This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Howard, G. Articles by Howard, V. J. Search for Related Content PubMed PubMed Citation Articles by Howard, G. Articles by Howard, V. J.

(Stroke. 1997;28:936-940.)
© 1997 American Heart Association, Inc.

Articles

Evaluation of Social Status as a Contributing Factor to the Stroke Belt Region of the United States

George Howard, DrPH; Roger Anderson, PhD; Norman J. Johnson, PhD; Paul Sorlie, PhD; Gregory Russell, MS; Virginia J. Howard, MSPH

From the Departments of Public Health Sciences (G.H., R.A., G.R.) and Neurology (G.H., V.J.H.), Bowman Gray School of Medicine, Winston-Salem, NC; the Bureau of the Census, Washington, DC (N.J.J.); and the National Heart, Lung, and Blood Institute, Bethesda, Md (P.S.).

Correspondence to Dr George Howard, DrPH, Department of Public Health Sciences, Bowman Gray School of Medicine, Medical Center Blvd, Winston-Salem, NC 27157-1063. E-mail ghoward{at}phs.bgsm.edu

	Abstract

Top Abstract Introduction Methods Results Discussion References

 
Background and Purpose The southeastern United States has stroke mortality rates above the national average. The causes for this excess mortality are unknown; however, lower socioeconomic status (SES) is a risk factor for stroke, and the lower SES in the Southeast is a potential cause. In this report we assess the proportion of the excess stroke mortality attributable to SES.

Methods The more than 400 000 participants in the National Longitudinal Mortality Study were categorized into three regions: the coastal plain region of North Carolina, South Carolina, and Georgia ("stroke buckle"); the remainder of these states plus five other southern states ("stroke belt"); and the remainder of the United States. The stroke mortality rates were calculated with and without adjustment for SES, and the proportion of the excess mortality attributable to SES was estimated.

Results In persons between the ages of 35 and 54 years, stroke mortality in the stroke buckle is estimated to be more than twice that of the rest of the nation and 1.7 times greater for ages 55 to 74 years. For persons in the stroke belt, the stroke mortality was 1.3 times greater than that in the rest of the nation for the ages of 35 to 54 and 55 to 74 years. Less than 16% of this excess stroke morality was attributable to SES.

Conclusions SES does not appear to be a major contributor to the excess mortality in the southeastern United States. Of additional concern is the stroke buckle region, which was shown to have stroke mortality rates substantially greater than those in the traditionally recognized stroke belt.

Key Words: cause of death • cerebrovascular disorders • southeastern United States • models, statistical

	Introduction

Top Abstract Introduction Methods Results Discussion References

 
While the stroke belt has been recognized since the 1960s¹ and has been present since the 1940s,² the geographic boundaries have remained ill defined and the underlying causes have remained mysterious. Perhaps the most widely accepted definition of the stroke belt is an 8- to 10-state region in the southeastern United States.^{2 3 4 5} Others investigators, however, have noted considerable variation in stroke mortality within this multistate region.^{6 7} Recently, the stroke belt was defined as a 153-county region along the coastal seaboard of North Carolina, South Carolina, and Georgia.⁸ Regardless of the definition, most investigators suggest that the stroke belt continues to exist in the southeastern region of the United States.^{3 6 8} However, it has been suggested that as part of the ongoing decline in stroke mortality,^{3 9 10} stroke mortality rates for the stroke belt region are declining more rapidly than in the rest of the United States, and the difference in rates between the stroke belt and the remainder of the United States is closing.^{11 12}

The causes underlying the existence and persistence of the stroke belt are not known. It has been suggested that the elevated risk in the stroke belt is due to a higher prevalence of risk factors, lack of access to health care, or factors associated with the geography of the region (such as water content).^{1 2 3 4 5 6 7 8} It has been convincingly argued that the stroke belt is not the result of geographic differences in stroke reporting,^{2 3 4 8} and there are reasonable arguments against most obvious causes that could give rise to the stroke belt.^{2 6 8}

