Donate Help Contact The AHA Sign In Home
American Heart Association
Stroke
Search: search_blue_button Advanced Search
« Previous Article | Table of Contents | Next Article »
Stroke. 1995;26:554-561

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

(Stroke. 1995;26:554-561.)
© 1995 American Heart Association, Inc.

Articles

Effects of Interstate Migration on the Geographic Distribution of Stroke Mortality in the United States

Douglas J. Lanska, MD, MS Patrick M. Peterson, MS, MA

From the Departments of Neurology (D.J.L., P.M.P.), Preventive Medicine and Environmental Health (D.J.L.), and Statistics (P.M.P.), and the Sanders Brown Center on Aging (D.J.L.), University of Kentucky Medical Center, and the Neurology Service (D.J.L.), Veterans Affairs Medical Center, Lexington, Kentucky.

Correspondence to Douglas J. Lanska, MD, Department of Neurology, Room E124, Kentucky Clinic, University of Kentucky, Lexington, KY 40536-0284.


*    Abstract
up arrowTop
 *Abstract
down arrowIntroduction
 down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Background and Purpose This study examines the effects of lifetime net interstate migration on the geographic distribution of stroke mortality in the United States.

Methods National Center for Health Statistics and Bureau of the Census data were used to map the geographic distribution of age-adjusted, race-, and race/sex-specific stroke mortality rates by interstate migration status for natives, outmigrants, nonmigrants, inmigrants, and residents in the United States for 1979 to 1981.

Results High age-adjusted stroke mortality rates were significantly clustered in the southeastern United States for both whites and blacks; in addition, for whites, low-rate states were concentrated in some Mountain and northeastern states. Migrant status did not change this large-scale pattern, but individual states showed significant migration effects, which varied in magnitude and direction. Among whites, states that benefited from migration, with markedly lower stroke mortality rates among residents than natives, included Arizona, Colorado, District of Columbia, and Florida, whereas states that suffered from migration included California, Idaho, Montana, North Dakota, Nevada, and Oklahoma. Among blacks, only Colorado showed an apparent large benefit from migration, whereas 21 states suffered from migration.

Conclusions Although the overall large-scale spatial distribution of resident stroke mortality rates cannot be explained by migration effects, some individual states had rates that were strongly influenced by migration. Patterns of mortality among migrant groups in Sun Belt retirement destination states probably result from differential selection effects for retirement migration in older adults. Patterns of mortality for black migrants to the North are probably influenced by "carryover" effects from their origin states.


Key Words: cerebrovascular disorders • demography • epidemiology • risk factors


*    Introduction
up arrowTop
 up arrowAbstract
 *Introduction
down arrowMethods
 down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Since at least 1940 there has been a consistent pattern of marked geographic variation in stroke mortality rates within the United States.1 2 3 Very high rates are reported in the southeast coastal plain, an area designated the "Stroke Belt," while very low rates are reported in the Mountain census division.1 2 3 These general patterns of geographic variation have been observed for both sexes and for whites and nonwhites, although stroke rates have been consistently declining in all geographic areas of the continental United States during this interval.1 The causes of this marked regional variation are unknown, but some aspects of the geographic distribution of stroke mortality may be due to selective migration, particularly for retirement destination states.1 2 This study examines the effect of lifetime net interstate migration on the geographic distribution of stroke mortality in the United States.


*    Methods
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 *Methods
down arrowResults
 down arrowDiscussion
 down arrowReferences
 
Vital Statistics Mortality Data
Data on deaths with cerebrovascular disease listed as an underlying or contributing cause of death were obtained from public-use, multiple-cause-of-death, machine-readable, US mortality data files created by the National Center for Health Statistics (NCHS).4 5 6 Data were tabulated for the 3-year period centered on the 1980 decennial census year for white and black persons whose birthplace and residence at death were in the coterminous United States. Studies on the validity of state of birth coding on death certificates are not available, but this information is common knowledge within families, and any errors are unlikely to produce significant systematic geographic variation in rates calculated by state of birth. Stroke deaths were defined as those coded to rubrics 430 through 438 according to the International Classification of Diseases, 9th Revision, Clinical Modification (ICD-9-CM).7

For multiple-cause tabulations, "record-axis" codes were used: These codes represent the best person-level description of all conditions listed in the medical certification portion of the death certificate. NCHS derived record-axis codes from "entity-axis codes" of the individual conditions reported with the use of traditional linkage and modification rules for mortality coding; each entity-axis code was examined and modified as appropriate to create a set of record-axis codes that are free of contradictions and duplications and are the most precise within the constraints of ICD-9-CM and the medical information on the record.8

Census Data
The racial classification used for data collected in the 1980 census differs from that used for vital statistics data and for data collected in previous censuses. Racial counts in the 1980 census were particularly affected by changes in reporting among the Hispanic population and by changes in coding and classification of racial groups. In censuses before 1980, virtually all persons of Hispanic origin were considered white, whereas in the 1980 census almost 6.7 million people, mostly of Hispanic origin, were reported in a residual "other (not specified)" category.9 Although vital statistics data do not include the residual category, routine 1980 census publications and summary tape files present unmodified data using the categories in which the data were collected.

