Disparities in Diabetes Management by Race or Ethnicity in a Primary Care Clinic in Central Texas
The Journal – November 2010
Tex Med. 2010;106(11):e1.
By Samuel N. Forjuoh, MD, DrPH; Jane N. Bolin, RN, JD, PhD;Manisha Gupta, MD, MPH; Charles Huber, PhD; Janet W. Helduser, MA; Sonia Holleman, BS; Anne Robertson; and Marcia G. Ory, PhD, MPH
Dr Forjuoh, Dr Gupta, Ms Holleman, and Ms Robertson, Department of Family & Community Medicine, Scott & White, College of Medicine; Dr Bolin and Ms Helduser, Department of Health Policy & Management, School of Rural Public Health; Dr Huber, Department of Epidemiology & Biostatistics, School of Rural Public Health; and Dr Ory, Department of Social & Behavioral Health, School of Rural Public Health; Texas A&M Health Science Center, Temple, Texas. Send correspondence to Samuel N. Forjuoh, MD, DrPH, Department of Family & Community Medicine, Scott & White Santa Fe Center, 1402 W Avenue H, Temple, TX 76504; e-mail: email@example.com.
We determined the nature and magnitude of extant health disparities in patients with type 2 diabetes (T2DM) by race and ethnicity. Data were abstracted from the electronic medical records and charts of all patients 18 years or older who had been diagnosed with T2DM and seen over a 1-year period in one primary care clinic. Data abstracted included patient demographics; provision of counseling on smoking cessation, diet, exercise, and home blood glucose monitoring (HBGM); health care utilization; laboratory measures; and clinical outcomes. No significant racial or ethnic differences were found in the rate of provision of counseling on smoking cessation, diet, exercise, and HBGM, which were all suboptimal according to American Diabetes Association recommendations. In addition, no significant differences were found in the mean number of hospital admissions, emergency room visits, and referrals for specialty care. However, the mean HbA1c levels for African Americans (9.9%) and Hispanics (9.0%) were significantly higher than that of whites (8.7%; P<.0001), even after controlling for body mass index and age. Explanation of the significant racial and ethnic differences found in HbA1c levels, despite similar diabetes self-management treatment protocols or health care utilization, calls for further research.
Diabetes is reaching epidemic proportions, affecting an estimated 23.6 million people (7.8%) of the US population.1 Of this number, 17.9 million people have diagnosed and 5.7 million people have undiagnosed diabetes.1 The estimated diabetes-related costs in the United States in 2007 totaled $174 billion: $116 billion in direct medical costs and $58 billion in indirect costs, including disability, work loss, and premature mortality.1 Even after adjusting for population age and sex differences, the average medical expenditure for people with diagnosed diabetes has been found to be 2 to 3 times higher than for those without diabetes.1 A major cause of morbidity and disability, diabetes is now the fifth leading cause of death in the United States.2
Significant health disparities are often reported among racial and ethnic minorities in access to health care, health outcomes, and quality of care despite concerted efforts over the past several decades to address these differences.3 Racial and ethnic minorities indeed bear a disproportionate burden of the diabetes epidemic; they have higher prevalence rates, poorer diabetes control, and higher rates of complications.4-8 For example, of the 23.5 million (10.7%) of the population 20 years or older with diabetes, 14.9 million (9.8%) of all non-Hispanic whites 20 years or older have diabetes, while 3.7 million (14.7%) of all non-Hispanic blacks 20 years or older have diabetes.1 Moreover, African Americans have 2 to 4 times more renal disease, blindness, amputations, and amputation-related mortality than do non-Hispanic whites.9-11 Similarly, Latinos have higher rates of renal disease and retinopathy.9,11 In addition, racial and ethnic minorities have higher rates of dyslipidemia and hypertension.12 The Institute of Medicine report titled "Unequal Treatment: Confronting Racial and Ethnic Disparities in Health Care" reviewed the literature and found a persistent residual gap between minorities and nonminorities in outcomes attributed to differences in the quality of care received.13
Several specific studies reveal a more complicated picture. Heisler and colleagues examined racial disparities in diabetes care processes, outcomes, and treatment intensity and found no racial differences in the receipt of HbA1c tests or foot examinations, but African Americans were less likely than whites to have had their low-density lipoprotein (LDL) cholesterol level checked in the previous 2 years (72% vs 80%, P <.05) and also to have had a dilated eye examination (50% vs 63%, P <.01). The racial disparities associated with LDL-cholesterol testing and eye examinations persisted even after adjusting for patients' age, education, income, insulin use, diabetes self-management, diabetes duration, diabetes severity, comorbidities, and health services utilization.14
