Association of preadmission insomnia symptoms with objective in-hospital sleep and clinical outcomes among hospitalized patients.

STUDY OBJECTIVES: To determine the prevalence of preadmission insomnia symptoms among hospitalized patients and assess the association of insomnia symptoms with objective in-hospital sleep and clinical outcomes. METHODS: We conducted a prospective cohort study of medicine inpatients (age ≥ 50, no previously diagnosed sleep disorders). Participants answered the Insomnia Severity Index (ISI) questionnaire to assess for preadmission insomnia symptoms (scored 0-28; higher scores suggest more insomnia symptoms). Sleep duration and efficiency were measured with actigraphy. Participants self-reported 30-day postdischarge readmissions and emergency department and/or urgent care visits. RESULTS: Of 568 participants, 49% had ISI scores suggestive of possible undiagnosed insomnia (ISI ≥ 8). Higher ISI scores were associated with shorter sleep duration [ß = -2.6, 95% confidence interval (CI) -4.1 to -1.1, P = .001] and lower sleep efficiency (ß = -0.39, 95% CI -0.63 to -0.15, P = .001). When adjusted for age, sex, body mass index, and comorbidities, higher ISI scores were associated with longer length of stay (incidence rate ratio 1.01, 95% CI 1.00-1.02, P = .011), increased risk of 30-day readmission (odds ratio 1.04, 95% CI 1.01-1.07, P = .018), and increased risk of 30-day emergency department or urgent care visit (odds ratio 1.04, 95% CI 1.00-1.07, P = .043). CONCLUSIONS: Among medicine inpatients, there was a high prevalence of preadmission insomnia symptoms suggestive of possible undiagnosed insomnia. Participants with higher ISI scores slept less with lower sleep efficiency during hospitalization. Higher ISI scores were associated with longer length of stay, increased risk of a 30-day postdischarge readmission, and increased risk of a 30-day postdischarge emergency department or urgent care visit. CITATION: Neborak JM, Press VG, Parker WF, et al. Association of preadmission insomnia symptoms with objective in-hospital sleep and clinical outcomes among hospitalized patients. J Clin Sleep Med. 2024;20(5):681-687.

Clinician Trust in Artificial Intelligence: What is Known and How Trust Can Be Facilitated.

Predictive analytics based on artificial intelligence (AI) offer clinicians the opportunity to leverage big data available in electronic health records (EHR) to improve clinical decision-making, and thus patient outcomes. Despite this, many barriers exist to facilitating trust between clinicians and AI-based tools, limiting its current impact. Potential solutions are available at both the local and national level. It will take a broad and diverse coalition of stakeholders, from health-care systems, EHR vendors, and clinical educators to regulators, researchers and the patient community, to help facilitate this trust so that the promise of AI in health care can be realized.

Scaling up a diagnostic pause at the ICU-to-ward transition: an exploration of barriers and facilitators to implementation of the ICU-PAUSE handoff tool.

OBJECTIVES: The transition from the intensive care unit (ICU) to the medical ward is a high-risk period due to medical complexity, reduced patient monitoring, and diagnostic uncertainty. Standardized handoff practices reduce errors associated with transitions of care, but little work has been done to standardize the ICU to ward handoff. Further, tools that exist do not focus on preventing diagnostic error. Using Human-Centered Design methods we previously created a novel EHR-based ICU-ward handoff tool (ICU-PAUSE) that embeds a diagnostic pause at the time of transfer. This study aims to explore barriers and facilitators to implementing a diagnostic pause at the ICU-to-ward transition. METHODS: This is a multi-center qualitative study of semi-structured interviews with intensivists from ten academic medical centers. Interviews were analyzed iteratively through a grounded theory approach. The Sittig-Singh sociotechnical model was used as a unifying conceptual framework. RESULTS: Across the eight domains of the model, we identified major benefits and barriers to implementation. The embedded pause to address diagnostic uncertainty was recognized as a key benefit. Participants agreed that standardization of verbal and written handoff would decrease variation in communication. The main barriers fell within the domains of workflow, institutional culture, people, and assessment. CONCLUSIONS: This study represents a novel application of the Sittig-Singh model in the assessment of a handoff tool. A unique feature of ICU-PAUSE is the explicit acknowledgement of diagnostic uncertainty, a practice that has been shown to reduce medical error and prevent premature closure. Results will be used to inform future multi-site implementation efforts.

