A Content Analysis of Text Messages in a Type 1 Diabetes Peer Mentoring Program—The Importance of Shared Interests

• Published in: Diabetes (New York, N.Y.) Vol. 67; no. Supplement_1
• Main Authors: WALKER, ASHBY F., JOHNSON, CATHRYN, ANEZ-ZABALA, CLAUDIA, DORVIL, SARAH R., HALLER, MICHAEL J., GURKA, MATTHEW J., BRUGGEMAN, BRITTANY S., GUIFFRE, DANIELLE, ATKINSON, MARK A., SCHATZ, ILYSSA, SCHATZ, DESMOND
• Format: Journal Article
• Language: English
• Published: 01-07-2018
• ISSN: 0012-1797; 1939-327X
• DOI: 10.2337/db18-844-P

A paucity of research exists regarding the feasibility of utilizing personalized, live-texting as outreach to improve health outcomes for low-income teenagers with type 1 diabetes (T1D). A peer mentoring program pairing college students with T1D with publicly insured teens ages 11-17 with T1D (gender matched) was piloted with the aim of improving outcomes for both groups (n=88; 44 control and 44 treatment). The program involved social events, clinic visits, automated daily texts for blood glucose monitoring (BGM), and weekly text exchanges. For consistency, mentors were trained in protocols for texts using a technique of reciprocal sharing and limited to one text exchange with their mentee a week at a time of day agreed on by parents. Mentors, mentees and parents were aware that all texts were monitored by study staff through IRB-approved processes. A content analysis of all texts for a 12-week period was conducted (47,475 words) using six coders with schemata that included presence/absence variables (e.g., did the mentor prompt for BGM), numeric variables (word counts), and categorical variables (major themes discussed). Texts included poignant examples of uninhibited sharing about daily life and excellent coaching by mentors. However, pairs that talked about a shared common interest outside of T1D used far more words about diabetes-specific and non-specific topics (both were athletes, both enjoyed drawing, shared an ethnic identity, etc.). Mean word use for pairs with a common interest was 2,902 compared to 1,537 with no common interest (p=.01; t-test). For mentors, as word use increased in texts there was a greater improvement in HbA1c (r = -.54, p=.03; Pearson’s correlation). This pilot study shows the feasibility of using texting in peer mentoring programs and reiterates the importance of finding common ground beyond T1D to facilitate rapport building. Additional efforts are needed to better explicate the role of shared interests in outcomes for peer mentoring programs.
Disclosure
A.F. Walker: None. C. Johnson: None. C. Anez-Zabala: None. S.R. Dorvil: None. M.J. Haller: None. M.J. Gurka: None. B.S. Bruggeman: None. D. Guiffre: None. M.A. Atkinson: Other Relationship; Self; Patent Issued. I. Schatz: None. D. Schatz: None.