Search PubMed⌕ Search

Biomedical subjects

J C Bailar

Publications and source records attributed to J C Bailar.

At least 19 recordsLinked to original sources
PubMed1988

Interactions between statisticians and biomedical journal editors.

The increased involvement of statisticians in the publication of biomedical research has resulted in increased communication between statisticians and biomedical journal editors. This paper considers ways to enhance positive interactions between statisticians and journal editors. Specific questions addressed are: What are the most serious statistical problems in manuscripts? What statistical design information should the methods include? How can editors identify papers that need statistical review? How can editors interpret what statistical reviewers say? How can editors identify statisticians who are willing to review manuscripts?

Humans↗
PubMed1988

One-hit models of carcinogenesis: conservative or not?

One-hit formulas are widely believed to be "conservative" when used to analyze carcinogenesis bioassays, in the sense that they will rarely underestimate risks of cancer at low exposures. Such formulas are generally applied to the lifetime incidence of cancer at a specific site, with risks estimated from animal data at zero dose (control), and two or more additional doses that are appreciable fractions of a maximum tolerated dose. No empirical study has demonstrated that the one-hit formula is conservative in the sense described. The Carcinogenesis Bioassay Database System contains data on 1212 separate bioassays of 308 chemical substances tested at exactly three evaluable doses. These provided sufficient data to examine 8432 specific combinations of cancer site with sex, species, and chemical. For each of these we fitted a one-hit formula to the zero and maximum dose data points, then examined the relation of the fitted curve to the incidence rate observed at the mid-dose, with and without adjustment for intercurrent mortality. Both underestimates and overestimates of risk at mid-dose occurred substantially more often than expected by chance. We cannot tell whether such underestimates would occur at lower doses, but offer six biological reasons why underestimates might be expected. In a high percentage of animal bioassays, the one-hit formula is not conservative when applied in the usual way to animal data. It remains possible that the one-hit formula may indeed be conservative at sufficiently low doses (below the observational range), but the usual procedure, applied to the usual dose range, can be nonconservative in estimating the slope of the formula at such low doses. Risk assessments for regulation of carcinogens should incorporate some measure of additional uncertainty.

Animals↗
PubMed1988

Guidelines for statistical reporting in articles for medical journals. Amplifications and explanations.

The 1988 edition of the Uniform Requirements for Manuscripts Submitted to Biomedical Journals includes guidelines for presenting statistical aspects of scientific research. The guidelines are intended to aid authors in reporting the statistical aspects of their work in ways that are clear and helpful to readers. We examine these guidelines for statistics using 15 numbered statements. Although the information presented relates to manuscript preparation, it will also help investigators in earlier stages make critical decisions about research approaches and protocols.

Periodicals as Topic↗
PubMed1987

Have we reduced the risk of getting cancer or of dying from cancer? An update.

We have examined trends in cancer mortality, incidence and survival in the United States to update our earlier work and respond to criticisms. [Bailar, J.C., Smith, E.M.: New Engl. J. Med. 314, 1226 (1986).] Here we concentrate on the years 1975-1984, and show that overall cancer mortality has increased, incidence has increased and case survival is virtually unchanged. This generally unfavorable picture is scarcely changed when lung cancer is excluded from the trends. While trends for individual cancers have been mixed, overall progress in both curative treatment and prevention has been minimal. This evaluation does not deny the marked progress in treating some uncommon forms of cancer, improved palliation, reduced extent or severity of treatment, or benefits of cancer research that can be applied in other areas of medicine. While our finding of limited progress is not new, we believe that it requires increased attention in setting the course of future research initiatives, demonstration programs, medical training and clinical practice.

Humans↗
PubMed1986

Progress against cancer?

We assessed the overall progress against cancer during the years 1950 to 1982. In the United States, these years were associated with increases in the number of deaths from cancer, in the crude cancer-related mortality rate, in the age-adjusted mortality rate, and in both the crude and the age-adjusted incidence rates, whereas reported survival rates (crude and relative) for cancer patients also increased. In our view, the best single measure of progress against cancer is change in the age-adjusted mortality rate associated with all cancers combined in the total population. According to this measure, we are losing the war against cancer, notwithstanding progress against several uncommon forms of the disease, improvements in palliation, and extension of the productive years of life. A shift in research emphasis, from research on treatment to research on prevention, seems necessary if substantial progress against cancer is to be forthcoming.

Age Factors↗
PubMed1986

Science, statistics, and deception.

