This pilot study will involve intensive training in self-management of diabetes mellitus. The intensive diabetes self-management program will involve an initial diabetes assessment with the nurse/diabetes educator, an initial evaluation by the Endocrinologist, Dr. Santen, at one of the Tri-Area CHC clinics; weekly or bi-weekly phone calls to review glucose and insulin data and 3-month, 4.5 month and 6-month follow-ups on telemedicine with Dr. Santen. A cloud based glucose monitoring system will utilize the Verizon based Telcare glucose meter with glucose test strips and the Glucommander-outpatientR algorithm to make insulin dosage recommendations. The goal will be to test blood glucose 4-7 times/day. The serial glucose levels are accessible on the meter itself but also can be accessed online by the entire diabetes team. Weekly to four weekly phone calls with the UVA endocrinologist ( and nurse/diabetes educator from Tri-Area as necessary) will be completed to discuss glucose patterns and make insulin adjustments. In addition, the patient will attend four diabetes tele-education programs to learn more about improved self-management of their diabetes. To facilitate nutritional education, a modified , patient specific, Nutrisystem R diet will be supplied , designed to be the exclusive source of nutrition. This diet will involve 1200 calories for women and 1800 calories for men. Both frozen and non-frozen food will be delivered to the patient. Nutrisystems will supply these " meal replacements" at a reduced cost based on this pilot program.

Introduction: Greater use of technology will allow patients with diabetes on insulin therapy to optimize their glucose levels, limit input from health care providers, and thus markedly reduce costs incurred by health care professional fees. A key component required to implement training in use of this technology is an intensive program of self-management which utilizes automated, cloud-based glucose data transfer and storage along with analysis and algorithm and endocrinologist based treatment recommendations. The hypothesis to be tested in this study is to demonstrate that inclusion of the technology based approach rather than standard of care will statistically significantly improve primary and secondary endpoints. A simple, technological approach is most suited to underserved populations where medical resources for patients with diabetes are scarce. A successful study will enhance the rationale for using Cloud-based meter and analysis systems.

Background: Diabetes mellitus is a very common disease with 22.3 million affected in the United states and with estimated annual costs of $245 billion(1). About 18 million people die every year from cardiovascular disease, for which diabetes and hypertension are major predisposing factors. The prevalence of diabetes in the USA is estimated to increase from 7.4% in 2014 to 12% in 2025. The increase is driven primarily by the increase in obesity .This is not exclusively a USA problem as globally the number of patients with diabetes will increase to 380 million from 246 million in 2007.(2) Much of this increase is attributable to the increase in obesity both globally and in the United States specifically (2) .

The American Diabetes Association recommends referral to an endocrinologist when patients do not achieve the desired goals of treatment, namely a HbA1C at or below 7.0% (3) However, in the standard care arm of the DCCT, the median Hb A1C documented during conventional treatment was 8.9% and only 12% of patients were at or below 7%.(ref).These data suggest that a large fraction of patients with diabetes should be referred to an endocrinologist for advice and/or management. However, workforce data report that only 15% % of patients with diabetes at present are evaluated by endocrinologists(4). Part of the reason is the gap in workforce numbers of Endocrinologists in the USA. There are the 5496 board certified adult endocrinologists in the USA currently(4) with an estimated gap of 1500 endocrinologists in 2014 .This gap which is estimated to grow to 2700 by 2025 unless more endocrinologists are trained. Diabetes. As diabetes care represents 46.1% of the coded visits of Endocrinologists(4) and there is a substantial workforce gap, it is obvious why involvement by endocrinologists is limited.

The data described above reveals that new approaches are needed to improve the care of patients with diabetes. This is perhaps more relevant for patients living in rural, undeserved areas where access to specialists, primary care teams focused on management of diabetes, and primary care providers is limited by professional, geographical and financial considerations(5) (ref) .

A proposed solution to the rural access problem is the use of telehealth and telemedicine. Telehealth is defined as access to providers and educational resources by video, smartphone, computer, telephone, cloud based glucose meters, text messaging and other means of rapid communication. Telemedicine specifically refers to patient encounters with providers via real time television interactions . These methodologies can alleviate the access problems due to geography, the availability of endocrinologists, and the cost of specific clinic visits. Pilot data from the Kaiser Permanente Health care system suggest that 90% of encounters with patients with diabetes can be replaced via telehealth methodology.

