- Email: [email protected]
Gender differences across racial and ethnic groups in the quality of care for diabetes
Women’s Health Issues 16 (2006) 56 – 65
GENDER DIFFERENCES ACROSS RACIAL AND ETHNIC GROUPS IN THE QUALITY OF CARE FOR DIABETES Rosaly Correa-de-Araujo, MD, MSc, PhDa*, Kelly McDermott, MAb, and Ernest Moy, MD, MPHa a
Agency for Healthcare Research and Quality, Rockville, Maryland b The Medstat Group, Inc, Rockville, Maryland
Received 4 October 2004; revised 24 February 2005; accepted 18 April 2005
High-quality care for diabetes is based on proper prevention, coordination of care among a multidisciplinary team of health care professionals, enhanced patient–provider relationships, and patient self-management skills. This paper discusses gender differences across racial and ethnic groups in the quality of care for type 2 diabetes according to 10 measures defined by the National Healthcare Quality Report and the National Healthcare Disparities Report. These measures include 5 process measures and one composite measure derived from the Medical Expenditure Panel Survey and 4 outcome measures derived from the Healthcare Cost and Utilization Project. National rates for 2 process measures—measurement of HbA1c (women 89.70% versus men 90.10%) and lipid profile (women 92.9% versus men 95.3%)—are high, but only 28.9% of women and 33.9% of men with diabetes received all 5 recommended process measures (HbA1c, lipid profile, eye exam, foot exam, and influenza immunization). Screening rates for retinal and foot exams and influenza immunization should be improved for all, but the need is particularly urgent for Hispanics and non-Hispanic blacks. Women and men have similar rates of hospital admissions for uncontrolled diabetes, but rates for lower extremity amputations were higher for men, particularly non-Hispanic blacks and Hispanics. Avoidable hospitalizations for diabetes decreased as income increased across racial/ethnic groups, but other factors (e.g., quality of primary care, age, relationship with providers, patients’ self-management skills) may influence such rates. Moreover, any improvements in the diabetes outcomes measures may lag many years behind any measurable improvements in quality of care. Well-designed interventions that reallocate resources for diabetes self-care should be developed to ensure that gender differences are addressed across racial/ethnic groups. Because much of this care involves the management of risk factors, self-management education should be tailored to the lifestyles and beliefs specific to gender and racial/ethnic groups.
Introduction Diabetes is the sixth leading cause of death in the United States, with mortality rates for adults with diabetes being twice that of the general population (American Diabetes Association, 2003a, 2003b). In 2005, 20.8 million people
The views expressed in this article are those of the authors and do not necessarily represent the views of the Agency for Healthcare Research and Quality or the Federal government * Correspondence to: Rosaly Correa-de-Araujo, MD, MSc, PhD, Director, Women’s Health and Gender-Based Research, Agency for Healthcare Research and Quality, 540 Gaither Road, Rockville, MD 20850. E-mail: [email protected]. Copyright © 2006 by the Jacobs Institute of Women’s Health. Published by Elsevier Inc.
had diabetes with 14.6 million people diagnosed and 6.2 million undiagnosed (Centers for Disease Control and Prevention [CDC], 2005). Diabetes affects 9.7 million women and 10.9 million men age 20 and over. Prevalence for the various racial/ethnic groups is: non-Hispanic whites (13.1 million), non-Hispanic blacks (3.2 million), Hispanic/Latino Americans (2.5 million), and American Indians/Alaska Natives (117,994). Prevalence is also high among people with lower educational levels. Diabetes prevalence in the general population is projected to increase by 44% by 2020: 107% for Hispanics and 56% for older adults (American Diabetes Association, 2002, 2003b). Diabetes-related mortality rates are higher among blacks, Native Americans, and Hispanics 1049-3867/06 $-See front matter. doi:10.1016/j.whi.2005.08.003
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
57
Table 1. NHQR/NHDR measure sets for diabetes Measure Process measures Hemoglobin A1c (HbA1c) measurement Lipid profile measurement Eye exam Foot exam Influenza immunization Composite measure of all 5 services Outcomes measures (avoidable hospitalizations) Uncontrolled diabetes Short-term complications Long-term complications Lower extremity amputations
Description
Data Sources National/State
Percent of adults with diabetes who had a hemoglobin A1c measurement at least once in the past year Percent of patients with diabetes who had a lipid profile in the past 2 years Percent of adults with diabetes who had a retinal eye exam in the past year Percent of adults with diabetes who had a foot exam in the past year Percent of adults with diabetes who had an influenza immunization in the past year Percent of adults who received all 5 services for diabetes in appropriate time frame
MEPS
Hospital admissions for uncontrolled diabetes per 100,000 population Hospital admissions for short-term complications of diabetes per 100,000 population Hospital admissions for long-term complications of diabetes per 100,000 population Hospital admissions for lower extremity amputations in patients with diabetes per 100,000 population
HCUP SID
MEPS MEPS MEPS MEPS MEPS
HCUP SID HCUP SID HCUP SID
Abbreviations: MEPS, Medical Expenditure Panel Survey, 2000 – 01; HCUP SID, Healthcare Cost and Utilization Project, State Inpatient Database, 2001.
(Mokdad et al., 2000). Diabetes is associated with a range of other illnesses and is a major risk factor for cardiovascular disease. People with diabetes are at increased risk for stroke, ischemic heart disease, peripheral vascular disease, and neuropathy (American Diabetes Association, 2002, 2003b). Blacks have higher rates of serious complications from diabetes, including higher rates of end-stage renal disease and lower extremity amputation (CDC, 1999; Guadagnoli, Ayanian, Gibbons, McNeil, & LoGerfo, 1995; Gornick et al., 1996). Diabetes is a public health and economic concern. The total cost of the disease in the United States for 2002 was estimated at $132 billion, of which $91.8 billion was attributed to direct medical costs and $40 billion to indirect costs owing to disability, work loss, and premature mortality (American Diabetes Association, 2003a). Diabetes is a preventable disease that can be effectively managed to delay or avoid its complications (CDC, 2004; Heisler, Vijan, Anderson, Ubel, Bernstein, & Hofer, 2003; Hill-Briggs, Cooper, Loman, Brancati, & Cooper, 2003). To identify gaps in care and avoid unnecessary expense, monitoring the ongoing quality of health care in patients with diabetes is crucial. Despite evidence currently available on the best practices in diabetes care, there is still wide variation in diagnostic evaluation, use of preventive services, and the quality and extent of disease management (American Diabetes Association, 2003b; Diabetes Prevention Program Research Group, 2002; Dallo & Weller, 2003). The purpose of this study is to investigate whether
gender differences across racial/ethnic groups exist in the quality of care received by people who suffer from type 2 diabetes. The quality of care for diabetes is evaluated according to 10 process and outcomes measures as defined by the National Healthcare Quality Report (NHQR) and the National Healthcare Disparities Report (NHDR) (Agency for Healthcare Research and Quality [AHRQ], 2004a, 2004b, 2005a, 2005b). The paper’s unique contribution is that it goes beyond the scope of the national reports by performing additional data analysis by gender within racial/ethnic groups. Our findings provide the basis for future development of gender- and/or race/ethnicity-specific strategies to help close the gaps in diabetes care.
Methods Data sources Medical Expenditure Panel Survey. The Medical Expenditure Panel Survey (MEPS) collects data through computer-assisted, in-person interviews of a nationally representative sample of the noninstitutionalized civilian population using a stratified multistage probability design. This analysis uses data from the MEPS Household Component as well as the Diabetes Care Survey supplement of the MEPS, which is a paperand-pencil questionnaire administered to household respondents who answered “yes” when asked whether
58
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
Table 2. Preventive measures for diabetes by gender across race/ethnicity: MEPS 2000 – 01 Non-Hispanic white (%)
Total (%) Measure* Percent of adults with diabetes who had a hemoglobin A1c measurement at least once in the past year Percent of patients with diabetes who had a lipid profile in past 2 years Percent of adults with diabetes who had a retinal eye examination in past year Percent of adults with diabetes who had a foot examination in past year Percent of adults with diabetes who had an influenza immunization in past year Percent of adults with diabetes who received all five recommended diabetes services in appropriate time frame
Non-Hispanic black (%)
Hispanic (%)
p-value NHW to NHB
p-value NHW to Hispanic
W
M
pvalue
W
M
pvalue
W
M
pvalue
W
M
pvalue
W
M
W
M
89.70
90.10
.822
91.00
91.50
.814
88.90
85.30
.471
83.90
84.60
.888
0.490
0.168
0.037
0.102
92.90
95.30
.035
92.30
96.30
.011
96.80
91.50
.077
89.70
92.10
.408
0.023
0.081
0.292
0.055
66.60
69.80
.158
67.40
73.10
.049
68.70
63.20
.338
58.00
57.40
.910
0.762
0.039
0.013
0.001
63.70
69.10
.021
64.20
71.30
.026
62.30
67.50
.295
63.30
60.90
.639
0.620
0.398
0.818
0.023
55.10
56.30
.655
57.90
61.40
.271
47.10
42.50
.439
54.60
41.00
.013
0.019
0.000
0.466
0.000
28.90
33.90
.094
30.60
37.40
.065
24.20
26.30
.701
26.70
20.50
.274
0.137
0.026
0.425
0.000
Bolded p value indicates a statistically significant difference between women and men or between racial/ethnic groups (p ⱕ .05). NHW, non-Hispanic whites; NHB, non-Hispanic blacks; W, women; M, men. Source: Agency for Healthcare Research and Quality. Center for Financing. Access and Cost Trends. Medical Expenditure Panel Survey combining 2000 and 2001 data. *Denominator for summary measure excludes missing values.
