Abstract
OBJECTIVE
Diabetes-dependent quality of life (DDQoL) is a salient health outcome for young adults with type 1 diabetes (T1D), shaped by diabetes distress and social support. This cross-sectional analysis examines 1) how diabetes distress subtypes relate to DDQoL, 2) whether online support is associated with DDQoL, and 3) whether online support moderates the relationship between diabetes distress and DDQoL.
RESEARCH DESIGN AND METHODS
We analyzed baseline data from a clinical trial of young adults with T1D (age 18–30 years, diabetes duration >1 year, HbA1c ≥7.5%). Participants completed the Diabetes Distress Scale, the Audit of Diabetes-Dependent Quality of Life, and a study-designed questionnaire on online support. Univariate and multivariate linear regressions were conducted.
RESULTS
Among 209 young adults (66% female, mean [SD] age 24.3 [3.7] years, 42.6% Hispanic or Latinx ethnicity), higher total diabetes distress and subscale scores were significantly associated with poorer DDQoL (average weighted impact score, P < 0.001), with emotional distress showing the strongest negative association (β –0.60, P < 0.001). Online support was also associated with lower DDQoL (β −0.17, P = 0.016). In the adjusted model, online support moderated the impact of interpersonal distress on DDQoL (interaction β 0.23, P = 0.026). Female gender was associated with better DDQoL (β 0.14, P = 0.031) and confounded key associations.
CONCLUSIONS
These findings revealed the differential associations between diabetes distress subtypes and DDQoL and the moderating role of online support in reducing the negative impact of interpersonal distress, underscoring the need for interventions targeting emotional distress and optimizing online support for young adults with T1D.
Introduction
Type 1 diabetes (T1D) is an autoimmune disease that requires lifelong self-management (1) and can adversely affect multiple domains of quality of life (QoL). Among various constructs of QoL, diabetes-dependent QoL (DDQoL) has been identified as a critical health outcome, reflecting both the comprehensive and individualized impacts of diabetes on individuals’ lives (2). Young adults, a population managing T1D with additional challenges due to increased life responsibilities and the potentially disrupted transitions from pediatric to independent adult health care (3), experience vulnerability of lowered DDQoL (4). Understanding the factors that shape DDQoL in this population is crucial. Research has found that diabetes distress is a significant determinant that negatively affects DDQoL (5), while social support serves as a protective factor, as reported in qualitative studies (6,7). However, several gaps remain in understanding the impact of these two constructs on DDQoL and their interaction in influencing DDQoL.
Diabetes distress refers to negative emotional responses to living with and managing diabetes (8) and is highly prevalent among young adults due to the developmental and health care transitions they experience (9). Higher levels of diabetes distress have been found to be associated with poorer self-management skills and reduced DDQoL in young adults with T1D (10). A commonly used conceptualization of diabetes distress parses it into four distinct dimensions: emotional, physician-related, regimen-related, and interpersonal distress (8). These subtypes offer nuanced insights into how individuals experience and respond to T1D-related distress (11), yet most studies have relied solely on the total score of diabetes distress when examining its associations with DDQoL (10) instead of subtype scores. As a result, little is known about how specific types of distress may differentially impact DDQoL. Addressing this gap is critical for designing tailored psychosocial interventions that target the most impactful domains of distress (12).
As mentioned, social support generally serves as a protective contextual factor of well-being in chronic conditions (6,7). According to the buffering hypothesis (13), social support influences health outcomes in two ways: by directly improving health outcomes (main effect) or by mitigating the harmful effects of stress on well-being (buffering effect). Manifesting the distinct effects of social support is crucial for decision-making in social support interventions (14).
Among individuals with diabetes, social support has been a targeted domain of self-care interventions to enhance DDQoL (15), indicating the direct benefits of general social support in QoL research. As digital technology becomes embedded in everyday life, online environments have emerged as prominent sources of social interaction and social support (16). Young adults with T1D are almost universally active on social media and online communities (17), where they may obtain support for diabetes management (18). However, studies have typically applied a qualitative approach in relatively small samples (e.g., ∼30 participants). Additionally, existing studies have yet to introduce a generic health outcome, such as the DDQoL, to explore the influence of online support.
In addition to the potential direct effect, currently, no studies have examined the buffering effect of online support on DDQoL in young adults with T1D. Drawing on the buffering hypothesis, we defined the buffering effect as the mechanism by which online support attenuates the negative impact of diabetes distress on DDQoL, such that individuals with similar levels of distress may experience different DDQoL outcomes depending on their use of online support. Addressing this gap is critical for informing digital support strategies tailored to this population, as it may clarify whether and how online support enhances DDQoL.
