Development of the Diabetes Index for Social Determinants of Health (DISDOH)

WHAT IS ALREADY KNOWN ON THIS TOPIC While social determinants of health (SDoH) are key risk factors in impacting diabetes outcomes, there are currently no diabetes-related measures that address SDoH-associated vulnerability impacting the ability of individuals to manage their diabetes, nor are any aligned with the American Diabetes Association’s scientifically identified SDoH domains affecting those living with diabetes.WHAT THIS STUDY ADDS This study provides the first validated measure that can be used to measure SDoH factors impacting the ability of individuals to manage their diabetes: the Diabetes Index for Social Determinants of Health (DISDOH).HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY Healthcare providers can integrate DISDOH into routine diabetes care to systematically identify and address social risk factors, personalize treatment plans, and enhance patient-centered interventions.Introduction According to the Centers for Disease Control and Prevention, an estimated 38.4 million Americans (11.6% of the US population) had diabetes, and an estimated 97.6 million Americans had prediabetes as of 2021. 1 Among those living with diabetes, the highest prevalences are historically among racial/ethnic minority populations, in individuals with lower educational levels, lower income levels, and who dwell in non-metropolitan or rural areas. These factors are among the social and environmental factors, known as social determinants of health (SDoH), that have been observed to be key contributors to health disparities and collectively accounting for 50%–60% of health outcomes.2 The American Diabetes Association (ADA) convened a writing committee to conduct a review of the state of the science examining the relationship of SDoH with diabetes risk and the corresponding outcomes. The scientific review established five categories of SDoH observed to have ‘a sufficient body of literature to demonstrate influence of the determinant on diabetes’.3 The five SDoH factors included socioeconomic status, neighborhood and physical environment, food environment, healthcare, and social context.A 2023 article by Jiang et al calls for the modernization of diabetes quality measures to help improve the health of people with diabetes.4 Currently, there are few measures that assess SDoH-associated vulnerability focused on the diabetes population. As a result, there are no validated or reliable measures available to clinicians, providers, community-based programmes, patients, and other stakeholders that address SDoH factors impeding optimal diabetes care. Additionally, the wide variety of general SDoH measures available for use do not focus on the critical factors identified by the ADA as influencing diabetes outcomes. Often, many of the factors in these general measures cannot feasibly be modified by individuals’ diabetes care teams or facilitators in diabetes education/support programmes. Thus, a diabetes SDoH measure should encompass the key SDoH domains while focusing on factors that healthcare professionals and diabetes programmes can help modify and improve by providing resources or interventions. Finally, many general SDoH measures require individuals to spend an extended amount of time to answer a long list of questions. These measures may not be feasible for clinical settings or in community-based programmes.The goal of this project was to develop an instrument to measure SDoH factors impacting the ability of individuals to manage diabetes. The Diabetes Index for Social Determinants of Health (DISDOH) aims to be implemented across clinical, education, and support settings as a tool to understand SDoH burden and highlight areas where resources can be provided to reduce that burden.Research design and methods Basic design The development of the DISDOH consisted of three different phases: Phase I: DISDOH Item Development and Phase II: Instrument Development began in early 2023; Phase III: Instrument Evaluation began in the Fall of 2024. See online supplemental material 1 for a detailed Gantt chart of the tool’s development and validation. Participants in all phases of the project gave informed consent.SP110.1136/bmjdrc-2025-005113.supp1Supplementary data Phase I: DISDOH item development Best practice guidelines for developing and validating scales for health, social, and behavioral research were followed. 5–7 Identification of domain and item generation Based on the scientific guidance from the ADA, the selected DISDOH items represented each of five SDoH categories: socioeconomic status, neighborhood and physical environment, food environment, healthcare, and social context. 