Weight loss from maximum body weight and mortality: the Third National Health and Nutrition Examination Survey Linked Mortality File.

OBJECTIVE: The aim of this longitudinal study is to examine the relationship between weight loss from maximum body weight, body mass index (BMI), and mortality in a nationally representative sample of men and women. DESIGN: Longitudinal cohort study. SUBJECTS: In all, 6117 whites, blacks, and Mexican-Americans 50 years and over at baseline who survived at least 3 years of follow-up, from the Third National Health and Nutrition Examination Survey Linked Mortality Files (1988-1994 with passive mortality follow-up through 2000), were included. MEASUREMENTS: Measured body weight and self-reported maximum body weight obtained at baseline. Weight loss (maximum body weight minus baseline weight) was categorized as <5%, 5-<15%, and >or=15%. Maximum BMI (reported maximum weight (kg)/measured baseline height (m)(2)) was categorized as healthy weight (18.5-24.9), overweight (25.0-29.9), and obese (>or=30.0). RESULTS: In all, 1602 deaths were identified. After adjusting for age, race, smoking, health status, and preexisting illness, overweight men with weight loss of 15% or more, overweight women with weight loss of 5-<15%, and women in all BMI categories with weight loss of 15% or more were at increased risk of death from all causes compared with those in the same BMI category who lost <5%; hazard ratios ranged from 1.46 to 2.70. Weight loss of 5-<15% reduced risk of death from cardiovascular diseases among obese men. CONCLUSIONS: Weight loss of 15% or more from maximum body weight is associated with increased risk of death from all causes among overweight men and among women regardless of maximum BMI.

Serum folate and cardiovascular disease mortality among US men and women.

BACKGROUND: Folate has been linked to cardiovascular disease (CVD) through its role in homocysteine metabolism. OBJECTIVE: To assess the relationship between serum folate and CVD mortality. DESIGN: In this prospective study, serum folate concentrations were measured on a subset of adults during the Second National Health and Nutrition Examination Survey (1976-1980) and vital status ascertained after 12 to 16 years. SETTING AND PATIENTS: A national probability sample consisting of 689 adults who were 30 to 75 years of age and did not have a history of CVD at baseline. MAIN OUTCOME MEASURE: Vital status was determined by searching national databases that contained information about US decedents. RESULTS: The associations between serum folate and CVD and all-cause mortality differed by diabetes status (P =.04 and P =.03, respectively). Participants without diabetes in the lowest compared with the highest serum folate tertile had more than twice the risk of CVD mortality after adjustment for age and sex (relative risk [RR], 2.64; 95% confidence interval [CI], 1.15-6.09). This increased risk for participants in the lowest tertile was attenuated after adjustment for CVD risk factors (RR, 2.28; 95% CI, 0.96-5.40). Serum folate tertiles were not significantly associated with total mortality, although the age- and sex-adjusted risk was increased for participants in the lowest compared with highest tertile (RR, 1.74; 95% CI, 0.96-3.15). Risk estimates for participants with diabetes were unstable because of the small sample size (n = 52). CONCLUSION: These data suggest that low serum folate concentrations are associated with an increased risk of CVD mortality among adults who do not have diabetes. Arch Intern Med. 2000;160:3258-3262.

Incidence of hypertension and educational attainment: the NHANES I epidemiologic followup study. First National Health and Nutrition Examination Survey.

Previous research has demonstrated the association between cardiovascular disease and education. However, few studies have described the incidence of hypertension, a risk factor for cardiovascular disease, by education or other socioeconomic status indicators. To examine the association between hypertension incidence and education, the authors analyzed data from the First National Health and Nutrition Examination Survey (NHANES I) Epidemiologic Followup Study (NHEFS) (1971-1984). The relative risk of hypertension incidence (blood pressure > or =160/95 and/or using antihypertensive medication) by education was calculated for non-Hispanic Whites (aged 25-64 years) and non-Hispanic Blacks (aged 25-44 years) normotensive at baseline using Cox proportional hazards models. The age-adjusted relative risk of hypertension incidence among persons with less than 12 years of education compared with those with more than 12 years was significant among non-Hispanic Whites aged 25-44 years (men: relative risk (RR) = 2.14, 95% confidence interval (CI): 1.29, 3.54; women: RR = 2.06, 95% CI: 1.39, 3.05) but not among non-Hispanic Blacks (RR = 1.16, 95% CI: 0.63, 2.14). Relative risks for non-Hispanic White men remained stable after adjusting for age, systolic blood pressure, body mass index, and region of residence; relative risks for non-Hispanic White women were reduced but remained significant. Non-Hispanic White men and women aged 45-64 years with less than 12 years of education were not at higher risk of developing hypertension compared with their more educated counterparts. These results demonstrate a significant interaction between age and education with an independent association between education and hypertension incidence among younger but not older non-Hispanic White men and women.

