Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2020 Jul 1.
Published in final edited form as: Am J Prev Med. 2019 May 24;57(1):111–116. doi: 10.1016/j.amepre.2019.02.013

Beyond Race Disparities: Accounting for Socioeconomic Status in Diabetes Self-Care

Lyndsay A Nelson 1,2, Michael T Ackerman 3, Robert A Greevy Jr 3, Kenneth A Wallston 4,5, Lindsay S Mayberry 1,2,5
PMCID: PMC6589128  NIHMSID: NIHMS1525376  PMID: 31130463

Abstract

Introduction:

Among patients with type 2 diabetes, racial disparities are prevalent across a variety of outcomes; however, inconsistent disparities in determinants of outcomes warrants exploring the impact of other, related factors. This study sought to examine whether disparities in health literacy, numeracy, self-care behaviors, and HbA1c persisted between non-Hispanic blacks and non-Hispanic whites after applying a robust adjustment for socioeconomic status (SES).

Methods:

From 2016 to 2018, adult patients with type 2 diabetes (N=444) were recruited from primary care clinics. Participants self-reported demographics, indicators of SES (i.e., income, education, health insurance, housing status, and financial strain), subjective health literacy and numeracy, and self-care behaviors. Participants also completed an HbA1c test. In 2018, differences were examined between non-Hispanic blacks and non-Hispanic whites in health literacy, numeracy, self-care, and HbA1c, first unadjusted and then adjusted using propensity score weighting.

Results:

In unadjusted analyses, compared with non-Hispanic whites, non-Hispanic blacks had lower health literacy (p=0.039) and numeracy (p<0.001), less medication adherence (p=0.009), use of information for dietary decisions (p=0.013), and problem eating behaviors (p<0.001; i.e., non-Hispanic blacks reported fewer problems), and higher HbA1c levels (p=0.005). After adjusting for all SES indicators, only the reverse disparity in problem eating behaviors (p=0.016) and the disparity in HbA1c (p=0.011) remained.

Conclusions:

Findings highlight the importance of considering SES when examining disparities in health-related skills and behaviors. Moving beyond education and income to inclusion of more comprehensive markers of SES can improve understanding of how SES may contribute to disparities and the ability to appropriately target factors leading to inequality.

INTRODUCTION

Approximately 10% of U.S. adults have diabetes1 which can lead to heart disease, stroke, and amputations.1 Costs of complications are substantial2 and forecasted to more than double in the next decade.3 Consistent performance of self-care behaviors can improve HbA1c and prevent complications.4,5 Non-Hispanic blacks/African Americans (NHBs) with type 2 diabetes (T2D) have more complications,6,7 more emergency room visits,7 and worse HbA1c levels8–10 as compared with non-Hispanic whites (NHWs). However, evidence is mixed as to whether NHBs have worse self-care. Rather, racial disparities in self-care behaviors tend to vary by sample and behavior.11,12 In addition, health literacy and numeracy are important determinants of diabetes outcomes and are often worse among NHBs.13–15

Race is a construct used to group individuals, but more genetic variation exists within racial categories than between.16,17 Rather than attributing differences between individuals to a racial category, it may be more informative to examine factors for which race may act as a proxy.18 Social and economic conditions vary considerably between races and gaps have persisted for decades19; in the U.S., NHBs have significantly less education and lower income than NHWs, and are twice as likely to live in poverty.19 Research examining the role of SES in racial disparities in diabetes self-care and health-related skills is limited.11 Looking beyond race to understand factors that contribute to diabetes disparities can identify appropriate strategies for reducing disparities. Therefore, this study seeks to examine differences in health literacy, numeracy, self-care behaviors, and HbA1c between NHBs and NHWs after adjustment for multiple SES indicators.

METHODS

Study Sample

This research was conducted as part of an RCT20; for the present analyses, baseline data was used from participants who self-reported NHW or NHB race. Data were collected between 2016 and 2018 from English-speaking adults diagnosed with T2D, prescribed at least one T2D medication, and receiving primary care at Vanderbilt University Medical Center or federally qualified health centers. Patients were excluded if they had an HbA1c <6.8% as the most recent value within 12 months. The Vanderbilt University IRB approved all study procedures.

