Health survey reliability in an active duty population.

We examined the reliability of data collected from the Health Enrollment Assessment Review (HEAR) survey, a self-report instrument administered by the Department of Defense styled after the Behavioral Risk Factor Surveillance System (BRFSS). Survey responses from a convenience sample of active duty service members who completed a HEAR survey on two occasions were examined. We measured test-retest reliability by comparing individuals' responses to several of the survey items and compared HEAR reliability patterns with the reliability patterns of similar studies conducted with the BRFSS. The majority of estimates reflected fair to excellent reliability. We found substantial agreement between the results of this investigation and similar BRFSS studies. Our findings support the reliability of responses from the HEAR survey for active duty military members. Our results were generally consistent with those of other studies despite differences in survey administration, respondent characteristics, and privacy guarantees.

Going the distance: how far will women travel to undergo free mammography?

This study examined the effect of travel distance on mammography facility choice, with mammography cost held constant. Women who were eligible to undergo free mammography at a military treatment facility (MTF) responded to a mail survey by indicating whether their last mammograms were performed at an MTF (for free) or at a civilian medical facility (for a fee). The distance from each respondent's home to the MTF was also recorded. Analyses indicated that a majority of the women who traveled to the MTF for a mammogram lived within 20 miles of the facility. Results of this study suggest that, even if mammography screening is offered free of charge, women may not use the service if they must travel more than 20 miles to receive it. Furthermore, any MTF trying to increase its TRI-CARE market share may want to focus its marketing efforts on individuals who live within a 20-mile radius.

Teleoncology in the Department of Defense: a tale of two systems.

Two telemedicine networks were developed for the purpose of conducting multidisciplinary oncology ("teleoncology") conferences. The infrastructure of each system differed: one system was Internet-based; the other was delivered via Integrated Services Digital Network (ISDN) lines. The purpose of this study was to describe the infrastructure and cost, consultative process, technical aspects, and conference format of the two teleoncology programs. The two systems' technical aspects, participant satisfaction with the systems, and conference participation were compared qualitatively. Assessment of the technical aspects of the systems suggested that each had distinct advantages. Survey results indicated that provider satisfaction with the technical and logistical aspects of each type of teleoncology conference was high. The present study may prove helpful for individuals who are considering implementing their own teleoncology programs.

Predictors of perceived memory impairment: do they differ in Alzheimer's disease versus normal aging?

Predictors of perceived memory impairment were investigated in 40 elderly normal adults and 28 individuals with Alzheimer's disease. Measures of perceived memory impairment, global cognitive functioning, memory, use of memory strategies, memory strategy efficacy, and depressive symptomatology were obtained for all participants. The elderly normal and Alzheimer's disease groups did not differ in the extent to which they reported perceived memory impairment. For both participant groups, more frequent use of memory strategies and lower perceived memory strategy efficacy were significant predictors of perceived memory impairment. Depressive symptomatology was an additional, significant determinant of perceived memory impairment for the elderly normal group.

Ethnic trends in survival curves and mortality.

The nature of secular trends in survival curves has been widely debated. Fries (1984) has argued for increasing rectangularization, while Myers and Manton (1984a; 1984b) have observed increases in mean age at death with little or no change in standard deviation--arguing against rectangularization. We hypothesize that ethnic differences in mortality trends may shed light on this argument. Using California population data for 1970, 1980, and 1990, we examined ethnic differences in rectangularization using both visual and means and standard deviations analyses. The resulting patterns varied by ethnicity, gender, and type of analyses. Nearly all female groups demonstrated modest rectangularization, regardless of mean age of death, while most of the male groups did not.