Machine-Learning Modeling to Predict Hospital Readmission Following Discharge to Post-Acute Care

Howard, Elizabeth; Morris, John N.; Schachter, Erez; Schwarzkopf, Ran; Shepard, Nicholas; Buchanan, Emily R.

doi:10.1016/j.jamda.2020.12.017

Cited by 10 publications

(8 citation statements)

References 18 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Our XGBoost models only achieved a minimum improvement compared with logistic regression in predicting hospitalization and ED visit, consistent with previous studies. 9,14,40,41 Interestingly, our models performed better than models used by Care Compare based on various metrics of performance. 5 There are multiple reasons why this may have occurred.…”

Section: Discussionmentioning

confidence: 84%

“…8,9 The extreme gradient boosting method (XGBoost) is a promising nonparametric tree-based algorithm. [10][11][12][13][14] Compared with regression-based algorithms (eg, logistic regression), XGBoost has several strengths that can benefit-risk-adjustment methods. In logistic regression, human decisions are needed to determine what factors go into the risk-adjustment model.…”

mentioning

confidence: 99%

See 1 more Smart Citation

Developing a Machine Learning Risk-adjustment Method for Hospitalizations and Emergency Department Visits of Nursing Home Residents With Dementia

Bowblis

Becerra

et al. 2023

Medical Care

View full text Add to dashboard Cite

Background: Long-stay nursing home (NH) residents with Alzheimer disease and related dementias (ADRD) are at high risk of hospital transfers. Machine learning might improve risk-adjustment methods for NHs. Objectives: The objective of this study was to develop and compare NH risk-adjusted rates of hospitalizations and emergency department (ED) visits among long-stay residents with ADRD using Extreme Gradient Boosting (XGBoost) and logistic regression. Research Design: Secondary analysis of national Medicare claims and NH assessment data in 2012 Q3. Data were equally split into the training and test sets. Both XGBoost and logistic regression predicted any hospitalization and ED visit using 58 predictors. NH-level risk-adjusted rates from XGBoost and logistic regression were constructed and compared. Multivariate regressions examined NH and market factors associated with rates of hospitalization and ED visits. Subjects: Long-stay Medicare residents with ADRD (N=413,557) from 14,057 NHs. Results: A total of 8.1% and 8.9% residents experienced any hospitalization and ED visit in a quarter, respectively. XGBoost slightly outperformed logistic regression in area under the curve (0.88 vs. 0.86 for hospitalization; 0.85 vs. 0.83 for ED visit). NH-level risk-adjusted rates from XGBoost were slightly lower than logistic regression (hospitalization=8.3% and 8.4%; ED=8.9% and 9.0%, respectively), but were highly correlated. Facility and market factors associated with the XGBoost and logistic regression-adjusted hospitalization and ED rates were similar. NHs serving more residents with ADRD and having a higher registered nurse-to-total nursing staff ratio had lower rates. Conclusions: XGBoost and logistic regression provide comparable estimates of risk-adjusted hospitalization and ED rates.

show abstract

Section: Discussionmentioning

confidence: 84%

mentioning

confidence: 99%

Developing a Machine Learning Risk-adjustment Method for Hospitalizations and Emergency Department Visits of Nursing Home Residents With Dementia

Bowblis

Becerra

et al. 2023

Medical Care

View full text Add to dashboard Cite

show abstract

“…For some older adults, postacute rehabilitation occurs after a healthy level of functioning is disrupted by an acute medical event (e.g., myocardial infarction, total joint replacement) and the person recovers to their previous level of health and function (Figure 1a). However, a large proportion of older adults discharged to postacute care present with medical complexities inclusive of multiple chronic conditions (56% with more than four chronic conditions), clinical frailty (66%), and moderate-severe cognitive impairment (23%; Howard et al, 2021; Kohler et al, 2020; Shier et al, 2022). For many of these individuals, an acute medical event occurred within the context of a trajectory characterized by previous declines in health and function (e.g., repeated falls, deconditioning, sepsis) and partial or incomplete postacute recovery with a high potential for hospital readmission (Figure 1b; Middleton et al, 2019).…”

Section: Postacute Rehabilitation and Psychosocial Dyadic Distressmentioning

confidence: 99%

Spilling over at the boiling point: A commentary on the need for dyadic approaches to psychosocial care with older adults and their care-partners in postacute rehabilitation.