3 A literature review was conducted to evaluate current SDoH measures. Those measures were then reviewed to determine if they included questions addressing the five selected SDoH categories. Initial items included in the DISDOH were either (1) Already included in the Health Extension for Diabetes (HED) programme registration questionnaire or (2) Modified from SDoH measures identified based on literature review. HED is an ADA-recognized diabetes education and support programme and licensed by Clemson University (see Intervention). See online supplemental material 2 for origination of each of the initial DISDOH items.SP210.1136/bmjdrc-2025-005113.supp2Supplementary data Content validity: Item revision through internal HED team input Initial items were constructed into a survey. Items considered for inclusion in the survey were reviewed via an iterative process during a weekly 1 hour HED research team meeting. Over the course of a 1 month period, iterations of the survey were reviewed based on literature review and comparison to existing instruments to assess DISDOH items and ensure each item accurately reflected the SDoH domain it was intended to measure among individuals living with diabetes. Consensus among authors was reached via discussion and review of all items considered for each of the five domains. Adjustments were made to items to improve clarity and readability until a consensus was reached with at least 80% of the internal study team.Number of participants The internal team for item development included diabetes research experts, certified diabetes care and education specialists, and HED programme facilitators. A total of 13 individuals were a part of the internal team.Data analysis An iterative qualitative process was used to assess DISDOH items and ensure each item accurately reflected the SDoH domain it was intended to measure until consensus was reached with at least 80% of the internal study team.Phase II: Instrument development Intervention: HED HED is a structured, 4-month programme designed to support adults managing type 1 or type 2 diabetes through education and support. It consists of eight biweekly sessions covering the Association of Diabetes Care and Education Specialists’ seven self-care behaviors, with additional support provided between sessions. The programme is available in both English and Spanish and is facilitated by trained Rural Health and Nutrition Extension Agents who undergo training in diabetes education and public health programme delivery. HED is a collaborative effort between Clemson University and Prisma Health, ensuring both academic and clinical expertise are integrated throughout the programme. Programme effectiveness is well-chronicled and in forthcoming research. 8 9 Pretesting questions: Evaluation of content validity through HED participants, programme facilitators and researchers’ feedback Initial DISDOH items were implemented into the ongoing HED collection procedures in the summer of 2023. All individuals who consented to participate in the HED programme answered the DISDOH items during programme registration. HED programme facilitators were asked to provide any input they received from participants regarding the DISDOH items in addition to their own feedback. Researchers from the HED programme were also asked to provide input on the usability of the tool. This feedback was collected in an ongoing fashion.Item revision: Implementation of HED participants, programme facilitators and researcher feedback Feedback from programme facilitators was used to iteratively adjust DISDOH items over a 1-year period. Facilitators reported on participants’ reactions to specific questions, including difficulties with comprehension, discomfort with sensitive content, or the need for additional context to respond accurately. Modifications to items were made until no further input was received from HED participants indicating need for item improvement. Input from programme facilitators also informed revisions of the DISDOH items to be more diabetes-related (ie, I have access to the healthcare services I need to effectively manage my diabetes). Facilitators reported this would be helpful in understanding the needs of their participants rather than having more general SDoH questions.For scoring of DISDOH domains, early phases of scoring were difficult. Researcher feedback about the scoring process reported that difficulty was due to the various response types in the initial questions (ie, yes/no, various Likert scales, and check all that apply response option choices). Items for the final DISDOH measure were revised so that item responses for the neighborhood and physical environment, food environment, healthcare, and social context