Abstract
Background
Diabetes poses a serious health burden, of which veterans have a disproportionate share. Few data exist regarding differences in self-care behaviors and provider-based quality of care indicators among a large sample of veterans. The objective of this study was to determine the effect of Veterans Affairs (VA) use on diabetes quality of care indicators among veterans.
Methods
A cross-sectional analysis was done on data from 36,525 veterans in the 2003 Behavioral Risk Factor Surveillance Survey. VA use was defined as receiving some or all health care from a VA facility in the previous 12 months. Diabetes quality indicators such as two or more provider visits, two or more hemoglobin A1c tests, and flu and pneumonia shots were compared between VA users and non-VA users. The independent effect of VA use on each quality indicator was analyzed with multiple regression using STATA version 10 (Stata Press, College Station, TX) to account for the complex survey design and yield population estimates.
Results
Among veterans with diabetes, 26.8% were VA users. The only significant difference between VA users and non-VA users was that VA users were significantly more likely to check their feet one or more times daily (75.7% vs. 68.5%, P = 0.015). In final adjusted models, VA users were at least twice as likely as non-VA users to have foot exams by a provider (odds ratio 2.59) and receive flu and pneumonia shots (odds ratio 2.30 and 2.05, respectively). VA users were also more likely to have two or more provider visits, dilated eye exams, and two or more hemoglobin A1c tests than non-VA users.
Conclusions
Key quality indicators for diabetes care were better among veterans getting some or all of their care from VA facilities, suggesting more effective care strategies. However, interventions should identify and perpetuate excellent self-care behaviors to more substantially impact adverse diabetes-related outcomes.
Introduction
Diabetes hastens death and disability through serious cardiovascular disease outcomes, such as myocardial infarction, stroke, peripheral arterial disease, blindness, and kidney damage.1 Approximately 23.6 million people in the United States or 7.8% of the population have diabetes,1 and a disproportionate burden is seen among the elderly, racial/ethnic minorities, and low-income populations.1,2 Estimates from 2007 indicate the total cost of diabetes is approximately $174 billion, including $116 billion in direct costs and $58 billion in indirect costs.1 After adjusting for age and sex, having diabetes is associated with a 2.3-fold increase in health expenditures compared to not having diabetes.1
Diabetes is equally highly prevalent among veterans. The annual incidence of diabetes in veterans is ∼2% per year,3 with prevalence rates increasing from 16.7% in 1998 to 19.6% in 2000. In the Veterans Affairs (VA) system, diabetes is the third most common diagnosis, affecting 16% of patients.4 The average age of VA patients with diabetes is 65.3 years.5 Published literature has demonstrated that nearly 4% of total VA expenditures are attributable to veterans with diabetes.6
Previous studies have shown that quality of care for many chronic conditions is higher in the VA system than in non-VA systems.7–9 However, few studies have used nationally representative samples to examine whether diabetes self-care and provider-based quality of care differ between those who use VA facilities for medical care and those who use non-VA facilities. We used the Behavioral Risk Factor Surveillance Survey (BRFSS) to examine whether self-care behaviors and provider-based quality of care differed significantly between veterans who received care from VA facilities compared to those that did not. We hypothesized that veterans with diabetes receiving care from VA facilities would have collectively better diabetes self-care and provider-based quality of care compared to those who received care from non-VA facilities.
Subjects and Methods
We examined self-reported data from respondents of the 2003 BRFSS, a state-based, random-digit dialing telephone survey designed to measure behavioral risk factors of the non-institutionalized, civilian population of the United States aged 18 years and older. Details about the BRFSS have been published previously.10 The BRFSS, initiated in 1984, is an ongoing data collection program with the objective of collecting uniform, state-specific data on preventive health practices and risk behaviors that are linked to chronic diseases, injuries, and preventable infectious diseases in the adult population. The BRFSS uses a complex sampling design involving stratification, clustering, and multistage sampling to yield nationally representative estimates. Weights were applied so that estimates reflect the non-institutionalized U.S. population.
Veterans' status and use of VA facilities
Subjects were asked, “Have you ever served on active duty in the United States Armed Forces, either in the regular military or in a National Guard or military reserve unit?” Subjects were also asked, “Which of the following best describes your service in the United States military?” Veterans were defined as those who served in a regular military or in a National Guard or military reserve unit, but were not currently on active duty, in National Guard or reserve units. Subjects who refused to answer the question or responded “don't know” or “not sure” were excluded. Users of VA facilities were defined as veterans who reported that in the last 12 months they had received some or all of their health care from VA facilities.