Factors related to SES are associated both with stroke risk factors (including smoking, hypertension, and diabetes) and obesity.^{13 14 15 16} In addition, SES is associated with reduced access to primary care,^{17 18} which may in turn be associated with increased stroke risk through a failure to manage the "warning signs" for stroke. For these reasons, it is not surprising that SES has been associated with an increase in stroke mortality.^{19 20 21} Whether by coincidence or by cause, the southeastern region of the United States, sometimes characterized as a stroke belt, has an average SES lower than the national average (eg, the median income of each of the eight stroke belt states is lower than the US median²² ). This leads to the hypothesis that the stroke belt may be due at least in part to factors associated with SES.^{6 7 19} In addition, geographic shifts in the stroke belt have been attributed to potential geographic shifts in the relative SES of regions.⁶

Reports linking the low SES of the stroke belt region to high stroke mortality have consisted of analyses performed on an "ecological" basis; that is, the average measures of SES for a region have been correlated with the stroke mortality for the region.^{6 19} Although ecological analysis offers important insights into the patterns of causation of disease, it cannot account for the variability of individuals within a region of either SES or stroke mortality risk. Ecological analyses do not address the question of whether individuals at lower SES are at higher stroke risk. The primary focus of this report is to address this shortcoming in the literature by assessing the magnitude of excess stroke mortality in the southeastern United States and by evaluating the impact of this excess through adjustment for SES at the individual level. A secondary goal of the study is to document the relative stroke risk in the 153-county region we have defined as the stroke belt and compare the results to a more commonly used definition that encompasses at least eight states.

	Methods

Top Abstract Introduction Methods Results Discussion References

 
The NLMS²³ is a prospective study of the noninstitutionalized population of the United States constructed by combining monthly CPS²⁴ samples to form a study cohort. Each monthly CPS is a complex, national probability sample of households surveyed to obtain demographic, economic, and social information, with particular emphasis on labor force characteristics. The surveys, collected by the US Bureau of the Census for the US Bureau of Labor Statistics, are conducted by personal and telephone interview and have a response rate of close to 96%. For this study of stroke, 10 monthly samples (consisting of approximately 418 000 persons aged 35 years and older surveyed during the period 1979 through 1985) were identified and, with the CPS survey information as a baseline measurement, then followed for mortality during the years 1979 through 1989. Mortality and cause-of-death information on individuals were determined for this period by computer matching the NLMS records to the National Death Index.²⁵ For those persons identified as having died during the matching process, cause-of-death information was obtained from the death certificate and centrally coded using the ICD-9. The average follow-up time for the study cohort was approximately 8.4 years.

A portion of the information collected by the CPS consists of an assessment of each person's education and family income. These two variables are used in each analysis to account for the broad nature and influence of SES. We have grouped education in three categories: less than high school graduation, high school graduation, and some college or more. Household income, adjusted to 1980 dollars, was grouped into categories of less than $10 000 dollars, $10 000 to $19 999 dollars, and $20 000 or more. Deaths with ICD-9 codes of 430 to 438 (cerebrovascular disease) were considered stroke deaths; deaths from other causes were censored from analyses.

We divided the United States into three mutually exclusive regions (see Fig 1). The first region, the "stroke buckle," was a 153-county region influenced by the stroke mortality maps of Wing and coworkers⁷ and defined in our previous report.⁸ These counties are along the eastern seaboard of North Carolina, South Carolina, and Georgia. The second region, called the stroke belt, is defined as an eight-state region including the counties in North Carolina, South Carolina, and Georgia not included in the buckle and all counties in the states of Tennessee, Alabama, Mississippi, Louisiana, and Arkansas. The remainder of the United States is the third region. Study participants were categorized into these areas according to their county of residence as recorded in the CPS.

View larger version (69K): [in this window] [in a new window]   Figure 1. Map of US counties showing the stroke buckle (dark shading), stroke belt (light shading), and the remainder of the United States (not shaded). Note that Alaska and Hawaii are included in the remainder of the United States but are not shown in the map.