Eating frequency and risk of colorectal cancer.

Case-control studies have found elevated risk of colorectal cancer with higher eating frequency. The present analyses, the first to examine this association using prospectively collected information, utilized nationally representative data from the Epidemiologic Follow-Up Study of the First National Health and Nutrition Examination Survey (NHEFS). The study population included 9,978 subjects followed from 1982-84 to 1992. Colorectal cancer cases (n = 141) were identified by self-report, hospital records, and death certificates. Interviews conducted in 1982-84 provided information on meal and snack frequencies and covariates of interest. Relative risk (RR) of colorectal cancer and 95% confidence intervals (CI) were estimated using Cox proportional hazards models adjusted for age, gender, and race and energy, alcohol, total fat, and fiber intake. Compared with those eating < 3 times/day, those eating 3-4 times/day had an RR of 0.66 (95% CI = 0.42-1.03) and those eating > 4 times/day had an RR of 0.74 (95% CI = 0.41-1.32). The association was due to decreased risk with more meals rather than more snacks per day. Given limitations of previous studies as well as the possibility of a protective effect of higher eating frequency through improved glycemic control, the present findings suggest that the influence of eating frequency on colorectal cancer risk is more complex than has been previously supposed and merits additional study.

Homicide rates among US teenagers and young adults: differences by mechanism, level of urbanization, race, and sex, 1987 through 1995.

CONTEXT: Homicide rates for persons 15 through 24 years old began to decline between 1993 and 1994, but recent trends in homicide rates by mechanism of homicide and urbanization group have not been described. OBJECTIVE: To examine homicide trends from 1987 through 1995 for persons 15 through 24 years old by urbanization level. DESIGN: Homicide rates by urbanization level were analyzed using the Compressed Mortality File, a county-level mortality and population database maintained by the National Center for Health Statistics, Centers for Disease Control and Prevention, and the rural-urban continuum codes developed by the Economic Research Service, US Department of Agriculture. SETTING: United States, 1987 through 1995, according to 5 urbanization strata: core, counties with the primary central city of a metropolitan statistical area (MSA) of 1 million or more; fringe, remaining counties within an MSA of 1 million or more; medium, counties within an MSA of 250000 to 999999; small, counties in an MSA of less than 250000; and nonmetropolitan, counties not in an MSA. SUBJECTS: All persons 15 through 24 years old by race whose cause of death was homicide (International Classification of Diseases, Ninth Revision codes E960-E969). MAIN OUTCOME MEASURES: Firearm and nonfirearm homicide rates and average annual percentage changes by 5 urbanization levels, race, and sex. RESULTS: From 1987 through 1991, the average annual firearm homicide rates among persons 15 through 24 years old among all 5 urbanization strata increased between 10.7% in small counties and 19.8% in fringe counties. From 1991 through 1993, the rates increased between 3.3% in core counties and 11.7% in small counties. From 1993 through 1995, the rates declined between 4.4% in fringe counties and 15.3% in medium counties. By 1995, firearm homicide rates among persons 15 through 24 years old ranged from 6.5 and 7.3 per 100000 in the nonmetropolitan and small counties, respectively, to 9.6 and 13.3 per 100000 in the fringe and medium strata, respectively, to 33.5 per 100000 in the core stratum. During 1987 through 1990, nonfirearm homicide rates either were stable or increased, and from 1990 through 1995, nonfirearm homicide rates declined in all 5 strata, on average 3.7% to 8.0% per year, with rates in 1995 ranging from 2.1 to 4.7 per 100000 across the strata. CONCLUSIONS: After increasing since 1987, firearm and nonfirearm homicide rates began declining between 1993 and 1995 among persons 15 through 24 years old. These declines are taking place across all urbanization strata and among white and black males and females.

Are symptoms of anxiety and depression risk factors for hypertension? Longitudinal evidence from the National Health and Nutrition Examination Survey I Epidemiologic Follow-up Study.