Suboptimal diabetes care has been found to exist for both minorities and nonminorities. A study of adherence to clinical guidelines for preventing coronary heart disease in patients with type 2 diabetes mellitus (T2DM) conducted by Kamyar et al found that overall provider adherence to guideline criteria was significantly lower for secondary prevention than for primary prevention (74.4% vs 80.1%, P<.05).15 Furthermore, a medical chart review of rural private practices reported the following adherence rates to diabetes guidelines: 15% for foot examinations, 23% for annual eye examinations, 33% for annual microalbuminuria testing, and 20% for annual HbA1c testing.16 Another study designed to determine if there was concordance of provider recommendations with American Diabetes Association guidelines found that many patients were not receiving all of the eight processes (best practices) of care studied, particularly those with adult-onset diabetes, such as elderly patients, African Americans, and Hispanics.17
The purpose of this study was to evaluate extant disparities in diabetes self-management treatment protocols, health care utilization, laboratory measures, and clinical outcomes by race and ethnicity in one primary care clinic in Central Texas. The clinic was chosen to help control for practice differences across settings. This information is intended to be one step in a larger study examining whether disparities exist in an integrated health care system, which has been evaluated as a high adherer to diabetes treatment guidelines.18,19
We conducted an electronic medical record (EMR) search, chart review, and data abstraction on extant data from one primary care clinic of a large university-affiliated, multispecialty group practice associated with a 186,000-member health maintenance organization (HMO) in Central Texas. The chart review and data abstraction were performed as part of a larger health disparities intervention study being conducted at the clinic and 12 others in Central Texas. The average annual total number of patients seen at this clinic is about 25,585, including more than 3000 patients with T2DM, type 1 diabetes, and gestational diabetes. The Scott & White Institutional Review Board reviewed and approved the study protocol.
The study participants included all patients 18 years or older who had been diagnosed with T2DM and were identified via an exhaustive EMR search as having been seen at the clinic between December 2006 and November 2007. A standardized form was used to abstract data from the EMRs and charts on the identified patients, including their demographics, laboratory measures, clinical outcomes, and any mention of provision of counseling related to diabetes care. The demographic data that were abstracted included patient age, sex, race and ethnicity, and health insurance status. The laboratory measures abstracted included total cholesterol, LDL-cholesterol, high-density lipoprotein (HDL) cholesterol, microalbumin, creatinine kinase (CK), and creatinine kinase-myocardial band isoenzyme (CK-MB). The clinical outcomes included HbA1c values; number of emergency room visits, hospital admissions, and referrals for specialty care; and number of podiatry visits and ophthalmology visits. Data on diabetes counseling abstracted included any mention of provision of counseling on smoking cessation, diet, exercise, and HBGM from the previous 3 years of individual subjects. If specific phrases such as "counseling on diet were provided" were mentioned or if a diabetes education class was attended and recorded by a diabetic educator, then the patient was considered as having received diabetes counseling.
Data were entered into an Excel file, which were then imported into an SPSS data file. Analyses were performed by using SPSS version 11.0 on a personal computer. Descriptive analyses were performed to determine frequencies and means. For preliminary bivariate analyses, two-tailed chi-square tests were used to determine significant differences between proportions. Finally, we used linear regression modeling to adjust for the observed significant racial and ethnic differences in HbA1c levels, controlling for body mass index (BMI) and age. The statistical significance level used was P<.05.
Values used to categorize laboratory findings were obtained from MedlinePlus (website of the National Institute of Health).20 The HDL-cholesterol values were classified as indicative of high risk (<40 mg/dL), average risk (40-60 mg/dL), and less-than-average risk (>60 mg/dL) for heart disease. Values for CK were classified as normal (38-120) and high (>120). The CK-MB values were classified as normal (≤3) and abnormal (>3). Other laboratory values were classified as shown in Table 1 [PDF].
Characteristics of study subjects
The final study sample (n=548) comprised 65.3% whites, 20.4% Hispanics, and 14.2% African Americans with mean ages of 61.3, 53.1, and 52.3 years, respectively. Their mean BMIs were 34.5, 34.7, and 39.2, respectively. Approximately one-third of the subjects were aged 55 to 64 years, 15% were 45 years or younger, and 12% were 75 years or older. Half were male (49.6%), and most had the HMO's health plan (52.4%), with the rest having some other insurance payer or no insurance.
