To examine the association between diabetes stigma, socioeconomic status, psychosocial variables, and substance use in adolescents and young adults (AYAs) with type 1 or type 2 diabetes.
This is a cross-sectional analysis of AYAs from the SEARCH for Diabetes in Youth study who completed a survey on diabetes-related stigma, generating a total diabetes stigma score. Using multivariable modeling, stratified by diabetes type, we examined the relationship of diabetes stigma with variables of interest.
Of the 1,608 AYAs who completed the diabetes-related stigma survey, 78% had type 1 diabetes, and the mean age was 21.7 years. Higher diabetes stigma scores were associated with food insecurity (P = 0.001), disordered eating (P < 0.0001), depressive symptoms (P < 0.0001), and decreased health-related (P < 0.0001) and diabetes-specific quality of life (P < 0.0001).
Diabetes stigma is associated with food insecurity, disordered eating, and lower psychosocial well-being.
Introduction
Diabetes mellitus is a chronic disease that is associated with an increased burden of psychosocial comorbidities (1–3). The physical and psychological aspects of diabetes care have been well described (1), but there is only a limited body of research on the social burden of diabetes, which may be examined by assessing diabetes-related stigma. Health-related stigma is defined as a social process or personal experience characterized by rejection, blame, or exclusion that results from an adverse social judgment about a person or group with a specific health condition (4). The adolescent and young adult (AYA) population may experience or perceive more consequences of stigma given their developmental stage (5).
In AYAs in the SEARCH for Diabetes in Youth study, diabetes-related stigma has been shown to be associated with the female sex, higher HbA1c, and some chronic complications, in addition to diabetic ketoacidosis and severe hypoglycemia in those with type 1 diabetes (6). There are two small prior studies on diabetes stigma and psychosocial well-being in AYAs, with one demonstrating that diabetes-related stigma was associated with depressive symptoms in Latino youth with type 1 diabetes (n = 65) (7) and the other showing diabetes-related stigma was associated with lower well-being and self-efficacy in diabetes management in adolescents with type 1 diabetes (n = 380) (8). In adults, higher levels of perceived diabetes-related stigma have been associated with higher levels of diabetes distress (9–12), increased depressive symptoms (10,11), increased anxiety symptoms (11), and lower quality of life (QoL) (10,12).
The aim of this study was to assess the association of diabetes-related stigma with socioeconomic status, psychosocial outcomes, and substance use in AYAs with type 1 or type 2 diabetes. We hypothesized that disordered eating, depressive symptoms, and substance use are positively associated with diabetes stigma, whereas health-related QoL measures are negatively associated.
Research Design and Methods
The SEARCH for Diabetes in Youth study is a multicenter study designed to follow the clinical course of individuals diagnosed with type 1 or type 2 diabetes in youth in the United States. Our study included SEARCH participants who were diagnosed with type 1 diabetes or type 2 diabetes before age 20 years, completed an in-person study visit, completed a questionnaire regarding diabetes-related stigma, and were 10 to 24.9 years of age at the time of the study visit (Fig. 1). Methods for this study were previously described (6) (Supplementary Methods).
Family history of diabetes was collected and included biological parents, full siblings, and half-siblings. Participants self-reported whether they were U.S. born or foreign born for themselves and their parents. Substance use was collected and defined as alcohol use in the last 30 days, marijuana use in the last 30 days, tobacco use in the last 30 days, former tobacco use beyond the last 30 days, and never using tobacco. Food insecurity was assessed with the 18-item United States Household Food Security Survey Module, with the first 10 questions pertaining to all households with or without youth and the last 8 questions specific to households with youth ages 0–17 years. Food insecure is defined as three or more affirmative responses (13). The 16-item Diabetes Eating Problems Survey–Revised (DEPS-R), which is validated in youth with diabetes, was used to assess disordered eating, with a positive screen defined as a score of ≥20 (14). The 20-item Center for Epidemiological Studies–Depression (CES-D) validated survey instrument was used to assess depressive symptoms, with a total score ranging from 0 to 60 and a cutoff score of ≥16 identifying people at risk for clinical depression (15). QoL measures included the 23-item PedsQL Young Adult QoL Inventory, which is validated for measuring health-related QoL (16), and the 33-item PedsQL Diabetes Module, which is validated for measuring diabetes-specific health-related QoL in youth and young adults with type 1 or type 2 diabetes (17,18).