To summarize, three key gaps remain: 1) the lack of research examining how different diabetes distress subtypes relate to DDQoL, 2) the unexplored direct association of online support with DDQoL, and 3) the unknown buffering effect, whereby online support may attenuate the negative impact of diabetes distress on DDQoL. To address these gaps, the current study uses secondary analysis of cross-sectional data from a randomized controlled trial involving young adults with T1D (19). Specifically, we aimed to 1) examine how different subtypes of diabetes distress (emotional, physician-related, regimen-related, and interpersonal distress) are associated with DDQoL, 2) investigate whether online support is directly associated with DDQoL, and 3) test whether online support buffers the negative association between diabetes distress and DDQoL in young adults with T1D.
Research Design and Methods
This cross-sectional analysis used the baseline data from the Resilient, Empowered, Active Living–Telehealth (REAL-T) study, a randomized controlled trial evaluating the efficacy of an occupational therapy intervention (19). Young adults (age 18–30 years) with T1D were recruited between October 2019 and November 2022 through in-person clinics in the greater Los Angeles area and via social media outreach. The full study protocol and methodology are outlined in a previous publication (20).
Eligible participants had a T1D diagnosis for >1 year (based on medical records and/or self-report), had an HbA1c level ≥7.5%, and were proficient in English or Spanish. Exclusion criteria included severe cognitive impairment or disability that would prevent study participation, current pregnancy or plans to become pregnant during the study period, and prior participation in a diabetes self-management intervention within the past 12 months, as assessed during structured eligibility screening via videoconferencing.
A total of 209 participants provided informed consent and completed baseline demographic and self-report questionnaires. The questionnaires were delivered in English, with Spanish versions available upon request, and all participants chose to respond in English. Demographic information included age, gender, race and ethnicity, education level, income, and diabetes-related characteristics, such as HbA1c and diabetes devices. Participants also reported diabetes-related health care utilization indicators, including the occurrence of ketoacidosis and severe hypoglycemia in the past month. Diabetic ketoacidosis was assessed as self-reported episodes of high blood glucose with ketones diagnosed by a physician and requiring hospital, emergency department, or other health care facility evaluation. Severe hypoglycemia was defined as episodes of low blood glucose resulting in loss of consciousness or seizure. All study procedures were approved by the Institutional Review Board of the University of Southern California.
Measures
Audit of DDQoL
The Audit of Diabetes Dependent Quality of Life-15 (ADDQoL-15) assesses the perceived impact of diabetes across 15 life domains, such as work, leisure, travel, finances, social relationships, dietary freedom, physical health, self-confidence, and feelings about the future (21). Adapted from the ADDQoL-19 by Health Psychology Research Ltd. to reduce respondent burden, the ADDQoL-15 rewords the dependency item for clarity. For each applicable domain, respondents rate the impact of diabetes (from −3 for very much negative to 1 for positive) and its importance (from 0 for not at all to 3 for very important). Weighted scores are calculated by multiplying impact by importance, and the overall ADDQoL score is the mean of all applicable domains, ranging from −9 to 3, with higher scores indicating better DDQoL. In our 15-domain version (excluding the dependency item due to missing data), confirmatory factor analysis supported good model fit (comparative fit index 0.97, Tucker-Lewis index 0.96, root mean square error of approximation 0.09 [90% CI 0.08, 0.12], standardized root mean square residual 0.09), and internal consistency was high (McDonald ω 0.89) (Supplementary Tables 1–3).
Diabetes Distress Scale
The Diabetes Distress Scale is a validated 17-item self-report measure that assesses the degree to which individuals perceive diabetes-related experiences as problems over the past month. It was designed to be relevant to both T1D and type 2 diabetes (T2D) (11). The Diabetes Distress Scale uses a 6-point ordinal scale (from 1, not a problem, to 6, a very serious problem [8]), yielding four reliable subscales based on item means, including emotional burden (α = 0.92), physician-related distress (α = 0.89), regimen-related distress (α = 0.91), and interpersonal distress (α = 0.89). The total diabetes distress score is derived as the mean of all 17 items, with excellent internal reliability reported in individuals with T1D (α = 0.92) (22).
Sources of Social Support
The study-created Sources of Social Support questionnaire is composed of four items to reflect whether participants have access to online support for diabetes management. The four items assessed the use of diabetes-related social media (e.g., platforms, hashtags), engagement with diabetes websites, and participation in in-person diabetes groups or events. Each item was answered with a binary response (yes or no). Participants were categorized as having online support if they reported using any diabetes-related social media or websites for their condition. Those who only participated in in-person diabetes groups or reported no online engagement were categorized as not having online support.
Statistical Analyses
Descriptive statistics included the frequency of responses for demographic and study variables. Distributions of continuous variables were examined using graphical methods (histograms and box plots) to identify skewness and outliers. A four-step regression analysis strategy was conducted to examine the associations among DDQoL, diabetes distress, and online support.
Step 1: Univariate Linear Regression Testing
We first conducted univariate linear regressions to explore the total diabetes distress score and its subscale scores (emotional, physician-related, regimen-related, and interpersonal distress) association with DDQoL. Additional univariate regression tested the association between DDQoL and online support.
