David M. Eddy

Screening for Colorectal Cancer

Considerable indirect evidence, based on the natural history of colorectal cancer and the ability of tests to detect adenomas and invasive cancers, suggests that screening for colorectal cancer reduces mortality. Without screening, a 50-year-old person at average risk has approximately a 530-in-10,000 chance of developing invasive colorectal cancer in the rest of his or her life and approximately a 250-in-10,000 chance of dying from it. Analysis of indirect evidence with a mathematic model indicates that screening persons for 25 years, from the age of 50 to the age of 75 years should reduce the chance of developing or dying from colorectal cancer by 10% to 75%, depending on which screening tests are used and how often screening is done. Screening for colorectal cancer is optional. A possible recommendation is that annual fecal occult blood tests and 65-cm flexible sigmoidoscopy every 3 to 5 years be done for average-risk men and women who are between 50 and 75 years of age. In addition to having annual fecal occult blood tests, persons with first-degree relatives with colorectal cancer can be offered barium enemas instead of sigmoidoscopies every 3 to 5 years.

Screening for Cervical Cancer

Indirect evidence indicates that cervical cancer screening should reduce the incidence and mortality of invasive cervical cancer by about 90%. In the absence of screening, a 20-year-old average-risk woman has about a 250 in 10,000 chance of developing invasive cervical cancer during the rest of her life, and about a 118 in 10,000 chance of dying from it. Screening at least every 3 years from 20 to 75 years of age will decrease these probabilities by about 215 in 10,000 and 107 in 10,000, respectively, and will increase a 20-year-old womans life expectancy by about 96 days. The particular age at which screening is begun (for example, 17 or 20 years), the requirement of several initial annual examinations before reducing the frequency, and screening every 1 or 2 years compared with every 3 years improves the effectiveness by less than 5%. Screening is recommended at least every 3 years from about age 20 to about age 65 years.

Model Transparency and Validation A Report of the ISPOR-SMDM Modeling Good Research Practices Task Force–7

Trust and confidence are critical to the success of health care models. There are two main methods for achieving this: transparency (people can see how the model is built) and validation (how well it reproduces reality). This report describes recommendations for achieving transparency and validation, developed by a task force appointed by the International Society for Pharmacoeconomics and Outcomes Research (ISPOR) and the Society for Medical Decision Making (SMDM). Recommendations were developed iteratively by the authors. A nontechnical description should be made available to anyone—including model type and intended applications; funding sources; structure; inputs, outputs, other components that determine function, and their relationships; data sources; validation methods and results; and limitations. Technical documentation, written in sufficient detail to enable a reader with necessary expertise to evaluate the model and potentially reproduce it, should be made available openly or under agreements that protect intellectual property, at the discretion of the modelers. Validation involves face validity (wherein experts evaluate model structure, data sources, assumptions, and results), verification or internal validity (check accuracy of coding), cross validity (comparison of results with other models analyzing same problem), external validity (comparing model results to real-world results), and predictive validity (comparing model results with prospectively observed events). The last two are the strongest form of validation. Each section of this paper contains a number of recommendations that were iterated among the authors, as well as the wider modeling task force jointly set up by the International Society for Pharmacoeconomics and Outcomes Research and the Society for Medical Decision Making.

Screening for breast cancer

There is very good evidence that screening for breast cancer reduces mortality in women older than 50 years and suggestive but inconsistent evidence that screening is effective in reducing long-term mortality in women younger than 50 years. The probability that an average-risk woman will be diagnosed with breast cancer in the coming 10 years is about 130 in 10,000 for a 40-year-old woman, 230 in 10,000 for a 55-year-old woman, and 280 in 10,000 for a 65-year-old woman. The chance of dying from breast cancer diagnosed in the coming 10 years is about 90 in 10,000, 123 in 10,000, and 120 in 10,000 for women age 40, 55, and 65, respectively. Mathematical models based on data from controlled trials of screening programs indicate that screening annually for 10 years with breast physical examination will decrease the probability of death from breast cancer by about 25 in 10,000 for women in the three age groups and increase life expectancy by about 20 days. Adding annual mammography will decrease the probability of death from breast cancer an additional 25 in 10,000 and increase life expectancy an additional 20 days. The actual reductions in mortality observed in controlled trials are slightly lower. If women are screened annually for 10 years with breast physical examination and mammography, the chance for a false-positive result over the 10-year period is approximately 2500 in 10,000. On the population level, if 25% of women age 40 to 75 are screened annually with both examinations, deaths from breast cancer would be decreased by about 4000 in the year 2000. Net annual costs would be approximately

Screening for Osteoporosis

1.3 billion. Recommending a screening strategy requires weighing the benefits against the risks and costs.

Screening for Lung Cancer

PURPOSE To review evidence that screening for osteoporosis by measuring bone mass in postmenopausal women would reduce fracture incidence. DATA IDENTIFICATION An English-language literature search using MEDLINE (1966 to 1989), bibliographic reviews of book chapters and review articles, technology assessments of bone mass measurement, and other publications. STUDY SELECTION We summarize prospective studies of fracture risk prediction done with widely used bone mass measurement techniques, and we document noncontroversial or peripheral points with recent papers and reviews. DATA EXTRACTION Without osteoporosis screening trials, no quantitative analysis is possible. Instead, we assess the ability of screening tests to measure bone mass and define fracture risk categories, the ability of risk categories to determine treatment, and the ability of treatment to reduce fracture incidence. RESULTS OF DATA SYNTHESIS Bone mass measurement meets many of the criteria for a screening test, and indirect evidence suggests that a screening program might reduce osteoporosis-related fracture incidence. No trial has shown this directly; however, and questions remain about overall benefits and costs of mass screening. CONCLUSIONS Although there are clinical indications for bone mass measurement, unselective screening for osteoporosis cannot be recommended until a specific program is formulated and justified.