No-show Prediction Model Performance Among People With HIV: External Validation Study.

BACKGROUND: Regular medical care is important for people living with HIV. A no-show predictive model among people with HIV could improve clinical care by allowing providers to proactively engage patients at high risk of missing appointments. Epic, a major provider of electronic medical record systems, created a model that predicts a patient's probability of being a no-show for an outpatient health care appointment; however, this model has not been externally validated in people with HIV. OBJECTIVE: We examined the performance of Epic's no-show model among people with HIV at an academic medical center and assessed whether the performance was impacted by the addition of demographic and HIV clinical information. METHODS: We obtained encounter data from all in-person appointments among people with HIV from January 21 to March 30, 2022, at the University of Chicago Medicine. We compared the predicted no-show probability at the time of the encounter to the actual outcome of these appointments. We also examined the performance of the Epic model among people with HIV for only HIV care appointments in the infectious diseases department. We further compared the no-show model among people with HIV for HIV care appointments to an alternate random forest model we created using a subset of seven readily accessible features used in the Epic model and four additional features related to HIV clinical care or demographics. RESULTS: We identified 674 people with HIV who contributed 1406 total scheduled in-person appointments during the study period. Of those, we identified 331 people with HIV who contributed 440 HIV care appointments. The performance of the Epic model among people with HIV for all appointments in any outpatient clinic had an area under the receiver operating characteristic curve (AUC) of 0.65 (95% CI 0.63-0.66) and for only HIV care appointments had an AUC of 0.63 (95% CI 0.59-0.67). The alternate model we created for people with HIV attending HIV care appointments had an AUC of 0.78 (95% CI 0.75-0.82), a significant improvement over the Epic model restricted to HIV care appointments (P<.001). Features identified as important in the alternate model included lead time, appointment length, HIV viral load >200 copies per mL, lower CD4 T cell counts (both 50 to <200 cells/mm3 and 200 to <350 cells/mm3), and female sex. CONCLUSIONS: For both models among people with HIV, performance was significantly lower than reported by Epic. The improvement in the performance of the alternate model over the proprietary Epic model demonstrates that, among people with HIV, the inclusion of demographic information may enhance the prediction of appointment attendance. The alternate model further reveals that the prediction of appointment attendance in people with HIV can be improved by using HIV clinical information such as CD4 count and HIV viral load test results as features in the model.

Improving Communication in Intensive Care Unit to Ward Transitions: Protocol for Multisite National Implementation of the ICU-PAUSE Handoff Tool.

