Semi-automating abstract screening with a natural language model pretrained on biomedical literature.

We demonstrate the performance and workload impact of incorporating a natural language model, pretrained on citations of biomedical literature, on a workflow of abstract screening for studies on prognostic factors in end-stage lung disease. The model was optimized on one-third of the abstracts, and model performance on the remaining abstracts was reported. Performance of the model, in terms of sensitivity, precision, F1 and inter-rater agreement, was moderate in comparison with other published models. However, incorporating it into the screening workflow, with the second reviewer screening only abstracts with conflicting decisions, translated into a 65% reduction in the number of abstracts screened by the second reviewer. Subsequent work will look at incorporating the pre-trained BERT model into screening workflows for other studies prospectively, as well as improving model performance.

Estimating Transmission Parameters for COVID-19 Clusters by Using Symptom Onset Data, Singapore, January-April 2020.

We estimated the generation interval distribution for coronavirus disease on the basis of serial intervals of observed infector-infectee pairs from established clusters in Singapore. The short mean generation interval and consequent high prevalence of presymptomatic transmission requires public health control measures to be responsive to these characteristics of the epidemic.

Lifetime trajectory simulation of chronic disease progression and comorbidity development.

INTRODUCTION: Comorbidity is common in elderly patients and it imposes heavy burden on both individual and the whole healthcare system. This study aims to gain insights of comorbidity development by simulating the lifetime trajectory of disease progression from single chronic disease to comorbidity. METHODS: Eight health states spanning from no chronic condition to comorbidity are considered in this study. Disease progression network is constructed based on the seven-year retrospective data of around 700,000 residents living in Singapore central region. Microsimulation is applied to simulate the process of aging and disease progression of a synthetic new-born cohort for the entire lifetime. RESULTS: Among the 40 unique trajectories observed from the simulation, the top 10 trajectories covers 60% of the cohort. Timespan of most trajectories from birth to death is 80 years. Most people progress to at risk at late 30 s, develop the first chronic condition at 50 s or 60 s, and then progress to complications at 70 s. It is also observed that the earlier one person develops chronic conditions, the more life-year-lost is incurred. DISCUSSION: The lifetime disease progression trajectory constructed for each person in the cohort describes how a person starts healthy, becomes at risk, then progresses to one or more chronic conditions, and finally deteriorates to various complications over the years. This study may help us have a better understanding of chronic disease progression and comorbidity development, hence add values to chronic disease prevention and management.

Singapore's regional health systems-a data-driven perspective on frequent admitters and cross utilization of healthcare services in three systems.

INTRODUCTION: With population health management being a priority in the Singapore, this paper aims to provide a data-driven perspective of the population health management initiatives to aid program planning and serves as a baseline for evaluation of future implemented programs. METHODS: A database with information on patient demographics, health services utilization, cost, diagnoses and chronic disease information from 2008 to 2013 for three regional health systems in Singapore was used for analysis. Patients with three or more inpatient admissions were considered as "Frequent Admitters." Health service utilization was quantified, and cross utilization of services was studied. One-year readmission rate for inpatients was studied, and a predictive model for readmission or death was developed. RESULTS: There were a total of 2.8 M patients in the database. Frequent admitters accounted for 0.9% of all patients with an average cost per patient of S$29 547. Of these, 89% had chronic diseases. Cross utilization of health services showed that 8.2% of the patients utilized services from more than one hospital with 19.6% utilizing hospital and polyclinic services in 2013. The highest risk of readmission or death was for those patients who had five or more inpatient episodes in each of the preceding 2 years. CONCLUSION: By understanding the profile of the patients and their utilization patterns in the three regional health systems, our study will help clinicians and decision makers design appropriate integrated care programs for patients with the aim of covering the healthcare needs for the enitre population across the healthcare spectrum in Singapore. Copyright © 2015 John Wiley & Sons, Ltd.

Analysis of patient waiting time governed by a generic maximum waiting time policy with general phase-type approximations.

Waiting time can affect patient satisfaction and quality of care in the emergency department (ED). Studies have shown that waiting time accounted for more than 50 % of total patient turnaround time at ED. The objective of this study is to examine a maximum waiting time policy such that patients who would experience a long wait are assumed to be processed in a threshold period. In particular, we are interested to investigate the associated factors of the policy such as new mean waiting time and the threshold period and their interaction. Under the policy, original patient waiting distribution is transformed to a piecewise distribution where one piecewise discontinuous and one piecewise continuous distributions are further investigated. Under the phase-type (PH) distribution assumption on the original waiting time, we establish closed-form expressions concerning new mean waiting time and time points of the threshold period. By fitting PH distributions to patient waiting data of an emergency department in Singapore, the factors are then estimated under various scenarios using the obtained analytical expressions. Specifically, for a given target mean waiting time, the threshold period needed in the policy is estimated. New mean waiting time is assessed with different choices of the threshold period. Analytical expressions in terms of the variance of the transformed waiting time and the threshold period are also presented.

Real-time prediction of waiting time in the emergency department, using quantile regression.

