PubMed Health⌕ Search

Biomedical subjects

D W Hosmer

Publications and source records attributed to D W Hosmer.

At least 19 recordsLinked to original sources
PubMed1992

Effects of alterations in fatty acid intake on the blood pressure of adolescents: the Exeter-Andover Project.

To judge the effect on blood pressure, the ratio of polyunsaturated to saturated fatty acids (P:S) of foods served to students at two boarding high schools was modified alternately at each school for one school year. The average P:S of the diet of males increased from 0.53 to 0.93 during the intervention whereas among females it increased from 0.64 to 0.98. Comparison of repeated systolic and diastolic blood pressure measurements near the end of the school year did not demonstrate a beneficial effect of the dietary fat changes on the blood pressure of these normotensive adolescents. Compared with the blood pressure patterns during control years, the dietary intervention resulted in slightly higher systolic (+0.88 mm Hg; 95% CI -0.66, +2.42) and diastolic (+1.23 mm Hg; 95% CI = +0.04, +2.42) blood pressure readings among males. Among females the intervention resulted in slightly lower systolic (-0.54 mm Hg; 95% CI = -1.95, +0.88) and diastolic (-0.80 mm Hg (95% CI -2.18, +0.58) blood pressure readings.

Adolescent↗
PubMed1992

Confidence interval estimation of interaction.

Relative excess risk due to interaction, the proportion of disease among those with both exposures that is attributable to their interaction, and the synergy index have been proposed as measures of interaction in epidemiologic studies. This paper presents the methodology for obtaining confidence interval estimates of these indices utilizing routinely available output from multiple logistic regression software.

Confidence Intervals↗
PubMed1992

The prevalence of risk factors for venous thromboembolism among hospital patients.

BACKGROUND: This study provides an estimate of the prevalence of risk factors for venous thromboembolism among hospital patients. METHODS: The presence of risk factors for venous thromboembolism was determined from a retrospective review of the medical records of 1,000 randomly selected patients in 16 acute care hospitals in central Massachusetts. RESULTS: The most common risk factors for venous thromboembolism were age 40 years (59%) or more, obesity (28%), and major surgery (23%). The average number of risk factors increased with increasing age. One or more risk factors for venous thromboembolism were present in 78% of hospital patients, two or more in 48%, three or more in 19%, four or more in 6%, and five or more in 1%. CONCLUSION: Risk factors for venous thromboembolism are common among hospital patients, suggesting that prophylaxis should be widely employed. The cost-effectiveness and risk benefit of prophylaxis is well established in patients undergoing major surgery. Further studies are needed to confirm the benefit of prophylaxis in patients with nonsurgical risk factors for venous thromboembolism.

Age Factors↗
PubMed1991

Physician practices in the prevention of venous thromboembolism.

OBJECTIVE: To determine the rate of use of prophylaxis for venous thromboembolism in high-risk hospital patients. DESIGN: A retrospective medical record review. SETTING: A community-wide study in 16 short-stay hospitals in central Massachusetts. PATIENTS: A total of 2017 patients with multiple risk factors for venous thromboembolism. MEASUREMENTS AND MAIN RESULTS: On the basis of age, length of hospitalization, and the presence of at least one additional major risk factor, 17% of 151,349 discharges (25,410 patients) were identified as being at high risk for venous thromboembolism. Eight percent of these discharges were randomly selected for medical record review. Prophylaxis for venous thromboembolism was received by 32% of these high-risk patients. Prophylaxis use among the 16 study hospitals varied widely, ranging from 9% to 56%, and was higher in teaching hospitals than in nonteaching hospitals (44% compared with 19%; P less than 0.001). One or more of the following methods of prophylaxis was used: low-dose heparin (78%), intermittent calf compression (13%), warfarin (12%), and inferior vena caval filter (3%). Use of prophylaxis increased with the number of risk factors identified (P less than 0.001). CONCLUSION: Prophylaxis for venous thromboemobolism is underused, particularly in nonteaching hospitals.

Adult↗
PubMed1991

The importance of assessing the fit of logistic regression models: a case study.