domains are on a 5-point Likert Scale (ie, 1—strongly disagree, 2—disagree, 3—neutral, 4—agree, 5—strongly agree). Through a combination of HED facilitator feedback and statistical findings, the final DISDOH was created containing 16 items.Face validity: Item validation through an expert stakeholder panel Experts from a variety of fields (eg, public health, health services research, medicine, and survey methods) were purposively sampled and received a recruitment email describing the study. Using a Qualtrics survey (Qualtrics, Provo, Utah, USA), experts were asked to evaluate the questions in an item-specific and global fashion. The overall evaluation assessed the experts’ appraisal of tool relevance, comprehensiveness, and comprehensibility using an adapted approach similar to the COSMIN (COnsensus-based Standards for the selection of health Measurement INstruments) methodology. 10 Informed by the expert review, items were slightly rearranged (eg, the insurance item moved to the beginning with other baseline items), edited, and adapted to a Lexile approximating a middle school reading level for accessibility.Number of participants Individuals with self-reported diabetes were recruited through participation in HED. A total of 440 HEDs were recruited and completed the initial DISDOH items. Feedback about items was provided by 13 HED facilitators. For face validity, a total of nine stakeholders responded to the request for feedback for the face validity survey.Data analysis Principal component analysis At the end of the 1-year data collection period (summer 2024), principal component analysis (PCA) was used to assess the initial 16 items of DISDOH. Suitability of data for PCA was assessed using the Kaiser–Meyer–Olkin (KMO) value and Bartlett’s test of sphericity prior to the PCA analysis. Data were assessed using IBM SPSS V.28.Face validity Face validity was assessed by distributing the Qualtrics questionnaire and DISDOH to experts for feedback on question content, tool structure, accessibility, and overall usability. The internal team reviewed and addressed the feedback through consensus. Descriptive statistics were used to analyze expert ratings on relevance, comprehensiveness, and comprehensibility.Phase III: Instrument evaluation In phase III, reliability and validity of the finalized 16-item DISDOH measure was assessed. Individuals with a self-reported history of diabetes were recruited using a crowdsourcing platform, Prolific. A priori power calculations were conducted for this phase’s procedures. For test-retest reliability, a baseline sample of at least 184 participants was required, accounting for a 20% attrition rate over approximately 2 weeks, resulting in a final sample of 148 to maintain strong statistical power. Similarly, construct validation components necessitated a minimum sample size of 200 participants.Number of participants A total of 303 individuals from Prolific were screened for being diagnosed with type 1 or type 2 diabetes and 215 individuals with diabetes were included in the study.Prolific Prolific is an online crowdsourcing platform designed for recruiting high-quality research participants. 11 The platform ensures that participants are real, engaged, and provide reliable data using rigorous screening and quality control measures. Prolific connects researchers with a diverse pool of participants from across the USA.Statistical analysis A maximum likelihood confirmatory factor analysis (CFA) was conducted to determine how well the items represent the constructs. A goodness-of-fit test was used to examine if the CFA model fits the data well with the χ 2/df ratio ≤3 rule.12 Reliability was assessed using Cronbach’s α reliability estimate for total score. Test-retest reliability was assessed using intraclass correlation coefficient (ICC). Internal consistency estimates were calculated for each domain. Pearson correlations were used to assess convergent and discriminant validity. Analyses completed were conducted in IBM SPSS Statistics V.28.Results Phase I: DISDOH item development Table 1 depicts the face validation expert feedback and results.Table 1Face validation item-specific and global DISDOH feedbackItem-specific feedback Concept Item before Selected expert feedback Item after Health context Do you have health insurance? It’s odd to include this in the middle of the Likert items.It might be better to provide a list of types of health insurance since the three types all are ‘equal’.May add something like ‘including Medicare, Medicaid, or Tricare insurance’. Many people with Medicaid don’t realize it is insurance. Moved to Q1 instead of Q12.