Sample characteristics
Demographic variables included age, race/ethnicity, education, income, marital status, employment, and health status. Age was divided into four categories: 18–34, 35–49, 50–64, and 65+ years. Race/ethnicity was defined as non-Hispanic whites, non-Hispanic blacks, Hispanics, and others based on self-report. Four education levels were created: less than high school graduate, high school graduate or general equivalency diploma, less than college graduate, and college graduate. Four income categories were used: <$25,000, <$50,000, <$75,000, and $75,000+. Marital status was dichotomized as married or not married, and employment status as employed or unemployed. Health status was dichotomized as excellent/very good/good versus fair/poor.
Diagnosis of diabetes and diabetes education
The diagnosis of diabetes was based on self-report of whether a doctor had ever told them they had diabetes other than during pregnancy. Diabetes education was based on “yes” response to the question, “Have you ever taken a course or class in how to manage your diabetes yourself?”
Diabetes self-care behaviors
Four diabetes self-care behaviors were assessed based on self-report: physical activity (PA), testing blood glucose at home, checking feet at home, and fruit and vegetable intake.
PA was computed based on questions about type, duration, and intensity of PA. Two categories of physical activity were created: meeting PA recommendations (defined as ≥30 min/day for ≥5 days per week of moderate activity or ≥20 min/day on ≥3 days of vigorous activity) and not meeting PA recommendations.
Home blood glucose testing was assessed by the question, “About how often do you check your blood for glucose or sugar? Include times when checked by a family member or friend, but do not include times when checked by a health professional.” A dichotomous variable for frequency of testing was created: one or more times versus less than once per day.
Home foot examination was assessed by asking respondents, “About how often do you check your feet for any sores or irritations? Include times when checked by a family member or friend, but do not include times when checked by a health professional.” A dichotomous variable for frequency of foot examination was created: one or more times versus less than once per day.
Fruit and vegetable intake was determined using a summary index item within the BRFSS database that calculated whether or not respondents consumed five or more servings of fruits/vegetables per day based on established algorithms. The variable was categorized as consumption of five or more servings of fruits and vegetables per day versus less than five servings per day.
Quality of diabetes care
Seven quality of diabetes care indicators were defined based on current American Diabetes Association guidelines,11 including biannual provider office visits, hemoglobin A1c testing, foot exam, dilated eye exam, aspirin use, and influenza and pneumonia vaccination.
Office visit to a health provider was determined by asking respondents, “About how many times in the past 12 months have you seen a doctor, nurse or other health professional for your diabetes?” A dichotomous variable was created to distinguish those who had two or more visits versus less than two visits in the last year (12 months).
Hemoglobin A1c testing by a health provider was assessed by asking respondents, “A test for hemoglobin A1C measures average level of blood sugar over the past 3 months. About how many times in the past 12 months has a doctor, nurse, or other health professional checked your A1C?” Frequency of hemoglobin A1c testing by health providers was categorized as less than two versus two or more times per year.
Foot exam by a health provider was assessed by asking respondents, “About how many times in the past 12 months has a health professional checked your feet for any sores or irritations?” Frequency of foot exam by health providers was categorized as less than one versus one or more times per year.
Dilated eye exam by a health provider was assessed by asking respondents, “When was the last time you had an eye exam in which the pupils were dilated? This would have made you temporarily sensitive to bright light.” Frequency of dilated eye exam by health providers was categorized as never, less than 1 year ago, 1–2 years ago, and 2+ years ago. A dichotomous variable categorized as less than one versus one or more times per year.
Aspirin use was assessed by asking respondents, “Do you take aspirin daily or every other day?”
Influenza vaccination was assessed by asking respondents whether they got the flu shot in the past 12 months. An individual was deemed to have received the flu shot if he or she responded “yes.”
Pneumonia vaccination was assessed by asking respondents whether they had ever received the pneumonia shot. An individual was deemed to have received the pneumonia shot if he or she responded “yes.”
