Patient related factors in frequent readmissions: the influence of condition, access to services and patient choice

Risk factors for hospitalisation in community-dwelling pre-frail and frail older people: results of a longitudinal study

BACKGROUND

Older adults account for a large proportion of hospital admissions. In this study we aim to bridge a gap between medical and psychosocial factors in predicting hospitalisation.

METHODS

Demographic and social characteristics of community-dwelling pre-frail and frail older people were collected by questionnaires every six months during a two year follow-up. Hospital admission within this period was dichotomised as yes/no. To define pre-frailty and frailty the Fried frailty criteria were used. Analysis of risk factors for hospitalisation was performed using multivariable logistic regression.

RESULTS

Hospitalised participants (n = 1803) were more often male and frail in comparison to not-hospitalised participants. They also experienced more chronic diseases (54.5% ≥ 4 chronic diseases), poorer self-perceived health (SPH) (76.4% fair to very poor) and lack of informal care (20.1%). In multivariable logistic regression male gender (Odds ratio (OR) 1.65, p < 0.001), frailty (vs. pre-frailty) (OR 1.66, p = 0.002), reporting lower SPH (OR 3.12, p < 0.001) and lacking informal care (OR 1.69, p < 0.001) showed significant associations with hospital admission. Subgroup analysis of pre-frail and frail participants, showed consistent associations between male gender (respectively OR 1.61, p < 0.001 ; OR 1.72, p = 0.085), lower SPH (OR 2.23, p = 0.001; OR 31.16, p < 0.001), lack of informal care (OR 1.64, p = 0.005; OR 2.63, p = 0.012) and hospitalisation.

CONCLUSION

Frailty, male gender, lower SPH and lack of informal care are risk factors for hospitalisation within community-dwelling older people, showing the need of a holistic approach to possibly prevent hospitalisation. Further research should focus on evaluating individual factors for hospitalisation, particularly targeting pre-frail individuals.

Health mediation does not reduce the readmission rate of frequent users of emergency departments living in precarious conditions: what lessons can be learned from this randomised controlled trial?

BACKGROUND

Severe overcrowding of emergency departments (EDs) affects the quality of healthcare. One factor of overcrowding is precariousness, but it has rarely been considered a key factor in designing interventions to improve ED care. Health mediation (HM) aims to facilitate access to rights, prevention, and care for the most vulnerable persons and to raise awareness among healthcare providers about obstacles in accessing healthcare. The primary aim was to determine whether HM intervention for frequent users of EDs (FUED) living in precarious conditions could reduce the readmission rate at 90 days.

METHODS

Between February 2019 and May 2022, we enrolled and interviewed 726 FUED in four EDs of southeastern France in this randomised controlled trial. The HM intervention started in the ED and lasted 90 days. In addition to the primary endpoint (first readmission at 90 days), secondary endpoints (readmission at 30 and 180 days, number of hospitalisations at 30, 90, 180 days, admissions for the same reasons as the first admission) were also studied. The outcomes were measured in the ED information systems. Statistical methods included an intention-to-treat analysis and a per-protocol analysis. Comparisons were adjusted for gender, age, ED, and health mediator.

RESULTS

46% of patients reported attending the ED because they felt their life was in danger, and 42% had been referred to the ED by the emergency medical dispatch centre or their GP; 40% of patients were considered to be in a serious condition by ED physicians. The proportion of patients who were readmitted at 90 days was high but did not differ between the control and the HM intervention groups (31.7% vs. 36.3%, p = 0.23). There was no significant difference in any of the secondary outcome measures between the control and HM intervention groups. Per-protocol analysis also showed no significant difference for the primary and secondary endpoints.

CONCLUSIONS

This randomised controlled trial did not show that our health mediation intervention was effective in reducing the use of emergency services by FUED living in precarious conditions. Some limitations are discussed: the duration of the intervention (90 days), the long-term effects (> 6 months), the involvement of the ED staff.

TRIAL REGISTRATION

Registered on clinicaltrials.gov as NCT03660215 on 4th September 2018.

Patients with repeated acute admissions to somatic departments: sociodemographic characteristics, disease burden, and contact with primary healthcare sector - a retrospective register-based case-control study

BACKGROUND

Healthcare systems face escalating capacity challenges and patients with repeated acute admissions strain hospital resources disproportionately. However, studies investigating the characteristics of such patients across all public healthcare providers in a universal healthcare system are lacking.

OBJECTIVE

To investigate characteristics of patients with repeated acute admissions (three or more acute admissions within a calendar year) in regard to sociodemographic characteristics, disease burden, and contact with the primary healthcare sector.

