HomeLarge Type Edition
HOME ARCHIVE SEARCH TABLE OF CONTENTS
This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
PubMed
Right arrow PubMed Citation
The Journals of Gerontology Series A: Biological Sciences and Medical Sciences 63:420-425 (2008)
© 2008 The Gerontological Society of America

The White–Black Disability Gap Revisited: Does an Incident Heart Attack Change This Gap?

Mihaela A. Popa, Laurence G. Branch and Ross Andel

1 H. Lee Moffitt Cancer Center and Research Institute, Health Outcomes and Behavior Program, Tampa, Florida.
2 College of Public Health and 3 School of Aging Studies, University of South Florida, Tampa.

Address correspondence to Mihaela A. Popa, MD, PhD, H. Lee Moffitt Cancer Center and Research Institute, 12902 Magnolia Drive, SA-PROGRAM, Tampa, FL 33612-9497. E-mail: mihaela.popa{at}moffitt.org


   Abstract
Top
 Abstract
Methods
 Results
 Discussion
 References
 
Background. A myocardial infarction (MI) results typically in abrupt functional deterioration immediately postevent, followed by recovery. The post-MI health disparities experienced by black older adults may be attributable to the social and health correlates of race. We explored patterns of change in functional status in a community-based sample of 243 older white and black persons hospitalized for an incident MI.

Methods. The study sample was drawn from the Established Populations for Epidemiologic Studies of the Elderly (EPESE). All older adults hospitalized for an incident MI between the first two waves of data collection were followed up yearly for two additional years. Nonlinear quadratic trajectories of functional status, as measured by disability in activities of daily living (ADL) and functional limitation (FL), were fit using mixed-effects models.

Results. Although there were no nonlinear differences in ADL trajectories, there was a faster nonlinear rate of change in FL in older blacks compared to whites, independent of other social and health factors. The baseline white–black gap in FL widened after the MI by the first follow-up, continued to widen at a less accelerated pace until the second follow-up, and narrowed by the third follow-up.

Conclusions. Disparities in relevant social and health factors did not account for the more abrupt deterioration in FL postevent or for the more substantial recovery in older blacks compared to older whites. Disparities in therapeutic strategies and the "survival of the fittest" may underlie the pattern of this white–black gap in FL after an incident MI.

Key Words: White–black • Disability gap • Heart attack

THE burden of disability and the pace of recovery in myocardial infarction (MI) survivors have unequal patterns in white and black older adults (1–3). For example, older blacks are more likely than whites to develop congestive heart failure in the 6 months after an MI (4,5), and older black women have lower cardiac and functional capacity 6 weeks after an MI (6) and are more likely to have a recurrent MI compared to white women (7). Understanding the health and social factors associated with differential disability outcomes in older white and black MI survivors may contribute to the achievement of the Healthy People 2010 (3) agenda of eliminating health disparities.

Race/ethnicity is associated with an array of social (8) and health correlates (2,4,8–11) that in turn affect the occurrence and sequelae of MIs. Specifically, social class is a strong predictor for no functional improvement 1 year post-MI (12), and socioeconomic status (SES) has been linked to disparities in preventive health care and evidence-based therapeutic procedures for MIs (7,13–16). In addition, smoking (10), overweight or obesity (8,9), and diabetes and high blood pressure (4,11), all well-known health risk factors for MIs (2), tend to be more common in older blacks than whites. In the rest of this article, we will refer to race/ethnicity simply as race.

The purpose of the present analysis is to establish what disparities in functional status that are associated with race still persist after controlling for major social and health correlates of race among survivors of an incident MI. According to the disablement model developed by Verbrugge and Jette (17), the pathological processes specific to an MI would progress toward impairments, then functional limitations (FL), and finally disability; this progression is influenced by a series of risk factors.

