Home-based Prediabetes Care in Acoma Pueblo - Study 1

Overview

The prevalence of overweight, obesity and T2D among American Indians (AIs) in the Albuquerque service area of Indian Health Services (IHS) serving all pueblos in New Mexico including Pueblo of Acoma is disproportionately elevated. Specifically, among AI, adults age 18 and over in the IHS Albuquerque Service Area, 51% have a BMI \> 30, which significantly exceeds the rate of obesity observed nationally (35.3%) and the Healthy People 2020 target of 30.5%. Likewise, the rate of T2D (22.8%) among the adult AI population in our Albuquerque service area is almost double the rate of the U.S. adult population (12.2%), and the age-adjusted diabetes mortality rate for AIs was 104.7 per 100,000 compared to 23.1 per 100,000 among non-Hispanic Whites in the region. At the same time, the median age of diagnosis of T2D among AI adults was much younger (42.2 years) than the national average (53.8 years). Our major goal of implementing educational interventions to slow the current rate of increase in diabetes in Native communities is aligned with NIH's (NIGMS) and NM INBRE's vision in reducing health disparity using innovative interventions. The investigators propose following aims: Aim 1: Recruit and Screen 300 community members in Acoma Pueblo, NM to identify incident cases of pre-diabetes for the proposed study of Home Based Diabetes Care (HBDC); Aim 2: Enroll 150 Acoma Natives aged 21-70 years, at risk for T2D (i.e., overweight, obese, and/or with at least one affected first degree relative or a history of gestational DM) and conduct HBDC for a 16-week lifestyle intervention in a longitudinal cohort study. Randomize household in a 1:1 allocation to enter either the intervention arm immediately or after a 12-month waiting list in control arm. Control participants will be treated with usual care. Participants randomized to the waiting list will enter the intervention group 12 months after entering the study. Both intervention groups will be followed longitudinally for total of 12 months. Compared with people who will receive "usual care (control group)", prediabetic participants receiving 4 months of the HBDC will exhibit improved risk factor profiles for diabetes, obesity and heart disease, improved Patient Activation Measures, improved adherence with medical treatment, and improved Quality of Life scores.