Persons under age 35 years are unlikely to die of a stroke and are excluded from these analyses. Other exclusions include persons not classified as either white or black and those missing data on education or family income (approximately 5%). After these exclusions, there were 8001 NLMS participants in the buckle region of whom 127 (1.28%) died of a stroke during follow-up, 42 927 NLMS participants in the belt region of whom 564 (1.15%) died of a stroke during follow-up, and 367 215 NLMS participants in the remainder of the United States of whom 4133 (0.90%) died of a stroke during the follow-up (Table).

View this table: [in this window] [in a new window]   Table 1. Description of Study Population Shown by Region

With an approach similar to that of previous reports,²⁰ two models were used to estimate the hazard ratio for age strata from 35 to 54, 55 to 74, and 75 years and over. The first of these models provided these estimates of the hazard ratio associated with the stroke buckle and stroke belt with adjustment for age, race, and sex but without adjustment for income and education. The later model provided estimates with adjustment for the demographic factors plus income and education. The difference in the estimated hazard ratio associated with the stroke belt or buckle before SES adjustment compared with after SES adjustment provides an estimate of the amount of the excess geographic risk attributable to differences in SES. For example, if the hazard ratio associated with the stroke buckle without adjustment for SES was 2.11 and after SES adjustment was 2.06, then approximately 3% ([(2.11-1.0)-(2.06-1.0)]/(2.11-1.0)=0.03) of the excess stroke mortality in this region is attributable to SES for this age. We have shown the magnitude of the stroke belt to be similar by race-sex strata,⁸ and we know of no report showing a differential stroke belt effect by race and sex. Nevertheless, it is possible that SES plays a differential role in the stroke belt by race and/or sex. To examine this possibility, the analysis was performed separately by race (and sex) strata, and substantial differences were not noted between results. Because the number of stroke events was relatively small for blacks (particularly in the stroke buckle), we thought it more appropriate to present the analysis after pooling and then adjusting for race and sex in the models.

	Results

Top Abstract Introduction Methods Results Discussion References

 
The Table provides a brief description of the NLMS population in each of the three study regions. No substantial differences in the mean age or percentage of males are observed among the regions. However, a graded relationship appears to exist in the proportion black and in the measures of SES in moving from the general population to the stroke belt and to the stroke buckle. For the NLMS data, the proportion of blacks in the stroke buckle population is more than four times that in the remainder of the United States. There is also an approximate 50% increase in the proportion of the population with less than a high school education and with incomes less than $10 000 in the stroke buckle region compared with the general population.

Fig 2 shows the unadjusted and SES-adjusted hazard ratio for each of the two major stroke regions by age group. For the stroke buckle, the largest increase in the risk of stroke mortality was present for the younger ages. For ages 35 to 54 years, the risk (hazard) is 2.1 times greater than in the remainder of the United States. The estimated hazard decreased to 1.7 times greater for ages 55 to 74 years and to only 1.08 times greater for residents 75 years and older. The pattern was somewhat different for the stroke belt, where the estimated hazard was approximately 1.3 times greater than in the reminder of the United States for both residents aged 35 to 54 and 55 to 74 years. As in the stroke buckle, the hazard decreased for residents 75 years and older to only 1.08 (compared with residents in the remainder of the United States).

View larger version (37K): [in this window] [in a new window]   Figure 2. Estimated hazard ratio for the stroke buckle (right bar in each age strata) and stroke belt (left bar in each age strata) relative to the remainder of the United States. Except for the stroke buckle at age 75 and over, the total height of the bar is the hazard ratio unadjusted for SES, and the portion of the bar with light shading is the hazard ratio after adjustment for SES. Hence, the dark-shaded top of the bars is the portion "explained" by SES. In the stroke buckle at age 75 and older (final right bar), the height of the light shading is the hazard without adjustment for SES, and the total bar is the risk after adjustment for SES (the dark-shaded region is the increase in risk for SES adjustment).