OBJECTIVE: To test the hypothesis that symptoms of anxiety and depression increase the risk of experiencing hypertension, using the National Health and Nutrition Examination I Epidemiologic Follow-up Study. DESIGN: A cohort of men and women without evidence of hypertension at baseline were followed up for 7 to 16 years. The association between 2 outcome measures (hypertension and treated hypertension) and baseline anxiety and depression was analyzed using Cox proportional hazards regression adjusting for hypertension risk factors (age; sex; education; cigarette smoking; body mass index; alcohol use; history of diabetes, stroke, or coronary heart disease; and baseline systolic blood pressure). Analyses were stratified by race and age (white persons aged 25-44 years and 45-64 years and black persons aged 25-64 years). SETTING: General community. PARTICIPANTS: A population-based sample of 2992 initially normotensive persons. MAIN OUTCOME MEASURES: Incident hypertension was defined as blood pressure of 160/95 mm Hg or more, or prescription of antihypertensive medications. Treated hypertension was defined as prescription of antihypertensive medications. RESULTS: In the multivariate models for whites aged 45 to 64 years, high anxiety (relative risk [RR], 1.82; 95% confidence interval [CI], 1.30-2.53) and high depression (RR, 1.80; 95% CI, 1.16-2.78) remained independent predictors of incident hypertension. The risks associated with treated hypertension were also increased for high anxiety (RR, 2.36; 95% CI, 1.73-3.23) and high depression (RR, 1.89; 95% CI, 1.25-2.85). For blacks aged 25 to 64 years, high anxiety (RR, 2.74; 95% CI, 1.35-5.53) and high depression (RR, 2.99; 95% CI, 1.41-6.33) remained independent predictors of incident hypertension. The risks associated with treated hypertension were also increased for high anxiety (RR, 3.24; 95% CI, 1.59-6.61) and high depression (RR, 2.92; 95% CI, 1.37-6.22). For whites aged 25 to 44 years, intermediate anxiety (RR, 1.62; 95% CI, 1.18-2.22) and intermediate depression (RR, 1.60; 95% CI, 1.17-2.17) remained independent predictors of treated hypertension only. CONCLUSION: Anxiety and depression are predictive of later incidence of hypertension and prescription treatment for hypertension.

Relation between residence in the southeast region of the United States and stroke incidence. The NHANES I Epidemiologic Followup Study.

For at least 50 years, stroke death rates have been higher in the southeast region of the United States than in other US regions. To test the hypotheses that stroke incidence is higher in the Southeast than in other regions and that higher levels of known stroke risk factors in the Southeast explain the difference in incidence, data were analyzed from a nationally representative, longitudinal cohort study of a sample drawn from the US population, the First National Health and Nutrition Examination Survey (NHANES I) Epidemiologic Followup Study (1971-1987). In white men and women aged 45-74 years, the risk of stroke was significantly higher in the Southeast than the Northeast or the West in men and the Midwest in women. In white men, this excess risk could not be explained by regional differences in multiple stroke risk factors (Northeast vs. Southeast risk-adjusted relative risk = 0.71, 95% confidence interval 0.52-0.98). In white women, some of the excess risk associated with residence in the Southeast compared with the Midwest could be explained by the regional differences in risk factors measured in NHANES I (Midwest vs. Southeast risk-adjusted relative risk = 0.73, 95% confidence interval 0.53-1.00). In blacks, regional differences that were statistically significant could not be demonstrated. However, a strong association of increased stroke risk with nonmetropolitan residence in blacks was demonstrated that was independent of region or other stroke risk factors. Higher stroke incidence rates in the Southeast contribute to the higher stroke mortality rates in that US region.

Serum transferrin saturation, stroke incidence, and mortality in women and men. The NHANES I Epidemiologic Followup Study. National Health and Nutrition Examination Survey.

Several studies have examined relatively large body iron stores and the risk of coronary heart disease with conflicting results. No reports of studies that associated body iron stores with stroke were found. To test the hypothesis that relatively high transferrin saturation is associated with increased stroke incidence and mortality in women and men, data from a follow-up study of a national cohort were examined. A total of 5,033 women and men aged 45-74 years from the First National Health and Nutrition Examination Survey Epidemiologic Followup Study who were free of stroke at baseline were followed an average of 12 years. Transferrin saturation (serum iron concentration divided by total iron binding capacity) was used as a measure of the amount of circulating iron available to tissues. In white women aged 45-74, after adjusting for age or for age and other risk variables, the authors observed a significant U-shaped association of transferrin saturation with risk of incident stroke (> 44% vs. 30-36%, relative risk = 1.96, 95% confidence interval 1.15-3.36; < 20% vs. 30-36%, relative risk = 1.80, 95% confidence interval 1.20-2.71). However, no significant associations were found in white men aged 45-74 after adjusting for other risk variables. Similar findings were observed for stroke mortality in whites, but no significant associations were seen in blacks. The significantly increased risk of stroke that was seen at both high and low levels of transferrin saturation in white women should be confirmed in other cohorts of women and men.