*Socioeconomic status What is your current employment status? Is this check all that apply or what best describes? eg, ‘student’ and ‘part-time’ could easily overlap.May separate part-time from temporary/seasonal. Added qualifying statement ‘(Select the option that describes you best)’.*Socioeconomic status What is the highest level of education you have completed? Would personalize the question more. Also, ‘some college’ and ‘associate degree’ could seem similar.Consider master’s degree or higher. Changed the option to master’s degree or higher.*Neighborhood and physical environment In the past 12 Months, I had a stable and secure place to sleep every night. What does stable and secure mean?What do you mean by ‘secure’? Safe? In the past 12 months, I had a safe and stable place to sleep every night.Food environment It is easy for me to access fresh fruits and vegetables when I need it. Can take out ‘produce’. It is easy for me to access fresh fruits and vegetables when I need it.Health context I can easily afford the medications I need. Would take out ‘easily’ since the scale indicates the degree they agree. I can afford the medications I need.Social context I regularly get together with family, friends, or neighbors each week. Take out ‘regularly’ since the scale indicates level. N/A Feedback overall Category Item Mean of experts(Likert Scale 1—Yes, 2—Somewhat, 3—No)RelevanceAre the questions addressing current SDoH aspects for diabetes management? 1.40 Do the included items reflect the priorities of patients with diabetes in managing their condition? 1.20 Are the included items relevant for the context of diabetes care? 1.00 Are the response options appropriate for each question? 1.60 ComprehensivenessDoes the questionnaire adequately capture many of the SDoH listed in the scientific review (based on table 2 of Felicia Hill-Briggs et al 2021)?3 1.00 Is the coverage of selected determinants sufficient for assessing the needs of patients with diabetes? 1.80 ComprehensibilityAre the terms and concepts used in the questionnaire easy to understand for individuals with varying levels of health literacy? 1.60 Are the instructions for filling out the questionnaire clear and concise? 1.40 Is the language in the questions accessible for diverse populations, including those with limited proficiency in English? 2.00 *: Denotes a stylistic action, this was not a substantive change to the question.DISDOH, Diabetes Index for Social Determinants of Health; Q1, Question 1; Q12, Question 12; SDoH, social determinants of health.Table 2Initial DISDOH Questionnaire (phase II) participant demographics and biometric informationVariablesParticipantsn=440Demographics, N (%)Biological sex  Female334 (75.9)  Male106 (24.1)Race  African American/black151 (34.3)  White231 (52.5)  Other55 (12.5)  Prefer not to answer3 (0.7)Ethnicity  Hispanic/Latino57 (13.0)  Non-Hispanic/Latino366 (83.2)  Other5 (1.1)  Prefer not to answer12 (2.7)Diabetes type  Type 124 (5.5)  Type 2383 (87.0)  Not diagnosed presently18 (4.1)  Don’t know15 (3.4)Biometric information, mean (SD)  Weight, lb 207.3 (±52.6)  BMI, lb/in2 34.3 (±8.3)  A1C, %7.4 (±2.3)BMI, body mass index; DISDOH, Diabetes Index for Social Determinants of Health.Phase II: Instrument development Participants Table 2 provides a description of the demographic and biometric information of the participants who responded to the initial DISDOH questionnaire. A majority of the participants were non-Hispanic (83.2%) and female (75.9%), and indicated they had type 2 diabetes (87.0%). Approximately half the participants were white (52.4%) and approximately a third of the participants were African American/black (34.3%). The average participant weight was 207.3 pounds (±52.6), average body mass index (BMI) was 34.3 (±8.3) and average A1C was 7.4 (±2.3).Principal component analysis The KMO value was observed to be 0.680, surpassing the recommended value of 0.6 and indicating the data were suitable for PCA. 13 A significant Bartlett’s test of sphericity (t( X2 (136) = 1815.32, p<0.001) was also observed, which supports the factorability of the correlation matrix.14 Thus, data were determined to be suitable for PCA. Initial PCA results revealed the presence of six eigenvectors with eigenvalues exceeding 1, explaining a total of 61% variance. Inspection of the scree plot suggested five eigenvectors. A Promax rotation was applied to the PCA as SDoH factors are expected to be aligned with one another.15 The question, ‘Do you have internet access at home?’ was removed due to its loading factor being below 0.4.A final PCA using a Promax rotation was used on the 15-item DISDOH and a 5-factor solution was revealed, explaining 58% of the variance. Component 1 contained four items and explained 22.2% variance representing Neighborhood and Physical