Some common scientific practices cannot quite be called lying, though they are potentially, and sometimes deliberately, deceptive. Some examples are the failure to explain to readers all the weaknesses in data, statistical testing of post hoc hypotheses, fragmentary or selective reporting of findings, and reporting as "negative" a study that had insufficient chance of detecting an effect. The first step toward controlling potential problems is a redefinition of ethical standards to bar readily avoidable as well as deliberate deception. Other remedies include substantially greater restraint in the use of questionable practices, full disclosure and justification each time these practices are used, and greater skepticism by readers. Pressures to publish tend to promote deception. A broadened concept of ethical standards in science should be reflected in training programs and in the structure of scientific rewards, including a sharply reduced emphasis on publication as an end in itself.

Research Design↗
PubMed1984

A classification for biomedical research reports.

Biomedical research uses a wide range of designs applied to problems in laboratory, clinical, and population settings. Whatever the nature of the study, a few key features--such as the admission rule, the method of allocating subjects to treatments, and the use of controls--largely determine the strength of scientific inferences. We used these and other features to classify the 332 Original Articles published in the New England Journal of Medicine during 1978-1979. This classification directs attention to critical aspects of study design and performance and can help in the choice of suitable research approaches and protocols. It emphasizes the critical role of the investigators' intent in performing and analyzing a study, and it alerts readers to important aspects of interpretation. We recommend that authors always report enough detail about their work for readers to apply this or a similar classification. Omission of such detail may limit the interpretation of a research study because a study that cannot be classified has probably been incompletely reported.

Clinical Trials as Topic↗
PubMed1984

Studies without internal controls.

Sometimes questions of clinical interest can be addressed only by investigations without concurrent controls that are under the supervision of the investigator (internal controls). Such studies nearly always make use of other types of comparisons (external controls), such as historical controls. In this paper we examine studies of clinical treatments that have weak internal controls or lack internal controls, as illustrated by recent examples from the Journal. These studies have a small but important and unique role in clinical investigation. Five interrelated features can add to the strength of such studies: (1) an intent by the investigator, expressed before the study, that the treatment will affect the outcomes reported; (2) planning of the analysis before the data are generated; (3) articulation of a plausible hypothesis before the results are observed; (4) a likelihood that the results would still have been of interest if they had been "opposite" in some sense; and (5) reasonable grounds for generalizing the results from the study subjects to a substantially broader group of patients. In spite of potential pitfalls, carefully selected and reported studies without internal controls can play a substantial part in the acquisition of scientific knowledge.

Animals↗
PubMed1984

Crossover and self-controlled designs in clinical research.

Crossover studies (clinical trials in which each patient receives two or more treatments in sequence) and self-controlled studies (in which each patient serves as his or her own control) can produce results that are statistically and clinically valid with far fewer patients than would otherwise be required. We investigated the use of the crossover design in the 13 crossover studies that appeared in the Journal during 1978 and 1979. We considered the following important features of design and analysis as they applied to these studies: the method by which patients were assigned to initial treatment (only 7 of 13 studies used random assignment); the determination of when to switch treatments (10 of the 13 used a time-dependent rule, and 3 a less appropriate disease-state-dependent rule); blinding of the crossover point (in only 3 of the 13 studies was the crossover point concealed, but in 4 of the remaining 10 concealment was impossible); assessment of the effects of the order of treatments (included in only 1 of the 13 studies); and the use of at least minimally acceptable statistical analysis (11 of the 13 studies had such an analysis). We also report briefly on 28 additional studies of a single treatment each, in which each patient served as his or her own control before or after treatment or both. The scientific issues were much the same as in crossover studies except that self-controlled comparisons of treatments tended to be less precisely designed and conducted and to focus on clinical problems and patient groups that are especially difficult to study.

Chronic Disease↗
PubMed1984

The multihit model of carcinogenesis: etiologic implications for colon cancer.

A new multihit model of carcinogenesis is developed for use in evaluating age-specific cancer incidence rates in human populations. The model allows for some heterogeneity in both risk (perhaps genetic) and pathway (number of hits). Fitting the model yields estimates of (1) levels of effect of background exposure to environmental agents, (2) tumor growth times after initiation of a malignant cell, and (3) relative sizes of high-risk groups in a human population. Maximum likelihood procedures are used to fit the model to the polyposis coli data of Veale and the colon cancer incidence data from the Third National Cancer Survey. Model estimates may be verified in some cases by review of independent data in the literature and results have both theoretical and practical implications. Findings are generally consistent with the adenoma-carcinoma etiologic sequence postulated by Hill, Morson and Bussey with one exception. A large proportion of the population may be at risk of four-hit colon tumors following a non-adenoma etiologic sequence.

Adolescent↗