Development of a predictive model to identify inpatients at risk of re-admission within 30 days of discharge (PARR-30)

Differential Effect of Patient- and Provider-Level Factors on Patient Satisfaction Scores in Academic General Internal Medicine Clinics

OBJECTIVE

Determine whether patient-level or provider-level factors have greater influence on patient satisfaction scores in an academic general internal medicine clinic.

METHODS

Two years of data (2017-2019) from the Clinician and Group Consumer Assessment of Healthcare Providers and Systems (CGCAHPS) surveys from ambulatory internal medicine clinic visits in an academic health center located in the Midwest United States were used. Patient satisfaction was measured using the overall provider satisfaction score (0-10), dichotomized with 9-10 defined as satisfactory and 0-8 as unsatisfactory. Provider-level independent variables included age, sex, race/ethnicity, provider type, service type, clinical effort, academic rank, and years since graduation. Patient-level factors included age, sex, race/ethnicity, education, and Epic Risk Score. Generalized mixed-effects logistic regression models were used to investigate associations between top-box satisfaction score and patient- and provider-level factors, accounting for the nesting of patients within providers.

RESULTS

Thirty-three providers and 4597 patients were included in the analysis. Male providers (OR, 1.57; 95% CI, 1.00, 2.47), minority group 2 (OR, 3.54; 95% CI, 1.24, 10.07) and minority group 3 (OR, 6.04; 95% CI, 1.45, 25.12), faculty (OR, 3.83; 95% CI, 1.56, 9.36), and primary care providers (OR, 5.60; 95% CI, 1.62, 19.34) had increased odds of having a top-box rating compared with females, minority group 1, advanced practice providers, and perioperative providers respectively. Age was the only patient independent correlate of top-box rating with a 3% increased odds of top-box rating for every year increase in age (OR, 1.03; 95% CI 1.02, 1.03).

CONCLUSIONS

In this academic general internal medicine clinic, top-box satisfaction scores were more strongly associated with provider-level factors, including provider race/ethnicity, provider type, and service type, as opposed to patient-level factors. Further research is needed to confirm these findings and identify potential system-level interventions.

A review of information sources and analysis methods for data driven decision aids in child and adolescent mental health services

OBJECTIVE

Clinical data analysis relies on effective methods and appropriate data. Recognizing distinctive clinical services and service functions may lead to improved decision-making. Our first objective is to categorize analytical methods, data sources, and algorithms used in current research on information analysis and decision support in child and adolescent mental health services (CAMHS). Our secondary objective is to identify the potential for data analysis in different clinical services and functions in which data-driven decision aids can be useful.

MATERIALS AND METHODS

We searched related studies in Science Direct and PubMed from 2018 to 2023(Jun), and also in ACM (Association for Computing Machinery) Digital Library, DBLP (Database systems and Logic Programming), and Google Scholar from 2018 to 2021. We have reviewed 39 studies and extracted types of analytical methods, information content, and information sources for decision-making.

RESULTS

In order to compare studies, we developed a framework for characterizing health services, functions, and data features. Most data sets in reviewed studies were small, with a median of 1,176 patients and 46,503 record entries. Structured data was used for all studies except two that used textual clinical notes. Most studies used supervised classification and regression. Service and situation-specific data analysis dominated among the studies, only two studies used temporal, or process features from the patient data. This paper presents and summarizes the utility, but not quality, of the studies according to the care situations and care providers to identify service functions where data-driven decision aids may be relevant.

CONCLUSIONS

Frameworks identifying services, functions, and care processes are necessary for characterizing and comparing electronic health record (EHR) data analysis studies. The majority of studies use features related to diagnosis and assessment and correspondingly have utility for intervention planning and follow-up. Profiling the disease severity of referred patients is also an important application area.

Predictive modeling for identification of older adults with high utilization of health and social services

AIM

Machine learning techniques have demonstrated success in predictive modeling across various clinical cases. However, few studies have considered predicting the use of multisectoral health and social services among older adults. This research aims to utilize machine learning models to detect high-risk groups of excessive health and social services utilization at early stage, facilitating the implementation of preventive interventions.

METHODS

We used pseudonymized data covering a four-year period and including information on a total of 33,374 senior citizens from Southern Finland. The endpoint was defined based on the occurrence of unplanned healthcare visits and the total number of different services used. Input features included individual's basic demographics, health status and past usage of healthcare resources. Logistic regression and eXtreme Gradient Boosting (XGBoost) methods were used for binary classification, with the dataset split into 70% training and 30% testing sets.

RESULTS

Subgroup-based results mirrored trends observed in the full cohort, with age and certain health issues, e.g. mental health, emerging as positive predictors for high service utilization. Conversely, hospital stay and urban residence were associated with decreased risk. The models achieved a classification performance (AUC) of 0.61 for the full cohort and varying in the range of 0.55-0.62 for the subgroups.

CONCLUSIONS

Predictive models offer potential for predicting future high service utilization in the older adult population. Achieving high classification performance remains challenging due to diverse contributing factors. We anticipate that classification performance could be increased by including features based on additional data categories such as socio-economic data.

Can prediction models for hospital readmission be improved by incorporating patient-reported outcome measures? A systematic review and narrative synthesis

PURPOSE

To investigate the roles, challenges, and implications of using patient-reported outcome measures (PROMs) in predicting the risk of hospital readmissions.

METHODS

We systematically searched four bibliometric databases for peer-reviewed studies published in English between 1 January 2000 and 15 June 2023 and used validated PROMs to predict readmission risks for adult populations. Reported studies were analysed and narratively synthesised in accordance with the CHARMS and PRISMA guidelines.

RESULTS

Of the 2858 abstracts reviewed, 23 studies met predefined eligibility criteria, representing diverse geographic regions and medical specialties. Among those, 19 identified the positive contributions of PROMs in predicting readmission risks. Seven studies utilised generic PROMs exclusively, eleven used generic and condition-specific PROMs, while 5 focussed solely on condition-specific PROMs. Logistic regression was the most used modelling approach, with 13 studies aiming at predicting 30-day all-cause readmission risks. The c-statistic, ranging from 0.54 to 0.84, was reported in 22/23 studies as a measure of model discrimination. Nine studies reported model calibration in addition to c-statistic. Thirteen studies detailed their approaches to dealing with missing data.

CONCLUSION

Our study highlights the potential of PROMs to enhance predictive accuracy in readmission models, while acknowledging the diversity in data collection methods, readmission definitions, and model evaluation approaches. Recognizing that PROMs serve various purposes beyond readmission reduction, our study supports routine data collection and strategic integration of PROMs in healthcare practices to improve patient outcomes. To facilitate comparative analysis and broaden the use of PROMs in the prediction framework, it is imperative to consider the methodological aspects involved.

Exploration of understanding of impactibility analysis and application through workshops to inform model design for population health management policy: a qualitative assessment

OBJECTIVES

Assess understanding of impactibility modelling definitions, benefits, challenges and approaches.

DESIGN

Qualitative assessment.

SETTING

Two workshops were developed. Workshop 1 was to consider impactibility definitions and terminology through moderated open discussion, what the potential pros and cons might be, and what factors would be best to assess. In workshop 2, participants appraised five approaches to impactibility modelling identified in the literature.

PARTICIPANTS

National Health Service (NHS) analysts, policy-makers, academics and members of non-governmental think tank organisations identified through existing networks and via a general announcement on social media. Interested participants could enrol after signing informed consent.

OUTCOME MEASURES

Descriptive assessment of responses to gain understanding of the concept of impactibility (defining impactibility analysis), the benefits and challenges of using this type of modelling and most relevant approach to building an impactibility model for the NHS.

RESULTS

37 people attended 1 or 2 workshops in small groups (maximum 10 participants): 21 attended both workshops, 6 only workshop 1 and 10 only workshop 2. Discussions in workshop 1 illustrated that impactibility modelling is not clearly understood, with it generally being viewed as a cross-sectional way to identify patients rather than considering patients by iterative follow-up. Recurrent factors arising from workshop 2 were the shortage of benchmarks; incomplete access to/recording of primary care data and social factors (which were seen as important to understanding amenability to treatment); the need for outcome/action suggestions as well as providing the data and the risk of increasing healthcare inequality.

CONCLUSIONS

Understanding of impactibility modelling was poor among our workshop attendees, but it is an emerging concept for which few studies have been published. Implementation would require formal planning and training and should be performed by groups with expertise in the procurement and handling of the most relevant health-related real-world data.

