Sequence Based Prediction of Hospital Readmissions

Agrawal, Surabhi; Hon, Chun Pan; Garg, Sonali; Sampath, A.; Sushmita, Shanu; Cock, Martine De

doi:10.1145/2975167.2985647

Cited by 2 publications

(2 citation statements)

References 6 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…gov/data-and-reports/request-data/) for historical documents including an algorithm guide, detailed mission and history, dataset formation, availability for public access, the data acquisition process, and peer reviewed publications. 37,40,41 We concluded that OSHPD would provide high quality administrative data to meet our aims to collect non-VA ED visits and hospitalization in California among veterans with SCI/D. Patient-level hospitalization data were available beginning 01/01/1999 in both VA and OSHPD dataset, but ED visit data were only available beginning 01/01/2005 when California OSHPD started collecting ED data.…”

Section: Identify Data Availabilitymentioning

confidence: 99%

Data acquisition process for VA and non-VA emergency department and hospital utilization by veterans with spinal cord injury and disorders in California using VA and state data

Cai

Bakerjian

Bang³

et al. 2020

The Journal of Spinal Cord Medicine

View full text Add to dashboard Cite

Context: To identify VA and non-VA Emergency Department (ED) and hospital utilization by veterans with spinal cord injury and disorders (SCI/D) in California. Design: Retrospective cohort study. Setting: VA and Office of Statewide Health Planning and Development (OSHPD) in California. Participants: Total 300 veterans admitted to the study VA SCI/D Center for initial rehabilitations from 01/01/1999 through 08/17/2014. Interventions: N/A. Outcome Measures: Individual-level ED visits and hospitalizations during the first-year post-rehabilitation.Results: Among 145 veterans for whom ED visit data available, 168 ED visits were identified: 94 (55.2%) at non-VA EDs and 74 (44.8%) at the VA ED, with a mean of 1.16 (±2.21) ED visit/person. Seventy-seven (53.1%) veterans did not visit any ED. Of 68 (46.9%) veterans with ≥ one ED visit, 20 (29.4%) visited the VA ED only, 34 (50.0%) visited non-VA EDs only, and 14 (20.6%) visited both VA and non-VA EDs. Among 212 Veterans for whom hospitalization data were available, 247 hospitalizations were identified: 82 (33.2%) non-VA hospitalizations and 165 (66.8%) VA hospitalization with a mean of 1.17 (±1.62) hospitalizations/person. One hundred-seven (50.5%) veterans had no hospitalizations. Of 105 veterans with ≥ one hospitalization, 58 (55.2%) were hospitalized at the study VA hospital, 15 (14.3%) at a non-VA hospital, and 32 (30.5%) at both VA and non-VA hospitals. Conclusion: Non-VA ED and hospital usage among veterans with SCI/D occurred frequently. The acquisition of non-VA healthcare data managed by state agencies is vital to accurately and comprehensively evaluate needs and utilization rates among veteran populations.

show abstract

Section: Identify Data Availabilitymentioning

confidence: 99%

Data acquisition process for VA and non-VA emergency department and hospital utilization by veterans with spinal cord injury and disorders in California using VA and state data

Cai

Bakerjian

Bang³

et al. 2020

The Journal of Spinal Cord Medicine

View full text Add to dashboard Cite

show abstract

“…For example, event sequence data, which are used to describe patient care pathways 6,7 or changes in a patient's health state, 8 are more difficult to represent in such a manner, as a scalar summary removes important information about temporal changes. A popular method for scalarising sequence data is to count the frequency of each unique subsequence of length n within a given sequence (known as the n-gram method 9,10 ) which results in K n variables summarising the sequence, where K is the number of potential states that the sequence may take. This results in an explosion in the number of variables required to adequately represent a single predictive feature, artificially inflating the dimensionality of the problem for any planned statistical analysis.…”

Section: Introductionmentioning

confidence: 99%

Distance-based Classification and Regression Trees for the analysis of complex predictors in health and medical research

Johns

Bernhardt

Чурилов

2021

Stat Methods Med Res

View full text Add to dashboard Cite

Predicting patient outcomes based on patient characteristics and care processes is a common task in medical research. Such predictive features are often multifaceted and complex, and are usually simplified into one or more scalar variables to facilitate statistical analysis. This process, while necessary, results in a loss of important clinical detail. While this loss may be prevented by using distance-based predictive methods which better represent complex healthcare features, the statistical literature on such methods is limited, and the range of tools facilitating distance-based analysis is substantially smaller than those of other methods. Consequently, medical researchers must choose to either reduce complex predictive features to scalar variables to facilitate analysis, or instead use a limited number of distance-based predictive methods which may not fulfil the needs of the analysis problem at hand. We address this limitation by developing a Distance-Based extension of Classification and Regression Trees (DB-CART) capable of making distance-based predictions of categorical, ordinal and numeric patient outcomes. We also demonstrate how this extension is compatible with other extensions to CART, including a recently published method for predicting care trajectories in chronic disease. We demonstrate DB-CART by using it to expand upon previously published dose–response analysis of stroke rehabilitation data. Our method identified additional detail not captured by the previously published analysis, reinforcing previous conclusions. We also demonstrate how by combining DB-CART with other extensions to CART, the method is capable of making predictions about complex, multifaceted outcome data based on complex, multifaceted predictive features.

show abstract

Sequence Based Prediction of Hospital Readmissions

Cited by 2 publications

References 6 publications

Data acquisition process for VA and non-VA emergency department and hospital utilization by veterans with spinal cord injury and disorders in California using VA and state data

Data acquisition process for VA and non-VA emergency department and hospital utilization by veterans with spinal cord injury and disorders in California using VA and state data

Distance-based Classification and Regression Trees for the analysis of complex predictors in health and medical research

Contact Info

Product

Resources

About