidating the characteristics in patients with cancer of the ovary who have survived 10 years compared with patients who have not survived 10 years, 161 patients with epithelial cancer and 15 patients with granulosa cell tumor who have survived 10 years, group A, have been compared with 157 patients with epithelial cancer and 14 patients with granulosa cell tumor, who have not survived 10 years, group B. The study showed that among epithelial tumors the stage of tumor and the histological picture in the form of "low potential malignance" or adenocarcinoma was of decisive importance for 10 years survival. No correlation between information such as age, marital status, profession, duration of symptoms, and the nature of symptoms and 10 years survival could be found in patients with tumor in stages I and II. Neither could the gynaecological examination, the nature of the surgical treatment nor radiation therapy be correlated with 10 years survival in stages I and II. 32 of the 152 patients with epithelial cancer in stages I and II, who have survived 10 years, have either died from their ovarian disease or have later on developed cancer localized to cervix ro corpus uteri. The author points out the risk of not removing both ovaries as well as the uterus and recommends that such patients are followed for many years after the treatment. In 29 patients with granulosa cell tumor neither the medical history, the stage of the tumor nor the treatment could be correlated with 10 years survival.     

Correction for attenuation and the equivalence of tests.

Block (1963) suggested that in estimating the equivalence of tests E should correct for unreliability to give a more appropriate estimate of the "conceptual equivalence" of the tests. In the present paper it is pointed out that: a distinction should be made between conceptual equivalence and functional equivalence; functional equivalence is of prior importance in research seeking relationships of other variables to the variable represented by the tests; in estimating functional equivalence it is not apropriate to correct for unreliability; and, in fact, correction for attenuation may often lead to over-generalization of results. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Successful aging calls for more than drugs

Paracetamol is a common cause of fatal self-poisoning in the UK every year. Despite this, it continues to be sold freely without medical supervision and can be found in quantity in most household medicine cabinets.     

Are dysphoric individuals more suggestible or less suggestible than nondysphoric individuals?

Dysphoric (n?=?75) and nondysphoric (n?=?64) university students completed the Gudjonsson Suggestibility Scale—Form 2 (GSS2; G. Gudjonsson, see record 1988-03263-001). The GSS2 measures interrogative suggestibility, the tendencies to yield to leading questions and to shift answers in response to interrogative pressure. Dysphoric participants were more suggestible than were nondysphoric individuals, yielding more to leading questions and shifting their answers more after receiving negative feedback. These results imply that people suffering from dysphoria may make highly suggestible therapy clients. (PsycINFO Database Record (c) 2010 APA, all rights reserved)     

Introduction to the special section: Contextualizing significance testing in clinical trials.

The presentation and interpretation of clinical trial data is of crucial importance to psychotherapy research and practice. This introduction briefly describes how this Special Section on significance testing in clinical trials came about, as well as some of the content included in the articles. Between the original theoretical article and the four invited comments, this Special Section provides a concise and accessible overview of current thinking regarding the limitations of clinical trial data, particularly significance testing, as well as improvements and supplements to these analyses that may benefit both psychotherapy research and those who use this information in applied practice. (PsycINFO Database Record (c) 2011 APA, all rights reserved)     

A computer program for designing clinical trials with arbitrary survival curves and group sequential testing

We present a program, mhghseq, for designing a clinical trial to compare two treatments. For a user-specified hypothesis and a user-specified alternative, mhghseq finds the number of subjects required to achieve a given size and power. It permits each treatment to have an arbitrary survival curve and it allows for group sequential testing. It provides solutions for small samples as well as for large samples. There may be a follow-up period after accrual is completed. Tests may be one or two-sided, either the log-rank or the Gehan test may be used, and either of two commonly used boundary forms--the O'Brien-Fleming or Pocock boundary forms--may be specified. Although mhghseq relies primarily on simulation to produce results, the large sample sequential boundary for the log-rank test optionally may be computed with an algorithm given here that is different from ones used in the past for this purpose.     

A comparative study of sibship tests of linkage and/or association

Population-based tests of association have used data from either case-control studies or studies based on trios (affected child and parents). Case-control studies are more prone to false-positive results caused by inappropriate controls, which can occur if, for example, there is population admixture or stratification. An advantage of family-based tests is that cases and controls are well matched, but parental data may not always be available, especially for late-onset diseases. Three recent family-based tests of association and linkage utilize unaffected siblings as surrogates for untyped parents. In this paper, we propose an extension of one of these tests. We describe and compare the four tests in the context of a complex disease for both biallelic and multiallelic markers, as well as for sibships of different sizes. We also examine the consequences of having some parental data in the sample.     

Crossover designs for clinical trials

I discuss three-period crossover designs for an efficient comparison of two test treatments with special application to clinical trials which often have many practical limitations. In this paper I specify a subset of three-period crossover designs so that the investigators are not left with the problematic two-period two-sequence design, should the trials be terminated after the second period. I show that there is a dramatic reduction in variability for estimating the direct and residual treatment effects in three-period designs compared to two-period designs. I also show that the universally optimal design with ABB and BAA sequences is unsuitable when a complex form of residual effects is suspected, such as the second-order residual effects or treatment by period interactions. The design with ABB, BAA, AAB, and BBA sequences is relatively robust to these uncertain model assumptions. I also discuss missing data problems and conclude that, even with a large proportion of missing values, the three-period design is far more efficient than the two-period design.     

Statistical principles for clinical trials

If a trial is to be well designed, and the conclusions drawn from it valid, a thorough understanding of the benefits and pitfalls of basic statistical principles is required. When setting up a trial, appropriate sample-size calculation is vital. If initial calculations are inaccurate, trial results will be unreliable. The principle of intent-to-treat in comparative trials is examined. Randomization as a method of selecting patients to treatment is essential to ensure that the treatment groups are equalized in terms of avoiding biased allocation in the mix of patients within groups. Once trial results are available the correct calculation and interpretation of the P-value is important. Its limitations are examined, and the use of the confidence interval to help draw valid conclusions regarding the clinical value of treatments is explored.