Increasing Social Determinants of Health Screening Rates Among Six Endocrinology Centers Across the United States: Results From the T1D Exchange Quality Improvement Collaborative

This QI study was conducted among five pediatric diabetes centers and one adult diabetes center in the T1DX-QI. All participating centers received local institutional review board approval to share aggregate data and participate in this project. Aggregate baseline data were collected and stratified by site (Table 1). No protected health information was transmitted outside of each clinic. This project was deemed nonhuman subject research by the Western Institutional Review Board.

The six participating T1DX-QI centers were Cincinnati Children’s Hospital Medical Center in Cincinnati, OH; Baylor College of Medicine in Houston, TX; Hassenfeld Children’s Hospital at NYU Langone in New York, NY; Children’s Mercy Research Institute Hospital in Kansas City, MO; and the adult and children’s diabetes centers at SUNY Upstate Medical University in Syracuse, NY. Participating centers each have physician champions, advanced practice nurses, CDCESs, psychologists, social workers, and patients as part of their team (Table 2). The six participating centers in this project serve a total of 17,917 patients with a mean age of 15.1 ± 5.6 years.

The centers used Plan-Do-Study-Act (PDSA) cycles to implement and scale up interventions to increase SDOH screening. PDSA cycles were implemented and communicated to the T1DX-QI coordinating office through regular QI coaching calls (Figure 1).

Table 3 summarizes the interventions that were tested and scaled. Among these were provider training on SDOH screening, use of an SDOH tab in the electronic health record (EHR) system, development of procedures suitable for both telemedicine and in-person visits, integration of repeated SDOH screening at every visit, use of paper forms and digital tablets for screening, revised language in the screener to emphasize confidentiality and communicate the availability of free resources, translation of the screener into multiple languages, implementation of an EHR best practice alert to prompt referrals for patients with positive screening results, and a streamlined process for making referrals for social work support when needed. Standardized SDOH assessment was also introduced for patients admitted to the hospital with diabetic ketoacidosis (DKA). This assessment examines factors contributing to the occurrence of the DKA episode that extend beyond medical and insulin-related matters (Supplementary Figure S1).

The primary QI outcome was the SDOH screening rate, measured as the total number of all patients with type 1 diabetes seen in the reporting month as the denominator and the total number of those patients who were screened in the reporting month as the numerator.

Participating clinics shared monthly aggregate data with the T1DX-QI coordinating office using a secure collaborative spreadsheet (www.smartsheet.com). Data were reported from July 2020 through October 2022.

Data were analyzed using control chart rules; a control chart is a type of process-behavior data chart that assesses shifts and evaluates QI project effectiveness (17). Shifts were identified if any point was beyond the specified three-sigma control limits, if two of three consecutive points fell beyond the two-sigma control limits, if four of five consecutive points fell beyond the one-sigma control limits, or if the run of eight consecutive points fell on either side of the center line (18). Control charts were created using SPC for Excel software (www.spcforexcel.com). We applied SQUIRE 2 (Revised Standards for Quality Improvement Reporting Excellence) guidelines when writing this article (19).

SDOH screening rates increased across all participating clinics, with improvements ranging from 41 to 70%. The mean SDOH screening rate across all six centers increased by 69% between July 2020 and October 2022 through targeted interventions specific to each center.

Two shifts were identified using control chart rules (17). The pre-intervention mean from July 2020 to January 2021 was 0.2%; the mean from February to December 2021 was 61%, and the mean from January to October 2022 was 70% (Figure 2). Statistical significance was calculated for each shift using a t test, and the P value was <0.001 (Table 4).

To our knowledge, this is the first multicenter QI initiative aimed at increasing screening for SDOH across multiple diabetes centers in the United States. Notably, all collaborating centers achieved an increase in their SDOH screening rate throughout the study.

SDOH screening has enabled clinical settings to identify the social and environmental conditions affecting patients and to address patient needs that extend beyond standard clinical care (10). In the absence of regular SDOH screening, health care practitioners frequently remain uninformed about challenges faced by many people with type 1 diabetes and their families. In our cohort, all participating centers addressed the social needs of patients with positive screening results by referring them to appropriate local community resources.