White subjects were significantly older (P<.0001) and less likely to subscribe to the HMO's health plan (P=.004). Nearly three-quarters of the subjects were clinically obese with this finding, being consistent across all three race and ethnicity categories. Although more African Americans were current smokers and more whites were successful in quitting smoking, no significant differences were found in smoking behaviors across race and ethnicity (Table 2) [PDF].
Provision of counseling
Overall rates for provision of counseling on smoking cessation for current smokers and provision of counseling on diet, exercise, and HBGM were 69.2%, 74.5%, 69.3%, and 70.1%, respectively, with no significant racial or ethnic differences (Figure 1) [PDF]. We also found that counseling was provided by different health professionals such as patients' primary care physicians (PCPs), other specialty physicians (eg, endocrinologists and cardiologists), and nurses (eg, registered nurses and nurse diabetes educators), although counseling was provided mainly by the PCPs. Other specialty physicians and nurses were more likely to provide counseling on diet compared with providing counseling on exercise and HBGM, although these differences were not statistically significant (Table 3) [PDF].
Significant differences were found in triglyceride levels across race and ethnicity categories, with whites and Hispanics showing higher levels compared with African Americans (P=.02). Total cholesterol levels were higher among Hispanic patients (14.1%) compared with whites (8.8%) and African Americans (11.3%), while LDL-cholesterol levels were higher among African Americans (6.8%) compared with whites (4.9%) and Hispanics (4.9%). However, none of these differences was statistically significant. Hispanics (3.9%) had higher urine microalbumin levels indicative of advanced kidney disease, but this finding was only marginally significant (Table 4) [PDF].
As shown in Figure 2 [PDF], the mean HbA1c levels for African Americans (9.9%) and Hispanics (9.0%) were significantly higher than those of whites (8.7%) (P<.0001). These significant racial and ethnic differences in HbA1c levels persisted after adjusting for BMI and age (data not shown). However, no significant differences were found in the mean number of hospital admissions, emergency room visits, ophthalmology visits, podiatry visits, and referrals for specialty care by race and ethnicity.
Healthy People 2010, which guides the disease prevention agenda for the United States, has set a goal of eliminating disparities in diabetes health outcomes by the year 2010.21 To address the problem of suboptimal diabetes care, the federal government, insurance companies, and health care delivery systems have invested heavily in quality improvement programs. The Veterans Health Administration, for example, has implemented a number of quality improvement initiatives and recently reported measures of diabetes care and diabetes health outcomes that are much higher than national averages, including annual rates for HbA1c testing and eye examinations at 93% and 91%, respectively, and rates for control of diabetes and dyslipidemia at 83% and 86%, respectively.22
In this study, we found significant differences in the extant clinical indicator of diabetes status, HbA1c levels, among race and ethnicity groups. These results are similar to those reported in other studies that have found racial and ethnic differences in clinical status.14,23 While the causes of these health disparities remain unclear, the differences have often been attributed to potential confounders such as gender, BMI, diabetes duration, diabetes therapy, patient adherence, and provider management.24 In this study, however, BMI and access to health care were not found to be related to HbA1c variations by race and ethnicity. This may be attributed to the fact that BMI was almost equal across the groups, and almost all our study subjects had health insurance. That differences in counseling or provider management could be a reason for observed differences in diabetes outcomes has been suggested,24 but no significant differences were observed in the rate for provision of counseling in this study. (Differences in patient compliance with clinical recommendations could also play a role in racial and ethnic disparities, despite receipt of the same rate for provision of counseling.6,25 The observed differences in glycemic control in our study could also be attributed to differences in medication adherence,26 which is difficult to measure because patients generally refill their prescriptions at different pharmacies, and collecting data on prescription refills was beyond the scope of this study. Understanding why this difference occurs, however, has important implications in the efforts to reduce morbidity and mortality in minority groups. Further studies are, therefore, required to explain the differences in disease management protocols contributing to the observed differences.
Compared with their white and Hispanic counterparts, African Americans were found to have high levels of LDL-cholesterol and high systolic and diastolic blood pressures. Similar findings were reported in a systematic qualitative review of 1993-2003 literature by Kirk and colleagues, who assessed ethnic disparities in the control of glycemia, blood pressure, and LDL-cholesterol for patients with diabetes and found that minorities have high rates of elevated blood pressure and elevated LDL-cholesterol.27,28 Thus, significant racial and ethnic differences do occur among diabetes clinical outcomes, and minimizing these differences is important to reduce mortality and morbidity in minority groups. Improved diabetes management will help address this issue.