Re-hospitalization factors and economic characteristics of urinary tract infected patients using machine learning

Objective

Urinary tract infection is one of the most prevalent bacterial infectious diseases in outpatient treatment, and 50-80% of women experience it more than once, with a recurrence rate of 40-50% within a year; consequently, preventing re-hospitalization of patients is critical. However, in the field of urology, no research on the analysis of the re-hospitalization status for urinary tract infections using machine learning algorithms has been reported to date. Therefore, this study uses various machine learning algorithms to analyze the clinical and nonclinical factors related to patients who were re-hospitalized within 30 days of urinary tract infection.

Methods

Data were collected from 497 patients re-hospitalized for urinary tract infections within 30 days and 496 patients who did not require re-hospitalization. The re-hospitalization factors were analyzed using four machine learning algorithms: gradient boosting classifier, random forest, naive Bayes, and logistic regression.

Results

The best-performing gradient boosting classifier identified respiratory rate, days of hospitalization, albumin, diastolic blood pressure, blood urea nitrogen, body mass index, systolic blood pressure, body temperature, total bilirubin, and pulse as the top-10 factors that affect re-hospitalization because of urinary tract infections. The 993 patients whose data were collected were divided into risk groups based on these factors, and the re-hospitalization rate, days of hospitalization, and medical expenses were observed to decrease from the high- to low-risk group.

Conclusions

This study showed new possibilities in analyzing the status of urinary tract infection-related re-hospitalization using machine learning. Identifying factors affecting re-hospitalization and incorporating preventable and reinforcement-based treatment programs can aid in reducing the re-hospitalization rate and average number of days of hospitalization, thereby reducing medical expenses.

Predictors of unplanned hospital readmission after non-cardiac surgery in Singapore: a 2-year retrospective review

INTRODUCTION

Unplanned hospital readmissions after surgery contribute significantly to healthcare costs and potential complications. Identifying predictors of readmission is inherently complex and involves an intricate interplay between medical factors, healthcare system factors and sociocultural factors. Therefore, the aim of this study was to elucidate the predictors of readmissions in an Asian surgical patient population.

METHODS

A two-year single-institution retrospective cohort study of 2744 patients was performed in a university-affiliated tertiary hospital in Singapore, including patients aged 45 and above undergoing intermediate or high-risk non-cardiac surgery. Unadjusted analysis was first performed, followed by multivariable logistic regression.

RESULTS

Two hundred forty-nine patients (9.1%) had unplanned 30-day readmissions. Significant predictors identified from multivariable analysis include: American Society of Anaesthesiologists (ASA) Classification grades 3 to 5 (adjusted OR 1.51, 95% CI 1.10-2.08, p = 0.01), obesity (adjusted OR 1.66, 95% CI 1.18-2.34, p = 0.04), asthma (OR 1.70, 95% CI 1.03-2.81, p = 0.04), renal disease (OR 2.03, 95% CI 1.41-2.92, p < 0.001), malignancy (OR 1.68, 95% CI 1.29-2.37, p < 0.001), chronic obstructive pulmonary disease (OR 2.46, 95% CI 1.19-5.11, p = 0.02), cerebrovascular disease (OR 1.73, 95% CI 1.17-2.58, p < 0.001) and anaemia (OR 1.45, 95% CI 1.07-1.96, p = 0.02).

CONCLUSION

Several significant predictors of unplanned readmissions identified in this Asian surgical population corroborate well with findings from Western studies. Further research will require future prospective studies and development of predictive risk modelling to further address and mitigate this phenomenon.

Identification of elderly patients at risk for 30-day readmission: Clinical insight beyond big data prediction

AIM

This study explores the potential benefit of combining clinicians' risk assessments and the automated 30-day readmission prediction model.

BACKGROUND

Automated readmission prediction models based on electronic health records are increasingly applied as part of prevention efforts, but their accuracy is moderate.

METHODS

This prospective multisource study was based on self-reported surveys of clinicians and data from electronic health records. The survey was performed at 15 internal medicine wards of three general Clalit hospitals between May 2016 and June 2017. We examined the degree of concordance between the Preadmission Readmission Detection Model, clinicians' readmission risk classification and the likelihood of actual readmission. Decision trees were developed to classify patients by readmission risk.

RESULTS

A total of 694 surveys were collected for 371 patients. The disagreement between clinicians' risk assessment and the model was 34.5% for nurses and 33.5% for physicians. The decision tree algorithms identified 22% and 9% (based on nurses and physicians, respectively) of the model's low-medium-risk patients as high risk (accuracy 0.8 and 0.76, respectively).

CONCLUSIONS

Combining the Readmission Model with clinical insight improves the ability to identify high-risk elderly patients.

IMPLICATIONS FOR NURSING MANAGEMENT

This study provides algorithms for the decision-making process for selecting high-risk readmission patients based on nurses' evaluations.

LACE Index to Predict the High Risk of 30-Day Readmission in Patients With Acute Myocardial Infarction at a University Affiliated Hospital

Background

The LACE index (length of stay, acuity of admission, comorbidity index, and emergency room visit in the past 6 months) has been used to predict the risk of 30-day readmission after hospital discharge in both medical and surgical patients. This study aimed to utilize the LACE index to predict the risk of 30-day readmission in hospitalized patients with acute myocardial infraction (AMI).

Methods

This was a retrospective study. Data were extracted from the hospital's electronic medical records of patients admitted with AMI between 2015 and 2019. LACE index was built on admission patient demographic data, and clinical and laboratory findings during the index of admission. The multivariate logistic regression was performed to determine the association and the risk prediction ability of the LACE index, and 30-day readmission were analyzed by receiver operator characteristic curves with C-statistic.

Results

Of the 3,607 patients included in the study, 5.7% (205) were readmitted within 30 days of discharge from the hospital. The adjusted odds ratio based on logistic regression of all baseline variables showed a statistically significant association with the LACE score and revealed an increased risk of readmission within 30 days of hospital discharge. However, patients with high LACE scores (≥10) had a significantly higher rate of emergency revisits within 30 days from the index discharge than those with low LACE scores. Despite this, analysis of the receiver operating characteristic curve indicated that the LACE index had favorable discrimination ability C-statistic 0.78 (95%CI; 0.75–0.81). The Hosmer–Lemeshow goodness- of-fit test P value was p = 0.920, indicating that the model was well-calibrated to predict risk of the 30-day readmission.

Conclusion

The LACE index demonstrated the good discrimination power to predict the risk of 30-day readmissions for hospitalized patients with AMI. These results can help clinicians to predict the risk of 30-day readmission at the early stage of hospitalization and pay attention during the care of high-risk patients. Future work is to be focused on additional factors to predict the risk of 30-day readmissions; they should be considered to improve the model performance of the LACE index with other acute conditions by using administrative data.

Comparison of LACE and HOSPITAL Readmission Risk Scores for CMS Target and Nontarget Conditions

This study evaluated the utility and performance of the LACE index and HOSPITAL score with consideration of the type of diagnoses and assessed the accuracy of these models for predicting readmission risks in patient cohorts from 2 large academic medical centers. Admissions to 2 hospitals from 2011 to 2015, derived from the Vizient Clinical Data Base and regional health information exchange, were included in this study (291 886 encounters). Models were assessed using Bayesian information criterion and area under the receiver operating characteristic curve. They were compared in CMS diagnosis-based cohorts and in 2 non-CMS cancer diagnosis-based cohorts. Overall, both models for readmission risk performed well, with LACE performing slightly better (area under the receiver operating characteristic curve 0.73 versus 0.69; P ≤ 0.001). HOSPITAL consistently outperformed LACE among 4 CMS target diagnoses, lung cancer, and colon cancer. Both LACE and HOSPITAL predict readmission risks well in the overall population, but performance varies by salient, diagnosis-based risk factors.

Ability of the LACE Index to Predict 30-Day Readmissions in Patients with Acute Myocardial Infarction

Aims: This study aimed to utilize the existing LACE index (length of stay, acuity of admission, comorbidity index and emergency room visit in the past six months) to predict the risk of 30-day readmission and to find the associated factors in patients with AMI. Methods: This was a retrospective study and LACE index scores were calculated for patients admitted with AMI between 2015 and 2019. Data were utilized from the hospital’s electronic medical record. Multivariate logistic regression was performed to find the association between covariates and 30-day readmission. The risk prediction ability of the LACE index for 30-day readmission was analyzed by receiver operating characteristic curves with the C statistic. Results: A total of 205 (5.7%) patients were readmitted within 30 days. The odds ratio of older age group (OR = 1.78, 95% CI: 1.54–2.05), admission via emergency ward (OR = 1.45; 95% CI: 1.42–1.54) and LACE score ≥10 (OR = 2.71; 95% CI: 1.03–4.37) were highly associated with 30-day readmissions and statistically significant. The receiver operating characteristic curve C statistic of the LACE index for AMI patients was 0.78 (95% CI: 0.75–0.80) and showed favorable discrimination in the prediction of 30-day readmission. Conclusion: The LACE index showed a good discrimination to predict the risk of 30-day readmission for hospitalized patients with AMI. Further study would be recommended to focus on additional factors that can be used to predict the risk of 30-day readmission; this should be considered to improve the model performance of the LACE index for other acute conditions by using the national-based administrative data.

