Factors Associated with Physician Prescribed Low-Value Care: The Case of Advanced Imaging Test

• Main Authors: Chieh-Jen Wu, 吳潔人
• Other Authors: 郭年真
• Format: Others
• Language: zh-TW
• Published: 2019
• Online Access: http://ndltd.ncl.edu.tw/handle/7x2j4c

碩士 === 國立臺灣大學 === 健康政策與管理研究所 === 107 === Background: One of the primary challenges in healthcare management is the elimination of low-value care, which is of limited benefit and can actually cause iatrogenic harm to patients. Taiwan’s National Health Insurance annually spends around 860 million U.S dollars in CT, MRI, and ultrasound. Unfortunately, little is known about the effectiveness of advanced imaging tests and whether the utilization of low-value imaging test is associated with physician characteristics. Objective: This population-based study obtained data from the Taiwan National Health Insurance (NHI) database for the years 2010 - 2014 to characterize the relationship between the use of low-value imaging tests and the characteristics of healthcare providers. Patients who were diagnosed with uncomplicated headache or syncope were included in the analysis. Methods: Three types of imaging tests were defined as low-value: head imaging for uncomplicated headache, head imaging for the evaluation of syncope, and carotid ultrasound imaging for simple syncope. Multilevel logistic regression was used to identify physician characteristics associated with the use of low-value imaging modalities. Physician characteristics included age, gender, specialty, charges for advanced imaging as a proportion of annual reimbursements for disease/symptom claims, and the number of consultations claimed per year. The models were controlled for hospital characteristics and patient characteristics. Results: The utilization of the three low-value diagnostic procedures was as follows: head imaging for uncomplicated headache (8.98%), head imaging for the evaluation of syncope (14.08%), and carotid ultrasound imaging for simple syncope (5.80%). After controlling for patient and hospital factors, our study revealed that the proportion of charges for advanced imaging in annual reimbursements (OR: 1.342-4.690) , gender (OR:0.825), age (OR:1.198-1.410) and Emergency department (OR: OR:2.746-4.127, except for carotid ultrasound imaging for simple syncope) were positively associated with the use of low-value imaging tests. Conclusions: This study revealed that after controlling for patient characteristics and hospital characteristics, the utilization of low-value imaging tests was associated with physician characteristics. The proportion of annual reimbursements attributed to advanced imaging was positively associated with the use of low-value imaging tests, which implies that the utilization of low-value care might be driven by financial incentives. Many interventions could be implemented to limit the use of low-value care, such as the Choosing Wisely scheme advocated in many countries. This study sheds light on important issues regarding the need for interventions aimed at reducing the use of low-value care options.