John Snow: the first hired gun?

The 1854 English cholera outbreak led to reform of Victorian public health legislation, including the Nuisances Removal and Diseases Prevention Act. The reforms threatened the closure of many factories whose fumes were considered hazardous to the public's health. The second witness to appear before the Parliamentary committee considering the reforms was Dr. John Snow. Snow testified on behalf of the manufacturers threatened by the reforms. He stated that the fumes from such establishments were not hazardous. He contended that the workers in these factories did not become ill as a result of their exposures, and therefore these fumes could not be a hazard to the general public's health. Snow also presented data from the 1854 cholera outbreak as the basis for his belief that epidemic diseases were transmitted by water, not air. Although the data concerned cholera, Snow extended the inference to all epidemic diseases. When the committee's report was published, The Lancet chastised Snow in a stinging editorial. Parliament subsequently revised the bill in favor of the manufacturers and passed it into law. The implications of this particular episode in the history of epidemiology are discussed.

Decreasing mortality from pulmonary embolism in the United States, 1979-1996.

BACKGROUND: Mortality trends and patterns for pulmonary embolism may yield clues to its aetiology. Previous investigations had identified several mortality contrasts in pulmonary embolism mortality among US residents. These findings had been made in the context of a trend of increasing rates during 1962-1984. METHODS: Annual age-specific and age-adjusted pulmonary embolism mortality rates for US residents during 1979-1996 were compiled from the US National Center for Health Statistics web site. These data were analysed for mortality contrasts and trends. RESULTS: For all racial-gender groups, age-adjusted mortality declined throughout the period. The greatest rate of decline was found among black men, followed by (in decreasing order) black women, white men, other men, white women, and other women. In 1996, the previously observed demographic contrasts of blacks experiencing the highest pulmonary embolism mortality, followed by whites and then others, and the male rate being higher than the female one were still present despite this decline. Age-specific mortality from pulmonary embolism in the US during 1996 also mirrored that reported for the 1970s, with mortality increasing during the life span (the risk of death doubling with each decade of life). Specifically, the annual age-adjusted pulmonary embolism mortality rate in 1996 for white men was 2.4 per 100 000 persons; white women, 2.3 per 100 000 persons; black men, 6.0 per 100 000 persons; black women, 4.8 per 100 000 persons; non-black non-white men, 1.0 per 100 000 persons; and non-black non-white women, 0.7 per 100 000 persons. CONCLUSIONS: Mortality from pulmonary embolism in the US declined significantly during 1979-1996. Several demographic contrasts, particularly an excess among men, continue to exist.

Mortality from primary pulmonary hypertension in the United States, 1979-1996.

STUDY OBJECTIVES: To determine whether primary pulmonary hypertension mortality in the United States increased since 1979 coincident with the introduction of anorexigens. DESIGN: Examination of annual age-adjusted and age-specific primary pulmonary hypertension mortality in the United States from 1979 through 1996 and in five selected states from 1992 through 1996. SETTING: The United States, from 1979 through 1996. PATIENTS OR PARTICIPANTS: Residents of the United States, from 1979 through 1996. INTERVENTIONS: None. MEASUREMENTS AND RESULTS: Annual age-adjusted mortality increased at different rates among white men and women and black men and women. The greatest increase was among black women (who also had the highest rates). Age-specific mortality showed a high rate among infants < 1 year old, a low rate in childhood, and an ascending rate throughout the remainder of life. Similar patterns were identified at the state level. CONCLUSIONS: Primary pulmonary hypertension mortality in the United States has increased notably since 1979. Some portion of this increase may be related to the introduction of anorexigens. Improvements in diagnostic recognition may also explain part of the increase in mortality. These results need to be confirmed in a diagnosis validation study, particularly because the same mortality data suggest that the disease may be more common in the elderly than has been previously reported.

Sensitivity and specificity of a screening questionnaire for dry eye.

We developed a Dry Eye Screening Questionnaire for the Dry Eye Epidemiology Projects (DEEP), a proposed large epidemiologic study. All persons who screen positive and a small sample of those who screen negative are to be invited for a diagnostic examination. Containing 19 questions, of which only 14 were used in the analysis, the questionnaire takes only a few minutes to administer on the telephone. To construct a discriminator function and thus a ROC curve, we used stepwise multiple regression on screening responses from a clinic series of 77 cases and 79 controls. Stepwise regression may incorporate into the predictor equation variables whose relation to the predicted is only accidental. Further, misclassification rates are underestimated by the resubstitution method, in which the proportion misclassified is obtained from the same dataset in which the discriminator function was fitted. To counter these problems, we randomly divided the data in half. We chose as predictors only those variables (Dry and Irritated) selected by stepwise regression in both data halves. We estimated unbiased misclassification rates using the unbiased test set method, in which the discriminator is fitted in one data half, and misclassification rates are calculated in the other half. Comparison of ROC curves arising from resubstitution and test set estimates indicates that resubstitution bias in misclassification rate estimation is negligible in our data. A resubstitution estimate made on the entire data is thus preferred. The resulting sensitivity/specificity values are reasonably high (e.g., 60%/94%), suggesting that the questionnaire will be a useful screening tool in the DEEP study. A second discriminator using the sum of all 14 responses is similar in its misclassification characteristics to the first discriminator. A second potentially significant error, arising from applying results from a clinical series to a general population, will be investigated as survey results in DEEP become available.

