Haifeng Xie

Emergency Readmission Criterion: A Technique for Determining the Emergency Readmission Time Window

A frequently chosen time window in defining readmission is 28 days after discharge. Yet in the literature, shorter and longer periods such as 14 days or 90-180 days have also been suggested. In this paper, we develop a modeling approach that systematically tackles the issue surrounding the appropriate choice of a time window as a definition of readmission. The approach is based on the intuitive idea that patients who are discharged from hospital can be broadly divided in to two groups-a group that is at high risk of readmission and a group that is at low risk. Using the national data (England), we demonstrate the usefulness of the approach in the case of chronic obstructive pulmonary disease (COPD), stroke, and congestive heart failure (CHF) patients, which are known to be the leading causes of early readmission. Our findings suggest that there are marked differences in the optimal width of the time window for COPD, stroke, and CHF patients. Furthermore, time windows and the probabilities of being in the high-risk group for COPD, stroke, and CHF patients for each of the 29 acute and specialist trusts in the London area indicate wide variability between hospitals. The novelty of this modeling approach lies in its ability to define an appropriate time window based on evidence objectively derived from operational data. Therefore, it can separately provide a unique approach in examining variability between hospitals, and potentially contribute to a better definition of readmission as a performance indicator.

A Semi-open Queueing Network Approach to the Analysis of Patient Flow in Healthcare Systems

In this paper, we present a modelling framework for patient flow in a healthcare system using semi-open queueing network models, which introduces a total bed constraint, above which new patients will be refused admission. Hence this model provides a realistic representation of a real system. This approach enables us to have access to a range of established methods that deals with queueing network models. We demonstrate the usefulness of the model in the context of a geriatric department and show that hospital managers can use this model to gain better understanding of the dynamics of patient flow and to study potential long-term impacts of policy changes.

A framework for predicting gross institutional long-term care cost arising from known commitments at local authority level

As the UK population ages, it is forecasted that there will be an unsustainable increase in the need for, and therefore in the costs of long-term care. Although several studies have been performed to estimate these costs, they do not take into account the impact of survival patterns on costs. Focussing only on residents already in care (known commitments), we have developed, in association with an English local authority, a framework for estimating the future gross cost incurred by this group, built around a survival model. We apply this framework to forecast the cost over a given period of time, of maintaining a group of individuals in residential and nursing care, funded by the local authority. One of the novelties in the model is that it translates survival inputs and unit fees for care into cost in a manner, which was useful and meaningful to decision makers.

Random effects models for operational patient pathways

Patient flow modeling is a growing field of interest in health services research. Several techniques have been applied to model movement of patients within and between health-care facilities. However, individual patient experience during the delivery of care has always been overlooked. In this work, a random effects model is introduced to patient flow modeling and applied to a London Hospital Neonatal unit data. In particular, a random effects multinomial logit model is used to capture individual patient trajectories in the process of care with patient frailties modeled as random effects. Intuitively, both operational and clinical patient flow are modeled, the former being physical and the latter latent. Two variants of the model are proposed, one based on mere patient pathways and the other based on patient characteristics. Our technique could identify interesting pathways such as those that result in high probability of death (survival), pathways incurring the least (highest) cost of care or pathways with the least (highest) length of stay. Patient-specific discharge probabilities from the health care system could also be predicted. These are of interest to health-care managers in planning the scarce resources needed to run health-care institutions.

A simple graphical decision aid for the placement of elderly people in long-term care

This paper describes the construction of a graphical decision tool to aid placement decisions of a multidisciplinary review panel for admissions to long-term care in a London borough in the UK. First we construct a prediction model of placement decisions based on an applicants attributes. Using data from the London borough, a composite model comprising syndromic decision rules followed by a two-stage hierarchical logistic regression model is proposed. The model proved to be robust in differentiating cases needing residential home care and nursing home care. Placement outcomes generated by the model are then represented graphically on a triangle plot. This approach could potentially be used as a decision support tool by managers of long-term care for continuous monitoring and assessment of the appropriateness of placements with respect to residents’ needs.

