Introduction & Objective: Digital chronic condition programs (DCCP) offer education, coaching, and self-management tools to improve health behaviors. Engagement is key for behavior change and improved outcomes, and personalized recommendations have shown to increase engagement. This study examined 1:1 coaching utilization with certified diabetes care and education specialists (CDCES) between DCCP members who received a personalized ‘next best action’ e-newsletter powered by predictive modeling.

Methods: Members with type 2 diabetes and no reported insulin use were included in a weekly e-newsletter pilot from 11/20/2023 - 1/29/2024 to evaluate predictive models that recommend next best action from member’s past engagement. Control and treatment groups were block randomized by age, gender, time in program, A1c, and historical email engagement rates. Control 1 (C1) received original design and content. C2 received new design and original content. Treatment 1 (T1) received new design and personalized activity graphic. T2 received new design and next best action. T3 received new design, activity graphic, and next best action. First time utilization of coaching compared between groups was measured by any coaching completed after first newsletter open date. Standard two sample independent proportions test was used to measure utilization differences. Results: Average age was 55.6 years old (SD 10.6) with each group 50% female. All treatment groups with next best action had increased utilization of coaching in members with no coaching engagement prior to pilot (N = 137,283) (C1 0.14%, C2 0.12%, T1 0.14%, T2 0.17%, T3 0.21%). Statistically significant increases were shown for T3 compared to both C1 (p=0.02) and C2 (p=0.007).

Conclusion: Use of predictive modeling to recommend next best action from past engagement in a DCCP was shown to be effective in increasing short-term coaching engagement that has been associated with improved glycemic control.

Disclosure

K. Kapoor: Employee; Teladoc Health. A. Ramasubramaniam: Employee; Teladoc Health. Y. Jiang: Employee; Teladoc Health. S. Boraston: Employee; Teladoc Health. M. Roach: Employee; Teladoc Health. A. Eiselt: Employee; Teladoc Health. Stock/Shareholder; Teladoc Health. Research Support; Teladoc Health. S. Lawton: None. T. Kompala: None. G. Yedapadi Rajaram: Employee; Teladoc Health.

Readers may use this article as long as the work is properly cited, the use is educational and not for profit, and the work is not altered. More information is available at http://www.diabetesjournals.org/content/license.