Objective: Conflicting research findings regarding the ability of tension or anxiety to predict incident coronary heart disease (CHD) have created uncertainty in the literature. In addition, there are no prospective studies relating these characteristics to the development of atrial fibrillation (AF). Methods: From 1984 to 1987, 3682 participants (mean age 48.5 ± 10.1 year; 52% women) of the Framingham Offspring Study were examined and followed for 10 years for the incidence of CHD, AF, and total mortality. Measures of tension, anxiety, and risk factors for CHD and AF were collected at the baseline examination. Results: After adjusting for age, systolic blood pressure, body mass index, current cigarette smoking, diabetes, and total cholesterol/high-density cholesterol in Cox proportional hazards models, increased tension was predictive of 10-year incidence of definite CHD (relative risk (RR) = 1.25 relative to a one SD difference; 95% confidence interval (CI), 1.05–1.49) and total mortality (RR = 1.23; 95% CI, 1.06–1.42) in men. After adjusting for AF risk factors, tension also predicted AF in men (RR = 1.24; 95% CI, 1.04–1.48). Anxiety in men (RR = 1.22; 95% CI, 1.08–1.38), and in women (RR = 1.27; 95% CI, 1.05–1.55) was significantly related to total mortality. Conclusions: Tension was observed to be an independent risk factor for incident CHD, AF, and mortality in men. Anxiety was a risk factor for total mortality in men and women. Our findings suggest that further research into the pathophysiology of the excess morbidity and mortality observed with tension and anxiety is merited. AF = atrial fibrillation; CHD = coronary heart disease; RR = relative risk; CI = confidence interval; BMI = body mass index.

This research elucidates the risk of institutional care and/or death of patients with Alzheimer's disease (AD) or other dementia (OD) compared with noncases. Community dwelling incident cases of AD (n = 240) or OD (n = 208) and age-matched noncases (n = 363) living in an enumerated population were included. The adjusted hazard ratio (HR) of being admitted to a nursing home compared with controls was 5.44 (95% confidence interval, CI: 3.68,8.05) for AD cases and 5.08 (95% CI: 3.38, 7.63) for OD cases. The adjusted RR of death compared to controls was 1.90 (95% CI: 1.36, 2.65) for AD cases and 2.27 (95% CI: 1.62, 3.18) for OD cases. This is the only population-based study in the U.S. documenting that AD/OD increases the risk of death independent of comorbid conditions. AD/OD patients have more comorbidities and are at greater risk of entering a nursing home than to die during the follow-up, compared with those without these diagnoses.

PURPOSE: Recent findings on medical care utilization among people with Alzheimer's disease (AD) or other dementia (OD) are conflicting. A population-based case-control study was designed to determine if patients with clinically diagnosed AD or OD have different medical care utilization patterns before and after diagnosis compared to age-matched controls. METHODS: All community dwelling incident cases of AD (n = 240) or OD (n = 208) diagnosed between July 1, 1992 and June 30, 1997, and age-matched controls (n = 363) living in an enumerated population, were included. Medical care services, defined by Physicians' Current Procedural Terminology Codes, were assessed the year prior to diagnosis and up to five years after diagnosis of AD/OD. RESULTS: Prior to diagnosis, AD/OD cases had a significantly greater age-adjusted average number of medical services per year, 33 and 29, respectively, compared to 20 for controls (p = 0.0001 and p = 0.0002, respectively). These differences were significant after adjusting for age, gender, follow-back time, comorbidities, and insurance status. After diagnosis, AD/OD cases still had significantly greater age-adjusted average number of medical services per year (34 services and 37 services, respectively) compared to controls with 27 services (p = 0.0064 and p = 0.0006, respectively). After adjusting for confounding variables (particularly comorbid conditions), these differences were no longer significant. CONCLUSIONS: This research demonstrates that higher medical care utilization by patients with AD/OD after diagnosis is not due to dementia per se, but explained by their poorer physical health compared to controls. Greater utilization prior to diagnosis of AD/OD is most likely due to the search for a diagnosis for the person developing dementia.