Measures

Participants self-reported age, gender, race, diabetes duration, number of prescribed diabetes medications, insulin status, and SES (education, annual income, housing status, insurance status, and financial strain using the Tool for Assessing Patients’ Stressors [TAPS]).21 The full 20-item TAPS scale asks about experiences of stressors (yes/no) in the last 12 months; for the current study, the four items relevant to low SES were included (i.e., not enough money for food, rent or mortgage, or clothes; difficulty paying for health-related expenses; stressful jobs/unemployment; and lack of affordable local transportation). Participants also completed validated measures of health literacy,22 numeracy,23 and self-care behaviors. Healthy diet was assessed with two separate subscales that assess “use of information for diet decision making” and “problem eating behavior.”24,25 The study assessed physical activity by calculating total MET minutes/week via the International Physical Activity Questionnaire,26,27 medication adherence with the Adherence to Refills and Medications Scale for Diabetes,28 self-monitoring of blood glucose with the Summary of Diabetes Self-Care Activities blood glucose testing subscale,29 and smoking status using Behavioral Risk Factor Surveillance System items.30 HbA1c was collected via venipuncture or point-of-care by patients’ clinic or using a mail-in HbA1c kit31,32 provided and analyzed by CoreMedica Laboratories (Lee’s Summit, MO).

Statistical Analysis

Data were analyzed in 2018. Differences in covariates and outcomes between NHB and NHW participants were compared using unweighted and weighted samples. The R package Survey, version 3.33–2, was used for the weighted cohort analysis. Propensity score weights were calculated from a logistic model of race by SES indicators (income, education, housing status, insurance status, financial strain) as well as covariates which could confound effects of interest (age, gender, clinic site [Vanderbilt University Medical Center versus federally qualified health center], diabetes duration, number of prescribed diabetes medications, insulin use, and HbA1c test type). To allow nonlinear associations with race, restricted cubic splines were fit for income, age, and diabetes duration. To allow for covariates with missing values, multiple imputation was used with the predictive mean in the R package MICE, version 2.46.0.33 Each participant’s propensity score weight equaled one over the predicted probability of being NHB or NHW based on his/her covariates.

The unweighted differences appear as they exist without covariate adjustment. The weighted differences appear as they would exist in an adjusted cohort where NHBs and NHWs both had covariate distributions resembling the unweighted sample as a whole (i.e., holding SES and covariates constant). Table 1 shows the covariate balance achieved in the weighted cohort between NHBs and NHWs. Table 2 shows differences in outcomes before and after weighting.

Table 1.