Step 2: Interaction Effect Testing
We examined whether online support moderated the relationship between diabetes distress and DDQoL. We included online support as an independent variable in each model and compared models with and without the interaction term (diabetes distress ∗ online support) using ANOVA. A Bonferroni correction was applied to control for multiple comparisons, with an α value of 0.01 (23). Interaction effects were probed using simple slopes analysis to interpret significant interactions.
Step 3: Confounder Testing
To identify potential confounders, we included demographic and clinical covariates in the model with the most significantly associated diabetes distress subscale and the significant interaction term. The covariates were supported by prior literature as relevant to diabetes distress and DDQoL in young adults with T1D, including age, gender, education, household income, HbA1c, diabetes device, and recruitment sites. A covariate was retained as a confounder if its inclusion changed the coefficients of the primary independent variables by ≥10% (24). The significant confounder was added to the final model to control for bias in explaining relationships among diabetes distress, online support, and DDQoL.
Step 4: Final Model Specification
The final regression model included the most strongly associated diabetes distress subscale, the online support variable, the significant interaction term, and identified confounders. This model reflected how diabetes distress dimensions, in interaction with online support, contribute to DDQoL.
The analyses were conducted using complete-case data. One participant with missing gender information data was excluded from the final regression model (<1% of the total sample). A significance level of α = 0.05 was used for all primary analyses, and Bonferroni correction was applied to adjust for multiple comparisons. All statistical analyses were performed using R 4.3.3 (25).
Data and Resource Availability
The data sets generated and analyzed during the current study will be deposited in the Inter-University Consortium for Political and Social Research repository by 31 December 2027 (26). Data are available from the corresponding author upon reasonable request.
Results
Descriptive Statistics
The study included 209 young adults with T1D (59.8% female, mean [SD] age 24.3 [3.7] years, 42.6% Hispanic or Latinx ethnicity) enrolled between October 2019 and November 2022. Baseline demographic, clinical, and psychological characteristics are summarized in Table 1. Income data were missing for 24.9% of participants, while all other key variables had complete data, including diabetes distress, online support, and DDQoL.
Table 1.
Sample demographics, clinical characteristics, and psychosocial variables (N = 209)
| Participated in an online support group | ||||
|---|---|---|---|---|
| Variable | All participants | Yes (n = 132) | No (n = 77) | P |
| Recruitment source | ||||
| Clinic | 90 (43) | 43 (33) | 47 (61) | <0.001 |
| Social media | 119 (57) | 89 (67) | 30 (39) | |
| Age (years), mean (SD) | 24.3 (3.7) | 24.9 (3.7) | 23.4 (3.7) | 0.007 |
| Gender | <0.001 | |||
| Male | 75 (36) | 34 (26) | 41 (53) | |
| Female | 125 (60) | 91 (69) | 34 (44) | |
| Other | 8 (3.8) | 6 (4.5) | 2 (2.6) | |
| Not available | 1 (0.5) | 1 (0.8) | 0 | |
| Diagnosis duration (years), mean (SD) | 12.8 (6.0) | 13.0 (6.1) | 12.4 (6.0) | 0.476 |
| Race and ethnicity | 0.099 | |||
| Hispanic or Latinx | 91 (44) | 51 (39) | 40 (52) | |
| White | 99 (47) | 70 (53) | 29 (38) | |
| Other | 19 (9.1) | 11 (8.3) | 8 (10.4) | |
| Educational attainment | 0.001 | |||
| College or graduate degree | 66 (32) | 51 (39) | 15 (20.0) | |
| Less than college | 89 (43) | 57 (43) | 32 (42) | |
| Less than high school | 53 (25) | 24 (18) | 29 (38) | |
| Unknown | 1 (0.5) | 0 (0) | 1 (1.3) | |
| Employment status (nonexclusive) | ||||
| Student | 71 (34) | 41 (31) | 30 (39) | 0.245 |
| Working | 147 (70) | 101 (73) | 46 (60) | 0.010 |
| On leave or unemployed | 27 (13) | 15 (11) | 12 (16) | 0.380 |
| Other | 14 (6.7) | 10 (7.6) | 4 (5.2) | 0.507 |
| Income | 0.237 | |||
| Low (<$24,999) | 40 (19) | 26 (20) | 14 (18) | |
| Middle ($24,999–$74,999) | 72 (34) | 47 (36) | 25 (33) | |
| High (>$74,999) | 45 (22) | 32 (24) | 13 (17) | |
| Unknown | 52 (25) | 27 (21) | 25 (33) | |
| Hypoglycemia | 0.335 | |||
| No | 198 (95) | 123 (93) | 75 (97) | |
| Yes | 11 (5.3) | 9 (6.8) | 2 (2.6) | |
| Ketoacidosis | 0.772 | |||
| No | 196 (94) | 123 (93) | 73 (95) | |
| Yes | 13 (6.2) | 9 (6.8) | 4 (5.2) | |
| Diabetes device | 0.016 | |||
| Insulin pump + CGM | 73 (35) | 53 (40) | 20 (26) | |
| Injection + CGM | 47(22) | 27 (21) | 20 (26) | |
| Insulin pump only | 19 (9.1) | 12 (9.1) | 7 (9.1) | |
| Injection only | 58 (28) | 29 (22) | 29 (38) | |
| Other combination | 12 (5.7) | 11 (8.3) | 1 (1.3) | |
| HbA1c, mean (SD) | 9.16 (1.53) | 9.02 (1.42) | 9.40 (1.69) | 0.105 |
| DDQoL score, mean (SD) | −3.39 (1.88) | −3.62 (1.81) | −2.98 (1.93) | 0.018 |
| Diabetes distress, mean (SD) | ||||
| Total score | 3.00 (1.05) | 3.10 (1.00) | 2.82 (1.14) | 0.085 |
| Emotional | 3.73 (1.43) | 3.86 (1.41) | 3.49 (1.44) | 0.077 |
| Physician-related | 1.74 (1.19) | 1.81 (1.15) | 1.62 (1.26) | 0.260 |
| Regimen-related | 3.59 (1.28) | 3.67 (1.20) | 3.45 (1.40) | 0.265 |
| Interpersonal | 2.45 (1.38) | 2.56 (2.56) | 2.25 (2.25) | 0.115 |
Data are n (%) unless otherwise indicated. CGM, continuous glucose monitoring.