To maintain comparability with the racial designations used in the vital statistics data and to obtain data on lifetime net interstate migration, it was necessary to use individual-level census data and modify the self-reported racial categories. For this purpose we used public-use microdata sample A (PUMS-A) for the 1980 census, which is an edited abstract of the census basic record tapes for a stratified random subsample of the full census sample that received long-form questionnaires. PUMS-A represents 5% of the US population and contains census records for more than 11 million persons.10 11

Before modification, the PUMS-A population estimates for whites were low (sometimes markedly so) compared with Census Bureau modified-race estimates9 12 in states with large Hispanic populations. As an approximation of the complex racial category modification algorithm used by the Bureau of the Census for the preparation of special modified-race summary data files,13 all individuals of Hispanic origin who were originally in the "other (not specified)" race category were reassigned to a modified white race group. After modification, all of the state population estimates for whites were within 1% of the Census Bureau modified-race estimates.9 12 Tables of state population totals by race are available through the National Auxiliary Publication Service (NAPS) for the original PUMS-A estimates and for two different modification approaches tried.

No modification of black census counts was performed for these analyses. For blacks, 14 states had very small black populations (<50 000) and are not considered further: Idaho, Iowa, Maine, Montana, Nebraska, New Hampshire, New Mexico, North Dakota, Oregon, Rhode Island, South Dakota, Utah, Vermont, and Wyoming. Three states had low black population estimates compared with Census Bureau modified-race estimates, with the percentage differences as follows: Massachusetts, 7.4%; Minnesota, 3.7%; and New York, 2.5%. Data presented below for blacks for these states are subject to potential bias due to likely distortions in the population estimates. Eight additional states had black population estimates that differed from Census Bureau modified-race estimates9 12 by 1% to 2%, but six of these were overestimates.

Coverage of the white population in the 1980 census was fairly complete, but blacks showed a net 5% omission rate.14 The vast majority of omissions among blacks, however, occurred among children aged younger than 5 years and males aged 20 to 54 years.14 As such, this undercount is unlikely to significantly distort age-adjusted stroke mortality rates, which typically receive their greatest contribution from older age groups. Regional data on coverage are not available.

Analyses
Lifetime net interstate migration was determined by a difference between state of birth and current place of residence for census data or by a difference between state of birth and state of residence at death for mortality data. For those whose state of residence at birth differed from their current residence or their residence at death, this is a valid indication of migration. The age at which migration took place, the influence of disease on migration, and the frequency and duration of intervening migrations were not available. Also, some migrants may have returned to their birthplace and would subsequently be designated as "nonmigrants."

We calculated age-, race-, and sex-specific rates by state for each migrant group using the appropriate migrant group population as the denominator. Average annual, age-adjusted, race-specific rates (per 100 000 population) were computed by the direct method, that is, by applying the age-specific death rates for stroke to the standard population distributed by age. The reference population used was the total US population enumerated on April 1, 1980. Statistical significance of differences in age-adjusted rates was determined according to formulas given by Kahn and Sempos.15 Rates for populations less than 50 000 are not shown.

Spatial autocorrelation of age-adjusted rates was assessed with the use of Moran's I spatial autocorrelation coefficient.16 The expected value of I for a random arrangement is of small magnitude and negative; positive values of I imply clustering, and extreme negative values imply dispersion. Since the range of possible values of I depends on the spatial structure of a particular study region, significance testing was performed under the randomization hypothesis.17 Given the specific observed values of age-adjusted mortality rates, the probability that they could have been arranged in the observed way by chance is determined by calculating a standard normal deviate from the calculated value, the expected value, and the standard deviation.


*    Results
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 *Results
down arrowDiscussion
 down arrowReferences
 
Separate maps display the geographic variation of mortality rates at a state level for each of the following: natives (nonmigrants and outmigrants), outmigrants, nonmigrants, inmigrants, and residents (nonmigrants and inmigrants). Underlying-cause mortality rate maps for whites are shown in Figs 1 through 5DownDownDownDownDown. Maps of underlying-cause mortality rates for blacks, underlying-cause mortality rates by race/sex group, and multiple-cause mortality rates by race and race/sex group are not shown but are available from NAPS. All of the maps for whites illustrate intense clustering of high stroke mortality rates in the southeastern United States (P<<.0001), either for underlying-cause rates (Figs 1 through 5DownDownDownDownDown) or multiple-cause rates. The spatial patterns for the multiple-cause rates were similar to the patterns for the underlying-cause rates, but the highest multiple-cause rates occurred almost exclusively east of the Mississippi River. The maps for blacks are more sparse as a result of small population sizes for many of the state-level migrant groups. Nevertheless, all of the maps for blacks also show statistically significant clustering of high stroke mortality rates in the southeastern United States.



View larger version (0K):
[in this window]
[in a new window]
  		Figure 1. Map of average annual, age-adjusted, underlying-cause-of-death, state stroke mortality rates for native whites for 1979 to 1981. Natives are defined as persons born in the state. District of Columbia (not shown) had a rate of 70.4/100 000 population per year. There is significant clustering (P<<.0001) of rates, with high rates in the Southeast and low rates in the West and Northeast. Note that the rate for Florida natives is similar to the rates of other states in the Southeast.



View larger version (0K):
[in this window]
[in a new window]
  		Figure 2. Map of average annual, age-adjusted, underlying-cause-of-death, state stroke mortality rates for outmigrant whites for 1979 to 1981. District of Columbia (not shown) had a rate of 71.4/100 000 population per year. There is significant clustering (P<<.0001) of rates, with high rates in the Southeast and low rates in the West and Northeast. The pattern of rates for outmigrants largely parallels the pattern for natives (compare Fig 1Up).