Unadjusted and Adjusted Differences in Participant Characteristics by Race

Unweighted sample Weighted samplea
M ± SD or n (%) p-value M ± SD or n (%) p–value
Characteristicsb NHW (n=242) NHB (n=202) NHW (nc=433) NHB (nc=450)
Age, yearsd 56.7 ± 9.0 55.5 ± 10.1 0.306 56.2 ± 9.3 56.4± 9.5 0.881
Sex, male 134 (55) 70 (35) <0.001*** 206 (48) 213 (47) 0.954
Education, years 0.063 0.991
 <HS degree 21 (9) 23 (12) 43 (10) 42 (10)
 HS degree or 68 (28) 72 (37) 137 (32) 140 (32)
 equivalent
 >HS degree 151 (63) 102 (52) 250 (58) 258 (59)
Annual household income, US$d <0.001*** 0.955
 <10,000 33 (14) 48 (27) 77 (19) 75 (19)
 10,000–24,999 59 (26) 55 (31) 112 (28) 120 (30)
 25,000–54,999 54 (24) 55 (31) 112 (28) 101 (25)
 >55,000 82 (36) 22 (12) 105 (26) 104 (26)
Housing status, homeless or unstable 30 (12) 21 (10) 0.534 50 (12) 52 (12) 0.997
 Insurance status 0.021* 0.992
 Private 131 (55) 95 (48) 221 (52) 229 (51)
 Public only 50 (21) 65 (32) 109 (26) 113 (25)
 Uninsured 59 (25) 40 (20) 98 (23) 105 (24)
Financial strain (TAPS-4) 1.1 ± 1.2 1.4 ± 1.3 0.002** 1.2 ± 1.22 1.2 ± 1.2 0.753
Clinic, VUMC patient 154(64) 102 (50) 0.005** 257 (59) 263 (58) 0.849
Diabetes duration, yearsd 10.7 ± 7.7 11.3 ± 8.1 0.447 11.0 ± 8.1 11.4 ± 8.3 0.694
Number of prescribed medicationsdiabetes 2.2 ± 0.9 1.9 ± 0.8 0.002** 2.1 ± 0.8 2.1 ± 0.8 0.898
Insulin status, taking insulin 120 (50) 100 (50) 0.986 214 (50) 226 (50) 0.888
HbA1c test type 0.972 0.944
 Venipuncture 171(72) 141 (72) 312 (73) 327 (75)
 Mail-in kit 51 (21) 42 (22) 88 (21) 84 (19)
 Point-of-care 16 (7) 12 (6) 26 (6) 28 (6)
Open in a new tab

Notes: Boldface indicates statistical significance (*p<0.05, **p<0.01, ***p<0.001).

a

Cohort constructed via propensity score weighting to be balanced on covariates.

b

There was10.6% missingness in income, less than 5% missingness in HbA1c test type, insurance status, education and housing status, and less than 2% missingness in the remaining variables. Missingness rates (not shown) were similar between NHWs and NHBs in the weighted cohort.

c

Sum of weights in the constructed cohort; the propensity model scales the distribution of each group to match the distribution of the sample as a whole across all variables included in the model.

d

Models fit age, income, and diabetes duration with restricted cubic splines using five knots.

NHW, non-Hispanic white; NHB, non-Hispanic black; HS, high school; TAPS, Tool for Assessing Patients’ Stressors; VUMC, Vanderbilt University Medical Center.

Table 2.

Unadjusted and Adjusted Differences in Outcomes by Race (N=444)

Unweighted sample Weighted samplea
M ± SD or n (%) p-value M ± SD or n (%) p-value
Outcomesb NHW (n=242) NHB (n=202) NHW (nc=433) NHB (nc=450)
Health literacy (BHLS) 13.33 ± 2.49 12.99 ± 2.45 0.039* 13.31 ± 2.50 13.05 ± 2.43 0.327
Numeracy (SNS-3) 4.66 ± 1.32 4.22 ± 1.17 <0.001*** 4.51 ± 1.37 4.42 ± 1.21 0.520
Healthy diet (PDQ)
 Use of information for dietary decisions 3.14 ± 1.64 2.77 ± 1.65 0.013* 3.09 ± 1.67 2.85 ± 1.65 0.170
 Problem eating behaviors 3.53 ± 1.06 3.19 ± 1.11 <0.001*** 3.48 ± 1.07 3.19 ± 1.12 0.016*
 Physical activity (IPAQ) 2,238.4 ± 2,800.3 2,298.6 ± 2,879.9 0.778 2,264.2 ± 2,844.5 2,252.2 ± 2,714.9 0.966
Medication adherence (ARMS-D) 40.42 ± 3.36 39.47 ± 3.84 0.009** 40.12 ± 3.47 39.55 ± 3.98 0.179
Self-monitoring of blood glucose (SDSCA) 3.50 ± 2.88 3.77 ± 2.76 0.371 3.41 ± 2.91 3.96 ± 2.72 0.065
Smoking status (current smoker) 46 (19) 52 (26) 0.095 95 (22) 114 (26) 0.458
HbA1c (%) 8.31 ± 1.63 8.86 ± 1.97 0.005** 8.33 ± 1.68 8.82 ± 1.88 0.011*
Open in a new tab

