1 January 2020 in Operations Research
Decision-makers increasingly have access to rich customer-specific data, providing an opportunity to make better, personalized service decisions. For example, in healthcare, doctors can personalize interventions based on a patient’s clinical history; in marketing, companies can target ads based on customer purchase history. However, the increased variety of potentially relevant customer data implies that an individual’s covariates may be high dimensional, which, in turn, poses statistical challenges for learning personalized decision-making policies. In “Online Decision-Making with High-Dimensional Covariates,” H. Bastani and M. Bayati introduce the LASSO Bandit, an adaptive decision-making algorithm that efficiently leverages high-dimensional user covariates by learning sparse models of decision rewards. The authors illustrate the practical relevance of such an approach by evaluating it against a personalized medication dosing problem, finding that the LASSO Bandit outperforms existing bandit methods and physicians in correctly dosing a majority of patients.
1 March 2021 in Management Science
The contextual bandit literature has traditionally focused on algorithms that address the exploration–exploitation tradeoff. In particular, greedy algorithms that exploit current estimates without any exploration may be suboptimal in general. However, exploration-free greedy algorithms are desirable in practical settings where exploration may be costly or unethical (e.g., clinical trials). Surprisingly, we find that a simple greedy algorithm can be rate optimal (achieves asymptotically optimal regret) if there is sufficient randomness in the observed contexts (covariates). We prove that this is always the case for a two-armed bandit under a general class of context distributions that satisfy a condition we term covariate diversity. Furthermore, even absent this condition, we show that a greedy algorithm can be rate optimal with positive probability. Thus, standard bandit algorithms may unnecessarily explore. Motivated by these results, we introduce Greedy-First, a new algorithm that uses only observed contexts and rewards to determine whether to follow a greedy algorithm or to explore. We prove that this algorithm is rate optimal without any additional assumptions on the context distribution or the number of arms. Extensive simulations demonstrate that Greedy-First successfully reduces exploration and outperforms existing (exploration-based) contextual bandit algorithms such as Thompson sampling or upper confidence bound. This paper was accepted by J. George Shanthikumar, big data analytics.
1 March 2019 in Management Science
Recent Medicare legislation seeks to improve patient care quality by financially penalizing providers for hospital-acquired infections (HAIs). However, Medicare cannot directly monitor HAI rates and instead relies on providers accurately self-reporting HAIs in claims to correctly assess penalties. Consequently, the incentives for providers to improve service quality may disappear if providers upcode, i.e., misreport HAIs (possibly unintentionally) in a manner that increases reimbursement or avoids financial penalties. Identifying upcoding in claims data is challenging because of unobservable confounders (e.g., patient risk). We leverage state-level variations in adverse event reporting regulations and instrumental variables to discover contradictions in HAI and present-on-admission (POA) infection reporting rates that are strongly suggestive of upcoding. We conservatively estimate that 10,000 out of 60,000 annual reimbursed claims for POA infections (18.5%) were upcoded HAIs, costing Medicare $200 million. Our findings suggest that self-reported quality metrics are unreliable and, thus, that recent legislation may result in unintended consequences. The online appendix is available at https://doi.org/10.1287/mnsc.2017.2996 . This paper was accepted by Vishal Gaur, operations management.
6 July 2021 in Journal of Medical Internet Research
Background: New technology adoption is common in health care, but it may elicit frustration if end users are not sufficiently considered in their design or trained in their use. These frustrations may contribute to burnout.Objective: This study aimed to evaluate and quantify health care workers’ frustration with technology and its relationship with emotional exhaustion, after controlling for measures of work-life integration that may indicate excessive job demands.Methods: This was a cross-sectional, observational study of health care workers across 31 Michigan hospitals. We used the Safety, Communication, Operational Reliability, and Engagement (SCORE) survey to measure work-life integration and emotional exhaustion among the survey respondents. We used mixed-effects hierarchical linear regression to evaluate the relationship among frustration with technology, other components of work-life integration, and emotional exhaustion, with adjustment for unit and health care worker characteristics.Results: Of 15,505 respondents, 5065 (32.7%) reported that they experienced frustration with technology on at least 3-5 days per week. Frustration with technology was associated with higher scores for the composite Emotional Exhaustion scale (r=0.35, P<.001) and each individual item on the Emotional Exhaustion scale (r=0.29-0.36, P<.001 for all). Each 10-point increase in the frustration with technology score was associated with a 1.2-point increase (95% CI 1.1-1.4) in emotional exhaustion (both measured on 100-point scales), after adjustment for other work-life integration items and unit and health care worker characteristics.Conclusions: This study found that frustration with technology and several other markers of work-life integration are independently associated with emotional exhaustion among health care workers. Frustration with technology is common but not ubiquitous among health care workers, and it is one of several work-life integration factors associated with emotional exhaustion. Minimizing frustration with health care technology may be an effective approach in reducing burnout among health care workers.
1 August 2024 in Management Science
In this paper, we study the design and analysis of experiments conducted on a set of units over multiple time periods in which the starting time of the treatment may vary by unit. The design problem involves selecting an initial treatment time for each unit in order to most precisely estimate both the instantaneous and cumulative effects of the treatment. We first consider nonadaptive experiments, in which all treatment assignment decisions are made prior to the start of the experiment. For this case, we show that the optimization problem is generally NP-hard, and we propose a near-optimal solution. Under this solution, the fraction entering treatment each period is initially low, then high, and finally low again. Next, we study an adaptive experimental design problem, in which both the decision to continue the experiment and treatment assignment decisions are updated after each period’s data are collected. For the adaptive case, we propose a new algorithm, the precision-guided adaptive experiment algorithm, which addresses the challenges at both the design stage and the stage of estimating treatment effects, ensuring valid post-experiment inference, accounting for the adaptive nature of the design. Using realistic settings, we demonstrate that our proposed solutions can reduce the opportunity cost of the experiments by more than 50%, compared with static design benchmarks. This paper was accepted by George Shanthikumar, data science. Funding: S. Athey and G. Imbens were supported by the Office of Naval Research [Grant N00014-19-1-2468]. M. Bayati was supported by the National Science Foundation [Grant CMMI 1554140]. Supplemental Material: The e-companion and data are available at https://doi.org/10.1287/mnsc.2023.4928 .