they were ever told by a doctor or other health professional that they had diabetes (AHRQ, 2004c). Each year, MEPS collects data from a new sample of households. For this article, data from 2000 and 2001 were pooled to create a larger sample size to assess subgroup populations. The total sample size for the 2 years was 2,365 adults with diabetes. Adjustments were made for pooled variance to calculate standard errors. Data for this report were suppressed if cell size
was ⬍100 or the relative standard error was ⬎30% of the estimate (AHRQ, 2004c). Because the MEPS data are self-reported, certain factors such as recall bias and social desirability in answering questions may limit the accuracy of the data. Healthcare cost and utilization project. The Healthcare Cost and Utilization Project (HCUP) is a family of health care databases and products developed
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
59
Table 3. Diabetes measures—logistic regressions: MEPS, 2000 – 01 Odds Ratio (95% Confidence Interval)
Characteristic Female vs male (reference) Race/ethnicity Hispanic Non-Hispanic black Non-Hispanic white (reference) Income Poor Near poor Middle High (reference) Education Less than high school High school graduate Any college (reference)
HbA1c in past year
Lipid profile in past 2 years
Retinal eye exam in past year
Foot exam in past year
Influenza vaccine in past year
All 5 services
1.04 (0.67, 1.60)
0.74 (0.48, 1.14)
0.92 (0.74, 1.15)
0.78 (0.62, 0.98)
0.97 (0.78, 1.21)
0.83 (0.64, 1.08)
0.71 (0.39, 1.27) 0.76 (0.47, 1.24)
0.87 (0.53, 1.42) 1.41 (0.72, 2.77)
0.73 (0.54, 0.99) 0.93 (0.67, 1.28)
0.82 (0.60, 1.13) 0.85 (0.63, 1.15)
0.81 (0.59, 1.11) 0.59 (0.44, 0.80)
0.73 (0.45, 1.17) 0.67 (0.48, 0.93)
1.00
1.00
1.00
1.00
1.00
1.00
0.68 (0.41, 1.14) 0.97 (0.56, 1.70) 0.95 (0.60, 1.53) 1.00
0.39 (0.18, 0.82) 0.55 (0.26, 1.20) 0.62 (0.33, 1.17) 1.00
0.65 (0.47, 0.91) 0.65 (0.49, 0.86) 0.71 (0.54, 0.93) 1.00
1.07 (0.73, 1.56) 1.08 (0.77, 1.51) 0.97 (0.71, 1.34) 1.00
0.78 (0.54, 1.14) 0.91 (0.67, 1.24) 0.79 (0.61, 1.03) 1.00
0.76 (0.44, 1.31) 1.09 (0.73, 1.62) 1.10 (0.78, 1.54) 1.00
0.58 (0.35, 0.96) 0.71 (0.43, 1.19) 1.00
0.68 (0.36, 1.28) 0.82 (0.46, 1.46) 1.00
0.64 (0.48, 0.84) 0.71 (0.55, 0.93) 1.00
0.87 (0.66, 1.15) 0.83 (0.65, 1.07) 1.00
0.67 (0.50, 0.89) 0.72 (0.53, 0.98) 1.00
0.64 (0.45, 0.91) 0.59 (0.42, 0.83) 1.00
Note: Models include age, insurance, race/ethnicity, income, education, and place of residence. Bolded values indicate a statistically significant difference (95% confidence interval does not include 1).
through a public–private partnership sponsored by AHRQ (Healthcare Cost and Utilization Project, 2004). Statewide data organizations, including state agencies, state hospital associations, and private data consortia, provide discharge-level data for all hospitals in their state to AHRQ for the development of information research databases. The HCUP State Inpatient Databases (SID) include records for all discharges for all hospitals in more than 35 states (AHRQ, 2004c). A file designed to provide national estimates on disparities was developed using a sample of community hospitals from the 2001 SID data for 22 states that participate in HCUP and collect data on race and ethnicity. In total, these states accounted for 65% of all hospital discharges in the United States in 2001. Hospitals were excluded from the sampling frame if the coding of patient race was suspect. Weights were developed to provide national estimates of disparities. Data were suppressed if the count in the denominator of the rates was ⬍70 cases to ensure that relative standard errors were ⬍30% (Coffey, Barrett, et al., 2004). NHQR and NHDR measure sets This paper uses 10 diabetes measures based on the NHQR and NHDR (Table 1). There are 5 process measures of services, a composite measure for receipt of all 5 such services, and 4 outcome measures of diabetes-related hospital admissions. The 5 process measures and the composite measure use MEPS and the 4 outcome measures use HCUP data. The process measures are part of the best standard of care for diabetes and are used to monitor disease progress,
help to control the disease, and avoid or delay hospitalizations due to short- and long-term complications. Statistical analysis Comparisons were made on data stratified by gender across racial/ethnic groups. Two-tailed t-tests were used to assess significance. Although p-values were considered statistically significant at an alpha level of .05, additional criteria were imposed on the comparisons to identify “important differences,” because the large sample sizes yielded highly statistically significant results even for small differences. Relative differences ⬎10% were defined as statistically important. Statistically important differences are indicated in the tables. Because racial and ethnic minorities are disproportionately likely to be of lower socioeconomic status, health care disparities among racial and ethnic minorities are often highly correlated with disparities that fall along socioeconomic lines. To begin to disaggregate racial, ethnic, and socioeconomic effects, multivariate models were developed for MEPS data. These logistic regression models included age, gender, race/ ethnicity, household income, education, insurance, and location of residence. HCUP uses the median household income of a patient’s community as a proxy for socioeconomic status, created by linking the patient’s zip code on the SID with income information obtained from Claritas, a statistical package that provides intercensal estimates and projections. Data include age- and gender-adjusted rates per 100,000 population stratified by area income using Zip code-level counts by age, gender,
60
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
Table 4. Avoidable diabetes-related hospitalizations by gender across race/ethnicity: HCUP, 2001 Population by gender and race/ethnicity
Measure Hospital admissions for uncontrolled diabetes per 100,000 population Hospital admissions for short-term complications of diabetes per 100,000 population Hospital admissions for long-term complications of diabetes per 100,000 population Hospital admissions for lower extremity amputations in patients with diabetes per 100,000 population
Total (per 100,000 population)
NonHispanic White (per 100,000 population)
Non-Hispanic Black (per 100,000 population)
Hispanic (per 100,000 population)
W
M
W
M
W
M
W
M
W
M
W
M
25
27
16
18
79
91
45
42
†
†
†
†
53
57
44
42
133
187
49
54
†
†
106
136
76
109
300
333
181
199
†
†
†
†
28
55
19
45
93
134
45
84
†
†
†
†
NHW to Black
NHW to Hispanic
†
W, women; M, men; NHW, non-Hispanic whites; NHB, non-Hispanic blacks. Rates reported by gender have been age-adjusted. Bolded values indicate the difference is important with a p-value ⱕ .05 and a relative difference ⬎10%, comparing women to men within each racial/ethnic group. †The difference is important with a p-value ⱕ .05 and a relative difference ⬎10% comparing whites to blacks or whites to hispanics by gender. Source: Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project, State Inpatient Databases, disparities analysis file, 2001.
race, and ethnicity from Claritas (Coffey, Barrett, et al., 2004). All analyses accounted for the complex survey design of the 2 databases. MEPS analyses used SUDAAN, and HCUP analyses used the SAS PROC SURVEYMEANS function.
Results Results are presented by gender for the following racial/ethnic groups: non-Hispanic whites, non-Hispanic blacks, and Hispanics. Gender analysis across other racial/ethnic groups (e.g., Asians, Pacific Islanders, and Native Americans/Alaska Natives) was not possible because data were found to be of nonreliable statistical significance (sample size inadequate). These populations were, therefore, excluded from the study.
Preventive measures Table 2 displays information on differences in preventive measures for diabetes care for the total population of women and men across racial/ethnic groups. Differences by gender. The percentage of adults receiving secondary preventive services for diabetes differed by gender for lipid profiles and foot exams. Among non-Hispanic whites, women were less likely to have a lipid profile or receive retinal eye and foot exams. Although such gender differences were not observed among non-Hispanic blacks or Hispanics, Hispanic men were less likely than Hispanic women to receive an influenza immunization. Differences by race/ethnicity. Overall, the receipt of the composite measure of all 5 recommended diabetes services was comparable among all subgroups of
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
61
Table 5. Avoidable diabetes-related hospitalizations by community income levels across race/ethnicity: HCUP, 2001 Total (per 100,000 population)
Measure Hospital admissions for uncontrolled diabetes per 100,000 population Hospital admissions for short-term complications of diabetes per 100,000 population Hospital admissions for long-term complications of diabetes per 100,000 population Hospital admissions for lower extremity amputations in patients with diabetes per 100,000 population
Non-Hispanic White (per 100,000 population)
Average Average $25,000– $35,000– decrease $25,000– $35,000– decrease ⬍$25,000 $34,999 $44,999 ⱖ$45,000 (%) ⬍$25,000 $34,999 $44,999 ⱖ$45,000 (%) 70
43
24
15
40
35
29
17
11
32
115
76
57
40
30
66
59
46
34
20
221
152
118
95
24
123
109
93
80
13
68
51
41
32
22
37
36
33
27
10
Bolded values indicate that the difference is important with a p value ⱕ .05 and a relative difference ⬎10% comparing each income level to the highest income level (ⱖ$45,000), within each racial/ethnic group. —Data do not meet the criteria for statistical reliability, data quality or confidentiality or are otherwise unavailable. Source: Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project, State Inpatient Databases, disparities analysis file, 2001.
women. However, compared to non-Hispanic white women, non-Hispanic black women received fewer lipid profiles and influenza immunizations, and Hispanic women received fewer HbA1c measurements and retinal eye exams. Compared to non-Hispanic white men, non-Hispanic black and Hispanic men were less likely to have eye exams, be immunized against influenza, and receive all 5 of the recommended diabetes services. Hispanic men also received fewer foot exams than non-Hispanic white men.
observed, but non-Hispanic white men were 43% more likely than non-Hispanic white women to be hospitalized for long-term complications and nonHispanic black men were nearly 41% more likely than non-Hispanic black women to be hospitalized for short-term complications. Moreover, men and women differed significantly overall and across all 3 racial/ ethnic groups in the rate of hospitalization for lower extremity amputation, with men being much more likely to receive the procedure.