Notes: Boldface indicates statistical significance (*p<0.05, **p<0.01, ***p<0.001). Possible score range: BHLS (3–15); SNS-3 (1–5); PDQ subscales (1–6); ARMS-D (11–44; items reverse-scored so higher scores indicate greater adherence); SDSCA (1–7).

a

Cohort constructed via propensity score weighting to be balanced on covariates.

b

There was less than 5% missingness in IPAQ scores and HbA1c, and less than 2% missingness in the remaining variables. Missingness rates (not shown) were similar between NHWs and NHBs in the weighted cohort.

c

Sum of weights in the constructed cohort.

NHW, non-Hispanic white; NHB, non-Hispanic black; BHLS, Brief Health Literacy Screen; SNS, Subjective Numeracy Scale; PDQ, Personal Diabetes Questionnaire; IPAQ, International Physical Activity Questionnaire; ARMS-D, Adherence to Refills and Medication Scale for Diabetes; SDSCA, Summary of Diabetes Self-Care Activities.

RESULTS

The sample (N=444) was nearly half (45%) NHB. In unweighted analyses (Table 2), there was no difference between NHBs and NHWs in MET minutes/week, self-monitoring of blood glucose, or smoking; however, compared with NHWs, NHBs had lower health literacy and numeracy, and less medication adherence, use of information for dietary decisions, and problem eating behaviors (such that NHBs reported fewer problems). In addition, NHBs had higher HbA1c values relative to NHWs. In weighted analyses, racial differences were attenuated for all outcomes, and only the problem eating behaviors reverse disparity and HbA1c disparity remained significant.

DISCUSSION

Among a diverse sample of adults diagnosed with T2D, racial disparities in determinants of diabetes outcomes present in unadjusted analyses largely disappeared after balancing the sample on multiple dimensions of SES. HbA1c, although attenuated, remained near 0.5% higher among NHBs relative to NHWs. This suggests additional factors (e.g., racial discrimination in health care)34,35 may contribute to the HbA1c disparity. The reverse disparity in problem eating could be due to cultural differences in the perception of eating behaviors as problematic,36,37 leading to differential interpretation of scale items; however, more research is needed.

This study highlights the importance of considering SES when examining disparities in determinants of diabetes outcomes. In past studies reporting worse self-care and health-related skills among NHBs,11,13,14 SES may have been a key omitted variable driving those disparities. In a recent systematic review on racial disparities in T2D self-management, only nine of 25 studies adjusted for even a single marker of SES.11 Seven of the nine studies reported a racial disparity persisted in at least one behavior, but almost all included only income and/or education in their SES adjustment. SES is a multidimensional construct and use of these more traditional indicators limits understanding of its role38,39 (e.g., years of education does not consider other types of training; income often fails to capture financial strain).39

Limitations

Limitations include use of self-reported measures that are vulnerable to social desirability and memory biases. In addition, findings were based on cross-sectional research limiting casual inferences. Relevant confounding variables were selected, but inclusion of other/additional confounders could affect results. Lastly, recruiting from clinics in a single county in Tennessee and self-selection bias may limit generalizability of the findings.

CONCLUSIONS

Many initiatives focus on eliminating racial disparities in health; the role of socioeconomic factors should be fully considered in these efforts to advance health equity. Specifically, multidimensional assessments of SES are needed to advance efforts to eliminate disparities. Because race is a social construct representing different meanings in different samples and over time,18,40 targeting upstream social determinants of health, which have remained consistently and markedly different between races, may be more effective in reducing health disparities.41

ACKNOWLEDGMENTS

The authors thank the Rapid Encouragement/Education And Communications for Health (REACH) team, our partnering clinics (i.e., Faith Family Medical Center, The Clinic at Mercury Courts, Connectus Health, Shade Tree Clinic, United Neighborhood Health Services, Vanderbilt Adult Primary Care), and the participants for their contributions to this research. The research presented in this paper is that of the authors and does not reflect the official policy of the NIH.

