Impact of diabetes quality assurance program on quality of diabetes care in primary care clinics

Abstract

Background: To improve the quality of diabetes care the Health Insurance Review and Assessment Service (HIRA) adopted the Diabetes Quality Assurance(DQA) program in 2011. The DQA program evaluates three components; continuity of care, prescription accuracy, and completion of examinations. Primary care clinics rated for high quality by the DQA program are reported to the public and receive additional payments. In this study, we evaluated the effect of the DQA program on DQA indicators and health outcomes associated with diabetes in primary care clinics. Methods: The National Health Insurance Service National Sample Cohort 2002 to 2015 (NHIS-NSC 2002–2015) database was used. Diabetic Patients with newly prescribed hypoglycemic medications with diagnostic codes from January 2008 to December 2010 and January 2013to December 2015 were included in this study. Patients with diabetes who mainly visited primary care clinics were defined as cases and patients who mainly visited general hospitals were defined as controls. Difference in differences analysis was used to investigate the impact of the DQA program on the quality of diabetes care. Results: A total of 16,232 cases and 8,669 controls were included in the study population. Two indicators of continuity of care, visit every quarter rate (OR 1.086, p<0.0001) and medication rate (OR 1.091, p<0.0001), were improved. However, prescription accuracy did not significantly change over period and institution type. The odds ratios for completion of examinations, HbA1c test rate (OR 1.175, p<0.0001), cholesterol profile test rate (OR 1.043, p<0.0001), and fundoscopy (OR 1.134, p<0.0001), were increased. There was no significant difference between diabetic nephropathy (OR 1.127, p=0.215) and neuropathy (OR 1.181, p=0.073), microvascular complications of diabetes, but diabetic retinopathy (OR 1.497, p<0.0001) significantly increased. In the subgroup analysis the odds ratios for diabetic retinopathy significantly increased in all groups, except for the group that lived in rural areas. Conclusion: The DQA program showed different effects for DQA indicators. It had a positive effect on the completion of examination indicators and the continuity of care indicators. Moreover, it had no significant positive effects on prescription accuracy indicators. The DQA program showed different effects on health outcome variables. Myocardial infarction, stroke, angina pectoris and diabetic retinopathy significantly increased considering period and institution type. Therefore, to reduce the occurrence of complications, it is desirable to include levels such as HbA1c, blood pressure, and lipids, which are intermediate outcome indicators, as indicators of the DQA program.