Developing and validating a predictive model for future emergency hospital admissions

Although many emergency hospital admissions may be unavoidable, a proportion of these admissions represent a failure of the care system. The adverse consequences of avoidable emergency hospital admissions affect patients, carers, care systems and substantially increase care costs. The aim of this study was to develop and validate a risk prediction model to estimate the individual probability of emergency admission in the next 12 months within a regional population. We deterministically linked routinely collected data from secondary care with population level data, resulting in a comprehensive research dataset of 190,466 individuals. The resulting risk prediction tool is based on a logistic regression model with five independent variables. The model indicated a discrimination of area under the receiver operating characteristic curve of 0.9384 (95% CI 0.9325-0.9443). We also experimented with different probability cut-off points for identifying high risk patients and found the model's overall prediction accuracy to be over 95% throughout. In summary, the internally validated model we developed can predict with high accuracy the individual risk of emergency admission to hospital within the next year. Its relative simplicity makes it easily implementable within a decision support tool to assist with the management of individual patients in the community.

Factors associated with accessing long-term adult social care in people aged 75 and over: a retrospective cohort study

Background

An ageing population and limited resources have put strain on state provision of adult social care (ASC) in England. With social care needs predicted to double over the next 20 years, there is a need for new approaches to inform service planning and development, including through predictive models of demand.

Objective

Describe risk factors for long-term ASC in two inner London boroughs and develop a risk prediction model for long-term ASC.

Methods

Pseudonymised person-level data from an integrated care dataset were analysed. We used multivariable logistic regression to model associations of demographic factors, and baseline aspects of health status and health service use, with accessing long-term ASC over 12 months.

Results

The cohort comprised 13,394 residents, aged ≥75 years with no prior history of ASC at baseline. Of these, 1.7% became ASC clients over 12 months. Residents were more likely to access ASC if they were older or living in areas with high socioeconomic deprivation. Those with preexisting mental health or neurological conditions, or more intense prior health service use during the baseline period, were also more likely to access ASC. A prognostic model derived from risk factors had limited predictive power.

Conclusions

Our findings reinforce evidence on known risk factors for residents aged 75 or over, yet even with linked routinely collected health and social care data, it was not possible to make accurate predictions of long-term ASC use for individuals. We propose that a paradigm shift towards more relational, personalised approaches, is needed.

A review of literature on risk prediction tools for hospital readmissions in older adults

PURPOSE

The purpose of this paper is to identify and analyse the readmission risk prediction tools reported in the literature and their benefits when it comes to healthcare organisations and management.

DESIGN/METHODOLOGY/APPROACH

Readmission risk prediction is a growing topic of interest with the aim of identifying patients in particular those suffering from chronic diseases such as congestive heart failure, chronic obstructive pulmonary disease and diabetes, who are at risk of readmission. Several models have been developed with different levels of predictive ability. A structured and extensive literature search of several databases was conducted using the Preferred Reporting Items for Systematic Reviews and Meta-analysis strategy, and this yielded a total of 48,984 records.

FINDINGS

Forty-three articles were selected for full-text and extensive review after following the screening process and according to the eligibility criteria. About 34 unique readmission risk prediction models were identified, in which their predictive ability ranged from poor to good (c statistic 0.5-0.86). Readmission rates ranged between 3.1 and 74.1% depending on the risk category. This review shows that readmission risk prediction is a complex process and is still relatively new as a concept and poorly understood. It confirms that readmission prediction models hold significant accuracy at identifying patients at higher risk for such an event within specific context.

RESEARCH LIMITATIONS/IMPLICATIONS

Since most prediction models were developed for specific populations, conditions or hospital settings, the generalisability and transferability of the predictions across wider or other contexts may be difficult to achieve. Therefore, the value of prediction models remains limited to hospital management. Future research is indicated in this regard.

ORIGINALITY/VALUE

This review is the first to cover readmission risk prediction tools that have been published in the literature since 2011, thereby providing an assessment of the relevance of this crucial KPI to health organisations and managers.

Miller N, L. Pecina J, D. Decker M, I. Wi C, Garrison GM. Association of Early and Late Hospital Readmissions with a Novel Housing-Based Socioeconomic Measure

Background

While socioeconomic status has been linked to hospital readmissions for several conditions, reliable measures of individual socioeconomic status are often not available. HOUSES, a new measure of individual socioeconomic status based upon objective public data about one's housing unit, is inversely associated with overall hospitalization rate but it has not been studied with respect to readmissions.

Purpose

To determine if patients in the lowest HOUSES quartile are more likely to be readmitted within 30 days (short-term) and 180 days (long-term).

Methods

A retrospective cohort study of 11 993 patients having 21 633 admissions was conducted using generalized linear mixed-effects models.

Results

HOUSES quartile did not show any significant association with early readmission. However, when compared to the lowest HOUSES quartile, the second quartile (OR = 0.90, 95%CI 0.83-0.98) and the third quartile (OR = 0.91, 95%CI 0.83-0.99) were associated with lower odds of late readmission while the highest quartile (OR = 0.91, 95%CI 0.82-1.01) was not statistically different.

Conclusion

HOUSES was associated with late readmission, but not early readmission. This may be because early readmissions are influenced by medical conditions and hospital care while late readmissions are influenced by ambulatory care and home-based factors. Since HOUSES relies on public county assessor data, it is generally available and may be used to focus interventions on those at highest risk for late readmission.

Bridging the impactibility gap in population health management: a systematic review

OBJECTIVES

Assess whether impactibility modelling is being used to refine risk stratification for preventive health interventions.

DESIGN

Systematic review.

SETTING

Primary and secondary healthcare populations.

PAPERS

Articles published from 2010 to 2020 on the use or implementation of impactibility modelling in population health management, reported with the terms 'intervenability', 'amenability', and 'propensity to succeed' (PTS) and associated with the themes 'care sensitivity', 'characteristic responders', 'needs gap', 'case finding', 'patient selection' and 'risk stratification'.

INTERVENTIONS

Qualitative synthesis to identify themes for approaches to impactibility modelling.

RESULTS

Of 1244 records identified, 20 were eligible for inclusion. Identified themes were 'health conditions amenable to care' (n=6), 'PTS modelling' (n=8) and 'comparison or combination with clinical judgement' (n=6). For the theme 'health conditions amenable to care', changes in practice did not reduce admissions, particularly for ambulatory care sensitive conditions, and sometimes increased them, with implementation noted as a possible issue. For 'PTS modelling', high costs and needs did not necessarily equate to high impactibility and targeting a larger number of individuals with disorders associated with lower costs had more potential. PTS modelling seemed to improve accuracy in care planning, estimation of cost savings, engagement and/or care quality. The 'comparison or combination with clinical judgement' theme suggested that models can reach reasonable to good discriminatory power to detect impactable patients. For instance, a model used to identify patients appropriate for proactive multimorbid care management showed good concordance with physicians (c-statistic 0.75). Another model employing electronic health record scores reached 65% concordance with nurse and physician decisions when referring elderly hospitalised patients to a readmission prevention programme. However, healthcare professionals consider much wider information that might improve or impede the likelihood of treatment impact, suggesting that complementary use of models might be optimum.

CONCLUSIONS

The efficiency and equity of targeted preventive care guided by risk stratification could be augmented and personalised by impactibility modelling.

Personalising the risk of conversion from laparoscopic to open hysterectomy in benign conditions: Development and external validation of risk prediction models

OBJECTIVE

To develop and validate novel prediction models to personalise the risk of conversion from laparoscopic to open hysterectomy in benign conditions.

DESIGN

Retrospective cohort study.

SETTINGS

English NHS Hospitals between 2011 and 2018.

POPULATION

68 752 women undergoing laparoscopic hysterectomy for benign conditions.

METHODS

We developed two multivariable logistic models using readily available clinical information, one for the pre-operative setting and another for operative decision-making using additional surgical information, using 2011-2016 data in five regions (24 806 women). We validated them (a) temporally in the same regions using 2017-2018 data (12 438 women); (b) geographically in the same time-period using data from three different regions (22 024 women); and (c) temporally and geographically using 2017-2018 data in three different regions (9484 women).

MAIN OUTCOME MEASURES

Conversion from laparoscopic to open hysterectomy.

