Sociodemographic Factors Associated With Engagement in Diabetes Self-management Education Among People With Diabetes in the United States

Abstract

Objective:

Research outside the United States shows that certain subgroups of patients (eg, those who are older, male, of low socioeconomic status, and uninsured) are less likely than others to report receiving diabetes self-management education (DSME); however, less is known about DSME uptake in the United States. We examined sociodemographic, patient, and behavioral characteristics associated with DSME in a nationally representative sample.

Methods:

We analyzed data from the 2011-2013 Behavioral Risk Factor Surveillance System for 84 179 adults who self-identified receiving a diagnosis of diabetes. We constructed weighted, multivariate logistic regression models to examine the associations between DSME and sociodemographic characteristics (age, sex, race/ethnicity, marital status, education, and annual household income), patient characteristics (body mass index, having a regular provider, health insurance status, health status, and insulin use), and self-management behaviors (home foot examination, home blood glucose testing, and physical activity).

Results:

More than half (n = 45 557, 53.7% [weighted]) of respondents reported engaging in DSME. Compared with non-Hispanic white adults, non-Hispanic black adults were more likely to engage in DSME (adjusted odds ratio [aOR] = 1.17; 95% confidence interval [CI], 1.07-1.29). Respondents were less likely to engage in DSME if they were male (aOR = 0.85; 95% CI, 0.80-0.91) or Hispanic (aOR = 0.81; 95% CI, 0.71-0.92), were a high school graduate (but no college; aOR = 0.71; 95% CI, 0.66-0.78) or less than a high school graduate (aOR = 0.51; 95% CI, 0.45-0.59), had an annual household income of $15 000-$24 999 (aOR = 0.81; 95% CI, 0.73-0.89) or <$15 000 (aOR = 0.70; 95% CI, 0.62-0.78), or had no health insurance (aOR = 0.87; 95% CI, 0.76-0.98). DSME was significantly associated with all 3 self-management behaviors.

Conclusions:

Increasing public health interventions aimed at educating people with diabetes about self-management could improve outcomes.

Keywords

type 2 diabetes diabetes education socioeconomic status demographics diabetes self-management

Diabetes affects 29.1 million people of all ages in the United States, about 9.3% of the total population.¹ Diabetes prevalence is rising, and an estimated 44 million people are anticipated to have diabetes by 2034.² Many comorbid conditions and complications are associated with diabetes, including, but not limited to, blindness and other eye problems, nervous system disorders including dementia, kidney disease, lower-extremity amputations, periodontal disease, heart disease, and stroke.^3,4 Comorbidities and complications can have a negative effect on diabetes self-care, health status, and quality of life.⁵

Between 50% and 80% of people with diabetes have limited knowledge about the disease and limited skills in diabetes management. Diabetes self-management education (DSME) is the process of imparting the knowledge and teaching the skills necessary to give people the ability to manage their diabetes.⁶ Effective diabetes self-management is essential to achieving optimal glycemic control and decreasing the morbidity and mortality associated with diabetes. The American Diabetes Association (ADA) suggests managing diabetes by maintaining a healthy diet, scheduling regular physician visits, and staying physically active.⁷ Blood glucose monitoring is the main tool to control diabetes.⁷ DSME has been shown to improve preventive care practices and clinical outcomes,⁸ reduce diabetes and associated costs,³ mitigate future complications, and improve overall health.⁸ DSME empowers people to manage their diabetes through education about nutrition, medication, insulin therapy, stress management, and preventive foot and eye care.⁸

According to the National Standards for Diabetes Self-management Education, DSME is an intricate component necessary for improving patient outcomes.⁹ Studies show that certain groups of patients are less likely than others to report receiving diabetes education.^10

-14 These groups include patients aged ≥50, males, patients with lower socioeconomic status, and patients without health insurance^10

-14; however, most of these studies were conducted outside of the United States. One study in the United States used nationally representative data but was published in 1994, more than 20 years ago.¹³ The objective of this study was to use recent data to describe (1) sociodemographic and patient characteristics associated with DSME in a nationally representative sample and (2) the association between self-monitoring behaviors and DSME engagement.