Diabetes Risk Calculator: A simple tool for detecting undiagnosed diabetes and pre-diabetes

Lung cancer is the commonest cause of death from cancer in both men and women, with approximately 152,000 new cases and 139,000 deaths in 1988. The incidence and mortality rates are increasing rapidly in women. Two main tests have been used to screen for lung cancer: chest roentgenography and sputum cytology. Four recent controlled trials and one case-control study failed, however, to show that screening reduces lung cancer mortality even in high-risk persons (smokers). In the Mayo Lung Project, for example, the lung cancer death rate in high-risk men offered sputum cytology and chest roentgenogram every 4 months was 3.1 per 1000 person-years, compared with 3.0 per 1000 person-years in a control group. Chest roentgenograms and sputum cytology lead to false-positive test results in smokers of approximately 5% and 0.5%, respectively. Because of the lack of evidence of benefit and because of its potential harms and costs, screening for lung cancer is not recommended.

Synthesis of Environmental Evidence: Nitrogen Dioxide Epidemiology Studies

OBJECTIVE—The objective of this study was to develop a simple tool for the U.S. population to calculate the probability that an individual has either undiagnosed diabetes or pre-diabetes. RESEARCH DESIGN AND METHODS—We used data from the Third National Health and Nutrition Examination Survey (NHANES) and two methods (logistic regression and classification tree analysis) to build two models. We selected the classification tree model on the basis of its equivalent accuracy but greater ease of use. RESULTS—The resulting tool, called the Diabetes Risk Calculator, includes questions on age, waist circumference, gestational diabetes, height, race/ethnicity, hypertension, family history, and exercise. Each terminal node specifies an individuals probability of pre-diabetes or of undiagnosed diabetes. Terminal nodes can also be used categorically to designate an individual as having a high risk for 1) undiagnosed diabetes or pre-diabetes, 2) pre-diabetes, or 3) neither undiagnosed diabetes or pre-diabetes. With these classifications, the sensitivity, specificity, positive and negative predictive values, and receiver operating characteristic area for detecting undiagnosed diabetes are 88%, 75%, 14%, 99.3%, and 0.85, respectively. For pre-diabetes or undiagnosed diabetes, the results are 75%, 65%, 49%, 85%, and 0.75, respectively. We validated the tool using v-fold cross-validation and performed an independent validation against NHANES 1999–2004 data. CONCLUSIONS—The Diabetes Risk Calculator is the only currently available noninvasive screening tool designed and validated to detect both pre-diabetes and undiagnosed diabetes in the U.S. population.

Model transparency and validation: a report of the ISPOR-SMDM Modeling Good Research Practices Task Force--7

The use of meta-analysis is becoming more common in the medical literature, but it is not common in the environmental literature. Although meta-analysis cannot combine a group of poorly executed, conflicting studies to get an unequivocal answer, there are certain situations where it can be helpful. The inability of studies to produce similar results may be a function of the power of the studies rather than a reflection of their quality. The literature on the effects of nitrogen dioxide on the odds of respiratory illness in children is such an example. Three quantitative methods for the synthesis of this evidence are presented. Although the methods produce slightly different results, the conclusion from all three methods is that the increase in the odds of respiratory illness in children exposed to a long-term increase of 30 micrograms/m3 (comparable to the increase resulting from exposure to a gas stove) is about 20 percent. This estimated increase is not sensitive to the method of analysis.

Individualized Guidelines: The Potential for Increasing Quality and Reducing Costs

Trust and confidence are critical to the success of health care models. There are two main methods for achieving this: transparency (people can see how the model is built) and validation (how well the model reproduces reality). This report describes recommendations for achieving transparency and validation developed by a taskforce appointed by the International Society for Pharmacoeconomics and Outcomes Research and the Society for Medical Decision Making. Recommendations were developed iteratively by the authors. A nontechnical description--including model type, intended applications, funding sources, structure, intended uses, inputs, outputs, other components that determine function, and their relationships, data sources, validation methods, results, and limitations--should be made available to anyone. Technical documentation, written in sufficient detail to enable a reader with necessary expertise to evaluate the model and potentially reproduce it, should be made available openly or under agreements that protect intellectual property, at the discretion of the modelers. Validation involves face validity (wherein experts evaluate model structure, data sources, assumptions, and results), verification or internal validity (check accuracy of coding), cross validity (comparison of results with other models analyzing the same problem), external validity (comparing model results with real-world results), and predictive validity (comparing model results with prospectively observed events). The last two are the strongest form of validation. Each section of this article contains a number of recommendations that were iterated among the authors, as well as among the wider modeling taskforce, jointly set up by the International Society for Pharmacoeconomics and Outcomes Research and the Society for Medical Decision Making.