RESULTS

Conversions occurred in 6.8% (1687/24 806) of hysterectomies in the development group, and in 5.5% (681/12 438) in the temporal, 5.9% (1297/22 204) in the geographical and 5.2% (488/9484) in the temporal and geographical validation groups. In the development cohort, the area under the receiver operating characteristic curve values for the pre-operative and operative models were 0.65 and 0.67, respectively. In the validation cohorts the corresponding values were 0.65 and 0.66 (temporal), 0.66 and 0.68 (geographical) and 0.65 and 0.68 (temporal and geographical), respectively. Factors predictive of conversion included age, Asian ethnicity, obesity, fibroids, adenomyosis, endometriosis and adhesions. Adhesions were the most predictive (odds ratio 2.54, 95% confidence interval 2.22-2.90).

CONCLUSION

The models for predicting conversions showed acceptable performance and transferability.

TWEETABLE ABSTRACT

Novel tool to personalise the risk of conversion from laparoscopic to open hysterectomy in benign conditions.

Predictive modeling for 14-day unplanned hospital readmission risk by using machine learning algorithms

BACKGROUND

Early unplanned hospital readmissions are associated with increased harm to patients, increased medical costs, and negative hospital reputation. With the identification of at-risk patients, a crucial step toward improving care, appropriate interventions can be adopted to prevent readmission. This study aimed to build machine learning models to predict 14-day unplanned readmissions.

METHODS

We conducted a retrospective cohort study on 37,091 consecutive hospitalized adult patients with 55,933 discharges between September 1, 2018, and August 31, 2019, in an 1193-bed university hospital. Patients who were aged < 20 years, were admitted for cancer-related treatment, participated in clinical trial, were discharged against medical advice, died during admission, or lived abroad were excluded. Predictors for analysis included 7 categories of variables extracted from hospital's medical record dataset. In total, four machine learning algorithms, namely logistic regression, random forest, extreme gradient boosting, and categorical boosting, were used to build classifiers for prediction. The performance of prediction models for 14-day unplanned readmission risk was evaluated using precision, recall, F1-score, area under the receiver operating characteristic curve (AUROC), and area under the precision-recall curve (AUPRC).

RESULTS

In total, 24,722 patients were included for the analysis. The mean age of the cohort was 57.34 ± 18.13 years. The 14-day unplanned readmission rate was 1.22%. Among the 4 machine learning algorithms selected, Catboost had the best average performance in fivefold cross-validation (precision: 0.9377, recall: 0.5333, F1-score: 0.6780, AUROC: 0.9903, and AUPRC: 0.7515). After incorporating 21 most influential features in the Catboost model, its performance improved (precision: 0.9470, recall: 0.5600, F1-score: 0.7010, AUROC: 0.9909, and AUPRC: 0.7711).

CONCLUSIONS

Our models reliably predicted 14-day unplanned readmissions and were explainable. They can be used to identify patients with a high risk of unplanned readmission based on influential features, particularly features related to diagnoses. The operation of the models with physiological indicators also corresponded to clinical experience and literature. Identifying patients at high risk with these models can enable early discharge planning and transitional care to prevent readmissions. Further studies should include additional features that may enable further sensitivity in identifying patients at a risk of early unplanned readmissions.

A prospective cohort study of two predictor models for 30-day emergency readmission in older patients

AIM

To undertake a prospective study of the accuracy of two models (LACE and BOOST) in predicting unplanned hospital readmission in older patients (>75 years).

METHODS

Data were collected from a single centre prospectively on 110 patients over 75 years old admitted to the acute medical unit. Follow-up was conducted at 30 days. The primary outcome was the c-statistic for both models.

RESULTS

The readmission rate was 32.7% and median age 82 years, and both BOOST and LACE scores were significantly higher in those readmitted compared with those who were not. C-statistics were calculated for both tools with BOOST score 0.667 (95% CI 0.559-0.775, P = .005) and LACE index 0.685 (95% CI 0.579-0.792, P = .002).

CONCLUSION

In this prospective study, both the BOOST and LACE scores were found to be significant yet poor, predictive models of hospital readmission. Recent hospitalisation (within the previous 6 months) was found to be the most significant contributing factor.

Implementing an Individual-Centric Discharge Process across Singapore Public Hospitals

Singapore is one of the first known countries to implement an individual-centric discharge process across all public hospitals to manage frequent admissions-a perennial challenge for public healthcare, especially in an aging population. Specifically, the process provides daily lists of high-risk patients to all public hospitals for customized discharge procedures within 24 h of admission. We analyzed all public hospital admissions (N = 150,322) in a year. Among four models, the gradient boosting machine performed the best (AUC = 0.79) with a positive predictive value set at 70%. Interestingly, the cumulative length of stay (LOS) in the past 12 months was a stronger predictor than the number of previous admissions, as it is a better proxy for acute care utilization. Another important predictor was the "number of days from previous non-elective admission", which is different from previous studies that included both elective and non-elective admissions. Of note, the model did not include LOS of the index admission-a key predictor in other models-since our predictive model identified frequent admitters for pre-discharge interventions during the index (current) admission. The scientific ingredients that built the model did not guarantee its successful implementation-an "art" that requires the alignment of processes, culture, human capital, and senior management sponsorship. Change management is paramount, otherwise data-driven health policies, no matter how well-intended, may not be accepted or implemented. Overall, our study demonstrated the viability of using artificial intelligence (AI) to build a near real-time nationwide prediction tool for individual-centric discharge, and the critical factors for successful implementation.

Published models that predict hospital readmission: a critical appraisal

INTRODUCTION

The number of readmission risk prediction models available has increased rapidly, and these models are used extensively for health decision-making. Unfortunately, readmission models can be subject to flaws in their development and validation, as well as limitations in their clinical usefulness.

OBJECTIVE

To critically appraise readmission models in the published literature using Delphi-based recommendations for their development and validation.

METHODS

We used the modified Delphi process to create Critical Appraisal of Models that Predict Readmission (CAMPR), which lists expert recommendations focused on development and validation of readmission models. Guided by CAMPR, two researchers independently appraised published readmission models in two recent systematic reviews and concurrently extracted data to generate reference lists of eligibility criteria and risk factors.

RESULTS

We found that published models (n=81) followed 6.8 recommendations (45%) on average. Many models had weaknesses in their development, including failure to internally validate (12%), failure to account for readmission at other institutions (93%), failure to account for missing data (68%), failure to discuss data preprocessing (67%) and failure to state the model's eligibility criteria (33%).

CONCLUSIONS

The high prevalence of weaknesses in model development identified in the published literature is concerning, as these weaknesses are known to compromise predictive validity. CAMPR may support researchers, clinicians and administrators to identify and prevent future weaknesses in model development.

The liver transplant risk score prognosticates the outcomes of liver transplant recipients at listing

BACKGROUND

We assessed if the risk of post-liver transplant mortality within 24 h could be stratified at the time of listing using the liver transplant risk score (LTRS). Secondary aims were to assess if the LTRS could stratify the risk of 30-day, 1-year mortality, and survival beyond the first year.

METHODS

MELD, BMI, age, diabetes, and the need for dialysis were the five variables used to calculate the LTRS during patients' evaluation for liver transplantation. Mortality rates at 24 h, 30 days, and 1-year were compared among groups of patients with different LTRS. Patients with ABO-incompatibility, redo, multivisceral, partial graft and malignancies except for hepatocellular carcinoma were excluded. Data of 48,616 adult liver transplant recipients were extracted from the Scientific Registry of Transplant Recipients between 2002 and 2017.

RESULTS

24-h mortality was 0.9%, 1.0%, 1.1%, 1.7%, 2.3%, 2.0% and 3.5% for patients with LTRS of 0,1,2,3,4, 5 and ≥ 6, respectively (P < 0.001). 30-day mortality was 3.5%, 4.2%, 4.9%, 6.2%, 7.6%, 7.2% and 10.1% respectively (P < 0.001). 1-year mortality was 8.6%, 10.8%, 12.9%, 13.9%, 18.5%, 20.3% and 28.6% respectively (P < 0.001). 10-year survival was 61%, 56%, 57%, 54%, 47%, and 31% for patients with 0, 1, 2, 3, 4, 5 and ≥ 6 points respectively (P < 0.001).

CONCLUSION

Perioperative mortality and long-term survival of patients undergoing LT can be accurately estimated at the time of listing by the LTRS.

Effect of a Real-Time Risk Score on 30-day Readmission Reduction in Singapore

OBJECTIVE

To develop a risk score for the real-time prediction of readmissions for patients using patient specific information captured in electronic medical records (EMR) in Singapore to enable the prospective identification of high-risk patients for enrolment in timely interventions.