Methods

We examined data from the 2011-2013 Behavioral Risk Factor Surveillance System (BRFSS).¹⁵ The BRFSS is an annual, state-based, random-digit-dial telephone survey of the civilian, noninstitutionalized US population aged ≥18 that collects data on health risk behaviors, preventive health practices, and health care access. Detailed information on the design and sampling methods used in the BRFSS is reported elsewhere.^15,16 We first included survey respondents who answered yes to the question “[Ever told] you have diabetes?” (n = 198 071). We excluded women who answered yes to having gestational diabetes and respondents who reported they had been told they have prediabetes or borderline diabetes (n = 13 502). After further excluding 100 390 participants who did not answer the question on diabetes education, our final sample size was 84 179.

Diabetes Education

We assessed DSME with the question, “Have you ever taken a course or class in how to manage your diabetes yourself?” Respondents were deemed to have received diabetes education if they responded yes to this question.

Sociodemographic and Patient Characteristics

We assessed the following sociodemographic and patient characteristics: sex (male or female), age (18-54, 55-64, and ≥65); race/ethnicity (non-Hispanic white, non-Hispanic black, Hispanic, and other [eg, Asian, Native Hawaiian or other Pacific Islander, American Indian/Alaska Native, and multiracial]); marital status (married/living with a partner, widowed/divorced/separated, and never married); body mass index measured in weight in kilograms divided by height in square meters (kg/m²; underweight, <18.0 kg/m²; normal weight, 18.0-24.9 kg/m²; overweight, 25.0-29.9 kg/m²; and obese, ≥30.0 kg/m²), annual household income (<$15 000, $15 000-$24 999, $25 000-$34 999, $35 000-$49 999, and ≥$50 000); education level (<high school graduate, high school graduate, some college, and ≥college graduate); health insurance (“Do you have any kind of health care coverage, including health insurance, prepaid plans such as [health maintenance organizations] or government plans such as Medicare?” [yes/no]); a regular health care provider (“Do you have one person you think of as your personal physician or health care provider?” [yes/no]); general health condition (excellent/very good, good, and fair/poor); and insulin use (“Are you now taking insulin?” [yes/no]).

Diabetes Self-management Behaviors

We assessed 3 diabetes self-management behaviors based on self-report: home blood glucose testing, home foot examination, and physical activity. We assessed home blood glucose testing behavior with 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.” The American Association of Diabetes Educators (AADE) and the ADA recommend daily self-monitoring of blood glucose for optimal self-care.⁷ We created a dichotomous variable for testing frequency (≥1 time daily and <1 time daily).

We assessed home foot care examination with the question, “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.” The AADE and ADA recommend daily home foot examinations for patients with diabetes.⁷ We created a dichotomous variable for foot examination frequency (≥1 time daily and <1 time daily).

We assessed physical activity with the question, “During the past month, other than your regular job, did you participate in any physical activities or exercises such as running, calisthenics, golf, gardening, or walking for exercise?” We considered a respondent to be physically active if he or she answered yes to this question.

Statistical Analysis

We performed analyses by using SAS version 9.4 procedures,¹⁷ which incorporate survey sampling weights to account for the complex sampling design used in the BRFSS and to provide representative estimates of the US population. We used a full-sample weight to calculate population estimates, and we used the jackknife variance estimation method to compute standard errors, which ensured valid inferences from the sample to the US population, correcting for nonresponse and noncoverage. We used weighted proportions to describe the characteristics of the study participants. We used weighted, bivariable, and multivariable logistic regression models using listwise deletion to examine sociodemographic characteristics (age, sex, race/ethnicity, marital status, education, and income level) and patient characteristics (body mass index, having a regular health care provider, health insurance status, health status, and insulin use) associated with DSME. We ran 3 separate weighted multiple logistic regression models to assess associations between the 3 self-management behaviors and DSME, while controlling for all sociodemographic and patient characteristics. We reported odds ratios (ORs) and 95% confidence intervals (CIs) for each variable. We used a 2-tailed test of significance, with P < .05 considered significant.