Participants completed a five-question survey on diabetes-related stigma (Supplementary Table 1) that was scored on a 6-point Likert scale to create a total diabetes-related stigma score, ranging from 5 to 30 points, with the higher number indicating greater perception or experience of diabetes-related stigma (6). Descriptive statistics were calculated for all variables of interest. Each measure was examined in the full participants’ data set and then stratified by type 1 diabetes and type 2 diabetes. Next, we examined a series of general linear models with β coefficients and P values to determine the relationship between diabetes-related stigma scores (the outcome in the model) and variables of interest, with analyses stratified by diabetes type. Analyses were performed using SAS (Version 9; SAS Institute, Cary, NC).
Results
The demographics of the 1,608 participants are shown in Table 1, along with the average CES-D, PedsQL Young Adult QoL, and PedsQL Diabetes Module scores. Of all participants, 25.5% had a positive DEPS-R score, and 24.2% had a positive CES-D score. Participants with type 1 diabetes had an average (SD) diabetes-related stigma score of 10.9 (5.4), and those with type 2 diabetes had a diabetes-related stigma score of 9.8 (5.6).
Characteristics of included SEARCH 4 participants
. | Type 1 diabetes . | Type 2 diabetes . | Total sample . |
---|---|---|---|
N (%) | 1,255 (78) | 353 (22) | 1,608 (100) |
Sex | |||
Female | 668 (53.2) | 235 (66.6) | 903 (56.2) |
Male | 587 (46.8) | 118 (33.4) | 705 (43.8) |
Age at cohort visit, mean (SD), years | 20.8 (5.01) | 24.6 (4.31) | 21.7 (5.11) |
Diabetes duration, mean (SD), years | 11.1 (3.36) | 10.3 (3.54) | 10.9 (3.42) |
Race and ethnic group | |||
American Indian or Alaska Native | 9 (0.7) | 29 (8.2) | 38 (2.4) |
Asian or Pacific Islander | 20 (1.6) | 6 (1.7) | 26 (1.6) |
Hispanic | 268 (21.4) | 86 (24.4) | 354 (22) |
Multiracial | 67 (5.3) | 11 (3.1) | 78 (4.9) |
Non-Hispanic Black | 179 (14.3) | 154 (43.6) | 333 (20.7) |
Non-Hispanic White | 712 (56.7) | 67 (19.0) | 779 (48.5) |
Highest education | |||
≥High school | 802 (63.9) | 293 (83.0) | 1,095 (68.1) |
<High school | 444 (35.4) | 59 (16.7) | 503 (31.3) |
≤18 years old | 9 (0.1) | 1 (0.3) | 10 (0.6) |
Employment status | |||
Employed | 615 (49.0) | 188 (53.2) | 803 (49.9) |
Unemployed | 136 (10.8) | 72 (20.4) | 208 (12.9) |
Student | 71 (5.6) | 13 (3.7) | 84 (5.2) |
Disabled | 19 (1.5) | 38 (10.8) | 57 (3.5) |
Other/do not know | 30 (2.4) | 19 (5.4) | 49 (3.0) |
≤18 years old | 384 (30.0) | 23 (6.5) | 407 (25.3) |
Insurance types | |||
Private | 909 (72.4) | 156 (44.2) | 1,065 (66.2) |
Public | 208 (16.6) | 105 (29.7) | 313 (19.5) |
None/unknown | 138 (11.0) | 92 (26.1) | 230 (14.3) |
BMI | |||
<25 kg/m2 | 640 (51.0) | 24 (6.8) | 664 (41.3) |
25–29.9 kg/m2 | 376 (30.0) | 65 (18.4) | 441 (27.4) |
≥30 kg/m2 | 238 (19.0) | 264 (74.8) | 502 (31.2) |
HbA1c, mean (SD)% (mmol/mol) | 9.1 (2.1) (76) | 9.5 (2.9) (80) | 9.2 (2.3) (77) |
Positive DEPS-R score | 276 (22.2) | 129 (37.2) | 405 (25.5) |
