mHealth: Can text message based education lower Hb A1C?

It’s the start of a hectic shift, and you have many patients waiting to be seen. Right now, you are discharging a 31 year-old obese, uninsured, Spanish speaking woman, who presented for the second time this month with polyuria and blurry vision. You sit and explain to her the importance of managing her diabetes, but you wish you could do more to help support her after you discharge her—and for the many other similar patients waiting to be seen.


As emergency departments become increasingly utilized by safety net populations for access to healthcare, emergency care providers confront a growing dilemma: patient education. In order to improve the health outcomes of this subset of patients, there is a desperate need to provide a means to support disease management until this population can obtain regular outpatient care.

Enter mobile health (mHealth for short). In the broadest sense, mHealth is the use of mobile devices to engage in healthcare—from education to research. Given the demonstrated saturation of mobile phones throughout all socioeconomic echelons of society, mHealth seems like an attractive way to reach safety net populations in their time of greatest vulnerability.1,2 In this June’s Issue of the Annals of Emergency Medicine, the most robust study to date put mHealth to the test in a population of type 2 diabetic patients in Los Angeles County.

Article Citation

Arora S et al. “Trial to Examine Text Message–Based mHealth in Emergency Department Patients With Diabetes (TExT-MED): A Randomized Controlled Trial.” Annals of Emergency Medicine,2014 Jun;63(6):745-754.e6. doi: 10.1016/j.annemergmed.2013.10.012.

Study Objectives

To determine whether a unidirectional, text-message based mHealth intervention could improve clinical and behavioral outcomes in a diabetic safety-net population.

Primary outcome: change in Hb A1C

Secondary outcomes: changes in medication adherence, self-efficacy, performance of self-care tasks, diabetes-specific knowledge, ED utilization, patient satisfaction, and quality of life.

Study Methods

  • Single center, open label RCT of TExT-MED mHealth intervention versus usual care at Los Angeles County Hospital of the University of Southern California
  • Patients
    • Age ≥ 18
    • Speak and read Spanish or English
    • Have DM2 with HbA1C ≥ 8%
  • Intervention
    • Patients in the TExT-MED arm received twice-daily text messages (9 am and 5 pm) with diabetes education, triggers for self-care behaviors, and medication adherence encouragement.
  • Course
    • Patients returned after 6 months for a non-clinical appointment to re-evaluate outcomes.


  • No statistically significant decrease in HbA1C by TExT-MED group
    • TExT-MED: 1.05% decrease
    • Control: 0.60% decrease
  • No statistically significant difference in secondary outcomes, however:
    • Medication adherence was increased in the TExT-MED group
    • Emergency services were used less frequently by the TExT-MED group
  • Patient satisfaction with the TExT-MED intervention was high
    • No subjects dropped out of the messaging service
    • 100% would recommend the program to friends and family


While there was no statistical significance to the observed trends of improved outcomes in the mHealth intervention arm, the authors believe TExT-MED to be successful because of the improved medication adherence among TExT-MED users—especially in the higher-risk Latino subset of their study population. This mixed-bag sort of result is par for the course in the limited studies available on compatible mHealth interventions. Unfortunately, the study’s biggest limitation prevents us from fully understanding this result.

Although subjects were asked to rate satisfaction with TExT-MED, there was no information collected about how the TExT-MED intervention was utilized. There was no basic means of assuring study participants even read the mHealth messages. In order to verify congruent mHealth services are being used meaningfully, further studies ought to explore and verify subjects’ interaction with the one-way message system. Potential new metrics might include questions about the number of times messages were viewed, the percentage of messages unopened, the duration of time messages were viewed for, and the delay between message receipt and viewing. Moving forward, this will be essential to refining mHealth approaches, and it might identify subset populations for which mHealth is a more effective intervention.

The promise of high patient satisfaction with the program is inspiring, but might not be as suggestive as the authors conclude. While no study participants dropped out of the messaging service during the study, it is unclear if they were reminded that they could do so during the 6-month period. Additionally, 21% of subjects opted to drop out of the program at the study’s conclusion, presumably after the $175 compensation for participating was awarded.

Final Thoughts

TExT-MED is a unique mHealth platform that demonstrates potential for cheap, large-scale deployment. Still, it suffers from an absence of validation. It could represent a powerful means of extending patient education to patients awaiting critical long-term follow up. It could, alternatively, fail to demonstrate efficacy or suffer from limited investigation like other mHealth initiatives. More rigorous studies need to be conducted to elucidate which path such mHealth interventions will take in the future of emergency medical care.

Programs like TExT-MED represent a powerful opportunity to expand the domain of emergency medicine and help us impact population health outcomes. With the shifting shores of ED utilization in healthcare reform, future mHealth initiatives may provide a means for Emergency Departments to better define their role in provision of healthcare for patients and society at large.

References / Further Reading

  1. Post LA, Vaca FE, Doran KM, Luco C, Naftilan M, Dziura J, Brandt C, Bernstein S, Jagminas L, D’Onofrio G. New Media Use by Patients Who Are Homeless: The Potential of mHealth to Build Connectivity. J Med Internet Res 2013;15(9):e195. doi: 10.2196/jmir.2724
  2. D. Keith McInnes, Alice E. Li, and Timothy P. Hogan. Opportunities for Engaging Low-Income, Vulnerable Populations in Health Care: A Systematic Review of Homeless Persons’ Access to and Use of Information Technologies. American Journal of Public Health 2013; 103(S2):e11-24. doi: 10.2105/AJPH.2013.301623
Scott Kobner is a medical student at NYU. Article edited by Alex Koyfman.

Leave a Reply

Your email address will not be published. Required fields are marked *