METHODS

Machine-learning models were built to estimate the probability of a patient being readmitted within 30 days of discharge. EMR of 25,472 patients discharged from the medicine department at Ng Teng Fong General Hospital between January 2016 and December 2016 were extracted retrospectively for training and internal validation of the models. We developed and implemented a real-time 30-day readmission risk score generation in the EMR system, which enabled the flagging of high-risk patients to care providers in the hospital. Based on the daily high-risk patient list, the various interfaces and flow sheets in the EMR were configured according to the information needs of the various stakeholders such as the inpatient medical, nursing, case management, emergency department, and postdischarge care teams.

RESULTS

Overall, the machine-learning models achieved good performance with area under the receiver operating characteristic ranging from 0.77 to 0.81. The models were used to proactively identify and attend to patients who are at risk of readmission before an actual readmission occurs. This approach successfully reduced the 30-day readmission rate for patients admitted to the medicine department from 11.7% in 2017 to 10.1% in 2019 (p < 0.01) after risk adjustment.

CONCLUSION

Machine-learning models can be deployed in the EMR system to provide real-time forecasts for a more comprehensive outlook in the aspects of decision-making and care provision.

Readmission within three months after inpatient geriatric care-Incidence, diagnosis and associated factors in a Swedish cohort

INTRODUCTION

Readmissions are very costly, in monetary terms but also for the individual patient's safety and health. Only by understanding the reasons and drivers of readmissions, it is possible to ensure quality of care and improve the situation. The aim of this study was to assess inpatient readmissions during the first three months after discharge from geriatric inpatient care regarding main diagnosis and frequency of readmission. Furthermore, the aim was to analyze association between readmission and patient characteristics including demography and socioeconomics, morbidity, physical function, risk screening and care process respectively.

METHODS

The study includes all individuals admitted for inpatient care at three geriatric departments operated by the Stockholm region during 2016. Readmission after discharge was studied within three different time intervals; readmission within 10 days after discharge, within 11-30 days and within 31-90 days, respectively. Main diagnosis at readmission was assessed.

RESULTS

One fourth of the individuals discharged from inpatient geriatric care was readmitted during the first three months after discharge. The most common main diagnoses for readmission were heart failure, chronic obstructive pulmonary disease and pneumonia. Statistically significant risk factors for readmission included age, sex, number of diagnoses at discharge, and to some extent polypharmacy and destination of discharge.

CONCLUSIONS

Several clinical risk factors relating to physical performance and vulnerability were associated with risk of readmission. Socioeconomic information did not add to the predictability. To enable reductions in readmission rates, proactive monitoring of frail individuals afflicted with chronic conditions is necessary, and an integrated perspective including all stakeholders involved is crucial.

Designing an Elderly Hospital Admission Risk Prediction Model in Iran's Hospitals

Background:

The identification of elderly at risk of new functional disabilities in activities of daily living at admission to the hospital may facilitate referral for purposive interventions to prevent decline and institutionalization. This study was aimed at designing a risk prediction model for identifying the elderly at risk of admission in Iran's hospitals.

Materials and Methods:

This is a cross-sectional descriptive study conducted in 2017. In order to formulate and validate a prediction model, the study was done in two development and validation cohort study. Functional decline was defined as a decline of at least one point on the Katz ADL index at follow-up compared with preadmission status.

Results:

In development cohort, the mean age was 71 years including 54% of men and 46% women, 22% of men and 17% of women experienced functional decline after 3 months. In the validation cohort, the mean age was 70 years, including 49% of men and 51% women, 19% of men and 15% of women, functional decline after 3 months was observed.

Conclusion:

On the basis of the findings, aging at risk of hospital admission can be identified by easy designed model with four questions: (1) Is the patient's age more than 85 years? (2) Does the patient's mini mental status <22? (3) Does the patient need help for using general transporting? (4) Has the patient lost weight <5% over the past 6 months and body mass index <18.5? And also geriatrics experts can use the designed model as a predictive tool in order to improve the quality level of healthcare services to elderly as a vulnerable and high risk group. The important point of model is easy to use even for nonspecialists.

Hospital readmissions of patients with heart failure from real world: timing and associated risk factors

AIMS

This study aims to investigate hospital readmissions and timing, as well as risk factors in a real world heart failure (HF) population.

METHODS AND RESULTS

All patients discharged alive in 2016 from Sahlgrenska University Hospital/Östra, Gothenburg, Sweden, with a primary diagnosis of HF were consecutively included. Patient characteristics, type of HF, treatment, and follow-up were registered. Time to first all-cause or HF readmission, as well as number of 1 year readmissions from discharge were recorded. In total, 448 patients were included: 273 patients (mean age 78 ± 11.8 years) were readmitted for any cause within 1 year (readmission rate of 60.9%), and 175 patients (mean age 76.6 ± 13.7) were never readmitted. Among readmissions, 60.1% occurred during the first quarter after index hospitalization, giving a 3 month all-cause readmission rate of 36.6%. HF-related 1 year readmission rate was 38.4%. Patients who were readmitted had significantly more renal dysfunction (52.4% vs. 36.6%, P = 0.001), pulmonary disease (25.6% vs. 15.4%, P = 0.010), and psychiatric illness (24.9% vs. 12.0%, P = 0.001). Number of co-morbidities and readmissions were significantly associated (P < 0.001 for all cause readmission rate and P = 0.012 for 1 year HF readmission rate). Worsening HF constituted 63% of all-cause readmissions. Psychiatric disease was an independent risk factor for 1 month and 1 year all-cause readmissions. Poor compliance to medication was an independent risk factor for 1 month and 1 year HF readmission.

CONCLUSIONS

In our real world cohort of HF patients, frequent hospital readmissions occurred in the early post-discharge period and were mainly driven by worsening HF. Co-morbidity was one of the most important factors for readmission.

A Predictive Risk Model for Infection-Related Hospitalization Among Home Healthcare Patients

Infection prevention is a high priority for home healthcare (HHC), but tools are lacking to identify patients at highest risk of developing infections. The purpose of this study was to develop and test a predictive risk model to identify HHC patients at risk of an infection-related hospitalization or emergency department visit. A nonexperimental study using secondary data was conducted. The Outcome and Assessment Information Set linked with relevant clinical data from 112,788 HHC admissions in 2014 was used for model development (70% of data) and testing (30%). A total of 1,908 patients (1.69%) were hospitalized or received emergency care associated with infection. Stepwise logistic regression models discriminated between individuals with and without infections. Our final model, when classified by highest risk of infection, identified a high portion of those who were hospitalized or received emergent care for an infection while also correctly categorizing 90.5% of patients without infection. The risk model can be used by clinicians to inform care planning. This is the first study to develop a tool for predicting infection risk that can be used to inform how to direct additional infection control intervention resources on high-risk patients, potentially reducing infection-related hospitalizations, emergency department visits, and costs.

Investigating the characteristics and needs of frequently admitting hospital patients: a cross-sectional study in the UK

OBJECTIVES

This study forms the user requirements phase of the OPTIMAL project, which, through a predictive model and supportive intervention, aims to decrease early hospital readmissions. This phase aims to investigate the needs and characteristics of patients who had been admitted to hospital ≥2 times in the past 12 months.

SETTING

This was a cross-sectional study involving patients from Croydon University Hospital (CUH), London, UK.

PARTICIPANTS

A total of 347 patients responded to a postal questionnaire, a response rate of 12.7%. To meet the inclusion criteria, participants needed to be aged ≥18 and have been admitted ≥2 times in the previous 12 months (August 2014-July 2015) to CUH.

PRIMARY AND SECONDARY OUTCOMES

To profile patients identified as frequent admitters to assess gaps in care at discharge or post-discharge. Additionally, to understand the patients' experience of admission, discharge and post-discharge care.

RESULTS

The range of admissions in the past 12 months was 2-30, with a mean of 2.8. At discharge 72.4% (n=231/347) were not given a contact for out-of-hours help. Regression analysis identified patient factors that were significantly associated with frequent admissions (>2 in 12 months), which included age (p=0.008), being in receipt of care (p=0.005) and admission due to a fall (p=0.01), but not receiving polypharmacy. Post-discharge, 41.8% (n=145/347) were concerned about being readmitted to the hospital. In the first 30 days after discharge, over half of patients (54.5% n=189/347) had no contact from a healthcare professional.

CONCLUSION

Considering that social care needs were more of a determinant of admission risk than medical needs, rectifying the lack of integration, communication and the under-utilisation of existing patient services could prevent avoidable problems during the transition of care and help decrease the likelihood of hospital readmission.

What Are They Worth? Six 30-Day Readmission Risk Scores for Medical Inpatients Externally Validated in a Swiss Cohort

BACKGROUND

Several clinical risk scores for unplanned 30-day readmission have been published, but there is a lack of external validation and head-to-head comparison.