BACKGROUND: The intensive care unit (ICU)-ward transfer poses a particularly high-risk period for patients. The period after transfer has been associated with adverse events and additional work for care teams related to miscommunication or omission of information. Standardized handoff processes have been found to reduce communication errors and adverse patient events in other clinical environments but are understudied at the ICU-ward interface. We previously developed an electronic ICU-ward transfer tool, ICU-PAUSE, which embeds the key elements and diagnostic reasoning to facilitate a safe transfer of care at ICU discharge. OBJECTIVE: The aim of this study is to evaluate the implementation process of the ICU-PAUSE handoff tool across 10 academic medical centers, including the rate of adoption and acceptability, as perceived by clinical care teams. METHODS: ICU-PAUSE will be implemented in the medical ICU across 10 academic hospitals, with each site customizing the tool to their institution's needs. Our mixed methods study will include a combination of a chart review, quantitative surveys, and qualitative interviews. After a 90-day implementation period, we will conduct a retrospective chart review to evaluate the rate of uptake of ICU-PAUSE. We will also conduct postimplementation surveys of providers to assess perceptions of the tool and its impact on the frequency of communication errors and adverse events during ICU-ward transfers. Lastly, we will conduct semistructured interviews of faculty stakeholders with subsequent thematic analysis with the goal of identifying benefits and barriers in implementing and using ICU-PAUSE. RESULTS: ICU-PAUSE was piloted in the medical ICU at Barnes-Jewish Hospital, the teaching hospital of Washington University School of Medicine in St. Louis, in 2019. As of July 2022, implementation of ICU-PAUSE is ongoing at 6 of 10 participating sites. Our results will be published in 2023. CONCLUSIONS: Our process of ICU-PAUSE implementation embeds each step of template design, uptake, and customization in the needs of users and key stakeholders. Here, we introduce our approach to evaluate its acceptability, usability, and impact on communication errors according to the tenets of sociotechnical theory. We anticipate that ICU-PAUSE will offer an effective handoff tool for the ICU-ward transition that can be generalized to other institutions. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/40918.

Care Quality for Patients with Chronic Obstructive Pulmonary Disease in the Readmission Penalty Era.

Rationale: Chronic obstructive pulmonary disease (COPD) is the fifth-leading cause of admissions and third-leading cause of readmissions among U.S. adults. Recent policies instituted financial penalties for excessive COPD readmissions. Objectives: To evaluate changes in the quality of care for patients hospitalized for COPD after implementation of the Hospital Readmissions Reduction Program (HRRP). Methods: We conducted a retrospective cohort study of patients older than 40 years of age hospitalized for COPD across 995 U.S. hospitals (Premier Healthcare Database). Measurements and Main Results: Quality of care before and after HRRP implementation was measured via adherence to recommended inpatient care treatments for acute exacerbations of COPD (recommended care, nonrecommended care, "ideal care" [all recommended and no nonrecommended care]). We included 662,842 pre-HRRP (January 2010-September 2014) and 285,508 post-HRRP (October 2014-December 2018) admissions. Recommended care increased at a rate of 0.16% per month pre-HRRP and 0.01% per month post-HRRP (P < 0.001). Nonrecommended care decreased at a rate of 0.15% per month pre-HRRP and 0.13% per month post-HRRP. Ideal care increased at a rate of 0.24% per month pre-HRRP and 0.11% per month post-HRRP (P < 0.001). Conclusions: The pre-HRRP trends toward improving care quality for inpatient COPD care slowed after HRRP implementation. This suggests that financial penalties for readmissions did not stimulate higher quality of care for patients hospitalized with COPD. It remains unclear what policies or approaches will be effective to ensure high care quality for patients hospitalized with COPD exacerbations.

Consumption of golden berries (Physalis peruviana L.) might reduce biomarkers of oxidative stress and alter gut permeability in men without changing inflammation status or the gut microbiota.

