Lessons learned from the hospital to home community care program in Singapore and the supporting AI multiple readmissions prediction model.

In a prior practice and policy article published in Healthcare Science, we introduced the deployed application of an artificial intelligence (AI) model to predict longer-term inpatient readmissions to guide community care interventions for patients with complex conditions in the context of Singapore's Hospital to Home (H2H) program that has been operating since 2017. In this follow on practice and policy article, we further elaborate on Singapore's H2H program and care model, and its supporting AI model for multiple readmission prediction, in the following ways: (1) by providing updates on the AI and supporting information systems, (2) by reporting on customer engagement and related service delivery outcomes including staff-related time savings and patient benefits in terms of bed days saved, (3) by sharing lessons learned with respect to (i) analytics challenges encountered due to the high degree of heterogeneity and resulting variability of the data set associated with the population of program participants, (ii) balancing competing needs for simpler and stable predictive models versus continuing to further enhance models and add yet more predictive variables, and (iii) the complications of continuing to make model changes when the AI part of the system is highly interlinked with supporting clinical information systems, (4) by highlighting how this H2H effort supported broader Covid-19 response efforts across Singapore's public healthcare system, and finally (5) by commenting on how the experiences and related capabilities acquired from running this H2H program and related community care model and supporting AI prediction model are expected to contribute to the next wave of Singapore's public healthcare efforts from 2023 onwards. For the convenience of the reader, some content that introduces the H2H program and the multiple readmissions AI prediction model that previously appeared in the prior Healthcare Science publication is repeated at the beginning of this article.

Two Singapore public healthcare AI applications for national screening programs and other examples.

This article explains how two AI systems have been incorporated into the everyday operations of two Singapore public healthcare nation-wide screening programs. The first example is embedded within the setting of a national level population health screening program for diabetes related eye diseases, targeting the rapidly increasing number of adults in the country with diabetes. In the second example, the AI assisted screening is done shortly after a person is admitted to one of the public hospitals to identify which inpatients-especially which elderly patients with complex conditions-have a high risk of being readmitted as an inpatient multiple times in the months following discharge. Ways in which healthcare needs and the clinical operations context influenced the approach to designing or deploying the AI systems are highlighted, illustrating the multiplicity of factors that shape the requirements for successful large-scale deployments of AI systems that are deeply embedded within clinical workflows. In the first example, the choice was made to use the system in a semi-automated (vs. fully automated) mode as this was assessed to be more cost-effective, though still offering substantial productivity improvement. In the second example, machine learning algorithm design and model execution trade-offs were made that prioritized key aspects of patient engagement and inclusion over higher levels of predictive accuracy. The article concludes with several lessons learned related to deploying AI systems within healthcare settings, and also lists several other AI efforts already in deployment and in the pipeline for Singapore's public healthcare system.

Chest radiograph-based artificial intelligence predictive model for mortality in community-acquired pneumonia.

BACKGROUND: Chest radiograph (CXR) is a basic diagnostic test in community-acquired pneumonia (CAP) with prognostic value. We developed a CXR-based artificial intelligence (AI) model (CAP AI predictive Engine: CAPE) and prospectively evaluated its discrimination for 30-day mortality. METHODS: Deep-learning model using convolutional neural network (CNN) was trained with a retrospective cohort of 2235 CXRs from 1966 unique adult patients admitted for CAP from 1 January 2019 to 31 December 2019. A single-centre prospective cohort between 11 May 2020 and 15 June 2020 was analysed for model performance. CAPE mortality risk score based on CNN analysis of the first CXR performed for CAP was used to determine the area under the receiver operating characteristic curve (AUC) for 30-day mortality. RESULTS: 315 inpatient episodes for CAP occurred, with 30-day mortality of 19.4% (n=61/315). Non-survivors were older than survivors (mean (SD)age, 80.4 (10.3) vs 69.2 (18.7)); more likely to have dementia (n=27/61 vs n=58/254) and malignancies (n=16/61 vs n=18/254); demonstrate higher serum C reactive protein (mean (SD), 109 mg/L (98.6) vs 59.3 mg/L (69.7)) and serum procalcitonin (mean (SD), 11.3 (27.8) µg/L vs 1.4 (5.9) µg/L). The AUC for CAPE mortality risk score for 30-day mortality was 0.79 (95% CI 0.73 to 0.85, p<0.001); Pneumonia Severity Index (PSI) 0.80 (95% CI 0.74 to 0.86, p<0.001); Confusion of new onset, blood Urea nitrogen, Respiratory rate, Blood pressure, 65 (CURB-65) score 0.76 (95% CI 0.70 to 0.81, p<0.001), respectively. CAPE combined with CURB-65 model has an AUC of 0.83 (95% CI 0.77 to 0.88, p<0.001). The best performing model was CAPE incorporated with PSI, with an AUC of 0.84 (95% CI 0.79 to 0.89, p<0.001). CONCLUSION: CXR-based CAPE mortality risk score was comparable to traditional pneumonia severity scores and improved its discrimination when combined.

Baseline measurement of running away among youth in foster care.

The current study evaluated the use of various behavioral measures of running away with regard to (a) the differential utility of interval- versus event-based measures, (b) the differential utility of rate versus duration measures, (c) the utility of correcting for occurrence opportunity, and (d) the influence of unit of analysis (i.e., single-subject vs. grouped data). Seven different baseline measures were calculated for 84 runaways, and a unit-size analysis was conducted by constructing groups of various sizes from the original sample. An expert panel evaluated the suitability of the baseline measures for treatment evaluation. Results demonstrate the utility of evaluating duration-based measures and correcting for occurrence opportunity. Results also indicate that single-subject baselines may often be unacceptable for treatment evaluations, regardless of the type of measure selected for use.

An evaluation of simultaneous and sequential presentation of preferred and nonpreferred food to treat food selectivity.

In the current investigation, we compared two methods of food presentation (simultaneous vs. sequential) to increase consumption of nonpreferred food for 3 children with food selectivity. In the simultaneous condition, preferred foods were presented at the same time as nonpreferred food (e.g., a piece of broccoli was presented on a chip). In the sequential condition, acceptance of the nonpreferred food resulted in presentation of the preferred food. Increases in consumption occurred immediately during the simultaneous condition for 2 of the 3 participants. For 1 participant, increases in consumption occurred in the simultaneous condition relative to the sequential condition, but only after physical guidance and re-presentation were added to treatment. Finally, consumption increased for 1 participant in the sequential condition, but only after several sessions. These results are discussed in terms of possible mechanisms that may alter preferences for food (i.e., establishing operations, flavor-flavor conditioning).