Post-stratification Adjustment Research Articles

Behavioral Health Risk Factors of United States Emergency Medical Technicians: The LEADS Project

Personal risk behaviors are modifiable. This report describes the 2002 national baseline of behavioral health risk factors of US emergency medical technicians (EMTs) that can guide policy and program development in improving EMT well-being. A 19-item Health Behavioral Risk Survey (Appendix) was added to the 2002 Longitudinal Emergency Medical Technician Demographic Study mail survey. Risk survey questions covering physical activity, tobacco use, and alcohol use were modeled after the Centers for Disease Control and Prevention 2002 Behavioral Risk Factor Surveillance System (BRFSS) questionnaire. Personal, non-work related seatbelt use and motor vehicle driving questions were adopted from the 2002 US National Highway Traffic Safety Administration (NHTSA) Motor Vehicle Occupant Safety Survey (MVOSS). Post-stratification adjustment factors were used to allow comparisons with BRFSS and MVOSS national estimates. A total of 1,919 EMT respondents were compared with 239,866 BRFSS and 5,220 MVOSS respondents. These comparisons indicate that EMT-Basics drove more slowly than paramedics; male EMTs drove faster, drank more, and wore their seatbelts less often than did female EMTs; female EMTs smoked more and engaged in vigorous exercise less than males. Those EMTs who reported to be in fair or poor health, smoked more and exercised less than those who reported to be in good or excellent health. Regardless of gender, age, or race, EMTs, on average, wore their seatbelts less often, drove faster than, and were less likely to engage in moderate physical exercise, compared to US adults. Stereotypical gender differences in risk behaviors exist among EMTs. An EMT's self-reported health positively correlates with smoking and exercising. Compared to US national estimates, except for smoking and vigorous exercise, EMTs have increased risk behaviors.

Modified balanced repeated replication for complex survey data

SUMMARY Balanced repeated replication, commonly used to estimate the variances of nonlinear estimators from stratified sampling designs, can run into problems in estimating a ratio of small domain totals or in the case of poststratification or unit nonresponse adjustment involving ratio weighting within cells if some cell sizes are small. This problem is mainly caused by a sharp perturbation of the weights to construct replicate estimates. To avoid extreme replicate estimates, Robert Fay proposed a modified balanced repeated replication method using gentler perturbation of weights. Theoretical properties of the modified method are studied here. The method is extended to the case of imputation for item nonresponse. Both smooth and nonsmooth estimators are studied.

Race as a risk factor in the severity of infragenicular occlusive disease: Study of an urban hospital patient population

We retrospectively studied the arteriograms of 135 men admitted for evaluation of lower extremity ischemia to examine whether race influences the severity of infragenicular occlusive disease. The scoring system prepared by the Ad Hoc Committee on Reporting Standards for the Society for Vascular Surgery and the International Society for Cardiovascular Surgery was used to grade the severity of stenosis in each of the upper, middle, and lower thirds of the anterior and posterior tibial and peroneal arteries (collectively called “infragenicular” arteries). The patients were divided into two groups: 83 blacks (140 arteriogram limbs) and 52 whites (87 arteriogram limbs). Disease severity scores between the groups were compared, and the existence of five known risk factors for atherosclerosis were considered for poststratification adjustment. Results showed that higher disease scores, indicating more severe disease, were found in the black population in every segment of the infragenicular arteries. The mean (± SE) score for all the infragenicular segments in blacks was significantly higher than that in the whites (2.08 ± 0.05 vs 1.57 ± 0.06, p < 0.001). The black and white groups were comparable in terms of age (65.2 vs 64.6 years), prevalence of diabetes (20% vs 25%), smoking history (93% vs 90%), and hypercholesterolemia (51% vs 63%). Hypertension was more prevalent among the black patients (51% vs 27%, p < 0.001). When only the non-hypertensive patients in both groups were considered, however; the mean severity score was still significantly higher in blacks (2.10 ± 0.06 vs 1.42 ± 0.06, p < 0.001). These findings provide evidence that there are racial differences in the severity of infragenicular occlusive disease. Further investigation is needed to determine whether this is due to a genetic predisposition or is a result of unidentified environmental factors.