Development, Validation and Deployment of a Real Time 30 Day Hospital Readmission Risk Assessment Tool in the Maine Healthcare Information Exchange

Hao, Shiying; Wang, Yue; Jin, Bo; Shin, Andrew; Zhu, Chunqing; Huang, Min; Zheng, Le; Luo, Jin; Hu, Zhongkai; Fu, Changlin; Dai, Dingwei; Wang, Yicheng; Culver, Devore S; Alfreds, Shaun T; Rogow, Todd; Stearns, Frank; Sylvester, Karl G.; Widen, Eric; Ling, Xuefeng B.

doi:10.1371/journal.pone.0140271

Cited by 54 publications

(61 citation statements)

References 28 publications

Supporting

Mentioning

Contrasting

Unclassified

Order By: Relevance

“…The analysis described herein builds on our previously validated machine-learning models to predict hospital readmission 35,36 to develop COVID-19 readmission risk prediction tool 37 . The algorithm described here enables different case management strategies for various risk score thresholds, and facilitates the incorporation of different assumptions about the impact of the intervention and quarantine ( Supplementary Figure 1).…”

Section: Discussionmentioning

confidence: 99%

Recurrence of SARS-CoV-2 PCR positivity in COVID-19 patients: a single center experience and potential implications

Huang

Zheng

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

IMPORTANCE How to appropriately care for patients who become PCR-negative for severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is still not known.Patients who have recovered from coronavirus disease 2019 (COVID-19) could profoundly impact the health care system if a subset were to be PCR-positive again with reactivated SARS-CoV-2.OBJECTIVE To characterize a single center COVID-19 cohort with and without recurrence of PCR positivity, and develop an algorithm to identify patients at high risk of retest positivity after discharge to inform health care policy and case management decision-making. DESIGN, SETTING, AND PARTICIPANTSA cohort of 414 patients with confirmed SARS-CoV-2 infection, at The Second Affiliated EXPOSURES Polymerase chain reaction (PCR) and IgM-IgG antibody confirmed SARS-CoV-2 infection. MAIN OUTCOMES AND MEASURES Univariable and multivariable statistical analysis of the clinical, laboratory, radiologic image, medical treatment, and clinical course of admission/quarantine/readmission data to develop an algorithm to predict patients at risk of recurrence of PCR positivity.RESULTS 16.7% (95CI: 13.0%-20.3%) patients retest PCR positive 1 to 3 times after discharge, despite being in strict quarantine. The driving factors in the recurrence prediction model included: age, BMI; lowest levels of the blood laboratory tests

show abstract

Section: Discussionmentioning

confidence: 99%

Recurrence of SARS-CoV-2 PCR positivity in COVID-19 patients: a single center experience and potential implications

Huang

Zheng

et al. 2020

Preprint

Self Cite

View full text Add to dashboard Cite

show abstract

“…The work of Hao et al is notable as an operationalized, externally validated model of readmission risk using data from a health information exchange in Maine. 18 Our plan for data and clinical validation presented a similar opportunity to prospectively evaluate model performance and to revise the model as needed. The need for model revision is not unexpected, but rather points to the critical nature of model validation.…”

Section: Revised Model Resultsmentioning

confidence: 99%

“…[14][15][16][17] Recent work by Hao et al demonstrates prospective validation of a 30-day readmission risk tool using health information exchange data from Maine. 18 Thirty-day hospital readmissions are a priority for all health care payers, including the federal government. The 2014 Military Health System (MHS) Review, a special report at the behest of the Secretary of Defense to address MHS access, quality, and safety, highlighted reducing readmissions as a priority for military treatment facilities (MTFs) and cited an all-cause MTF readmission rate of 8.8%.…”

