Racial and Ethnic Disparities in Diabetes Care in the United States

Racial and Ethnic Disparities in Diabetes Care in the United States Ronny Bell, PhD, MS Professor, Division of Public Health Sciences Director, Maya Angelou Research Center on Minority Health Wake Forest University School of Medicine Overview   To describe the outcomes of a recently completed research project designed to document racial and ethnic disparities in diabetes care and outcomes among U.S. adults To discuss future directions for the work produced from this project The Burden of Diabetes in the United States     Approximately 21 million Americans have diabetes Diabetes is the sixth leading cause of death in the United States, contributing to approximately 225,000 deaths annually Direct and indirect cost associated with diabetes (2007) exceed $174 billion Diabetes accounts for 19% of health care expenditures in the US Source: American Diabetes Association Percentage of the U.S. population > 20 years of age with diagnosed diabetes, undiagnosed diabetes, and IFG 20.8 Million Americans Percent with diagnosed diabetes, undiagnosed diabetes, or IFG Diagnosed diabetes Undiagnosed diabetes IFG 50 45 40 35 30 25 20 15 10 5 0 8.3% 6.5 2.8 7.9% 5.2 2.7 14.6% 11.0 3.6 10.4 3.0 13.4% 31.6 26 26.1 17.7 All Races Non-Hispanic White Non-Hispanic Black Mexican American Data from the National Health and Nutrition Examination Survey, 1999-2002. Source: Cowie et al. Prevalence of diabetes and impaired fasting glucose in adults in the U.S. population. Diabetes Care 2006;29:1263-1268. Prevalence of Self-Reported Diabetes among U.S. Adults, by Race/ethnicity and Sex, 2005 Source: Centers for Disease Control and Prevention, Behavioral Risk Factor Surveillance System Prevalence of Diabetes Among Strong Heart Participants 80 64.7 60 40 20 0 Arizona Oklahoma Dakota Total 36.4 40.8 32.4 43.0 43.5 70.9 Men Women 52.4 % with Diabetes, 45-74 Years of Age, By Sex and Center From: Welty, et al. Cardiovascular Disease Risk Factors among American Indians: The Strong Heart Study. American Journal of Epidemiology 1995;142:269-287 Prevalence of Diabetes Macrovascular and Microvascular Complications Diagnosed diabetes Normal blood sugar levels Percentage with complications 30 27.8 22.9 18.9 20 10 9.8 1.8 9.5 1.7 9.1 2.1 Coronary heart disease 7.9 1.1 Congestive heart failure 10 6.6 1.8 Stroke Chronic kidney disease Foot problems Eye damage 6.1 0 Heart attack Chest pain Macrovascular Microvascular American Association of Clinical Endocrinologists. State of Diabetes Complications in America Report. Available at: http://www.aace.com/newsroom/press/2007/images/DiabetesComplicationsReport_FINAL.pdf. Relative Risks for Incidence of Lower Extremity Amputation among US Veterans with Diabetes Race/Ethnicity White African American Hispanic Adjusted Odds Ratio (95% Confidence Interval) Referent 1.41 (1.34 – 1.48) 1.28 (1.20 – 1.38) Native American Asian 1.74 (1.39 – 2.18) 0.31 (0.19 – 0.50) Source: Young et al., Diabetes Care 26:495-501, 2003 Cardiovascular Disease in Type 2 Diabetes     CVD is the leading cause of death among people with diabetes, attributing to 65 – 70% of all death CHD risk increased 2-4 fold among persons with diabetes 8-27% of CHD cases may be attributable to diabetes  Higher in high risk populations Probability of survival for persons with MI is reduced for persons with diabetes Effect of Diabetes on Life Expectancy With and Without Diabetes at Age 50 Years – Framingham Heart Study 40 35 30 25 20 15 10 5 0 No 22.0 14.2 6.8 LE with CVD 28.8 LE without CVD 34.7 6.6 -8.2 years 26.5 6.8 -7.5 years 21.3 7.1 LE, Years 28.0 19.6 Men Yes No Women Yes Presence of Diabetes at Baseline Franco et al, Arch Intern Med 2007;167:1145-1151 Hazard Ratios and 95% Confidence Intervals for Ethnicity (Reference: Whites) From Cox Proportional Hazard Models of Diabetic Complications Karter, A. J. et al. JAMA 2002;287:2519-2527. Copyright