The National Institute of Diabetes and Digestive and Kidney Diseases (R01 DK100694, PI Mayberry) supported the research and all authors’ work. Dr. Mayberry was also supported by a Mentored Career Development Award from the National Institute of Diabetes and Digestive and Kidney Diseases (K01 DK106306). The study sponsor had no role in study design; collection, analysis, interpretation of data; writing the report, or the decision to submit the report for publication.

LAN led the study design and wrote the manuscript. MTA conducted the analyses and edited the manuscript. RAG planned and oversaw the analyses and edited the manuscript. KAW helped plan the analyses and reviewed and edited the manuscript. LSW oversaw the parent study, planned analyses, and reviewed and edited the manuscript.

KAW is on the Advisory Board of EdLogics, Inc. No other conflicts of interest were reported by the others of this paper.

Footnotes

Publisher's Disclaimer: This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final citable form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

REFERENCES

  • 1.CDC. National diabetes statistics report, 2017: estimates of diabetes and its burden in the United States; 2017. www.cdc.gov/diabetes/pdfs/data/statistics/national-diabetes-statistics-report.pdf. Published 2017. Accessed February 15, 2019.
  • 2.American Diabetes Association. Economic costs of diabetes in the U.S. in 2017. Diabetes Care. 2018;41(5):917–928. 10.2337/dci18-0007. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Bommer C, Sagalova V, Heesemann E, et al. Global economic burden of diabetes in adults: projections from 2015 to 2030. Diabetes Care. 2018;41(5):963–970. 10.2337/dc17-1962. [DOI] [PubMed] [Google Scholar]
  • 4.American Association of Diabetes Educators. AADE7™ self-care behaviors. www.diabeteseducator.org/docs/default-source/legacy-docs/_resources/pdf/publications/aade7_position_statement_final.pdf?sfvrsn=4. Published December 3, 2014. Accessed February 15, 2019. [Google Scholar]
  • 5.Beck J, Greenwood DA, Blanton L, et al. 2017 National Standards for Diabetes Self-Management Education and Support. Diabetes Care. 2017;40(10):1409–1419. 10.2337/dci17-0025. [DOI] [PubMed] [Google Scholar]
  • 6.Fayfman M, Vellanki P, Alexopoulos AS, et al. Report on racial disparities in hospitalized patients with hyperglycemia and diabetes. J Clin Endocrinol Metab. 2016;101(3):1144–1150. 10.1210/jc.2015-3220. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Hazel-Fernandez L, Li Y, Nero D, et al. Racial/ethnic and gender differences in severity of diabetes-related complications, health care resource use, and costs in a medicare population. Popul Health Manag. 2015;18(2):115–122. 10.1089/pop.2014.0038. [DOI] [PubMed] [Google Scholar]
  • 8.Herman WH, Dungan KM, Wolffenbuttel BH, et al. Racial and ethnic differences in mean plasma glucose, hemoglobin A1c, and 1,5-anhydroglucitol in over 2000 patients with type 2 diabetes. J Clin Endocrinol Metab. 2009;94(5):1689–1694. 10.1210/jc.2008-1940. [DOI] [PubMed] [Google Scholar]
  • 9.Weinstock RS, Teresi JA, Goland R, et al. Glycemic control and health disparities in older ethnically diverse underserved adults with diabetes: five-year results from the Informatics for Diabetes Education and Telemedicine (IDEATel) study. Diabetes Care. 2011;34(2):274–279. 10.2337/dc10-1346. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Kirk JK, D’Agostino RB, Bell RA, et al. Disparities in HbA1c levels between African-American and non-Hispanic white adults with diabetes: a meta-analysis. Diabetes Care. 2006;29(9):2130–2136. 10.2337/dc05-1973. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Mayberry LS, Bergner EM, Chakkalakal RJ, Elasy TA, Osborn CY. Self-care disparities among adults with type 2 diabetes in the USA. Curr Diab Rep. 2016;16(11):113 10.1007/s11892-016-0796-5. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Peyrot M, Egede LE, Funnell MM, et al. U.S. ethnic group differences in self-management in the 2nd Diabetes Attitudes, Wishes and Needs (DAWN2) study. J Diabetes Complications. 2018;32(6):586–592. 10.1016/j.jdiacomp.2018.03.002. [DOI] [PubMed] [Google Scholar]