Physical activity and stroke incidence in women and men. The NHANES I Epidemiologic Follow-up Study.

To test the hypothesis that physical inactivity is associated with increased stroke risk in women and men, the authors analyzed data from a longitudinal cohort study with three follow-up data collection waves. In the Nation Health and Nutrition Examination Survey I (NHANES I) Epidemiologic Follow-up Study, 7,895 white persons and black persons aged 45-74 years were examined in 1971-1975 as part of NHANES I. Included in this analysis were 5,852 persons without a history of stroke (fatal and nonfatal) or missing data. The average follow-up was 11.6 years (maximum, 16.4 years). Incident stroke (fatal or nonfatal) was the main outcome measure. Events were ascertained from cause of death information coded from death certificates and from discharge diagnoses coded from hospital and nursing home records during the follow-up period (1971 through 1987). Participants were asked to characterize their level of habitual physical activity as low, moderate, of high. The relative risk for stroke was estimated by Cox proportional hazards regression analysis, comparing persons reporting low with those reporting high physical activity at baseline and persons in the upper with those in the lower tertile of resting pulse rate. There were 249 incident cases of stroke identified in white women, 270 in white men, and 104 in blacks. In white women aged 65-74 years, low nonrecreational activity was associated with an increased risk of stroke (relative risk = 1.82,95% confidence interval 1.10-3.02) after adjusting for the baseline risk factors of age, smoking, history of diabetes, history of heart disease, education, systolic blood pressure, serum total cholesterol, body mass index, and hemoglobin concentration. Similar associations were seen for men and for blacks and for low recreational activity in women. A higher resting pulse rate was associated with an increased risk of stroke in blacks but not in whites. A consistent association of reported low physical activity with an increased risk of stroke was observed in white women. Regular physical activity may be of benefit in preventing stroke in women as well as men.

Relation between serum albumin concentration and stroke incidence and death: the NHANES I Epidemiologic Follow-up Study.

Relatively high serum albumin levels have been associated with reduced cardiovascular mortality and coronary heart disease incidence. No prospective studies have examined serum albumin and stroke mortality and incidence. Therefore, data from the First National Health and Nutrition Examination Survey (NHANES I) Epidemiologic Follow-up Study were examined to assess serum albumin level as a risk factor for stroke. White men aged 65-74 years with serum albumin concentrations of > 4.4 g/dl had a risk of stroke incidence over a follow-up period of 9-16 years of only about two-thirds that of men with serum albumin concentrations of < 4.2 g/dl. This effect persisted after controlling for multiple stroke risk variables (relative risk = 0.61, 95% confidence interval 0.41-0.89). A similar association with stroke death was found in white men aged 65-74 years. Serum albumin was not associated with stroke risk in white women aged 65-74 years. In blacks aged 45-74 years, serum albumin concentrations of > 4.4 g/dl were associated with a risk of stroke incidence only one-half and a risk of stroke death only one-fourth that seen at levels < 4.2 g/dl after controlling other risk variables. Further studies are needed to confirm these findings and to elucidate mechanisms for the effect of serum albumin on stroke incidence and death.

Statistical issues in analyzing the NHANES I Epidemiologic Followup Study. Series 2: Data evaluation and methods research.

This report presents alternative strategies for analysis of data from the NHANES I Epidemiologic Followup Study (NHEFS) using Cox proportional hazards and person-time logistic regression models. Analytic issues related to the complex survey design of the NHANES I and the variable length of followup of NHEFS participants are discussed.

White blood cell count and stroke incidence and death. The NHANES I epidemiologic follow-up study.

A 1982 report (J Chronic Dis 1982;35:703-14) that a relatively high white blood cell (WBC) count predicted increased incidence of cerebral thrombosis could not establish whether this association was independent of smoking. Therefore, the authors examined data from the First National Health and Nutrition Examination Survey (NHANES I) Epidemiologic Follow-up Study, conducted in 1971-1987, to assess WBC count as a risk factor for stroke in a sample of the US population. White men with a WBC count of > 8,100 cells/mm3 had a 39% increase in age-adjusted stroke incidence compared with those with a WBC count of < 6,600 cells/mm3. However, controlling for cigarette smoking reduced the association and rendered it statistically nonsignificant (relative risk = 1.26, 95% confidence interval 0.93-1.70). No significant associations of WBC count with stroke incidence were seen in white women or in blacks. In white men, elevated WBC count may be a mediator of cardiovascular effects of smoking, an indicator of smoking exposure, or both. Further studies are needed to confirm these findings and to elucidate mechanisms for the effect of smoking and WBC count on stroke incidence and death.