Of the sample, 132 patients (63.2%) reported using at least one online platform to obtain diabetes-related social support, most commonly Instagram (n = 87) and Facebook (n = 66). Frequently mentioned online communities included Breakthrough T1D (formerly JDRF) (n = 9) and Beyond Type 1 (n = 8), with common hashtags such as #T1D, #Insulin4All, and #T1DLooksLikeMe. As shown in Table 1, online support users differed significantly from nonusers in age, gender, education, and diabetes device use (all P < 0.05). Although race and ethnicity did not reach statistical significance (P = 0.10), Hispanic or Latinx patients were less represented among online support users (39%) compared with nonusers (52%).
Inferential Statistics
Step 1: Univariate Linear Regression Testing
The total diabetes distress score was significantly associated with lower DDQoL (β −0.51 [95% CI −1.12, −0.70], P < 0.001). Among the distress subtypes, emotional distress showed the strongest negative association with DDQoL (β −0.60 [95% CI −0.93, −0.64], P < 0.001), followed by regimen distress, interpersonal distress, and physician-related distress (Table 2). Individuals who reported using online support had significantly lower DDQoL scores compared with those who did not (β −0.17 [95% CI −1.17, −0.12], P = 0.016).
Table 2.
Univariate linear regression models (DDQoL as the dependent variable)
| Independent variable | β | 95% CI | F statistic | P |
|---|---|---|---|---|
| Diabetes distress (total) | −0.51 | −1.12, −0.70 | 73.37 | <0.001 |
| Emotional distress | −0.60 | −0.93, −0.64 | 115.8 | <0.001 |
| Physician distress | −0.22 | −0.56, −0.13 | 10.36 | 0.001 |
| Regimen distress | −0.37 | −0.73, −0.35 | 32.27 | <0.001 |
| Interpersonal distress | −0.37 | −0.67, −0.32 | 31.86 | <0.001 |
| Online support (yes) | −0.17 | −1.17, −0.12 | 5.92 | 0.016 |
Step 2: Interaction Effect Testing
Among the models tested (Supplementary Table 4), a significant interaction was observed between interpersonal distress and online support (β 0.42 [95% CI 0.12, 0.83], P = 0.010), with the overall model comparison yielding a significant ANOVA test result (ΔF 4.72, P = 0.010). Simple slope analysis (Supplementary Table 5 and Fig. 1) indicated that interpersonal distress was negatively associated with DDQoL in both groups, but the association was weaker among patients with online support (β −0.31 [95% CI −0.52, −0.09], P = 0.005) compared with those without online support (β −0.77 [95% CI −1.05, −0.49], P < 0.001), suggesting a buffering effect. Although interaction plots for total diabetes distress and other subtypes visually suggest slight differences in slope or crossover patterns, the corresponding interaction terms were not statistically significant after Bonferroni correction. Therefore, these visual differences should not be interpreted as evidence of moderation.
Figure 1.
Plots of the interaction between Diabetes Distress Scale scores and online support. A: Interaction of total score with online support. B: Interaction of emotional distress with online support. C: Interaction of physician-related distress with online support. D: Interaction of regimen distress with online support. E: Interaction of interpersonal distress with online support. Shading indicates the 95% CI.
Step 3: Confounder Testing
The percentage changes after adding each demographic covariate to the model are shown in Supplementary Table 6. Gender substantially influenced interpersonal distress (20%), online support (21%), and interaction term (13%) coefficients. Education, income, and diabetes device each caused a ≥10% change in the interpersonal distress coefficient (8–16%), while education also approached the threshold for online support (9.9%). Other covariates, such as age, HbA1c, and recruitment sites, resulted in minimal changes (<10%).
