PubMed Health⌕ Search

Biomedical subjects

D M Eddy

Publications and source records attributed to D M Eddy.

At least 73 records · Page 4Linked to original sources
PubMed1987

Screening for colorectal cancer in a high-risk population. Results of a mathematical model.

A mathematical model was used to estimate the cost-effectiveness of colorectal cancer screening strategies for people who are at high risk because of a first-degree relative with colorectal cancer. The model uses indirect evidence about such factors as cancer incidence, sensitivity and specificity of different tests, and treatment effectiveness. The analysis indicates that for screening people over 40 yr old an annual fecal occult blood test may reduce colorectal cancer mortality by about one-third, either colonoscopy or barium enema may reduce mortality by approximately 85%, a 3-5-yr frequency for endoscopies or barium enemas preserves 70%-90% of the effectiveness of an annual frequency, and beginning screening at age 50 reduces effectiveness by 5%-10%. Although both barium enemas and colonoscopies appear to be effective in reducing mortality, the lower cost of the barium enema makes it a more cost-effective strategy. All of these estimates depend on the baseline estimates of each of the factors incorporated in the model; the conclusions are most sensitive to assumptions about the natural history of adenomatous polyps, the bleeding of adenomas and presymptomatic cancers, and the sensitivity of the fecal occult blood test. Recommendations about colorectal cancer screening must also consider factors such as discomfort, inconvenience, and the availability of various technologies.

Colonic Neoplasms↗
PubMed1987

CAN*TROL: a computer model for designing national cancer control strategies.

Planning national and regional cancer control strategies is difficult. At present, cancer experts and planners rely on global subjective judgment to estimate the effectiveness and cost of different programs, and to set priorities. The complexity of the problem makes this approach vulnerable to oversimplification and error. Computer models can be very powerful aids to planning, enabling decision makers to break a problem into parts for which data exist, and reconstructing the parts to estimate the effect of a virtually limitless variety of cancer control activities, on a large number of important clinical and economic outcomes. CAN*TROL has been developed to serve this need. CAN*TROL has been used in the U.S., Chile and India, and is scheduled for use in several other countries.

Chile↗
PubMed1986

Setting priorities for cancer control programs.

This paper describes a simple method for comparing the effectiveness and costs of different cancer control activities and illustrates use of the method by evaluating priorities for controlling oral cancer in developing countries. The method estimates the long-term effect of prevention, screening, detection, treatment, and support activities (e.g., pain control) on morbidity, mortality, measures of quality of life, and cost for a specified population. It can be used to compare the cost effectiveness of various combinations of activities for one or more cancers and to help set priorities for cancer control programs. An analysis of two primary prevention activities, two screening activities, and three treatment activities to control oral cancer in Sri Lanka indicates that highest priority should be given to primary prevention activities such as anti-tobacco education and to screening.

Adult↗
PubMed1986

Secondary prevention of cancer: an overview.

Secondary prevention of cancer (screening) involves the use of tests to detect a cancer before the appearance of signs or symptoms. Before starting such a programme, the available evidence should be analysed to estimate the effectiveness of the proposed activities. Essential requirements are an understanding of the natural history of the particular cancer, availability of a test that can detect it, effective treatment for it, good evidence that early detection reduces the incidence and/or mortality, and that the expected benefits of screening outweigh the risks and costs. A screening programme should be limited to significant cancers and applied selectively, and should be integrated into the total health care programme. Programmes should take into account the risks, costs and expected benefits; provide quality assurance as well as facilities to follow, diagnose, and treat people with positive test results; maintain all records; and keep costs to a minimum. Ideally the effectiveness of screening should be demonstrated by randomized controlled trials showing a reduction in mortality, but this type of evidence exists for few cancers. Often an estimate of the effectiveness of screening must rest on other types of evidence, such as observations that the tests can detect the cancer before the appearance of signs or symptoms; that the tests can find a greater proportion of cancers in early stages; and that the patients with cancers detected through screening have higher survival rates after diagnosis and treatment although it must be recognized that these observations may be biased. This article discusses the available evidence on the effectiveness of screening for eight cancers, and gives estimates of the potential impact of secondary prevention for the year 2000.

Humans↗
PubMed1985

Screening for cancer in adults.

With careful selection of procedures, frequencies and risk groups, the periodic screening of adults for cancer can be a highly cost-effective medical activity. This paper summarizes the evidence of effectiveness of screening, the expected reduction in mortality, the risks and costs of screening, and the cost-effectiveness of screening strategies for cancer of the breast, colon, lung, cervix, endometrium and oral cavity. Some of the most pressing questions to be discussed concern the value of mammography, the frequency of the Pap smear, sigmoidoscopy, the value of chest X-rays and sputum cytology in the detection of lung cancer, and the current status of screening for cancers of the endometrium and oral cavity.

Adult↗
PubMed1985

Artificial heart and assist devices: directions, needs, costs, societal and ethical issues.