Incident MIs are associated with onset of disability (18) and with an increase in levels of prevalent activities of daily living (ADL) disability (19,20). Based on previous research (4–6) it is possible that an MI will have a greater initial effect on blacks than whites, leading to a wider disability gap in the year following the attack. Considering the acute and catastrophic nature of an MI, a nonlinear trajectory of functional status is expected, where a more substantial deterioration in the year postevent will be followed by some recovery in the following year. Using the disablement model as the conceptual framework, this study addresses the following research question: Are there any residual longitudinal differences in functional status between white and black older adults hospitalized for an incident MI, independent of other social and health factors? Specifically, we hypothesized that blacks would start out with a lower functional status at baseline, as suggested by previous findings from the general population (21,22), experience a steeper deterioration in the year postevent, and have a less substantial recovery thereafter compared to whites. We also hypothesized that there would be no residual disparities in functional status after adjusting for blacks' unfavorable social and health profile (19).


   METHODS
Top
 Abstract
 Methods
Results
 Discussion
 References
 
Data Source
We used the baseline data plus three annual follow-ups from Established Populations for Epidemiologic Studies of the Elderly (EPESE). The EPESE samples were designed to be representative of community-dwelling older adults 65 years old or older (23). Baseline survey data were collected from four geographic sites: Iowa and Washington Counties, Iowa (1981–1982), New Haven, Connecticut (1982), East Boston, Massachusetts (1982–1983), and five counties in north central North Carolina (1986–1987). Follow-up surveys were administered in person annually as long as participants were still community dwelling. The baseline and third follow-up were in-person interviews, whereas the first and second follow-ups were telephone interviews. Data regarding mortality status were available for 6 years postbaseline.

Although more waves were available, we limited our analyses to four waves based on the patterns of selective survival and the clinical course of MI. Most deaths after an MI occur in the first 3 or 4 months, and reports of total mortality in the year post-MI range from 8%–10% (24) to 40% (25). Because longer survival post-MI may characterize the fittest one and hence introduce a bias, we considered a 3-year follow-up to be more appropriate than a longer follow-up.

Of the 14,456 participants at baseline, we selected all participants who reported hospitalization for an incident MI between baseline and the first follow-up assessment (N = 248). Five cases (four whites and one black) had missing data on both education and income and were deleted, leading to a final study sample of 243 participants. During the follow-up 46 participants died (37 whites and 9 blacks). The participants who died during the follow-up were kept in the sample because the analytic method used to fit the trajectories of functional status creates longitudinal estimates for participants with missing data without case-wise deletion.

Self-reported income had about 15% missing values, body mass index (BMI) had 11%, and the rest of the covariates had <5% missing values. We imputed missing values only for income. Assuming values were missing at random, we first ran a regression model using age (measured as 5-year age groups), gender, race, site, education, and marital status as predictors and income as the outcome. This analysis indicated that age, race, education, and marital status were all associated with income (p <.05). Therefore, we replaced the missing values for income with the corresponding mode for the each 5-year age group, race, education, and marital status subgroup.

Measures
We measured two dimensions of functional status, the outcome of this study, with self-reported ADL disability and FL. We included a measure of FL to capture more accurately changes in functional status after an MI because previous studies reported only small increases in overall disability level (26) and no deterioration in discrete disability tasks such as walking (27). Disability was measured using an adaptation of the Katz scale (28), which includes six ADL items (e.g., bathing, dressing, walking across a small room, toileting, eating, and transferring), each having three response options (e.g., "do not need help" = 0; "need help" = 1, and "unable to do" = 2). These items were summed to obtain an ADL disability index, with scores ranging from 0 to 12. FL was measured using the Rosow–Breslau physical functioning scale (29) including three items (e.g., ability to do heavy work around the house, ability to walk up and downs the stairs, and ability to walk half a mile), each with two answer choices (i.e., "yes, able to" = 0 and "no, not able to" = 1). The FL index (range 0–3) resulted from the summation of these three items.

The covariates in this study reflect baseline social and health risk factors for disability as conceptualized by the disablement model (17). Demographics are represented by age measured in 5-year groups (from 65–69 to 85+ years old) and gender. Social factors include marital status, educational attainment, and income level. Health factors are represented by BMI, smoking, and being diagnosed with previous MI, high blood pressure, and/or diabetes. Mortality was measured by a dichotomous variable indicating whether participant was dead or alive during the follow-up.

Analyses
We used mixed-effects models (30) in the SAS (version 9; SAS Institute, Cary, NC) procedure MIXED using the REPEATED statement (30,31). Our mixed-effects model yielded estimates of overall fixed effects, or average effects for the group (31). The estimation method was maximum likelihood. This method allowed us to model linear and nonlinear changes in functional status over time, and to compare functional status by race over time after controlling for the health and social factors that characterize the disparities between older blacks and whites.