restrictions may apply. Project Objectives  Association of Teachers of Preventive Medicine (CDC) issued a Request for Applications in 2001 for projects to understand racial and ethnic disparities in diabetes care in the United States  Racial/Ethnic Gaps in Diabetes Burden, Care and Outcomes – Julienne Kirk, Principal Investigator Project Objectives    To systematically search publication databases and governmental and organizational online resources to identify and obtain the published literature pertaining to racial and ethnic disparities in diabetes burden in the U.S. To grade literature obtained according to predefined quality criteria. To create evidence tables summarizing methodology and key findings from all sources which meet minimum quality criteria. Project Objectives   To produce manuscripts of publishable quality that detail disparities, theories of causes, and propose a model To perform an analysis of diabetes-related morbidity and mortality by race/ethnicity in a retrospective cohort of elderly Medicare beneficiaries with diabetes (1994-1999) Advisory Committee      Tom Arcury – Family and Community Medicine Alain Bertoni – Public Health Sciences David Goff, Jr. – Public Health Sciences Sara Quandt – Public Health Sciences Program Assistant – Carol Hildebrandt Model – Racial/Ethnic Gaps in Diabetes Burden, Care & Outcomes Race/Ethnicity PatientΨ Characteristics Diabetes Development Elements of Quality of Care† Control Outcomes¥ Self Mgt Outcomes‡ Complications Provider & Group Structural Aspects* Race/Ethnicity Key ΨPatient Characteristics = SES, Culture, Perceptions, Locus of Control, Social Support, Psychosocial/Depression, Language, Immigration Status, Health Literacy, Insurance, Communication with Provider *Provider & Group Structural Aspects = Racism, Community, Rural/Urban/Inner City, Policy Issues, Immigration, Access to Care, Cultural Competence, Barriers/Gaps, ¥Control Outcomes = Blood Pressure, Lipids, Microalbuminuria, HbA1c, Eye Exams, Quality of Life, Adherence, DQIP ‡Self Management Outcomes = Diet, Exercise, Foot Checks, Self-Monitoring of Blood Glucose, Self Care †Elements of Quality of Care = Diabetes Screening, Education Processes, Treatment/Rx Processes, Provider Perceptions, Vaccination Literature Ascertainment and Grading     Abstracts of manuscripts were obtained using search terms defined by the committee Assistance provided by WFUSM Carpenter Library staff Over 1,700 abstracts pulled initially Abstracts reviewed by JK and RB for relevance to project work Literature Ascertainment and Grading    Abstract citations entered into Excel spreadsheet Keywords given to each citation Full-text of relevant citations were pulled and systematically graded (JK, RB)  Input on qualitative studies provided by Drs. Arcury and Quandt  Evidence tables created in Excel (CH), and citations entered into RefMan Project Productivity    Eight total publications in peer-reviewed journals Five presentations at national scientific meetings Follow-up grants under development A Qualitative Review of Studies of Diabetes Preventive Care among Minority Patients in the United States, 1993-2003 Julie Kirk, Ronny Bell, Alain Bertoni, Tom Arcury, Sara Quandt, David Goff, Jr. Wake Forest University School of Medicine KM Venkat Narayan Centers for Disease Control and Prevention American Journal of Managed Care 2005;11:349-360 Presented at the 19th National Conference on Chronic Disease Prevention and Control, Atlanta, Georgia, March 2005 Study Methodology  Database search of studies published from 1993 – 2003  Medline, Web of Science, Education Resources Information Center, Cumulative Index to Nursing and Allied Health, Combined Health Information Database, Cochrane Library  Studies limited to adult populations with diabetes (excluding pre-diabetes and gestational diabetes) Study Methodology  Database strategy     “Exploded” diabetes mellitus Free text