View larger version (0K):
[in this window]
[in a new window]
  		Figure 3. Map of average annual, age-adjusted, underlying-cause-of-death, state stroke mortality rates for nonmigrant whites for 1979 to 1981. Nonmigrants are defined as persons born and currently residing in the state. District of Columbia (not shown) had a nonmigrant population of less than 50 000. There is significant clustering (P<<.0001) of rates, with high rates in the Southeast and low rates in the West and Northeast. The pattern of rates for nonmigrants largely parallels the pattern for natives (compare Fig 1Up).



View larger version (0K):
[in this window]
[in a new window]
  		Figure 4. Map of average annual, age-adjusted, underlying-cause-of-death, state stroke mortality rates for inmigrant whites for 1979 to 1981. District of Columbia (not shown) had a rate of 56.8/100 000 population per year. There is significant clustering (P<<.0001) of rates, with high rates in the Southeast, West, and Northwest and low rates in the southeastern Mountain states and the Northeast. Note particularly the extremely low rate for inmigrants to Florida and the high rates for inmigrants to some of the western states.



View larger version (0K):
[in this window]
[in a new window]
  		Figure 5. Map of average annual, age-adjusted, underlying-cause-of-death, state stroke mortality rates for resident whites for 1979 to 1981. District of Columbia (not shown) had a rate of 58.5/100 000 population per year. There is significant clustering (P<<.0001) of rates, with high rates in the Southeast and low rates in the southern Mountain states, parts of the Midwest, and the Northeast. Note the low rate for Florida, which parallels the extremely low rate among inmigrants (compare Fig 4Up); contrast this with the high rate among Florida natives and nonmigrants (compare Figs 1Up and 3Up). Note also the relatively high rates in many of the western states, which parallel the rates for inmigrants to these states (compare Fig 4Up); contrast these with the low rates among natives, outmigrants, and nonmigrants (compare Figs 1 through 3UpUpUp).

To judge the overall effect of inmigration and outmigration on state stroke mortality rates, we compared age-adjusted, underlying-cause rates for natives and residents (Figs 1Up and 5Up; an additional table available from NAPS lists rates for each migrant group by state, race, and sex, along with statistical significance of rate differences between migrant and nonmigrant populations).

Among whites, states that apparently benefited most from migration (ie, resident underlying-cause rates <90% of native rates) included the following (with native and resident rates listed, respectively, in parentheses): Arizona (63.4 versus 56.2), Colorado (68.2 versus 60.7), District of Columbia (70.4 versus 58.5), and Florida (72.2 versus 64.5). Florida benefited from an influx of persons with significantly lower stroke rates than natives. In Colorado, inmigrant rates were similar to nonmigrant rates, but outmigrant rates were significantly higher than nonmigrant rates. District of Columbia has an extremely small nonmigrant white population (25 900 in 1980), making comparisons between migrants and nonmigrants problematic, but outmigrants had higher rates than inmigrants (P=.019). Differences between migrant and nonmigrant populations for Arizona were not significant.

Among whites, states that apparently suffered the most from migration (ie, resident underlying-cause rates >10% greater than native rates) included the following (with native and resident rates listed, respectively, in parentheses): California (68.2 versus 75.1), Idaho (55.8 versus 73.0), Montana (60.2 versus 74.0), North Dakota (62.1 versus 68.9), Nevada (35.6 versus 71.9), and Oklahoma (69.2 versus 81.3). California, Idaho, and Oklahoma suffered from an influx of persons with significantly higher stroke rates. In Nevada, outmigrants had significantly lower rates than nonmigrants, while the rates for inmigrants and nonmigrants were almost identical. The differences between migrants and nonmigrants for Montana and North Dakota were not significant.

Among blacks, only one state—Colorado—apparently benefited markedly from migration (ie, resident underlying-cause rates <90% of native rate), with the native and resident rates being 108.6 and 81.3, respectively. In contrast, 21 states suffered markedly from migration (ie, resident underlying-cause rates >10% greater than native rates). These states were (with the native and resident rates listed, respectively, in parentheses) Arkansas (111.7 versus 133.2), California (45.2 versus 98.0), Connecticut (52.9 versus 74.6), Florida (107.6 versus 121.5), Georgia (124.7 versus 140.4), Illinois (65.1 versus 87.7), Indiana (89.4 versus 110.3), Kansas (83.0 versus 105.4), Massachusetts (62.1 versus 71.9), Michigan (39.2 versus 96.1), North Carolina (111.3 versus 126.7), New Jersey (69.5 versus 94.4), New York (53.2 versus 77.5), Ohio (79.6 versus 90.6), Oklahoma (96.1 versus 106.4), Pennsylvania (69.6 versus 87.8), South Carolina (125.5 versus 139.8), Tennessee (114.8 versus 131.7), Virginia (106.4 versus 123.2), West Virginia (83.5 versus 105.2), and Washington (25.7 versus 100.4). Eight states suffered as a result of an influx of persons with significantly higher stroke rates than natives: Arkansas, California, Illinois, Michigan, New Jersey, New York, Pennsylvania, and South Carolina. Outmigrants from Florida, Georgia, North Carolina, Tennessee, and Virginia had significantly lower rates than nonmigrants. In the other states, differences between migrants and natives were not significant (in some because of small population sizes among some migrant groups).