We followed the analytic strategy outlined by Littell and colleagues (31). Our analytic plan included three main steps: (i) testing the model fit by means of the covariance matrix structure; (ii) specifying the main and interaction effects of time within the estimation model to obtain a good estimate of the trajectory of change based on the data; and (iii) estimating the rate of change in functional status. First, we tested the model fit using the Akaike's Information Criterion (AIC). The "unstructured" covariance matrix structure yielded the best fit for ADL, and the "compound symmetry" covariance matrix structure yielded the best fit for FL.

Second, we tested the trajectory of change in functional status. Based on our hypothesis, we were interested in whether nonlinear trajectories of change in functional status would provide a good fit for our data. To test the fit of the model to the data, we included in the model variables for the linear (Time) and nonlinear (Time x Time) effect of time. These predictors test the significance of linear and nonlinear quadratic changes in the outcome over the course of the study. Finally, we tested whether there were any differences in rate of change in ADL or FL by race after controlling for various health and social factors. To test linear and nonlinear changes in ADL or FL over time by race, we included in the model interaction terms between race and time (i.e., Time x Race, Time x Time x Race). A significant nonlinear effect of time indicates that nonlinear trajectories do provide a good fit for our data. In addition, a significant interaction between time and race indicates the existence of race differences over time. If nonlinear time changes and race differences over time are evidenced, then a test of quadratic relationship between time and race (i.e., Time x Time x Race) is preferred to test the hypotheses.

After the nonlinear time effect and the race differences over time are confirmed, the terms measuring the main effects of time (Time and Time x Time) are omitted from the estimation model to obtain results for each race separately as well as the by-race comparisons (31).

We entered the covariates in the estimation model in a hierarchical fashion as follows. First we entered only main and interaction terms for race and time; then we added (sequentially) demographics, social factors, and health factors. Time was centered at the mean to avoid multicollinearity between linear and quadratic regression coefficients. Convergence criteria were met for all models. Level of significance was set at a two-tailed 0.05 level.


   RESULTS
Top
 Abstract
 Methods
 Results
Discussion
 References
 
Of the 243 participants at baseline, 79% were white and 21% were black. Blacks were significantly more likely to be younger, female, to have a higher BMI, lower levels of education and income, higher prevalence of hypertension and diabetes, and higher levels of ADL disability (Table 1).


View this table:
[in this window]
[in a new window]

  		Table 1. Sample Characteristics at Baseline.

 
The ADL disability trajectories did not change nonlinearly over time after controlling for social and health factors, as indicated by the lack of significance for the term Time x Time (F = 0.19, p =.662). Also, there were no significant white–black differences in ADL disability over time after controlling for social and health factors, as indicated by the nonsignificant interaction terms of Time x Race (F = 0.20, p =.656) and Time x Time x Race (F = 3.76, p =.054).

However, there were significant residual white–black nonlinear longitudinal differences in FL after controlling for social and health factors, as indicated by the significance for the term Time x Time (F = 25.99, p <.001). In addition, these changes were different for whites and blacks after controlling for social and health factors, as indicated by the significant interaction term for Time x Race (F = 3.92, p =.048). Based on these results from the initial specification model, we then examined further nonlinear FL trajectories for whites and for blacks separately while sequentially adding covariates (Table 2).


View this table:
[in this window]
[in a new window]

  		Table 2. Parameter Estimates and Test for Fixed Effects for Functional Limitation Modeled as a Function of Quadratic Time.

