Utilizing Off-the-shelf Technology to Improve Diabetes Management in a Rural Population

Intervening in a rural primary care population with a clinically validated telehealth tool (Epharmix Diabetes) that algorithmically triages patients and facilitates patient-provider communication will result in a statistically significant improvement in the indicators of diabetes control.

Study Design

• Study Type: Experimental, non-controlled crossover study (intervention vs. standard of care for clinic)
• Primary aim: decrease average self-tested pre-prandial blood glucose level and HgbA1C
• Secondary aim: decreased number of disease-related emergency room (ER) visits, decreased number of self-reported blood glucose measurements outside of goal
• Duration: 3 months + 1 month enrollment period
• Safety: Protocol will be under direction of University of Missouri Institutional Review Board (IRB). All patients will be referred to emergency department if unsafe blood glucose levels are reported. All information is encrypted and HIPAA compliant.
• Sample Size: Clinic flow analysis estimates 61 patients eligible for study (5 providers, assuming 30% of eligible patients who meet inclusion criteria during 4-week enrollment period interested), with 50 patients likely to stay in study (80% retention rate).
• Power analysis: Based on pilot data and assuming an alpha error = 0.05, this study will have have greater than 80% statistical power of detecting differences in populations using only 20 patients with a minimum detectable effect size (decrease in HgbA1C) of 0.7%

Methodology

• The investigator will supply family practice clinics at Mercy with a one-page summary and advertisement of the intervention. At some point during a clinical encounter with a patient with an existing or new diagnosis of diabetes, providers will invite the patient to participate in this Community Integration Project. Providers will maintain a list of patients interested in the study with corresponding phone numbers and the investigator will regularly add patients to the study.
• Over a 3-month study period, patients will interact with the telehealth system twice a day, entering blood glucose levels when prompted. As recommended by physicians at Mercy, the investigator will message the physicians once a week with a digest of their patients at risk through the existing electronic medical record (EMR) inbox to more easily integrate with existing clinic flow.
• HgbA1c will be measured prior to starting intervention and will be measured within 7 days of completing intervention.
• Providers will be notified by the system when patients are triaged to be "high risk," that is, out of a pre-selected range of self-reported blood glucose levels. Providers can then be automatically connected with these patients to titrate blood glucose control.
• At the conclusion of study, new average self-reported blood glucose responses will be compared against those averages prior to intervention and to clinic averages for patients with diabetes. Additionally, HgbA1C levels will be compared pre and post -intervention.
• Results will be analyzed and conclusions will be presented to participating stakeholders in a poster session.