For both the stroke belt and buckle, adjustment for SES played a very minor role in the reduction of the estimated hazard ratio (Fig 2). In the stroke buckle, adjustment for SES reduced the estimated hazard ratio only 3% (from 2.12 to 2.06) for those aged 35 to 54 and only 7% (from 1.74 to 1.69) for those aged 55 to 74 years. In the stroke belt, adjustment for SES reduced the hazard 7% (1.27 to 1.25) for those aged 35 to 54 and 16% (from 1.31 to 1.26) for those aged 55 to 74 years. For residents aged 75 and over, the excess stroke risk without SES adjustment was only 1.08; as such, discussing the degree of reduction associated with SES adjustment is not meaningful. Hence, for ages for which there is an elevated stroke risk associated with the stroke belt or buckle, 16% or less of the variation is explained by SES adjustment.

	Discussion

Top Abstract Introduction Methods Results Discussion References

 
Finding causes that are likely not to be responsible for the stroke belt is much easier than making constructive and reasonable arguments for factors that could be responsible. The lower average SES of the southeastern United States has long been considered a possible explanatory cause for the stroke belt, with SES acting to raise risk through an association with traditional stroke risk factors or through an association between SES and access to health care.^{1 2 3 4 5 6 7 8} Hypotheses on these causes were strengthened by the recent report of Casper et al,⁶ who found that the stroke belt may be shifting westward. They noted that such shifts make it difficult to support hypotheses focusing on factors associated with the geology of the region. Casper et al appropriately suggest that investigators refocus their attention on hypotheses regarding the causes of the stroke belt, including "a variety of medical, socioeconomic, and behavioral conditions... ."⁶ Our present report suggests that SES is not likely to be a major contributor to the excess mortality associated with the stroke belt; rather than advancing the understanding of the causes of the stroke belt, this report only removes a potential "promising" cause from the list of factors associated with the excess mortality.

Other investigators have associated stroke risk in a region (county, state economic area, or state) with the average SES status of that region.^{19 21} These studies have found a strong relationship between the average SES and stroke deaths and have suggested that the stroke belt may be attributable to SES. This is the first analysis for which SES and stroke mortality data have been collected on individual subjects, and results suggest that little of the excess risk of the stroke belt region could be attributed to SES. The divergence of our results from those previously reported could have at least two explanations. The first (and in our opinion the most likely) reason for the discrepant results is the "ecological fallacy," by which analyses on groups of individuals provide misleading indications of results for individuals.²⁶ Second, the lower average SES of the stroke belt and stroke buckle could underlie a poorer level of resources available to all residents of the region (poorer health care, emergency services, etc). This poorer level of resources could impact all residents of the region, affecting both those with high and low SES in the region. If this is the case, the ecological analysis may reflect truth, while data on individuals may be misleading. A recent analysis of the NLMS data on effects of area and individual SES on all-cause mortality displayed appreciable area-SES effects that were independent of an individual's SES.²⁷ However, the risks associated with the area-SES effects were smaller in magnitude than those associated with the individual's SES.

The geographic boundaries of the stroke belt, and questions about the continued existence of the stroke belt, are a matter of debate.^{8 11 12} The stroke mortality data provided by Wing et al⁷ and information using the Geographical Information Systems analysis²⁸ both suggest that the coastal plain region of the North Carolina, South Carolina, and Georgia has a higher stroke mortality than the rest of the commonly defined stroke belt region and that this excess has persisted over time.⁸ The present study confirms this coastal plain region as having elevated stroke risk even compared with the commonly defined stroke belt region, with a stroke mortality rate twice as high as the eight-state region traditionally considered as the stroke belt. This observation should not diminish the drive to discover the causes of the stroke belt region defined as the eight southeastern states; rather, it should draw particular attention to a region meriting truly extreme concern. We propose that this region be called the "stroke buckle," compared with the eight-state region commonly defined as the "stroke belt." The consistency and magnitude of the excess stroke mortality in the stroke buckle region should draw the attention of healthcare providers and public policy administrators and motivate an even greater effort to identify the underlying causes of the excess stroke mortality in this region.

In the NLMS, there was clear evidence that the magnitude of excess risk in the stroke buckle declined with increasing age. This pattern has not been observed in vital statistics data from the National Center for Health Statistics,⁸ and we do not have an adequate explanation for why this declining pattern exists in these (but not vital statistics) data.