 
The interaction term Time x Time x Race yielded significant results for both whites and blacks in all four models after controlling for social and health factors (Table 2). The negative β coefficients for these interaction terms indicate that the magnitude of the change in FL was greater at earlier times compared to later times during the follow-up (31). Moreover, adding health factors did not change the estimates for nonlinear change in FL in whites, but it led to an increase in the estimate for Time x Time x Race in blacks by about 14%. Finally, the test of fixed effects for race indicated that the rates of nonlinear change in FL between whites and blacks were significantly different in Model I (F = 14.61, p <.001) and remained essentially unchanged in the other three models including the full model, which controls for the social and health factors (Model II: F = 14.45, p <.001; Model III: F = 14.93, p <.001; Model IV: F = 13.78, p <.001). The trajectories for FL by race after adjusting for social and health covariates are illustrated in Figure 1. The figure illustrates that blacks had a more accelerated increase in FL between baseline and first follow-up (after the MI) that was not attributable to social and health factors compared to whites, leading to a widening gap in FL. The gap continued to increase between the first and second follow-up. Between the second and the third follow-up, both whites and blacks experienced decreases in FL. However, blacks appeared to have a more rapid rate of decrease, resulting in a narrower gap compared to the previous wave.


Figure 01
View larger version (5K):
[in this window]
[in a new window]

  		Figure 1. Trajectories of functional limitation (FL) by race/ethnicity. Whites (triangles); blacks (squares)

 

   DISCUSSION
Top
 Abstract
 Methods
 Results
 Discussion
References
 
Independent of a number of social and health correlates of race, blacks responded more poorly than whites to an MI with respect to functional status in the initial year postevent but recovered at a faster rate than whites during the second year postevent. This finding may indicate that other factors that were found to affect MI survival differentially in whites and blacks, such as delay time from onset of symptoms to arrival at the hospital (32) and evidence-based pharmacological and interventional therapies (15,16), may be involved in blacks' deeper deterioration in functional status postevent. Also, blacks' greater recovery may be the result of the "survival of the fittest" phenomenon. That is, blacks' earlier age of onset of an initial MI, higher rates of death at earlier ages from an MI, and lower rates of survival among those hospitalized for an MI compared to whites (33) suggest that those who do survive may be the fit ones, with lower risk profile and more favorable clinical outcomes.

The rate of decline and recovery in ADL disability was similar across blacks and whites, after controlling for social and health factors. Findings from the general population 65 years old or older consistently indicate higher levels of disability in blacks compared to whites after adjusting for social and health factors cross-sectionally (21) but inconsistencies in the longitudinal patterns (32,34). Similar to findings reported by Kelley-Moore and Ferraro (35) for the entire EPESE North Carolina sample and consistent with our hypothesis, we found that blacks had more ADL disability at baseline compared to whites, but there were no residual white–black differences in longitudinal changes after controlling for social and health factors.

The incongruence between results for ADL disability and FL indicates that the dysfunctions in various organs and systems brought about by an MI are severe enough to affect tasks such as doing heavy work around the house, walking stairs, or walking half a mile, but not enough to affect ADL tasks, which tend to be less physically taxing.

The major strength of this study is its unique sample, which includes exclusively white and black older adults hospitalized for an incident MI that happened between the baseline and the first follow-up assessment. Virtually no data exist on disability trajectories in older whites and blacks after an incident MI. Another unique feature of our study is our ability to parcel out the residual effects of race after controlling for many of the known social and health correlates that underpin the health disparities research literature and the clinical literature on the functional status sequelae post-MI. Our main finding reflects a widening white–black residual gap in FL (but not in ADL) in survivors of incident MI in the first year postevent, followed by narrowing of the gap in the subsequent year, independent of social and health factors. Still, the results should be interpreted with caution. We were able to identify only 192 white and 51 black survivors of an incident MI with onset after the age of 65. The limited sample size may have made it difficult to detect statistically significant differences by race, particularly in the covariate-adjusted models, and may have affected generalizability of results. In addition, it would have been preferable to test differences in racial groups that were more equal in terms of sample size. However, the fact that we did find statistical differences in the rate of recovery by race with our sample gives more credence to our findings. Additionally, we were unable to control for other important correlates of race which would be expected to influence residual functional status, such as disparities in access to and quality of medical care, delay time from onset of symptoms to arrival at the hospital, and evidence-based pharmacological and interventional therapies. Last, we did not know the exact date of the MI, which could have occurred at any time within the first year of the study. Given the importance of the immediate post-MI period for overall health status, such information would be desirable.

Conclusion
We found a residual white–black difference in FL but not ADL after an incident MI, after adjusting for social and health factors. These findings suggest that black MI survivors are particularly vulnerable to postevent functional impairment and may need to be targeted with respect to rehabilitative care and education. Studies with larger samples and more comprehensive medical assessment data, such as health care access and types of treatment received in the hospital and after discharge, are needed.