Migration had significant effects on three of the four principal Sun Belt retirement destination states (ie, Florida, Arizona, and California) as detailed above. In contrast, migration had little effect on rates for Texas, currently the fourth leading destination state for elderly interstate migrants.18 Underlying-cause stroke mortality rates for white residents and natives of Texas were almost identical (79.4 versus 79.1 per 100 000 population per year, respectively), and rates for white migrants to and from Texas did not differ from rates for white nonmigrants. Rates for black residents and natives of Texas were also similar (111.6 versus 108.1, respectively).

In some situations not detailed here, migrants differed significantly from nonmigrants of either the origin or destination state, but there was little if any difference in rates for natives and residents of those states (see additional tables filed with NAPS); in most of these situations, either the migrant groups were too small to influence the overall state rates, or the effect of inmigrants was counterbalanced by an opposite effect of outmigrants.


*    Discussion
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 *Discussion
down arrowReferences
 
The nonrandom distribution of stroke mortality across the United States, the large magnitude of the difference between high- and low-rate areas, the persistence of the pattern over more than 50 years, the similarity of the distribution for different race/sex groups, and lack of delimitation by administrative or political boundaries suggest that the pattern of excess stroke mortality is not an artifact of different diagnostic and reporting practices.1 2 Moreover, in the 1960s national cooperative studies confirmed the apparent large differences in stroke mortality rates among geographic areas in the United States19 20 ; these large variations in mortality rates could not be explained by differences in certification practices (such as choice of underlying cause of death when multiple causes contributed to death), the frequency that clinical stroke diagnoses were listed on certificates, differences in accuracy of stroke diagnosis, or variations in the standards of medical care.19 20 21 22 23 Furthermore, the large regional differences parallel geographic differences in stroke incidence24 and hospital utilization for stroke.2 25

The present results indicate that the overall large-scale spatial distribution of (resident) stroke mortality rates cannot be explained by migration effects. Nevertheless, migration effects vary considerably among states in magnitude and direction. The effects of migration on state rates are complex and probably depend on several factors, including size and age distribution of the migrant group, selection factors for migration into and out of various states, risk-modifying influences at origin and destination, and possibly selective mortality in various migrant and nonmigrant groups. While the ultimate determination of the relative contributions of these various factors must await studies that examine specific state-to-state migration streams and account for differences in socioeconomic and health status, available information can explain many of the observed differences among various migrant and nonmigrant groups.

Selection Effects: Retirement Migration to the Sun Belt
Selection effects on migrant mortality rates occur when outmigrants are differentially selected from among natives on some basis related to health, producing rates that are either lower or higher than for nonmigrants, independent of the destination. Selection for positive health status occurs particularly when migration is determined by positive factors at the destination, whereas selection for negative health status occurs particularly when migration is determined by negative factors at the place of origin.26

Although the elderly represent a relatively small proportion of all interstate migrants,27 28 29 elderly migrants have a large effect on stroke mortality rates, particularly in some retirement destination states, because elderly migrants have markedly higher stroke mortality rates than younger groups, because migration in the elderly is strongly associated with health status, and because elderly migration patterns are so strongly channelized. Elderly interstate migration generally differentially selects independent, healthy individuals in the upper income and educational ranks who are seeking amenities at the retirement destination; in contrast, local mobility differentially selects more dependent elderly, many of whom are moving because of a decline in independence or health status.18 30 31 32 33 34 35 36 37 Elderly interstate retirement migration is also remarkably channelized: In the 1970 and 1980 censuses, elderly interstate migration to Florida, California, Arizona, and Texas represented nearly half of all elderly interstate migration in the two preceding 5-year periods.27 35 38 In all four of these states there was net inmigration of the elderly, and in Florida and Arizona elderly retirement inmigration overwhelmed the usually much larger "labor force peak" in the migration schedules, producing dominant "retirement peaks."27 Except for a high frequency of return migration to Texas, very few of these elderly persons migrating to the principal Sun Belt retirement states were returning to their state of birth.18 39 41 However, return migration from the Sun Belt to the "Snow Belt" may contribute to the selectivity of lifetime net interstate migration to the Sun Belt, since return migration of the elderly to their state of birth is often associated with a decline in health status ("mobility for assistance").26 28 40 42 43 44 45

For more than four decades Florida has been the leading destination of elderly interstate migrants, receiving approximately a quarter of all elderly migrants.27 29 36 38 Inmigrants to Florida make up a phenomenally high proportion (approximately 93%) of the resident white elderly population of that state in the 1980 census,46 with recent inmigrants (within the previous 5 years) accounting for approximately 20% of the elderly population.36 These inmigrants are almost exclusively white, and a large majority are healthy, independent, well educated, and financially secure.18 29 36 45 47

Based on analyses of state stroke mortality rates from 1939-1941 to 1979-1981, Lanska1 suggested that patterns of elderly interstate migration may strongly influence rates in Florida, and possibly other retirement destination states, as a result of migration-associated selection by health status. Lanska found that stroke mortality rates for white residents of Florida became significantly lower than the national rate after World War II and have remained low, in contrast to persistently significantly high rates among black residents of Florida and in contrast to persistently high rates among both whites and blacks in other South Atlantic states during this period.1

The present study confirms the previous prediction1 that migration effects have strongly influenced the observed stroke mortality rates among residents of Florida. Stroke mortality rates among white residents of Florida were 11% less than those for natives of the state. This difference was due to an influx of persons with significantly lower stroke rates than nonmigrants, whereas rates for outmigrants and nonmigrants did not differ. In contrast, stroke mortality rates among black residents of Florida were 13% greater than those for natives of the state. This difference was due in part to an exodus of persons with significantly lower stroke rates than nonmigrants.