Statistical analyses
All analyses took into account the complex survey design and weighted sampling probabilities of the data source and were performed using STATA version 10 (Stata Press, College Station, TX).12 All statistical tests were two-tailed, and significance was set at an α of 0.05. Three sets of statistical analyses were performed. In the first set of analyses, demographic and clinical characteristics of participants were compared by VA user status using χ2 test. In the second set of analyses, diabetes self-care and quality of care indicators were compared by VA user status using the χ2 test. In the third set of analyses, 11 separate multiple logistic regression models were run for the four self-care behaviors (PA, fruit/vegetable intake, home glucose testing, and home foot care) and seven quality of care indicators (number of office visits, hemoglobin A1c testing by provider, foot exam by provider, dilated eye exam by provider, aspirin use, receipt of flu shot, and receipt of pneumonia shot) to assess the independent association between use of VA facilities and these outcome variables. In each model, self-care behaviors and quality of care indicators were entered as dependent variables, VA user status as the primary independent variable, and demographics, access to care, health status, and attendance at diabetes education classes as covariates. All variables included in the models were conceptually related to the outcomes of interest and differed significantly by VA user status.
Results
Veterans comprised 14.2% (n = 36,525) of the BRFSS sample in 2003. Approximately 12.4% reported being diagnosed with diabetes (n = 4,501). Among veterans with diabetes, 26.8% (n =1,342) received some or all of their care from a VA facility. Table 1 shows the characteristics of the sample by VA user status. Overall, the majority of the sample was 50 years of age or older, male, mostly non-Hispanic white, and married. Similarly, the vast majority was insured and saw a healthcare provider in the last 12 months. Notable differences between VA users and non-users was that VA users were more likely to be non-Hispanic black, less educated, and unemployed, to have lower income, and to report poorer health and were less likely to have health insurance.
Table 1.
Demographic Characteristics of Veterans with Diabetes by VA User Status
| All (n = 4,501) | Non-VA users (n = 3,159) | VA users (n = 1,342) | P value | |
|---|---|---|---|---|
| Age (years) | 0.618 | |||
| 18–34 | 0.87 | 1.04 | 0.40 | |
| 35–49 | 7.67 | 7.59 | 7.89 | |
| 50–64 | 36.0 | 36.3 | 35.0 | |
| ≥65 | 55.6 | 55.1 | 56.7 | |
| Gender/sex | 0.558 | |||
| Female | 2.77 | 2.83 | 2.46 | |
| Male | 97.2 | 97.1 | 97.5 | |
| Race/ethnicity | 0.028* | |||
| Non-Hispanic white | 78.8 | 80.8 | 73.1 | |
| Non-Hispanic black | 9.69 | 8.20 | 13.8 | |
| Hispanic | 6.28 | 6.42 | 5.91 | |
| Other | 5.27 | 4.54 | 7.26 | |
| Marital status | <0.001* | |||
| Married | 74.8 | 76.6 | 69.8 | |
| Not married | 25.2 | 23.4 | 30.2 | |
| Educational level | 0.002* | |||
| Less than HS graduate | 11.4 | 10.3 | 14.5 | |
| HS graduate | 31.7 | 31.7 | 32.0 | |
| Some college | 29.0 | 27.1 | 34.2 | |
| College graduate | 27.9 | 31.0 | 19.4 | |
| Employment status | <0.001* | |||
| Employed | 29.3 | 32.5 | 20.6 | |
| Unemployed | 70.7 | 67.5 | 79.4 | |
| Annual income level | <0.001* | |||
| <$25,000 | 36.7 | 30.7 | 52.5 | |
| $25,000–$49,999 | 34.2 | 34.4 | 33.4 | |
| $50,000–$74,999 | 13.0 | 15.0 | 7.74 | |
| $75,000+ | 16.1 | 19.8 | 6.34 | |
| Self-rated health status | 0.001* | |||
| Excellent/very good/good | 56.7 | 60.1 | 47.1 | |
| Fair/poor | 43.4 | 39.9 | 52.9 | |
| Has health insurance | 0.001* | |||
| Yes | 94.2 | 95.4 | 90.9 | |
| No | 5.81 | 4.61 | 9.10 | |
| Saw healthcare provider in the last 12 months | 0.150 | |||
| Yes | 93.5 | 94.2 | 91.8 | |
| No | 6.48 | 5.83 | 8.25 | |
| Received diabetes education | 0.034* | |||
| Yes | 55.3 | 53.4 | 60.2 | |
| No | 44.8 | 47.0 | 39.9 |
All numbers represent percentages. HS, high school.