   Footnotes
Top
 Abstract
 Methods
 Results
 Discussion
 References
 
Decision Editor: Luigi Ferrucci, MD, PhD

Received November 15, 2006

Accepted July 16, 2007


   References
Top
 Abstract
 Methods
 Results
 Discussion
 References
 

  1. Administration on Aging (AOA). Facts and Figures: Statistics on Minority Aging in the US. 2005. Available at www.aoa.gov/prof/Statistics/minority_aging/facts_minority_aging. Last accessed: August 22, 2005.
  2. American Heart Association (AHA). Older Americans and cardiovascular diseases: Statistics. 2006. Available at: http://www.americanheart.org/downloadable/heart/1136584495498OlderAm06.pdf. Last accessed September 5, 2006.U.S. Department of Health and Human Services (USDHHS). Healthy People 2010: Understanding and Improving Health. 2nd Ed. Washington, DC: U.S. Government Printing Office; 2000.
  3. Tofler GH, Stone PH, Muller JE, et al. Effects of gender and race on prognosis after myocardial infarction: adverse prognosis for women, particularly Black women. J Am Coll Cardiol. 1987;9:473-482.[Abstract]
  4. Haywood LJ. Coronary heart disease mortality/morbidity and risk in Blacks. I: Clinical manifestations and diagnostic criteria: the experience with the beta blocker MI trial. Am Heart J. 1984;108:787-793.[Medline]
  5. Rankin SH. Women recovering from acute myocardial infarction: psychosocial and physical functioning outcomes for 12 months after acute myocardial infarction. Heart Lung. 2002;31:399-410.[Medline]
  6. Jha AK, Varosy PD, Kanaya AM, et al. Differences in medical care and disease outcomes among black and white women with heart disease. Circulation. 2003;108:1089-1094.[Abstract/Free Full Text]
  7. US Department of Health and Human Services (USDHHS). Health, United States, 2005. 2005. Available at: http://www.cdc.gov/nchs/data/hus/hus05.pdf. Last accessed: August 12, 2006.
  8. Chang VW, Christakis NA. Income inequality and weight status in US metropolitan areas. Soc Sci Med. 2005;61:83-96.[Medline]Centers for Disease Control and Prevention (CDC). Trends in health and aging. Smoking status by age, sex, and race/ethnicity: United States, 1997-2004. NHIS (NHISS04). 2006. Available at: http://209.217.72.36/aging/TableViewer/tableView/aspx. Last accessed: June 2, 2006.
  9. Rosenberg L, Palmer JR, Rao RS, Adams-Campbell LL. Risk factors for coronary heart disease in African American women. Am J Epidemiol. 1999;150:904-909.[Abstract/Free Full Text]
  10. Ickovics JR, Viscoli CM, Horwitz RI. Functional recovery after myocardial infarction in men: the independent effects of social class. Ann Intern Med. 1997;127:518-525.[Abstract/Free Full Text]
  11. Ford ES, Cooper RS. Racial/ethnic differences in health care utilization of cardiovascular procedures: a review of the evidence. Health Serv Res. 1995;30:237-252.[Medline]
  12. Holmes JS, Arispe IE, Moy E. Heart disease and prevention: race and age differences in heart disease prevention, treatment, and mortality. Med Care. 2005;43:I-33-I-41.[Medline]
  13. Peterson ED, Pollack CV, Roe MT, et al. Early use of glycoprotein IIb/IIIa inhibitors in non-ST-elevation acute myocardial infarction: observations from the National Registry of Myocardial Infarction. J Am Coll Cardiol. 2003;42:45-53.[Abstract/Free Full Text]
  14. Sonel AF, Good CB, Mulgund J, et al. Racial variations in treatment and outcomes of black and white patients with high-risk non-ST-elevation acute coronary syndromes: insights from CRUSADE (Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes With Early Implementation of the ACC/AHA Guidelines?). Circulation. 2005;111:1225-1232.[Abstract/Free Full Text]
  15. Verbrugge LM, Jette AM. The disablement process. Soc Sci Med. 1994;38:1-14.[Medline]