  • 13.Ginde AA, Clark S, Goldstein JN, Camargo CA Jr. Demographic disparities in numeracy among emergency department patients: evidence from two multicenter studies. Patient Educ Couns. 2008;72(2):350–356. 10.1016/j.pec.2008.03.012. [DOI] [PubMed] [Google Scholar]
  • 14.Weekes CV. African Americans and health literacy: a systematic review. ABNF J. 2012;23(4):76–80. [PubMed] [Google Scholar]
  • 15.Osborn CY, Cavanaugh K, Wallston KA, et al. Health literacy explains racial disparities in diabetes medication adherence. J Health Commun. 2011;16(suppl 3):268–278. 10.1080/10810730.2011.604388. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Yudell M, Roberts D, DeSalle R, Tishkoff S. Taking race out of human genetics. Science. 2016;351(6273):564–565. 10.1126/science.aac4951. [DOI] [PubMed] [Google Scholar]
  • 17.Rosenberg NA, Pritchard JK, Weber JL, et al. Genetic structure of human populations. Science. 2002;298(5602):2381–2385. 10.1126/science.1078311. [DOI] [PubMed] [Google Scholar]
  • 18.Williams DR. Race and health: basic questions, emerging directions. Ann Epidemiol. 1997;7(5):322–333. 10.1016/S1047-2797(97)00051-3. [DOI] [PubMed] [Google Scholar]
  • 19.Pew Research Center. On views of race and inequality, blacks and whites are worlds apart. http://assets.pewresearch.org/wp-content/uploads/sites/3/2016/06/ST_2016.06.27_Race-Inequality-Final.pdf. Published June 27, 2016. Accessed February 15, 2019.
  • 20.Nelson LA, Wallston KA, Kripalani S, et al. Mobile phone support for diabetes self-care among diverse adults: protocol for a three-arm randomized controlled trial. JMIR Res Protoc. 2018;7(4):e92 10.2196/resprot.9443. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Osborn CY, Mayberry LS, Wagner JA, Welch GW. Stressors may compromise medication adherence among adults with diabetes and low socioeconomic status. West J Nurs Res. 2014;36(9):1091–1110. 10.1177/0193945914524639. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Wallston KA, Cawthon C, McNaughton CD, Rothman RL, Osborn CY, Kripalani S. Psychometric properties of the brief health literacy screen in clinical practice. J Gen Intern Med. 2014;29(1):119–126. 10.1007/s11606-013-2568-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.McNaughton CD, Cavanaugh KL, Kripalani S, Rothman RL, Wallston KA. Validation of a short, 3-item version of the subjective numeracy scale. Med Decis Making. 2015;35(8):932–936. 10.1177/0272989X15581800. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Stetson B, Schlundt D, Rothschild C, Floyd JE, Rogers W, Mokshagundam SP. Development and validation of The Personal Diabetes Questionnaire (PDQ): a measure of diabetes self-care behaviors, perceptions and barriers. Diabetes Res Clin Pract. 2011;91(3):321–332. 10.1016/j.diabres.2010.12.002. [DOI] [PubMed] [Google Scholar]