Section: Background and Significancementioning

confidence: 99%

Development and Prospective Validation of a Machine Learning-Based Risk of Readmission Model in a Large Military Hospital

et al. 2019

View full text Add to dashboard Cite

Background Thirty-day hospital readmissions are a quality metric for health care systems. Predictive models aim to identify patients likely to readmit to more effectively target preventive strategies. Many risk of readmission models have been developed on retrospective data, but prospective validation of readmission models is rare. To the best of our knowledge, none of these developed models have been evaluated or prospectively validated in a military hospital. Objectives The objectives of this study are to demonstrate the development and prospective validation of machine learning (ML) risk of readmission models to be utilized by clinical staff at a military medical facility and demonstrate the collaboration between the U.S. Department of Defense's integrated health care system and a private company. Methods We evaluated multiple ML algorithms to develop a predictive model for 30-day readmissions using data from a retrospective cohort of all-cause inpatient readmissions at Madigan Army Medical Center (MAMC). This predictive model was then validated on prospective MAMC patient data. Precision, recall, accuracy, and the area under the receiver operating characteristic curve (AUC) were used to evaluate model performance. The model was revised, retrained, and rescored on additional retrospective MAMC data after the prospective model's initial performance was evaluated. Results Within the initial retrospective cohort, which included 32,659 patient encounters, the model achieved an AUC of 0.68. During prospective scoring, 1,574 patients were scored, of whom 152 were readmitted within 30 days of discharge, with an all-cause readmission rate of 9.7%. The AUC of the prospective predictive model was 0.64. The model achieved an AUC of 0.76 after revision and addition of further retrospective data. Conclusion This work reflects significant collaborative efforts required to operationalize ML models in a complex clinical environment such as that seen in an integrated health care system and the importance of prospective model validation.

show abstract

“…59 In addition, and generally distinct from the directed approach, querybased exchange's use of aggregate and longitudinal data naturally lends itself to population-level analytics and identification of highrisk patients for use by case managers and care coordinators. 60,61 Given the factors listed above, we hypothesize that query-based exchange will be associated with a lower probability of potentially avoidable use of health care services.…”

Section: Vest Et Almentioning

confidence: 99%

“…Alternatively, query‐based exchange can support better communication of information during transitions of care by giving ambulatory care providers access to diagnostic information from inpatient settings or by facilitating information sharing with specialists . In addition, and generally distinct from the directed approach, query‐based exchange's use of aggregate and longitudinal data naturally lends itself to population‐level analytics and identification of high‐risk patients for use by case managers and care coordinators . Given the factors listed above, we hypothesize that query‐based exchange will be associated with a lower probability of potentially avoidable use of health care services.…”

Section: Introductionmentioning

confidence: 99%

The associations between query‐based and directed health information exchange with potentially avoidable use of health care services

Vest

Unruh

Shapiro

et al. 2019

Health Services Research

View full text Add to dashboard Cite

Objective To quantify the impact of two approaches (directed and query‐based) to health information exchange (HIE) on potentially avoidable use of health care services. Data Sources/Study Setting Data on ambulatory care providers’ adoption of HIE were merged with Medicare fee‐for‐service claims from 2008 to 2014. Providers were from 13 counties in New York served by the Rochester Regional Health Information Organization (RHIO). Study Design Linear regression models with provider and year fixed effects were used to estimate changes in the probability of utilization outcomes for Medicare beneficiaries attributed to providers adopting directed and/or query‐based HIE compared with beneficiaries attributed to providers who had not adopted HIE. Data Collection Providers’ HIE adoption status was determined through Rochester RHIO registration records. RHIO and claims data were linked via National Provider Identifiers. Principal Findings Query‐based HIE adoption was associated with a 0.2 percentage point reduction in the probability of an ambulatory care sensitive hospitalization and a 1.1 percentage point decrease in the likelihood of an unplanned readmission. Directed HIE adoption was not associated with any outcome. Conclusions The Centers for Medicare & Medicaid Services’ (CMS) EHR certification criteria includes requirements for directed HIE, but not query‐based HIE. Pending further research, certification criteria should place equal weight on facilitating query‐based and directed exchange.

show abstract

Development, Validation and Deployment of a Real Time 30 Day Hospital Readmission Risk Assessment Tool in the Maine Healthcare Information Exchange

Cited by 54 publications

References 28 publications

Recurrence of SARS-CoV-2 PCR positivity in COVID-19 patients: a single center experience and potential implications

Recurrence of SARS-CoV-2 PCR positivity in COVID-19 patients: a single center experience and potential implications

Development and Prospective Validation of a Machine Learning-Based Risk of Readmission Model in a Large Military Hospital

The associations between query‐based and directed health information exchange with potentially avoidable use of health care services

Contact Info

Product

Resources

About