Using bivariate analyses, we examined the association between DSMES use and age, sex, subscriber status, metropolitan status, and CCI score category. For individuals who utilized any DSMES during the 2-year period, we estimated a binary logit model to estimate the probability that an individual chose Omada, controlling for other demographic and health status factors. Data analysis was performed using SAS 9.4 (SAS Institute Inc).
RESULTS
Communications Strategy Development and Deployment
To promote the new O4D program, SEGIP administrators developed a communications strategy in collaboration with Omada and the Minnesota Department of Health, along with input from University of Minnesota–based research team members. During the first year O4D was offered, SEGIP coordinated 2 communication campaigns to promote engagement in both Omada programs.
Promotional messages were informed by 6 employee focus groups conducted virtually in early 2021 with 30 total participants. Minnesota Department of Health staff and the research partner developed focus group questions to identify effective messaging and communication outreach strategies to promote the DSMES benefit, understand participation barriers, and test specific messages to promote O4D, including potential differences for enrollees in urban vs rural locations. The research partner conducted thematic analyses of focus group data.
The focus groups demonstrated the importance of using language that emphasized the collective vs individual (“we” vs “me”), underscored what can be learned from engaging in O4D (eg, what knowledge/skills are gained from participating), and highlighted that communications should use positive vs fear-based framing related to managing diabetes. Focus group participants also noted the importance of clearly articulating DSMES insurance coverage, specifying who to contact with questions, providing a website link with more information, ensuring that messages were culturally adapted, and using simple messaging without too many graphics.
In May and June 2021, SEGIP conducted its first communications campaign; the second campaign occurred from January through March 2022. Communications varied in 4 dimensions: messenger (SEGIP or Omada), recipient population (broad or narrow), modality (website, email communication to employee, mail to home address), and message framing.
Table 1 summarizes communications tactics to promote O4D. During the first campaign, SEGIP communicated with its enrollee population about Omada using 4 distinct communication approaches. It incorporated information about Omada on its benefits web page using language to motivate engagement through maximizing one’s benefits and potential financial incentives through earning points as part of its incentive-based well-being program. SEGIP also integrated information about Omada into its Advantage Value for Diabetes communications sent to members with newly diagnosed diabetes. Third, links to Omada’s website were provided by SEGIP’s plan administrators. Fourth, SEGIP introduced employee members to O4D through a wellness email that covered multiple topics. This final communication was delivered to 53,352 individuals. SEGIP administrators reported that 46% opened this message and 5% clicked on at least 1 of 13 links included in the message. These rates are similar to SEGIP’s median open and click rates for health-related messaging overall and higher than other state and local median benchmark data (25% and 4%, respectively).17,18
Omada, in coordination with SEGIP administrators, also provided communications to targeted enrollee segments during this first campaign. Specifically, Omada emailed SEGIP members who previously applied for the Omada for Prevention program (targeting those with prediabetes) but were ineligible because of a preexisting diabetes diagnosis. Although this communication had a lower open rate (39%), Omada reported a higher click rate (17%). Also, Omada sent a SEGIP cobranded postal mailer about O4D to 4030 members who had a recent claim for diabetes medications, utilizing established research partnerships between SEGIP, Omada, and SEGIP’s PBM.
Table 1 also summarizes communication efforts from the second campaign in early 2022, which coincided with the common practice of setting New Year health behavior resolutions. Strategies included a virtual meeting with SEGIP wellness champions, encouraging them to promote the O4D program within state agencies, and email communications to all employee members about SEGIP’s focus on health and well-being, including direct promotion of Omada. Additionally, a separate follow-up email communication was made to employee members residing in nonurban communities (n = 13,671). This communication was motivated by a preliminary analysis of SEGIP claims data suggesting lower DSMES utilization among individuals in nonmetro areas (authors’ unpublished analysis for the Minnesota Department of Health). Communication to the nonmetro audience yielded a 38% open rate and 2% click rate—considerably lower than the overall SEGIP population. Finally, Omada distributed a second round of postal mailers to SEGIP members with diabetes based on the PBM data (n = 3965).
Omada Program Engagement and Communications Timing
The Figure shows monthly counts of the number of applicants to O4D and Omada for Prevention combined, the number eligible for O4D, and the number who enrolled and completed at least 1 lesson based on Omada administrative data. Omada uses a single application system and screener for both programs to triage individuals appropriately. Eligibility for the O4D program is based on applicants’ responses to questions about HbA1c or fasting blood glucose laboratory values and self-reported diabetes diagnosis. Between May 2021 and April 2022, Omada received 940 applications from SEGIP members for its diabetes prevention program and O4D. Of these, 272 individuals were eligible for O4D, and 216 enrolled (defined as having completed at least 1 session). As seen in the Figure, the number of applications increased immediately following both communication campaigns.