BACKGROUND: The logistic regression model is being used with increasing frequency in all areas of public health research. In the calendar year 1989, over 30% of the articles published in the American Journal of Public Health employed some form of logistic regression modeling. In spite of this increase, there has been no commensurate increase in the use of commonly available methods for assessing model adequacy. METHODS: We review the current status of the use of logistic regression modeling in the American Journal of Public Health. We present a brief overview of currently available and easily used methods for assessing the adequacy of a fitted logistic regression model. RESULTS: An example is used to demonstrate the methods as well as a few of the adverse consequences of failing to assess the fit of the model. One important adverse consequence illustrated in the example is the inclusion of variables in the model as a result of the influence of one subject. CONCLUSIONS: Failure to address model adequacy may lead to misleading or incorrect inferences. Recommendations are made for the use of methods for assessing model adequacy and for future editorial policy in regard to the review of articles using logistic regression.

Bias↗
PubMed1991

A population-based perspective of the hospital incidence and case-fatality rates of deep vein thrombosis and pulmonary embolism. The Worcester DVT Study.

A community-wide study was conducted in 16 short-stay hospitals in metropolitan Worcester, Mass, to examine the incidence and case-fatality rates of deep vein thrombosis and pulmonary embolism in patients hospitalized between July 1, 1985, and December 31, 1986. The average annual incidence of deep vein thrombosis alone was 48 per 100,000, while the incidence of pulmonary embolism with or without deep vein thrombosis was 23 per 100,000. The incidence rates of deep vein thrombosis and pulmonary embolism increased exponentially with age. The in-hospital case-fatality rate of venous thromboembolism was 12%. Among patients discharged from the hospital, the long-term case-fatality rates were 19%, 25%, and 30% at 1, 2, and 3 years after hospital discharge. Extrapolation of the data from this population-based study suggests that there are approximately 170,000 new cases of clinically recognized venous thromboembolism in patients treated in short-stay hospitals in the United States each year, and 99,000 hospitalizations for recurrent disease. Because of the silent nature of this disease and the low rate of autopsy in the United States, the total incidence, prevalence, and mortality rates of venous thromboembolism remain elusive.

Adolescent↗
PubMed1989

Effects on blood pressure of a decrease in sodium use in institutional food preparation: the Exeter-Andover Project.

To judge the effect on blood pressure, sodium intake of students at two boarding high schools was reduced by 15-20% through changes in food purchasing and in preparation practices in the schools' kitchens. Students were not asked to change their usual eating habits. Each school served alternately as the control or intervention school for one school year. Blood pressure was monitored among 341 subjects during control years and 309 subjects during intervention years. Analysis of blood pressure differences between early in the school year and near the end of the school year, with adjustment for sex and initial blood pressure, showed the effect of the dietary intervention to be -1.7 mmHg for systolic (95% CI = -0.6, -2.9, p = 0.003) and -1.5 mmHg for diastolic pressure (95% CI = -0.6, -2.5, p = 0.002). Such modest and easily attainable changes in sodium intake, if maintained, could have a significant effect on the future risk of essential hypertension among young people.

Adolescent↗
PubMed1989

Birthweight of single livebirths and weight specific early neonatal mortality in Hungary and Norway.

Comparisons of birthweights of single livebirths in Hungary and Norway reveal distributions to have similar shapes; however, in the case of Hungary the distribution is shifted to the left, i.e. towards lower weights. In the registration of pregnancy outcomes, almost identical definitions are applied in the two countries, and the observed difference in distributions of birthweights is taken to reflect that Norwegian livebirths, are on average about 300 g heavier than Hungarian livebirths. Employing the method of analysis of birthweight and perinatal mortality suggested by Wilcox & Russell, it can be demonstrated that the proportions of births in the residual distributions of birthweights in the two countries are of the same magnitude and that the relative differences in first week mortality risks are similar for all birthweights. These results are taken to support the conclusion that to use a cut-off point of 2500 g in defining low birthweight, which will result in a two fold higher proportion of such infants in Hungary compared to Norway, is unwarranted, as it will falsely convey the impression of relatively more obstetric and paediatric problems in Hungary.

Birth Weight↗
PubMed1988

Sample size requirements for studies estimating odds ratios or relative risks.

This paper presents formulae for determining the number of subjects necessary, in either a case-control or a cohort study, to estimate the odds ratio or relative risk, respectively, to within a selected percentage (epsilon) of the true population value with some specified probability. This approach differs somewhat from previous comparable work that estimated the log odds ratio within a stated fixed distance rather than as a percentage of the actual odds ratio. Comparable development for relative risk has not previously appeared in the literature. These formulae provide guidelines for determination of study size that does not depend on hypothesis testing considerations.