Even with use of telehealth technology, the number of providers, endocrinologists , educators, and members of multi-disciplinary teams with a diabetes focus are limited. A solution to this problem, is to teach the patient self-management. It is hypothesized that an intensive, 6 month, educational program in self-management will markedly reduce the need for health care providers and endocrinologists long term and will result in improved glucose control and a reduction is complications related to poor control.

Published data on the ability to improve glucose control with use of telehealth have been both positive and negative(3;6;7;8;9-12;13-17;17-25) and the conclusions regarding clinical utility controversial. However, a recent meta-analysis identified the key features necessary for statistically significant improvement in HbA1C level and other parameters. The greater the number of the seven key components implemented, the greater the success in lowering HbA1C. The 7 key elements include (1) Patient education (2) provider education (3) structured SMBG-self monitoring blood glucose (4) the establishment of specific SMBG goals (5) diet and exercise data feedback (6) glucose data used to modify treatment (7) interactive communication or shared decision making. Each of these components will be built into the proposed study.

The overall goal of this study will be to demonstrate that a 6 month, intensive self-management program using telehealth methodology will result in long term improvement of glucose control and ultimately, in rate of complications this will involve. Several barriers to achievement of these goals, based on published data, are anticipated which influenced protocol design. These include: lack of a home computer or smartphone; lack of knowledge, use and /or training regarding computers, cost of glucose test strips, problems with reimbursement , problems with communication with electronic medical records , and , effective rapid feedback mechanisms.

Our Preliminary data: The Telcare meter, cloud based glucose monitoring system has been beta tested by the project team and selected for use. Initially, the availability of several smartphone, Bluetooth-based electronic data transfer systems were evaluated and rejected on the basis of complexity. Later the Livongo system was evaluated and I t was found that the provider T-moble had limited to no coverage in the area served by the patient population ( i.e. patients served by the Tri-Area and Bland County Community Health systems. A geographical drive through survey of the area revealed that the Verizon signal, assessed on a Samsung Smartphone, appeared to have adequate signal. Two nurses then utilized the Verizon based Telcare system and documented its utility in the clinic and catchment areas. Finally a test patient has utilized the Telcare system for 2 months with use of the 7, 14, 21, and 30 day record software and demonstrated ease of use, transmission of data to EPIC (The EMR used by the university of Virginia-UVA) and feedback adjustment of insulin doses. On these bases, the Telcare system has been chosen for utilization. The glucommander -outpatient algorithm is directly linked to the Telcare dashboard and will make recommendations about adjustment of insulin doses. The dose changes will be overseen by the endocrinologist.

The diabetes tele- educational segment of this project has been developed and conducted by the Virginia Center for Diabetes Prevention and Education (VCDPE) at UVA over the past ten years as part of a State of Virginia funded project. As detailed in the appendix, this consists of educational classes delivered to rural sites ( Federally Qualified Health Center and local health departments) via-teleconferencing with components of didactic material presentations and question and answer periods. These diabetes tele-educational classes will be used with minor modification in this project.

Unique nature of program: This program is designed to meet a unique need of patients in underserved areas where there is no access to Endocrinologist, Certified Diabetes Educators, or sufficient number of health care providers to meet the needs of the increasing number of patients with diabetes. Use of cloud based-glucose monitoring and telemedicine will prove an efficient means of managing patients at reduced cost and greater impact Rationale: There are an insufficient number of Endocrinologists Certified Diabetes Educators, and sufficiently trained health care providers to facilitate excellent glucose control in patients with insulin requiring diabetes mellitus in rural areas. A key solution to this problem is to intensively educate patients in diabetes self-management in order for them to manage their own disease with markedly reduced professional input. Greater self-management will thus reduce requirements for health care provider input and overall costs long term. Patients with diabetes in underserved areas have limited access to health care providers and would benefit from an intensive, diabetes self-management program based on practical, easy to use, cloud-based internet technology with Endocrinologist -driven therapeutic recommendations combined with diabetes tele-education.