Table 2 includes self-reported demographic and health status summary statistics shared by Omada. Enrollees in O4D were more likely to be female (57.4%) and aged 45 to 60 years (56.5%). Of the enrollees, 77.8% identified as White, 6.5% identified as Black, 5.1% identified as Asian or Pacific Islander, 3.7% identified as Hispanic, and 6.9% identified as other race/not available. Among enrollees, 72.2% reported a body mass index (BMI) of 30 or greater, and 18.5% had a BMI of 25 to 29.9.
Omada Engagement, DSMES Utilization, and Urban-Rural Disparities
Next, we examined demographic and clinical characteristics associated with engagement with O4D alongside traditional DSMES offerings (in-person and telehealth combined) using claims data (Table 3). Across this 2-year period, 19.0% of individuals with a diabetes diagnosis utilized DSMES from a traditional provider, 2.9% engaged with the O4D program, and 78.1% did not engage in any DSMES services. When comparing DSMES utilization in the year before with the year after the launch of O4D, the percentage of individuals with at least 1 claim for traditional DSMES or O4D increased modestly, from 11.7% to 13.9%. This increase was driven mainly by individuals who had not used DSMES during 2020, the year before O4D was launched.
Table 3 also shows that age, female sex, subscriber status, and metropolitan residence are all statistically associated (P < .05) with use of DSMES. We did not observe a relationship between DSMES status and overall health risk as captured by the CCI score. Variation in estimated prevalence of DSMES use (O4D, traditional DSMES, no DSMES) suggests that SEGIP enrollees who are older, male, and living in nonmetropolitan areas are less likely to engage in either DSMES delivery type.
Finally, we estimated a parsimonious binary logit model of an enrollee’s decision of whether to engage with Omada as a function of demographic and health status factors. Among the subset of individuals with at least some DSMES use during the 2-year period (n = 1525), only female sex and employee subscriber status were positive predictors of O4D program use. No other factors demonstrated statistically significant relationships.
DISCUSSION
Ever-evolving population health management strategies used by employers and insurers now include adopting digital health solutions to expand enrollees’ choices for chronic condition management, including DSMES in the context of diabetes management. However, evidence about implementation, uptake, and program effectiveness in real-world settings is limited.
Our evaluation generated some key lessons for employers and other stakeholders related to implementing a digitally based DSMES program. First, know your audience. Implementation of disease management solutions requires partnerships to provide deliberate, coordinated communications and positive messaging to encourage engagement by enrollees. SEGIP, its third-party administrators, and Omada utilized a variety of passive and active communication strategies that varied by population targeted, modality, and message framing. Although program implementation efforts did not include communications with SEGIP’s other in-network providers, this could also serve as a viable pathway to increase awareness of available resources. However, for some providers who also deliver DSMES, stand-alone digital solutions may be perceived as potential competition.
Second, define success for your population. Employers and plans should establish expectations for potential demand for digitally based disease management that account for existing alternatives and market conditions. Despite a well-developed communication strategy, SEGIP did not experience high take-up of this digital program or greatly expanded overall DSMES use. Demand for the program may have been lower than expected if, for example, enrollees faced digital access barriers (eg, lower comfort levels with mobile health solutions).19 Our analysis also did not reveal strong spillover effects. That is, offering and aggressively promoting a new digitally based DSMES option did not result in significant increases in traditional DSMES use. One possible reason is that traditional DSMES use requires a provider referral, which may create a potential barrier. DSMES provider capacity constraints could also reduce access even if an individual wanted care. Another important consideration is that this intervention was deployed in an employer setting with a robust set of offerings to support individuals with diabetes, including established DSMES coverage, medication therapy management, and a value-based insurance design to reduce financial barriers. With these existing investments, eligible individuals may have perceived lower incremental benefits from this new program.
When considering potential demand, there may also be value in evaluating the market environment. Many health care delivery organizations and health plans in Minnesota have made significant investments to promote effective diabetes management because of public reporting initiatives (eg, Minnesota Community Measurement). Rates of DSMES utilization in this population are already higher than what is reported nationally, so it may be more difficult to further increase DSMES utilization rates.20 Future work could examine whether take-up rates vary across types of employer settings (eg, health benefits generosity) or geographic markets with varied traditional DSMES provider capacity or lower DSMES baseline utilization.