PedsQL Young Adult Quality of Life score, mean (SD) | 79.1 (14.9) | 76.1 (17.0) | 78.4 (15.4) |
PedsQL Diabetes Module score, mean (SD) | 70.6 (14.4) | 70.8 (16.9) | 70.7 (15.0) |
CES-D score, mean (SD) | 10.2 (9.0) | 13.1 (9.1) | 10.8 (9.1) |
Total diabetes stigma score, mean (SD) | 10.9 (5.4) | 9.8 (5.6) | 10.7 (5.5) |
Q1: Blame and Judgement* | 3.2 (1.7) | 2.7 (1.8) | 3.1 (1.7) |
Q2: Treated Differently* | 1.8 (1.3) | 1.6 (1.3) | 1.8 (1.3) |
Q3: Self-Stigma* | 2.5 (1.7) | 2.3 (1.8) | 2.4 (1.7) |
Q4: Diabetes Management* | 1.7 (1.3) | 1.6 (1.3) | 1.7 (1.3) |
Q5: Social Engagement* | 1.6 (1.2) | 1.6 (1.3) | 1.6 (1.2) |
. | Type 1 diabetes . | Type 2 diabetes . | Total sample . |
---|---|---|---|
N (%) | 1,255 (78) | 353 (22) | 1,608 (100) |
Sex | |||
Female | 668 (53.2) | 235 (66.6) | 903 (56.2) |
Male | 587 (46.8) | 118 (33.4) | 705 (43.8) |
Age at cohort visit, mean (SD), years | 20.8 (5.01) | 24.6 (4.31) | 21.7 (5.11) |
Diabetes duration, mean (SD), years | 11.1 (3.36) | 10.3 (3.54) | 10.9 (3.42) |
Race and ethnic group | |||
American Indian or Alaska Native | 9 (0.7) | 29 (8.2) | 38 (2.4) |
Asian or Pacific Islander | 20 (1.6) | 6 (1.7) | 26 (1.6) |
Hispanic | 268 (21.4) | 86 (24.4) | 354 (22) |
Multiracial | 67 (5.3) | 11 (3.1) | 78 (4.9) |
Non-Hispanic Black | 179 (14.3) | 154 (43.6) | 333 (20.7) |
Non-Hispanic White | 712 (56.7) | 67 (19.0) | 779 (48.5) |
Highest education | |||
≥High school | 802 (63.9) | 293 (83.0) | 1,095 (68.1) |
<High school | 444 (35.4) | 59 (16.7) | 503 (31.3) |
≤18 years old | 9 (0.1) | 1 (0.3) | 10 (0.6) |
Employment status | |||
Employed | 615 (49.0) | 188 (53.2) | 803 (49.9) |
Unemployed | 136 (10.8) | 72 (20.4) | 208 (12.9) |
Student | 71 (5.6) | 13 (3.7) | 84 (5.2) |
Disabled | 19 (1.5) | 38 (10.8) | 57 (3.5) |
Other/do not know | 30 (2.4) | 19 (5.4) | 49 (3.0) |
≤18 years old | 384 (30.0) | 23 (6.5) | 407 (25.3) |
Insurance types | |||
Private | 909 (72.4) | 156 (44.2) | 1,065 (66.2) |
Public | 208 (16.6) | 105 (29.7) | 313 (19.5) |
None/unknown | 138 (11.0) | 92 (26.1) | 230 (14.3) |
BMI | |||
<25 kg/m2 | 640 (51.0) | 24 (6.8) | 664 (41.3) |
25–29.9 kg/m2 | 376 (30.0) | 65 (18.4) | 441 (27.4) |
≥30 kg/m2 | 238 (19.0) | 264 (74.8) | 502 (31.2) |
HbA1c, mean (SD)% (mmol/mol) | 9.1 (2.1) (76) | 9.5 (2.9) (80) | 9.2 (2.3) (77) |
Positive DEPS-R score | 276 (22.2) | 129 (37.2) | 405 (25.5) |
PedsQL Young Adult Quality of Life score, mean (SD) | 79.1 (14.9) | 76.1 (17.0) | 78.4 (15.4) |
PedsQL Diabetes Module score, mean (SD) | 70.6 (14.4) | 70.8 (16.9) | 70.7 (15.0) |
CES-D score, mean (SD) | 10.2 (9.0) | 13.1 (9.1) | 10.8 (9.1) |
Total diabetes stigma score, mean (SD) | 10.9 (5.4) | 9.8 (5.6) | 10.7 (5.5) |
Q1: Blame and Judgement* | 3.2 (1.7) | 2.7 (1.8) | 3.1 (1.7) |
Q2: Treated Differently* | 1.8 (1.3) | 1.6 (1.3) | 1.8 (1.3) |
Q3: Self-Stigma* | 2.5 (1.7) | 2.3 (1.8) | 2.4 (1.7) |
Q4: Diabetes Management* | 1.7 (1.3) | 1.6 (1.3) | 1.7 (1.3) |
Q5: Social Engagement* | 1.6 (1.2) | 1.6 (1.3) | 1.6 (1.2) |
Data are given as n (%) unless otherwise indicated. DEPS-R measures disordered eating. CES-D assesses depressive symptoms.