The Women's Health Alliance Intervention Study is a community-based intervention study designed to determine if county-based coalitions can increase breast and cervical cancer screening rates among women aged 40 and older in four rural Wisconsin counties. This article describes the intervention process and demonstrates the feasibility of rural volunteer coalitions to promote breast and cervical cancer screening in communities. An accompanying article in this issue describes the effectiveness of these interventions.

The Women's Health Alliance Intervention Study is a quasi-experimental intervention designed to test if county-based coalitions can increase breast and cervical cancer screening compliance among women aged 40 years and older living in rural communities. A number of interventions were designed and implemented by coalitions in four counties in north-central Wisconsin during a 2-year period. Four control counties in southwestern Wisconsin were identified for comparison. Judging from the results of this study, community-based intervention efforts can increase breast and cervical cancer screening compliance significantly among women living in rural communities.

Objective: to study the impact of serum cholesterol concentrations on the total risk of mortality in older people.Design: retrospective cohort study with a follow-up of 8-10 years,Subjects: a total of 989 subjects (367 men and 622 women) aged 65 and over, living in the Marshfield Epidemiologic Study Area at the time of their first complete serum lipid assessment.Methods: we calculated sex-specific mean levels of serum total cholesterol, low-density lipoprotein, high-density Lipoprotein and triglycerides, and the ratio of total cholesterol to high-density lipoprotein, for subjects who died of all causes and for those who survived to the end of follow-up, with adjustment for relevant covariates. We obtained estimates of the risk factor-adjusted sex-specific relative risk for all-cause mortality with approximate quartiles of serum cholesterol concentrations by proportional hazards regression models. We also evaluated the possible combined effects of age, sex and cholesterol on all-cause mortality.Results: a high level of high-density lipoprotein was significantly associated with a low total risk of mortality in older men. Conversely an elevated ratio of total cholesterol to high-density Lipoprotein was directly related to an increased total risk of mortality in older men. Age and high-density lipoprotein level had a significant synergistic effect on all-cause mortality for the elderly men. We found little or no association in women between all-cause mortality and any of the Lipid measures studied.Conclusions: an increased ratio of total cholesterol to high-density lipoprotein appears to be associated with an increase in risk for all-cause mortality in men aged 65 and over, while an elevated level of high-density lipoprotein, considered alone, seems to be protective against mortality from all causes in men aged 65-74 years, but this effect diminishes over the age of 75.

This study examines the relationship between family history of breast cancer and current compliance with mammography screening guidelines. A random telephone survey of women age 40 or older living in rural Wisconsin (N = 2398) was conducted to determine mammography screening knowledge, family history of breast cancer, attitudes, intentions, physician recommendation, and compliance with screening guidelines. Compared with women without a family history of breast cancer, women with a family history were significantly more likely to demonstrate correct knowledge (p = 0. 01); express intentions in compliance with recommended screening guidelines (p < 0.001); report having been advised by a physician to obtain a mammogram (p < 0.001); and be in current compliance with mammography screening guidelines (p < 0.001). Results of simultaneous and individual logistic regression suggest that the effects of family history on compliance with screening guidelines are directly mediated through the combination of women's knowledge, women's intentions, and physician recommendation. Thus, programs to increase compliance with mammography screening guidelines should address both women and providers.

Results from a randomized controlled trial demonstrated that a physician reminder letter combined with telephone counseling from a health educator significantly increased women's use of both mammograms and Pap tests in a low-income population in a managed care setting. This article presents results from a process evaluation and cost analysis of the intervention. An average of 35 minutes was spent preparing each of 304 intervention letters for mailing, including the time needed to secure signatures from 110 physicians. The results of an economic analysis suggested that this intervention cost $11.44 per recipient and $28.93 per screening test received above expected. However, intervention costs can be reduced significantly if one physician signs all letters (rather than each woman's own primary care physician) and if the health educator labor costs are diminished (e.g., by using student interns). Overall, the women under study reported that they are comfortable with both mailed and telephone reminders when they are post due for a clinical preventive service.