Migration-associated selection by health status among white migrants to Florida may also be responsible for the lower than expected rate of hospitalization for stroke in Florida. The stroke hospitalization rate for Florida in 1989 was 0.84/100 Medicare eligibles—only 91% of the predicted value based on the age, sex, race, and socioeconomic distributions of the Florida Medicare population2 ; in contrast, almost all other states in the East South Central and South Atlantic census divisions had significantly higher hospitalization rates than predicted.2

It is not yet possible to fully exclude an origin effect as a contributor to observed rate differences between inmigrants to Florida and nonmigrant Floridians, although a large effect seems unlikely. The origins of most white inmigrants to Florida are the East North Central, Middle Atlantic, and southern New England states.35 36 New York contributes the largest share of elderly migrants to Florida (29% during 1975 to 1980),27 35 and New York has consistently had stroke mortality rates significantly lower than both the national rate1 and rates for nonmigrant Floridians. However, rates for other major migrant contribution states have been mixed,1 and as a group these states are unlikely to have produced a large origin effect. A destination effect is also unlikely to account for these data for whites given that rates for white inmigrants were both significantly lower than rates for nonmigrant white Floridians and lower than rates for whites (or white nonmigrants) in most principal origin states (see tables filed with NAPS).1

Similar to the situation in Florida, interstate inmigrants to Arizona constituted a phenomenally high proportion (approximately 94%) of the resident white elderly population of that state in the 1980 census, with recent inmigrants (within the previous 5 years) accounting for more than a fifth of the elderly population.34 38 These inmigrants are a highly selected group: almost exclusively white and with a large majority healthy, independent, well educated, and financially secure.18 29 35 36 45 47 In contrast to the situation in Florida, however, elderly inmigrants to Arizona have been similar to the in-place elderly population in various demographic, socioeconomic, and housing quality indexes,32 35 as well as in their stroke mortality rates. Arizona's elderly inmigrants originated primarily from the Midwest,18 with particularly heavy flows from Michigan and Illinois32 ; these states have had average or significantly high stroke mortality rates for whites in recent decades,1 and therefore an origin effect cannot explain the very low rates among migrants to Arizona, but a destination effect contribution cannot be excluded.

Although elderly interstate migration to California has progressively declined for three decades, California still ranks second to Florida in attracting elderly interstate migrants.31 35 However, unlike the selection for positive health status in Florida and Arizona, elderly inmigrants to California are selected for negative health status,18 32 35 a situation reflected in higher rates of dependency, disability, institutionalization, and stroke mortality among inmigrants to California compared with nonmigrants. Neither an origin nor a destination effect is likely to account for these data, at least for whites, given that stroke mortality rates for white inmigrants were both significantly higher than rates for nonmigrant white Californians and higher than rates for whites (or white nonmigrants) in almost all of the major origin states.1 35

Although Texas was the fourth leading destination of elderly interstate migrants in 1980 and the fifth leading destination in 1970,27 38 for several reasons the impact of migration is much lower than for Florida, California, and Arizona. First, elderly lifetime interstate inmigrants make up a much smaller percentage (35%) of the elderly population of Texas than they do for the more popular Sun Belt retirement destination states. In addition, Texas ranks only 24th among states in the percentage of its elderly who were recent inmigrants.38 These migrants, while not uncommonly disabled or dependent,18 are as a group not unlike the elderly residents of Texas.35 Moreover, the majority of the migrants to Texas are from other Sun Belt states,36 which are heterogeneous in their stroke mortality rates,1 some higher and some lower than Texas, making a strong origin effect (across the entire group of inmigrants) unlikely.

Origin Effects: Black Migrants to the North
Origin or "carry-over"48 effects occur when outmigrants carry with them some of the risk of their place of birth, as may occur if the genetic makeup of the migrants or early-life (before migration) environmental exposures influence outcome or if important lifestyle factors are carried with the migrants and maintained at their destination.

Before 1920 the vast majority of US blacks lived in rural areas of the southeastern United States. In the 1920s and then again after 1940 through the early 1970s, there was extensive black migration from the rural South to inner cities of large metropolitan areas in the North and also the West.36 47 49 50 51 Approximately one third of blacks living in the Northeast and the Midwest and almost half of those living in the West were born in the South.53 Available evidence suggests that these migrants carried with them the higher stroke mortality risks of their states of birth in the South—an origin effect.48 53 54 Northern-born nonmigrant blacks had lower rates than inmigrant blacks from the South, presumably reflecting different socioenvironmental influences, which the inmigrants either did not have or acquired too late to sufficiently decrease their stroke mortality risk. An early study, for example, showed that age-adjusted stroke mortality rates in 1960 were higher for black residents of Baltimore born in the United States outside of Maryland than for those born in Maryland.53 Inmigrant men had adjusted rates 25% greater than nonmigrant men, and inmigrant women had adjusted rates 12% greater than nonmigrant women. Inmigrants came predominantly from the Stroke Belt areas of the southeastern United States. The authors acknowledged the relatively small sample and the possible contribution of other factors, such as socioeconomic status. A larger and somewhat more detailed analysis of stroke mortality among black residents of Ohio, aged 45 to 64 years, who died from 1960 to 1967 showed a marked difference in average annual, age-adjusted mortality rates when statified by place of birth for both men and women.48 Individuals born in the South had rates approximately twice the rates for those nonmigrants born in Ohio. Inmigrants from the Northeast also had higher rates than nonmigrants, being 33% greater for men and 59% greater for women. The reported rates for those born in the Midwest are difficult to interpret, since they included (and in fact were largely composed of) the blacks born in Ohio.

