Learn More
BACKGROUND Although animal studies have raised the possibility that prolactin-elevating dopamine antagonists used to treat psychotic disorders may initiate and promote breast cancers, epidemiologic studies in humans have been limited and inconsistent. METHODS A retrospective cohort study was conducted of 52 819 women exposed and 55 289 not exposed to(More)
BACKGROUND There are few previous epidemiologic studies of gallbladder cancer, a rare but nearly always lethal gastrointestinal cancer with a demonstrated greater frequency in adult women and older subjects of both sexes, and also in the members of populations throughout central and eastern Europe and certain racial groups such as native American Indians.(More)
Multilayer neural networks have been faulted for functioning as "black boxes" and for failing to assess the relative importance of the input factors. The aim of this paper is to illustrate how neural networks can classify individuals. The authors investigated the role of weights in the formation of neural networks' decision surfaces and decision regions.(More)
Observational studies often provide the only available information about treatment effects. Control of confounding, however, remains challenging. The authors compared five methods for evaluating the effect of tissue plasminogen activator on death among 6,269 ischemic stroke patients registered in a German stroke registry: multivariable logistic regression,(More)
Periconceptional folic acid supplementation reduces the risk of neural tube defects (NTDs). To determine whether periconceptional exposure to folic acid antagonists (FAAs) might therefore increase the risk of NTDs, the authors examined data from an ongoing case-control study of birth defects (1979-1998) in the United States and Canada. They compared data on(More)
BACKGROUND In observational research, propensity score techniques can be used to account for baseline differences between compared therapies. Although propensity scores are used increasingly often, their limitations in settings without complete data may not be recognized. OBJECTIVES We sought to evaluate the ability of propensity score matching to(More)
The authors developed and cross-validated prediction models for newly diagnosed cases of liver disorders by using logistic regression and neural networks. Computerized files of health care encounters from the Fallon Community Health Plan were used to identify 1,674 subjects who had had liver-related health services between July 1, 1992, and June 30, 1993. A(More)
Hydroxymethyl glutaryl coenzyme A inhibitors ("statins") are carcinogenic in rodents and an increased incidence of breast cancer was reported among pravastatin users in one randomised trial. We conducted a case-control study in the General Practice Research Database to evaluate the risk of breast cancer among 50- to 79-year old women treated with statins(More)
The frequency of hospital admission for perforated ulcer was not measurably affected by concurrent use of non-steroidal anti-inflammatory drugs (NSAID) during nearly 30 million person-days of NSAID use at Group Health Cooperative of Puget Sound. Whether patients had ever used cimetidine or antacids, drugs which indicate the presence of ulcer disease or(More)