Quality of care of Diabetes Quality-based Payment Initiative (DQBPI) in Taiwan- patients' perspectives

Background: A national-wide DQBPI, a disease management intervention, was launched on November 2001 to provide financial incentive to encourage patient-centered comprehensive and coordinated care. Few study exam the quality of care of DQBPI from patient's perspectives.

Objectives: The aim of this study was to assess the quality of care from patients' perspectives who participated in DQBPI.

Methods: A post-test only with control group design was applied and a computer-assisted telephone interview survey (CATI) was conducted in 2004 on diabetic patients who participated in DQBPI. Those who have received comprehensive care (DQBPI group) in a year were compared quality of care with those who were interrupted (Interrupt group) and the brand new participants (control group). The latter were served as control group due to the lacking of telephone numbers of the non-participants. A national representative stratified (by types of providers and regions) random sampling were conducted on the above three groups. The final samples yielded 1,238, 552, and 558 patients respectively (total 2,348). Quality of care were measured by receiving necessary screening tests, satisfaction on diabetes quality of care (average score of 11 items measured by 5 points Likert scale), and compliance on diabetes self-care in a week (average score of 7 items). Multiple linear regression and logistic regression were applied with controlling age, gender, education, marital status, occupation, household income, Charlsons' comorbidity index, self-reported health, patterns of diabetes treatment, and providers' characteristic.

Results: The average age of 2,348 patients were 62 years, about 54% were female; 57% had at least one comorbidity; average self-reported health score was 73; 13% used insulin, 91% took oral medicine. The three groups were comparable in all characteristics except age and patterns of diabetes treatment. After controlling for the confounders mentioned above, DQBPI group and the control group significantly received more screening tests: including eye, blood sugar, blood pressure, lipid, and urine tests compared with Interrupt group. In addition, the patients in DQBPI group were more likely to have regular source of care, performed significantly better in compliance on diabetes self-care than the interrupt group, and perceived better quality of care than the control group.

Conclusion: The findings suggest that DQBPI might provide better quality of care.