'Best clinical practice': assessment of processes of care and of outcomes in the US Military Health Services System.

The National Quality Management Program of the Military Health Services System of the United States has undertaken a series of projects whose objective is the active, on-going monitoring and improvement of the effectiveness and efficiency of the care provided to a broad population that encompasses troops on active duty, retirees and dependents. The analytic activities consist of (1) identification by clinical panels of conditions and procedures of interest; (2) collection of data from electronic repositories and from charts to characterize the patients, how they are managed, the clinical outcomes they experience, the resource costs their care entails, and, from questionnaires, their functional status and level of satisfaction, and (3) generation of 'report cards' that inform organizational units down to the level of the hospital of the characteristics of their patients, their practices, and the risk-adjusted outcomes they achieve. The patterns of care employed by the hospitals that obtain the best risk-adjusted outcomes and resource utilization ('best clinical practice') are identified and made known. In addition, (4) a systematic process of developing outcomes-based practice guidelines has been devised. It intent is to serve as a decision-support tool for clinicians. Initial estimates have been obtained of the probable consequences of the application of this tool to operative interventions in childbirth. Use of the tool would result in a higher occurrence of elective Caesarean sections, a reduced rate of emergency Caesarean sections and much lower use of forceps, with an overall improvement in outcomes and lower resource costs. This program is currently in the early phases of implementation. The two principal requirements for the immediate future are (1) education of the clinical and administrative communities in the use of the data and the decision-support tools and (2) evaluation of the consequences of the use of the data by the clinical and administrative communities.

Disposal-related sharps injuries at a New York City Teaching Hospital.

OBJECTIVE: To characterize disposal-related sharps injuries. DESIGN: A three-part study including (a) descriptive analysis of disposal-related injuries in a 1-year period, (b) 4:1 matched case-control study of nurses injured while using sharps disposal containers, and (c) survey to solicit opinions of users of containers. SETTING: An 1,181-bed teaching hospital in New York City. PARTICIPANTS: For epidemiologic analyses, persons with self-reported injuries identified via New York State and Occupational Safety and Health Administration forms and control nurses without self-reported injuries. For survey, convenience sample of hospital nurses, laboratory workers, and maintenance workers. MAIN OUTCOME MEASURES: Circumstances of injuries determined by study questionnaires. Employee opinions obtained by questionnaires and discussions during small group sessions. RESULTS: Three hundred sixty-one persons reported sharps injuries, of whom 72 (20%) had disposal-related injuries. Persons with disposal-related injuries included four hospital visitors and one patient. Of 67 disposal-related injuries among employees, 25 (37%) directly involved use of a sharps disposal container. Significant risk factors for injury included container height greater than 4 ft above the floor, distance less than 5 ft from site of sharp object use to nearest container, and lack of attendance at universal precautions inservice classes. Survey groups involved 69 employees who identified a variety of preferred features for sharps disposal containers. CONCLUSIONS: Disposal of sharp objects is an important cause of sharps injuries. Ergonomic factors, worker education, and appropriate container design should be considered in injury prevention strategies. Relevant guidelines and regulations are lacking and are needed.

Projected neurodegenerative disease mortality among minorities in the United States, 1990-2040.

Between 1992 and 2040, the United States nonwhite elderly population is expected to grow from 3.3 to 14.1 million. In order to assess the implications of this increase on the mortality from neurodegenerative diseases in the United States, we used Census Bureau population estimates to formulate projections of the annual number of deaths from neurodegenerative diseases and from six comparison conditions (liver cirrhosis, colon cancer, lung cancer, cancer of the female breast, multiple sclerosis, and malignant melanoma), assuming that the United States disease-age-gender-specific death rates for 1985-1988 remain constant between 1990 and 2040. We find that neurodegenerative disease mortality increases by 281-524%, depending on the model of population growth used. For the 'middle' population growth model, the increase in annual neurodegenerative disease mortality is 373%. The major component of this increase is the rise in deaths attributed to dementia. For the six comparison diseases, the increases in mortality range from 130 (multiple sclerosis) to 288% (colon cancer). Given the current level of underascertainment of neurodegenerative disease mortality, particularly among minorities, and the conservative nature of the Census Bureau estimates of future population, it is likely that these projections are under-estimates. The implications of these data are discussed.