See Supplementary Table 1 for list of questions. Questions were scored on a six-point Likert scale: never (one point), less than once a year (two points), a few times a year (three points), a few times a month (four points), at least once a week (five points), or almost daily (six points).
Socioeconomic Variables
Table 2 shows the association of sociodemographic variables, including being a non-U.S.-born person and having a family member with diabetes, with diabetes-related stigma scores. Food insecurity was associated with a 3.26-point higher diabetes-related stigma score in those with type 1 diabetes (P < 0.001) and a 2.12-point higher diabetes-related stigma score in those with type 2 diabetes (P = 0.0017). As previously reported, female sex (P < 0.001 for AYAs with type 1 diabetes; P = 0.0026 for AYAs with type 2 diabetes) and household income less than $25,000 for AYAs with type 1 diabetes (P = 0.031) were associated with higher diabetes-related stigma scores, but race, ethnicity, education, health insurance, and employment status were not associated with diabetes-related stigma scores (Supplementary Table 2) (6).
Multivariable linear models of continuous total diabetes-related stigma score and socioeconomic and psychosocial variables stratified by diabetes type
. | Total diabetes-related stigma score . | |||
---|---|---|---|---|
. | Type 1 diabetes . | Type 2 diabetes . | ||
Variables . | β (SE) . | P value . | β (SE) . | P value . |
Non–U.S.-born people | ||||
Patient non–U.S. born | 0.70 (1.16) | 0.5451 | −2.23 (1.72) | 0.1978 |
Mother non–U.S. born | 1.02 (0.75) | 0.1728 | 0.77 (2.18) | 0.7259 |
Father non–U.S. born | −0.23 (0.54) | 0.6657 | 0.35 (1.25) | 0.7816 |
Patient and parents are U.S. born | ref | ref | ||
≥1 family member with diabetes | 0.001 (0.34) | 0.9963 | −0.46 (0.65) | 0.4781 |
Positive DEPS-R score | 4.20 (0.36) | <0.0001 | 3.40 (0.62) | <0.0001 |
PedsQL Young Adult Quality of Life score: linear | −0.18 (0.008) | <0.0001 | −0.17 (0.01) | <0.0001 |
PedsQL Diabetes Module score: linear | −0.22 (0.009) | <0.0001 | −0.21 (0.01) | <0.0001 |
CES-D score: linear | 0.27 (0.01) | <0.0001 | 0.24 (0.03) | <0.0001 |
Food insecurity | 3.26 (0.41) | <0.0001 | 2.12 (0.67) | 0.0017 |
. | Total diabetes-related stigma score . | |||
---|---|---|---|---|
. | Type 1 diabetes . | Type 2 diabetes . | ||
Variables . | β (SE) . | P value . | β (SE) . | P value . |
Non–U.S.-born people | ||||
Patient non–U.S. born | 0.70 (1.16) | 0.5451 | −2.23 (1.72) | 0.1978 |
Mother non–U.S. born | 1.02 (0.75) | 0.1728 | 0.77 (2.18) | 0.7259 |
Father non–U.S. born | −0.23 (0.54) | 0.6657 | 0.35 (1.25) | 0.7816 |
Patient and parents are U.S. born | ref | ref | ||
≥1 family member with diabetes | 0.001 (0.34) | 0.9963 | −0.46 (0.65) | 0.4781 |
Positive DEPS-R score | 4.20 (0.36) | <0.0001 | 3.40 (0.62) | <0.0001 |
PedsQL Young Adult Quality of Life score: linear | −0.18 (0.008) | <0.0001 | −0.17 (0.01) | <0.0001 |
PedsQL Diabetes Module score: linear | −0.22 (0.009) | <0.0001 | −0.21 (0.01) | <0.0001 |
CES-D score: linear | 0.27 (0.01) | <0.0001 | 0.24 (0.03) | <0.0001 |
Food insecurity | 3.26 (0.41) | <0.0001 | 2.12 (0.67) | 0.0017 |
All models are adjusted for sex, race, ethnicity, age, clinic site, duration of diabetes, treatment plan, continuous HbA1c, education level, and insurance status. DEPS-R measures disordered eating. CES-D assesses depressive symptoms. ref, reference.