Plys,

Bannon,

Keeney

et al. 2023

Rehabilitation Psychology

View full text Add to dashboard Cite

Purpose/Objective: Older adults typically receive rehabilitation services following an acute medical event within the context of declining health. Post-acute rehabilitation often serves as a “boiling point” for health needs and is met with numerous shared emotional stressors related to recovery, long-term care, and end-of-life considerations among older patients and their care-partners—referred to together as a dyad. Psychosocial intervention that targets patients and care-partners separately misses the opportunity to support dyads who typically navigate health-related emotional challenges together. In the context of chronic and serious illness, dyadic interventions (i.e., patient and informal care-partner, together) can successfully reduce emotional distress as well as improve communication and collaborative illness management. However, this approach has yet to be applied to older post-acute rehabilitation patients and their care-partners. Research Method/Design: In this commentary, we outline the need for dyadic psychosocial intervention approaches with older adults and their family care-partners in the context of post-acute rehabilitation. Results: First, we provide evidence for the potential benefits of a dyadic approach to psychosocial care. Next, we review theoretical models as well as clinically relevant confounding factors that can inform dyadic psychosocial case conceptualization and intervention. Finally, we offer a real-world clinical case example that demonstrates the opportunity for dyadic intervention to address common psychosocial challenges seen by psychologists in post-acute rehabilitation settings. Conclusions/Implications: Our goal is to encourage rehabilitation psychologists to view dyadic intervention as a first-line approach to psychosocial care with the growing population of older adults and their family care-partners in post-acute rehabilitation settings.

show abstract

“…Based on existing evidence and state of the science, various machine learning algorithms already helped create predictive equations for standard functional measures after inpatient rehabilitation for stroke: Functional Independence Measure (FIM), 10-m walk test, 6-min walk test and Berg Balance Scale ( 58 ). Moreover, machine-learning modeling predicted 30-day hospital readmissions after discharge to post-acute care, using patient SDH and other characteristics ( 59 ).…”

Section: Leveraging Big Data and Expanding Machine Learning In Physiatrymentioning

confidence: 99%

Social Determinants of Health in Physiatry: Challenges and Opportunities for Clinical Decision Making and Improving Treatment Precision

Conic

Geis

Vincent

2021

Front. Public Health

View full text Add to dashboard Cite

Physiatry is a medical specialty focused on improving functional outcomes in patients with a variety of medical conditions that affect the brain, spinal cord, peripheral nerves, muscles, bones, joints, ligaments, and tendons. Social determinants of health (SDH) play a key role in determining therapeutic process and patient functional outcomes. Big data and precision medicine have been used in other fields and to some extent in physiatry to predict patient outcomes, however many challenges remain. The interplay between SDH and physiatry outcomes is highly variable depending on different phases of care, and more favorable patient profiles in acute care may be less favorable in the outpatient setting. Furthermore, SDH influence which treatments or interventional procedures are accessible to the patient and thus determine outcomes. This opinion paper describes utility of existing datasets in combination with novel data such as movement, gait patterning and patient perceived outcomes could be analyzed with artificial intelligence methods to determine the best treatment plan for individual patients in order to achieve maximal functional capacity.

show abstract

Machine-Learning Modeling to Predict Hospital Readmission Following Discharge to Post-Acute Care

Cited by 10 publications

References 18 publications

Developing a Machine Learning Risk-adjustment Method for Hospitalizations and Emergency Department Visits of Nursing Home Residents With Dementia

Developing a Machine Learning Risk-adjustment Method for Hospitalizations and Emergency Department Visits of Nursing Home Residents With Dementia

Spilling over at the boiling point: A commentary on the need for dyadic approaches to psychosocial care with older adults and their care-partners in postacute rehabilitation.

Social Determinants of Health in Physiatry: Challenges and Opportunities for Clinical Decision Making and Improving Treatment Precision

Contact Info

Product

Resources

About