Logistic regression analysis. Table 3 shows results of the logistic regression models. These showed that socioeconomic status is an important determinant of diabetes care. People with low incomes were less likely to receive retinal eye exams and lipid profiles, and those with less education were less likely to receive HbA1c measurements, retinal eye exams, influenza immunizations, and all 5 services for diabetes. After controlling for age, insurance, income, education, and place of residence, some racial, ethnic, and gender disparities persisted. Non-Hispanic blacks were less likely than non-Hispanic whites to be immunized against influenza and receive all 5 services for diabetes, whereas Hispanics received fewer retinal eye exams. Regarding gender differences, women were less likely than men to receive foot exams.
Differences by racial/ethnic groups. Non-Hispanic black and Hispanic women and men had higher rates than their white counterparts for all but 1 measure of diabetes complications and hospitalizations. The exception was hospitalizations for short-term complications in non-Hispanic white women compared to Hispanic women. Table 5 shows community income levels across racial/ethnic groups. Rates of hospitalization decreased as income level increased for each racial/ ethnic group. For the total population, hospital admissions for uncontrolled diabetes decreased by approximately 40% at each increasing level of community income. In addition, for all racial/ethnic groups combined, hospital admissions across the 4 avoidable hospitalization measures were statistically different for income groups of ⬍$45,000 per year compared to those of ⱖ$45,000 per year. Relative risk ratios calculated using data from Table 5 show that, across all income categories, the relative risk of hospital admissions for non-Hispanic blacks over non-Hispanic whites was approximately 3.5 for long-term complications and 3.7 for short-term complications and lower extremity amputations. Compared to non-Hispanic
Avoidable hospitalizations Table 4 displays information on avoidable diabetesrelated hospitalizations per 100,000 population for women and men across racial/ethnic groups. Differences by gender. Overall, no gender differences in hospital admissions for uncontrolled diabetes were
62
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
Table 5. Continued Non-Hispanic Black (per 100,000 population)
Hispanic (per 100,000 population) Average decrease (%)
⬍$25,000
$35,000– $44,999
131.00
110.00
68
56
24
—
58
36
28
20
244.00
181.00
148
116
22
71
65
48
41
16
410.00
366.00
292
252
15
259
235
168
162
14
131.00
134.00
105
94
10
82
77
56
54
12
ⱖ$45,000
whites, the risk of hospitalization for uncontrolled diabetes was 2.5 times greater for Hispanics and 5 times greater for non-Hispanic blacks.
Discussion Diabetes is a complex chronic disease requiring comprehensive quality care. Studies have shown that when appropriate care is provided, lower diabetesrelated stress and fewer emergency room and doctor’s office visits are reported (CDC, 2004; Heisler et al., 2003; Hill-Briggs et al., 2003). Although research on the management of diabetes has not concentrated on men or women specifically (Shojania, McDonald, Wachter & Owens, 2004), the disparities in quality of care between genders and across racial and ethnic groups provide us with an opportunity to specifically target populations to improve diabetes management. Providing preventive services is crucial to identify and manage people whose diabetes is uncontrolled with test results outside the normal limits. In addition, test results provide health care professionals and patients who are not doing well with the opportunity to improve their relationship and work together toward controlling the disease. Keeping diabetes under control is essential for delaying or reducing the progression of microvascular complications. Despite the high rates of HbA1c and lipid profile measurements for both women and men, there is still room for improvement. Similarly, the percentages of retinal eye and foot exams and influenza immunizations should be increased for both women and men across all racial/ethnic groups. One major concern is the low percentages of women and men with diabetes who reported receiving all 5 recommended services for the disease in the appropriate time frame (28.9% for women and 33.9% for men). MEPS data show a significant increase in retinal eye
⬍$25,000
$25,000– $34,999
$35,000– $44,999
Average decrease (%)
$25,000– $34,999
ⱖ$45,000
exams for high school graduates with diabetes from 2000 (61%) to 2001 (70%), but gaps in retinal eye exams continue to exist for low-income people with diabetes. The disparities observed in relation to influenza immunization also have been reported in previous studies and have been associated with a variety of factors. For example, in a Medicare population of patients with diabetes, black patients with less than a high school education had fewer influenza vaccinations and HbA1c measurements (Chin, Zhang, & Merrell, 1998) than those with none education; in another study, receipt of an annual influenza immunization was independently associated with race but related to better glycemic control (de Rekeneire et al., 2003). Overall, rates of hospital admissions for uncontrolled diabetes have significantly decreased since 1994 from 40.7 per 100,000 population to the current rates of 25 for women and 27 for men per 100,000 population. The highest hospitalization rates are linked to long-term complications of diabetes, particularly the high rates for non-Hispanic blacks and Hispanics. Another challenge involves reducing the number of hospitalizations for lower extremity amputations. Amputation rates for men were notably high for non-Hispanic blacks (134 per 100,000 population) and Hispanics (84 per 100,000 population); among women, they were 5 times higher for non-Hispanic blacks and 2 times higher for Hispanics than for whites. If the quality of health services for diabetes improves, it may still take several years to detect improvements in the rates of long-term diabetes complications including amputations. Generally, people with lower incomes and less education have worse health status and experience worse health care than those with higher incomes and more education (Pamuk, Makuc, Heack, Reuben, & Lockner, 1998). We found that avoidable hospitalizations for diabetes complications decreased as income
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
increased across racial/ethnic groups. Low socioeconomic status seems to be clearly related to the gender and racial/ethnic differences in hospitalization rates. However, factors such as quality of primary care, age, relationship with providers, and patients’ self-management skills also play a role. In fact, within the complexity of factors affecting the management of diabetes, numerous other patient-associated factors (e.g., job benefits, child care, transportation) may lead to missing scheduled medical appointments and consequent variability in the number of services patients receive. These factors reduce the continuity and effectiveness of health care delivery, cause lapses in the appropriate monitoring of health status, increase the chances of developing poor disease outcomes, and elevate the cost of health services (Karter et al., 2004). Problems in reliable evaluation of the effects of income, education, and insurance on gender and racial/ethnic groups constitute a major weakness in this study and were a major challenge during the development of the NHQR/NHDR (Arispe, Holmes, & Moy, 2005). First, because numerous factors can affect the outcomes for the measures used here, solid interpretation of the relationships between such outcomes and the quality of diabetes care has proven to be quite difficult (Moy, Arispe, Holmes, & Andrews, 2005). Second, data on specific racial, ethnic, and socioeconomic groups frequently are not collected or are insufficient for generating consistent and credible estimates (Moy et al., 2005). Enhanced data collection, therefore, is crucial to better data analysis and interpretation and supports the development of strategies targeting the elimination of disparities in diabetes care. Learning about the gaps in the delivery of diabetes secondary preventive services is very important but has limited value if the next step (i.e., management of individuals who do not have their diabetes under control) is not properly addressed. The diabetes outcome measures discussed in this paper are likely to be very distant outcomes from simple physician’s office testing such as foot exams, lipid profiles, and HbA1c measurement. These tests are often completed in the same office visit. Differences in rates of completion of these health services may affect both clinician and patient education. Because of the link between care and outcomes and the existence of high-quality evidence on the effectiveness of prevention and treatment of diabetes, the disease is a prototypical model for self-management of chronic diseases. Therefore, high-quality care for diabetes should encompass an enhanced patient–provider relationship to help improve patients’ self-management skills. Effective patient–provider communication, particularly when associated with agreement on management goals and strategies, has been shown to contribute to higher patient self-efficacy and self-management skills
63
(Heisler et al., 2003). Several other studies also have shown that patient education improves self-monitoring of glycemic levels and control (Cowie & Harris, 1997; Harris, Eastman, Cowie, Flegal, & Eberhardt, 1999). More recently, coverage of diabetes blood glucose monitors by health maintenance organizations resulted in more patients self-monitoring their blood glucose, increased regularity in the use of anti-diabetes medications, and ultimately reduced blood glucose levels (Soumerai et al., 2004). In addition, the disparities observed here require action toward the development of well-designed interventions that reallocate resources for diabetes selfcare while ensuring that gender differences are addressed across racial/ethnic groups. Because much of this care involves the management of risk factors, self-management education should be tailored to the lifestyles and beliefs specific to gender and racial/ ethnic groups. One effort undertaken by AHRQ in disseminating information about diabetes that can help with the development of targeted well-designed quality improvement strategies to eliminate disparities is a resource guide and workbook for state-level action to improve diabetes care. The guide and workbook, companions to the NHQR and NHDR, use data compiled in the reports and walk users through data procedures, examining the finer points of data collection and analysis and also presenting broader case studies of state-level efforts. The model emphasizes data collection as the first step toward quality improvement and makes suggestions for implementation (Coffey, Matthews, & McDermott, 2004; Kass, 2004). It provides state leaders with the opportunity to monitor their progress in the quality of diabetes care and to exchange experiences about their successes and failures.