The association between weight change and psychological well-being in women.

Previous studies on the effects of weight change on psychological well-being in clinical samples have yielded inconsistent results. We examined the relationship between weight change and psychological well-being as measured by the General Well-Being (GWB) scale in 3747 women aged 50 years or less at baseline using data from the NHANES I Epidemiologic Follow-up Study. Multiple logistic regression was used to estimate odds ratios while adjusting for potential confounders. Recent weight gain was associated with poorer well-being in both overweight and non-overweight women and recent weight loss with poorer well-being in non-overweight women. These findings were unchanged by controlling for age, race, marital status, employment status, education, physical activity level, number of medical conditions, alcohol use and extroversion. Thus, maintenance of stable weight may contribute to psychological well-being in women.

White blood cell count, coronary heart disease, and death: the NHANES I Epidemiologic Follow-up Study.

To confirm the reported association of elevated WBC count with increased risk of CHD incidence and death in white men and to determine whether such associations exist for CHD incidence and death in women and blacks, data were examined from the NHANES I Epidemiologic Follow-up Study. Over a follow-up period of 7 to 16 years, WBC counts > 8100 cells/mm3 compared with WBC counts < 6600 cells/mm3 were associated with increased risk of CHD incidence in white men (RR = 1.31; 95% CL 1.07, 1.61) and in white women (RR = 1.31; 1.05, 1.63) aged 45 to 74 after adjustment for baseline risk factors. The association was found in white female but not in white male subjects who had never smoked. Increases in risk of death from all causes, cardiovascular diseases, and noncardiovascular diseases were also seen in all white men in the sample. RRs for death for all causes at ages 45 to 74, which compared the upper and lower strata of WBC counts, were 1.43 (95% CL 1.22, 1.68) in all white men and 1.33 (95% CL 1.00, 1.78) in subjects who had never smoked after adjustment for baseline risk factors. Similar increases in risk of death from all causes were seen in blacks aged 45 to 64 despite small sample size. Thus this analysis failed to clearly establish an increased risk of CHD incidence in white men with relatively elevated WBC counts who never smoked cigarettes, although such an association was evident in white women. The increased risk of death from all causes in men appeared to be only partially due to effects of smoking.(ABSTRACT TRUNCATED AT 250 WORDS)

Leukocyte count and cardiovascular risk factors.

In view of the growing evidence for leukocyte count as an important cardiovascular risk factor, data from the first National Health and Nutrition Examination Survey (NHANES I) were examined to determine the association of leukocyte count with a number of other cardiovascular risk factors. Complete data were available for 5586 persons. Multiple linear regression analyses revealed hemoglobin concentration and height in white male nonsmokers and hemoglobin, cigarettes/day, and pulse rate in white male smokers were associated with leukocyte count. In white female nonsmokers, age, pulse rate, systolic blood pressure, height, and uric acid concentration were associated with leukocyte count. In white female smokers, hemoglobin, cigarettes/day, pulse rate, and height were associated with leukocyte count. However, associated variables explained only 8% of the variation in leukocyte count. Other measured risk factors were not associated with leukocyte count. Future analyses of leukocyte count and cardiovascular disease should control for hemoglobin concentration and pulse rate in addition to smoking, blood pressure, and cholesterol.

Firearm homicide among black teenage males in metropolitan counties. Comparison of death rates in two periods, 1983 through 1985 and 1987 through 1989.

