Background-The emergence of the World Wide Web in the last decade has made it feasible for the Internet to be a vehicle for chronic disease management.

The literature on self-management in diabetes consists predominantly of investigations on compliance, behavioral modification techniques, assessment of health beliefs, and cognitive theories. Little systematic research has explored the actual experience of applying and adapting to a diabetes regimen. This qualitative study reports a diabetes self-management model (DSMM) based on interview data from 18 adults with Type I diabetes. As described and confirmed by the study participants, the DSMM is composed of sequential phases and stages that evolve over time and is influenced by four major factors: personal considerations, monitoring activities, specific cognitive skills for diabetes problem solving, and definition of control. This study provides a model that affirms the experience and effort of learning self-management and recognizes the importance of individualized regimens developed from personal experiences and perceptions of what "works for me."

The care for people with diabetes and most chronic illness suffers if it is acute, reactive, and fragmentary. We report the first 5 years of a comprehensive, integrated approach to diabetes care at Group Health Cooperative of Puget Sound, a large group model Health Maintenance Organization in Washington State. The program is population-based, evidence-based, and patient-centered. Primary care teams receive support in the form of electronic diabetes registries, evidence-based guidelines, patient self-management support, and decentralized onsite consultation with a diabetes expert team (a physician and nurse specialist). In a population of more than 15,000 patients with diabetes, by 1998 more than 70% had had a dilated retinal examination and microalbuminuria test in the previous 12 months, 82% had received a foot examination, and 68% of patients have a hemoglobin A 1c (Hb A 1c ) under 8.0%. Patient satisfaction improved while costs and utilization decreased. Overall costs decreased by $62 per member per month, despite a slight increase in pharmacy costs. In conclusion we have shown that an integrated and proactive approach to diabetes care improves health outcomes and patient satisfaction and decreases overall costs of care. 75

The purpose of this study was to investigate the effectiveness of a Web-based intervention on self-reported blood sugars for older adults with diabetes. Participants who were randomized in the first phase to the intervention group (n = 15) aged 60 and older with diabetes accessed a study Web site that provided them access to an on-line library, advice and counseling from a nurse via e-mail, a personal electronic log of self-management activities, and weekly on-line problem-solving group discussions. Longitudinal mixed-effects modeling was used to investigate the effects of the Web-based intervention on blood sugar levels over a 6-month period. Participants who had fewer than six self-reported medical comorbidities (n = 8) at baseline were more likely to experience a linear decline on blood sugar levels than participants reporting more than six comorbidities. A significant interaction appeared among the time of day blood sugar readings were reported, baseline A1c, and the number of comorbidities. Participants who had more than six comorbidities (n = 7) experienced increased blood sugar levels over the study period, specifically with AM fasting and before bedtime reporting times. Participants with fewer than six comorbidities experienced a decline in blood sugar levels during the same testing times. Findings show that a Web-based intervention was not effective in improving blood sugar readings among adults with six or more comorbidities at baseline. Future research needs to further investigate the role comorbidities may play with the outcomes of Web-based interventions.