Environment. Component 2 contained four items and explained 11.9% variance representing Health Context. Component 3 contained two items and explained 10.6% variance representing Food Environment. Component 4 contained three items and explained 8.2% variance representing Socioeconomic Status (SES). Component 5 contained two items and explained 7.1% variance representing Social Context.Phase III: Instrument evaluation Participants Table 3 provides a description of the Prolific participants’ demographic and biometric information. A majority of participants were non-Hispanic (92.6%) and white (80.5%), and indicated they had type 2 diabetes (78.1%). About half of the participants were female (51.6%). Participants’ average weight was 198.7 (±64.9) kg, average BMI was 32.0 (±8.8), and average A1C was 7.1 (±1.6). Of the 215 individuals that participated in the baseline assessment of DISDOH, 183 individuals responded to the 14-day retest. The median time to complete DISDOH was 8.13 min.Table 3Prolific (phase III) participant demographics and biometric informationVariablesParticipantsn=215Demographics, N (%)Biological sex  Female111 (51.6)  Male103 (47.9)  Prefer not to answer1 (0.5)Race  African American/black22 (10.2)  Multiracial2 (0.9)  White173 (80.5)  Other10 (4.7)  Prefer not to answer8 (3.7)Ethnicity  Hispanic/Latino7 (3.3)  Non-Hispanic/Latino199 (92.6)  Prefer not to answer1 (0.5)Diabetes type  Type 147 (21.9)  Type 2168 (78.1)Biometric information, mean (SD)  Weight, lb 198.7 (±64.9)  BMI, lb/in2 32.0 (±8.8)  A1C, %7.1 (±1.6)BMI, body mass index.Confirmatory factor analysis The maximum likelihood CFA revealed the items loaded on five factors with eigenvalues exceeding 1, explaining a total of 69% variance. This model matched our proposed five domains for DISDOH. The combination of the χ 2/df ratio was observed to be 1.214 (χ2/df ratio ≤ 3 rule, X2 = 60.694, df=50), and the non-significant χ2 goodness-of-fit test (p=0.143) allowed us to fail to reject the null hypothesis and conclude the CFA model had an acceptable fit to the data. Table 4 outlines the loading factor of each item within the DISDOH domains.Table 4Confirmatory factor analysisFactorsFood environmentNeighborhood and physical environmentHealth contextSocial contextSocioeconomic statusdisdoh_ses1−0.0150.2610.0040.238 0.579 disdoh_ses20.2180.117−0.0040.316 0.308 disdoh_ses30.2450.2660.0200.371 0.627 disdoh_envi10.404 0.483 0.1900.1950.171disdoh_envi20.351 0.895 −0.107−0.052−0.01disdoh_envi30.401 0.789 −0.147−0.044−0.095disdoh_envi40.267 0.450 0.007−0.2950.009disdoh_envi50.357 0.342 −0.163−0.227−0.075disdoh_food1 0.999 −0.009−0.0030.0010.000test_disdoh_food2 0.673 0.170.1150.0910.074disdoh_health10.2050.049 0.307 −0.0910.084disdoh_health20.3930.15 0.783 −0.2090.113disdoh_health30.5450.122 0.592 −0.208−0.026disdoh_health40.4810.254 0.498 −0.284−0.087disdoh_social10.0610.2680.395 0.706 −0.266disdoh_social20.1710.2390.307 0.386 −0.278Bolded values denote which variables were included within each factor based on the analysis.Reliability analysis At baseline, the Cronbach’s α reliability estimate for the 16-item total score was 0.805. At the 14-day retest, the Cronbach α reliability estimate for the 16-item total score was 0.818. The test-retest reliability of the 16-item scale was excellent, as indicated by an average ICC of 0.901, suggesting strong consistency in responses over the 14-day period. The five DISDOH domains demonstrated acceptable internal consistency estimates: Domain 1: Socioeconomic status ( a=0.660), Domain 2: Neighborhood and physical environment (a=0.812), Domain 3: Food environment (a=0.801), Domain 4: Health context (a=0.812), and Domain 5: Social context (a=0.708).Convergent and discriminant validity To evaluate convergent validity, Pearson correlations were assessed between DISDOH domains and items measuring similar constructs on the Protocol for Responding to and Assessing Patients’ Assets, Risks and Experiences (PRAPARE) assessment and the Centers for Medicare & Medicaid Services (CMS) Accountable Health Communities (AHC) Health-Related Social Needs (HRSN) screening tool. Convergent validity was observed to have the following correlation for each DISDOH domain: DISDOH socioeconomic status items and PREPARE socioeconomic status items were statistically significant and moderate in magnitude (r=0.53, p<0.001), DISDOH neighborhood and physical environment items and CMS neighborhood and physical environment items were statistically significant and moderate in magnitude (r=0.55, p<0.001), DISDOH food environment items and CMS food environment items were statistically significant and strong in magnitude (r=0.68, p<0.001), DISDOH health context items and PREPARE health context items were