Predicting 30-day Hospital Readmission with Publicly Available Administrative Database. A Conditional Logistic Regression Modeling Approach

A Predictive Decision Analytics Approach for Primary Care Operations Management: A Case Study of Double-Booking Strategy Design and Evaluation

Primary care plays a vital role for individuals and families in accessing care, keeping well, and improving quality of life. However, the complexities and uncertainties in the primary care delivery system (e.g., patient no-shows/walk-ins, staffing shortage, COVID-19 pandemic) have brought significant challenges in its operations management, which can potentially lead to poor patient outcomes and negative primary care operations (e.g., loss of productivity, inefficiency). This paper presents a decision analytics approach developed based on predictive analytics and hybrid simulation to better facilitate management of the underlying complexities and uncertainties in primary care operations. A case study was conducted in a local family medicine clinic to demonstrate the use of this approach for patient no-show management. In this case study, a patient no-show prediction model was used in conjunction with an integrated agent-based and discrete-event simulation model to design and evaluate double-booking strategies. Using the predicted patient no-show information, a prediction-based double-booking strategy was created and compared against two other strategies, namely random and designated time. Scenario-based experiments were then conducted to examine the impacts of different double-booking strategies on clinic's operational outcomes, focusing on the trade-offs between the clinic productivity (measured by daily patient throughput) and efficiency (measured by visit cycle and patient wait time for doctor). The results showed that the best productivity-efficiency balance was derived under the prediction-based double-booking strategy. The proposed hybrid decision analytics approach has the potential to better support decision-making in primary care operations management and improve the system's performance. Further, it can be generalized in the context of various healthcare settings for broader applications.

Predictive Modeling of Hospital Readmission: Challenges and Solutions

Hospital readmission prediction is a study to learn models from historical medical data to predict probability of a patient returning to hospital in a certain period, e.g. 30 or 90 days, after the discharge. The motivation is to help health providers deliver better treatment and post-discharge strategies, lower the hospital readmission rate, and eventually reduce the medical costs. Due to inherent complexity of diseases and healthcare ecosystems, modeling hospital readmission is facing many challenges. By now, a variety of methods have been developed, but existing literature fails to deliver a complete picture to answer some fundamental questions, such as what are the main challenges and solutions in modeling hospital readmission; what are typical features/models used for readmission prediction; how to achieve meaningful and transparent predictions for decision making; and what are possible conflicts when deploying predictive approaches for real-world usages. In this paper, we systematically review computational models for hospital readmission prediction, and propose a taxonomy of challenges featuring four main categories: (1) data variety and complexity; (2) data imbalance, locality and privacy; (3) model interpretability; and (4) model implementation. The review summarizes methods in each category, and highlights technical solutions proposed to address the challenges. In addition, a review of datasets and resources available for hospital readmission modeling also provides firsthand materials to support researchers and practitioners to design new approaches for effective and efficient hospital readmission prediction.

Early liaison psychiatry consultations and general hospital readmission: A retrospective cohort study

BACKGROUND

Readmission rates are under growing scrutiny as an indicator of quality of care as much as a potential source of savings. Patients with comorbid psychiatric conditions are more likely to be readmitted, so Consultation-Liaison Psychiatry (CLP) may play a role in lowering readmission rates.

METHOD

In this retrospective cohort study conducted in a general hospital in Paris, France, all consecutive adult inpatients referred for the first time to CLP from January 2008 to December 2016, were included. The main outcomes were 30-day and 7-day readmissions in the same hospital, excluding iterative and planned stays. The objective of this study is to determine whether the timing of psychiatric consultations is associated with 30-day and 7-day readmission rates.

RESULTS

A total of 4498 inpatients (2298(51·1%) women, age = 59·8(±19·3) years) were referred to CLP. Adjusting for age, sex, place of residence, year of admission, type of ward, psychiatric diagnosis and disease severity, later consultation was associated with higher 30-day and 7-day readmission rates (adjusted Odds Ratio [95% confidence interval]:1.21[1·10-1·33] and 1·26[1·11-3·13], respectively). Further adjusting for length of stay, the association remained significant for 7-day readmission (1.28[1·05-1·57]). After stratification on the length of stay, for stays in the highest tercile (i.e., >21 days) an intervention after day 3 (versus before) was associated with 30-day and 7-day readmission rates of 15·8% versus 8·6%(1·81 [1·11-3·13]) and 4·9% versus 1·8%(2·98[1·16-9·88]), respectively.