As with any study, this study has limitations. While our results provide evidence of discrete, identifiable clinical disparities along racial and ethnic lines, the study was conducted at a single (albeit large urban) clinic; therefore, the results cannot be generalized. Furthermore, measurement of the rate for provision of counseling abstracted from chart reviews has its limitations, ie, a difference might exist between actual counseling and documentation of that counseling with regard to its content and extent. For example, recommending that patients follow a healthy diet differs from providing them with specific choices for a healthy diet, utilizing a 5-A behavioral counseling framework that has been shown to be effective in changing lifestyle behaviors.29 In our study, provision of counseling was reported in the affirmative even if the counseling was documented for just a single visit out of the total visits for the previous 3 years. More consistent and elaborate counseling could have had a greater impact on outcomes rather than that from counseling just once in 3 years.
We acknowledge that counseling during physician office visits is not as structured or elaborate as that at diabetes education classes, as indicated in studies revealing that primary care physicians spend very little time counseling about healthy lifestyle behaviors.30 On one hand, from 60% to 75% of patients received some sort of counseling in this study, a rate that is higher than those found in many studies.31,32 However, given that counseling should occur for every patient with diabetes at each medical encounter, the less-than-universal rate of counseling found in this study calls for additional efforts on strategies for increasing counseling. This is in line with chronic care guidelines stressing self-management33 as well as with obesity guidelines.34
We conclude that even in an integrated health care system such as our study setting with a very high adherence rate to diabetes treatment guidelines, significant racial and ethnic differences were found to exist in HbA1c levels. This is in spite of the insignificant racial and ethnic differences in diabetes self-management treatment protocols or health care utilization. These findings underscore the need for more research to examine and explain these observed differences. We reiterate that the less-than-universal rate of counseling found in this study calls for additional efforts on strategies to increase counseling given that diabetes counseling should be provided for every patient with diabetes.
This work was supported in part by a US NIH grant P20MD002295 awarded for the Program for Rural and Minority Health Disparities Research, which is a research collaborative between the Center for the Study of Health Disparities at the Texas A&M University and the Center for Community Health Development at the Texas A&M Health Science Center School of Rural Public Health. The authors acknowledge Phyllis Davis for administrative assistance.
- Centers for Disease Control and Prevention. National Diabetes Fact Sheet: General Information and National Estimates on Diabetes in the United States, 2007. Atlanta, GA: US Dept of Health and Human Services; Centers for Disease Control and Prevention, 2008. http://www.cdc.gov/diabetes/pubs/pdf/ndfs_2007.pdf [PDF]. Accessed February 25, 2009.
- American Diabetes Association. Many Faces of Diabetes. http://www.diabetes.org/pr-american-diabetes-month-alert-092607.jsp. Accessed February 25, 2009.
- Agency for Healthcare Research and Quality. National Healthcare Disparities Report, 2007. http://www.ahrq.gov/qual/nhdr07/nhdr07.pdf [PDF]. Accessed May 27, 2009.
- Peek ME, Cargill A, Huang ES. Diabetes health disparities: a systematic review of health care interventions. Med Care Res Rev. 2007;64(5 suppl):101S-156S.
- McBean AM, Li S, Gilbertson DT, Collins AJ. Differences in diabetes prevalence, incidence, and mortality among the elderly of four racial/ethnic groups: whites, blacks, Hispanics, and Asians. Diabetes Care. 2004;27(10):2317-2324.
- Mokdad AH, Bowman BA, Ford ES, Vinicor F, Marks JS, Koplan JP. The continuing epidemics of obesity and diabetes in the United States. JAMA. 2001;286(10):1195-1200.
- Sequist TD, Fitzmaurice GM, Marshall R, Shaykevich S, Safran DG, Ayanian JZ. Physician performance and racial disparities in diabetes mellitus care. Arch Intern Med. 2008;168(11):1145-1151
- Marshall MC Jr. Diabetes in African Americans. Postgrad Med J. 2005;81(962):734-740.
- Carter JS, Pugh JA, Monterrosa A. Non-insulin-dependent diabetes mellitus in minorities in the United States. Ann Intern Med. 1996;125(3):221-232.
- Lavery LA, Houtum WH, Ashry HR, Armstrong DG, Pugh JA. Diabetes-related lower-extremity amputations disproportionately affect blacks and Mexican Americans. South Med J. 1999;92(6):593-599.
- Lanting LC, Joung IM, Mackenbach JP, Lamberts SW, Bootsma AH. Ethnic differences in mortality, end-stage complications, and quality of care among diabetic patients: a review. Diabetes Care. 2005;28(9):2280-2288.
- Sundquist J, Winkleby MA, Pudaric S. Cardiovascular disease risk factors among older black, Mexican-American, and white women and men: an analysis of NHANES III, 1988-1994. Third National Health and Nutrition Examination Survey. J Am Geriatr Soc 2001;49(2):109-116.