Step 4: Final Model Specification
A final multiple linear regression model was constructed (Table 3) to examine the associations among emotional distress, interpersonal distress, online support, and their interaction in relation to DDQoL, adjusting for demographic and clinical characteristics. Emotional distress was significantly associated with lower DDQoL (β −0.55 [95% CI −0.73, −0.36], P < 0.001). Interpersonal distress was not significantly associated with DDQoL (β −0.16 [95% CI −0.42, 0.10], P = 0.226). Online support was also associated with lower DDQoL (β −0.16 [95% CI −0.60, −0.08], P = 0.001). A significant interaction between interpersonal distress and online support (β 0.23 [95% CI 0.03, 0.54], P = 0.026) indicated that online support buffered the negative effect of interpersonal distress on DDQoL. Among the covariates, identifying as female was positively associated with better DDQoL (β 0.14 [95% CI 0.03, 0.56], P = 0.031). Educational attainment, income, and diabetes device use were not significantly associated with DDQoL after adjustment.
Table 3.
Final multiple linear regression model (DDQoL as dependent variable, adjusting for demographic and clinical characteristics)
| Independent variable | β | 95% CI | P |
|---|---|---|---|
| Emotional distress | −0.55 | −0.73, −0.36 | <0.001 |
| Interpersonal distress | −0.16 | −0.42, 0.10 | 0.226 |
| Online support | −0.16 | −0.60, −0.08 | 0.001 |
| Interpersonal diabetes distress ∗ online support | 0.23 | 0.03, 0.54 | 0.026 |
| Gender (ref. male) | |||
| Female | 0.14 | 0.03, 0.56 | 0.031 |
| Other | 0.06 | −0.30, 0.97 | 0.298 |
| Education (ref. college graduate) | |||
| Less than college | −0.07 | −0.43, 0.15 | 0.338 |
| Less than high school | −0.08 | −0.52, 0.18 | 0.339 |
| Income (ref. high) | |||
| Low | 0.02 | −0.47, 0.36 | 0.783 |
| Middle | 0.06 | −0.22, 0.48 | 0.465 |
| Unknown | −0.06 | −0.52, 0.25 | 0.490 |
| Diabetes device (ref. injection + CGM) | |||
| Injection only | −0.08 | −0.52, 0.14 | 0.437 |
| Insulin pump only | −0.01 | −0.50, 0.40 | 0.991 |
| Other combination | −0.08 | −0.87, 0.19 | 0.306 |
| Pump + CGM | 0.02 | −0.29, 0.39 | 0.778 |
R2 = 0.34; adjusted R2 = 0.29; F(15,192) = 6.58; P < 0.001. CGM, continuous glucose monitoring; ref., reference group.
Conclusions
Emotional distress emerged as the strongest contributor to lower DDQoL. While online support was independently associated with lower DDQoL, it buffered the negative impact of interpersonal distress on DDQoL, indicating a potential protective role that may vary by gender.
Emotional distress played the most prominent role in reducing DDQoL, reinforcing prior evidence that emotional aspects of diabetes distress strongly affect QoL in young adults (10). This highlights the need for interventions addressing emotional responses to diabetes management, especially during the transition to independent care among young adults.
In both unadjusted and adjusted models, obtaining online support was associated with lower DDQoL. Since online support was not randomly assigned in our study, this association may reflect self-selection in real-life settings in which individuals experiencing poorer DDQoL may seek online communities as a coping strategy, consistent with prior work linking reduced well-being to increased help-seeking behaviors (27). A similar pattern has been reported in a cross-sectional study among individuals with T2D, which also observed a negative association between social support and health-related QoL (HRQoL) (28). Our findings extend this evidence by showing how such coping manifests in digital spaces. Online support in chronic conditions often requires active searching, digital literacy, and sustained engagement (29), which are behaviors especially relevant for young adults with T1D (18). Alternatively, engagement with online spaces may sometimes expose users to emotionally charged or stigmatizing content (30), which could undermine well-being. Thus, while online support holds potential, its effects depend on how individuals engage with the communities and integrate them into daily life, suggesting the need for future research that more clearly describes patterns of engagement and real-world use.
Online support buffered the negative impact of interpersonal distress on DDQoL (β 0.23, P = 0.026). Individuals using online support reported higher DDQoL than nonusers when interpersonal distress was moderate to high (Diabetes Distress Scale score >3.5). This supports the buffering hypothesis of social support and suggests that online communities may compensate for strained offline relationships. To our knowledge, the current study is the first to demonstrate a protective role of online support in the context of diabetes distress among young adults with T1D. Notably, several studies examining the interactive role of social support in the relationship between diabetes distress and HRQoL have been conducted among individuals with T2D. Onu et al. (28) reported that social support moderates the associations between overall diabetes distress and HRQoL, as well as between most distress subscales and HRQoL, with the exception of emotional distress. Similarly, another study in adults with T2D found that general social support buffered the association between physician-related distress and HRQoL but did not attenuate the impact of interpersonal distress (31). In contrast, the current study identified a buffering effect specifically for interpersonal distress. These distinctive findings suggest that the source of support may matter: Diabetes-related online support may be particularly well suited to addressing interpersonal distress, whereas broader support resources may be more relevant for navigating health care–related stressors. This contrast also points to potentially different support needs across populations. Young adults with T1D who are managing developmental transitions alongside intensive self-management may especially benefit from peer-involved, identity-affirming communities online (32).