Golden berry (Physalis peruviana) is a tropical fruit rich in antioxidants that has been proposed to be able to control the lipid profile in hypercholesterolemic patients. Dyslipidemia is an independent risk factor for cardiometabolic diseases. The gut microbiota is strongly associated with cardiometabolic risk and is involved in redox balance, intestinal permeability, and inflammation. However, the impacts of golden berry on some of these factors, including the human gut microbiota, have never been tested, and there are no tools for compliance monitoring or dietary intake assessment regarding nutritional interventions with this fruit. In the pre-post quasi-experimental nutritional intervention presented here, 18 adult men (27-49 years old) consumed golden berries (Dorada variety) for three weeks. We evaluated putative biomarkers of exposure through an untargeted metabolomics approach (liquid chromatography-mass spectrometry LC-MS), quantified the biomarkers of oxidative stress, gut permeability, and inflammation in plasma, and assessed the effects of fruit intake on the gut microbiota through 16S rRNA gene sequencing of feces (Illumina MiSeq V2). First, syringic acid and kaempferol were identified as putative biomarkers of golden berry consumption. Intervention with this fruit promoted physiological changes in the participants after three weeks, reducing the level of the oxidative stress marker 8-isoprostane (-148 pg/ml; 36.1 %; p = 0.057) and slightly altering gut permeability by increasing the plasma levels of LBP (2.91 µg/ml; 54.6 %; p = 0.0005) and I-FABP (0.15, 14.7 %, p = 0.04) without inducing significant inflammation; i.e., the levels of IL-1ß, TNF-α and IL-8 changed by 0.7 (2.0 %), -4.0 (-9.6 %) and -0.4 (-1.8 %) pg/ml, respectively. Notably, the consumption of golden berries did not affect the gut microbiota of the individuals consistently but instead shifted it in a personalized manner. The compositions of the gut microbiota of a given individual at the end of intervention and one month after the end of intervention were statistically more similar to their own baseline than to a corresponding sample from a different individual. This intervention identified putative biomarkers of golden berry intake along with potential benefits of its consumption relevant to cardiometabolic disease risk reduction. Golden berries are likely to positively modulate redox balance, although this effect must be proven in a future controlled clinical trial.

The Evaluation of a Clinical Decision Support Tool Using Natural Language Processing to Screen Hospitalized Adults for Unhealthy Substance Use: Protocol for a Quasi-Experimental Design.

BACKGROUND: Automated and data-driven methods for screening using natural language processing (NLP) and machine learning may replace resource-intensive manual approaches in the usual care of patients hospitalized with conditions related to unhealthy substance use. The rigorous evaluation of tools that use artificial intelligence (AI) is necessary to demonstrate effectiveness before system-wide implementation. An NLP tool to use routinely collected data in the electronic health record was previously validated for diagnostic accuracy in a retrospective study for screening unhealthy substance use. Our next step is a noninferiority design incorporated into a research protocol for clinical implementation with prospective evaluation of clinical effectiveness in a large health system. OBJECTIVE: This study aims to provide a study protocol to evaluate health outcomes and the costs and benefits of an AI-driven automated screener compared to manual human screening for unhealthy substance use. METHODS: A pre-post design is proposed to evaluate 12 months of manual screening followed by 12 months of automated screening across surgical and medical wards at a single medical center. The preintervention period consists of usual care with manual screening by nurses and social workers and referrals to a multidisciplinary Substance Use Intervention Team (SUIT). Facilitated by a NLP pipeline in the postintervention period, clinical notes from the first 24 hours of hospitalization will be processed and scored by a machine learning model, and the SUIT will be similarly alerted to patients who flagged positive for substance misuse. Flowsheets within the electronic health record have been updated to capture rates of interventions for the primary outcome (brief intervention/motivational interviewing, medication-assisted treatment, naloxone dispensing, and referral to outpatient care). Effectiveness in terms of patient outcomes will be determined by noninferior rates of interventions (primary outcome), as well as rates of readmission within 6 months, average time to consult, and discharge rates against medical advice (secondary outcomes) in the postintervention period by a SUIT compared to the preintervention period. A separate analysis will be performed to assess the costs and benefits to the health system by using automated screening. Changes from the pre- to postintervention period will be assessed in covariate-adjusted generalized linear mixed-effects models. RESULTS: The study will begin in September 2022. Monthly data monitoring and Data Safety Monitoring Board reporting are scheduled every 6 months throughout the study period. We anticipate reporting final results by June 2025. CONCLUSIONS: The use of augmented intelligence for clinical decision support is growing with an increasing number of AI tools. We provide a research protocol for prospective evaluation of an automated NLP system for screening unhealthy substance use using a noninferiority design to demonstrate comprehensive screening that may be as effective as manual screening but less costly via automated solutions. TRIAL REGISTRATION: ClinicalTrials.gov NCT03833804; https://clinicaltrials.gov/ct2/show/NCT03833804. INTERNATIONAL REGISTERED REPORT IDENTIFIER (IRRID): DERR1-10.2196/42971.