Acknowledgments The authors would like to acknowledge the statewide data organizations that participate in the 2001 HCUP Nationwide Inpatient Sample (NIS): Arizona Department of Health Services; California Office of Statewide Health Planning & Development; Colorado Health & Hospital Association; Connecticut–Chime, Inc.; Florida Agency for Health Care Administration; Georgia Hospital Association; Hawaii Health Information Corporation; Illinois Health Care Cost Containment Council; Iowa Hospital Association; Kansas Hospital Association; Kentucky Department for Public Health; Maine Health Data Organization; Maryland Health Services Cost Review Commission; Massachusetts Division of Health Care Finance and Policy; Michigan Health and Hospital Association; Minnesota Hospital Association; Missouri Hospital Industry Data Institute; Nebraska Hospital Association; New Jersey Department of Health & Senior Services; New York State Department of Health; North Carolina Department of Health and Human Services; Ore-
64
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
gon Association of Hospitals & Health Systems; Pennsylvania Health Care Cost Containment Council; Rhode Island Department of Health; South Carolina State Budget & Control Board; Tennessee Hospital Association; Texas Health Care Information Council; Utah Department of Health; Vermont Association of Hospitals and Health Systems; Virginia Health Information; Washington State Department of Health; West Virginia Health Care Authority; Wisconsin Department of Health & Family Services.
References Agency for Healthcare Research and Quality. (2005a). National Healthcare Disparities Report. (in press) Agency for Healthcare Research and Quality. (2005b). National Healthcare Quality Report. (in press) Agency for Healthcare Research and Quality. (2004a). 2004 National Healthcare Disparities Report. Rockville, MD: U.S. Department of Health and Human Services. Agency for Healthcare Research and Quality. (2004b). 2004 National Healthcare Quality Report. Rockville, MD: U.S. Department of Health and Human Services. Agency for Healthcare Research and Quality. (2004c). 2004 National Healthcare Quality Report, Measure Specifications Appendix. Rockville, MD: U.S. Department of Health and Human Services. American Diabetes Association. (2002). Diabetes fact sheet. Available: http://www.diabetes.org. Accessed July 22, 2004. American Diabetes Association. (2003a). Economic costs of diabetes in the US in 2002. Diabetes Care, 26, 917–932. American Diabetes Association. (2003b). Standards of medical care for patients with diabetes mellitus. Diabetes Care, 17, 1514 –1522. Arispe, I. E., Holmes, J. S., & Moy, E. (2005). Measurement challenges in developing the National Healthcare Quality Report and the National Healthcare Disparities Report. Medical Care, 43, I17–I23. Centers for Disease Control and Prevention (CDC). (1999). Diabetes Surveillance, 1999. Atlanta: National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention. Available: http://www.cdc.gov. Accessed July 23, 2004. Centers for Disease Control and Prevention (CDC). (2005). National diabetes fact sheet, 2005. Atlanta: U.S. Department of Health and Human Services. Centers for Disease Control and Prevention. Available: http://www.cdc.gov. Accessed July 23, 2004. Centers for Disease Control and Prevention (CDC). (2004). At a glance, diabetes: Disabling, deadly, and on the rise. Atlanta: National Center for Chronic Disease Prevention and Health Promotion, CDC. Available: http://www.cdc.gov. Accessed August 2, 2004. Chin, M. H., Zhang, J. X., & Merrell, K. (1998). Diabetes in the African-American population. Morbidity, quality of care, and resource utilization. Diabetes Care, 21, 1090 –1095. Coffey R, Barrett M, Houchens B, Moy E, Andrews R, Kelley E., et al. (2004). Methods Applying AHRQ Quality Indicators to Healthcare Cost and Utilization Project (HCUP) Data for the National Healthcare Disparities Report. Draft. Coffey, R. M., Matthews, T. L., & McDermott, K. (2004). Diabetes care quality improvement: A resource guide for State action (Prepared by The Medstat Group, Inc. and The Council of State Governments under Contract No. 290-00-0004; AHRQ Pub. No. 04-0072). Rockville, MD: Agency for Healthcare Research and Quality, Department of Health and Human Services. Cowie, C. C., & Harris, M. (1997). Ambulatory medical care for non-Hispanic whites, African Americans, and Mexican-Americans with NIDDM in the U.S. Diabetes Care, 20, 142–147.
Dallo, F. J., & Weller, S. C. (2003). Effectiveness of diabetes mellitus screening recommendations. Proceedings of the National Academies of Science, 100, 10574 –10579. de Rekeneire, N., Roos, R. N., Simonsick, E. M., Shorr, R. I., Kuller, L. H., Scwartz, A. V., et al. (2003). Racial differences in glycemic control in a well-functioning older diabetic population. Diabetes Care, 26, 1986 –1992. Diabetes Prevention Program Research Group. (2002). The Diabetes Prevention Program (DPP): Description of lifestyle intervention. Diabetes Care, 25, 2165–2171. Gornick, M. E., Eggers, P. W., Reilly, T. W., Mentnech, R. M., Fitterman, L. K., Kucken, L. E., et al. (1996). Effects of race and income on mortality and use of services among Medicare beneficiaries. New England Journal of Medicine, 335, 791–799. Guadagnoli, E., Ayanian, J. Z., Gibbons, G., McNeil, B. J., & LoGerfo, F. W. (1995). The influence of race on the use of surgical procedures for treatment of peripheral vascular disease of the lower extremities. Archives of Surgery, 130, 381–386. Harris, M. I., Eastman, R. C., Cowie, C. C., Flegal, K. M., & Eberhardt, M. S. (1999). Racial and ethnic differences in glycemic control of adults with type two diabetes. Diabetes Care, 22, 403– 408. Healthcare Cost and Utilization Project (HCUP). (2004). HCUP Databases. Rockville, MD: Agency for Healthcare Research and Quality. Available: www.hcup-us.ahrq.gov. Accessed July 23, 2004. Heisler, M., Vijan, S., Anderson, R. M., Ubel, P. A., Bernstein, S. J., & Hofer, T. P. (2003). When do patients and their physicians agree on diabetes treatment goals and strategies, and what difference does it make? Journal of General Internal Medicine, 18, 893–902. Hill-Briggs, F., Cooper, D. C., Loman, K., Brancati, F. L., & Cooper, L. A. (2003) A qualitative study of problem solving and diabetes control in type 2 diabetes self-management. Diabetes Education, 29, 1018 –1028. Karter, A. J., Parker, M. M., Moffet, H. H., Ahmed, A. T., Ferrara, A., Liu, J.Y., et al. (2004). Missed appointments and poor glycemic control: An opportunity to identify high-risk diabetic patients. Medical Care, 42, 110 –115. Kass, B. (2004, September). Diabetes care quality improvement: A workbook for state action (AHRQ Pub No. 04-0073). Rockville, MD: Agency for Healthcare Research and Quality, Department of Health and Human Services. Mokdad, A. H., Ford, E. S., Bowman, B. A., Nelson, D. E., Engelgau, M. M., Vinicor F., et al. (2000). Diabetes trends in the U.S.: 1990 –1998. Diabetes Care, 23, 1278 –1283. Moy, E., Arispe, I., Holmes, J., & Andrews, R. (2005). Preparing the National Healthcare Disparities Report: Gaps in data for assessing racial, ethnic, and socioeconomic disparities in health care. Medical Care, 43, I9 –I16. Pamuk, E., Makuc, D., Heack, K., Reuben, C., & Lockner, K. (1998). Socioeconomic status and health chartbook: Health, United States, 1998. Hyattsville, MD: National Center for Health Statistics. Shojania, K. G., McDonald, K. M., Wachter, R. M., & Owens, D. K. (2004) Closing the quality gap: A critical analysis of diabetes care strategies (AHRQ Publication No. 04-0051-2). Rockville, MD: Agency for Healthcare Research and Quality. Soumerai, S. B., Mah, C., Zhang, F., Adams, A., Barton, M., Fajtova V., et al. (2004). Effects of health maintenance organization coverage of self-monitoring devices on diabetes self care and glycemic control. Archives of Internal Medicine, 164, 645– 652.
Author Description Rosaly Correa-de-Araujo, MD, MSc, PhD, is a cardiovascular pathologist trained at the National Heart, Lung, and Blood Institute. As the Agency for Healthcare Research and Quality’s Director of Women’s
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
Health and Gender-Based Research, Dr. Correa oversees the development of a national research agenda for women in consultation with prominent members of the research community and other government agencies. Her main areas of interest include gender-based research and analysis particularly related to chronic diseases, medication use outcomes and safety, and disparities in health care.
65
Kelly McDermott, MA, is currently a predoctoral student in health services research at the University of Washington in Seattle. Ernest Moy, MD, MPH, is a Senior Service Fellow with the Center for Quality Improvement and Patient Safety in the Agency for Healthcare Research and Quality. Dr. Moy leads the development of the National Healthcare Disparities Report.