  16. Spiers NA, Matthews RJ, Jagger C, et al. Diseases and impairments as risk factors for onset of disability in the older population in England and Wales: findings from the Medical Research Council Cognitive Function and Ageing Study. J Gerontol A Biol Sci Med Sci. 2005;60A:248-254.[Abstract/Free Full Text]
  17. Kamper AM, Stott DJ, Hyland M, Murray HM. Ford I for the PROSPER Study Group. Predictors of functional decline in elderly people with cardiovascular risk factors or disease. Age Ageing. 2005;34:450-455.[Abstract/Free Full Text]
  18. van Jaarsveld CHM, Sanderman R, Miedema I, Ranchor AV, Kempen GIJM. Changes in health-related quality of life in older patients with acute myocardial infarction or congestive heart failure: a prospective study. J Am Geriatr Soc. 2001;49:1052-1058.[Medline]
  19. Hayward MD, Crimmins EM, Miles TP, Yang Y. The significance of socioeconomic status in explaining the racial gap in chronic health conditions. American Sociological Review. 2000;65:910-930.
  20. Zsembik BA, Peek MK, Peek CW. Race and ethnic variation in the disablement process. J Aging Health. 2000;12:229-249.[Abstract/Free Full Text]
  21. Cornoni-Huntley J, Brock DB, Ostfeld AM, Taylor JO, Wallace RB. Established Populations for Epidemiologic Studies of the Elderly: Resource Data Book. National Institute on Aging, US Department of Health and Human Services, National Institutes of Health, NIH Publication No. 86-2443.
  22. Vaccarino V, Berkman LF, Krumholz HM. Long-term outcome of myocardial infarction in women and men: a population perspective. Am J Epidemiol. 2000;152:965-973.[Abstract/Free Full Text]
  23. Corti MC, Guralnik JM, Ferruci L, et al. Evidence for a black-white cross-over in all-cause and coronary heart disease mortality in an older population: The North Carolina EPESE. Am J Public Health. 1999;89:308-314.[Abstract/Free Full Text]
  24. Neill WA, Branch LG, De Jong G, et al. Cardiac disability: the impact of coronary heart disease on patients' daily activities. Arch Intern Med. 1985;145:1642-1647.[Abstract/Free Full Text]
  25. Guralnik JM, Ferucci L, Balfour JL, Volpato S, Di Iorio A. Progressive versus catastrophic loss of ability to walk: implications for the prevention of mobility loss. J Am Geriatr Soc. 2001;49:1463-1470.[Medline]
  26. Katz S. Assessing self-maintenance: activities of daily living, mobility and instrumental activities of daily living. J Am Geriatr Soc. 1983;31:721-726.[Medline]
  27. Rosow I, Breslau N. A Guttman health scale for the aged. J Gerontol. 1966;21:556-559.
  28. SAS Institute. SAS System for Microsoft Windows (version 9). Cary, NC: SAS Institute Inc.; 2003.
  29. Littell RC, Milliken GA, Stroup WW, Wolfinger RD. SAS system for mixed models. Cary, NC: SAS Publishing; 2005:87–135.
  30. Cooper RS, Simmons B, Castaner A, Prasad R, Franklin C, Ferlinz J. Survival rates and prehospital delay during myocardial infarction among black persons. Am J Cardiol. 1986;57:208-211.[Medline]
  31. Roig E, Castaner A, Simmons B, Patel R, Ford E, Cooper R. In-hospital mortality rates from acute myocardial infarction by race in US hospitals: findings from the National Hospital Discharge Survey. Circulation. 1987;76:280-288.[Abstract/Free Full Text]
  32. Mendes de Leon CF, Barnes LL, Bienias JL, Skarupski KA, Evans DA. Racial disparities in disability: recent evidence from self-reported and performance-based disability measures in a population-based study of older adults. J Gerontol B Psychol Sci Soc Sci. 2005;60:263-271.
  33. Kelley-Moore JA, Ferraro KF. The black/white disability gap: persistent inequality in later life? J Gerontol Soc Sci. 2004;59B:S34-S43.[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Services
Right arrow Download to citation manager
PubMed
Right arrow PubMed Citation

HOME ARCHIVE SEARCH TABLE OF CONTENTS