Using Machine Learning to Predict Likelihood and Cause of Readmission After Hospitalization for Chronic Obstructive Pulmonary Disease Exacerbation.

Background: Chronic obstructive pulmonary disease (COPD) is a leading cause of hospital readmissions. Few existing tools use electronic health record (EHR) data to forecast patients' readmission risk during index hospitalizations. Objective: We used machine learning and in-hospital data to model 90-day risk for and cause of readmission among inpatients with acute exacerbations of COPD (AE-COPD). Design: Retrospective cohort study. Participants: Adult patients admitted for AE-COPD at the University of Chicago Medicine between November 7, 2008 and December 31, 2018 meeting International Classification of Diseases (ICD)-9 or -10 criteria consistent with AE-COPD were included. Methods: Random forest models were fit to predict readmission risk and respiratory-related readmission cause. Predictor variables included demographics, comorbidities, and EHR data from patients' index hospital stays. Models were derived on 70% of observations and validated on a 30% holdout set. Performance of the readmission risk model was compared to that of the HOSPITAL score. Results: Among 3238 patients admitted for AE-COPD, 1103 patients were readmitted within 90 days. Of the readmission causes, 61% (n = 672) were respiratory-related and COPD (n = 452) was the most common. Our readmission risk model had a significantly higher area under the receiver operating characteristic curve (AUROC) (0.69 [0.66, 0.73]) compared to the HOSPITAL score (0.63 [0.59, 0.67]; p = 0.002). The respiratory-related readmission cause model had an AUROC of 0.73 [0.68, 0.79]. Conclusion: Our models improve on current tools by predicting 90-day readmission risk and cause at the time of discharge from index admissions for AE-COPD. These models could be used to identify patients at higher risk of readmission and direct tailored post-discharge transition of care interventions that lower readmission risk.

Cocreating the ICU-PAUSE Tool for Intensive Care Unit-Ward Transitions.

Background: Intensive care unit (ICU)-ward patient transfers are inherently high risk, and clinician miscommunication has been linked to adverse events and negative outcomes. Despite these risks, few educational tools exist to improve resident handoff communication at ICU-ward transfer. Objective: We used human-centered design (HCD) methods to cocreate a novel electronic health record ICU-ward transfer tool alongside Internal Medicine residents at three academic hospitals. Methods: We conducted HCD workshops at each hospital, performing process mapping, brainstorming, and rapid prototyping. We performed thematic analysis on verbatim-transcribed workshop audio recordings to inform development and adaptation of the final resident prototype into the ICU-PAUSE tool. Results: ICU-PAUSE focuses on reasons for ICU admission and problem-based ICU course (I); Code status, goals of care, and family contacts (C); a diagnostic pause acknowledging Uncertainty (U); Pending tests (P); Active consultants (A); high-risk medications, including medications to be Unprescribed (U); Summary of problems and to-dos (S); and a current physical Exam (E). Conclusion: We used HCD to cocreate a novel, more user-friendly electronic ICU-ward transfer tool, ICU-PAUSE, alongside Internal Medicine trainees. Future steps will involve formal usability testing, evidence-driven implementation, and clinical evaluation of ICU-PAUSE across multiple hospitals.

Analysis and comparison of image processing and artificial intelligence algorithms to detect AFB in pulmonary tuberculosis images.

Pulmonary tuberculosis (TB) is one of the top 10 causes of death worldwide caused by an infection. TB is curable with an adequate diagnosis, normally performed through bacilloscopies. Automate TB diagnosis implies bacilli detection and counting usually based on smear images processing and artificial intelligence. Works reported in the literature usually consider images with similar coloring characteristics, which are difficult to obtain due to the Ziehl - Neelsen staining method variations (excess or deficiency of coloration), provoking errors in the bacilli segmentation. This paper presents an image preprocessing technique, based on simple, fast and well-known processing techniques, to improve and standardize the contrast in the Acid-Fast Bacilli (AFB) images used to diagnose TB; these techniques are used previously to the segmentation stage to obtain accurate results. The results are validated with and without the preprocessing stage by the Jaccard index, pixel detection accuracy and UAC obtained in an Artificial Neural Network (ANN) and a Bayesian classifier with Gaussian mixture model (GMM). Obtained results indicate that the proposed approach can be applied to automate the Tuberculosis diagnostic.