statistically significant and moderate in magnitude (r=0.50, p<0.001), and DISDOH social context items and PREPARE social context items were statistically significant and moderate in magnitude (r=0.48, p<0.001). These results indicate that each DISDOH domain is capturing the aspect of SDoH it aims to be associated with.For divergent validity, items within PRAPARE and CMS-AHC-HRSN assessing different aspects of social determinants of health were compared with DISDOH domain items. Divergent items included CMS-AHC-HRSN items on disability, physical activity, and substance use. PREPARE items were related to safety. While there are some statistically significant differences between the DISDOH domains and the divergent items being compared, the actual correlation between them was observed to be very weak—weak, indicating DISDOH accurately captures its intended concepts. Divergent validity is presented in table 5.Table 5Divergent validity of the DISDOH measurement tool: comparisons with PRAPARE and CMS-AHC-HRSN screening toolsConvergent variablesDivergent variablesDISDOH SESDISDOH neighborhoodDISDOH foodDISDOH healthDISDOH social CMS disability   Pearson correlation −0.05−0.225** 0.370** −0.141* 0.006   Sig. (two-tailed)0.465<0.001<0.0010.0410.925   N 212212212212212 CMS physical activity   Pearson correlation 0.377** 0.065−0.039−0.040.185**   Sig. (two-tailed)<0.0010.3490.570.5610.007   N 212212212212212CMS substance use   Pearson correlation 0.253** −0.0550.171* −0.0260.091   Sig. (two-tailed)<0.0010.4290.0130.710.186   N 212212212212212 PRAPARE safety   Pearson correlation −0.105−0.329** 0.273** −0.132−0.133   Sig. (two-tailed)0.126<0.001<0.0010.0560.053   N 212212212212212Asterisks indicate statistical significance: p < .05 (*), p < .01 (**), based on two-tailed Pearson correlation tests.AHC, Accountable Health Communities; CMS, Centers for Medicare and Medicaid Services; DISDOH, Diabetes Index for Social Determinants of Health; HRSN, Health-Related Social Needs; PRAPARE, Protocol for Responding to and Assessing Patients’ Assets, Risks and Experiences.Conclusions Over the course of 2 years, we developed, tested, and validated the DISDOH. The DISDOH is a concise 16-item assessment that provides an overview of the ADA’s five critical SDoH factors influencing individuals with diabetes: socioeconomic status, neighborhood and physical environment, food environment, healthcare access, and social context. No similar measure exists presently. We established initial reliability and validity psychometric properties for the tool. As such, providers, patients, healthcare organizations, and other potential users may trust the interpretation of its results when assessing the SDoH vulnerability of individuals living with diabetes.While psychometric properties should be established in tool creation processes, pragmatism must also be embraced to foster better uptake and implementation of these tools.16 Concise measures such as the DISDOH are associated with a reduction of respondent time burden and higher response rate. While intrinsic adverse factors may play a role, as noted earlier by Garg et al,17 many advocates for SDoH screening tools have argued18 that recognizing them is a step in the right direction for better healthcare delivery—a claim supported by strong evidence.19 Of important note, users of tools like the DISDOH must be prepared and equipped to have conversations addressing the vulnerabilities these tools uncover, ensuring meaningful engagement and effective interventions.The DISDOH is one of the first concise, diabetes-specific SDoH measures that is designed to be both actionable and feasible for use in real-world settings, addressing a critical gap in diabetes care. Because of its psychometric properties, we are confident that the DISDOH tool is an effective screening option for those involved with caring for persons living with diabetes to foster more holistic health assessments and better health equity, and equip to better address the socially determined health outcomes associated with this chronic illness. To support practical implementation, we developed domain-specific and total score interpretation ranges that classify individuals into four levels of SDoH burden (high, moderate, minimal, and low), providing clear guidance for clinical decision-making. Lower DISDOH scores indicate a greater burden of SDoH that may hinder an individual’s ability to effectively manage their diabetes. In other words, individuals in the high SDoH burden category (lower DISDOH scores) are likely experiencing more SDoH challenges than those in the low SDoH burden category (higher DISDOH scores). These thresholds help identify individuals with elevated social needs and support