Mismanaging Diagnostic Accuracy Under Congestion

• Published in: Operations research Vol. 71; no. 3; pp. 895 - 916
• Main Authors: Kremer, Mirko, de Véricourt, Francis
• Format: Journal Article
• Language: English
• Published: Linthicum INFORMS 01.05.2023; Institute for Operations Research and the Management Sciences
• Subjects: Accuracy; Algorithms; Congestion; Data processing; Decision making; diagnostic accuracy; Diagnostic systems; experiments; Medical diagnosis; Operations research; partially observable Markov decision process; path-dependent decision making; Special Issue on Behavioral Queueing Science; System dynamics; task completion bias; Tradeoffs; Traffic congestion; undertesting
• ISSN: 0030-364X; 1526-5463
• DOI: 10.1287/opre.2022.2292

Diagnostic processes are difficult to manage because they require the decision maker (DM) to dynamically balance the benefit of acquiring more diagnostic information against the cost of doing so. When additional and unattended diagnostic tasks build up over time, making this tradeoff becomes especially challenging. In their study “Mismanaging Diagnostic Accuracy Under Congestion,” Kremer and de Véricourt uncover different biases to which DMs are subject when making diagnostic decisions while unattended diagnostic tasks accumulate over time. The authors find that, in their experiments, DMs are overall insufficiently sensitive to congestion. As a result, DMs acquire too little information at low congestion levels, but too much at high levels, compared with an optimal normative benchmark. This in fact increases both the diagnostic errors and congestion levels in the system. The authors disentangle the underlying mechanisms for these effects and suggests different approaches to debias the DMs. To study the effect of congestion on the fundamental tradeoff between diagnostic accuracy and speed, we empirically test the predictions of a formal sequential testing model in a setting where the gathering of additional information can improve diagnostic accuracy but may also take time and increase congestion as a result. The efficient management of such systems requires a careful balance of congestion-sensitive stopping rules. These include diagnoses made based on very little or no diagnostic information and the stopping of diagnostic processes while waiting for information. We test these rules under controlled laboratory conditions and link the observed biases to system dynamics and performance. Our data show that decision makers (DMs) stop diagnostic processes too quickly at low congestion levels where information acquisition is relatively cheap. However, they fail to stop quickly enough when increasing congestion requires the DM to diagnose without testing or diagnose while waiting for test results. Essentially, DMs are insufficiently sensitive to congestion. As a result of these behavioral patterns, DMs manage the system with both lower-than-optimal diagnostic accuracy and higher-than-optimal congestion cost, underperforming on both sides of the accuracy/speed tradeoff. History: This paper has been accepted for the Operations Research Special Issue on Behavioral Queueing Science. Funding: This research was partially funded by the Deutsche Forschungsgemeinschaft [Grant VE 897/4-1]. Supplemental Material: The online appendix is available at https://doi.org/10.1287/opre.2022.2292 .