Framework for Integrating Equity Into Machine Learning Models: A Case Study.

Predictive analytic models leveraging machine learning methods increasingly have become vital to health care organizations hoping to improve clinical outcomes and the efficiency of care delivery for all patients. Unfortunately, predictive models could harm populations that have experienced interpersonal, institutional, and structural biases. Models learn from historically collected data that could be biased. In addition, bias impacts a model's development, application, and interpretation. We present a strategy to evaluate for and mitigate biases in machine learning models that potentially could create harm. We recommend analyzing for disparities between less and more socially advantaged populations across model performance metrics (eg, accuracy, positive predictive value), patient outcomes, and resource allocation and then identify root causes of the disparities (eg, biased data, interpretation) and brainstorm solutions to address the disparities. This strategy follows the lifecycle of machine learning models in health care, namely, identifying the clinical problem, model design, data collection, model training, model validation, model deployment, and monitoring after deployment. To illustrate this approach, we use a hypothetical case of a health system developing and deploying a machine learning model to predict the risk of mortality in 6 months for patients admitted to the hospital to target a hospital's delivery of palliative care services to those with the highest mortality risk. The core ethical concepts of equity and transparency guide our proposed framework to help ensure the safe and effective use of predictive algorithms in health care to help everyone achieve their best possible health.

Integrating Physicians Into Lean Quality Improvement Through a Structured Educational Program: The ECLIPSE Program.

In the increasingly complex health care system, physicians require skills and knowledge to participate with multidisciplinary team members in quality improvement (QI) that adds value to health care organizations. The Educational and Clinical Leaders Improving Performance with Structured E3L training (ECLIPSE) program was developed to address this challenge. Clinically relevant components of lean management were leveraged to create an online, flipped-classroom curriculum, and this was paired with Kaizen adapted specifically for physicians and multidisciplinary clinicians to promote experiential skills utilization. The focus of each adapted Kaizen was a topic of institutional QI priority, such as improving patient throughput or reducing readmission rates. Participants were awarded certification in the E3 Leadership management system-a patient-centered, equity-focused system based on lean principles. After 4 years, 50 E3 Leadership certificates were awarded to multidisciplinary clinicians, including 30 to physicians; participants scored an average 85% on module quizzes. The ECLIPSE program has improved physician participation in multidisciplinary QI projects with institutional alignment.

Zooming into Focus Groups: Strategies for Qualitative Research in the Era of Social Distancing.

Qualitative research methods are important and have become increasingly prominent in medical education and research. The reason is simple: many pressing questions in these fields require qualitative approaches to elicit nuanced insights and additional meaning beyond standard quantitative measurements in surveys or observatons. Among the most common qualitative data collection methods are structured or semistructured in-person interviews and focus groups, in which participants describe their experiences relevant to the research question at hand. In the era of physical and social distancing because of the novel coronavirus disease (COVID-19) pandemic, little guidance exists for strategies for conducting focus groups or semistructured interviews. Here we describe our experience with, and recommendations for, conducting remote focus groups and/or interviews in the era of social distancing. Specifically, we discuss best practice recommendations for researchers using video teleconferencing programs to continue qualitative research during the COVID-19 pandemic.

Variation in Ventilator Allocation Guidelines by US State During the Coronavirus Disease 2019 Pandemic: A Systematic Review.