Hypothesis: This study is based on the hypothesis that automatic recording and cloud-based data transfer and analysis of glucose data with peer based monitoring will markedly enhance compliance and improve glucose control. Manual rather than automatic recording of glucose logs and submission to health care providers is associated with poor compliance and thus inferior outcomes. Finally an intensive program of education in self- management will result in better long term glucose control and reduction of complications.

Elimination/amelioration of published barriers: Several steps will be taken to ameliorate the published barriers describe above. Patients will be provided with Telcare meters and glucose test strips. These meters uplink glucose data automatically to the cloud via Verizon and eliminate the need for smartphones and home computers. Use of the University of Virginia Telemedicine Office and telephone dictation capability will allow clinical and lab data to be entered into the EPIC electronic medical record. Weekly or fortnightly telephone and telemedicine encounters will be utilized for rapid feedback. The Federally funded Community Heath Center programs will assist with financial difficulties.

Description of Study

Study sites:

• Federally Funded Community Health Centers programs -Tri-Area Community Health Center ( Laurel Fork, Floyd, and Ferrum)

Technology Components: Several aspects of technology will facilitate success of this program. These include

• Internet based glucose monitoring program utilizing the Telcare meter, glucose strips, online data processing, and glucommander-outpatient dashboard capability.
• Telemedicine consultation visits with Endocrinologist from UVA with assistance from an onsite Nurse/Diabetes educator who will be with the patient
• Tele- educational sessions
• Vimeo based program description and introduction
• Dex Com continuous glucose monitor ( one week use for baseline information)

Technology Training: Each patient will be evaluated on site by Dr. Richard Santen, the study endocrinologist and at that time, he will supervise certain educational elements with onsite nurse/diabetes educator and patient . These will include:

• Instruction in using the Jabber program to maintain HIPPA confidentiality
• Demonstrate use of the Video Interface
• Viewing of a Vimeo movie describing the full program and all of its elements. Diet: The concept of "meal replacements" for the improvement of the glucose control of patients with diabetes mellitus has been studied previously but never applied to populations of patients in underserved, rural areas. The nutrisystem plan D provides a balanced, diabetic diet with 1800 calories for men and 1200 for women. Both frozen and non-frozen components are shipped each week. A counselor from Nutrisystems is available for coaching and instruction in use of the diet. The PI of this pilot project has beta tested the system and found it to be user friendly and exceedingly well designed with color coding and provision of low calorie, hunger reducing snacks.

Diabetes Self- Management Education:

• Utilization of comprehensive tele-education sessions developed by VCDPE at the University of Virginia ( see appendix for list) .These will be provided by tele-conferencing at monthly intervals and will include educational resources that have been validated and are currently in use.

Inclusion Criteria:

• Persistent poorly controlled diabetes mellitus

• ages 18-75

• renal function at least 50 % of normal or above

• Patient competent to use meter

  o HbA1C > 8.0% continuously for > 1 year with at least 2 HbA1C values Exclusion criteria: ( from VA study 2016 -Crowley et al Telemed and Telehealth )

• active alcohol/substance abuse

• active cancer therapy

• HIV/AIDS

• Organ transplant

• Cirrhosis of liver

  • hearing , speech or cognitive impairment'
  • dementia or psychosis
  • inability to speak English or interact effectively with an interpreter
  • life threatening illness recent cardiovascular event or stroke
  • prior hypoglycemic sezure or coma
  • Refusal to perform self monitoring of glucose
  • Use of subcutaneous insulin infusion pumps
  • > 75years of age or < age 18 Nursing home or extended facility residents

Duration of self-management education component:

• 6 months Number of patients

  • 3 patients per month over a 24 month period or 72 patients in total
  • Primary study endpoints :

• Hb1C at 3 and 6 months

• average blood sugars from records obtained at the end of six months of study

Secondary endpoints:

• Lipid levels
• Weight/BMI
• Normality of comprehensive panel data
• Urine albumin creatinine ratio
• LDL cholesterol

Inclusion Criteria:

Hemoglobin A1C > 8.0 on two occasions renal function at least 50% of normal persistent, poorly controlled diabetes -

Exclusion Criteria:

alcohol/substance abuse, active cancer therapy, HIV/AIDS, organ transplant, cirrhosis of liver, hearing or speech impediment, dementia or psychosis, cannot speak English or interact appropriately with interpreter

-

poorly controlled diabetes mellitus