OBJECTIVE: To identify US counties (1) that had either significantly high or significantly low firearm homicide rates among black males 15 through 19 years of age in 1983 through 1985 and in 1987 through 1989, and/or (2) that experienced a significant increase in the firearm homicide rate between 1983 through 1985 and 1987 through 1989. DESIGN: Using the Compressed Mortality File, a county-level mortality and population database maintained by the National Center for Health Statistics, Centers for Disease Control, Hyattsville, Md, county-level firearm homicide rates are analyzed. SETTING: Eighty counties with a population of at least 10,000 black males 15 through 19 years of age in 1987 through 1989. SUBJECTS: Black males 15 through 19 years of age whose underlying cause of death was classified as firearm homicide (E965.0 through E965.4, or E970) in the ICD-9 (International Statistical Classification of Diseases, Injuries, and Causes of Death, Ninth Revision). MAIN OUTCOME MEASURE: County-specific firearm homicide rate. RESULTS: In 1983 through 1985 and in 1987 through 1989, seven and 13 counties, respectively, were identified that had significantly high firearm homicide rates. Firearm homicide rates were significantly high in both time periods in the following counties: Los Angeles, California; Wayne, Michigan; Kings, New York; St Louis City, Missouri; and Baltimore City, Maryland. Firearm homicide rates increased significantly between 1983 through 1985 and 1987 through 1989 in 34 of the 80 counties. Twenty counties had significantly low rates in both time periods. Several counties with low rates in 1983 through 1985 experienced significant increases and by 1987 through 1989 were among those with high rates. CONCLUSIONS: Surveillance of firearm homicide rates at the county levels in counties with high and with low rates is a necessary first step in the development of successful violence prevention programs. Those counties where rates are high and increasing are the counties that are in greatest need for intervention strategies. Knowledge of the incidence of nonfatal firearm injuries is also needed.

Firearm and nonfirearm homicide among persons 15 through 19 years of age. Differences by level of urbanization, United States, 1979 through 1989.

OBJECTIVE: To examine trends (1979 through 1989) and current status in firearm and nonfirearm homicide rates by level of urbanization among persons 15 through 19 years of age. DESIGN: The Compressed Mortality File, a county-level mortality and population database maintained by the National Center for Health Statistics, Centers for Disease Control, Hyattsville, Md, and the 1980 Human Resource Profile County Codes are used to analyze age-, sex-, and race-specific firearm and nonfirearm homicide rates by urbanization level. SETTING: United States, 1979 through 1989. SUBJECTS: Black and white males and females 15 through 19 years of age whose underlying cause of death was either firearm homicide (E965.0 through E965.4 or E970) or nonfirearm homicide (E960 through E964, E965.5 through E969, or E971 through E978) in the ICD-9 (International Statistical Classification of Diseases, Injuries, and Causes of Death, Ninth Revision). MAIN OUTCOME MEASURES: Urbanization level-specific firearm and nonfirearm homicide rates. RESULTS: The 1989 firearm homicide rate in metropolitan counties was nearly five times the rate in nonmetropolitan counties (13.7 vs 2.9 deaths per 100,000 population). Firearm homicide rates were highest in core metropolitan counties, 27.7 per 100,000 population; rates were higher for black males than for any other race-sex group in each of five county urbanization strata for 1979 through 1989. Nonfirearm homicide rates are considerably lower, with smaller urban differentials; the rate in metropolitan counties was 1.4 times the rate in nonmetropolitan counties (2.6 vs 1.8 per 100,000 population). From 1979 through 1984, firearm homicide rates declined in each of the county strata. From 1984 through 1987, firearm homicide rates increased, and from 1987 through 1989 they increased rapidly, from 23% to 35% per year in the four metropolitan strata. From 1979 through 1989, nonfirearm homicide rates declined or remained stable. CONCLUSIONS: Large urbanization differentials in firearm homicide and smaller differentials in nonfirearm homicide are identified. Firearm homicide rates are highest and increasing the fastest among black teenage males in the core, fringe, and medium metropolitan strata.

Health service areas for the United States.

The objectives of this report are to document methods used to identify health service areas for the United States and to describe and evaluate these areas. A health service area is defined as one or more counties that are relatively self-contained with respect to the provision of routine hospital care. Service areas that include more than one county are characterized by travel between the counties for routine hospital care.

The use of health service areas for measuring provider availability.

Measurement of the availability of health care providers in a geographic area is a useful component in assessing access to health care. One of the problems associated with the county provider-to-population ratio as a measure of availability is that patients frequently travel outside their counties of residence for health care, especially those residing in nonmetropolitan counties. Thus, in measuring the number of providers per capita, it is important that the geographic unit of analysis be a health service area. We have defined health care service areas for the coterminous United States, based on 1988 Medicare data on travel patterns between counties for routine hospital care. We used hierarchical cluster analysis to group counties into 802 service areas. More than one half of the service areas include only nonmetropolitan counties. The service areas vary substantially in the availability of health care resources as measured by physicians and hospital beds per 100,000 population. For almost all of the service areas, the majority of hospital stays by area residents occur within the service area. In contrast, for 39 percent of counties, the majority of hospital stays by county residents occur outside the county. Thus, the service areas are a more appropriate geographic unit than the county for measuring the availability of health care.