Results

A total of 84 179 respondents reported being diagnosed with diabetes, of whom 45 557 (53.7%) had engaged in DSME. Most respondents were non-Hispanic white (n = 58 508, 65.6%), married or living with a partner (n = 39 518, 54.5%), and obese (n = 42 047, 54.0%); had health insurance (n = 77 333, 89.2%) and a regular health care provider (n = 79 325, 92.9%); and did not use insulin (n = 58 091, 68.6%). Sixty-three percent of respondents reported a home foot examination (n = 51 596) and checking blood glucose ≥1 time daily (n = 52 879; Table).

Table.

Characteristics of respondents with diabetes and weighted, adjusted logistic regression model estimating engagement in diabetes self-management education, Behavioral Risk Factor Surveillance System, 2011-2013 (n = 84 179)^a

Characteristic	No. (Weighted %)^b	OR (95% CI)	aOR^c (95% CI)	P Value^d
Engaged in diabetes self-management education
Yes	45 557 (53.7)
No	38 622 (46.3)
Self-management behaviors
Home foot examination
≥1 time daily	51 596 (63.0)
<1 time daily	29 558 (37.0)
Home blood glucose testing
≥1 daily	52 879 (62.9)
<1 daily	29 930 (37.1)
Physical activity
Yes	48 179 (59.2)
No	33 039 (40.8)
Age group, y
18-54	17 138 (31.7)	1.00 (Reference)	1.00 (Reference)
55-64	23 665 (27.3)	1.07 (0.99-1.15)	1.03 (0.94-1.13)	.50
≥65	43 376 (41.0)	0.80 (0.75-0.85)	0.82 (0.76-0.89)	<.001
Sex
Female	34 899 (49.3)	1.00 (Reference)	1.00 (Reference)
Male	49 280 (50.7)	0.90 (0.86-0.95)	0.85 (0.80-0.91)	<.001
Race/ethnicity
Non-Hispanic white	58 508 (65.6)	1.00 (Reference)	1.00 (Reference)
Non-Hispanic black	12 207 (16.8)	1.09 (1.01-1.17)	1.17 (1.07-1.29)	.008
Hispanic	7139 (12.1)	0.65 (0.59-0.72)	0.81 (0.71-0.92)	.002
Non-Hispanic other^e	6325 (5.5)	0.87 (0.77-0.97)	0.92 (0.80 -1.05)	.21
Marital status
Married/living with a partner	39 518 (54.5)	1.00 (Reference)	1.00 (Reference)
Widowed/divorced/separated	34 895 (32.2)	0.83 (0.79-0.88)	0.94 (0.87-1.00)	.06
Never married	9389 (13.3)	0.93 (0.85-1.02)	0.98 (0.88-1.10)	.72
Education
College graduate	18 859 (16.0)	1.00 (Reference)	1.00 (Reference)
Some college	22 340 (27.8)	0.94 (0.88-1.01)	0.99 (0.91-1.08)	.79
High school graduate	29 711 (33.8)	0.64 (0.60-0.69)	0.71 (0.66-0.78)	<.001
<High school graduate	12 941 (22.4)	0.41 (0.38-0.45)	0.51 (0.45-0.59)	<.001
Annual household income, $
≥50 000	18 736 (28.2)	1.00 (Reference)	1.00 (Reference)
35 000-49 999	9972 (13.8)	0.83 (0.76-0.90)	0.90 (0.81-0.99)	.04
25 000-34 999	9565 (13.2)	0.72 (0.66-0.79)	0.86 (0.77-0.95)	.006
15 000-24 999	18 213 (25.5)	0.64 (0.60-0.70)	0.81 (0.73-0.89)	<.001
<15 000	14 862 (19.3)	0.53 (0.49-0.58)	0.70 (0.62-0.78)	<.001
Body mass index, kg/m²
Normal weight or underweight (<25.0)	12 280 (14.7)	1.00 (Reference)	1.00 (Reference)
Overweight (25.0-29.9)	25 535 (31.3)	1.05 (0.97-1.14)	1.03 (0.93-1.14)	.62
Obese (≥30.0)	42 047 (54.0)	1.18 (1.00-1.28)	1.06 (0.96-1.16)	.27
Regular health care provider
Yes	79 325 (92.9)	1.00 (Reference)	1.00 (Reference)
No	4597 (7.1)	0.73 (0.65-0.83)	0.90 (0.77-1.06)	.21
Health insurance
Yes	77 333 (89.2)	1.00 (Reference)	1.00 (Reference)
No	6610 (10.8)	0.76 (0.69-0.83)	0.87 (0.76-0.98)	.02
General health
Excellent/very good	14 852 (17.6)	1.00 (Reference)	1.00 (Reference)
Good	29 582 (35.3)	0.96 (0.89-1.03)	0.97 (0.89-1.06)	.50
Fair/poor	39 303 (47.1)	0.86 (0.81-0.93)	0.92 (0.85-1.01)	.08
Insulin use
No	58 091 (68.6)	1.00 (Reference)	1.00 (Reference)
Yes	25 987 (31.4)	2.00 (1.89-2.11)	2.10 (1.95-2.25)	<.001