Third, many digital health solutions, including Omada, are evidence-based in their designs, with many positive features to encourage effective self-management of diabetes. Yet, the evidence base for quantifying specific health and economic effects of such programming in the postpandemic period remains limited. Evaluations may also need to consider, when defining success, the presence of heterogeneous effects in outcomes considered (eg, for whom do these programs work most effectively, assessing differences in how much employees value choice) and whether these programs mitigate or exacerbate socioeconomic-based disparities in clinical diabetes or other related outcomes. Low take-up rates and lack of information on enrollees’ race, ethnicity, or language within specific employer organizations may create challenging conditions for undertaking rigorous quantitative evaluations with sufficient statistical power to detect improvements and measure disparities.
Limitations
Our study was subject to a few key limitations. First, insights were based on experiences of a single large employer with a mature set of diabetes-related programming. Experiences may differ in other employer contexts. Second, we examined implementation during the program’s initial year (May 2021-April 2022), which coincided with the second year of an active public health emergency. Although higher participation rates might have been expected, given individuals’ greater familiarity with telehealth and digital technologies, other factors may have influenced this outcome, such as fatigue around digital connectivity or prioritization of other issues. Third, due to data limitations, we were unable to examine differences in enrollees’ engagement by income or race/ethnicity.
CONCLUSIONS
This study examined a large, self-insured employer’s experience with implementing a digitally delivered DSMES solution that complemented existing diabetes prevention and management solutions offered by the employer. Despite a comprehensive communications campaign, including targeted messaging to eligible enrollees in nonmetropolitan areas, program enrollment in absolute terms was low compared with the size of the eligible population, with female and employee subscribers more likely to enroll in the digitally based DSMES program. Overall, DSMES use increased modestly, from 11.7% prior to O4D launch to 13.9% afterward; however, we did not find evidence to suggest significantly higher rates of use among members in rural areas following O4D launch. With the diffusion of digital health solutions for chronic condition management, including diabetes, many opportunities exist for future evaluations that can inform best practices related to implementation and measuring program effectiveness on population health and economic outcomes.
Acknowledgments
The authors gratefully acknowledge the assistance and feedback of Josh Fangmeier, Carrie Suplick Benton, and Dawn Cvengros, as well as Bunchung Ly and Anna Granias from the Wilder Foundation.
Author Affiliations: University of Minnesota (JMA, HMP), Minneapolis, MN; Minnesota Department of Health (TA, MC, RSMK), St Paul, MN.
Source of Funding: Minnesota Department of Health (Prime: Centers for Disease Control and Prevention Cooperative Agreement 1NU58DP006611).
Author Disclosures: The authors report no relationship or financial interest with any entity that would pose a conflict of interest with the subject matter of this article.
Authorship Information: Concept and design (JMA, TA, MC, RSMK, HMP); acquisition of data (MC); analysis and interpretation of data (JMA, TA, MC, RSMK, HMP); drafting of the manuscript (JMA, TA, RSMK, HMP); critical revision of the manuscript for important intellectual content (JMA, TA, RSMK, HMP); statistical analysis (JMA, HMP); obtaining funding (TA); administrative, technical, or logistic support (JMA, TA, RSMK); and supervision (JMA, TA).
Address Correspondence to: Jean M. Abraham, PhD, University of Minnesota, 420 Delaware St SE, MMC 729, Minneapolis, MN 55455. Email: abrah042@umn.edu.