Importance: During the coronavirus disease 2019 pandemic, there may be too few ventilators to meet medical demands. It is unknown how many US states have ventilator allocation guidelines and how these state guidelines compare with one another. Objective: To evaluate the number of publicly available US state guidelines for ventilator allocation and the variation in state recommendations for how ventilator allocation decisions should occur and to assess whether unique criteria exist for pediatric patients. Evidence Review: This systematic review evaluated publicly available guidelines about ventilator allocation for all states in the US and in the District of Columbia using department of health websites for each state and internet searches. Documents with any discussion of a process to triage mechanical ventilatory support during a public health emergency were screened for inclusion. Articles were excluded if they did not include specific ventilator allocation recommendations, were in draft status, did not include their state department of health, or were not the most up-to-date guideline. All documents were individually assessed and reassessed by 2 independent reviewers from March 30 to April 2 and May 8 to 10, 2020. Findings: As of May 10, 2020, 26 states had publicly available ventilator guidelines, and 14 states had pediatric guidelines. Use of the Sequential Organ Failure Assessment score in the initial rank of adult patients was recommended in 15 state guidelines (58%), and assessment of limited life expectancy from underlying conditions or comorbidities was included in 6 state guidelines (23%). Priority was recommended for specific groups in the initial evaluation of patients in 6 states (23%) (ie, Illinois, Maryland, Massachusetts, Michigan, Pennsylvania, and Utah). Many states recommended exclusion criteria in adult (11 of 26 states [42%]) and pediatric (10 of 14 states [71%]) ventilator allocation. Withdrawal of mechanical ventilation from a patient to give to another if a shortage occurs was discussed in 22 of 26 adult guidelines (85%) and 9 of 14 pediatric guidelines (64%). Conclusions and Relevance: These findings suggest that although allocation guidelines for mechanical ventilatory support are essential in a public health emergency, only 26 US states provided public guidance on how this allocation should occur. Guidelines among states, including adjacent states, varied significantly and could cause inequity in the allocation of mechanical ventilatory support during a public health emergency, such as the coronavirus disease 2019 pandemic.

Accuracy of Clinicians' Ability to Predict the Need for Intensive Care Unit Readmission.

Rationale: Determining when an intensive care unit (ICU) patient is ready for discharge to the ward is a complex daily challenge for any ICU care team. Patients who experience unplanned readmissions to the ICU have increased mortality, length of stay, and cost compared with those not readmitted during their hospital stay. The accuracy of clinician prediction for ICU readmission is unknown.Objectives: To determine the accuracy of ICU physicians and nurses for predicting ICU readmissionsMethods: We conducted a prospective study in the medical ICU of an academic hospital from October 2015 to September 2017. After daily rounding for patients being transferred to the ward, ICU clinicians (nurses, residents, fellows, and attendings) were asked to report the likelihood of readmission within 48 hours (using a 1-10 scale, with 10 being "extremely likely"). The accuracy of the clinician prediction score (1-10) was assessed for all clinicians and by clinician type using sensitivity, specificity, and area under the curve (AUC) for the receiver operating characteristic curve for predicting the primary outcome, which was ICU readmission within 48 hours of ICU discharge.Results: A total of 2,833 surveys was collected for 938 ICU-to-ward transfers, of which 40 (4%) were readmitted to the ICU within 48 hours of transfer. The median clinician likelihood of readmission score was 3 (interquartile range, 2-4). When physician and nurse likelihood scores were combined, the median clinician likelihood score had an AUC of 0.70 (95% confidence interval [CI], 0.62-0.78) for predicting ICU readmission within 48 hours. Nurses were significantly more accurate than interns at predicting 48-hour ICU readmission (AUC, 0.73 [95% CI, 0.64-0.82] vs. AUC, 0.60 [95% CI, 0.49-0.71]; P = 0.03). All other pairwise comparisons were not significantly different for predicting ICU readmission within 48 hours (P > 0.05 for all comparisons).Conclusions: We found that all clinicians surveyed in our ICU, regardless of the level of experience or clinician type, had only fair accuracy for predicting ICU readmission. Further research is needed to determine if clinical decision support tools would provide prognostic value above and beyond clinical judgment for determining who is ready for ICU discharge.