The present results support the concept that black inmigrants from the South to northern states carried with them some of the excess risk of stroke mortality associated with their places of birth. Black inmigrants to northern states came predominantly from the South, and these inmigrants typically had age-adjusted mortality rates that were higher than native or nonmigrant rates among blacks in the northern states. However, some contribution of selection and destination effects cannot be excluded based on available data.

Destination Effects
Destination effects on migrant mortality rates occur when individuals moving into an area change their risk because of the new location. Destination effects may be spurious if healthcare systems in the origin and destination areas have different thresholds and accuracies for recognition and reporting of disease. Destination effects may also reflect important late-life influences on outcome due to factors such as differing management practices among healthcare providers, acculturation and modification of lifelong habits and lifestyle, and pervasive environmental influences. That environmental and sociocultural changes associated with migration may modify the risk of stroke incidence and mortality has been well demonstrated among Japanese immigrants to Hawaii and California.55 56 57 58 59 In the present study intense clustering of stroke mortality rates among inmigrants, in a pattern very similar to that for nonmigrants, suggests that destination effects may be important for intranational migration. Available data do not indicate significant regional variation in either diagnostic accuracy or death certificate reporting of stroke,19 21 suggesting that the observed results are not spurious. Further studies are needed, but it appears that environmental and sociocultural modification associated with migration can significantly affect stroke mortality.55 56 57 58 59


*    Acknowledgments
 
This study was supported in part by a clinical investigator development award from the National Institutes of Health (1-K08-NS01549-01) (Dr Lanska), by Research Advisory Group funding from the Office of Research and Development of the Department of Veterans Affairs (Dr Lanska), and by the philanthropic support of Jayne Bolotin (Dr Lanska). Readers can obtain supplemental tables and maps (103 pages or 2 microfiche) from the National Auxiliary Publications Service, c/o Microfiche Publications, PO Box 3513, Grand Central Station, New York, NY 10163-3513. Request document 05201. Remit with your order (not under separate cover), in US funds only, $32.65 for photocopies or $6 for microfiche. Outside the United States and Canada, add postage of $4 for the first 20 pages and $1 for each 10 pages of material thereafter, or $1.75 for the first microfiche and $.50 for each fiche thereafter. There is a $15 invoicing charge on all orders filled before payment.

Received October 20, 1994; revision received January 9, 1995; accepted January 9, 1995.


*    References
up arrowTop
 up arrowAbstract
 up arrowIntroduction
 up arrowMethods
 up arrowResults
 up arrowDiscussion
 *References
 

