PubMed Health⌕ Search

Biomedical subjects

D Salsburg

Publications and source records attributed to D Salsburg.

15 recordsLinked to original sources
PubMed1999

Why analysis of variance is inappropriate for multiclinic trials.

Violations of the assumptions behind analysis of variance (ANOVA) models do not tend to affect the alpha-level but can greatly decrease the power of a clinical trial to detect treatment effects. The very nature of multiclinic studies guarantees the violation of some of these assumptions. In this article, I explore the reduction in power that results from two of these violations--heterogeneity of variance across sites and the existence of "floor" and "ceiling" effects. I propose other methods of statistical analysis that avoid this loss of power.

Analysis of Variance↗
PubMed1993

The use of statistical methods in the analysis of clinical studies.

It is "standard" to analyze data from a clinical trial using a narrowly defined probabilistic mathematical model. This paper examines the ways in which mathematical models, in general, can be used in clinical research, the meaning of probability in the examination of clinical trials, and the philosophical flaws in the current "standard" method. An alternative formulation is proposed which is more flexible and which comes closer to meeting the needs of medical science. In this alternative formulation, significance tests are applied to the data from a study only as a first step to determine whether the data are worth further examination. After that, clinically relevant questions are answered with 50 and 95% confidence bounds. The initial significance test is tailored to be directed at a narrow class of hypotheses that, in turn, are dictated by clinical expectations.

Clinical Trials as Topic↗
PubMed1991

Situations where formal confirmatory analysis is inappropriate for clinical research.

When new drugs are developed for indications where no effective therapies have been established, such as emphysema, useful measures of clinical activity have to be derived from a large number of proposed measures. It often happens that most of the proposed measures are of little use in tracking the disease and that the numbers of patients needed to establish whether the drug is effective is very large. Thus, the first study in man may be the size of the usual Phase III "confirmatory" study, and the expense and the time needed to complete it are so great that decisions about the future development of the drug will depend upon this one large study. The situation of such innovative therapies is examined, and methods for controlling the error rate when almost all analyses have to be exploratory are proposed.

Biopharmaceutics↗
PubMed1990

Estimating dose response in chronic toxicity studies: 24-month administration of acrylonitrile in drinking water of Fischer 344 rats.

It is shown that low-dose extrapolation based on observed numbers of animals with tumors cannot be used within the standard frameworks of either toxicological or statistical methods since the conclusions cannot be examined critically with additional experiments. The traditional methods of examining dose response in toxicology are considered, and it is shown that they cannot be used as long as the studies are carried out to the animals' old age and the controls have senile lesions. Traditional multivariate statistical methods are proposed to examine the effect of dose on the shift in senile lesions that is induced by treatment. One method (using standard software for the computation of linear discriminant functions) is then applied to the tumorous and non-tumorous lesions found in this study. It is shown that doses as low as 3 ppm have a statistically significant effect, and 1 ppm is proposed as the "no mean effect level".

Acrylonitrile↗
PubMed1989

Use of restricted significance tests in clinical trials: beyond the one- versus two-tailed controversy.

Whether to use "one-tailed" or "two-tailed" tests for clinical trials is examined in the more general context of restricted tests, developed by J. Neyman. This is complimented by a discussion of theoretical work of Savage and Bahadur. It is concluded that all tests of statistical significance have to be "restricted" in some sense and that failure to recognize the restrictions implied by "standard" methods of analysis can have a negative effect on the ability to detect treatment effects of new drugs in phase II and early phase III clinical trials.

Analysis of Variance↗
PubMed1989

Does everything "cause" cancer: an alternative interpretation of the "carcinogenesis" bioassay.

R. S. Chhabra, J. Haseman, A. Hall, and C. Baskin (Fundam. Appl. Toxicol. 11, 685-690, 1988) have raised the question of whether the apparent antitumorigenic effect of 4-hexylresorcinol suggests its possible use as an antineoplastic agent. In response to that paper, it is noted that over 20% of the compounds subjected to the "carcinogenesis" bioassay of the NCI/NTP program show apparent antitumorigenic effects of a similar sort. A thought experiment is proposed that shows that any compound subjected to these test conditions should produce a shift in patterns of tumors, regardless of whether the compound is a true "carcinogen" or a true "antineoplastic" agent. This thesis is illustrated with an alternative statistical analysis of six AZO dye studies that were conducted in the same lab during the same period, and the existence of apparent effects for three other other relatively innocuous substances is noted. Short-term experiments are proposed to test this thesis. Finally, the implications of this thesis on the interpretation of "carcinogenesis" bioassays is examined.

Animals↗
PubMed1986

Alternative hypotheses for the effects of drugs in small-scale clinical studies.

New drugs that will be investigated in the future are expected to deal with chronic diseases, where the number of patients available for controlled clinical trials will be small and where the long-term sequelae that it is hoped will be ameliorated take a long time to occur. Thus, it would be useful to construct powerful tests of hypotheses that can be used to compare subtle intermediate measures of change in condition. This paper examines the probabilistic characteristics of these measures and the changes in them that might be expected from drug therapy.

Aged↗
PubMed1985

A statistical examination of historical controls for mouse bone marrow cytogenetic assays.

Data from 1,111 controls from assays run over 11 years are examined to determine a most powerful statistical procedure for detecting a mutagenic effect. It is concluded that the data do not show a constant probability of chromosomal abnormalities and that the data do not fit a simple Poisson or binomial distribution. Empirical Bayes techniques are used to derive a test that declares an effect if three or more cells in a group of 50 tested contain abnormalities.

Analysis of Variance↗
PubMed1981

Development of statistical analysis for single dose bronchodilators.

When measurements developed for the diagnosis of patients are used to detect treatment effects in clinical trials with chronic disease, problems in definition of response and in the statistical distributions of those measurements within patients have to be resolved before the results of clinical studies can be analyzed. An example of this process is shown in the development of the analysis of single-dose bronchodilator trials.

Bronchodilator Agents↗
PubMedDate not supplied

A comparison of statistical methods for low dose extrapolation utilizing time-to-tumor data.

The assessment of health risks due to low levels of exposure to potential environmental hazards based on the results of toxicological experiments necessarily involves extrapolation of results obtained at relatively high doses to the low dose region of interest. In this paper, different statistical extrapolation procedures which take into account both time-to-response and the presence of competing risks are compared using a large simulated data base. The study was designed to cover a range of plausible dose response models as well as to assess the effects of competing risks, background response, latency and experimental design on the performance of the different extrapolation procedures. It was found that point estimates of risk in the low dose region may differ from the actual risk by a factor of 1000 or more in certain situations, even when precise information on the time of occurrence of the particular lesion of interest is available. Although linearized upper confidence limits on risk can be highly conservative when the underlying dose response curve is sublinear in the low dose region, they were found not to exceed the actual risk in the low dose region by more than a factor of 10 in those cases where the underlying dose response curve was linear at low doses.

Animals↗
PubMedDate not supplied

The lifetime feeding study in mice and rats--an examination of its validity as a bioassay for human carcinogens.

Currently used designs for carcinogenic bioassay (lifetime feeding studies in mice or rats using maximum tolerated doses of the test compound) are examined to see if they meet the requirements of a bioassay, using the results of 170 compounds reported on as of June, 1980, by the National Cancer Institute. It is concluded that the lifetime feeding study has never been subjected to proper validation as an assay for human carcinogens. When an attempt is made to validate it on the basis of these reported studies and those in the literature, it appears to lack acceptable specificity and sensitivity. It is suggested that a drastically different design is needed and that such redesigning of the assay will require proper validation.

Animals↗