There are several shortcomings of these analyses. Although there are more than 400 000 participants in the NLMS, the number of stroke deaths in the stroke belt and buckle regions is not sufficient to provide highly reliable estimates by race-sex strata (ie, for black males, black females, white males, and white females). The reported analyses were adjusted for race and sex; the widely recognized effects of race and sex on stroke mortality are incorporated in the analysis. In other reports, the magnitude of the stroke belt and buckle regions has been relatively consistent across race and sex^{2 8} ; when race-sex specific analyses were performed on these data, the effects were largely supportive of the pooled data (analyses not shown). A second shortcoming is the lack of data describing traditional risk factors (smoking, hypertension, diabetes, etc) in the NLMS. Because of this absence, the role of these factors in the pathway between SES and stroke mortality could be not be assessed. In addition, while the death certificates retrieved using the National Death Index were centrally coded, the determination of cause of death was largely dependent on information available from the death certificate. Although there are acknowledged shortcomings in death-certificate data,²⁹ we feel it is unlikely that these are confounded by either geography or SES. Finally, we have used family income and education as indexes of SES. While these are widely used indexes, they do not fully reflect the complexity of SES. It is possible that if this complexity were more fully reflected in the analyses, more of the excess mortality in the stroke belt could be attributed to SES.

In conclusion, the existence of the stroke belt was confirmed in these prospective data, and the existence of a region in the coastal plain of North Carolina, South Carolina, and Georgia with even higher stroke mortality (the "stroke buckle") was shown. In both the stroke belt and stroke buckle, SES accounted for only a small portion of the excess stroke mortality; hence, while an excess of stroke mortality in the southeastern United States continues to be observed, there is little understanding of its causes.

	Selected Abbreviations and Acronyms

 

CPS = Current Population Survey ICD-9 = International Classification of Diseases, 9th Revision NLMS = National Longitudinal Mortality Study SES = socioeconomic status

	Acknowledgments

 
This study was supported by National Institutes of Health–National Heart Lung and Blood Institute grant R03-53339.

Received February 3, 1997; revision received March 5, 1997; accepted March 5, 1997.

	References

Top Abstract Introduction Methods Results Discussion References

 