terms not included in MeSH headings (e.g., literacy, immigration status, locus of control, insurance) Search terms – Preventive measures, processes of care, quality of care “Race/ethnicity/ethnology”, “ethnic groups”, “minority groups” Study Methodology  Process Measures        Glycemia testing Eye examination rates Foot examination rates Lipid profile assessment Influenza vaccination Nephropathy assessment Counseling referrals for smoking cessation Results    450 abstracts initially reviewed 390 full-text articles retrieved 36 studies had data specific to ethnic minority populations  One study did not have specific rates provided  Studies included those with specific ethnic comparisons as well as those with data for ethnic minority groups Results Process Measure Glycemia testing Eye examination rates Foot examination rates Lipid profile assessment Influenza vaccinations Nephropathy assessment Smoking cessation counseling Number of Studies 15 28 18 15 8 7 4 Results    Majority of studies showed low rates of adherence to diabetes preventive care Major racial/ethnic disparities found in eye examination, influenza vaccination and lipid profile testing among African Americans and Hispanics compared to non-Hispanic whites Limited comparative data for American Indians Racial/Ethnic Disparities in Eye and Foot Examinations, 2001 100 80 Percent Non-Hispanic Whites African Americans Hispanics 68.3 63.8 70.2 60.0 61.4 54.7 60 40 20 0 Eye Examination Source: MMWR, 2001 Foot Examination Trends in the Quality of Care and Racial Disparities in Medicare Managed Care 1999 White Eye Exam HbA1c Testing HbA1c Control 64 75 71 Black 55 71 67 White 72 90 82 2003 Black 70 88 75 LDL Testing LDL Control 70 36 61 23 94 73 92 66 Triveldi et al., N Engl J Med. 2005 Aug 18;353:692-700 Trends in the Quality of Care and Racial Disparities in Medicare Managed Care 14 12 Disparity (% Difference) 1999 2003 9 7 4 2 2 4 2 9 13 10 8 6 4 2 0 7 Eye Examination Glycemia Testing Glycemia Control LDL Testing LDL Control Triveldi et al., N Engl J Med. 2005 Aug 18;353:692-700 Conclusions   Rates generally low regardless of population measured Variations in measures of assessment   Frequency of measurement Source of data (self-report vs. chart review)   Many studies have a small sample size with limited representativeness More population-based data needed, particularly for some ethnic minority groups Ethnic Disparities in the Control of Glycemia, Blood Pressure and LDL Cholesterol Among U.S. Adults: A Systematic Review Julie Kirk, Ronny Bell, Alain Bertoni, Tom Arcury, Sara Quandt, David Goff, Jr. Wake Forest University School of Medicine KM Venkat Narayan Centers for Disease Control and Prevention Annals of Pharmacotherapy 2005;39:1489-1501 Presented at the28th Annual Centers for Disease Control and Prevention, Division of Diabetes Translation Conference, Miami, Florida, May, 2005 Study Methods    Publication abstraction similar to Process Measures paper (1993 – 2003) Quality of care indicators include control of glycemia, LDL-cholesterol, and blood pressure Quality measures included those from the ADA, JNC VI/VII, ATP II/III, HEDIS Results     450 abstracts initially reviewed 390 full-text articles retrieved 77 studies had data specific to ethnic minority populations Studies included those with single ethnic minority groups and those with comparisons to whites Results Outcome Measure Single Ethnic Minority Group Ethnic Comparisons Glycemia LDL-Cholesterol Blood Pressure 31 8 25 25 13 22 Racial/Ethnic Disparities in Glycemic Control Harris et al. (NHANES III) HbA1c >7% Whites Blacks Hispanics 55.1% 58.2% 65.5% McBean et al. (Medicare) HbA1c >9.5% 32.0% 40.4% 35.8% Source: Harris et al., Diabetes Care 1999;22:403-8; McBean et al., Diabetes Care 2003;26:3250-6. Racial/Ethnic Disparities in Diabetes Care, Insulin Resistance Atherosclerosis Study 100 80 65 69 54 66 38 34 44 Whites Whites Blacks Hispanics