Females experienced marginally better DDQoL than males (β = 0.14, P = 0.031), contrasting prior studies in which females typically reported poorer QoL (33). Contextual factors may explain this discrepancy. In line with gendered help-seeking norms, where females tend to seek emotional connection and males prefer task-oriented information exchanges (34), females’ greater engagement with online support may help mitigate interpersonal distress more effectively. This pattern suggests that observed gender differences in QoL among adults with T1D may reflect differences in coping strategies and social engagement rather than gender itself, highlighting the importance of examining how online community engagement differentially supports males and females in managing diabetes.
To our knowledge, this is among the first cross-sectional studies to quantitatively examine online support and its association with an overall well-being outcome in diverse young adults with T1D, as prior evidence has largely relied on qualitative approaches with predominantly White samples (35,36). Our sample, with 42.6% identifying as Hispanic or Latinx ethnicity, offers a more diverse representation than is typical in this literature. Our sample also differs notably from the general population with T1D, in which White individuals are disproportionately diagnosed (37). Although race and ethnicity was not significantly associated with online support use in our sample (P = 0.10), the directional patterns, with Hispanic or Latinx participants less represented among online support users, point to an area for further investigation. Given the racial and ethnic inequities in young adults with T1D, including lower technology uptake and higher HbA1c among Hispanic or Latinx individuals (38), further research should examine how online support needs and engagement patterns may differ across racial and ethnic groups to ensure that online support strategies are equitably designed.
This study had several limitations. First, online support was measured dichotomously (yes or no), rather than on a continuous or multidimensional scale. As Cohen and Wills (13) noted, both structural indicators of social support (presence or absence) and global functional (perceived availability) indicators may overlook nuanced processes involved in buffering effects. Our binary measure limited the assessment of the amount, intensity, and quality of online support, underscoring the need for more validated measures in future research. Second, the cross-sectional design precludes causal inference. Although we explored potential interpretations, including both direct and reverse-causation pathways, these relationships cannot be confirmed without longitudinal data. Third, the relatively modest sample size (N = 209) and unequal group distribution (132 with online support vs. 77 without) may have limited statistical power to detect interaction effects. Moderation analyses generally require larger, more balanced samples than main-effects models, and insufficient power increases the risk of type II error. Future studies with larger and more balanced samples are needed to more definitively determine the buffering effect of online support. Finally, data collection occurred during the coronavirus disease 2019 pandemic, a period marked by increased reliance on digital use (39) and elevated diabetes distress (40), which may limit the generalizability of these findings to nonpandemic contexts.
In conclusion, we revealed that emotional distress was the strongest contributor to reduced DDQoL among diabetes distress subtypes in young adults with T1D. While online support was independently associated with lower DDQoL, it buffered the negative impact of interpersonal distress, suggesting its potential role as a compensatory resource under specific psychosocial conditions. The observed gender effect in the final model highlights the importance of accounting for psychosocial context in QoL research. These findings have significant translational implications, underscoring the need for psychosocial interventions that specifically target emotional distress and highlighting the potential of leveraging online support platforms as part of comprehensive care strategies. Future research should explore how to optimize the use of online communities to promote diabetes-related QoL in young adults with chronic conditions.
This article contains supplementary material online at https://doi.org/10.2337/figshare.32888456.
Acknowledgments
Article Information
Acknowledgments
The authors acknowledge the use of the ADDQoL questionnaire, developed by Prof. Clare Bradley of and licensed by Health Psychology Research Ltd. The manuscript was reviewed by Health Psychology Research Ltd. (https://healthpsychologyresearch.com/) prior to submission.
During the preparation of this work, the authors used Grammarly for language editing; no artificial intelligence tools were used for content generation or data analysis. Following the use of this tool, the authors formally reviewed the content for its accuracy and edited it as necessary. The authors take full responsibility for all the content of this article.
Duality of Interest
No potential conflicts of interest relevant to this article were reported.
Author Contributions
Y.M. wrote the first draft of the manuscript. Y.M. and E.A.P. were involved in the conceptualization, design, analysis, and interpretation of the results. P.-J.L. curated the data set. G.G., J.B., and E.A.P. were involved in the study conduct and data collection. All authors edited, reviewed, and approved the final version of the manuscript. E.A.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.