Our project revealed that language barriers are an obstacle for non–English-speaking patients in our clinics. The screening rate increased at two participating centers that translated the SDOH screening tool into other languages such as Spanish, Arabic, and Nepali. SDOH screening rates increased among Hispanic people with type 1 diabetes when two of the participating centers revised the screening questions and added a sentence to emphasize confidentiality, highlight the voluntary nature of participation, and impart that available referral resources are free of charge. Further research is needed to establish a connection between racial/ethnic minority status and diminished frequency of SDOH screenings.

One of the centers in our study found that Black and Hispanic patients were disproportionately affected by SDOH-related barriers (20). Years of research have demonstrated that diabetes affects racial/ethnic minorities and low-income populations disproportionately, with higher risks of diabetes complications and mortality in these populations (4,20).

QI methodology is feasible in improving outcomes for people with type 1 diabetes. Five T1DX-QI centers applied QI tools and developed interventions to target barriers to insulin pump therapy among 12- to 26-year-old patients. After a series of rapid tests of change over 22 months, insulin pump uptake increased by 13% from baseline across the centers (21). Prahalad et al. (22) described QI interventions in 10 clinics in the T1DX-QI to increase the use of continuous glucose monitoring (CGM). Through targeted interventions specific to each center, CGM use increased by 21% over 20 months. QI methods also have been successfully used to improve SDOH screening in other areas of medicine (23–25).

In a recent T1DX-QI study aimed at increasing CGM use by non-Hispanic Black and Hispanic people with type 1 diabetes, the proactive identification and resolution of barriers identified through SDOH screening proved to be instrumental in increasing CGM uptake. Median CGM use increased by 12% among non-Hispanic Black people with type 1 diabetes and 15% among Hispanic people with type 1 diabetes (26).

Screening for and addressing SDOH barriers should be tailored to each center’s ability to change and adapt its workflows and should be cost-effective and culturally appropriate to promote health equity (5). As health care systems continue to address this need, it is important to establish best practices for implementing SDOH screening and making necessary referrals. Future studies should explore follow-up processes for individuals who have positive SDOH screening results.

A strength of this project was that participating clinics had the flexibility to create their own PDSA cycles based on their specific priorities, procedures, and policies and their patient population’s needs. This multicenter project also provided an opportunity for centers to learn from each other during monthly coaching calls.

A limitation of this project was that participating centers were all academic settings with varying levels of QI infrastructure and capacity; thus, some centers conducted more PDSA cycles than others. Some of the centers relied on paper documentation, which may have led to underreporting of SDOH screening. All participating centers screened for SDOH, but there was no agreement on which specific domains to include in the screening. Finally, this was an observational study, and most participating centers implemented multiple interventions. Therefore, we were unable to assess the success of a single intervention in increasing SDOH screening.

We applied QI methodology and principles to execute this project. Participating centers tested changes through rapid PDSA cycles. Once successful interventions were identified, they were expanded and sustained over time. This project demonstrated that increasing SDOH screening and facilitating connections between people with positive screening results and potentially helpful resources is feasible.

We learned the following key lessons through this project: 1) clinic processes and policies differ among participating centers, allowing interventions to be customized according to existing guidelines and procedures and facilitating favorable outcomes; 2) regular team meetings with multidisciplinary team members are a valuable tool for sharing improvement ideas and fostering learning; 3) involving patients/parents in brainstorming change ideas is important and aids in recognizing potential obstacles and contributing factors; and 4) the timely provision of data reports and the enthusiasm of the QI team accelerate the achievements of QI initiatives.

The authors thank the Leona M. and Harry B. Helmsley Charitable Trust for funding the T1DX-QI. The authors also acknowledge the contributions of patients, families, diabetes care teams, and collaborators within the T1DX-QI who continually seek to improve care and outcomes for people living with diabetes.

O.E. is a compensated Health Equity Advisory Board member for Medtronic Diabetes and serves as principal investigator for investigator-led projects sponsored by Abbott, Eli Lilly, Insulet, and Medtronic. He is compensated through his organization, the T1D Exchange. No other potential conflicts of interest relevant to this article were reported.

O.O. wrote the first draft of the manuscript. O.E. conceptualized the study, served as its principal investigator, and substantially reviewed and edited the manuscript. All authors critically edited the manuscript and approved the final version for submission. O.O. 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.

A portion of the data included in this article was presented in abstract form at the American Diabetes Association’s 83rd Scientific Sessions, 23–26 June 2023, in San Diego, CA.