CONCLUSION

Earlier psychiatric consultation was associated with fewer 30-day and 7-day readmissions. Interventional studies are needed to show that proactive CLP teams could help general hospitals to improve quality of care and make significant economic savings.

The path from big data analytics capabilities to value in hospitals: a scoping review

BACKGROUND

As the uptake of health information technologies increased, most healthcare organizations have become producers of big data. A growing number of hospitals are investing in the development of big data analytics (BDA) capabilities. If the promises associated with these capabilities are high, how hospitals create value from it remains unclear. The present study undertakes a scoping review of existing research on BDA use in hospitals to describe the path from BDA capabilities (BDAC) to value and its associated challenges.

METHODS

This scoping review was conducted following Arksey and O'Malley's 5 stages framework. A systematic search strategy was adopted to identify relevant articles in Scopus and Web of Science. Data charting and extraction were performed following an analytical framework that builds on the resource-based view of the firm to describe the path from BDA capabilities to value in hospitals.

RESULTS

Of 1,478 articles identified, 94 were included. Most of them are experimental research (n=69) published in medical (n=66) or computer science journals (n=28). The main value targets associated with the use of BDA are improving the quality of decision-making (n=56) and driving innovation (n=52) which apply mainly to care (n=67) and administrative (n=48) activities. To reach these targets, hospitals need to adequately combine BDA capabilities and value creation mechanisms (VCM) to enable knowledge generation and drive its assimilation. Benefits are endpoints of the value creation process. They are expected in all articles but realized in a few instances only (n=19).

CONCLUSIONS

This review confirms the value creation potential of BDA solutions in hospitals. It also shows the organizational challenges that prevent hospitals from generating actual benefits from BDAC-building efforts. The configuring of strategies, technologies and organizational capabilities underlying the development of value-creating BDA solutions should become a priority area for research, with focus on the mechanisms that can drive the alignment of BDA and organizational strategies, and the development of organizational capabilities to support knowledge generation and assimilation.

Value and Applicability of Large Administrative Healthcare Databases in Prosthetics and Orthotics Outcomes Research

The goal of health economics and outcomes research is to improve healthcare decision making. In the absence of high-value clinical data, the availability and quality of administrative healthcare data could be vital in the generation of evidence for orthotics and prosthetics services. The purpose of this article is to provide a stronger understanding of administrative healthcare data analysis, an area that has been scarcely examined within prosthetics and orthotics despite the wealth of information available within such data. Examples of common datasets in this arena currently available are provided, as well as an overview of the limitations and advantages of studies utilizing such datasets.

Persistent Heart Failure Symptoms at Hospital Discharge Predicts 30-Day Clinical Events

Background

The relationship between heart failure (HF) symptoms at hospital discharge and 30-day clinical events is unknown. Variability in HF symptom assessment may affect ability to predict readmission risk.

Objective

The aim of this study was to describe HF symptom profiles and burden at hospital discharge. A secondary aim was to examine the relationship between symptom burden at discharge and 30-day clinical events.

Methods

An exploratory descriptive design was used. Patients with HF (n = 186) were enrolled 24 to 48 hours pre hospital discharge. The HF Somatic Perception Scale quantified 18 HF physical signs and symptoms. Scores were divided into tertiles (0-10, 11-19, and 20 and higher). The Patient Health Questionnaire-9 quantified depressive symptoms. Self-assessed health, comorbid illnesses, and 30-day clinical events were documented. Chi-square and logistic regression were used to examine clinical events.

Results

The sample (n = 186) was predominantly White (87.6%), male (59.1%), elderly (mean [SD], 74.2 [12.5]), and symptomatic (92.5%) at discharge. Heart Failure Somatic Perception Scale scores ranged from 0 to 53, with a mean (SD) of 13.7 (10.1). Symptoms reported most frequently were fatigue (67%), nocturia (62%), need to rest (53%), and inability to do usual activities due to shortness of breath (52%). Thirty-day event rate was 28%, with significant differences between Heart Failure Somatic Perception Scale tertiles (9.4% vs 37.7% in the second and third tertiles, respectively; [chi]22(N = 186) = 16.73, P < .001). Heart Failure Somatic Perception Scale tertile 2 or 3 (odds ratio [OR], 5.7; P = .003; and OR, 4.3; P = .021), self-assessed health (OR, 2.6; P = .029), and being in a relationship predicted clinical events.