OBJECTIVE

Retrospective replication of six clinical risk scores (LACE, HOSPITAL, SEMI, RRS, PARA, Tsui et al.)f DESIGN: Models were fitted with the original intercept and beta coefficients as reported. Otherwise, a logistic model was refitted (SEMI and Tsui et al). We performed subgroup analyses on main admission specialty. This report adheres to the TRIPOD statement for reporting of prediction models.

PARTICIPANTS

We used our prospective cohort of 15,639 medical patients from a Swiss tertiary care institution from 2016 through 2018.

MAIN MEASURES

Thirty-day readmission rate and area under the curve (AUC < 0.50 worse than chance, > 0.70 acceptable, > 0.80 excellent) CONCLUSIONS: Among several readmission risk scores, HOSPITAL, PARA, and the score from Tsui et al. showed the best predictive abilities and have high potential to improve patient care. Interventional research is now needed to understand the effects of these scores when used in clinical routine.

KEY RESULTS

Among the six risk scores externally validated, calibration of the models was overall poor with overprediction of events, except for the HOSPITAL and the PARA scores. Discriminative abilities (AUC) were as follows: LACE 0.53 (95% CI 0.50-0.56), HOSPITAL 0.73 (95% CI 0.72-0.74), SEMI 0.47 (95% CI 0.46-0.49), RRS 0.64 (95% CI 0.62-0.66), PARA 0.72 (95% CI 0.72-0.74), and the score from Tsui et al. 0.73 (95% CI 0.72-0.75). Performance in subgroups did not differ from the overall performance, except for oncology patients in the PARA score (0.57, 95% CI 0.54-0.60), and nephrology patients in the SEMI index (0.25, 95% CI 0.18-0.31), respectively.

A machine learning model for predicting risk of hospital readmission within 30 days of discharge: validated with LACE index and patient at risk of hospital readmission (PARR) model

The objective of this study was to design and develop a predictive model for 30-day risk of hospital readmission using machine learning techniques. The proposed predictive model was then validated with the two most commonly used risk of readmission models: LACE index and patient at risk of hospital readmission (PARR). The study cohort consisted of 180,118 admissions with 22,565 (12.5%) of actual readmissions within 30 days of hospital discharge, from 01 Jan 2015 to 31 Dec 2016 from two Auckland-region hospitals. We developed a machine learning model to predict 30-day readmissions using the model types XGBoost, Random Forests, and Adaboost with decision stumps as a base learner with different feature combinations and preprocessing procedures. The proposed model achieved the F1-score (0.386 ± 0.006), sensitivity (0.598 ± 0.013), positive predictive value (PPV) (0.285 ± 0.004), and negative predictive value (NPV) (0.932 ± 0.002). When compared with LACE and PARR(NZ) models, the proposed model achieved better F1-score by 12.7% compared with LACE and 23.2% compared with PARR(NZ). The mean sensitivity of the proposed model was 6.0% higher than LACE and 41% higher than PARR(NZ). The mean PPV was 15.9% and 14.6% higher than LACE and PARR(NZ) respectively. We presented an all-cause predictive model for 30-day risk of hospital readmission with an area under the receiver operating characteristics (AUROC) of 0.75 for the entire dataset. Graphical abstract.

Medication-related harm in older adults following hospital discharge: development and validation of a prediction tool

OBJECTIVES

To develop and validate a tool to predict the risk of an older adult experiencing medication-related harm (MRH) requiring healthcare use following hospital discharge.

DESIGN, SETTING, PARTICIPANTS

Multicentre, prospective cohort study recruiting older adults (≥65 years) discharged from five UK teaching hospitals between 2013 and 2015.

PRIMARY OUTCOME MEASURE

Participants were followed up for 8 weeks in the community by senior pharmacists to identify MRH (adverse drug reactions, harm from non-adherence, harm from medication error). Three data sources provided MRH and healthcare use information: hospital readmissions, primary care use, participant telephone interview. Candidate variables for prognostic modelling were selected using two systematic reviews, the views of patients with MRH and an expert panel of clinicians. Multivariable logistic regression with backward elimination, based on the Akaike Information Criterion, was used to develop the PRIME tool. The tool was internally validated.

RESULTS

1116 out of 1280 recruited participants completed follow-up (87%). Uncertain MRH cases ('possible' and 'probable') were excluded, leaving a tool derivation cohort of 818. 119 (15%) participants experienced 'definite' MRH requiring healthcare use and 699 participants did not. Modelling resulted in a prediction tool with eight variables measured at hospital discharge: age, gender, antiplatelet drug, sodium level, antidiabetic drug, past adverse drug reaction, number of medicines, living alone. The tool's discrimination C-statistic was 0.69 (0.66 after validation) and showed good calibration. Decision curve analysis demonstrated the potential value of the tool to guide clinical decision making compared with alternative approaches.

CONCLUSIONS

The PRIME tool could be used to identify older patients at high risk of MRH requiring healthcare use following hospital discharge. Prior to clinical use we recommend the tool's evaluation in other settings.

Journey to multimorbidity: longitudinal analysis exploring cardiovascular risk factors and sociodemographic determinants in an urban setting

OBJECTIVE

To study the social determinants and cardiovascular risk factors for multimorbidity and the acquisition sequence of multimorbidity.

DESIGN

Longitudinal study based on anonymised primary care data.

SETTING

General practices in an urban multiethnic borough in London, UK.

PARTICIPANTS

332 353 patients aged ≥18 years.

MAIN OUTCOME MEASURES

Clinical and sociodemographic characteristics of patients with multimorbidity, defined as ≥3 of 12 long-term conditions (LTCs) selected according to high predicted healthcare use. Multilevel logistic regression was used to model social determinants and cardiovascular risk factors. Alluvial plots were constructed to illustrate multimorbidity acquisition sequences according to age, ethnicity and social deprivation.

RESULTS

5597 (1.7%) patients had ≥3 selected LTCs, the 'multimorbidity cohort'. The the most common LTCs were diabetes (63.0%) and chronic pain (CP) (42.8%). Social deprivation and ethnicity were independent determinants of multimorbidity: most compared with the least deprived quintile (adjusted OR (AOR) 1.56 (95% CI 1.41 to 1.72)); South Asian compared with white ethnicity (AOR 1.44 (95% CI 1.29 to 1.61)); and black compared with white ethnicity (AOR 0.86 (95% CI 0.80 to 0.92)). The included cardiovascular risk factors were relatively strong determinants of multimorbidity: hypertension (AOR 5.05 (95% CI 4.69 to 5.44)), moderate obesity (AOR 3.41 (95% CI 3.21 to 3.63)) and smoking (AOR 2.30 (95% CI 2.16 to 2.45)). The most common initial onset conditions were diabetes and depression; diabetes particularly in older and black ethnic groups; and depression particularly in younger, more deprived and white ethnicity groups. CP was less common as an initial condition.

CONCLUSION

Our findings confirm the importance of age, social deprivation and ethnicity as determinants of multimorbidity. Smoking, obesity and hypertension as cardiovascular risk factors were stronger determinants of multimorbidity than deprivation or ethnicity. The acquisition sequence of multimorbidity is patterned by sociodemographic determinants. Understanding onset conditions of multimorbidity and cardiovascular cardiovascular risk factors may lead to the development of interventions to slow the progression of multimorbidity.

Prospectively identifying older adults at risk for potentially avoidable hospitalizations in Korea using population-based data

OBJECTIVE

The purpose of this study is to develop and validate a prediction model to identify older adults at risk for potentially avoidable hospitalizations (PAHs) using variables readily available in routinely collected health administrative data.

DESIGN

Population-based, retrospective observational study using National Health Insurance Service-National Sample Cohort (NHIS-NSC) data.

SETTING

Primary care settings in South Korea.

PARTICIPANTS

The sample includes 113 612 older NHI beneficiaries in the year 2011, among which 2856 had one or more PAHs in 2012.

METHODS

We examined multi-dimensional risk factors of PAHs in the base year and developed and validated prediction models of hospitalization risk using the following year. The predictive power of the developed models was examined with samples generated by the bootstrap method. The performance of the final model was evaluated with further test statistics at different risk thresholds.

MAIN OUTCOME MEASURE

Predicting PAHs.

RESULTS

The c-statistic of the final predictive model was 0.784 (CI: 0.769-0.799). The final model including all selected predictors showed the greatest marginal improvement (integrated discrimination improvement of 365%) compared to the base model. The positive predictive value was good.

CONCLUSIONS

Our prediction model based on health administrative data can assist the insurer and practitioners to proactively identify and intervene with older adults at risk for a PAH.

