Although anaesthesiologists strive to avoid hypoxemia during surgery, reliably predicting future intraoperative hypoxemia is not currently possible. Here, we report the development and testing of a machine-learning-based system that, in real time during general anaesthesia, predicts the risk of hypoxemia and provides explanations of the risk factors. The system, which was trained on minute-by-minute data from the electronic medical records of over fifty thousand surgeries, improved the performance of anaesthesiologists when providing interpretable hypoxemia risks and contributing factors. The explanations for the predictions are broadly consistent with the literature and with prior knowledge from anaesthesiologists. Our results suggest that if anaesthesiologists currently anticipate 15% of hypoxemia events, with this system’s assistance they would anticipate 30% of them, a large portion of which may benefit from early intervention because they are associated with modifiable factors. The system can help improve the clinical understanding of hypoxemia risk during anaesthesia care by providing general insights into the exact changes in risk induced by certain patient or procedure characteristics.
This paper tests the assumption that evaluators are biased to positively evaluate high status individuals, irrespective of quality. Using unique data from Major League Baseball umpires' evaluation of pitch quality, which allow us to observe the difference in a pitch's objective quality and in its perceived quality as judged by the umpire, we show that umpires are more likely to over-recognize quality by expanding the strike zone, and less likely to under-recognize quality by missing pitches in the strike zone for high status pitchers. Ambiguity and the pitcher's reputation as a "control pitcher" moderate the effect of status on umpire judgment. Further, we show that umpire errors resulting from status bias lead to actual performance differences for the pitcher and team.Keywords: Status, Bias, Performance Forthcoming in Management Science2 Social scientists have long understood status to be an indicator of hierarchical position and prestige that helps individuals and organizations procure resources and opportunities for advancement (e.g., Whyte 1943;Podolny 2001;Sauder, Lynn, and Podolny 2012). Status markers, like ranking systems in education (Espeland and Sauder 2007) or awards and prizes (Rossman et al. 2010), accentuate quality differences among actors and create greater socio-economic inequality. Past research has advanced our understanding of how status hierarchies emerge and persist (Berger, Rosenholtz, and Zelditch 1980;Webster and Hysom 1998;Ridgeway and Correll 2006), on the psychological rewards of status attainment (e.g., Willer 2009;Anderson et al. 2012), on the association of status with characteristics such as race and gender (Ridgeway 1991), or on status outcomes, such as the accumulation of power, price and wage differentials, and other economic and social benefits (Podolny 1993;Benjamin and Podolny 1999;Thye 2000;Correll et al. 2007;Stuart, Hoang, and Hybels 1999;Bothner, Kim, and Smith 2011;Pearce 2011;Waguespack and Sorenson, 2011).In economic sociology, the classic statement of self-reproducing advantages is Robert Merton's (1968) theory of the Matthew Effect. Underlying the Matthew Effect is an assumption that individuals are biased to positively evaluate high status individuals irrespective of quality, and that this bias, rather than actual quality differences, perpetuates inequality between high status and low status actors. Scholars in status characteristics theory have made the case that this bias results from individuals' expectations and prior beliefs about competence and quality (Anderson et al. 2001;Ridgeway and Correll 2006;Berger et al. 1977;Ridgeway and Berger 1986), and experimental and audit studies have identified status characteristics as a mechanism that underlies discriminatory evaluations of low status individuals (e.g., Correll et al. 2007).Despite the evidence emanating from the experimental research in status characteristics studies, scholars studying the Matthew Effect in real world settings have had a more difficult time demonstrating that status bias is a mech...
One Sentence Summary: We present a new machine learning based system called Prescience that provides interpretable real-time predictions to help anesthesiologists prevent hypoxemia during surgery.Abstract: Hypoxemia causes serious patient harm, and while anesthesiologists strive to avoid hypoxemia during surgery, anesthesiologists are not reliably able to predict which patients will have intraoperative hypoxemia. Using minute by minute EMR data from fifty thousand surgeries we developed and tested a machine learning based system called Prescience that predicts real-time hypoxemia risk and presents an explanation of factors contributing to that risk during general anesthesia. Prescience improved anesthesiologists' performance when providing interpretable hypoxemia risks with contributing factors. The results suggest that if anesthesiologists currently anticipate 15% of events, then with Prescience assistance they could anticipate 30% of events or an estimated additional 2.4 million annually in the US, a large portion of which may be preventable because they are attributable to modifiable factors. The prediction explanations are broadly consistent with the literature and anesthesiologists' prior knowledge. Prescience can also improve clinical understanding of hypoxemia risk during anesthesia by providing general insights into the exact changes in risk induced by certain patient or procedure characteristics. Making predictions of complex medical machine learning models (such as Prescience) interpretable has broad applicability to other data-driven prediction tasks in medicine.peer-reviewed)
Income gains in the top 1 percent are the primary cause for the rapid growth in U.S. inequality since the late 1970s. Managers and executives of firms account for a large proportion of these top earners. Chief executive officers (CEOs), in particular, have seen their compensation increase faster than the growth in firm size. We propose that changes in the macro patterns of the distribution of CEO compensation resulted from a process of diffusion within localized networks, propagating higher pay among corporate executives. We compare three possible explanations for diffusion: director board interlocks, peer groups, and educational networks. The statistical results indicate that corporate director networks facilitate social comparisons that generate the observed pay patterns. Peer and education network effects do not survive a novel endogeneity test that we execute. A key implication is that local diffusion through executive network structures partially explains the changes in macro patterns of income distribution found in the inequality data.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.
hi@scite.ai
10624 S. Eastern Ave., Ste. A-614
Henderson, NV 89052, USA
Copyright © 2024 scite LLC. All rights reserved.
Made with 💙 for researchers
Part of the Research Solutions Family.