Conclusions

Heart failure symptom burden at discharge predicted 30-day clinical events. Comprehensive symptom assessment is important when determining readmission risk.

Predictive modeling to identify potential participants of a disease management program hypertension

BACKGROUND

Based on the premise of limited health-care resources, decision-makers pursue to allocate disease management programs (DMP) more targeted.

METHODS

Based on routine data from a private health insurance company, a prediction model was developed to estimate the individual risk for future in-patient stays of patients eligible for a DMP Hypertension. The database included anonymous claims data of 38,284 policyholders with a diagnosis in the year 2013. A cutoff point of ≥70% was used for selecting candidates with a risk for future hospitalization. Using a logistic regression model, we estimated the model's prognostic power, the occurrence of clinical events, and the resource use.

RESULTS

Overall, the final model shows acceptable prognostic power (detection rate = 64.3%; sensitivity = 68.7%; positive predictive value (PPV) = 64.1%, area under the curve (AUC) = 0.72). The comparison between the selected hypothetical DMP-group with a predicted (LOH) ≥70% showed additional costs of about 69% for the DMP-group compared to insure with a LOH <70%.

CONCLUSION

The predictive analytical approach may identify potential DMP participants with a high risk of increased health services utilization and in-patient stays.

Values, challenges and future directions of big data analytics in healthcare: A systematic review

The emergence of powerful software has created conditions and approaches for large datasets to be collected and analyzed which has led to informed decision-making towards tackling health issues. The objective of this study is to systematically review 804 scholarly publications related to big data analytics in health in order to identify the organizational and social values along with associated challenges. Key principles of Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology were followed for conducting systematic reviews. Following a research path, we present the values, challenges and future directions of the scientific area using indicative examples from relevant published articles. The study reveals that one of the main values created is the development of analytical techniques which provides personalized health services to users and supports human decision-making using automated algorithms, challenging the power issues in the doctor-patient relationship and creating new working conditions. A main challenge to data analytics is data management and security when processing large volumes of sensitive, personal health data. Future research is directed towards the development of systems that will standardize and secure the process of extracting private healthcare datasets from relevant organizations. Our systematic literature review aims to provide to governments and health policy-makers a better understanding of how the development of a data driven strategy can improve public health and the functioning of healthcare organizations but also how can create challenges that need to be addressed in the near future to avoid societal malfunctions.

The impact of creating mathematical formula to predict cardiovascular events in patients with heart failure

Since our retrospective study has formed a mathematical formula, α = f(x₁, …, x₂₅₂), where α is the probability of cardiovascular events in patients with heart failure (HF) and x₁ is each clinical parameter, we prospectively tested the predictive capability and feasibility of the mathematical formula of cardiovascular events in HF patients. First of all, to create such a mathematical formula using limited number of the parameters to predict the cardiovascular events in HF patients, we retrospectively determined f(x) that formulates the relationship between the most influential 50 clinical parameters (x) among 252 parameters using 167 patients hospitalized due to acute HF; the nonlinear optimization could provide the formula of α = f(x₁, …, x₅₀) which fitted the probability of the actual cardiovascular events per day. Secondly, we prospectively examined the predictability of f(x) in other 213 patients using 50 clinical parameters in 3 hospitals, and we found that the Kaplan–Meier curves using actual and estimated occurrence probabilities of cardiovascular events were closely correlated. We conclude that we created a mathematical formula f(x) that precisely predicted the occurrence probability of future cardiovascular outcomes of HF patients per day. Mathematical modelling may predict the occurrence probability of cardiovascular events in HF patients.

Claims data-driven modeling of hospital time-to-readmission risk with latent heterogeneity

Hospital readmission risk modeling is of great interest to both hospital administrators and health care policy makers, for reducing preventable readmission and advancing care service quality. To accommodate the needs of both stakeholders, a readmission risk model is preferable if it (i) exhibits superior prediction performance; (ii) identifies risk factors to help target the most at-risk individuals; and (iii) constructs composite metrics to evaluate multiple hospitals, hospital networks, and geographic regions. Existing work mainly addressed the first two features and it is challenging to address the third one because available medical data are fragmented across hospitals. To simultaneously address all three features, this paper proposes readmission risk models with incorporation of latent heterogeneity, and takes advantage of administrative claims data, which is less fragmented and involves larger patient cohorts. Different levels of latent heterogeneity are considered to quantify the effects of unobserved factors, provide composite measures for performance evaluation at various aggregate levels, and compensate less informative claims data. To demonstrate the prediction performances of the proposed models, a real case study is considered on a state-wide heart failure patient cohort. A systematic comparison study is then carried out to evaluate the performances of 49 risk models and their variants.

