To evaluate the effectiveness of two different diabetes self-management approaches (Personal Digital Assistant-based intervention & Chronic Disease Self-Management Program) to reduce health disparities in minority, rural residents, and other underserved populations with type 2 diabetes in Central Texas. We hypothesise that: 1) Racial/ethnic minority patients with T2DM will be found to experience disparities in diabetes self-management treatment protocols and clinical outcomes, which persist even when controlling for age, gender, obesity, and insurance status; 2) Patients with T2DM who reside in more rural areas will be found to experience disparities in diabetes self-management treatment protocols and clinical outcomes as compared to more urban counterparts, controlling for age, gender, race/ethnicity, obesity, and insurance status; 3) The introduction of CSDMP and HIT protocols will improve diabetes-related self management behaviors, reduce HBA1c values, and increase quality of life in persons with T2DM as compared to controls. A combined intervention approach will result in the greatest reductions; 4) Health improvements following the introduction of CDSMP, HIT or CDSMP/HIT protocols in persons with T2DM compared to controls will be more marked in racial/ethnic minority patients and those patients residing in rural areas; 5) The introduction of self-management interventions will be cost-effective in reducing HbA1c values over time, and associated health care utilization including overall reduction in ER and acute care hospital admissions; 6) Although there is little prior research in this area to guide specific hypotheses, we hypothesize that, overall, there will be no significant cost-effective differential in CDSMP as compared to HIT approaches, although the cost-effective ratio may be stronger in particular subpopulations. The combined approach will have higher costs, but is also anticipated to have a higher cost-benefit ratio for minority populations; 7) The majority of clinicians will be willing to let their patients enroll in the study and will reinforce intervention protocols; and 8) These interventions can be embedded into existing health care structures. At the end of the study, Scott and White will institutionalize cost-effective treatment protocols.

Despite concerted federal and state attempts to reduce health disparities over the past decades substantial disparities in reported rates of chronic disease for minorities still exist. In particular, African Americans and Hispanics experience higher rates of Type 2 diabetes (T2DM), and cardiovascular disease (CVD) than do other segments of the U.S. population. The objectives of this proposed research project are to test two different diabetes self-management (DSM) programs in a large multi-site health care organization in Central Texas that serves large populations of minority and rural residents, comparing outcomes in order to evaluate their efficacy for reducing health disparities. Our specific aims are to: 1) document the nature and magnitude of extant health disparities in diabetes treatment processes and outcomes; 2) evaluate different DSM intervention approaches on behavioral and clinical outcomes, with attention to differential effects by patient and environmental characteristics; 3) examine the cost-effectiveness of these different approaches to DSM education in minority and rural populations; and 4) explore the reach of our intervention efforts and the broader organizational impacts of DSM education, including feedback loops to clinicians and organizational receptivity to self-management approaches. Our study will employ four different activities: 1) an initial electronic chart review of 1300 records of adults; 2) a 2 by 2 open 24 month randomized clinical trial of behaviorally and technologically based DSM interventions with 400 adults age 21 and older who have type 2 diabetes (T2DM); 3) a cost-effectiveness analysis of the different treatment approaches; and 4) surveys of primary care providers and health care administrators. While our primary outcome will be reductions in hemoglobin A1c (HbA1c), our conceptual model includes clinical, behavioral, economic and organizational outcomes. We will also assess the extent to which our interventions reduce health disparities by examining differential treatment success. This study is innovative in its comparison of both behavioral and technological intervention approaches, its attention to the public health impact and cost-effectiveness of different intervention approaches, and its concern with organizational responses to intervention sustainability. A noteworthy significance will be the strengthening of the linkages between clinical and community treatment approaches and the identification of successful treatment strategies in different settings and populations.

Inclusion Criteria:

• Patients with T2DM, including those who require insulin therapy, aged >18 years (eliminates the need to obtain assent for minors who are also dependent on their parents).
• Last measured HbA1c value of > 7.5% (this study hopes to show an improvement in the control of patient's diabetes, and not focused on patients who already show evidence of good disease control).
• Willingness and ability to attend one initial research visit and semi-annual routine follow-up visits over a 24-month period. The follow-up visits include height, weight, and blood pressure measurement and a survey. Surveys may be conducted by phone interview or mail when a follow-up visit can not be scheduled.
• Ability to read, write, and speak English at least at a grade 8 level so as to be able to engage in self-monitoring and use the commercial diabetes management software program (Diabetes Pilot), which is available only in English. For those with lower-literacy, assistance in filling out forms and understanding required intervention protocols will be provided, and use of a "buddy" will be recommended.

Exclusion Criteria:

• Not willing to sign an informed consent or be randomized to any of the four treatment/control groups, (we want to minimize any upfront treatment biases, while adhering to human subject protocols).
• Currently, documented severe alcoholism or drug abuse that is < 6 months ago (concerns that this problem is likely to significantly affect their ability and likelihood to comply with the study requirements over the course of the 24 months).
• Female patients who are pregnant or planning to become pregnant within 12 months (in pregnancy, type 2 diabetes is managed in a completely different manner than in non-pregnant patients).

Differential dropout across interventions. Failure to obtain 50% minority and 50% non-minority participants, preventing further analyses regarding race/ethnicity differences in outcome. Could only provide information in an exploratory manner.