Psychosocial Measures
A positive DEPS-R score was associated with a 4.2-point higher diabetes-related stigma score in participants with type 1 diabetes (P < 0.0001) and a 3.4-point higher diabetes-related stigma score in those with type 2 diabetes (P < 0.0001). The CES-D score was positively associated with diabetes-related stigma scores for all participants (P < 0.0001). The PedsQL Young Adult QoL and PedsQL Diabetes Module scores were negatively associated with diabetes-related stigma scores for all participants (P < 0.0001).
Substance Use
Fewer participants completed the surveys on substance use (n = 537), as shown in Table 3. In participants with type 1 diabetes, current tobacco use was associated with a 1.57-point higher diabetes-related stigma score compared with never using tobacco (P = 0.03), whereas alcohol use in the last 30 days was associated with a 1.68-point lower diabetes-related stigma score (P = 0.012). There was no significant association between diabetes-related stigma scores and marijuana use for all participants. For participants with type 2 diabetes, there were no significant associations between diabetes-related stigma scores and tobacco or alcohol use.
Multivariable linear model of continuous total diabetes-related stigma score and substance use measurements
. | Total diabetes-related stigma score . | |||
---|---|---|---|---|
. | Type 1 diabetes (n = 418) . | Type 2 diabetes (n = 119) . | ||
Substance . | β (SE) . | P value . | β (SE) . | P value . |
Tobacco | ||||
Current | 1.57 (0.72) | 0.03 | −1.17 (1.42) | 0.41 |
Former | 0.11 (0.62) | 0.86 | −1.10 (1.36) | 0.42 |
Never | ref | ref | ||
Alcohol in last 30 days | −1.68 (0.67) | 0.012 | 0.15 (1.22) | 0.90 |
Marijuana in last 30 days | 0.31 (0.53) | 0.56 | −0.34 (1.18) | 0.77 |
. | Total diabetes-related stigma score . | |||
---|---|---|---|---|
. | Type 1 diabetes (n = 418) . | Type 2 diabetes (n = 119) . | ||
Substance . | β (SE) . | P value . | β (SE) . | P value . |
Tobacco | ||||
Current | 1.57 (0.72) | 0.03 | −1.17 (1.42) | 0.41 |
Former | 0.11 (0.62) | 0.86 | −1.10 (1.36) | 0.42 |
Never | ref | ref | ||
Alcohol in last 30 days | −1.68 (0.67) | 0.012 | 0.15 (1.22) | 0.90 |
Marijuana in last 30 days | 0.31 (0.53) | 0.56 | −0.34 (1.18) | 0.77 |
Model adjusted for age, sex, race, ethnicity, clinic site, duration of diabetes, treatment plan, continuous HbA1c, education, and insurance status. ref, reference.