Identifying Important Attributes for Prognostic Prediction in Traumatic Brain Injury Patients

Background: Generally, traumatic brain injury (TBI) patients do not have a stable condition, particularly after the first week of TBI. Hence, indicating the attributes in prognosis through a prediction model is of utmost importance since it helps caregivers with treatment-decision options, or prepares the relatives for the most-likely outcome. Objectives: This study attempted to determine and order the attributes in prognostic prediction in TBI patients, based on early clinical findings. A hybrid method was employed, which combines a decision tree (DT) and an artificial neural network (ANN) in order to improve the modeling process. Methods: The DT approach was applied as the initial analysis of the network architecture to increase accuracy in prediction. Afterwards, the ANN structure was mapped from the initial DT based on a part of the data. Subsequently, the designed network was trained and validated by the remaining data. 5-fold cross-validation method was applied to train the network. The area under the receiver operating characteristic (ROC) curve, sensitivity, specificity, and accuracy rate were utilized as performance measures. The important attributes were then determined from the trained network using two methods: change of mean squared error (MSE), and sensitivity analysis (SA). Results: The hybrid method offered better results compared to the DT method. The accuracy rate of 86.3 % vs. 82.2 %, sensitivity value of 55.1 % vs. 47.6 %, specificity value of 93.6 % vs. 91.1 %, and the area under the ROC curve of 0.705 vs. 0.695 were achieved for the hybrid method and DT, respectively. However, the attributes’ order by DT method was more consistent with the clinical literature. Conclusions: The combination of different modeling methods can enhance their performance. However, it may create some complexities in computations and interpretations. The outcome of the present study could deliver some useful hints in prognostic prediction on the basis of early clinical findings for TBI patients.

Contribution of temporal data to predictive performance in 30-day readmission of morbidly obese patients

Background

Reduction of readmissions after discharge represents an important challenge for many hospitals and has attracted the interest of many researchers in the past few years. Most of the studies in this field focus on building cross-sectional predictive models that aim to predict the occurrence of readmission within 30-days based on information from the current hospitalization. The aim of this study is demonstration of predictive performance gain obtained by inclusion of information from historical hospitalization records among morbidly obese patients.

Methods

The California Statewide inpatient database was used to build regularized logistic regression models for prediction of readmission in morbidly obese patients (n = 18,881). Temporal features were extracted from historical patient hospitalization records in a one-year timeframe. Five different datasets of patients were prepared based on the number of available hospitalizations per patient. Sample size of the five datasets ranged from 4,787 patients with more than five hospitalizations to 20,521 patients with at least two hospitalization records in one year. A 10-fold cross validation was repeted 100 times to assess the variability of the results. Additionally, random forest and extreme gradient boosting were used to confirm the results.

Results

Area under the ROC curve increased significantly when including information from up to three historical records on all datasets. The inclusion of more than three historical records was not efficient. Similar results can be observed for Brier score and PPV value. The number of selected predictors corresponded to the complexity of the dataset ranging from an average of 29.50 selected features on the smallest dataset to 184.96 on the largest dataset based on 100 repetitions of 10-fold cross-validation.

Discussion

The results show positive influence of adding information from historical hospitalization records on predictive performance using all predictive modeling techniques used in this study. We can conclude that it is advantageous to build separate readmission prediction models in subgroups of patients with more hospital admissions by aggregating information from up to three previous hospitalizations.

Big Data and Analytics in Healthcare

This editorial is part of the Focus Theme of Methods of Information in Medicine on "Big Data and Analytics in Healthcare". The amount of data being generated in the healthcare industry is growing at a rapid rate. This has generated immense interest in leveraging the availability of healthcare data (and "big data") to improve health outcomes and reduce costs. However, the nature of healthcare data, and especially big data, presents unique challenges in processing and analyzing big data in healthcare. This Focus Theme aims to disseminate some novel approaches to address these challenges. More specifically, approaches ranging from efficient methods of processing large clinical data to predictive models that could generate better predictions from healthcare data are presented.