A general overview of Mantel-Haenszel methods: applications and recent developments.
Explore the source record for details and available documents.
Clinical Trials as Topic↗
Biomedical subjects
J R Landis
Publications and source records attributed to J R Landis.
A generalized Mantel-Haenszel analysis of the regression of blood pressure on blood lead using NHANES II data.
This paper proposes an alternative analysis of the statistically significant blood pressure/blood lead relationship reported for males, ages 12 to 74, based on data from the second National Health and Nutrition Examination Survey. Because of the substantial decline, both for blood lead levels and blood pressures, during the 4-year survey period, there is considerable interest in the extent to which this association can be attributed to concurrent secular trends. The statistical methods illustrate the use of a randomization model-based approach to testing the statistical significance of the partial correlation between blood lead level and diastolic blood pressure, adjusting for age, body mass index, and the 64 sampling sites. The resulting analyses confirm that the significant linear association between blood lead levels and diastolic blood pressures cannot be dismissed as a spurious association due to concurrent secular trends in the two variables across the 4-year survey period. In a conservative approach to this investigation, a randomization model-based test statistic, using the actual level of the natural log of blood lead and diastolic blood pressure, remained statistically significant at the 5% level, even when averaging the association across 478 subgroups formed by the cross-classification of age, body mass index, and the 64 sampling sites.
Adolescent↗
Clinical characteristics and survival experience of out-of-hospital cardiac arrest victims without coronary heart disease.
Of 270 patients successfully resuscitated from out-of-hospital cardiac arrest, 16% had no evidence of coronary heart disease. In these 43 patients, other forms of heart disease were found in 81% (35/43): cardiomyopathy in 18 patients, valvular disease in six, congenital heart disease in two, and primary arrhythmia in nine. Seven patients had evidence of only pulmonary disease and one had pancreatitis as his precipitating event. Nineteen of the 43 patients (44%) had serum potassium less than 3.6 mEq l-1 in the initial blood sample after cardiac arrest. One- and two-year mortalities were 30% and 43%, respectively, for the group, which is similar to one-year (20%) and two-year (35%) mortalities of the 227 resuscitated patients with coronary heart disease. Patients who survive a sudden death experience and who have no evidence of coronary artery disease are a unique but heterogeneous group who usually have identifiable cardiac or pulmonary disease.
Cardiac Complexes, Premature↗
Education for clinical research: an experiment at the University of Michigan.
Explore the source record for details and available documents.
Curriculum↗
Attributable risk estimation from matched case-control data.
A methodology is proposed for obtaining summary estimators, variances, and confidence intervals for attributable risk measures from data obtained through a case-control study design where one or more controls have been matched to each case. The sampling design for obtaining these data is conceptualized as a simple random sample of cases being equivalent to a simple random sample of matched sets. By combining information across the strata determined by the matched sets, this approach provides all of the benefits associated with the Mantel-Haenszel procedure for the estimators of attributable risk among the exposed and population attributable risk. Asymptotic variances are derived under the assumption that the frequencies of the unique response patterns follow the multinomial distribution. Simulation results indicate that these methods fare very well with respect to bias and coverage probability.
Analysis of Variance↗
Attributable risk ratio estimation from matched-pairs case-control data.
Explicit formulas are provided for estimating the attributable risk ratio among the exposed and the entire target population utilizing matched-pairs data. Large-sample standard errors and corresponding confidence intervals are provided. These estimates can be obtained from the cross-classification frequencies of matched pairs by disease and exposure status in the usual 2 X 2 table. The key to the development of these formulas lies in recognizing that attributable risk among the exposed is a direct function of the odds ratio, and population attributable risk is a direct function of the odds ratio and exposure prevalence among only the cases (assuming a rare disease). The formulas presented in this paper require only a calculator for computation. The methodology is illustrated with data from a matched-pairs case-control study of oral conjugated estrogens and endometrial cancer.
Contraceptives, Oral↗
Re: "Exposure opportunity in case-control studies".
Explore the source record for details and available documents.
Epidemiologic Methods↗
Factors associated with glucose tolerance in adults in the United States.