The investigators are conducting a prospective, randomized clinical trial of 'Home Base Diabetes Care (HBDC)' versus usual care (delayed HBDC beginning 12 months after enrollment) among 150 Acoma citizens with prediabetes from approximately 60 households. The household will serve as the unit of randomization to prevent cross-contamination of the study groups. The investigators hope to demonstrate that HBDC will result in improved patient activation and adherence, improved health-related quality of life (HRQOL), and ultimately, improved diabetes-related health outcomes. Standard, usual care is an appropriate comparison group because it is widely available in Acoma through IHS and because there are a number of other existing resources in Acoma that participants can access to achieve their healthcare goals if they choose to do so. Improved Patient Activation may drive increased utilization of these existing resources. The study will be non-blinded. This is a 5-year study with screening and rolling enrollment beginning during the first year. Participants randomized to the HBDC group will receive 16 weeks of the intervention and will subsequently have one additional data collection visit 12 months after the intervention (total participation approximately 16 months). Participants randomized to the usual care group (delayed HBDC) will begin the HBDC intervention at 12 months. The intervention will be 16 weeks and a final data collection visit will be conducted 12 months after the intervention (total participation approximately 28 months). During the HBDC intervention, both groups will receive 8 home visits over 16 weeks (every other week) and 3 text messages per week. Data analysis and dissemination will be conducted during the final year of the study. The investigators will use community health representatives (CHRs) in collaboration with licensed personnel at the University of New Mexico Health Sciences Center (UNM HSC) to provide a comprehensive home-based intervention to prevent diabetes and improve health through counseling around lifestyle interventions such as diet and exercise. CHRs will make regularly scheduled home visits and they will have a lap top computer with internet access and telemedicine contact with personnel at UNM HSC (Dr. Bouchonville, in his role as Medical Director/Facilitator of the weekly Endo ECHO sessions) for ongoing review of care plans, and management of prediabetes, hypertension, dyslipidemia, and kidney disease according to current American Diabetes Association guidelines. In a collaborative manner, the participant and healthcare team will set appropriately realistic treatment goals. Professional personnel from UNM Clinical and Translational Science Center (e.g., physicians and bio-nutritionists) and HSC (Endo ECHO) will serve as consultants to the primary dyad. All subjects randomized to the HBDC arm will be visited by a CHR in their home at least every two weeks for the duration of the 16 weeks intervention. All enrolled members of a household unit will attend these visits. Given 25-35 households per study arm (comprising 75 participants each) and two full time CHRs working on the study, each CHR will be responsible for approximately 15 households and up to 30 home visits per month, although this latter estimate is high given that rolling enrollment will occur over a 2-3 year period. Each visit will last one hour and will cover curriculum materials agreed upon by the study team and prioritized by the participants. Other members of the household who do not have prediabetes or who are not enrolled in the study will also be allowed to participate in the sessions if they wish, and all subjects will be encouraged to keep their IHS appointments. Primary outcome - To demonstrate that prediabetes risk profiles will improve with HBDC as compared to usual care: To evaluate the effectiveness of the HBDC intervention, the investigators will assess between-group changes in the T2D risk profile of study participants using change in Body Mass Index and A1c as our primary outcome parameters. To reproduce real-life community healthcare conditions, A1c will be assessed in both study groups with POC as an indicator of 90 day glucose homeostasis. BMI will be determined using a SECA 876 Floor Scale (SECA Inc., Chino, CA), and height will be assessed using SECA 217 Portable Stadiometer with floor scale adapter. Study participants in the intervention arm will be counseled to lose 5% of their body weight over the course of the 12-month study. This is a reasonable weight-loss goal based on the following: (1) in our previous extended family-based studies, 45.9% were obese, and all of the randomized participants will have BMI \>27 kg/m2; (2) in our previous study, participants in the lifestyle intervention successfully met weight goals over a relatively short period of time; and (3) eliminating or expending enough energy to burn 500 calories/day, while making no other changes, will result in a weight loss of up to 0.5 kg per week. Biological variables include: (a) T2D incidence determined at each outcome assessment according to the criteria established by the American Diabetes Association (FPG ≥ 126 mg/dL and/or A1c \> 6.4%); (b) Waist-to-hip ratio following the protocol used in the Diabetes Prevention Program; (c) Fasting blood glucose (i-STAT system, Abbott Laboratories, Princeton, NJ); (d) Blood pressure determined using 2 measures after 5 minutes of sitting using an electronic implement (Prestige Medical, Northridge CA); and (e) Fasting lipid panel (HDL, LDL, total cholesterol, and triglycerides) at baseline, 16 weeks, and 16 months (HBDC group) and baseline, 12 months, 16 months, and 28 months (Usual Care/delayed HBDC group). Dietary change will be assessed by comparing the results of a food frequency questionnaire (FFQ) from baseline to month 12 and at post-intervention follow-up. The study will use a FFQ developed for and validated in American Indians. Participants in the intervention arm will be asked to complete a three-day food record once per month over the three days before the next home visit. Weekly activity logs, as well as pedometer data, will be collected over a one week period prior to assessment at 0, 8, and 16 weeks. Secondary outcomes: The PAM is the secondary outcome variable of Aim 2, and a PAM score will be obtained at 0, 6, and 12 weeks and at 12 month follow-up in all participants. In all cases, data will be collected within a two week window of the target date, except for the 12 month post-intervention time point, which will occur within a one month window of the 12 month target date. The investigators will use a short version of PAM using a 13-item survey tool designed to assess a person's knowledge, skill, and confidence in managing his or her own healthcare. The response options for the 13 PAM questions use a categorical agreement scale with 4 response options: (i) strongly disagree, (ii) disagree, (iii) agree, (iv) strongly agree, and N/A. The raw score will be calculated by adding responses to the 13 questions. If all questions are answered (i.e., no "N/A" is used), the range of raw scores would be 13 to 52. If there is at least 1 item with a response of N/A, the total score will be divided by the number of items completed and multiplied by 13 to yield a normalized raw score. A nomogram provided by Dr. Hibbard converts raw scores to an "activation score," ranging from 0 to 100. Activation score cut points derived from Dr. Hibbard's preliminary data will assign participants into 1 of 4 stages of progressive activation: (i) believes an active role is important; (ii) has the confidence and knowledge to take action; (iii) is taking action; and (iv) is able to maintain course under stress. The survey has been fully developed and validated. Higher PAM scores are associated with healthy behaviors, a higher likelihood of practicing self-management behaviors, and higher medication adherence. Patients scoring at the lower end of the activation scale may believe that the physician is the one to "fix" them, while mid-range scores indicate that patients recognize that they must be involved in their care but lack the knowledge to do so effectively. Patients with high PAM scores possess confidence in their ability to self-manage and make healthy changes. Thus, progression in PAM score is an ideal indicator to reflect improvement in health literacy, self-efficacy, and self-determination among at-risk, health-disparate peoples such as the Acoma. In Aim 2, the investigators will demonstrate that participants in the HBDC group will advance at least one level in activation as compared to participants in the control group. Adherence to treatment is another secondary outcome of Aim 2. Although adherence can be difficult to measure in a community based study such as this, the investigators will employ a comprehensive approach to informing patients about the importance of their treatment plans and to assessing adherence by combining education with behavior modification strategies (24). Indicators of adherence will include pill counts, process-oriented information (refill records), sessions attended, appointments kept, and diet and activity log books kept. For the purposes of data analysis, the investigators have devised a simple three-point system for determining adherence. Participants will receive one point for each of three benchmarks achieved: (a) \>80% of prescribed medications taken over the past 30 days (medications counted will include anti-hypertensive medications, lipid-lowering medications, and glucose lowering medications); (b) \>80% of all health-related appointments kept during the past six months; and (c) 100% of activity log books kept over the past two weeks. Data on "appointments kept" will include a composite of scheduled IHS clinic appointments.