  1. Lanska DJ. Geographic distribution of stroke mortality in the United States: 1939-41 to 1979-81. Neurology. 1993;43:1839-1851. [Abstract/Free Full Text]
  2. Lanska DJ, Kryscio R. The geographic distribution of hospital admissions, case fatality, and mortality from stroke among Medicare enrollees. Neurology. 1994;44:1541-1550. [Abstract/Free Full Text]
  3. Lanska DJ, Mi X. Decline in US stroke mortality in the era before antihypertensive therapy. Stroke. 1993;24:1382-1388. [Abstract/Free Full Text]
  4. National Center for Health Statistics. Vital Statistics, Mortality, Multiple Cause-of-Death, Detail, 1979 [machine-readable data file]. Springfield, Va: National Technical Information Service; 1982.
  5. National Center for Health Statistics. Vital Statistics, Mortality, Multiple Cause-of-Death, Detail, 1980 [machine-readable data file]. Springfield, Va: National Technical Information Service; 1983.
  6. National Center for Health Statistics. Vital Statistics, Mortality, Multiple Cause-of-Death, Detail, 1981 [machine-readable data file]. Springfield, Va: National Technical Information Service; 1984.
  7. US Department of Health and Human Services. The International Classification of Diseases, 9th Revision, Clinical Modification: ICD-9-CM, Vol 1. 3rd ed. Washington, DC: US Government Printing Office; 1989. US Dept of Health and Human Services publication PHS 89-1260.
  8. National Center for Health Statistics. Vital Statistics Mortality Data: Multiple Cause Detail, 1979 [tape contents and documentation package]. Springfield, Va: National Technical Information Service; 1982.
  9. Bureau of the Census. Census of Population and Housing, 1980: County Population by Age, Sex, Race, and Spanish Origin (Preliminary OMB-Consistent Modified Race) [tape technical documentation]. Washington, DC: US Bureau of the Census; 1983.
  10. Bureau of the Census. Census of Population and Housing, 1980: Public Use Microdata Sample A [machine-readable data file]. Washington, DC: US Bureau of the Census; 1983.
  11. Bureau of the Census. Census of Population and Housing, 1980: Public Use Microdata Samples [technical documentation]. Washington, DC: US Dept of Commerce; 1983.
  12. Bureau of the Census. Census of Population and Housing, 1980: County Population by Age, Sex, Race, and Spanish Origin (Preliminary OMB-Consistent Modified Race) [machine-readable data file]. Washington, DC: US Bureau of the Census; 1983.
  13. Ward DL. Population Estimates by Race and Hispanic Origin for States, Metropolitan Areas, and Selected Counties: 1980 to 1985. Washington, DC: US Bureau of the Census; 1989. Current Population Reports, Series P-25, No. 1040-RD-1.
  14. Passel JS, Siegel JS, Robinson JG. Coverage of the National Population in the 1980 Census, by Age, Sex, and Race: Preliminary Estimates by Demographic Analysis. Washington, DC: Bureau of the Census; 1982. Current Population Reports, Special Studies, P-23, No. 115.
  15. Kahn HA, Sempos CT. Statistical Methods in Epidemiology. New York, NY: Oxford University Press; 1989:92-95.
  16. Moran PAP. Notes on continuous stochastic phenomena. Biometrika. 1950;37:17-23. [Free Full Text]
  17. Ebdon D. Statistics in Geography. 2nd ed. New York, NY: Basil Blackwell Inc; 1985.
  18. Biggar JC. The Graying of the Sunbelt: A Look at the Impact of U.S. Elderly Migration. Washington, DC: Population Reference Bureau, Inc; 1984:1-17.
  19. Kuller LH, Bolker A, Saslaw MS, Paegel BL, Sisk C, Borhani N, Wray JA, Anderson H, Peterson D, Winkelstein W Jr, Cassel J, Spiers P, Robinson AG, Curry H, Lilienfeld AM, Seltser R. Nationwide cerebrovascular disease mortality study. Am J Epidemiol. 1969;90:536-578. [Abstract/Free Full Text]
  20. Nefzger MD, Acheson RM, Heyman A. Mortality from stroke among U.S. veterans in Georgia and 5 western states, I: study plan and death rates. J Chronic Dis. 1973;26:393-404. [Medline] [Order article via Infotrieve]
  21. Acheson RM, Nefzger MD, Heyman A. Mortality from stroke among U.S. veterans in Georgia and 5 western states, II: quality of death certification and clinical records. J Chronic Dis. 1973;26:405-415. [Medline] [Order article via Infotrieve]
  22. Acheson RM, Heyman A, Nefzger MD. Mortality from stroke among U.S. veterans in Georgia and 5 western states, III: hypertension and demographic characteristics. J Chronic Dis. 1973;26:417-429. [Medline] [Order article via Infotrieve]
  23. Heyman A, Nefzger MD, Acheson RM. Mortality from stroke among U.S. veterans in Georgia and 5 western states, IV: clinical observations. J Chronic Dis. 1973;26:431-446. [Medline] [Order article via Infotrieve]
  24. Kuller L, Anderson H, Peterson D, Cassel J. Nationwide cerebrovascular disease morbidity study. Stroke. 1970;1:86-98. [Abstract/Free Full Text]
  25. Centers for Disease Control. Cerebrovascular disease mortality and Medicare hospitalization—United States, 1980-1990. JAMA. 1992;268:858-859. [Medline] [Order article via Infotrieve]
  26. Lee ES. A theory of migration. Demography. 1966;3:47-57.
  27. Rogers A, Watkins J. General versus elderly interstate migration and population redistribution in the United States. Res Aging. 1987;9:483-529. [Abstract]
  28. Longino CF Jr. Geographical distribution and migration. In: Binstock RH, George LK, eds. Handbook of Aging and the Social Sciences. San Diego, Calif: Academic Press, Inc; 1990:45-63.
  29. Shryock HS Jr. Population Mobility Within the United States. Chicago, Ill: University of Chicago; 1964.
  30. Biggar JC. Who moved among the elderly, 1965 to 1970: a comparison of types of older movers. Res Aging. 1980;2:73-91. [Abstract]
  31. Biggar JC. Reassessing elderly Sunbelt migration. Res Aging. 1980;2:177-190. [Abstract]
  32. Biggar JC, Longino CF Jr, Flynn CB. Elderly interstate migration: impact on sending and receiving states, 1965 to 1970. Res Aging. 1980;2:217-232. [Abstract]
  33. Patrick CH. Health and migration of the elderly. Res Aging. 1980;2:233-241. [Abstract]
  34. Longino CF Jr, Biggar JC. The impact of retirement migration on the South. Gerontologist. 1981;21:283-290. [Medline] [Order article via Infotrieve]
  35. Longino CF Jr, Biggar JC, Flynn CB, Wiseman RF. The Retirement Migration Project: A Final Report to the National Institute on Aging. Coral Gables, Fla: University of Miami; 1984.
  36. Biggar JC. The sunning of America: migration to the Sunbelt. Popul Bull. 1979;34:1-44. [Medline] [Order article via Infotrieve]
  37. Findley SE. The directionality and age selectivity of the health-migration relation: evidence from sequences of disability and mobility in the United States. Int Migration Rev. 1988;22:4-29.
  38. Flynn CB, Longino CF Jr, Wiseman RF, Biggar JC. The redistribution of America's older population: major national migration patterns for three census decades, 1960-1980. Gerontologist. 1985;25:292-296. [Medline] [Order article via Infotrieve]
  39. Serow WJ. Return migration of the elderly in the USA: 1955-1960 and 1965-1970. J Gerontol. 1978;33:288-295. [Medline] [Order article via Infotrieve]
  40. Longino CF Jr. Going home: aged return migration in the United States 1965-1970. J Gerontol. 1979;34:736-745. [Medline] [Order article via Infotrieve]
  41. White SE. Interstate return migration: regional differences and implications. Soc Sci J. 1992;29:347-362.
  42. Speare A Jr, Meyer JW. Types of elderly residential mobility and their determinants. J Gerontol. 1988;43:S74-S81. [Medline] [Order article via Infotrieve]
  43. Wiseman RF, Roseman CC. A typology of elderly migration based on the decision making process. Economic Geography. 1979;55:324-447. [Medline] [Order article via Infotrieve]
  44. Speare A Jr, Avery R, Lawton L. Disability, residential mobility, and changes in living arrangements. J Gerontol. 1991;46:S133-S142. [Medline] [Order article via Infotrieve]
  45. Litwak E, Longino CF Jr. Migration patterns among the elderly: a developmental perspective. Gerontologist. 1987;27:266-272. [Medline] [Order article via Infotrieve]
  46. Rogers A. Return migration to region of birth among retirement-age persons in the United States. J Gerontol. 1990;45:S128-S134. [Medline] [Order article via Infotrieve]
  47. Longino CF Jr, Smith KJ. Black retirement migration in the United States. J Gerontol. 1991;46:S125-S132. [Medline] [Order article via Infotrieve]
  48. Mancuso TF, Redmond CK. Heart disease mortality among black migrants: a study of Ohio residents (1960-1967). Am Heart J. 1975;90:380-388. [Medline] [Order article via Infotrieve]
  49. Ostfeld AM, D'Atri DA. Rapid Sociocultural Change and High Blood Pressure. Basel, Switzerland: Karger; 1977:20-37. Advances in Psychosomatic Medicine Monograph No. 9.
  50. Hamilton CH. The Negro leaves the South. Demography. 1964;1:273-295.
  51. Johnson DM. Black Migration in America: A Social Demographic History. Durham, NC: Duke University Press; 1981.
  52. Bureau of the Census. Census of Population 1970: Subject Reports: State of Birth: Final Report PC(2)-2A. Washington, DC: US Bureau of the Census; 1973. Tables 23-24.
  53. Kuller L, Seltzer R. Cerebrovascular disease mortality in Maryland. Am J Epidemiol. 1967;86:442-450. [Free Full Text]
  54. Moriyama I, Krueger DE, Stamler J. Cardiovascular Diseases in the United States. Cambridge, Mass: Harvard University Press; 1971.
  55. Winkelstein W, Kagan S, Kato H, Sacks T. Epidemiologic studies of coronary heart disease and stroke in Japanese men living in Japan, Hawaii and California: blood pressure distributions. Am J Epidemiol. 1975;102:502-513. [Abstract/Free Full Text]
  56. Worth RM, Kato H, Rhoads GG, Kagan A, Syme SL. Epidemiologic studies of coronary heart disease and stroke in Japanese men living in Japan, Hawaii and California: mortality. Am J Epidemiol. 1975;102:481-490. [Abstract/Free Full Text]
  57. Marmot MG, Syme SL, Kagan A, Kato H, Cohen JB, Belsky J. Epidemiologic studies of coronary heart disease and stroke in Japanese men living in Japan, Hawaii and California: prevalence of coronary and hypertensive heart disease and associated risk factors. Am J Epidemiol. 1975;102:514-525. [Abstract/Free Full Text]
  58. Kagan A, Poppen J, Rhoads GG. Epidemiologic studies of coronary heart disease and stroke in Japanese men living in Japan, Hawaii, and California: prevalence of stroke. In: Scheinberg P, ed. Cerebrovascular Diseases: Tenth Princeton Conference. New York, NY: Raven Press Publishers; 1976:267-277.
  59. Marmot MG, Syme SL. Acculturation and coronary heart disease in Japanese-Americans. Am J Epidemiol. 1976;104:225-247.[Abstract/Free Full Text]