GENDER DIFFERENCES ACROSS RACIAL AND ETHNIC GROUPS IN THE QUALITY OF CARE FOR DIABETES Rosaly Correa-de-Araujo, MD, MSc, PhDa*, Kelly McDermott, MAb, and Ernest Moy, MD, MPHa a
Agency for Healthcare Research and Quality, Rockville, Maryland b The Medstat Group, Inc, Rockville, Maryland
Received 4 October 2004; revised 24 February 2005; accepted 18 April 2005
High-quality care for diabetes is based on proper prevention, coordination of care among a multidisciplinary team of health care professionals, enhanced patient–provider relationships, and patient self-management skills. This paper discusses gender differences across racial and ethnic groups in the quality of care for type 2 diabetes according to 10 measures defined by the National Healthcare Quality Report and the National Healthcare Disparities Report. These measures include 5 process measures and one composite measure derived from the Medical Expenditure Panel Survey and 4 outcome measures derived from the Healthcare Cost and Utilization Project. National rates for 2 process measures—measurement of HbA1c (women 89.70% versus men 90.10%) and lipid profile (women 92.9% versus men 95.3%)—are high, but only 28.9% of women and 33.9% of men with diabetes received all 5 recommended process measures (HbA1c, lipid profile, eye exam, foot exam, and influenza immunization). Screening rates for retinal and foot exams and influenza immunization should be improved for all, but the need is particularly urgent for Hispanics and non-Hispanic blacks. Women and men have similar rates of hospital admissions for uncontrolled diabetes, but rates for lower extremity amputations were higher for men, particularly non-Hispanic blacks and Hispanics. Avoidable hospitalizations for diabetes decreased as income increased across racial/ethnic groups, but other factors (e.g., quality of primary care, age, relationship with providers, patients’ self-management skills) may influence such rates. Moreover, any improvements in the diabetes outcomes measures may lag many years behind any measurable improvements in quality of care. Well-designed interventions that reallocate resources for diabetes self-care should be developed to ensure that gender differences are addressed across racial/ethnic groups. Because much of this care involves the management of risk factors, self-management education should be tailored to the lifestyles and beliefs specific to gender and racial/ethnic groups.
Introduction Diabetes is the sixth leading cause of death in the United States, with mortality rates for adults with diabetes being twice that of the general population (American Diabetes Association, 2003a, 2003b). In 2005, 20.8 million people
The views expressed in this article are those of the authors and do not necessarily represent the views of the Agency for Healthcare Research and Quality or the Federal government * Correspondence to: Rosaly Correa-de-Araujo, MD, MSc, PhD, Director, Women’s Health and Gender-Based Research, Agency for Healthcare Research and Quality, 540 Gaither Road, Rockville, MD 20850. E-mail: [email protected]. Copyright © 2006 by the Jacobs Institute of Women’s Health. Published by Elsevier Inc.
had diabetes with 14.6 million people diagnosed and 6.2 million undiagnosed (Centers for Disease Control and Prevention [CDC], 2005). Diabetes affects 9.7 million women and 10.9 million men age 20 and over. Prevalence for the various racial/ethnic groups is: non-Hispanic whites (13.1 million), non-Hispanic blacks (3.2 million), Hispanic/Latino Americans (2.5 million), and American Indians/Alaska Natives (117,994). Prevalence is also high among people with lower educational levels. Diabetes prevalence in the general population is projected to increase by 44% by 2020: 107% for Hispanics and 56% for older adults (American Diabetes Association, 2002, 2003b). Diabetes-related mortality rates are higher among blacks, Native Americans, and Hispanics 1049-3867/06 $-See front matter. doi:10.1016/j.whi.2005.08.003
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
57
Table 1. NHQR/NHDR measure sets for diabetes Measure Process measures Hemoglobin A1c (HbA1c) measurement Lipid profile measurement Eye exam Foot exam Influenza immunization Composite measure of all 5 services Outcomes measures (avoidable hospitalizations) Uncontrolled diabetes Short-term complications Long-term complications Lower extremity amputations
Description
Data Sources National/State
Percent of adults with diabetes who had a hemoglobin A1c measurement at least once in the past year Percent of patients with diabetes who had a lipid profile in the past 2 years Percent of adults with diabetes who had a retinal eye exam in the past year Percent of adults with diabetes who had a foot exam in the past year Percent of adults with diabetes who had an influenza immunization in the past year Percent of adults who received all 5 services for diabetes in appropriate time frame
MEPS
Hospital admissions for uncontrolled diabetes per 100,000 population Hospital admissions for short-term complications of diabetes per 100,000 population Hospital admissions for long-term complications of diabetes per 100,000 population Hospital admissions for lower extremity amputations in patients with diabetes per 100,000 population
HCUP SID
MEPS MEPS MEPS MEPS MEPS
HCUP SID HCUP SID HCUP SID
Abbreviations: MEPS, Medical Expenditure Panel Survey, 2000 – 01; HCUP SID, Healthcare Cost and Utilization Project, State Inpatient Database, 2001.
(Mokdad et al., 2000). Diabetes is associated with a range of other illnesses and is a major risk factor for cardiovascular disease. People with diabetes are at increased risk for stroke, ischemic heart disease, peripheral vascular disease, and neuropathy (American Diabetes Association, 2002, 2003b). Blacks have higher rates of serious complications from diabetes, including higher rates of end-stage renal disease and lower extremity amputation (CDC, 1999; Guadagnoli, Ayanian, Gibbons, McNeil, & LoGerfo, 1995; Gornick et al., 1996). Diabetes is a public health and economic concern. The total cost of the disease in the United States for 2002 was estimated at $132 billion, of which $91.8 billion was attributed to direct medical costs and $40 billion to indirect costs owing to disability, work loss, and premature mortality (American Diabetes Association, 2003a). Diabetes is a preventable disease that can be effectively managed to delay or avoid its complications (CDC, 2004; Heisler, Vijan, Anderson, Ubel, Bernstein, & Hofer, 2003; Hill-Briggs, Cooper, Loman, Brancati, & Cooper, 2003). To identify gaps in care and avoid unnecessary expense, monitoring the ongoing quality of health care in patients with diabetes is crucial. Despite evidence currently available on the best practices in diabetes care, there is still wide variation in diagnostic evaluation, use of preventive services, and the quality and extent of disease management (American Diabetes Association, 2003b; Diabetes Prevention Program Research Group, 2002; Dallo & Weller, 2003). The purpose of this study is to investigate whether
gender differences across racial/ethnic groups exist in the quality of care received by people who suffer from type 2 diabetes. The quality of care for diabetes is evaluated according to 10 process and outcomes measures as defined by the National Healthcare Quality Report (NHQR) and the National Healthcare Disparities Report (NHDR) (Agency for Healthcare Research and Quality [AHRQ], 2004a, 2004b, 2005a, 2005b). The paper’s unique contribution is that it goes beyond the scope of the national reports by performing additional data analysis by gender within racial/ethnic groups. Our findings provide the basis for future development of gender- and/or race/ethnicity-specific strategies to help close the gaps in diabetes care.
Methods Data sources Medical Expenditure Panel Survey. The Medical Expenditure Panel Survey (MEPS) collects data through computer-assisted, in-person interviews of a nationally representative sample of the noninstitutionalized civilian population using a stratified multistage probability design. This analysis uses data from the MEPS Household Component as well as the Diabetes Care Survey supplement of the MEPS, which is a paperand-pencil questionnaire administered to household respondents who answered “yes” when asked whether
58
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
Table 2. Preventive measures for diabetes by gender across race/ethnicity: MEPS 2000 – 01 Non-Hispanic white (%)
Total (%) Measure* Percent of adults with diabetes who had a hemoglobin A1c measurement at least once in the past year Percent of patients with diabetes who had a lipid profile in past 2 years Percent of adults with diabetes who had a retinal eye examination in past year Percent of adults with diabetes who had a foot examination in past year Percent of adults with diabetes who had an influenza immunization in past year Percent of adults with diabetes who received all five recommended diabetes services in appropriate time frame
Non-Hispanic black (%)
Hispanic (%)
p-value NHW to NHB
p-value NHW to Hispanic
W
M
pvalue
W
M
pvalue
W
M
pvalue
W
M
pvalue
W
M
W
M
89.70
90.10
.822
91.00
91.50
.814
88.90
85.30
.471
83.90
84.60
.888
0.490
0.168
0.037
0.102
92.90
95.30
.035
92.30
96.30
.011
96.80
91.50
.077
89.70
92.10
.408
0.023
0.081
0.292
0.055
66.60
69.80
.158
67.40
73.10
.049
68.70
63.20
.338
58.00
57.40
.910
0.762
0.039
0.013
0.001
63.70
69.10
.021
64.20
71.30
.026
62.30
67.50
.295
63.30
60.90
.639
0.620
0.398
0.818
0.023
55.10
56.30
.655
57.90
61.40
.271
47.10
42.50
.439
54.60
41.00
.013
0.019
0.000
0.466
0.000
28.90
33.90
.094
30.60
37.40
.065
24.20
26.30
.701
26.70
20.50
.274
0.137
0.026
0.425
0.000
Bolded p value indicates a statistically significant difference between women and men or between racial/ethnic groups (p ⱕ .05). NHW, non-Hispanic whites; NHB, non-Hispanic blacks; W, women; M, men. Source: Agency for Healthcare Research and Quality. Center for Financing. Access and Cost Trends. Medical Expenditure Panel Survey combining 2000 and 2001 data. *Denominator for summary measure excludes missing values.