Statistically significant, P < 0.05.
Self-care behaviors among this sample of veterans with diabetes are shown in Table 2. The only significant difference between VA users and non-VA users was that VA users were significantly more likely to check their feet one or more times daily (75.7% vs. 68.5%, respectively; P = 0.015). There were no significant differences between VA users versus non-VA users with regards to PA, eating fruits and vegetables, and doing self-foot exams.
Table 2.
Self-Care Behaviors by VA Use Among Veterans with Diabetes
| All | Non-VA users | VA users | P value | |
|---|---|---|---|---|
| PA level | 0.853 | |||
| Meets PA recommendations | 36.8 | 37.0 | 36.4 | |
| Insufficient or no PA | 63.2 | 63.0 | 63.6 | |
| Fruit and vegetable intake | 0.136 | |||
| 5+ servings per day | 22.3 | 23.3 | 19.7 | |
| <5 servings per day | 77.7 | 76.7 | 80.3 | |
| Blood sugar testing | 0.924 | |||
| 1+ times daily | 56.0 | 55.9 | 56.2 | |
| <1 time daily | 44.0 | 44.1 | 43.8 | |
| How often you check your feet | 0.015* | |||
| 1+ times daily | 70.5 | 68.5 | 75.7 | |
| <1 time daily | 29.5 | 31.5 | 24.3 |
All numbers represent percentages.
Statistically significant, P < 0.05.
Table 3 shows provider-based quality of care (i.e., proportions of veterans reporting receipt of diabetes quality of care measures generally performed by healthcare providers) by VA user status. Veterans using VA facilities were significantly more likely than non-VA users to have at least two office visits within 12 months (84% vs. 78%, respectively; P = 0.011). Veterans receiving care at a VA facility were significantly more likely to have had their hemoglobin A1c levels checked two or more times in the last 12 months (75% for VA users vs. 68% for non-VA users, P < 0.001). VA users were more likely to have a clinical foot exam in the last 12 months (85%) compared to 72% of non-VA users (P = 0.005). More VA users (80%) received a dilated eye exam in the last 12 months than non-VA users (72%) (P = 0.005). Similarly, VA users were more likely to have received an influenza vaccination in the last 12 months (77% vs. 65%, respectively; P < 0.001) and ever received pneumonia vaccination (72% vs. 56%, respectively; P < 0.001). There was no statistically significant difference in aspirin use between the two groups (P = 0.945).
Table 3.
Quality of Care by VA Use Among Veterans with Diabetes
| All | Non-VA users | VA users | P value | |
|---|---|---|---|---|
| Office visits in the last 12 months | 0.011* | |||
| 2+ visits | 79.9 | 78.1 | 84.5 | |
| <2 visits | 20.1 | 21.9 | 15.5 | |
| Dilated eye exam in the last 12 months | 0.005* | |||
| ≥1 exams | 74.5 | 72.2 | 80.5 | |
| <1 exam | 25.5 | 27.8 | 19.5 | |
| Hemoglobin A1c testing in the last 12 months | 0.023* | |||
| ≥2 tests | 69.9 | 68.0 | 75.0 | |
| <2 tests | 30.1 | 32.0 | 25.0 | |
| Foot exam in the last 12 months | <0.001* | |||
| ≥1 exams | 75.7 | 72.1 | 85.2 | |
| <1 exam | 24.3 | 28.0 | 14.8 | |
| Daily aspirin use | 0.945 | |||
| Yes | 64.6 | 64.7 | 64.5 | |
| No | 35.4 | 35.3 | 35.5 | |
| Received flu shot in the last 12 months | <0.001* | |||
| Yes | 68.6 | 65.5 | 77.2 | |
| No | 31.4 | 34.5 | 22.8 | |
| Ever received pneumonia vaccine | <0.001* | |||
| Yes | 60.4 | 56.1 | 72.0 | |
| No | 39.6 | 43.9 | 28.0 |
All numbers represent percentages.
Statistically significant, P < 0.05.