- Nelson A. Unequal treatment: confronting racial & ethnic disparities in health care. J Natl Med Assoc. 2002;94(8):666-668.
- Heisler M, Smith DM, Hayward RA, Krein SL, Kerr EA. Racial disparities in diabetes care processes, outcomes, and treatment intensity. Med Care. 2003;41(11):1221-1232.
- Kamyar M, Johnson BJ, McAnaw JJ, Lemmens-Gruber R, Hudson SA. Adherence to clinical guidelines in the prevention of coronary heart disease in type II diabetes mellitus. Pharm World Sci. 2008;30(1):120-127.
- Kirkman MS, Williams SR, Caffrey HH, Marrero DG. Impact of a program to improve adherence to diabetes guidelines by primary care physicians. Diabetes Care. 2002;25(11):1946-1951.
- Morritt Taub LF. Concordance of provider recommendations with American Diabetes Association's guidelines. J Am Acad Nurse Pract. 2006;18(3):124-133.
- Texas Diabetes: the newsletter of the Texas Diabetes Council. http://www.dshs.state.tx.us/diabetes/PDF/newsletter/spring06.pdf [PDF]. Accessed March, 18, 2009.
- Scott and White Health Plan. http://www.swhp.org/homepage/tamu/geninfo. Accessed March 18, 2009.
- Laboratory Tests. MedlinePlus. US National Library of Medicine (http://www.nlm.nih.gov/medlineplus/laboratorytests.html) and American Association for Clinical Chemistry (http://www.labtestsonline.org/understanding/conditions/diabetes.html).
- US Department of Health and Human Services. Understanding and improving health. In: Healthy People: 2010. 2nd ed. I and II. Washington, DC: US Dept of Health and Human Services; 2000. http://www.healthypeople.gov/Publications. Accessed March 18, 2009.
- Kerr EA, Gerzoff RB, Krein SL, et al. Diabetes care quality in the Veterans Affairs Health Care System and commercial managed care: the TRIAD Study. Ann Intern Med. 2004;141(4):272-281.
- Summerson JH, Konen JC, Dignan MB. Race-related differences in metabolic control among adults with diabetes. South Med J. 1992;85(10):953-956.
- Rhee MK, Ziemer DC, Caudle J, Kolm P, Phillips LS. Use of a uniform treatment algorithm abolishes racial disparities in glycemic control. Diabetes Educ. 2008;34(4):655-663.
- Arnold-Worner N, Holle R, Rathmann W, Mielck A. The importance of specialist treatment, treatment satisfaction and diabetes education for the compliance of subjects with type 2 diabetes – results from a population-based survey. Exp Clin Endocrinol Diabetes. 2008;116(2):123-128.
- Rhee MK, Ziemer DC, Caudie J, et al. Patient adherence improves glycemic control. Diabetes Educ. 2005;31(2):240-250.
- Kirk JK, Bell RA, Bertoni AG, et al. Ethnic disparities: control of glycemia, blood pressure, and LDL cholesterol among US adults with type 2 diabetes. Ann Pharmocother. 2005;39(9):1489-1501.
- Kirk JK, Bell RA, Bertoni AG, et al. A qualitative review of studies of diabetes preventive care among minority patients in the United States, 1993-2003. Am J Manag Care. 2005;11(6):349-360.
- Sturt JA, Whitlock S, Fox C, et al. Effects of the Diabetes Manual 1:1 structured education in primary care. Diabet Med. 2008;25(6):722-731.
- Ory MG, Peck BM, Browning C, Forjuoh SN. Lifestyle discussions during doctor-older patient interactions: the role of time in the medical encounter. MedGenMed. 2007;9(4):48.
- Peek ME, Tang H, Alexander GC, Chin MH. National prevalence of lifestyle counseling or referral among African-Americans and whites with diabetes. J Gen Intern Med. 2008;23(11):1858-1864.
- Lin SX, Larson E. Does provision of health counseling differ by patient race? Fam Med. 2005;37(9):650-654.
- Bodenheimer T, Wagner EH, Gumbach K. Improving primary care for patients with chronic illnesses. JAMA. 2002;288(14):1775-1779.
- Screening for Obesity in Adults, Topic Page. December 2003. US Preventive Services Task Force. Agency for Healthcare Research and Quality, Rockville, MD. http://www.uspreventiveservicestaskforce.org/uspstf/uspsobes.htm. Accessed March 18, 2009.
November 2010 Texas Medicine Contents
Texas Medicine Main Page