Person-centred, integrated and pro-active care for multi-morbid elderly with advanced care needs: a propensity score-matched controlled trial

BACKGROUND

Person-centred care (PCC) focusing on personalised goals and care plans derived from "What matters to you?" has an impact on single disease outcomes, but studies on multi-morbid elderly are lacking. Furthermore, the combination of PCC, Integrated Care (IC) and Pro-active care are widely recognised as desirable for multi-morbid elderly, yet previous studies focus on single components only, leaving synergies unexplored. The effect of a synergistic intervention, which implements 1) Person-centred goal-oriented care driven by "What matters to you?" with 2) IC and 3) pro-active care is unknown.

METHODS

Inspired by theoretical foundations, complexity science, previous health service research and a patient-driven evaluation of care quality, we designed the Patient-Centred Team (PACT) intervention across primary and secondary care. The PACT team collaborate with the patient to make and deliver a person-centred, integrated and proactive multi-morbidity care-plan. The control group receives conventional care. The study design is a pragmatic six months prospective, controlled clinical trial based on hospital electronic health record data of 439 multi-morbid frail elderly at risk for emergency (re) admissions referred to PACT and 779 propensity score matched controls in Norway, 2014-2016. Outcomes are emergency admissions, the sum of emergency inpatient bed days, 30-day readmissions, planned and emergency outpatient visits and mortality at three and six months follow-up.

RESULTS

The Rate Ratios (RR) for emergency admissions was 0,9 (95%CI: 0,82-0,99), for sum of emergency bed days 0,68 (95%CI:0,52-0,79) and for 30-days emergency readmissions 0,72 (95%CI: 0,41-1,24). RRs were 2,3 (95%CI: 2,02-2,55) and 0,9 (95%CI: 0,68-1,20) for planned and emergency outpatient visits respectively. The RR for death at 3 months was 0,39 (95% CI: 0,22-0,70) and 0,57 (95% CI: 0,34-0,94) at 6 months.

CONCLUSION

Compared with propensity score matched controls, the care process of frail multi-morbid elderly who received the PACT intervention had a reduced risk of high-level emergency care, increased use of low-level planned care, and substantially reduced mortality risk. Further study of process differences between groups is warranted to understand the genesis of these results better.

TRIAL REGISTRATION

ClinicalTrials.gov (identifier: NCT02541474 ), registered Sept 2015.

Emergence of digital biomarkers to predict and modify treatment efficacy: machine learning study

OBJECTIVES

Development of digital biomarkers to predict treatment response to a digital behavioural intervention.

DESIGN

Machine learning using random forest classifiers on data generated through the use of a digital therapeutic which delivers behavioural therapy to treat cardiometabolic disease. Data from 13 explanatory variables (biometric and engagement in nature) generated in the first 28 days of a 12-week intervention were used to train models. Two levels of response to treatment were predicted: (1) systolic change ≥10 mm Hg (SC model), and (2) shift down to a blood pressure category of elevated or better (ER model). Models were validated using leave-one-out cross validation and evaluated using area under the curve receiver operating characteristics (AUROC) and specificity- sensitivity. Ability to predict treatment response with a subset of nine variables, including app use and baseline blood pressure, was also tested (models SC-APP and ER-APP).

SETTING

Data generated through ad libitum use of a digital therapeutic in the USA.

PARTICIPANTS

Deidentified data from 135 adults with a starting blood pressure ≥130/80, who tracked blood pressure for at least 7 weeks using the digital therapeutic.

RESULTS

The SC model had an AUROC of 0.82 and a sensitivity of 58% at a specificity of 90%. The ER model had an AUROC of 0.69 and a sensitivity of 32% at a specificity at 91%. Dropping explanatory variables related to blood pressure resulted in an AUROC of 0.72 with a sensitivity of 42% at a specificity of 90% for the SC-APP model and an AUROC of 0.53 for the ER-APP model.

CONCLUSIONS

Machine learning was used to transform data from a digital therapeutic into digital biomarkers that predicted treatment response in individual participants. Digital biomarkers have potential to improve treatment outcomes in a digital behavioural intervention.

Effect on secondary care of providing enhanced support to residential and nursing home residents: a subgroup analysis of a retrospective matched cohort study

BACKGROUND

Thirteen residential care homes and 10 nursing homes specialising in older people in Rushcliffe, England, participated in an improvement programme. The enhanced support provided included regular visits from named general practitioners and additional training for care home staff. We assessed and compared the effect on hospital use for residents in residential and nursing homes, respectively.

METHODS

Using linked care home and administrative hospital data, we examined people aged 65 years or over who moved to a participating care home between 2014 and 2016 (n=568). We selected matched control residents who had similar characteristics to the residents receiving enhanced support and moved to similar care homes not participating in the enhanced support (n=568). Differences in hospital use were assessed for residents of each type of care home using multivariable regression.

RESULTS

Residents of participating residential care homes showed lower rates of potentially avoidable emergency admissions (rate ratio 0.50, 95% CI 0.30 to 0.82), emergency admissions (rate ratio 0.60, 95% CI 0.42 to 0.86) and Accident & Emergency attendances (0.57, 95% CI 0.40 to 0.81) than matched controls. Hospital bed days, outpatient attendances and the proportion of deaths that occurred out of hospital were not statistically different. For nursing home residents, there were no significant differences for any outcome.

CONCLUSIONS

The enhanced support was associated with lower emergency hospital use for older people living in residential care homes but not for people living in nursing homes. This might be because there was more potential to reduce emergency care for people in residential care homes. In nursing homes, improvement programmes may need to be more tailored to residents' needs or the context of providing care in that setting.

Observational study to determine the utility of hospital administrative data to support case finding of English patients at higher risk of severe healthcare-related harm

OBJECTIVES

To identify ways of using routine hospital data to improve the efficiency of retrospective reviews of case records for identifying avoidable severe harm DESIGN: Development and testing of thresholds and criteria for two indirect indicators of healthcare-related harm (long length of stay (LOS) and emergency readmission) to determine the yield of specified harms coded in Hospital Episode Statistics (HES).

SETTING

Acute National Health Service hospitals in England.

PARTICIPANTS

HES for acute myocardial infarction (AMI), bowel cancer surgery and hip replacement admissions from 2014 to 2015.

INTERVENTIONS

Case-mix-adjusted linear regression models were used to determine expected LOS. Different thresholds were examined to determine the association with harm. Screening criteria for readmission included time to readmission, length of readmission and diagnoses in initial admission and readmission. The association with harm was examined for each criterion.

RESULTS

The proportions of AMI cases with a harm code increased from 14% among all cases to 47% if a threshold of three times the expected LOS was used. For hip replacement the respective increase was from 10% to 51%. However as the number of patients at these higher thresholds was small, the overall proportion of harm identified is relatively small (15%, 19%, 9% and 8% among AMI, urgent bowel surgery, elective bowel surgery and hip replacement cohorts, respectively). Selection of the time to readmission had an effect on the yield of harms but this varied with condition. At least 50% of surgical patients had a harm code if readmitted within 7 days compared with 21% of patients with AMI.

CONCLUSIONS

Our approach would select a substantial number of patients for case record review. Many of these cases would contain no evidence of healthcare-related harm. In practice, Trusts may choose how many reviews it is feasible to do in advance and then select random samples of cases that satisfy the screening criteria.

Family Medicine Patients Have Shorter Length of Stay When Cared for on a Family Medicine Inpatient Service

Introduction: Hospitalists have been shown to have shorter lengths of stays than physicians with concurrent outpatient practices. However, hospitalists at academic medical centers may be less aware of local resources that can support the hospital to home transition for local primary care patients. We hypothesized that local family medicine patients admitted to a family medicine inpatient service have shorter length of stay than those admitted to general hospitalist services which also care for tertiary patients at an academic medical center. Methods: A retrospective cohort study was conducted at an academic medical center with a department of family medicine providing primary care to over 80 000 local patients. A total of 3100 consecutive family medicine patients admitted to either the family medicine inpatient service or a general medicine inpatient service over 3 years were studied. The primary outcome was length of stay, which was adjusted using multivariate linear regression for demographics, prior utilization, diagnosis, and disease severity. Results: Adjusted length of stay was 33% longer (95% CI 24%-44%) for local family medicine patients admitted to general medicine inpatient services as compared with the family medicine inpatient service. Readmission rates within 30 days were not different (19% vs 16%, P = .14). Conclusions: Local primary care patients were safely discharged from the hospital sooner on the family medicine inpatient service than on general medicine inpatient services. This is likely because the family physicians staffing their inpatient service are more familiar with outpatient resources that can be effectively marshaled to help local patients with the transition from hospital to home.