A Working Group appointed by the Director of the National Heart, Lung, and Blood Institute (NHBLI) has reviewed the current status of mechanical circulatory support systems (MCSS), and has examined the potential need for such devices, their cost, and certain societal and ethical issues related to their use. The media have reported the limited clinical investigative use of pneumatically energized total artificial hearts (which actually replace the patient's heart) and left ventricular assist devices (which support or replace the function of the left ventricle by pumping blood from the left heart to the aorta with the patient's heart in place). However, electrically energized systems, which will allow full implantation, permit relatively normal everyday activity, and involve battery exchange or recharge two or three times a day, are currently approaching long-term validation in animals prior to clinical testing. Such long-term left ventricular assist devices have been the primary goal of the NHLBI targeted artificial heart program. Although the ventricular assist device is regarded as an important step in the sequence of MCSS development, the Working Group believes that a fully implantable, long-term, total artificial heart will be a clinical necessity and recommends that the mission of the targeted program include the development of such systems. Past estimates of the potential usage of artificial hearts have been reviewed in the context of advances in medical care and in the prevention of cardiovascular disease. In addition, a retrospective analysis of needs was carried out within a defined population. The resulting projection of 17,000-35,000 cases annually, in patients below age 70, falls within the general range of earlier estimates, but is highly sensitive to many variables. In the absence of an actual base of data and experience with MCSS, projection of costs and prognoses was carried out using explicit sets of assumptions. The total cost of a left ventricular assist device, its implantation and maintenance for a projected average of 4 1/2 years of survival might be approximately $150,000 (in 1983 dollars). The gross annual cost to society could fall in the range of $2.5-$5 billion. Ethical issues associated with use of the artificial heart are not unique. For individual patients these relate primarily to risk-benefit, informed consent, patient selection, and privacy. However, for society as a whole, the larger concern relates to the distribution of national resources.(ABSTRACT TRUNCATED AT 400 WORDS)

Assisted Circulation↗
PubMed1985

The yield of the routine admission electrocardiogram. A study in a general medical service.

To evaluate the usefulness of the admission electrocardiogram, we prospectively studied 1410 patients admitted to a general medical service. The patients' history and physical examination were analyzed, and then the electrocardiogram was evaluated to determine whether it added new information. Fifty-two admission electrocardiograms added unique information; 31 established a new diagnosis, 28 of which changed management. Twenty-one electrocardiograms suggested a new diagnosis, 13 of which proved correct. Only age and a clinically evident cardiac abnormality predicted the yield of the admission electrocardiogram. Among patients with cardiac abnormalities, electrocardiogram yield was 8.5% in those 45 years or older and 2.5% in those younger. Among patients without cardiac abnormality, electrocardiogram yield was 1.0% overall and its cost-effectiveness was $24 000 per year of life saved. An admission electrocardiogram infrequently added new information to the clinical evaluation, but was useful when it did. Admission electrocardiograms are as cost-effective as many accepted medical practices.

Adult↗
PubMed1983

Finding cancer in asymptomatic people. Estimating the benefits, costs and risks.

There are no absolute criteria for determining the effect of early detection on cancer mortality, and evidence from many sources must be examined. The best evidence is provided by randomized controlled trials that use mortality as an endpoint (RCTMs), but this type of evidence is available for only two cancers, i.e., of the breast and colon. The decision to recommend a test for a particular person must take into account the strength of the evidence that early detection reduces mortality, the expected amount of the reduction, the chance the person will ever get the cancer, and the anticipated costs and risks. This paper describes the evidence that early detection of various cancers reduces mortality, and estimates the expected reduction in mortality to be achieved by various tests. Three specific examples are discussed: breast self-examination, mammography in women younger than age 50 years, and chest x-rays for male smokers over age 40.

Adult↗
PubMed1983

A mathematical model for timing repeated medical tests.

This paper presents a mathematical model that can be used to estimate the clinical and economic outcomes of monitoring patients with periodic examinations. The model can compare the consequences of monitoring for different disorders, with different tests, at different frequencies. The paper describes formulas that incorporate information about the incidence and natural history of disorders, the effectiveness of tests, the effectiveness of treatment, and the order and frequency of monitoring, to calculate the probability of detecting a disorder, the method and timing of detection, the earliness (e.g., stage) with which a disorder is detected, and the clinical and economic outcomes. The application of the model is described through a hypothetical example. The model has been used to analyze several cancer screening problems involving multiple disorders and multiple tests.

Cost-Benefit Analysis↗
PubMed1982

The art of diagnosis: solving the clinicopathological exercise.

We analyzed the psychological process by which physicians solve complicated diagnostic problems, such as those posed in clinicopathological exercises. The challenge of differential diagnosis is to select the most probable cause of a patient's condition, yet the size of the problem, the nature of medical information, and the notorious inability of human beings to manipulate probabilities in their heads all conspire against the diagnostician to make it virtually impossible to employ Bayes' theorem in routine diagnosis. Unable to estimate the desired probabilities explicitly, physicians recast the problem into a form that uses one of their most effective mental skills--that of comparing patterns. A study of 50 clinicopathological conferences published in the Journal suggests that the following six steps are taken to arrive at a diagnosis: aggregation of groups of findings into patterns, selection of a "pivot" or key finding, generation of a cause list, pruning of the cause list, selection of a diagnosis, and validation of the diagnosis. Although the clinicopathological conference differs in some important ways from real-life diagnostic problems, we believe that the principles described here closely resemble those used in practice. Properly selected clinicopathological conferences are excellent windows through which to study diagnostic reasoning.

Diagnosis↗