they were ever told by a doctor or other health professional that they had diabetes (AHRQ, 2004c). Each year, MEPS collects data from a new sample of households. For this article, data from 2000 and 2001 were pooled to create a larger sample size to assess subgroup populations. The total sample size for the 2 years was 2,365 adults with diabetes. Adjustments were made for pooled variance to calculate standard errors. Data for this report were suppressed if cell size
was ⬍100 or the relative standard error was ⬎30% of the estimate (AHRQ, 2004c). Because the MEPS data are self-reported, certain factors such as recall bias and social desirability in answering questions may limit the accuracy of the data. Healthcare cost and utilization project. The Healthcare Cost and Utilization Project (HCUP) is a family of health care databases and products developed
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
59
Table 3. Diabetes measures—logistic regressions: MEPS, 2000 – 01 Odds Ratio (95% Confidence Interval)
Characteristic Female vs male (reference) Race/ethnicity Hispanic Non-Hispanic black Non-Hispanic white (reference) Income Poor Near poor Middle High (reference) Education Less than high school High school graduate Any college (reference)
HbA1c in past year
Lipid profile in past 2 years
Retinal eye exam in past year
Foot exam in past year
Influenza vaccine in past year
All 5 services
1.04 (0.67, 1.60)
0.74 (0.48, 1.14)
0.92 (0.74, 1.15)
0.78 (0.62, 0.98)
0.97 (0.78, 1.21)
0.83 (0.64, 1.08)
0.71 (0.39, 1.27) 0.76 (0.47, 1.24)
0.87 (0.53, 1.42) 1.41 (0.72, 2.77)
0.73 (0.54, 0.99) 0.93 (0.67, 1.28)
0.82 (0.60, 1.13) 0.85 (0.63, 1.15)
0.81 (0.59, 1.11) 0.59 (0.44, 0.80)
0.73 (0.45, 1.17) 0.67 (0.48, 0.93)
1.00
1.00
1.00
1.00
1.00
1.00
0.68 (0.41, 1.14) 0.97 (0.56, 1.70) 0.95 (0.60, 1.53) 1.00
0.39 (0.18, 0.82) 0.55 (0.26, 1.20) 0.62 (0.33, 1.17) 1.00
0.65 (0.47, 0.91) 0.65 (0.49, 0.86) 0.71 (0.54, 0.93) 1.00
1.07 (0.73, 1.56) 1.08 (0.77, 1.51) 0.97 (0.71, 1.34) 1.00
0.78 (0.54, 1.14) 0.91 (0.67, 1.24) 0.79 (0.61, 1.03) 1.00
0.76 (0.44, 1.31) 1.09 (0.73, 1.62) 1.10 (0.78, 1.54) 1.00
0.58 (0.35, 0.96) 0.71 (0.43, 1.19) 1.00
0.68 (0.36, 1.28) 0.82 (0.46, 1.46) 1.00
0.64 (0.48, 0.84) 0.71 (0.55, 0.93) 1.00
0.87 (0.66, 1.15) 0.83 (0.65, 1.07) 1.00
0.67 (0.50, 0.89) 0.72 (0.53, 0.98) 1.00
0.64 (0.45, 0.91) 0.59 (0.42, 0.83) 1.00
Note: Models include age, insurance, race/ethnicity, income, education, and place of residence. Bolded values indicate a statistically significant difference (95% confidence interval does not include 1).
through a public–private partnership sponsored by AHRQ (Healthcare Cost and Utilization Project, 2004). Statewide data organizations, including state agencies, state hospital associations, and private data consortia, provide discharge-level data for all hospitals in their state to AHRQ for the development of information research databases. The HCUP State Inpatient Databases (SID) include records for all discharges for all hospitals in more than 35 states (AHRQ, 2004c). A file designed to provide national estimates on disparities was developed using a sample of community hospitals from the 2001 SID data for 22 states that participate in HCUP and collect data on race and ethnicity. In total, these states accounted for 65% of all hospital discharges in the United States in 2001. Hospitals were excluded from the sampling frame if the coding of patient race was suspect. Weights were developed to provide national estimates of disparities. Data were suppressed if the count in the denominator of the rates was ⬍70 cases to ensure that relative standard errors were ⬍30% (Coffey, Barrett, et al., 2004). NHQR and NHDR measure sets This paper uses 10 diabetes measures based on the NHQR and NHDR (Table 1). There are 5 process measures of services, a composite measure for receipt of all 5 such services, and 4 outcome measures of diabetes-related hospital admissions. The 5 process measures and the composite measure use MEPS and the 4 outcome measures use HCUP data. The process measures are part of the best standard of care for diabetes and are used to monitor disease progress,
help to control the disease, and avoid or delay hospitalizations due to short- and long-term complications. Statistical analysis Comparisons were made on data stratified by gender across racial/ethnic groups. Two-tailed t-tests were used to assess significance. Although p-values were considered statistically significant at an alpha level of .05, additional criteria were imposed on the comparisons to identify “important differences,” because the large sample sizes yielded highly statistically significant results even for small differences. Relative differences ⬎10% were defined as statistically important. Statistically important differences are indicated in the tables. Because racial and ethnic minorities are disproportionately likely to be of lower socioeconomic status, health care disparities among racial and ethnic minorities are often highly correlated with disparities that fall along socioeconomic lines. To begin to disaggregate racial, ethnic, and socioeconomic effects, multivariate models were developed for MEPS data. These logistic regression models included age, gender, race/ ethnicity, household income, education, insurance, and location of residence. HCUP uses the median household income of a patient’s community as a proxy for socioeconomic status, created by linking the patient’s zip code on the SID with income information obtained from Claritas, a statistical package that provides intercensal estimates and projections. Data include age- and gender-adjusted rates per 100,000 population stratified by area income using Zip code-level counts by age, gender,
60
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
Table 4. Avoidable diabetes-related hospitalizations by gender across race/ethnicity: HCUP, 2001 Population by gender and race/ethnicity
Measure Hospital admissions for uncontrolled diabetes per 100,000 population Hospital admissions for short-term complications of diabetes per 100,000 population Hospital admissions for long-term complications of diabetes per 100,000 population Hospital admissions for lower extremity amputations in patients with diabetes per 100,000 population
Total (per 100,000 population)
NonHispanic White (per 100,000 population)
Non-Hispanic Black (per 100,000 population)
Hispanic (per 100,000 population)
W
M
W
M
W
M
W
M
W
M
W
M
25
27
16
18
79
91
45
42
†
†
†
†
53
57
44
42
133
187
49
54
†
†
106
136
76
109
300
333
181
199
†
†
†
†
28
55
19
45
93
134
45
84
†
†
†
†
NHW to Black
NHW to Hispanic
†
W, women; M, men; NHW, non-Hispanic whites; NHB, non-Hispanic blacks. Rates reported by gender have been age-adjusted. Bolded values indicate the difference is important with a p-value ⱕ .05 and a relative difference ⬎10%, comparing women to men within each racial/ethnic group. †The difference is important with a p-value ⱕ .05 and a relative difference ⬎10% comparing whites to blacks or whites to hispanics by gender. Source: Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project, State Inpatient Databases, disparities analysis file, 2001.
race, and ethnicity from Claritas (Coffey, Barrett, et al., 2004). All analyses accounted for the complex survey design of the 2 databases. MEPS analyses used SUDAAN, and HCUP analyses used the SAS PROC SURVEYMEANS function.
Results Results are presented by gender for the following racial/ethnic groups: non-Hispanic whites, non-Hispanic blacks, and Hispanics. Gender analysis across other racial/ethnic groups (e.g., Asians, Pacific Islanders, and Native Americans/Alaska Natives) was not possible because data were found to be of nonreliable statistical significance (sample size inadequate). These populations were, therefore, excluded from the study.
Preventive measures Table 2 displays information on differences in preventive measures for diabetes care for the total population of women and men across racial/ethnic groups. Differences by gender. The percentage of adults receiving secondary preventive services for diabetes differed by gender for lipid profiles and foot exams. Among non-Hispanic whites, women were less likely to have a lipid profile or receive retinal eye and foot exams. Although such gender differences were not observed among non-Hispanic blacks or Hispanics, Hispanic men were less likely than Hispanic women to receive an influenza immunization. Differences by race/ethnicity. Overall, the receipt of the composite measure of all 5 recommended diabetes services was comparable among all subgroups of
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
61
Table 5. Avoidable diabetes-related hospitalizations by community income levels across race/ethnicity: HCUP, 2001 Total (per 100,000 population)
Measure Hospital admissions for uncontrolled diabetes per 100,000 population Hospital admissions for short-term complications of diabetes per 100,000 population Hospital admissions for long-term complications of diabetes per 100,000 population Hospital admissions for lower extremity amputations in patients with diabetes per 100,000 population
Non-Hispanic White (per 100,000 population)
Average Average $25,000– $35,000– decrease $25,000– $35,000– decrease ⬍$25,000 $34,999 $44,999 ⱖ$45,000 (%) ⬍$25,000 $34,999 $44,999 ⱖ$45,000 (%) 70
43
24
15
40
35
29
17
11
32
115
76
57
40
30
66
59
46
34
20
221
152
118
95
24
123
109
93
80
13
68
51
41
32
22
37
36
33
27
10
Bolded values indicate that the difference is important with a p value ⱕ .05 and a relative difference ⬎10% comparing each income level to the highest income level (ⱖ$45,000), within each racial/ethnic group. —Data do not meet the criteria for statistical reliability, data quality or confidentiality or are otherwise unavailable. Source: Agency for Healthcare Research and Quality, Healthcare Cost and Utilization Project, State Inpatient Databases, disparities analysis file, 2001.
women. However, compared to non-Hispanic white women, non-Hispanic black women received fewer lipid profiles and influenza immunizations, and Hispanic women received fewer HbA1c measurements and retinal eye exams. Compared to non-Hispanic white men, non-Hispanic black and Hispanic men were less likely to have eye exams, be immunized against influenza, and receive all 5 of the recommended diabetes services. Hispanic men also received fewer foot exams than non-Hispanic white men.
observed, but non-Hispanic white men were 43% more likely than non-Hispanic white women to be hospitalized for long-term complications and nonHispanic black men were nearly 41% more likely than non-Hispanic black women to be hospitalized for short-term complications. Moreover, men and women differed significantly overall and across all 3 racial/ ethnic groups in the rate of hospitalization for lower extremity amputation, with men being much more likely to receive the procedure.