In final adjusted models (Table 4), with non-VA users as the reference group and controlling for age, sex, race/ethnicity, education, income, insurance, and diabetes education, the same provider-based quality of care measures remained significant. VA users were more than twice as likely to have received foot exams by a provider (odds ratio [OR] 2.59, 95% confidence interval [CI] 1.76–3.83), ever had a pneumonia shot (OR 2.30, 95% CI 1.68–3.14), and had a flu shot (OR 2.05, 95% CI 1.44–2.92). In addition, VA users had a 60–70% greater likelihood of having a dilated eye exam (OR 1.68, 95% CI 1.14–2.49), two or more hemoglobin A1c tests (OR 1.65, 95% CI 1.19–2.28), and two or more provider visits (OR 1.61, 95% CI 1.08–2.39) in the previous 12 months.
Table 4.
Logistic Regression of Self-Care Behaviors and Quality of Care Indicators Among VA Users with Diabetes
| OR | 95% CI | |
|---|---|---|
| Self-care behaviors | ||
| Meets physical activity recommendations | 1.15 | 0.87, 1.52 |
| ≥5 daily servings of fruits, vegetables | 0.92 | 0.66, 1.29 |
| At least once daily blood sugar testing | 1.06 | 0.79, 1.40 |
| At least once daily checking of feet | 1.35 | 0.98, 1.85 |
| Quality of care indicators | ||
| At least office visits in the last 12 months | 1.61* | 1.09, 2.39 |
| At least one exam of feet in the last 12 months | 2.59* | 1.76, 3.82 |
| ≥2 hemoglobin A1c testing in the last 12 months | 1.65* | 1.19, 2.28 |
| Dilated eye exam in the last 12 months | 1.69* | 1.14, 2.49 |
| Daily aspirin use | 0.99 | 0.69, 1.42 |
| Flu shot received in the last 12 months | 2.05* | 1.44, 2.92 |
| Ever received pneumonia vaccine | 2.30* | 1.68, 3.14 |
Adjusted for age, gender, race, education, income, insurance status, and diabetes education. The reference group comprises veterans with diabetes who are non-VA users.
Statistically significant when range of the CI excludes 1.00.
Discussion
Among this national sample of veterans, important differences in diabetes quality of care were demonstrated between those who use VA facilities for their medical care and those who do not. Veterans with diabetes who were VA users had significantly better quality of care, particularly for those measures targeted by provider performance initiatives within the VA healthcare system. In adjusted models, VA users were significantly more likely to have clinical foot exams and to receive preventive care (flu shot, pneumonia vaccination, office visit, and eye exam) than non-VA users. Additionally, VA users had a significantly higher proportion receiving diabetes education than non-VA users (60% vs. 53%, respectively; P = 0.03). However, diabetes self-care behaviors were not significantly different between the two groups. These findings suggest that the VA healthcare system provides better diabetes quality of care; however, there remains a need to improve diabetes self-management strategies.
Studies that have examined diabetes processes of care demonstrate a higher quality of care with respect to provider-specific measures.4,9 Two studies have reported a higher level of diabetes processes of care ranging from 72% to 95% of patients within the VA healthcare system compared to those using non-federal care systems.9,13 Our findings are consistent with these studies such that most diabetes processes of care remained statistically significantly greater among VA users compared to non-VA users: provider foot checks (OR 2.59), pneumonia vaccination (OR 2.30), flu shot (OR 2.05), dilated eye exam (OR 1.69), hemoglobin A1c testing (OR 1.65), and office visits (OR 1.61). Therefore, research findings continue to consistently support substantial improvement in processes of care within the VA health system.7–9,13
For diabetes self-care behaviors, VA users were significantly more likely to perform self-foot checks than non-VA users (76% vs. 68%, respectively; P = 0.01) and were not inferior in other self-care measures. However, in adjusted models for quality of care, none of the diabetes self-care behaviors was different by VA user status. A recent analysis of diabetes self-management revealed that only 6% of adults with diabetes performed all four self-care behaviors (PA, fruit and vegetable intake, home glucose testing, and home foot exams) at recommended levels.14 Similar to our results, these authors also found adults with diabetes were much more likely to perform one self-care behavior (∼90%) with more attention paid to home foot exams (69%) and home glucose testing (55%), whereas very low rates of recommended diet and activity patterns were achieved (26% and 31%, respectively).