Epidemiologic Methods↗
PubMed1988

Association of electronic fetal monitoring during labor with cesarean section rate and with neonatal morbidity and mortality.

Data from the 1980 National Natality Survey by the National Center for Health Statistics were used to assess the relation of electronic fetal monitoring (EFM) during labor with cesarean section rates and neonatal morbidity and mortality. In univariate analyses, EFM was associated with higher cesarean section rates, lower five-minute Apgar scores, and a higher rate of respiratory distress. Logistic regression analysis controlling for other risk factors for poor neonatal outcome indicated that the association of EFM with higher cesarean section rates persisted (odds ratio 1.45, 95% CI 1.16, 1.81), except in certain pregnancies at very high risk for cesarean section. EFM was associated with an Apgar score less than 6 at five minutes only if delivery was by cesarean section. EFM was not found to be independently associated with respiratory distress. Neither univariate nor multivariate analyses found an association of EFM with neonatal mortality. These results suggest that EFM may identify hypoxic infants, who are frequently delivered by cesarean section. The lack of association of EFM with beneficial neonatal outcomes is consistent either with lack of effect of EFM or with uncontrolled selection bias.

Apgar Score↗
PubMed1988

Spontaneous abortion and general illness symptoms among semiconductor manufacturers.

The risk of adverse reproductive outcomes was examined among semiconductor manufacturers. Personal interviews were conducted with manufacturing workers, spouses of male manufacturers, and an internal comparison group of non-manufacturing workers. Elevated spontaneous abortion ratios were observed for females working in the "diffusion" (38.9%; relative risk = 2.18, 95% confidence interval estimate = 1.1, 3.6) and photolithographic process (31.1%; relative risk = 1.75, 95% confidence interval estimate = 0.8, 3.3). Analysis of potential confounding did not substantially alter the findings. The potential for recall bias was also assessed; although the confidence interval for the comparison of diffusion and non-manufacturing workers no longer excluded 1.0, an excess risk was still reflected. No substantive differences in other reproductive outcomes were identified. Various general health symptoms were examined and reported more frequently among manufacturers than non-exposed. These results should be viewed as tentative until studies with larger numbers and more detailed exposure data are carried out.

Abortion, Spontaneous↗
PubMed1987

Factors related to patient smoking status.

To investigate those factors associated with patients' cigarette smoking status, 455 consecutive patients seen in two specialty clinics and one general medicine clinic at a university medical center were studied. Patient's age, sex, health status, and number of previous cessation attempts discriminated current from ex-smokers. A strong interaction was observed between sex and disease status with females showing a greater impact of smoking-related disease on smoking behavior than males.

Adult↗
PubMed1987

Sources of sodium, potassium, and energy in the diets of adolescents.

Analysis of food diaries of 200 students in two boarding high schools indicated that 30% of their average daily sodium intake was contributed by bakery foods and ready-to-eat cereals. The next most important source of sodium (24%) was a category made up of food adjuncts, snack-type foods, soups, and beverages. The largest proportion of sodium in most foods had been added during food manufacture or preparation. Salt added at the table contributed less than 1% of the sodium in the diets of the teenagers. All students did not eat foods from all categories on their day of recording; 70% or more ate bakery products, dairy foods, food adjuncts, and meat, fish, poultry, and eggs. Changes in recipe formulations can decrease sodium content for a variety of bakery products, cereals, mixed dishes, snack-type foods, soups, and processed meats. It is therefore possible to make large reductions in dietary sodium without disturbing other nutrient values.

Adolescent↗
PubMed1979

The effect of feeding on fat deposition in early infancy.

Trained nutritionists surveyed 92 female infants and their mothers in Western Massachusetts at monthly intervals from birth to age 6 months. A diet history, a three-day record of food intake and 11 anthropometric measurements were obtained at each home visit. Skinfold measurements showed an earlier plateau than in other published studies. There were no significant differences in fat thicknesses when infants were classified by method of feeding until age 2 months, but formula-fed infants with solids started before age 2 months had the largest mean skinfolds which peaked at age 3 months and then decreased and became similar to those of the other groups by age 5 months.

Bottle Feeding↗
PubMed1978

A computer program for stepwise logistic regression using maximum likelihood estimation.

A computer program has been written which performs a stepwise selection of variables for logistic regression using maximum likelihood estimation. The selection procedure is based on likelihood ratio tests for the coefficients. These tests are used in a forward selection and a backward elimination at each step. The use of the program is illustrated by several examples.

Computers↗