Conclusions
In this cross-sectional analysis, we found that, in AYAs with type 1 or type 2 diabetes food insecurity, disordered eating, depressive symptoms, and decreased QoL are associated with diabetes stigma. Additionally, in AYAs with type 1 diabetes, tobacco use in the last 30 days was positively associated and alcohol use was negatively associated with diabetes stigma.
Diabetes stigma may be an important element to address in comprehensive diabetes care, as it is associated with worse clinical outcomes (6) and interrelates with psychosocial functioning in AYAs with type 1 or type 2 diabetes. Given the strong associations of diabetes-related stigma with psychological measures, addressing diabetes-related stigma may improve psychosocial functioning in AYAs with type 1 or type 2 diabetes. This can be done at the societal level with public health education on living with diabetes, in addition to at the individual level by building one’s resiliency and self-esteem.
Tobacco-related stigma has been well established (19). Therefore, AYAs with types 1 diabetes who use tobacco may experience or perceive a compounded tobacco-related stigma and diabetes-related stigma. The intersection of these stigmas should be further examined in a longitudinal study to further elucidate the temporal relationships and driving factors between diabetes stigma and tobacco stigma. Unexpectedly, alcohol use in the last 30 days was negatively associated with diabetes-related stigma scores. This may be due to social alcohol consumption decreasing the social burden of diabetes, stress, or depression (20).
Additionally, food insecurity is associated with diabetes stigma. These findings suggest that AYAs with more socioeconomic stressors are at high risk of experiencing diabetes-related stigma. This finding can help guide future screening and interventions, as well as support efforts to address social determinants of health such as food insecurity.
Limitations of this study include the cross-sectional design, fewer AYAs with type 2 diabetes, a smaller sample for substance use measurements, and an unvalidated diabetes-related stigma survey. Next steps may include conducting a larger prospective and longitudinal assessment of diabetes-related stigma to further elucidate the temporal relationships between diabetes stigma, diabetes self-care, and psychosocial functioning. Discussing diabetes stigma when providing comprehensive diabetes care, particularly in the AYA period when social relationships and personal identity are developing, may lead to improved psychosocial functioning and well-being.
This article contains supplementary material online at https://doi.org/10.2337/figshare.24612357.
Article Information
Funding. Grant support for SEARCH 4 is as follows: The SEARCH for Diabetes in Youth cohort study (1R01DK127208-01, 1UC4DK108173) is funded by the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases, and supported by the Centers for Disease Control and Prevention. The Population Based Registry of Diabetes in Youth Study (1U18DP006131, U18DP006133, U18DP006134, U18DP006136, U18DP006138, and U18DP006139) is funded by the Centers for Disease Control and Prevention (DP-15-002) and supported by the National Institutes of Health, National Institute of Diabetes and Digestive and Kidney Diseases. Grant support for SEARCH 1, 2, and 3 is as follows: SEARCH for Diabetes in Youth is funded by the Centers for Disease Control and Prevention (PA numbers 00097, DP-05-069, and DP-10-001) and supported by the National Institute of Diabetes and Digestive and Kidney Diseases. Support was also provided by Kaiser Permanente Southern California (U48/CCU919219, U01 DP000246, and U18DP002714), University of Colorado Denver (U48/CCU819241-3, U01 DP000247, and U18DP000247-06A1), Cincinnati's Children’s Hospital Medical Center (U48/CCU519239, U01 DP000248, and 1U18DP002709), University of North Carolina at Chapel Hill (U48/CCU419249, U01 DP000254, and U18DP002708), Seattle Children’s Hospital (U58/CCU019235-4, U01 DP000244, and U18DP002710-01), and Wake Forest University School of Medicine (U48/CCU919219, U01 DP000250, and 200-2010-35171). This study was also supported by 1R01DK117461.
Duality of Interest. No potential conflicts of interest relevant to this article were reported.
Author Contributions. K.B.E., A.J.R., R.D., C.E.B., R.A.B., A.B., A.C., D.D., L.M.D., E.T.J., A.D.L., K.R., S.M.M., and C.P. were involved in the conception, design, and conduct of the study and the analysis and interpretation of the results. K.B.E. wrote the first draft of the manuscript, and all authors edited, reviewed, and approved the final version of the manuscript. C.P. is the guarantor of this work and, as such, had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.