Modelling Decisions of a Multidisciplinary Panel for Admission to Long-Term Care

This paper describes a modelling study of a multidisciplinary review panel which is responsible for matching levels of long-term care to the needs of older people. The study aims to understand the decision making process of the review panel and to predict placement decisions based on an applicants attributes. Data were collected from cases notes presented to the London Borough of Merton review panel. A model predicting placement of an individual to residential home, nursing home or long-stay nursing care was built using logistic regression, and correctly predicts 78% of placement decisions. The model can be used as a means of checking the consistency of the review panels placement decisions.

Patients Flow: A Mixed-Effects Modelling Approach to Predicting Discharge Probabilities

A mixed effects approach hereby introduced to patients flow and length of stay modelling. In, particular, a class of generalized linear mixed models has been used to demonstrate the usefulness of this approach. This modelling technique is used to capture individual patients experience during the process of care as represented by their pathways through the system. The approach could predict the probability of discharge from the system, as well as detect where the system may be going wrong.

Determining Readmission Time Window Using Mixture of Generalised Erlang Distribution

The absence of a unified definition of readmissions has motivated the development of a modelling approach, to systematically tackle the issue surrounding the appropriate choice of a time window which defines readmission. The population of discharged patients can be broadly divided in two groups - a group at high risk of readmission and a group at low risk. This approach extends previous work by the authors, without restricting the number of stages, that patients may experience in the community. Using the national data (UK), we demonstrate its usefulness in the case of chronic obstructive pulmonary disease (COPD) which is known to be one of the leading causes of readmission. We further investigate variability in the definition of readmission among 10 strategic health authorities (SHAs) in England and observe that there are differences in the estimated time window across SHAs. The novelty of this modelling approach is the ability of capturing time to readmission that exhibit a non-zero mode and to estimate an appropriate time window based on evidence objectively derived from operational data.

A software tool to aid long-term care budget planning at local authority level

OBJECTIVE Local authorities face real challenges when it comes to annual budget planning for funding the system of long-term care. Uncertainty about the long-term cost of caring for current residents in the system, in addition to unknown future admissions, have made the tasks of local authority budget managers very complex and demanding. In this paper, we present a software implementation of a novel forecasting framework developed by the authors to provide useful information to local authority budget planners involved in long-term care. METHODS The tool is built upon a forecasting framework, which combines unit costs of care with an estimated underlying survival model for publicly funded residents in long-term care, to provide forecasts of the cost of maintaining the group of elderly people who are currently in long-term care (referred to as known commitments) for a period of time. A prototype version of the software tool, which was created and tested in collaboration with an English borough, allows user interaction via a friendly graphical interface that guides through a set of screens of options in a familiar wizard fashion. RESULTS AND DISCUSSION Feedback from care planners and managers show that the tool helps them gain better understanding of the patterns of length-of-stay of residents under their care, and provides quantitative inputs into their decision making on budget planning for long-term care. The development of the software tool brings advanced modelling techniques out of research papers into the hands of decision makers in the public sector and contributes to improving the delivery of long-term care.

A Method for Determining an Emergency Readmission Time Window for Better Patient Management

This paper introduces a modelling approach to determining the appropriate width of a time window within which an admission is classified as a readmission. The approach is based on an intuitive idea that patients, who are discharged from hospital, can be broadly considered as consisting of two groups - a group that is at high risk of readmission and a group that is at low risk. Using national data from the London area (UK), we demonstrate its usefulness in the case of chronic obstructive pulmonary diseases (COPD), one of the leading causes of early readmission. Although marked regional differences exist for the optimal width of the time window for COPD patients, our findings are largely inline with figures used by the government, hence provide some support for the use of 28 days as the time window for defining COPD readmissions. The novelty of this modelling approach lies in its ability to estimate an appropriate time window based on evidence objectively derived from operational data. Therefore, it can provide a means of monitoring performance for hospitals, and can potentially contribute to the better management of patient care