Percent 60 40 20 0 52 60 58 61 62 Poor Glycemic Poor LDL Control Control HbA1c >7 LDL >3.36 mmol/l Poor BP Control BP >130/85 Source: Bonds et al., Diabetes Care 2003;26:1040-1046 Conclusions     Rates generally low regardless of population examined Ethnic disparities most prominent for glycemic control Wide variation in study design, sample size, and assessment of control Much of the data represents time period prior to widespread use of effective medications Disparities in HbA1c between African American and NonHispanic White adults with Type 2 diabetes: A meta-analysis Julie Kirk, Ralph D’Agostino Jr, Ronny Bell, Passmore LV, Bonds DE, Karter AJ, Venkat Narayan KM Diabetes Care 2006;29:2130-2136 Methods     Literature retrieved providing data on glycemic control in African American and non-Hispanic white populations (1993 – 2005) 78 studies had data on A1c that met the initial criteria for inclusion 11 studies included in analyses For four studies, additional data provided by authors Methods  Study Designs  3 prospective cohort, 8 cross-sectional Managed Care, Non-Managed Care Retrieved from medical record review Blood draw from study participation  Health Care Setting   Data Collection   African Americans higher HbA1c non-His panic Whites higher HbA1c Bell et al, 2001*‡ Bonds et al, 2003†₤ Cook et al, 2000*‡ de Rekeneire et al, 2003 †₤ Gary et al, 2004*‡ Harris et al, 1999*₤ Kart er et al, 2002†‡ Sharma et al, 2001 *‡ Summerson et al, 1996 *₤ Weat herspoon et al, 1994 *‡ Wing et al, 1996†₤ Summary Effect -P rimary Summary Effect -Cross Sec.* Summary Effect -Cohort † Summary Effect -Chart Review‡ Summary Effect -Blood Draw₤ -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 Standardized Effect Men Bell et al, 2001 (20) Cook et al, 2000 (30) de Rekeneire et al, 2003 (28) Harris et al, 1999 (21) Summerson et al, 1996 (32) Weat herspoon et al, 1994 (33) (-0.29) -1.4 -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 Summary Effect Standard Effect with 95% Confidence Interval Bell et al, 2001 (20) Cook et al, 2000 (30) Women de Rekeneire et al, 2003 (28) Harris et al, 1999 (21) Summerson et al, 1996 (32) Weathersp oon et al, 1994 (33) Summary Effect (-0.36) (-0.36) (-0.29) -1.2 -1 -0.8 -0.6 -0.4 -0.2 0 0.2 Standard Effe ct with 95% Confide nce Inte rval Conclusions   Consistent disparities in glycemic control exist between African Americans and nonHispanic whites Summary effect size of -0.32 translated into an estimated difference of 0.65% in A1c levels between the two groups (slightly higher for women than men) Disparities in A1C Levels between Hispanics and Non-Hispanic White Adults with Diabetes: A Meta-Analysis Julie Kirk, Leah Passmore, Ronny Bell, Venkat Narayan, Ralph D’Agostino Jr., Tom Arcury, Sara Quandt Diabetes Care 2008;31:240–246 Presented at the Meharry-Vanderbilt Alliance Annual National Health Disparities Conference: “Diabetes Health Disparities: Determinants, Prevention, Treatment, and Policy,” Nashville, TN, November 12-13, 2007 Methods     Literature retrieved providing data on glycemic control in Hispanics and nonHispanic white populations (1993 – 2007) 41 studies on diabetes in Hispanic populations 11 studies included in analyses For three studies, additional data provided by authors Methods  Study Designs  3 prospective cohort, 8 cross-sectional Managed Care, Non-Managed Care Retrieved from medical record review Blood draw from study participation  Health Care Setting   Data Collection    Studies included a diversity of Hispanic populations Conclusions  Hispanics have a difference in A1c of approximately 0.5%  10.5% reduction in vascular complications  Results were consistent regardless of study design, health care setting and data collection method Conclusions I  In elderly patients with diabetes in US Medicare:      South blacks were at lower risk for CVD outcomes than non-South blacks South BF had a 5-20% lower risk for various CVD outcomes than WF South BM had a 7-31% lower risk for various CVD outcomes than WM Non-South BF had a 6-12% higher risk than WM Non-South BM had a similar risk than WM except for a 17% lower risk for AMI BF=Black Females, BM=Black Males, CVD=Cardiovascular Disease, WF=White Females WM=White Males Other Findings of Interest  Bertoni AG, Kirk JK, Case LD, Kay, C, Goff DC Jr, Venkat Narayan KM, Bell RA. Race, region, and cardiovascular disease incidence in the elderly with diabetes. Diabetes Care 2005;28:2620-2625.  