Glucose tolerance data from the second National Health and Nutrition Examination Survey were analyzed to determine factors predicting fasting plasma glucose levels and glucose tolerance in a representative US population. Central adiposity (subscapular skinfolds), age, and family history of diabetes were the major predictors of fasting levels and of glucose tolerance. For women, having a diabetic mother was significantly related to fasting glucose and glucose tolerance, but for men the proportion of siblings with diabetes was positively related. Multiple regression analyses, after adjustment for age and subscapular skinfolds, identified white cell count, systolic blood pressure, natural logarithm total iron binding capacity, and family history variables as being significant predictors of glucose tolerance in both sexes. In addition, several other variables were predictive for men or women, but not for both. These data confirm the importance of the major predictors of glucose tolerance and suggest provocative new associations in the general population.
Adult↗
Strategies for the analysis of imputed data from a sample survey. The National Medical Care Utilization and Expenditure Survey.
Missing data in sample surveys is virtually unavoidable, whether it is an entire unit that is missing or only an item for a responding unit. Compensation for unit nonresponse is usually made through the assignments of weights to responding units; for item nonresponse, the compensation often is by an imputation procedure. This paper reviews the extent of missing data in a large federal survey, the National Medical Care Utilization and Expenditure Survey, and the imputation procedures used to compensate for item missing data. The effects of imputation on several types of estimates from the survey are examined. In addition, several methods for analyzing survey data with imputed values are reviewed, and recommendations about preferred strategies are made for selected circumstances.
Data Collection↗
Analysis of cardiac symptoms preceding cardiac arrest.
Prodromal symptoms and cardiac history were examined in 227 patients with coronary artery disease who were successfully resuscitated after out-of-hospital cardiac arrest. Cardiac arrest was sudden--with either no symptoms or symptoms for less than 1 hour--in 71% of the patients. Nonsudden death--death occurring after more than 1 hour of symptoms--occurred in 29% of the patients. A history of cardiovascular disease was present in 85% of patients with sudden cardiac arrest and in 83% with nonsudden arrest. Cardiac arrest occurred without symptoms in 38% of the patients with sudden cardiac arrest and was the first expression of coronary artery disease in 4% of the entire study group. This study indicates that cardiac arrest usually occurs with symptoms and almost always in the setting of a history of cardiovascular disease.
Angina Pectoris↗
Disability, utilization, and costs associated with musculoskeletal conditions.
In this report, data from the 1980 National Medical Care Utilization and Expenditure Survey are used to present health characteristics, types and quantities of services used, and the charges for these services for persons with musculoskeletal diseases. Slightly more than 44 million people, or 19.8 percent of the U.S. civilian noninstitutionalized population, were reported in the survey to have at least one musculoskeletal disorder. These data are generally consistent with those from other health surveys, which show that the prevalence of musculoskeletal disorders increases for successive age categories, that such disorders are more common among women than among men, and that they are less prevalent among black persons than among persons of other races. In terms of both functional limitation and perceived health status, persons with musculoskeletal conditions are, with some notable exceptions, in relatively poor health. Persons having back problems in addition to problems with peripheral joints (such as the knee, hip, or shoulder) were more likely to rate their health as "fair" or "poor" compared with persons having only back problems or compared with persons in the civilian noninstitutionalized population as a whole. Musculoskeletal disorders accounted for a considerable proportion of all disability days reported by the total civilian noninstitutionalized population: 13 percent of restricted-activity days, 8.8 percent of bed-disability days, and 11.2 percent of all work-loss days were directly attributable to musculoskeletal conditions. The disabling effects of musculoskeletal problems pose a significant economic burden; they accounted for a total of $3.9 billion in lost productivity costs during 1980 for employed persons in the work force and for homemakers. For persons with musculoskeletal problems, the mean number of ambulatory visits per year was nearly twice the rate of 5.2 for the general civilian noninstitutionalized population. Of ambulatory visits made to all health care providers by persons with these conditions, 35.6 percent were related in some way to the treatment of their musculoskeletal problems. Musculoskeletal conditions are somewhat different from many other illnesses because their treatment is within the professional domain of several types of health care providers. Approximately 13 percent of persons with any type of musculoskeletal disorder received care from chiropractors during the year and this figure rose to nearly 30 percent for back problems only. However, nearly 33 percent of persons with musculoskeletal problems made no visits for treatment of their condition.(ABSTRACT TRUNCATED AT 400 WORDS)
Absenteeism↗
Costs of illness: United States, 1980.