An Integrated Framework for Reducing Hospital Readmissions using Risk Trajectories Characterization and Discharge Timing Optimization

When patients leave the hospital for lower levels of care, they experience a risk of adverse events on a daily basis. The advent of value-based purchasing among other major initiatives has led to an increasing emphasis on reducing the occurrences of these post-discharge adverse events. This has spurred the development of new prediction technologies to identify which patients are at risk for an adverse event as well as actions to mitigate those risks. Those actions include pre-discharge and post-discharge interventions to reduce risk. However, traditional prediction models have been developed to support only post-discharge actions; predicting risk of adverse events at the time of discharge only. In this paper we develop an integrated framework of risk prediction and discharge optimization that supports both types of interventions: discharge timing and post-discharge monitoring. Our method combines a kernel approach for capturing the non-linear relationship between length of stay and risk of an adverse event, with a Principle Component Analysis method that makes the resulting estimation tractable. We then demonstrate how this prediction model could be used to support both types of interventions by developing a simple and easily implementable discharge timing optimization.

Regional variations in trajectories of long-term readmission rates among patients in England with heart failure

BACKGROUND

We aimed to compare the characteristics and types of heart failure (HF) patients termed "high-impact users", with high long-term readmission rates, in different regions in England. This will allow clinical factors to be identified in areas with potentially poor quality of care.

METHODS

Patients with a primary diagnosis of heart failure (HF) in the period 2008-2009 were identified using nationally representative primary care data linked to national hospital data and followed up for 5 years. Group-based trajectory models and sequence analysis were applied to their readmissions.

RESULTS

In each of the 8 NHS England regions, multiple discrete groups were identified. All the regions had high-impact users. The group with an initially high readmission rate followed by a rapid decline in the rate ranged from 2.5 to 11.3% across the regions. The group with constantly high readmission rate compared with other groups ranged from 1.9 to 12.1%. Covariates that were commonly found to have an association with high-impact users among most of the regions were chronic respiratory disease, chronic renal disease, stroke, anaemia, mood disorder, and cardiac arrhythmia. Respiratory tract infection, urinary infection, cardiopulmonary signs and symptoms and exacerbation of heart failure were common causes in the sequences of readmissions among high-impact users in all regions.

CONCLUSION

There is regional variation in England in readmission and mortality rates and in the proportions of HF patients who are high-impact users.

Development of an iterative validation process for a 30-day hospital readmission prediction index

PURPOSE

A study was conducted to determine if an iterative validation process could maintain or improve the discriminative and predictive capabilities of a 30-day hospital readmission prediction index over 2.5 years.

METHODS

Patient admissions were retrospectively identified using the electronic medical record. The receiver operating characteristic curve was used to assess model discrimination. Prediction index specificity, sensitivity, and positive and negative predictive values were also assessed. A rolling iterative validation process was developed in which patient admissions were divided into 3-month cohorts. Each cohort was analyzed individually and then included into the cumulative patient cohort and analyzed again.

RESULTS

From 121,277 patient visits, an iterative validation approach maintained the discrimination (0.71 to 0.72), predictive validity, and overall accuracy (80.9% to 81.7%) of the 30-day readmission prediction index over 2.5 years. Index sensitivity and negative predictive value increased from baseline while specificity and positive predictive value remained largely unchanged. None of the assessed index parameters diminished or became less useful over the course of the study.

CONCLUSION

An internal iterative validation process based on frequentist statistics maintained the discriminative ability and accuracy of a readmission index over 2.5 years despite numerous changes in the variables associated with readmission in the patient population.

Development and Prospective Validation of a Machine Learning-Based Risk of Readmission Model in a Large Military Hospital

BACKGROUND

Thirty-day hospital readmissions are a quality metric for health care systems. Predictive models aim to identify patients likely to readmit to more effectively target preventive strategies. Many risk of readmission models have been developed on retrospective data, but prospective validation of readmission models is rare. To the best of our knowledge, none of these developed models have been evaluated or prospectively validated in a military hospital.

OBJECTIVES

The objectives of this study are to demonstrate the development and prospective validation of machine learning (ML) risk of readmission models to be utilized by clinical staff at a military medical facility and demonstrate the collaboration between the U.S. Department of Defense's integrated health care system and a private company.

METHODS

We evaluated multiple ML algorithms to develop a predictive model for 30-day readmissions using data from a retrospective cohort of all-cause inpatient readmissions at Madigan Army Medical Center (MAMC). This predictive model was then validated on prospective MAMC patient data. Precision, recall, accuracy, and the area under the receiver operating characteristic curve (AUC) were used to evaluate model performance. The model was revised, retrained, and rescored on additional retrospective MAMC data after the prospective model's initial performance was evaluated.

RESULTS

Within the initial retrospective cohort, which included 32,659 patient encounters, the model achieved an AUC of 0.68. During prospective scoring, 1,574 patients were scored, of whom 152 were readmitted within 30 days of discharge, with an all-cause readmission rate of 9.7%. The AUC of the prospective predictive model was 0.64. The model achieved an AUC of 0.76 after revision and addition of further retrospective data.

CONCLUSION

This work reflects significant collaborative efforts required to operationalize ML models in a complex clinical environment such as that seen in an integrated health care system and the importance of prospective model validation.

Defining a 'frequent admitter' phenotype among patients with repeat heart failure admissions

AIMS

We aimed to identify a 'frequent admitter' phenotype among patients admitted for acute decompensated heart failure (HF).

METHODS AND RESULTS

We studied 10 363 patients in a population-based prospective HF registry (2008-2012), segregated into clusters based on their 3-year HF readmission frequency trajectories. Using receiver-operating characteristic analysis, we identified the index year readmission frequency threshold that most accurately predicts HF admission frequency clusters. Two clusters of HF patients were identified: a high frequency cluster (90.9%, mean 2.35 ± 3.68 admissions/year) and a low frequency cluster (9.1%, mean 0.50 ± 0.81 admission/year). An index year threshold of two admissions was optimal for distinguishing between clusters. Based on this threshold, 'frequent admitters', defined as patients with ≥ 2 HF admissions in the index year (n = 2587), were of younger age (68 ± 13 vs 69 ± 13 years), more often male (58% vs. 54%), smokers (38.4% vs. 34.4%) and had lower left ventricular ejection fraction (37 ± 17 vs. 41 ± 17%) compared to 'non-frequent admitters' (< 2 HF admissions in the index year; n = 7776) (all P < 0.001). Despite similar rates of advanced care utilization, frequent admitters had longer length of stay (median 4.3 vs. 4.0 days), higher annual inpatient costs (€ 7015 vs. € 2967) and higher all-cause mortality at 3 years compared to the non-frequent admitters (adjusted odds ratio 2.33, 95% confidence interval 2.11-2.58; P < 0.001).

CONCLUSION

'Frequent admitters' have distinct clinical characteristics and worse outcomes compared to non-frequent admitters. This study may provide a means of anticipating the HF readmission burden and thereby aid in healthcare resource distribution relative to the HF admission frequency phenotype.

Development of a risk predictive scoring system to identify patients at risk of representation to emergency department: a retrospective population-based analysis in Australia

OBJECTIVE

To examine the characteristics of frequent visitors (FVs) to emergency departments (EDs) and develop a predictive model to identify those with high risk of a future representations to ED among younger and general population (aged ≤70 years).

DESIGN AND SETTING

A retrospective analysis of ED data targeting younger and general patients (aged ≤70 years) were collected between 1 January 2009 and 30 June 2016 from a public hospital in Australia.

PARTICIPANTS

A total of 343 014 ED presentations were identified from 170 134 individual patients.

MAIN OUTCOME MEASURES

Proportion of FVs (those attending four or more times annually), demographic characteristics (age, sex, indigenous and marital status), mode of separation (eg, admitted to ward), triage categories, time of arrival to ED, referral on departure and clinical conditions. Statistical estimates using a mixed-effects model to develop a risk predictive scoring system.

RESULTS

The FVs were characterised by young adulthood (32.53%) to late-middle (26.07%) aged patients with a higher proportion of indigenous (5.7%) and mental health-related presentations (10.92%). They were also more likely to arrive by ambulance (36.95%) and leave at own risk without completing their treatments (9.8%). They were also highly associated with socially disadvantage groups such as people who have been divorced, widowed or separated (12.81%). These findings were then used for the development of a predictive model to identify potential FVs. The performance of our derived risk predictive model was favourable with an area under the receiver operating characteristic (ie, C-statistic) of 65.7%.

CONCLUSION

The development of a demographic and clinical profile of FVs coupled with the use of predictive model can highlight the gaps in interventions and identify new opportunities for better health outcome and planning.