Dishonesty in Healthcare Practice : A Behavioral Experiment on Upcoding in Neonatology

We introduce a controlled behavioral experiment framed in a neonatal care context to analyze the effect of introducing a random audit and fines on individuals' honesty in a simple reporting task. Our behavioral data provide new evidence on dishonesty and upcoding in health care. We find that introducing audits combined with a fine significantly reduces dishonesty on aggregate. The effect is driven by a significant reduction in upcoding. At the same time, dishonest choices that cannot be detected as fraudulent by an audit (partial dishonesty) increase. We also find evidence that individual characteristics such as gender, medical background, and integrity are related to dishonest behavior