the use of targeted referrals or interventions. Importantly, the DISDOH was developed in collaboration with programme facilitators to ensure that the questions focus on aspects of SDoH that providers or other care team members can help modify or improve—such as connecting individuals to food banks, support groups, or free clinics—making it a practical tool for driving tailored support and resource linkage.A key strength of this study is the systematic and iterative nature of the development and validation. By incorporating feedback at various points and from various parties (diabetes care experts, programme facilitators, survey researchers, and individuals with diabetes), the DISDOH was refined to enhance its validity, reliability, and usability for clinical and community-based settings. The DISDOH has additional potential utility for researchers. Its concise format allows for efficient integration into large-scale data collection efforts, enabling the identification of social needs across different diabetes populations. Furthermore, the focus on modifiable domains supports targeted intervention planning and facilitates evaluation of programme impact on health equity and social risk reduction.Limitations Despite these strengths, several limitations should be noted. First, the majority of participants were recruited through structured programmes such as HED and Prolific, which may not fully capture the breadth of experiences across settings frequented by those living with diabetes. Future validation efforts should include larger and more demographically diverse populations, including individuals receiving care in different healthcare systems or participating in diabetes education programmes outside of structured interventions. Given that the SDoH impact varies across demographic groups, this broader sampling will be critical to enhance external validity and ensure the tool’s applicability across underrepresented populations. Another consideration is the reliance on self-reported diabetes data, which may introduce recall bias or social desirability bias in responses. In a similar vein, some individuals may find certain SDoH-related questions sensitive, which could impact response accuracy. The DISDOH was developed as a brief, pragmatic tool for assessing SDoH among individuals with diabetes. Given its short-form structure, certain domains (eg, food environment, social context) contain a limited number of items. As such, the DISDOH may not fully capture the breadth and complexity of each domain and should be considered a starting point if a more comprehensive understanding of an individual’s SDoH is required. It is recommended that implementations of tools like the DISDOH be accompanied by mechanisms that can help when issues or vulnerabilities arise. Future studies should consider evaluating the tool’s predictive validity and responsiveness to interventions aimed at addressing SDoH-related barriers to diabetes care and management. We are eager to collect longitudinal data to further assess DISDOH applicability and long-term impact.Clinical applicability This study reports on the development and validation of the 16-item DISDOH assessment. The DISDOH focuses on ADA’s five critical SDoH factors influencing individuals with diabetes: socioeconomic status, neighborhood and physical environment, food environment, healthcare access, and social context. Results support the validity and reliability of the measure in individuals with diabetes. By offering a concise, yet comprehensive, tool, the DISDOH reduces the burden of current SDoH assessments, making it more feasible for use in clinical settings. Healthcare providers may integrate the DISDOH into routine diabetes care to systematically identify and address social risk factors, personalize treatment plans, and enhance patient-centered interventions. The DISDOH could also inform policy and resource allocation to improve diabetes outcomes at both individual and population levels.While the DISDOH shows promise as a feasible and actionable tool for use in clinical settings, we acknowledge that these conclusions are preliminary and based on psychometric validation rather than implementation data. Future research should examine the DISDOH’s integration into clinical workflows, including its compatibility with electronic health record systems and its potential to support value-based care initiatives. Pilot studies assessing provider uptake, usability, and impact on care planning and patient outcomes will be critical for determining its real-world utility. These next steps will help establish the DISDOH not only as a valid assessment tool, but also as a practical mechanism for addressing SDoH in routine diabetes care.