Logistic regression analysis. Table 3 shows results of the logistic regression models. These showed that socioeconomic status is an important determinant of diabetes care. People with low incomes were less likely to receive retinal eye exams and lipid profiles, and those with less education were less likely to receive HbA1c measurements, retinal eye exams, influenza immunizations, and all 5 services for diabetes. After controlling for age, insurance, income, education, and place of residence, some racial, ethnic, and gender disparities persisted. Non-Hispanic blacks were less likely than non-Hispanic whites to be immunized against influenza and receive all 5 services for diabetes, whereas Hispanics received fewer retinal eye exams. Regarding gender differences, women were less likely than men to receive foot exams.
Differences by racial/ethnic groups. Non-Hispanic black and Hispanic women and men had higher rates than their white counterparts for all but 1 measure of diabetes complications and hospitalizations. The exception was hospitalizations for short-term complications in non-Hispanic white women compared to Hispanic women. Table 5 shows community income levels across racial/ethnic groups. Rates of hospitalization decreased as income level increased for each racial/ ethnic group. For the total population, hospital admissions for uncontrolled diabetes decreased by approximately 40% at each increasing level of community income. In addition, for all racial/ethnic groups combined, hospital admissions across the 4 avoidable hospitalization measures were statistically different for income groups of ⬍$45,000 per year compared to those of ⱖ$45,000 per year. Relative risk ratios calculated using data from Table 5 show that, across all income categories, the relative risk of hospital admissions for non-Hispanic blacks over non-Hispanic whites was approximately 3.5 for long-term complications and 3.7 for short-term complications and lower extremity amputations. Compared to non-Hispanic
Avoidable hospitalizations Table 4 displays information on avoidable diabetesrelated hospitalizations per 100,000 population for women and men across racial/ethnic groups. Differences by gender. Overall, no gender differences in hospital admissions for uncontrolled diabetes were
62
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
Table 5. Continued Non-Hispanic Black (per 100,000 population)
Hispanic (per 100,000 population) Average decrease (%)
⬍$25,000
$35,000– $44,999
131.00
110.00
68
56
24
—
58
36
28
20
244.00
181.00
148
116
22
71
65
48
41
16
410.00
366.00
292
252
15
259
235
168
162
14
131.00
134.00
105
94
10
82
77
56
54
12
ⱖ$45,000
whites, the risk of hospitalization for uncontrolled diabetes was 2.5 times greater for Hispanics and 5 times greater for non-Hispanic blacks.
Discussion Diabetes is a complex chronic disease requiring comprehensive quality care. Studies have shown that when appropriate care is provided, lower diabetesrelated stress and fewer emergency room and doctor’s office visits are reported (CDC, 2004; Heisler et al., 2003; Hill-Briggs et al., 2003). Although research on the management of diabetes has not concentrated on men or women specifically (Shojania, McDonald, Wachter & Owens, 2004), the disparities in quality of care between genders and across racial and ethnic groups provide us with an opportunity to specifically target populations to improve diabetes management. Providing preventive services is crucial to identify and manage people whose diabetes is uncontrolled with test results outside the normal limits. In addition, test results provide health care professionals and patients who are not doing well with the opportunity to improve their relationship and work together toward controlling the disease. Keeping diabetes under control is essential for delaying or reducing the progression of microvascular complications. Despite the high rates of HbA1c and lipid profile measurements for both women and men, there is still room for improvement. Similarly, the percentages of retinal eye and foot exams and influenza immunizations should be increased for both women and men across all racial/ethnic groups. One major concern is the low percentages of women and men with diabetes who reported receiving all 5 recommended services for the disease in the appropriate time frame (28.9% for women and 33.9% for men). MEPS data show a significant increase in retinal eye
⬍$25,000
$25,000– $34,999
$35,000– $44,999
Average decrease (%)
$25,000– $34,999
ⱖ$45,000
exams for high school graduates with diabetes from 2000 (61%) to 2001 (70%), but gaps in retinal eye exams continue to exist for low-income people with diabetes. The disparities observed in relation to influenza immunization also have been reported in previous studies and have been associated with a variety of factors. For example, in a Medicare population of patients with diabetes, black patients with less than a high school education had fewer influenza vaccinations and HbA1c measurements (Chin, Zhang, & Merrell, 1998) than those with none education; in another study, receipt of an annual influenza immunization was independently associated with race but related to better glycemic control (de Rekeneire et al., 2003). Overall, rates of hospital admissions for uncontrolled diabetes have significantly decreased since 1994 from 40.7 per 100,000 population to the current rates of 25 for women and 27 for men per 100,000 population. The highest hospitalization rates are linked to long-term complications of diabetes, particularly the high rates for non-Hispanic blacks and Hispanics. Another challenge involves reducing the number of hospitalizations for lower extremity amputations. Amputation rates for men were notably high for non-Hispanic blacks (134 per 100,000 population) and Hispanics (84 per 100,000 population); among women, they were 5 times higher for non-Hispanic blacks and 2 times higher for Hispanics than for whites. If the quality of health services for diabetes improves, it may still take several years to detect improvements in the rates of long-term diabetes complications including amputations. Generally, people with lower incomes and less education have worse health status and experience worse health care than those with higher incomes and more education (Pamuk, Makuc, Heack, Reuben, & Lockner, 1998). We found that avoidable hospitalizations for diabetes complications decreased as income
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
increased across racial/ethnic groups. Low socioeconomic status seems to be clearly related to the gender and racial/ethnic differences in hospitalization rates. However, factors such as quality of primary care, age, relationship with providers, and patients’ self-management skills also play a role. In fact, within the complexity of factors affecting the management of diabetes, numerous other patient-associated factors (e.g., job benefits, child care, transportation) may lead to missing scheduled medical appointments and consequent variability in the number of services patients receive. These factors reduce the continuity and effectiveness of health care delivery, cause lapses in the appropriate monitoring of health status, increase the chances of developing poor disease outcomes, and elevate the cost of health services (Karter et al., 2004). Problems in reliable evaluation of the effects of income, education, and insurance on gender and racial/ethnic groups constitute a major weakness in this study and were a major challenge during the development of the NHQR/NHDR (Arispe, Holmes, & Moy, 2005). First, because numerous factors can affect the outcomes for the measures used here, solid interpretation of the relationships between such outcomes and the quality of diabetes care has proven to be quite difficult (Moy, Arispe, Holmes, & Andrews, 2005). Second, data on specific racial, ethnic, and socioeconomic groups frequently are not collected or are insufficient for generating consistent and credible estimates (Moy et al., 2005). Enhanced data collection, therefore, is crucial to better data analysis and interpretation and supports the development of strategies targeting the elimination of disparities in diabetes care. Learning about the gaps in the delivery of diabetes secondary preventive services is very important but has limited value if the next step (i.e., management of individuals who do not have their diabetes under control) is not properly addressed. The diabetes outcome measures discussed in this paper are likely to be very distant outcomes from simple physician’s office testing such as foot exams, lipid profiles, and HbA1c measurement. These tests are often completed in the same office visit. Differences in rates of completion of these health services may affect both clinician and patient education. Because of the link between care and outcomes and the existence of high-quality evidence on the effectiveness of prevention and treatment of diabetes, the disease is a prototypical model for self-management of chronic diseases. Therefore, high-quality care for diabetes should encompass an enhanced patient–provider relationship to help improve patients’ self-management skills. Effective patient–provider communication, particularly when associated with agreement on management goals and strategies, has been shown to contribute to higher patient self-efficacy and self-management skills
63
(Heisler et al., 2003). Several other studies also have shown that patient education improves self-monitoring of glycemic levels and control (Cowie & Harris, 1997; Harris, Eastman, Cowie, Flegal, & Eberhardt, 1999). More recently, coverage of diabetes blood glucose monitors by health maintenance organizations resulted in more patients self-monitoring their blood glucose, increased regularity in the use of anti-diabetes medications, and ultimately reduced blood glucose levels (Soumerai et al., 2004). In addition, the disparities observed here require action toward the development of well-designed interventions that reallocate resources for diabetes selfcare while ensuring that gender differences are addressed across racial/ethnic groups. Because much of this care involves the management of risk factors, self-management education should be tailored to the lifestyles and beliefs specific to gender and racial/ ethnic groups. One effort undertaken by AHRQ in disseminating information about diabetes that can help with the development of targeted well-designed quality improvement strategies to eliminate disparities is a resource guide and workbook for state-level action to improve diabetes care. The guide and workbook, companions to the NHQR and NHDR, use data compiled in the reports and walk users through data procedures, examining the finer points of data collection and analysis and also presenting broader case studies of state-level efforts. The model emphasizes data collection as the first step toward quality improvement and makes suggestions for implementation (Coffey, Matthews, & McDermott, 2004; Kass, 2004). It provides state leaders with the opportunity to monitor their progress in the quality of diabetes care and to exchange experiences about their successes and failures.