Abbreviations: aOR, adjusted odds ratio; OR, crude odds ratio.

^a Data source: Centers for Disease Control and Prevention.¹⁵

^b Numbers may not add up to totals because of missing data. Weighted percentages corrected for nonresponse and noncoverage bias.

^c Model adjusted for age, sex, marital status, education, income, body mass index, regular provider, health insurance status, general health, and insulin use.

^d A 2-tailed test of significance was used, with P < .05 considered significant.

^e Includes Asian, Native Hawaiian or other Pacific Islander, American Indian/Alaska Native, and multiracial.

Sociodemographic and Patient Characteristics Associated With DSME

Age, sex, race/ethnicity, education, income, insurance status, and insulin use were significantly associated with DSME. Compared with non-Hispanic white respondents, non-Hispanic black respondents were more likely to engage in DSME (adjusted OR [aOR] = 1.17; 95% CI, 1.07-1.29), and Hispanic respondents were less likely to engage in DSME (aOR = 0.81; 95% CI, 0.71-0.92). Men were less likely than women to engage in DSME (aOR = 0.85; 95% CI, 0.80-0.91), and respondents aged ≥65 were more likely than respondents aged 18-54 to engage in DSME (aOR = 0.82; 95% CI, 0.76-0.89). Respondents without health insurance were less likely than respondents with health insurance to engage in DSME (aOR = 0.87; 95% CI, 0.76-0.98), and respondents who used insulin were significantly more likely than respondents who did not use insulin to engage in DSME (aOR = 2.10; 95% CI, 1.95-2.25; Table).

Education predicted engagement in DSME. The likelihood of engaging in DSME declined steadily across education levels and was significantly worse for respondents who did not graduate from high school (aOR = 0.51; 95% CI, 0.45-0.59) than for college graduates. Respondents whose annual household income was $15 000-$24 999 (aOR = 0.81; 95% CI, 0.73-0.89) or <$15 000 (aOR = 0.70; 95% CI, 0.62-0.78) were significantly less likely to engage in DSME than respondents whose annual household income was ≥$50 000 (Table).

Diabetes Self-management Behaviors and DSME

DSME was significantly associated with all 3 self-management behaviors. The crude odds were similar to the adjusted odds. In the adjusted logistic regression analyses, respondents who reported engaging in DSME were significantly more likely than respondents who did not engage in DSME to be physically active (aOR = 1.46; 95% CI, 1.37-1.56), to conduct home foot examinations ≥1 time daily (aOR = 1.37; 95% CI, 1.28-1.45), and to conduct home blood glucose testing ≥1 time daily (aOR = 1.59; 95% CI, 1.48-1.70).

Discussion

Comprehensive DSME is necessary to successfully implement and sustain lifestyle changes and to ensure that patients are using appropriate health services. A joint position statement of the ADA, the AADE, and the Academy of Nutrition and Dietetics also supports and recommends the use of DSME to help improve outcomes.¹⁸ Therefore, understanding factors associated with participation in DSME may help to identify points of intervention for improving participation rates. We found that patients with diabetes were more likely to engage in DSME if they were younger, female, or African American; had ≥college education; had a higher annual household income; had health insurance; or used insulin. Our results are consistent with characteristics identified by the few studies examining determinants of DSME.^11,13,14,19