While most elements of provider care are highly related to diabetes monitoring and the logistics of treatment adjustment, clinical care factors have a small impact on diabetes outcomes.15 Daily behaviors and activities (diet, PA, and medication adherence) are key in maintaining good glycemic control and therefore are critical for diabetes management. Patient-level factors have been shown to account for at least 95% of patient care and health management.15,16 The findings of this and other investigations support a dire need for successful lifestyle interventions and patient education. From a research standpoint, patient-level interventions should be the primary method by which behavior change is directed and better diabetes self-management is achieved. From the clinical standpoint, providers should place greater emphasis on the importance of self-examination and self-care among patients with diabetes, thus reducing complications. In addition, another strategy could focus on educating providers about how to empower patients to better manage their diabetes.17–19 As reported by Reiber et al.,13 diabetes self-management education is a necessary component of care within the VA health system, and, as such, the VA has a number of diabetes education strategies in place. However, the low proportions of veterans in this study receiving these educational strategies (ranging from 7% to 48%) suggest the need for VA administration to rethink the quality and effectiveness of these services.
This study demonstrated substantial differences in diabetes quality of care in a comparable sample of VA users and non-VA users. In each of the aforementioned studies the demographics of veteran populations were similar: older, mostly male, lower socioeconomic status, and higher diabetes burden. This study comprised typical veterans who use the VA healthcare system for their medical care7,20 but differs in that we analyzed data from one large single source, allowing for greater comparability within the veteran population. Therefore, these results have a higher likelihood of more accurately representing diabetes processes of care and self-care behaviors among veterans. However, secondary analyses have inherent limitations that deserve mention. First, BRFSS data are self-reported, thus lending the results to recall bias. A growing number of studies show a high correlation between self-reported data and objective measures. Second, very few female veterans use VA facilities for medical care,20 so results cannot be generalized to them. Third, the majority of veterans in this sample were elderly, so the study has limited generalizability to younger veterans. Finally, we analyzed secondary data so the inherent limitations of existing data apply. We cannot account for potential confounding factors such as diabetes severity, glycemic control, and veterans who utilize medical services at both VA and non-VA facilities.
In summary, veterans with diabetes who receive at least some care within the VA health system have a high quality of diabetes care. The lack of differences in self-care behaviors between VA users and non-VA users suggests that greater efforts need to target patients' self-management behaviors. Diabetes educators should implement more creative strategies to actively engage patients with diabetes in self-care potentially through a nutrition education intervention that demonstrates practical changes (cooking skills) or use of technological resources (electronic food diary). Health professionals should also integrate different self-care components to provide more comprehensive education (PA along with nutrition education). Patients are likely the principal drivers of diabetes outcomes through their daily lifestyle habits. Therefore, interventions should identify and perpetuate excellent self-care behaviors to more substantially impact adverse diabetes-related outcomes.
Author Disclosure Statement
No competing financial interests exist.
References
- 1.National Institute of Diabetes and Digestive and Kidney Diseases. National Diabetes Statistics, 2007 Fact Sheet. Bethesda, MD: U.S. Department of Health and Human Services, National Institutes of Health; 2008. [Google Scholar]
- 2.Cowie CC. Rust KF. Byrd-Holt DD. Eberhardt MS. Flegal KM. Engelgau MM. Saydah SH. Williams DE. Geiss LS. Gregg EW. Prevalence of diabetes and impaired fasting glucose in adults in the U.S. population: National Health And Nutrition Examination Survey 1999–2002. Diabetes Care. 2006;29:1263–1268. doi: 10.2337/dc06-0062. [DOI] [PubMed] [Google Scholar]
