Does the pay-for-performance programme reduce the emergency department visits for hypoglycaemia in type 2 diabetic patients?

OBJECTIVE The pay-for-performance (P4P) programme for diabetes care was implemented in Taiwan to promote holistic care for diabetic patients. This study investigated the effect of P4P on the need for emergency care for diabetic hypoglycaemia. RESEARCH DESIGN AND METHODS The Taiwan National Health Insurance Research Database was used to gather nationwide data on patients with new-onset type 2 diabetes in 2001-09. Using the propensity score matching method, diabetes patients in the P4P programme and those not in P4P were matched, resulting in 199 626 people in each group. A Cox proportional hazards model was applied to analyse the risk of requiring emergency care for diabetic hypoglycaemia in P4P patients. RESULTS A total of 5519 patients sought emergency care for diabetic hypoglycaemia. Of them, 2097 (1.97%) were enrolled in regular P4P treatment, 1671 (1.79%) were enrolled in P4P and irregular treatment, whereas 1751 (0.88%) were not. The Cox proportional hazards model revealed that the risk of seeking emergency care for diabetic hypoglycaemia was higher in P4P patients, with an hazard ratio of 1.90 (95% CI: 1.73-2.08) for regular treatment and 1.32 (95% CI: 1.20-1.45) for irregular treatment. Others with a higher risk of diabetic hypoglycaemia included those older than 25 years, those with lower salaries, those living in less urban areas, those with catastrophic illness and those with a higher comorbidity or higher diabetes complication. Those treated in a non-public hospital were at a higher risk of diabetic hypoglycaemia than those in a public hospital. Patients treated in the higher service volume hospital were at a higher risk as well. Emergency visits due to diabetic hypoglycaemia after P4P were significantly higher than those before P4P. CONCLUSION When treating P4P patients, special care should be taken to help them avoid diabetic hypoglycaemia severe enough to require emergency care.