The total costs of illness and injury in the U.S. civilian noninstitutionalized population in 1980 amounted to $381.7 billion. The direct costs of illness and injury--resource expenditures for the diagnosis, treatment, and management of medical and dental conditions--were $153.9 billion, or 40.3 percent of total costs. Indirect costs--economic losses from morbidity and mortality--were $227.9 billion, or 59.7 percent of total costs. Of indirect costs, $104.9 billion resulted from productivity losses because of morbidity, and $123.0 billion represent the present value of lost productivity from premature mortality based on a net effective discount rate of 4 percent. These estimates, based on data from the 1980. National Medical Care Utilization and Expenditure Survey (NMCUES), differ from other estimates of the costs of illness and injury in 1980 (Gibson and Waldo, 1982; Rice, Hodgson, and Kopstein, 1985). The differences, which can be resolved, are attributable to two major factors: (1) NMCUES includes only the civilian noninstitutionalized population, but the other estimates include the institutionalized population and the military; and (2) NMCUES indirect cost estimates for the population unable to work include persons who were retired for health reasons in 1979 and 1980, disabled homemakers, and other persons who were disabled for the entire year 1980 but were not retired for health reasons in 1979, but the Rice et al. estimates do not include the last two categories in the population unable to work. The principal NMCUES findings on the total costs of illness in the civilian noninstitutionalized population reinforce the importance of considering distributional effects. Persons 65 years of age and over represent one-tenth of this population yet account for more than one-fourth of direct costs and more than their share of total costs, even though the institutionalized elderly are excluded. More than two-thirds of total costs for this age category are accounted for by direct costs. Direct costs also account for more than two-thirds of total costs for people under 17 years of age. However, this youngest age category, which constitutes over one-fourth of the civilian noninstitutionalized population, generates only 12.3 percent of direct costs. In contrast, indirect costs account for well more than 60 percent of total costs for the working-age population (17-64 years of age). Within the working-age population, per capita direct costs are highest among persons who are not full participants in the work force, many of whom are not working full time or at all because of injury or ill health.(ABSTRACT TRUNCATED AT 400 WORDS)
Adolescent↗
Incidence, utilization, and costs associated with acute respiratory conditions, United States, 1980.
Acute respiratory conditions are common causes of health disturbance in the general population. They are generally self-limiting, although occasionally recurrent, and seldom result in large health care costs for each episode of illness. The National Medical Care Utilization and Expenditure Survey (NMCUES), conducted during 1980, provided an opportunity to assess the effect of acute respiratory conditions on utilization of medical services and on functional capability as well as the cost of related medical care. Acute respiratory conditions were reported by survey respondents and separated into five subgroups: colds, influenza, nasopharyngitis, otitis media, and lower respiratory infections. Allergic conditions and chronic respiratory disorders (tuberculosis, chronic obstructive pulmonary disease, and pneumoconioses) were excluded. The subgroupings of acute respiratory conditions appear to separate the disorders in a manner consistent with the epidemiologic characteristics of each condition. About one-half (50.4 percent) of the U.S. civilian noninstitutionalized population had one or more acute respiratory conditions during 1980. The highest rates for upper respiratory conditions (colds, influenza, nasopharyngitis, and otitis media) were reported for those under 18 years of age, and rates were lower in successively older groups. Lower respiratory infection rates were higher in the youngest and oldest groups. Despite a high incidence in the general population, most symptomatic episodes of colds, influenza, and nasopharyngitis did not result in ambulatory care visits or hospital admissions. Otitis media and lower respiratory infections were more often associated with medical visits. Acute respiratory conditions were associated with lower disability levels than the average for the U.S. civilian noninstitutionalized population during 1980 (5.9 restricted-activity days for acute respiratory conditions, compared with an overall average of 13.8 restricted-activity days). Persons with upper respiratory conditions (colds, influenza, otitis media, and nasopharyngitis) averaged 2.3 to 5.4 restricted-activity days, but persons with lower respiratory infections experienced an average of 8.2 restricted-activity days. Indirect costs attributed to acute respiratory conditions in 1980 were $7.7 billion for employed persons and $698 million for homemakers, for a total of $8.4 billion, about the same as total direct costs ($8.3 billion). These indirect costs were several times larger than the annual indirect costs estimated for either cardiovascular diseases or musculoskeletal diseases, two common chronic or recurrent condition groups. The high indirect costs reflect the high frequency of episodes in the general population during 1980 and the greater likelihood of associated bed-disability and work-loss days than for other conditions.(ABSTRACT TRUNCATED AT 400 WORDS)
Adolescent↗
Blood lead and blood pressure. Relationship in the adolescent and adult US population.