Prior Presentation
Parts of this study were presented in poster form at the 46th Annual Meeting and Scientific Sessions of the Society of Behavioural Medicine, San Francisco, CA, 26–29 March 2025, and the 85th Scientific Sessions of the American Diabetes Association, Chicago, IL, 20–23 June 2025.
Funding Statement
This research was supported by the National Institute of Diabetes and Digestive and Kidney Diseases at the National Institutes of Health (grant R01DK116719). The Southern California Clinical and Translational Science Institute supported the use of REDCap in this study (National Institutes of Health National Center for Advancing Translational Sciences grants UL1TR001855 and UL1TR000130).
Supporting information
References
- 1. Livingstone SJ, Levin D, Looker HC, et al.; Scottish Renal Registry . Estimated life expectancy in a Scottish cohort with type 1 diabetes, 2008–2010. JAMA 2015;313:37–44 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Speight J, Holmes-Truscott E, Hendrieckx C, Skovlund S, Cooke D.. Assessing the impact of diabetes on quality of life: what have the past 25 years taught us? Diabet Med 2020;37:483–492 [DOI] [PubMed] [Google Scholar]
- 3. Rasmussen B, Ward G, Jenkins A, King SJ, Dunning T.. Young adults’ management of type 1 diabetes during life transitions. J Clin Nurs 2011;20:1981–1992 [DOI] [PubMed] [Google Scholar]
- 4. Shishkova YA, Motovilin OG, Surkova EV, Divisenko SI, Mayorov AY.. Quality of life in young patients with type 1 diabetes mellitus. Diabetes Mellit 2010;13:43–47 [Google Scholar]
- 5. Vigen CLP, Carandang K, Blanchard J, et al. Psychosocial and behavioral correlates of A1C and quality of life among young adults with diabetes. Diabetes Educ 2018;44:489–500 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Johansen CB, Pouwer F, Beck-Nielsen H, Rothmann MJ.. Positive facilitators of diabetes management in emerging adults with type 1 diabetes—a qualitative analysis of blogs. Endocrinol Diabetes Metab 2020;3:e00161. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7. Skedgell KK, Cao VT, Gallagher KA, Anderson BJ, Hilliard ME.. Defining features of diabetes resilience in emerging adults with type 1 diabetes. Pediatr Diabetes 2021;22:345–353 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Polonsky WH, Fisher L, Earles J, et al. Assessing psychosocial distress in diabetes: development of the diabetes distress scale. Diabetes Care 2005;28:626–631 [DOI] [PubMed] [Google Scholar]
- 9. Leung JMWS, Al-Yahyawi NY, Choi HS, et al. Diabetes distress and transition readiness in youths with type 1 diabetes transitioning from pediatric to adult care. Pediatr Diabetes 2023;2023:5580180. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Sattoe J, Peeters M, Bronner M, van Staa A.. Transfer in care and diabetes distress in young adults with type 1 diabetes mellitus. BMJ Open Diabetes Res Care 2021;9:e002603. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Schmitt A, Reimer A, Kulzer B, Haak T, Ehrmann D, Hermanns N.. How to assess diabetes distress: comparison of the Problem Areas in Diabetes Scale (PAID) and the Diabetes Distress Scale (DDS). Diabet Med 2016;33:835–843 [DOI] [PubMed] [Google Scholar]
- 12. Sturt J, Dennick K, Due-Christensen M, McCarthy K.. The detection and management of diabetes distress in people with type 1 diabetes. Curr Diab Rep 2015;15:101. [DOI] [PubMed] [Google Scholar]
- 13. Cohen S, Wills TA.. Stress, social support, and the buffering hypothesis. Psychol Bull 1985;98:310–357 [PubMed] [Google Scholar]
- 14. Boumans NPG, Landeweerd JA.. The role of social support and coping behaviour in nursing work: main or buffering effect? Work Stress 1992;6:191–202 [Google Scholar]
- 15. Davidson P, LaManna J, Davis J, et al. The effects of diabetes self-management education on quality of life for persons with type 1 diabetes: a systematic review of randomized controlled trials. Sci Diabetes Self Manag Care 2022;48:111–135 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Trepte S, Dienlin T, Reinecke L.. Influence of social support received in online and offline contexts on satisfaction with social support and satisfaction with life: a longitudinal study. Media Psychol 2015;18:74–105 [Google Scholar]
- 17. Maxwell TK, Branka LA, Adam ATIF, Weitzman ER, Laffel LM.. Survey of social media use in young adults (YA) with type 1 diabetes (T1D). Diabetes 2024;73(Suppl. 1):1853-LB [Google Scholar]