To capitalize on Marshfield Clinic's advantages for population-based health research, we developed the Marshfield Epidemiologic Study Area (MESA). Marshfield Clinic is an integrated system consisting of a large multispecialty clinic and 23 affiliated clinics. Clinic physicians provide virtually all of the medical care, both inpatient and outpatient, for residents of the area. MESA consists of 14 ZIP codes in which over 95% of the 50,000 residents and most significant health events are captured in Marshfield Clinic databases, including all deaths, 94% of hospital discharges, and 92% of medical outpatient visits. MESA exemplifies the research potential of integrated medical care systems and the efforts required to realize that potential, Because it is representative of a defined population and provides an unselected sample of patients, MESA is well suited for epidemiologic research and research elucidating the clinical spectrum and natural history of diseases and the effectiveness of treatment.

Using linear regression, the authors demonstrated a strong association between State-specific coronary heart disease mortality rates and State prevalence of sedentary lifestyle (r2 = 0.34; P = 0.0002) that remained significant after controlling for the prevalence of diagnosed hypertension, smoking, and overweight among the State's population. This ecologic analysis suggests that sedentary lifestyle may explain State variation in coronary heart disease mortality and reinforces the need to include physical activity promotion as a part of programs in the States to prevent heart disease.

Objectives This study assessed whether a state public health department could effectively implement an affordable nutrition intervention program at the community level.Design Cross-sectional data were collected via telephone surveys of 9,839 adults, aged 18 years or older, in 1987, 1989, and 1991 in two South Carolina communities. Nutrition education programs began in 1988 in one community. The other community served as a comparison site. We assessed and compared changes in community levels of dietary fat and weekly meat consumption salt use, and nutrition promotion awareness with analysis of covariance regression techniques that included race, sex, and age as covariates.Results We observed favorable changes in most eating behaviors and levels of awareness in both communities. The intervention community experienced greater absolute changes than the comparison community in use of animal fats (-8.9% vs -4.0%; P=.02) and liquid or soft vegetable fats (+8.4% vs +3.6%; P=.04), and in awareness of restaurant nutrition information (+33.0% vs + 19.4%; P=.0001). Although the primary type of dietary fat used differed between black and white respondents, we observed significant change among both groups.Conclusions These results suggest that community-wide nutrition education programs may have augmented regional or national changes in dietary behavior among white and black adults in the intervention community.

The purpose of this analysis was to describe the association of leisure-time physical activity and high-density lipoprotein cholesterol levels among a large community sample of African-American and white men and women. Physical assessment of high-density lipoprotein cholesterol, weight, height, waist, and hip measurements were obtained for 3121 randomly selected community participants (969 white men, 1409 white women, 205 African-American men, and 538 African-American women). Leisure-time physical activity was assessed from responses to a standard series of questions about participation in various leisure-time activities. We found an inverse relationship between leisure-time activities and high-density lipoprotein cholesterol levels in African Americans, in contrast to a positive association observed in whites; however, these associations failed to reach statistical significance after controlling for age and body composition. These results confirm a strong association of body composition with high-density lipoprotein cholesterol and support continued interventions designed to reduce overall body weight, particularly upper body weight. The results further suggest that the relationships between some life-style factors and HDL levels may be different among minorities.

A pilot study of eighty-two caregivers was conducted in South Carolina in 1991 to identify positive and negative factors associated with caregiving. Through home visits, interviewers obtained data on a variety of physical and mental health measures, including two new scales designed to measure perceived ''positive and ''negative'' events that had occurred in the previous month. The Center for Epidemiologic Studies Depression scale was used as a measure of depressive symptomatology. For the new scales, only items that were significantly correlated with depressive symptomatology (p < 0.01) were retained. The new ''positive'' event scale (8 items) and the new ''negative'' event scale (16 items) had alpha coefficients of 0.79 and 0.86, respectively. These scales may be useful to researchers in sorting out mediating factors related to the burden of caregiving and in providing points for intervention.