Kaiser Permanente Uses AHRQ Automated Phone System to Support Diabetes Self-Management Program

Prevention and Care Management
March 2009

Kaiser Permanente Southern California's regional office and Kaiser's Riverside Medical Center in Riverside, California, are using the product of an AHRQ-funded grant, the Automated Telephone Self-Management System (ATSM), as a support tool for diabetes patients in the self-management of their illness.

The research of AHRQ grantee and University of California San Francisco Professor Dean Schillinger, MD, and his colleagues at San Francisco General Hospital's Center for Vulnerable Populations showed the value of the ATSM model. It was more effective than both the usual care and group medical visits for the delivery of population-based self-management support for diabetes care. Schillinger's research also showed that the ATSM model was particularly effective at reaching individuals with limited literacy and limited English proficiency; it also promoted patient safety.

The Kaiser project, which builds on Schillinger's work, was launched in Riverside in May 2008. The key goal of Kaiser's program is to coach and support diabetes patients in the self-titration of their insulin and oral diabetes medications under a pharmaceutical care protocol.

The program, explains Roger Benton, PhD, Practice Leader for Clinical Operations at Kaiser, is not designed according to the model of traditional health education, in which a patient is given information about his or her disease and condition. Rather, the model used is a behavioral model with accountability—which is concerned with habit formation with respect to key target behaviors through coaching from a live nurse. The idea is to teach patients a set of necessary tasks, encouraging them to adhere through frequent telephone follow-up, and coaching them through the barriers they will inevitably face.

Kaiser patients whose diabetes is poorly controlled are given the option of participating in this telephone program in addition to their regular medical visits. They are placed under the care of a team that includes their physician, a registered nurse, a licensed vocational nurse (LVN), and "Sally," the interactive voice messaging system. The program starts with a face-to-face orientation to the program with a nurse, followed by a live telephone call the first week to address questions and reinforce participation with Sally, and regular call-backs from an LVN or RN in response to data received during Sally's calls.

The program targets three key behaviors by patients: testing blood sugar daily, taking medications as prescribed, and reporting their "numbers" into the Automated Telephone Self-Management System. The three numbers the patients are asked to report to Sally are their morning blood sugar, their evening blood sugar, and their level of confidence in managing their diabetes. This information is then fed into a database that is used by a nurse during live follow-up calls with the patient. A set of clinical criteria determines who gets a call and who makes the call (LVN or RN). Patients can also call, give feedback on their experiences, and ask to speak to a physician or nurse.

The telephone system records data on each patient, providing a record of the progress of their adherence to the program and their success in diabetes control. The program had a 50-percent participation rate in the first week. Participation dropped off to about 25 percent over the first eight weeks, and then reached a plateau at that level that extended beyond the 20-week point. Benton is encouraged by the progress of the one patient out of four who is engaged long-term in the program. An initial analysis shows a reduction of 1.2 percent in Hemoglobin A1c (HbA1c) levels—a substantial improvement.

Benton notes that Kaiser chose to implement an automated telephone system in part because it was relatively inexpensive—a cost of $.50 to $1.00 per weekly automated telephone call compared to about $13 for a call from a nurse. The relatively low cost combined with demonstrated positive health outcomes should make the program attractive to large health plans like Kaiser, which need to consider whether a program can be scaled up to reach its membership. Riverside Medical Center alone treats an estimated 200,000 to 300,000 Kaiser Permanente members, of which about 10 percent have diabetes.

According to Benton, the program has been designed as a quality improvement initiative. This structure allows changes to be made in program design as experience is gained. Benton and his management team have already revised their patient orientation program to make it more focused, less information-concentrated, and easier to grasp. Additional training sessions have also been held for the patient care teams.

Enrollment ended in December 2008, with 333 patients participating. The next step will be to re-target early participants who dropped out of the program, using the new and improved orientation to Sally, a shorter IVR script, and shorter length of program commitment in the hope that participation rates can be improved.

Impact Case Study Identifier: 
AHRQ Product(s): Research
Topics(s): Health Literacy, Patient Safety, Chronic Care
Geographic Location: California
Implementer: Kaiser Permanente Southern California and Riverside Medical Center
Date: 03/01/2009

Schillinger D, Handley M, Wang F, Hammer H. Effects of self-management support on structure, process and outcomes among vulnerable patients with diabetes: a 3-arm practical clinical trial. Diabetes Care 2009; 32:559-566. (HS014864)

Handley M, Shumway M, Schillinger D. Cost-effectiveness of an automated telephone self-management support intervention with nurse care management among vulnerable patients with type-2 diabetes. Annals of Family Medicine 2008; 6:512-518.

Page last reviewed October 2014