REFERENCES
1. Keisler-Starkey K, Bunch LN, Lindstrom RA. Health Insurance Coverage in the United States: 2022. Current Population Reports. US Census Bureau. September 2023. Accessed June 6, 2024. https://www.census.gov/content/dam/Census/library/publications/2023/demo/p60-281.pdf
2. Buttorff C, Ruder T, Bauman M. Multiple Chronic Conditions in the United States. RAND Corporation. 2017. Accessed October 22, 2022. https://www.rand.org/content/dam/rand/pubs/tools/TL200/TL221/RAND_TL221.pdf
3. Claxton G, Rae M, Winger A, Wager E. Employer Health Benefits: 2023 Annual Survey. KFF. 2023. Accessed December 27, 2023. https://files.kff.org/attachment/Employer-Health-Benefits-Survey-2023-Annual-Survey.pdf
4. The prevalence of diagnosed diabetes, pre-diabetes, and gestational diabetes among the ESI population, 2008-2012. Health Care Cost Institute. December 2013. Accessed August 12, 2023. https://healthcostinstitute.org/images/easyblog_articles/117/Diabetes-Prevalence-2008-2012.pdf
5. National Diabetes Statistics Report. CDC. Accessed December 17, 2022. https://www.cdc.gov/diabetes/php/data-research/index.html
6. Sen A, Gordon BS, Valencia Z, Pupino A. Privately insured individuals with diabetes spend twice as much out-of-pocket on health care as those without diabetes. Health Care Cost Institute. July 13, 2022. Accessed May 26, 2023. https://healthcostinstitute.org/hcci-originals-dropdown/topics/diabetes-and-insulin/privately-insured-individuals-with-diabetes-spend-twice-as-much-out-of-pocket-on-health-care
7. AADE. AADE7 self-care behaviors. Diabetes Educ. 2008;34(3):445-449. doi:10.1177/0145721708316625
8. Powers MA, Bardsley JK, Cypress M, et al. Diabetes self-management education and support in adults with type 2 diabetes: a consensus report of the American Diabetes Association, the Association of Diabetes Care & Education Specialists, the Academy of Nutrition and Dietetics, the American Academy of Family Physicians, the American Academy of PAs, the American Association of Nurse Practitioners, and the American Pharmacists Association. Diabetes Care. 2020;43(7):1636-1649. doi:10.2337/dci20-0023
9. Whitehouse CR, Haydon-Greatting S, Srivastava SB, et al. Economic impact and health care utilization outcomes of diabetes self-management education and support interventions for persons with diabetes: a systematic review and recommendations for future research. Sci Diabetes Self Manag Care. 2021;47(6):457-481. doi:10.1177/26350106211047565
10. Luo H, Bell RA, Winterbauer NL, et al. Trends and rural-urban differences in participation in diabetes self-management education among adults in North Carolina: 2012-2017. J Public Health Manag Pract. 2022;28(1):E178-E184. doi:10.1097/PHH.0000000000001226
11. Adjei Boakye E, Varble A, Rojek R, et al. Sociodemographic factors associated with engagement in diabetes self-management education among people with diabetes in the United States. Public Health Rep. 2018;133(6):685-691. doi:10.1177/0033354918794935
12. Mendez I, Lundeen EA, Saunders M, Williams A, Saaddine J, Albright A. Diabetes self-management education and association with diabetes self-care and clinical preventive care practices. Sci Diabetes Self Manag Care. 2022;48(1):23-34. doi:10.1177/26350106211065378
13. Rutledge SA, Masalovich S, Blacher RJ, Saunders MM. Diabetes self-management education programs in nonmetropolitan counties – United States, 2016. MMWR Surveill Summ. 2017;66(10):1-6. doi:10.15585/mmwr.ss6610a1
14. Chronic Conditions Data Warehouse. Accessed August 12, 2023. https://www2.ccwdata.org/web/guest/home/
15. Rural-urban commuting area codes – documentation. US Department of Agriculture. Updated March 22, 2023. Accessed April 12, 2023. https://www.ers.usda.gov/data-products/rural-urban-commuting-area-codes/documentation/
16. Deyo RA, Cherkin DC, Ciol MA. Adapting a clinical comorbidity index for use with ICD-9-CM administrative databases. J Clin Epidemiol. 1992;45(6):613-619. doi:10.1016/0895-4356(92)90133-8
17. 2021 Civic Engagement Benchmark Report. Granicus. Accessed May 22, 2023. https://granicus.com/pdfs/2021_Granicus_Benchmark.pdf
18. 2022 Public Sector Digital Communications Benchmark Report. Granicus. Accessed May 22, 2023.
https://granicus.com/pdfs/guide_2022_Digital_Comms_Benchmark.pdf
19. Minnesota study of telehealth expansion and payment parity. Minnesota Department of Health. Updated June 30, 2023. Accessed March 28, 2024. https://www.health.state.mn.us/data/economics/telehealth/index.html
20. Li R, Shrestha SS, Lipman R, Burrows NR, Kolb LE, Rutledge S; Centers for Disease Control and Prevention (CDC). Diabetes self-management education and training among privately insured persons with newly diagnosed diabetes—United States, 2011–2012. MMWR Morb Mortal Wkly Rep. 2014;63(46):1045-1049.