Acknowledgments The authors would like to acknowledge the statewide data organizations that participate in the 2001 HCUP Nationwide Inpatient Sample (NIS): Arizona Department of Health Services; California Office of Statewide Health Planning & Development; Colorado Health & Hospital Association; Connecticut–Chime, Inc.; Florida Agency for Health Care Administration; Georgia Hospital Association; Hawaii Health Information Corporation; Illinois Health Care Cost Containment Council; Iowa Hospital Association; Kansas Hospital Association; Kentucky Department for Public Health; Maine Health Data Organization; Maryland Health Services Cost Review Commission; Massachusetts Division of Health Care Finance and Policy; Michigan Health and Hospital Association; Minnesota Hospital Association; Missouri Hospital Industry Data Institute; Nebraska Hospital Association; New Jersey Department of Health & Senior Services; New York State Department of Health; North Carolina Department of Health and Human Services; Ore-
64
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
gon Association of Hospitals & Health Systems; Pennsylvania Health Care Cost Containment Council; Rhode Island Department of Health; South Carolina State Budget & Control Board; Tennessee Hospital Association; Texas Health Care Information Council; Utah Department of Health; Vermont Association of Hospitals and Health Systems; Virginia Health Information; Washington State Department of Health; West Virginia Health Care Authority; Wisconsin Department of Health & Family Services.
References Agency for Healthcare Research and Quality. (2005a). National Healthcare Disparities Report. (in press) Agency for Healthcare Research and Quality. (2005b). National Healthcare Quality Report. (in press) Agency for Healthcare Research and Quality. (2004a). 2004 National Healthcare Disparities Report. Rockville, MD: U.S. Department of Health and Human Services. Agency for Healthcare Research and Quality. (2004b). 2004 National Healthcare Quality Report. Rockville, MD: U.S. Department of Health and Human Services. Agency for Healthcare Research and Quality. (2004c). 2004 National Healthcare Quality Report, Measure Specifications Appendix. Rockville, MD: U.S. Department of Health and Human Services. American Diabetes Association. (2002). Diabetes fact sheet. Available: http://www.diabetes.org. Accessed July 22, 2004. American Diabetes Association. (2003a). Economic costs of diabetes in the US in 2002. Diabetes Care, 26, 917–932. American Diabetes Association. (2003b). Standards of medical care for patients with diabetes mellitus. Diabetes Care, 17, 1514 –1522. Arispe, I. E., Holmes, J. S., & Moy, E. (2005). Measurement challenges in developing the National Healthcare Quality Report and the National Healthcare Disparities Report. Medical Care, 43, I17–I23. Centers for Disease Control and Prevention (CDC). (1999). Diabetes Surveillance, 1999. Atlanta: National Center for Chronic Disease Prevention and Health Promotion, Centers for Disease Control and Prevention. Available: http://www.cdc.gov. Accessed July 23, 2004. Centers for Disease Control and Prevention (CDC). (2005). National diabetes fact sheet, 2005. Atlanta: U.S. Department of Health and Human Services. Centers for Disease Control and Prevention. Available: http://www.cdc.gov. Accessed July 23, 2004. Centers for Disease Control and Prevention (CDC). (2004). At a glance, diabetes: Disabling, deadly, and on the rise. Atlanta: National Center for Chronic Disease Prevention and Health Promotion, CDC. Available: http://www.cdc.gov. Accessed August 2, 2004. Chin, M. H., Zhang, J. X., & Merrell, K. (1998). Diabetes in the African-American population. Morbidity, quality of care, and resource utilization. Diabetes Care, 21, 1090 –1095. Coffey R, Barrett M, Houchens B, Moy E, Andrews R, Kelley E., et al. (2004). Methods Applying AHRQ Quality Indicators to Healthcare Cost and Utilization Project (HCUP) Data for the National Healthcare Disparities Report. Draft. Coffey, R. M., Matthews, T. L., & McDermott, K. (2004). Diabetes care quality improvement: A resource guide for State action (Prepared by The Medstat Group, Inc. and The Council of State Governments under Contract No. 290-00-0004; AHRQ Pub. No. 04-0072). Rockville, MD: Agency for Healthcare Research and Quality, Department of Health and Human Services. Cowie, C. C., & Harris, M. (1997). Ambulatory medical care for non-Hispanic whites, African Americans, and Mexican-Americans with NIDDM in the U.S. Diabetes Care, 20, 142–147.
Dallo, F. J., & Weller, S. C. (2003). Effectiveness of diabetes mellitus screening recommendations. Proceedings of the National Academies of Science, 100, 10574 –10579. de Rekeneire, N., Roos, R. N., Simonsick, E. M., Shorr, R. I., Kuller, L. H., Scwartz, A. V., et al. (2003). Racial differences in glycemic control in a well-functioning older diabetic population. Diabetes Care, 26, 1986 –1992. Diabetes Prevention Program Research Group. (2002). The Diabetes Prevention Program (DPP): Description of lifestyle intervention. Diabetes Care, 25, 2165–2171. Gornick, M. E., Eggers, P. W., Reilly, T. W., Mentnech, R. M., Fitterman, L. K., Kucken, L. E., et al. (1996). Effects of race and income on mortality and use of services among Medicare beneficiaries. New England Journal of Medicine, 335, 791–799. Guadagnoli, E., Ayanian, J. Z., Gibbons, G., McNeil, B. J., & LoGerfo, F. W. (1995). The influence of race on the use of surgical procedures for treatment of peripheral vascular disease of the lower extremities. Archives of Surgery, 130, 381–386. Harris, M. I., Eastman, R. C., Cowie, C. C., Flegal, K. M., & Eberhardt, M. S. (1999). Racial and ethnic differences in glycemic control of adults with type two diabetes. Diabetes Care, 22, 403– 408. Healthcare Cost and Utilization Project (HCUP). (2004). HCUP Databases. Rockville, MD: Agency for Healthcare Research and Quality. Available: www.hcup-us.ahrq.gov. Accessed July 23, 2004. Heisler, M., Vijan, S., Anderson, R. M., Ubel, P. A., Bernstein, S. J., & Hofer, T. P. (2003). When do patients and their physicians agree on diabetes treatment goals and strategies, and what difference does it make? Journal of General Internal Medicine, 18, 893–902. Hill-Briggs, F., Cooper, D. C., Loman, K., Brancati, F. L., & Cooper, L. A. (2003) A qualitative study of problem solving and diabetes control in type 2 diabetes self-management. Diabetes Education, 29, 1018 –1028. Karter, A. J., Parker, M. M., Moffet, H. H., Ahmed, A. T., Ferrara, A., Liu, J.Y., et al. (2004). Missed appointments and poor glycemic control: An opportunity to identify high-risk diabetic patients. Medical Care, 42, 110 –115. Kass, B. (2004, September). Diabetes care quality improvement: A workbook for state action (AHRQ Pub No. 04-0073). Rockville, MD: Agency for Healthcare Research and Quality, Department of Health and Human Services. Mokdad, A. H., Ford, E. S., Bowman, B. A., Nelson, D. E., Engelgau, M. M., Vinicor F., et al. (2000). Diabetes trends in the U.S.: 1990 –1998. Diabetes Care, 23, 1278 –1283. Moy, E., Arispe, I., Holmes, J., & Andrews, R. (2005). Preparing the National Healthcare Disparities Report: Gaps in data for assessing racial, ethnic, and socioeconomic disparities in health care. Medical Care, 43, I9 –I16. Pamuk, E., Makuc, D., Heack, K., Reuben, C., & Lockner, K. (1998). Socioeconomic status and health chartbook: Health, United States, 1998. Hyattsville, MD: National Center for Health Statistics. Shojania, K. G., McDonald, K. M., Wachter, R. M., & Owens, D. K. (2004) Closing the quality gap: A critical analysis of diabetes care strategies (AHRQ Publication No. 04-0051-2). Rockville, MD: Agency for Healthcare Research and Quality. Soumerai, S. B., Mah, C., Zhang, F., Adams, A., Barton, M., Fajtova V., et al. (2004). Effects of health maintenance organization coverage of self-monitoring devices on diabetes self care and glycemic control. Archives of Internal Medicine, 164, 645– 652.
Author Description Rosaly Correa-de-Araujo, MD, MSc, PhD, is a cardiovascular pathologist trained at the National Heart, Lung, and Blood Institute. As the Agency for Healthcare Research and Quality’s Director of Women’s
R. Correa-de-Araujo et al. / Women’s Health Issues 16 (2006) 56 – 65
Health and Gender-Based Research, Dr. Correa oversees the development of a national research agenda for women in consultation with prominent members of the research community and other government agencies. Her main areas of interest include gender-based research and analysis particularly related to chronic diseases, medication use outcomes and safety, and disparities in health care.
65
Kelly McDermott, MA, is currently a predoctoral student in health services research at the University of Washington in Seattle. Ernest Moy, MD, MPH, is a Senior Service Fellow with the Center for Quality Improvement and Patient Safety in the Agency for Healthcare Research and Quality. Dr. Moy leads the development of the National Healthcare Disparities Report.