1. Borhani NO. Changes and geographic distribution of mortality from cerebrovascular disease. Am J Public Health. 1965;55:673-681.
2. Lanska DJ. Geographic distribution of stroke mortality in the United States: 1939-1941 to 1979-1981. Neurology. 1993;43:1839-1851.[Abstract/Free Full Text]
3. Feinleib M, Ingster L, Rosenberg H, Maurer J, Sing G, Kochanek K. Time trends, cohort effects, and geographic patterns in stroke mortality: United States. Ann Epidemiol. 1993;3:458-465.[Medline] [Order article via Infotrieve]
4. Lanska DJ, Kryscio R. Geographic distribution of hospitalization rates, case fatality, and mortality from stroke in the United States. Neurology. 1994;44:1541-1550.[Abstract/Free Full Text]
5. Miller JP, Perry HM Jr, Rossiter JE, Baty JD, Carmody SE, Sambhi MP. Regional differences in mortality during 15-year follow-up of 11,936 hypertensive veterans. Hypertension. 1994;23:431-438.[Abstract/Free Full Text]
6. Casper ML, Wing S, Anda RF, Knowles M, Pollard RA. The shifting stroke belt: changes in the geographic pattern of stroke mortality in the United States, 1962 to 1988. Stroke. 1995;26:755-760.[Abstract/Free Full Text]
7. Wing S, Casper M, Davis WB, Pellom A, Riggan W, Tyroler HA. Stroke mortality maps: United States whites aged 35-74 years, 1962-1982. Stroke. 1988;19:1506-1513.
8. Howard G, Evans GW, Pearce K, Howard VJ, Bell RA, Mayer EJ, Burke GL. Is the stroke belt disappearing? An analysis of racial, temporal, and age effects. Stroke. 1995;26:1153-1158.[Abstract/Free Full Text]
9. Whisnant JP. The decline of stroke. Stroke. 1984;15:160-168.[Abstract/Free Full Text]
10. Cooper R, Sempos C, Hsieh SC, Kovar MG. Slowdown in the decline of stroke mortality in the United States, 1978-1986. Stroke. 1990;21:1274-1279.[Abstract/Free Full Text]
11. Lanska DJ, Peterson PM. Comparison of additive and multiplicative models of regional variation in the decline of stroke mortality in the United States. Stroke. 1996;27:1055-1059.[Abstract/Free Full Text]
12. Lanska DJ, Peterson PM. Geographic variation in the decline of stroke mortality in the United States. Stroke. 1995;26:1159-1165.[Abstract/Free Full Text]
13. Gliksman MD, Kawachi I, Hunter D, Colditz GA, Manson JE, Stampfer MJ, Speizer FE, Willett WC, Hennekens CH. Childhood socioeconomic status and risk of cardiovascular disease in middle aged US women: a prospective study. J Epidemiol Community Health. 1995;49:10-15.[Abstract]
14. Kaplan GA, Keil JE. Socioeconomic factors and cardiovascular disease: a review of the literature. Circulation. 1993;88:1973-1993.[Abstract/Free Full Text]
15. Unwin N, Binns D, Elliott K, Kelly WF. The relationship between cardiovascular risk factors and socio-economic status in people with diabetes. Diabet Med. 1996;13:72-79.[Medline] [Order article via Infotrieve]
16. Lynch J, Kaplan GA, Salonen R, Cohen RD, Salonen JT. Socioeconomic status and carotid atherosclerosis. Circulation. 1995;92:1786-1792.[Abstract/Free Full Text]
17. Gornick ME, Eggers PW, Reilly TW, Mentnech RM, Fitterman LK, Kucken LE, Vladeck BC. Effects of race and income on mortality and use of services among Medicare beneficiaries. N Engl J Med. 1996;335:791-799.[Abstract/Free Full Text]
18. Adler NE, Boyce WT, Chesney MA, Folkman S, Syme SL. Socioeconomic inequalities in health: no easy solution. JAMA. 1993;269:3140-3145.[Abstract]
19. Casper M, Wing S, Strogatz D. Variation in the magnitude of black-white differences in stroke mortality by community occupation structure. J Epidemiol Community Health.. 1991;45:302-306.[Abstract]
20. Howard G, Russell GB, Anderson R, Evans GW, Morgan T, Howard VJ, Burke GL. Role of social class in excess stroke mortality. Stroke. 1995;26:1759-1763.[Abstract/Free Full Text]
21. Modan B, Wagener DK. Some epidemiological aspects of stroke: mortality/morbidity trends, age, sex, race, and socioeconomic status. Stroke. 1992;23:1230-1236.[Abstract/Free Full Text]
22. Statistical Abstract of the United States: 1993. 113th ed. Washington, DC: US Bureau of the Census; 1993.
23. Howard G, Anderson R, Sorlie P, Andrews V, Backlund E, Burke GL. Ethnic differences in stroke mortality between non-Hispanic whites, Hispanics, and blacks: the National Longitudinal Mortality Study. Stroke. 1994;25:2120-2125.[Abstract]
24. Current Population Survey: Design and Methodology. Washington DC: US Bureau of the Census; 1978. Technical paper 40.
25. National Center for Health Statistics. User's Manual: the National Death Index. Washington, DC: Public Health Service, US Government Printing Office: 1990. US Dept of Health and Human Services publication PHS 90-1148.
26. Kleinbaum DG, Kupper LL, Morgenstern H. Epidemiologic Research: Principals and Quantitative Methods. London, UK: Lifetime Learning Publications; 1982.
27. Anderson RT, Sorlie P, Backlund E, Johnson N, Kaplan GA. Mortality effects of community socioeconomic status. Epidemiology. 1997;8:42-47.[Medline] [Order article via Infotrieve]
28. Howard G, Howard VJ, Evans GW, Burke GL. A new approach to assess geographical patterns in stroke mortality. Stroke. 1996;27:165. Abstract.
29. Kelsey JL, Thompson WD, Evans AS. Methods in Observational Epidemiology. New York, NY: Oxford Press; 1986.