In a randomized trial, attending a DSME program at diagnosis led to improved understanding of the illness, and these changes in illness beliefs were correlated with metabolic changes.²⁰ However, our study found that only 54% of respondents reported engaging in DSME. Although this percentage is higher than the national estimate of 35% published in a study in 1994,¹³ the rate of engagement in DSME is low. Several factors might be contributing to this low rate of engagement, including patient unwillingness to participate, programs’ location and hours of operation, languages of service used, insurance coverage,^21,22 lack of referral to DSME by some health professionals,²³ and/or lack of discussion about the benefits of attending DSME programs.²⁴ Whatever the reasons, the low rate of engagement in DSME is worrisome given the success of DSME in improving outcomes for patients with diabetes.²⁰ DSME should use a patient-centered approach and should be an ongoing effort from the time of diabetes diagnosis.¹⁸

Community education programs could be used to increase engagement in DSME, especially because evidence suggests that group DSME programs can improve self-efficacy and diabetes knowledge.^19,25 In our study, respondents who were older (vs younger), Hispanic (vs non-Hispanic white), or poor (vs higher income); had lower levels of education (vs higher levels of education); and were uninsured (vs insured) were less likely to engage in DSME. Culturally sensitive education and materials written at an appropriate literacy level for education and language should be used to accommodate people whose first language may not be English and/or who have a low level of education.^26,27 Moreover, DSME programs should be expanded in poor neighborhoods or offered during the evening or on weekends to reach low-income patients who could not otherwise access these programs. For example, telemedicine could be used to deliver DSME in underserved rural communities with small populations and limited resources.²⁸ Other technologies could also be used to improve access and provide education for low-income and/or uninsured patients, such as the internet, text messaging, or automated telephone calls.^29
-31

In addition, increasing awareness among health care professionals of DSME program services and/or facilitating and coordinating referral processes may increase DSME engagement. Health care providers should recommend DSME when they visit patients in the hospital or emergency department, because these visits may be the only opportunities to get patients to engage in DSME, especially patients who do not have regular health care providers or health insurance. These interventions may help increase DSME engagement and optimize care among disadvantaged patients, who are at increased risk for diabetes complications. Systems-level and/or policy changes could also be instituted as a means of improving DSME engagement. Programs should be implemented to decrease barriers for those with low incomes or without health care. Expansion of health insurance coverage and health care access (ie, making DSME low cost or free) might help increase DSME engagement.

In the adjusted model, black respondents were more likely to engage in DSME and Hispanic respondents were less likely to engage in DSME than non-Hispanic white respondents. Although these findings are consistent with those of a previous study,¹³ the racial/ethnic distribution of DSME engagement in our study is interesting. It is possible that, because of the high prevalence of diabetes, the African American community is commonly targeted for DSME interventions, whereas fewer interventions are targeted toward non-Hispanic white or Hispanic patients with diabetes. Alternatively, physicians may be more likely to refer non-Hispanic black patients with diabetes to DSME than they are to refer non-Hispanic white patients. Understanding the referral patterns for DSME is likely as important for understanding participation in DSME as is understanding individual sociodemographic characteristics. We also found that respondents who used insulin were more likely than respondents who did not use insulin to engage in DSME, perhaps because diabetes patients requiring insulin—especially patients who use premeal insulin—receive additional education on administering medication, self-monitoring, and eating.

We found that older, less educated, and low-income respondents were less likely than younger, more educated, and higher-income respondents, respectively, to engage in a DSME program, which is consistent with the findings of previous studies.^10,11,32 A study in Canada, where access and barriers to care are reduced because of Canada’s universal health care system, found disparities between age and lower socioeconomic status and DSME engagement, which means less educated and low-income respondents face disparities other than access and insurance.¹⁰ For example, older, less educated, and low-income respondents may have lower levels of health literacy, be less able to navigate the health care system, or face more financial barriers to attending DSME programs than younger, more educated, and higher-income respondents. Therefore, it is important that interventions to increase DSME engagement focus on older, less educated, and low-income respondents.