  • 25.Akohoue SA, Wallston KA, Schlundt DG, Rothman RL. Psychometric evaluation of the short version of the Personal Diabetes Questionnaire to assess dietary behaviors and exercise in patients with type 2 diabetes. Eat Behav. 2017;26:182–188. 10.1016/j.eatbeh.2017.04.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Lee PH, Macfarlane DJ, Lam TH, Stewart SM. Validity of the International Physical Activity Questionnaire Short Form (IPAQ-SF): a systematic review. Int J Behav Nutr Phys Act. 2011;8:115 10.1186/1479-5868-8-115. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 27.Kim Y, Park I, Kang M. Convergent validity of the international physical activity questionnaire (IPAQ): Meta-analysis. Public Health Nutr. 2013;16(3):440–452. 10.1017/S1368980012002996. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Mayberry LS, Gonzalez JS, Wallston KA, Kripalani S, Osborn CY. The ARMS-D out performs the SDSCA, but both are reliable, valid, and predict glycemic control. Diabetes Res Clin Pract. 2013;102(2):96–104. 10.1016/j.diabres.2013.09.010. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Toobert DJ, Hampson SE, Glasgow RE. The Summary of Diabetes Self-Care Activities measure: results from 7 studies and a revised scale. Diabetes Care. 2000;23(7):943–950. 10.2337/diacare.23.7.943. [DOI] [PubMed] [Google Scholar]
  • 30.CDC. Behavioral Risk Factor Surveillance System Survey questionnaire. Atlanta, GA: CDC, HHS; 2017. [Google Scholar]
  • 31.Lorig K, Ritter PL, Moreland C, Laurent DD. Can a box of mailed materials achieve the triple aims of health care? The Mailed Chronic Disease Self-Management Tool Kit study. Health Promot Pract. 2015;16(5):765–774. 10.1177/1524839915571633. [DOI] [PubMed] [Google Scholar]
  • 32.Jones TG, Warber KD, Roberts BD. Analysis of hemoglobin A1c from dried blood spot samples with the Tina-quant® II immunoturbidimetric method. J Diabetes Sci Technol. 2010;4(2):244–249. 10.1177/193229681000400203. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Harrell FE. Regression Modeling Strategies With Applications to Linear Models, Logistic and Ordinal Regression, and Survival Analysis. 2nd ed. New York, NY: Springer; 2015. [Google Scholar]
  • 34.Peek ME, Wagner J, Tang H, Baker DC, Chin MH. Self-reported racial discrimination in health care and diabetes outcomes. Medical Care. 2011;49(7):618–625. 10.1097/MLR.0b013e318215d925. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Nelson A Unequal treatment: confronting racial and ethnic disparities in health care. J Natl Med Assoc. 2002;94(8):666–668. [PMC free article] [PubMed] [Google Scholar]
  • 36.Ricciardelli LA, McCabe MP, Williams RJ, Thompson JK. The role of ethnicity and culture in body image and disordered eating among males. Clin Psychol Rev. 2007;27(5):582–606. 10.1016/j.cpr.2007.01.016. [DOI] [PubMed] [Google Scholar]
  • 37.Gluck ME, Geliebter A. Racial/ethnic differences in body image and eating behaviors. Eat Behav. 2002;3(2):143–151. 10.1016/S1471-0153(01)00052-6. [DOI] [PubMed] [Google Scholar]
  • 38.Herd P, Goesling B, House JS. Socioeconomic position and health: the differential effects of education versus income on the onset versus progression of health problems. J Health Soc Behav. 2007;48(3):223–238. 10.1177/002214650704800302. [DOI] [PubMed] [Google Scholar]
  • 39.Shavers VL. Measurement of socioeconomic status in health disparities research. J Natl Med Assoc. 2007;99(9):1013. [PMC free article] [PubMed] [Google Scholar]
  • 40.Lee C “Race” and “ethnicity” in biomedical research: How do scientists construct and explain differences in health? Soc Sci Med. 2009;68(6):1183–1190. 10.1016/j.socscimed.2008.12.036. [DOI] [PubMed] [Google Scholar]
  • 41.Thornton RLJ, Glover CM, Cené CW, Glik DC, Henderson JA, Williams DR. Evaluating strategies for reducing health disparities by addressing the social determinants of health. Health Affairs (Millwood). 2016;35(8):1416–1423. 10.1377/hlthaff.2015.1357. [DOI] [PMC free article] [PubMed] [Google Scholar]

RESOURCES