Prevalence of aortic aneurysms in the Twin Cities metropolitan area, 1979-84.

The discharge summaries for Minneapolis-St. Paul metropolitan area residents hospitalized during 1979-84 were reviewed for diagnoses of aortic aneurysms. Annual age-specific and age-adjusted sex-specific hospital discharge diagnosis rates were calculated for all aortic aneurysms, dissecting aortic aneurysms, thoracic aortic aneurysms (nondissecting), and abdominal aortic aneurysms (nondissecting). For each aortic aneurysm type, hospital discharge diagnosis rates were found to increase with age for both men and women. Abdominal aortic aneurysms were the most common type reported (age-adjusted annual rates for men varied between 40.6 and 49.3 per 100,000 population; for women, between 6.8 and 12.0 per 100,000 population). Men were noted to have higher rates for each aneurysm type. An increasing temporal trend was observed for all aortic aneurysms and abdominal aortic aneurysms among men. These findings are reviewed in light of recent data on mortality from aortic aneurysms in the United States.

Projected neurodegenerative disease mortality in the United States, 1990-2040.

Between 1990 and 2040, the United States elderly population is expected to grow from 31.6 to 68.1 million. In order to assess the implications of this increase on the mortality from neurodegenerative diseases in the United States, we used Census Bureau population estimates to formulate projections of the annual number of deaths from neurodegenerative diseases and from six comparison conditions (liver cirrhosis, colon cancer, lung cancer, cancer of the female breast, multiple sclerosis, and malignant melanoma), assuming that the United States disease-age-gender-race-specific death rates for 1985-1988 remain constant between 1990 and 2040. We find that neurodegenerative disease mortality increases by 119-231%, depending on the model of population growth used. For the 'middle' population growth model, the increase in annual neurodegenerative disease mortality is 166%. The major component of this increase is the rise in deaths attributed to dementia. For the six comparison diseases, the increases in mortality range from 52 (multiple sclerosis) to 130% (colon cancer). Given the current level of under ascertainment of neurodegenerative disease mortality and the conservative nature of the Census Bureau estimates of future population, it is likely that these projections are underestimates. The implications of these data are discussed.

Geographic distribution of pulmonary embolism mortality rates in the United States, 1980 to 1984.

The identification of individuals who are at high risk for developing PE has focused on clinically identified groups. Epidemiologic characterization of high-risk populations has been less successful. To provide a demographic basis for further epidemiologic inquiry into groups at high risk for PE, we investigated the geographic distribution of PE mortality rates in the United States from 1980 to 1984. We found that for most sections of the United States, PE mortality rate patterns mirrored those that were observed previously for the nation. Men had greater PE mortality rates than women in most regions, and nonwhites had greater PE mortality rates than did whites. The exception to this pattern was the Pacific region, where the PE mortality rate was lower than rates in other parts of the country, particularly among nonwhites. This pattern probably reflects the different racial compositions in the regional populations of the United States, with corresponding PE risk factor differences. However, the specific risk factors that are responsible for these patterns have not been identified. Further inquiry into these geographic patterns may provide a means for the prevention of high PE morbidity and mortality rates.

Childhood asthma and passive smoking. Urinary cotinine as a biomarker of exposure.

To assess the relationship between passive smoking and asthma, we investigated (1) whether passive smoking was more prevalent among asthmatic than control children and (2) whether exposure to tobacco smoke was higher in acute asthma than in nonacute asthma. Three groups were recruited into a case-control study: 72 acute asthmatic children from the emergency room (ER), 35 nonacute asthmatic children from the asthma clinic, and 121 control children from the ER. Both questionnaire and urinary cotinine/creatinine ratio (CCR) were used to assess passive smoking. Levels of CCR greater than or equal to 30 ng/mg were used to identify children exposed at home. Mean CCR was also computed. Acute and nonacute asthmatic children had similar prevalences of passive smoking at home. Acute cases showed a higher mean CCR than nonacute cases, but this was not significant. In comparing all asthmatic to control children, smoking by the maternal caregiver was more prevalent among asthmatic children (odds ratio OR = 2.0, 95% CI 1.1, 3.4). This was confirmed by CCR greater than or equal to 30 ng/mg (OR = 1.9, 95% CI 1.04, 3.35) and by the difference in mean CCR (43.6 versus 25.8 ng/mg, p = 0.06). We conclude that smoking by the maternal caregiver is associated with clinically significant asthma in children. We could not show that it is a trigger of acute asthma attacks.

