Internet Intervention to Improve Rural Diabetes Care

• A1c [ Time Frame: January 2005-present ] [ Designated as safety issue: No ]
  
• Blood Pressure [ Time Frame: January 2005-present ] [ Designated as safety issue: No ]
  
• Lipids [ Time Frame: January 2005-present ] [ Designated as safety issue: No ]
  

• Eye screening [ Time Frame: January 2005-present ] [ Designated as safety issue: No ]
  
• Foot exam [ Time Frame: January 2005-present ] [ Designated as safety issue: No ]
  
• Kidney disease monitored [ Time Frame: January 2005-present ] [ Designated as safety issue: No ]
  
• Dietary or exercise advice [ Time Frame: January 2005-present ] [ Designated as safety issue: No ]
  
• Smoking cessation advice [ Time Frame: January 2005-present ] [ Designated as safety issue: No ]
  

This 4-year randomized trial of an Internet-based intervention aims to improve guideline adherence by rural physicians caring for adult patients with type 2 diabetes.

Objectives. (1) Assess barriers to implementation of diabetes guidelines and identify solutions through focus groups and case-based vignette surveys; (2) Develop and implement an interactive Internet intervention including individualized physician performance feedback; (3) Evaluate the intervention in a randomized controlled trial; and (4) Examine the sustainability of improved guideline adherence once feedback ceases.

Methods. In partnership with the University of Alabama rural medicine program, we will randomize 200 rural physician offices to an intervention or comparison arm. Our 18-month intervention, customized to the individual physician in real-time, consists of Internet learning modules with case-based education, performance feedback, and benchmarks. The comparison group will receive a text-based, non-interactive Internet posting of publicly available resources. Nurse practitioners and physician assistants from the offices of study physicians may also participate in the Internet modules. Outcomes will be based on previously developed and validated quality measures for diabetes. The intervention will cover screening, diagnosis, treatment, and prevention. Performance feedback will include 10-15 charts per intervention physician. The main analysis, conducted at the physician level, will compare differential improvement in guideline adherence between the study arms. Ancillary analyses will examine the effects of physician characteristics, other providers in the office, and patient characteristics (e.g., comorbidities, ethnicity, gender, age, and socioeconomic status). Multivariable techniques will adjust for repeated measures, clustering of patients within physicians, and multiple providers within a single office.

Study Population. RDOC will enroll 200 physicians from rural Alabama, Mississippi, Georgia, Tennessee, Florida, Arkansas, Kentucky, North Carolina, South Carolina, Missouri, and West Virginia. Participating physicians will be randomized to receive the intervention or to a comparison group, which will receive traditional, text-based continuing medical education (CME). Each participating physician will allow two rounds of medical record abstraction to be conducted in his/her office. Results from the medical record review: (1) be used in the intervention to provide physicians with personal performance feedback and (2) used to examine change in performance for the intervention and comparison physicians as an evaluation of the entire study.

Significance. This study offers a technologically advanced, theory-grounded intervention for improving care of a high-risk, underserved population. With expertise in translating research into practice, rural medicine, behavioral medicine, health informatics, and clinical diabetes, our multidisciplinary team has a proven record of collaboration. This project will produce an evidence-based and replicable intervention that can be sustained in the "real world," and easily modified for other diseases. This project is substantially improved after making important changes recommended in the second review.