Limitations and Strengths

This study had several limitations. First, the BRFSS relied on self-reported data, which may have been limited by recall bias or misclassification of diabetes status and participation in DSME. Engagement in DSME was determined by using a yes/no question in the BRFSS without eliciting further details, such as the language in which the course was conducted, the availability of interpreter services, the duration and content of the course, and the time elapsed since the DSME course. Similarly, self-management activities (eg, testing for blood glucose) were determined on the basis of self-report rather than by using validated methods. Second, the BRFSS is a cross-sectional survey and, therefore, it was not possible to establish causal relationships. Third, we were not able to capture data on the nature and source of DSME. Fourth, we could not determine how many times respondents participated in DSME. Fifth, we could not differentiate respondents who had type 1 diabetes from those with type 2 diabetes. Sixth, we did not know how well respondents’ diabetes was controlled; level of control could have affected whether or not they received DSME. Finally, the study was subject to residual confounding; that is, we did not adjust for important variables (eg, health literacy or language proficiency) that could have affected the findings, because BRFSS does not collect these data.

This study also had several strengths. It is one of the few studies to have investigated this topic using current data in the United States. The BRFSS is a long-standing, nationally representative survey; its representativeness increases the generalizability of results to the entire US population. The findings of this study are especially relevant to public health because diabetes is a chronic disease and its prevalence is increasing; effective management is necessary to reduce its burden. Much of the burden, including complications, could be prevented by increasing engagement in DSME. Our study identified special populations for whom interventions could be developed to improve DSME engagement and prevent future complications.

Conclusion

Health care providers should encourage older, male, poorer, less educated respondents, and respondents without health insurance to participate in DSME, given the substantial benefits of DSME in improving outcomes. Future studies should investigate the effectiveness of various diabetes education programs (online, in person, community groups) and of including other factors (eg, health literacy, duration of course, and/or language proficiency) in assessing correlates of DSME engagement.

Footnotes

Acknowledgment

A portion of this work was presented at the at 21st Annual International Meeting of the International Society for Pharmacoeconomics and Outcomes Research, May 2016, Washington, DC.

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Centers for Disease Control and Prevention. National diabetes statistics report, 2014: estimates of diabetes and its burden in the United States. https://www.cdc.gov/diabetes/pdfs/data/2014-report-estimates-of-diabetes-and-its-burden-in-the-united-states.pdf. Accessed July 16, 2018.

Huang

Basu

O’Grady

Capretta

. Projecting the future diabetes population size and related costs for the U.S. Diabetes Care. 2009;32(12):2225–2229.

Piette

Kerr

. The impact of comorbid chronic conditions on diabetes care. Diabetes Care. 2006;29(3):725–731.

National Association of Managed Care Physicians. Co-morbidities. 2008. http://www.namcp.org/diabetes/html/info/comorbidities.html. Accessed August 21, 2016.

Beverly

Wray

Chiu

LaCoe

. Older adults’ perceived challenges with health care providers treating their type 2 diabetes and comorbid conditions. Clin Diabetes. 2014;32(1):12–17.

Funnell

Brown

Childs

. National standards for diabetes self-management education. Diabetes Care. 2009;32(suppl 1):S87–S94.

American Diabetes Association. Standards of medical care in diabetes—2014. Diabetes Care. 2014;37(suppl 1):S14–S80.

Balamurugan

Rivera

Jack

Jr Allen

Morris

. Barriers to diabetes self-management education programs in underserved rural Arkansas: implications for program evaluation. Prev Chronic Dis. 2006;3(1):A15.

Haas

Maryniuk

Beck

. National standards for diabetes self-management education and support. Diabetes Care. 2014;37(suppl 1):S144–S153.

10.

Cauch-Dudek

Victor

Sigmond

Shah

. Disparities in attendance at diabetes self-management education programs after diagnosis in Ontario, Canada: a cohort study. BMC Public Health. 2013;13:85.

11.

Shah

Booth

. Predictors and effectiveness of diabetes self-management education in clinical practice. Patient Educ Couns. 2009;74(1):19–22.

12.

Heisler

Piette

Spencer

Kieffer

Vijan

. The relationship between knowledge of recent HbA1c values and diabetes care understanding and self-management. Diabetes Care. 2005;28(4):816–822.

13.