- 3.Miller YD. Dunstan DW. The effectiveness of physical activity interventions for the treatment of overweight and obesity and type 2 diabetes. J Sci Med Sport. 2004;7(1 Suppl):52–59. doi: 10.1016/s1440-2440(04)80278-0. [DOI] [PubMed] [Google Scholar]
- 4.Reiber GE. Au D. McDonell M. Fihn SD. Diabetes quality improvement in Department of Veterans Affairs Ambulatory Care Clinics: a group-randomized clinical trial. Diabetes Care. 2004;27(Suppl 2):B61–B68. doi: 10.2337/diacare.27.suppl_2.b61. [DOI] [PubMed] [Google Scholar]
- 5.Young BA. Pugh JA. Maynard C. Reiber G. Diabetes and renal disease in veterans. Diabetes Care. 2004;27(Suppl 2):B45–B49. doi: 10.2337/diacare.27.suppl_2.b45. [DOI] [PubMed] [Google Scholar]
- 6.Maciejewski ML. Maynard C. Diabetes-related utilization and costs for inpatient and outpatient services in the Veterans Administration. Diabetes Care. 2004;27(Suppl 2):B69–B73. doi: 10.2337/diacare.27.suppl_2.b69. [DOI] [PubMed] [Google Scholar]
- 7.Asch SM. McGlynn EA. Hogan MM. Hayward RA. Shekelle P. Rubenstein L. Keesey J. Adams J. Kerr EA. Comparison of quality of care for patients in the Veterans Health Administration and patients in a national sample. Ann Intern Med. 2004;141:938–945. doi: 10.7326/0003-4819-141-12-200412210-00010. [DOI] [PubMed] [Google Scholar]
- 8.Jha AK. Perlin JB. Kizer KW. Dudley RA. Effect of the transformation of the Veterans Affairs Health Care System on the quality of care. N Engl J Med. 2003;348:2218–2227. doi: 10.1056/NEJMsa021899. [DOI] [PubMed] [Google Scholar]
- 9.Kerr EA. Gerzoff RB. Krein SL. Selby JV. Piette JD. Curb JD. Herman WH. Marrero DG. Narayan KM. Safford MM. Thompson T. Mangione CM. Diabetes care quality in the Veterans Affairs Health Care System and commercial managed care: the TRIAD study. Ann Intern Med. 2004;141:272–281. doi: 10.7326/0003-4819-141-4-200408170-00007. [DOI] [PubMed] [Google Scholar]
- 10.Centers for Disease Control and Prevention: Behavioral Risk Factor Surveillance System Survey Data. Atlanta: Centers for Disease Control and Prevention, U.S. Department of Health and Human Services; 2003. [Google Scholar]
- 11.American Diabetes Association. Executive summary: standards of medical care in diabetes—2009. Diabetes Care. 2009;32(Suppl 1):S6–S12. doi: 10.2337/dc09-S006. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Stata Version 10.0 [computer program] College Station, TX: Stata Press; 2007. [Google Scholar]
- 13.Reiber GE. Koepsell TD. Maynard C. Haas LB. Boyko EJ. Diabetes in nonveterans, veterans, and veterans receiving Department of Veterans Affairs health care. Diabetes Care. 2004;27(Suppl 2):B3–B9. doi: 10.2337/diacare.27.suppl_2.b3. [DOI] [PubMed] [Google Scholar]
- 14.Nwasuruba C. Khan M. Egede LE. Racial/ethnic differences in multiple self-care behaviors in adults with diabetes. J Gen Intern Med. 2007;22:115–120. doi: 10.1007/s11606-007-0120-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Tuerk PW. Mueller M. Egede LE. Estimating physician effects on glycemic control in the treatment of diabetes: methods, effects sizes, and implications for treatment policy. Diabetes Care. 2008;31:869–873. doi: 10.2337/dc07-1662. [DOI] [PubMed] [Google Scholar]
- 16.McLaughlin S. Meeting the challenges: diabetes care in special settings. Diabetes Spectr. 2005;18:143–145. [Google Scholar]
- 17.Egede LE. Lifestyle modification to improve blood pressure control in individuals with diabetes: is physician advice effective? Diabetes Care. 2003;26:602–607. doi: 10.2337/diacare.26.3.602. [DOI] [PubMed] [Google Scholar]
- 18.Funnell MM. Kruger DF. Spencer M. Self-management support for insulin therapy in type 2 diabetes. Diabetes Educ. 2004;30:274–280. doi: 10.1177/014572170403000220. [DOI] [PubMed] [Google Scholar]
- 19.Funnell MM. Nwankwo R. Gillard ML. Anderson RM. Tang TS. Implementing an empowerment-based diabetes self-management education program. Diabetes Educ. 2005;31 doi: 10.1177/0145721704273166. 53, 55–56, 61. [DOI] [PubMed] [Google Scholar]
- 20.Frayne SM. Yu W. Yano EM. Ananth L. Iqbal S. Thrailkill A. Phibbs CS. Gender and use of care: planning for tomorrow's Veterans Health Administration. J Womens Health (Larchmt) 2007;16:1188–1199. doi: 10.1089/jwh.2006.0205. [DOI] [PubMed] [Google Scholar]