STUDY OBJECTIVE: Emergency department (ED) waiting times can affect patient satisfaction and quality of care. We develop and validate a model that predicts an individual patient's median and 95th percentile waiting time by using only data available at triage. METHODS: From the existing ED information system, we extracted date and time of triage completion, start time of emergency physician consultation, and patient acuity category (1=most urgent, 3=least urgent). Quantile regression was applied for model development and parameter estimation by using visits from January 2011. We assessed absolute prediction error, defined as the median difference between the 50th percentile (median) predicted waiting time and actual waiting time, and the proportion of underestimated prediction, defined as the percentage of patients whose actual waiting time exceeded the 95th percentile prediction. The model was validated retrospectively with June 2010 data and prospectively with data from April to June 2011 after integration with the existing ED information system. RESULTS: The derivation set included 13,200 ED visits; 903 (6.8%) were patient acuity category 1, 5,530 (41.9%) were patient acuity category 2, and 6,767 (51.3%) were patient acuity category 3. The median and 95th percentile waiting times were 17 and 57 minutes for patient acuity category 2 and 21 and 89 minutes for patient acuity category 3, respectively. The final model used predictors of patient acuity category, patient queue sizes, and flow rates only. In the retrospective validation, 5.9% of patient acuity category 2 and 5.4% of category 3 waiting times were underestimated. The median absolute prediction error was 11.9 minutes (interquantile range [IQR] 5.9 to 22.1 minutes) for patient acuity category 2 and 15.7 minutes (IQR 7.5 to 30.1 minutes) for category 3. In prospective validation, 4.3% of patient acuity category 2 and 5.8% of category 3 waiting times were underestimated. The median absolute prediction error was 9.2 minutes (IQR 4.4 to 15.1 minutes) for patient acuity category 2 and 12.9 minutes (IQR 6.5 to 22.5 minutes) for category 3. CONCLUSION: Using only a few data elements available at triage, the model predicts individual patients' waiting time with good accuracy.

Estimating ICU bed capacity using discrete event simulation.

PURPOSE: The intensive care unit (ICU) in a hospital caters for critically ill patients. The number of the ICU beds has a direct impact on many aspects of hospital performance. Lack of the ICU beds may cause ambulance diversion and surgery cancellation, while an excess of ICU beds may cause a waste of resources. This paper aims to develop a discrete event simulation (DES) model to help the healthcare service providers determine the proper ICU bed capacity which strikes the balance between service level and cost effectiveness. DESIGN/METHODOLOGY/APPROACH: The DES model is developed to reflect the complex patient flow of the ICU system. Actual operational data, including emergency arrivals, elective arrivals and length of stay, are directly fed into the DES model to capture the variations in the system. The DES model is validated by open box test and black box test. The validated model is used to test two what-if scenarios which the healthcare service providers are interested in: the proper number of the ICU beds in service to meet the target rejection rate and the extra ICU beds in service needed to meet the demand growth. FINDINGS: A 12-month period of actual operational data was collected from an ICU department with 13 ICU beds in service. Comparison between the simulation results and the actual situation shows that the DES model accurately captures the variations in the system, and the DES model is flexible to simulate various what-if scenarios. ORIGINALITY/VALUE: DES helps the healthcare service providers describe the current situation, and simulate the what-if scenarios for future planning.

Analysis of factors causing long patient waiting time and clinic overtime in outpatient clinics.

This paper is focused on the factors causing long patient waiting time/clinic overtime in outpatient clinics and how to mitigate them using discrete event simulation. A two-week period of data collection is conducted in an outpatient clinic of a Singapore government hospital. Detailed time study from patient arrival to patient departure is conducted, and the possible factors causing long patient waiting time/clinic overtime are discussed. A discrete simulation model is constructed to illustrate how to improve the clinic performance by mitigating the detected factors. Simulation and implementation results show that significant improvement is achieved if the factors are well addressed.

Queueing for healthcare.

Patient queues are prevalent in healthcare and wait time is one measure of access to care. We illustrate Queueing Theory-an analytical tool that has provided many insights to service providers when designing new service systems and managing existing ones. This established theory helps us to quantify the appropriate service capacity to meet the patient demand, balancing system utilization and the patient's wait time. It considers four key factors that affect the patient's wait time: average patient demand, average service rate and the variation in both. We illustrate four basic insights that will be useful for managers and doctors who manage healthcare delivery systems, at hospital or department level. Two examples from local hospitals are shown where we have used queueing models to estimate the service capacity and analyze the impact of capacity configurations, while considering the inherent variation in healthcare.

Intelligent analysis of acute bed overflow in a tertiary hospital in Singapore.

Hospital beds are a scarce resource and always in need. The beds are often organized by clinical specialties for better patient care. When the Accident & Emergency Department (A&E) admits a patient, there may not be an available bed that matches the requested specialty. The patient may be thus asked to wait at the A&E till a matching bed is available, or assigned a bed from a different specialty, which results in bed overflow. While this allows the patient to have faster access to an inpatient bed and treatment, it creates other problems. For instance, nursing care may be suboptimal and the doctors will need to spend more time to locate the overflow patients. The decision to allocate an overflow bed, or to let the patient wait a bit longer, can be a complicated one. While there can be a policy to guide the bed allocation decision, in reality it depends on clinical calls, current supply and waiting list, projected supply (i.e. planned discharges) and demand. The extent of bed overflow can therefore vary greatly, both in time dimension and across specialties. In this study, we extracted hospital data and used statistical and data mining approaches to identify the patterns behind bed overflow. With this insight, the hospital administration can be better equipped to devise strategies to reduce bed overflow and therefore improve patient care. Computational results show the viability of these intelligent data analysis techniques for understanding and managing the bed overflow problem.

Practical operations research applications for healthcare managers.

Operations research (OR) focuses on the application of analytical methods to facilitate better decision-making. Despite its usefulness and proliferation of papers in the academic literature, there are still major issues around getting OR models widely accepted and used as part of mainstream decision-making by clinicians, health managers and policy-makers. This article aims to raise the awareness of healthcare managers with regard to practical OR applications.