This article has been cited by other articles:

				R. Reklaitiene, V. Janilionis, M. Noreika, A. Tamosiunas, D. Virviciute, and D. Sopagiene Effects of age, period and cohort on stroke mortality among a middle-aged Lithuanian urban population from 1980 to 2004 Scand J Public Health, August 1, 2008; 36(6): 573 - 579. [Abstract] [PDF]

				M. M. Glymour, M. Avendano, and L. F. Berkman Is the 'Stroke Belt' Worn From Childhood?: Risk of First Stroke and State of Residence in Childhood and Adulthood Stroke, September 1, 2007; 38(9): 2415 - 2421. [Abstract] [Full Text] [PDF]

				H. Kelley and K. Betsalel Travels in Stroke Country: An Ethnographic Photo Essay Journal of Contemporary Ethnography, April 1, 2005; 34(2): 170 - 179. [Abstract] [PDF]

				L. B. Goldstein Editorial Comment--Prenatal Influences on Stroke Mortality in England and Wales Stroke, July 1, 2003; 34(7): 1602 - 1603. [Full Text] [PDF]

				G. Howard, V. J. Howard, C. Katholi, M. K. Oli, S. Huston, and K. Asplund Decline in US Stroke Mortality: An Analysis of Temporal Patterns by Sex, Race, and Geographic Region Editorial Comment: An Analysis of Temporal Patterns by Sex, Race, and Geographic Region Stroke, October 1, 2001; 32(10): 2213 - 2220. [Abstract] [Full Text] [PDF]

				D. Jakovljevic, C. Sarti, J. Sivenius, J. Torppa, M. Mahonen, P. Immonen-Raiha, E. Kaarsalo, K. Alhainen, K. Kuulasmaa, J. Tuomilehto, et al. Socioeconomic Status and Ischemic Stroke : The FINMONICA Stroke Register Stroke, July 1, 2001; 32(7): 1492 - 1498. [Abstract] [Full Text] [PDF]

				R. Grieve, J. Hutton, A. Bhalla, D. Rastenyte, D. Ryglewicz, C. Sarti, M. Lamassa, M. Giroud, R. Dundas, and C.D.A. Wolfe A Comparison of the Costs and Survival of Hospital-Admitted Stroke Patients Across Europe Stroke, July 1, 2001; 32(7): 1684 - 1691. [Abstract] [Full Text] [PDF]

				W. D. Rosamond, A. R. Folsom, L. E. Chambless, C.-H. Wang, P. G. McGovern, G. Howard, L. S. Copper, and E. Shahar Stroke Incidence and Survival Among Middle-Aged Adults : 9-Year Follow-Up of the Atherosclerosis Risk in Communities (ARIC) Cohort Stroke, April 1, 1999; 30(4): 736 - 743. [Abstract] [Full Text] [PDF]

				C. T. M. van Rossum, H. van de Mheen, M. M. B. Breteler, D. E. Grobbee, and J. P. Mackenbach Socioeconomic Differences in Stroke Among Dutch Elderly Women : The Rotterdam Study Stroke, February 1, 1999; 30(2): 357 - 362. [Abstract] [Full Text] [PDF]

				K. C. Johnston, J. Y. Li, P. D. Lyden, S. K. Hanson, T. E. Feasby, R. J. Adams, R. E. Faught Jr, and E. C. Haley Jr Medical and Neurological Complications of Ischemic Stroke : Experience From the RANTTAS Trial Stroke, February 1, 1998; 29(2): 447 - 453. [Abstract] [Full Text] [PDF]

This Article Abstract Alert me when this article is cited Alert me if a correction is posted Citation Map Services Email this article to a friend Similar articles in this journal Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Request Permissions Citing Articles Citing Articles via HighWire Citing Articles via Google Scholar Google Scholar Articles by Howard, G. Articles by Howard, V. J. Search for Related Content PubMed PubMed Citation Articles by Howard, G. Articles by Howard, V. J.