High level of burden of diabetes-related CVD among older adults, with some variation by race and region Co-Morbidities by Race White ESRD Renal Disease Hypertension Neuropathy† 0.6% 6.1% 62.2% 15.4% Black 2.3% 10.7% 75.6% 14.8% Retinopathy AMI IHD CHF‡ CVA 15.0% 5.1% 39.9% 22.4% 18.3% 17.3% 4.1% 31.2% 22.6% 19.0% All differences p<0.001 except†p<0.05 and ‡(NS) Incidence of IHD IHD South At risk 15799 5689 9929 2092 Incident 7280 2333 4884 822 Rate 95% CI 15.5-16.2 13.2-14.3 17.8-18.8 13.1-15.1 White women Black women White men Black men IHD Other region 15.9 13.7 18.3 14.1 White women Black women White men Black men 30821 3103 19249 1467 14485 1565 9736 699 17.0 19.7 19.6 19.3 16.7-17.3 18.7-20.7 19.2-20.0 17.9-20.8 Other Findings of Interest  Kirk JK, Graves DE, Bell RA, Hildebrandt CA, Venkat Narayan KM. Ethnic Disparities in Self-Monitoring of Blood Glucose among US Adults: A Qualitative Review. Ethnicity and Disease 2007;17:135 - 142.    Review of 10 studies shows low rates of self-monitoring of blood glucose, with generally lower rates among ethnic minorities English fluency has an influence on SMBG in some studies Limited data on Asian Americans and American Indians Overall Conclusions   Racial and ethnic disparities exist in process and outcome indicators necessary for optimal diabetes management These disparities may explain to some extent the excess in diabetes morbidity and mortality in these populations Future Efforts    Need to further elucidate the cause of ethnic disparities in diabetes process and outcome measures Need to include more data on other racial and ethnic minority groups What types of interventions will best address these complex issues? References Bertoni AG, Kirk JK, Goff DC, Jr., Wagenknecht LE. Excess mortality associated with diabetes mellitus among Medicare beneficiaries. Ann Epidemiol 2004:14:362 – 367. Kirk JK, Bell RA, Bertoni AG, Arcury TA, Quandt SA, Goff DC Jr, Narayan KMV. A Qualitative Review of Studies of Diabetes Preventive Care among Minority Patients in the United States, 1993-2003. Am J Manag Care. 2005;11:349-360. Bertoni AG, Kirk JK, Case LD, Kay, C, Goff DC Jr, Venkat Narayan KM, Bell RA. Race, region, and cardiovascular disease incidence in the elderly with diabetes. Diabetes Care 2005;28:2620-2625. Kirk JK, Bell RA, Bertoni AG, Arcury TA, Quandt SA, Goff DC Jr, Narayan KMV. Ethnic Disparities: Control of Glycemia, Blood Pressure, and LDL-Cholesterol among US adults with Type 2 diabetes. Ann Pharmacother 2005;39:14891501. References Kirk JK, D’Agostino RB Jr, Bell RA, Passmore LV, Bonds DE, Karter AJ, Venkat Narayan KM. Disparities in HbA1c between African American and Non-Hispanic White adults with Type 2 diabetes: A meta-analysis. Diabetes Care 2006;29:2130-2136. Kirk JK, Graves DE, Bell RA, Hildebrandt CA, Venkat Narayan KM. Ethnic Disparities in Self-Monitoring of Blood Glucose among US Adults: A Qualitative Review. Ethnicity and Disease 2007;17:135 - 142. Kirk JK, Bertoni AG, Case LD, Bell RA, Goff, DC, Jr., Venkat Narayan, KM. Racial/Ethnic differences in predicted risk of coronary heart disease among persons with Type 2 diabetes. Coronary Artery Disease 2007;18:595-600. Kirk JK, Passmore LV, Bell RA, Venkat Narayan KM, D’Agostino RB, Jr., Arcury TA, Quandt SA. Disparities in A1C Levels between Hispanics and Non-Hispanic White Adults with Diabetes: A Meta-Analysis. Diabetes Care 2008;31:240–246.