- 18. Elnaggar A, Ta Park V, Lee SJ, Bender M, Siegmund LA, Park LG.. Patients’ use of social media for diabetes self-care: systematic review. J Med Internet Res 2020;22:e14209. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Pyatak EA, Lee P-J, Nnoli ND, et al. Telehealth occupational therapy improves psychosocial well-being but not glycemia among young adults with type 1 diabetes: the Resilient, Empowered, Active Living-Telehealth (REAL-T) randomized controlled trial. Diabetes Res Clin Pract 2025;221:112005. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Pyatak EA, Ali A, Khurana AR, et al. Research design and baseline participant characteristics of the Resilient, Empowered, Active Living with Diabetes - Telehealth (REAL-T) study: a randomized controlled trial for young adults with type 1 diabetes. Contemp Clin Trials 2023;135:107386. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Bradley C, Todd C, Gorton T, Symonds E, Martin A, Plowright R.. The development of an individualized questionnaire measure of perceived impact of diabetes on quality of life: the ADDQoL. Qual Life Res 1999;8:79–91 [DOI] [PubMed] [Google Scholar]
- 22. Joensen LE, Tapager I, Willaing I.. Diabetes distress in type 1 diabetes—a new measurement fit for purpose. Diabet Med 2013;30:1132–1139 [DOI] [PubMed] [Google Scholar]
- 23. Armstrong RA. When to use the Bonferroni correction. Ophthalmic Physiol Opt 2014;34:502–508 [DOI] [PubMed] [Google Scholar]
- 24. Talbot D, Diop A, Lavigne-Robichaud M, Brisson C.. The change in estimate method for selecting confounders: a simulation study. Stat Methods Med Res 2021;30:2032–2044 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25. R Core Team . R: A Language and Environment for Statistical Computing. Vienna, R Foundation for Statistical Computing, 2024 [Google Scholar]
- 26. Pyatak EA. REAL-T study: RCT of diabetes management intervention, 2019–2025. OpenICPSR. Accessed 29 April 2026. Available from https://www.openicpsr.org/openicpsr/project/240494/version/V1/view
- 27. Göz F, Karaoz S, Goz M, Ekiz S, Cetin I.. Effects of the diabetic patients’ perceived social support on their quality-of-life. J Clin Nurs 2007;16:1353–1360 [DOI] [PubMed] [Google Scholar]
- 28. Onu DU, Ifeagwazi CM, Prince OA.. Social support buffers the impacts of diabetes distress on health-related quality of life among type 2 diabetic patients. J Health Psychol 2022;27:2305–2317 [DOI] [PubMed] [Google Scholar]
- 29. Verweel L, Newman A, Michaelchuk W, Packham T, Goldstein R, Brooks D.. The effect of digital interventions on related health literacy and skills for individuals living with chronic diseases: a systematic review and meta-analysis. Int J Med Inform 2023;177:105114. [DOI] [PubMed] [Google Scholar]
- 30. Malik FS, Panlasigui N, Gritton J, Gill H, Yi-Frazier JP, Moreno MA.. Adolescent perspectives on the use of social media to support type 1 diabetes management: focus group study. J Med Internet Res 2019;21:e12149. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Mahato S, Wen S, Misra R.. The moderating role of social support in the association between diabetes distress and health-related quality of life in rural adults with type 2 diabetes. MRAJ 2025;13:e6918 [Google Scholar]
- 32. Berry E, Cleal B, Morrissey E, et al. The role of social media on psychological wellbeing from the perspectives of young people with type 1 diabetes and their caregivers: a qualitative study. Int J Adolesc Youth 2024;29:2298083 [Google Scholar]
- 33. Castellano-Guerrero AM, Guerrero R, Ruiz-Aranda D, et al. Gender differences in quality of life in adults with long-standing type 1 diabetes mellitus. Diabetol Metab Syndr 2020;12:64. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34. Reddin JA, Sonn CC.. Masculinity, social support, and sense of community: the men’s group experience in Western Australia. J Mens Stud 2003;11:207–223 [Google Scholar]
- 35. Maxwell T, Branka L, Asher N, Commissariat P, Laffel L.. Young adults with type 1 diabetes and their perspectives on diabetes-related social media: qualitative study. JMIR Diabetes 2025;10:e69243. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Montali L, Zulato E, Cornara M, Ausili D, Luciani M.. Barriers and facilitators of type 1 diabetes self-care in adolescents and young adults. J Pediatr Nurs 2022;62:136–143 [DOI] [PubMed] [Google Scholar]
- 37. Fang M, Wang D, Selvin E.. Prevalence of type 1 diabetes among US children and adults by age, sex, race, and ethnicity. JAMA 2024;331:1411–1413 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 38. Agarwal S, Kanapka LG, Raymond JK, et al. Racial-ethnic inequity in young adults with type 1 diabetes. J Clin Endocrinol Metab 2020;105:e2960–e2969 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 39. Sujan MSH, Tasnim R, Islam MS, et al. COVID-19-specific diabetes worries amongst diabetic patients: the role of social support and other co-variates. Prim Care Diabetes 2021;15:778–785 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 40. García-Lara RA, Gómez-Urquiza JL, Membrive-Jiménez MJ, et al. Anxiety, distress and stress among patients with diabetes during COVID-19 pandemic: a systematic review and meta-analysis. J Pers Med 2022;12:1412. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.