No association between serum cholesterol and growth in preadolescent American children.

To determine the relation between serum cholesterol level (SCL) and growth in preadolescent children (2-12 years of age) in the United States, we examined the association of SCL and three measures of growth in a national sample of American children from 1971 to 1974. We investigated three parameters: sitting height, standing height, and weight and undertook separate analyses for sex and race. We found no consistent statistically significant associations with any of these three measures of growth. This result did not change when we adjusted the data for age alone or for age, poverty index, serum protein level, and hematocrit. We conclude that growth in this population group is not dependent on SCL.

Morbidity from congestive and hypertrophic cardiomyopathy in the Minneapolis-St. Paul metropolitan area: 1979-1984.

In order to develop an epidemiologic profile of congestive and hypertrophic cardiomyopathy, we reviewed hospital discharge diagnosis data for the Minneapolis-St. Paul metropolitan area from 1979 to 1984. For congestive cardiomyopathy, we found statistically significant increases in the age-adjusted hospital discharge diagnosis rate for both men (from 63 to 125/100,000 population/year) and women (from 35 to 58/100,000 population/year). We also found that the age-specific rates increased significantly among older men (55-64 years of age, from 86 to 198/100,000 population/year; 65-74 years of age, from 182 to 322/100,000 population/year) and older women (65-74 years of age, from 106 to 196/100,000 population/year). The rates increased with advancing age for both genders. Men had greater rates than women. For hypertrophic cardiomyopathy, no temporal trends could be discerned. Women were observed to have greater age-specific rates than men; the age-adjusted rates were, however, approximately the same for men and women. The hospital discharge diagnosis rate increased with advancing age among both men and women. The possible reasons for these observed patterns, including the major impact of improvements in diagnosis during the observation period, are discussed. Additional research into these disorders is needed.

Asbestos-associated pleural mesothelioma in school teachers: a discussion of four cases.

The causal relationship between malignant mesothelioma and exposure to asbestos is well established. In part as a result of that association, much public attention has focused on asbestos abatement in buildings, such as public schools, in which that asbestos was used as a construction material. The present communication is a report of four cases of malignant mesothelioma in school teachers whose only apparent exposure to asbestos was in the schools in which they taught. The concerns raised by this report are also discussed.

The silence: the asbestos industry and early occupational cancer research--a case study.

To gain insight into corporate activities regarding the identification of occupational carcinogens earlier in this century, the actions of one industry, the asbestos industry, were reviewed. This industry, in concert with many of its insurers, systematically developed and then suppressed information on the carcinogenicity of asbestos. The development of warnings for those exposed to the asbestos was delayed. As a result, millions of workers were exposed to the carcinogen and hundreds of thousands died. These events are placed into the context of similar activities in other industries during this time.

Parkinson's and motoneuron disease morbidity in the Twin Cities metropolitan area: 1979-1984.

To better characterize the morbidity from Parkinson's disease and motoneuron disease, we examined hospital discharge diagnosis data for all acute care facilities (except for the Veterans Administration Medical Center) in the Minneapolis-St. Paul area in each year from 1979 to 1984 for persons aged 30-74 years. For each person in whom the discharge diagnosis included either Parkinson's disease or motoneuron disease, the age, gender, and year of admission were recorded. Annual age-gender-specific and age-adjusted gender-specific hospital discharge diagnosis rates were calculated. For Parkinson's disease, for both, men and women, the age-adjusted hospital discharge diagnosis rate declined by 26-27%. Component age-specific data for Parkinson's disease showed more variable temporal changes. For motoneuron disease, no temporal changes were observed, although the number of observations was small. These data suggest that Parkinson's disease is declining in prevalence in the hospitalized population. The implications for motoneuron disease are less clear. Further studies in this area are needed.

Mortality from pulmonary embolism in the United States: 1962 to 1984.

To examine the effect of advances in the prevention of and therapy for PE, we reviewed mortality for PE in the United States from 1962 to 1984. Age-adjusted PE mortality increased by 67 to 100 percent between 1962 and 1974 for white and non-white men and women. From 1975 to 1984, these rates declined by 20 to 28 percent. Non-white PE mortality was greater than white PE mortality; men had a greater risk of PE death than women. Age-specific patterns (more than 40 years of age) of PE mortality followed those of the age-adjusted death rates, with increases noted in all groups between 1962 and 1974 and declines during the 1975-1984 period. These patterns might reflect improved ascertainment of cases and better prevention of disease. The magnitude of the rates suggests that the list of indications for prophylactic anticoagulation should be re-examined for possible expansion.