Coonrod

Betschart

Harris

. Frequency and determinants of diabetes patient education among adults in the U.S. population. Diabetes Care. 1994;17(8):852–858.

14.

Graziani

Rosenthal

Diamond

. Diabetes education program use and patient-perceived barriers to attendance. Fam Med. 1999;31(5):358–363.

15.

Centers for Disease Control Prevention. Behavioral Risk Factor Surveillance System: 2012 annual survey data. http://www.cdc.gov/brfss/annual_data/annual_data.htm. Accessed August 25, 2017.

16.

Mokdad

Stroup

Giles

. Public health surveillance for behavioral risk factors in a changing environment. Recommendations from the behavioral risk factor surveillance team. MMWR Recomm Rep. 2003;52(RR-9):1–12.

17.

SAS Institute, Inc. SAS Version 9.4. Cary, NC: SAS Institute, Inc; 2013.

18.

Powers

Bardsley

Cypress

. Diabetes self-management education and support in type 2 diabetes: a joint position statement of the American Diabetes Association, the American Association of Diabetes Educators, and the Academy of Nutrition and Dietetics. Clin Diabetes. 2016;34(2):70–80.

19.

Rygg

Rise

Gronning

Steinsbekk

. Efficacy of ongoing group based diabetes self-management education for patients with type 2 diabetes mellitus: a randomised controlled trial. Patient Educ Couns. 2012;86(1):98–105.

20.

Davies

Heller

Skinner

. Effectiveness of the Diabetes Education and Self Management for Ongoing and Newly Diagnosed (DESMOND) programme for people with newly diagnosed type 2 diabetes: cluster randomised controlled trial. BMJ. 2008;336(7642):491–495.

21.

Shaw

Killeen

Sullivan

Bowman

. Disparities in diabetes self-management education for uninsured and underinsured adults. Diabetes Educ. 2011;37(6):813–819.

22.

Gucciardi

Chan

VWS

Fortugno

Khan

Horodezny

Swartzack

. Primary care physician referral patterns to diabetes education programs in southern Ontario, Canada. Can J Diabetes. 2011;35(3):262–268.

23.

Winkley

Evwierhoma

Amiel

Lempp

Ismail

Forbes

. Patient explanations for non-attendance at structured diabetes education sessions for newly diagnosed type 2 diabetes: a qualitative study. Diabet Med. 2015;32(1):120–128.

24.

Sunaert

Vandekerckhove

Bastiaens

. Why do GPs hesitate to refer diabetes patients to a self-management education program: a qualitative study. BMC Fam Pract. 2011;12:94.

25.

Steinsbekk

Rygg

Lisulo

Rise

Fretheim

. Group based diabetes self-management education compared to routine treatment for people with type 2 diabetes mellitus: a systematic review with meta-analysis. BMC Health Serv Res. 2012;12:213.

26.

Anderson-Loftin

Barnett

Bunn

Sullivan

Hussey

Tavakoli

. Soul food light: culturally competent diabetes education. Diabetes Educ. 2005;31(4):555–563.

27.

Martin

. Changes and consistencies in diabetes education over 5 years: results of the 2010 National Diabetes Education Practice Survey. Diabetes Educ. 2012;38(1):35–46.

28.

Davis

Hitch

Salaam

Herman

Zimmer-Galler

Mayer-Davis

. TeleHealth improves diabetes self-management in an underserved community: diabetes TeleCare. Diabetes Care. 2010;33(8):1712–1717.

29.

Mulvaney

Rothman

Wallston

Lybarger

Dietrich

. An internet-based program to improve self-management in adolescents with type 1 diabetes. Diabetes Care. 2010;33(3):602–604.

30.

Osborn

Mayberry

Mulvaney

Hess

. Patient web portals to improve diabetes outcomes: a systematic review. Curr Diab Rep. 2010;10(6):422–435.

31.

Mulvaney

Ritterband

Bosslet

. Mobile intervention design in diabetes: review and recommendations. Curr Diab Rep. 2011;11(6):486–493.

32.

Forjuoh

Huber

Bolin

, et al. Provision of counseling on diabetes self-management: are there any age disparities? Patient Educ Couns. 2011;85(2):133–139.