Heavy lead exposure has been connected to cardiovascular disease, but modest exposures encountered in the general environment have not been associated previously with disease risk. The relationship between blood lead levels and blood pressures was examined using data from the second National Health and Nutrition Examination Survey. A direct relationship was found between blood lead levels and systolic and diastolic pressures for men and women and for white and black persons aged 12 to 74 years. Blood lead levels were significantly higher in younger men and women (aged 21 to 55 years) with high blood pressure, but not in older men or women (aged 56 to 74 years). In multiple regression analyses, the relationship of blood lead to blood pressure was independent of other variables for men, but not for women. Dietary calcium and serum zinc levels were inversely related to blood pressure.
Adolescent↗
The effect of bystander CPR on survival of out-of-hospital cardiac arrest victims.
The effect of bystander cardiopulmonary resuscitation (CPR) was studied in 2142 emergency medical service (EMS) cardiac arrest runs. When bystander CPR was administered to cardiac arrest victims, 22.9% of the victims survived until they were admitted to the hospital and 11.9% were discharged alive. In comparison, the statistics for cardiac arrest victims who did not receive bystander CPR were 14.6% and 4.7%, respectively (p less than 0.001). A critical factor in patient survival was the amount of time that elapsed before the EMS personnel arrived and administered CPR. Patients who received bystander CPR were more likely to have ventricular fibrillation when the EMS arrived. Other factors relating to patient survival were the location of the victim at the time of the cardiac arrest and the age of the victim. Understanding these factors is important in developing community strategies to treat patients with cardiac arrest out of hospital.
Adult↗
The reliability of Hertel exophthalmometry. Observer variation between physician and lay readers.
In order to quantify and characterize the interobserver variation in exophthalmometry and to assess the impact of experience on that variation, four observers of varying levels of experience obtained an exophthalmometer reading from 100 subjects' right eyes and 97 subjects' left eyes. Observers were masked to the measurements of their colleagues. The subjects measured were referrals to the eye plastic and orbital service over a three week interval. Substantial interobserver variation was demonstrated. Fully one-fourth to one-third of all measurements by the three less-experienced observers differed enough from the senior observer to be of concern to him. The direction of disagreement varied with experience. The least-experienced observer consistently underread whereas the more experienced observer slightly overread exophthalmometry, relative to the senior observer. As the study progressed, the least-experienced observer's reliability increased. Interobserver agreement for this observer was influenced by the degree of proptosis presenting in the eye being measured.
Analysis of Variance↗
The relationship between blood lead levels and blood pressure and its cardiovascular risk implications.
The relationship between blood pressure and blood lead levels in the second National Health and Nutrition Examination Survey (1976-1980) has been examined for white males aged 40-59 years. After adjustment for age, body mass index, nutritional factors, and blood biochemistries in a multiple linear regression model, the relationship of systolic and diastolic blood pressures to blood lead levels was statistically significant (p less than 0.01). There was no evidence of a threshold blood lead level for this relationship. Although these data alone do not prove a casual relationship between low blood lead levels and blood pressure, the findings are consistent with current epidemiologic and animal studies, indicating that a causal relationship is probable. To examine the potential health risks, the multiple logistic risk factor coefficients from the Pooling Project and Framingham studies were used to predict the impact of the 37% decrease in mean blood lead levels which occurred in adult white males from 1976 to 1980. As a result of this blood lead decrease, the calculations predicted a 4.7% decrease in the incidence of fatal and nonfatal myocardial infarction over 10 years, a 6.7% decrease in the incidence of fatal and nonfatal strokes over 10 years, and a 5.5% decrease in the incidence of death from all causes over 11.5 years. In addition, as a result of this blood lead decrease, the predicted number of white males in this age group with hypertension (diastolic blood pressure greater than or equal to 90 mmHg) decreased by 17.5%.
Adult↗
Some factors influencing interobserver variation in classifying simple pneumoconiosis.
Three experienced physician readers assessed the chest radiographs of 743 men from a coal mining community in West Virginia for the signs of simple pneumoconiosis, using the ILO U/C 1971 Classification of Radiographs of the Pneumoconioses. The number of films categorised by each reader as showing evidence of simple pneumoconiosis varied from 63 (8.5%) to 114 (15.3%) of the 743 films classified. The effect of film quality and obesity on interobserver agreement was assessed by use of kappa-type analytic procedures for measuring agreement on categorical data. Poor film quality and obesity both affected agreement adversely. Poor quality films were disproportionately frequent in obese individuals, as defined by the Quetelet index. On control of film quality by stratification, the effect of obesity on interobserver profusion agreement was no longer evident.
Coal Mining↗