This article has been cited by other articles:


Home page
 Prog Hum GeogrHome page
P. Boyle
Population geography: migration and inequalities in mortality and morbidity
Progress in Human Geography, December 1, 2004; 28(6): 767 - 776.
[PDF]

Home page
 StrokeHome page
D. T. Lackland, D. L. Bachman, T. D. Carter, D. L. Barker, S. Timms, and H. Kohli
The Geographic Variation in Stroke Incidence in Two Areas of the Southeastern Stroke Belt : The Anderson and Pee Dee Stroke Study
Stroke, October 1, 1998; 29(10): 2061 - 2068.
[Abstract] [Full Text] [PDF]

Home page
 StrokeHome page
L. W. Pickle, M. Mungiole, and R. F. Gillum
Geographic Variation in Stroke Mortality in Blacks and Whites in the United States
Stroke, August 1, 1997; 28(8): 1639 - 1647.
[Abstract] [Full Text]

Home page
 StrokeHome page
D. J. Lanska
Geographic Distribution of Stroke Mortality Among Immigrants to the United States
Stroke, January 1, 1997; 28(1): 53 - 57.
[Abstract] [Full Text]

Home page
 NEJMHome page
R. F. Gillum
The Epidemiology of Cardiovascular Disease in Black Americans
N. Engl. J. Med., November 21, 1996; 335(21): 1597 - 1599.
[Full Text]

Home page
 StrokeHome page
D. J. Lanska and P. M. Peterson
Geographic Variation in Reporting of Stroke Deaths to Underlying or Contributing Causes in the United States
Stroke, November 1, 1995; 26(11): 1999 - 2003.
[Abstract] [Full Text]

Home page
 StrokeHome page
D. J. Lanska and L. H. Kuller
The Geography of Stroke Mortality in the United States and the Concept of a Stroke Belt
Stroke, July 1, 1995; 26(7): 1145 - 1149.
[Full Text]

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

Stroke Home | Subscriptions | Archives | Feedback | Authors | Help | AHA Journals Home | Search
Copyright © 1995 American Heart Association, Inc. All rights reserved. Unauthorized use prohibited.