METHODS

This matched register-based case-control study investigated repeated acute admissions from 1 January 2014 to 31 December 2018, among individuals, who resided in four Danish municipalities. The study included 6169 individuals with repeated acute admissions, matched 1:4 to individuals with no acute admissions and one to two acute admissions, respectively. Group comparisons were conducted using conditional logistic regression.

RESULTS

Receiving social benefits increased the odds of repeated acute admissions 9.5-fold compared with no acute admissions (odds ratio (OR) 9.5; 95% confidence interval (CI) 8.5; 10.6) and 3.4-fold compared with one to two acute admissions (OR 3.4; 95% CI 3.1; 3.7). The odds of repeated acute admissions increased with the number of used medications and chronic diseases. Having a mental illness increased the odds of repeated acute admissions 5.8-fold when compared with no acute admissions (OR 5.7; 95% CI 5.2; 6.4) and 2.3-fold compared with one to two acute admissions (OR 2.3; 95% CI 2.1; 2.5). Also, high use of primary sector services (e.g. nursing care) increased the odds of repeated acute admissions when compared with no acute admissions and one to two acute admissions.

CONCLUSIONS

This study pinpointed key factors encompassing social status, disease burden, and healthcare utilisation as pivotal markers of risk for repeated acute admissions, thus identifying high-risk patients and facilitating targeted intervention.

Predictors of frequency of 1-year readmission in adult patients with diabetes

Diabetes mellitus (DM) is the third most common chronic condition associated with frequent hospital readmissions. Predictors of the number of readmissions within 1 year among patients with DM are less often studied compared with those of 30-day readmission. This study aims to identify predictors of number of readmissions within 1 year amongst adult patients with DM and compare different count regression models with respect to model fit. Data from 2008 to 2015 were extracted from the electronic medical records of the National University Hospital, Singapore. Inpatients aged ≥ 18 years at the time of index admission with a hospital stay > 24 h and survived until discharge were included. The zero-inflated negative binomial (ZINB) model was fitted and compared with three other count models (Poisson, zero-inflated Poisson and negative binomial) in terms of predicted probabilities, misclassification proportions and model fit. Adjusted for other variables in the model, the expected number of readmissions was 1.42 (95% confidence interval [CI] 1.07 to 1.90) for peripheral vascular disease, 1.60 (95% CI 1.34 to 1.92) for renal disease and 2.37 (95% CI 1.67 to 3.35) for Singapore residency. Number of emergency visits, number of drugs and age were other significant predictors, with length of stay fitted as a zero-inflated component. Model comparisons suggested that ZINB provides better prediction than the other three count models. The ZINB model identified five patient characteristics and two comorbidities associated with number of readmissions. It outperformed other count regression models but should be validated before clinical adoption.

Health mediation: an intervention mode for improving emergency department care and support for patients living in precarious conditions

BACKGROUND

Severe overcrowding of emergency departments (EDs) impacts the quality of healthcare. One factor of this overcrowding is precariousness, but it has rarely been considered a key factor in designing interventions to improve ED care. Health mediation (HM) aims to facilitate access to rights, prevention, and care for the most vulnerable persons and to raise awareness among healthcare providers about obstacles in accessing healthcare. We here present the results of an ancillary qualitative study to explore the prospects regarding a health mediation intervention implemented in EDs for deprived persons who are frequent ED users, from professionals' and patients' perspectives.

METHODS

Design, data collection, and data analysis were done according to a psychosocial approach, based on thematic content analysis and semi-structured interviews of 16 frequent ED users and deprived patients exposed to HM and of 14 professionals in 4 EDs of South-eastern France.

RESULTS

All patients reported multifactorial distress. Most of them expressed experiencing isolation and powerlessness, and lacking personal resources to cope with healthcare. They mentioned the use of ED as a way of quickly meeting a professional to respond to their suffering, and recognized the trustworthy alliance with health mediators (HMrs) as a means to put them back in a healthcare pathway. The presence of HMrs in EDs was appreciated by ED professionals because HMrs responded to requests they were not able to access and were perceived as an efficient support for caring for deprived persons in emergency contexts.

CONCLUSIONS

Our results are in favour of health mediation in EDs as a promising solution, requested by patients and ED professionals, to cope with frequent ED users and deprived patients. Our results could also be used to adapt other strategies for the most vulnerable populations to reduce the frequency of ED readmissions. At the interface of the patients' health experience and the medico-social sector, HM could complete the immediate responses to medical needs given in EDs and contribute in alleviating the social inequalities of health.

The Path to Sustainable Healthcare: Implementing Care Transition Teams to Mitigate Hospital Readmissions and Improve Patient Outcomes

Introduction Hospital readmissions within 30 days suggest care quality issues and increased mortality risks. They result from ineffective initial treatment, poor discharge planning, and inadequate post-acute care. These high readmission rates harm patient outcomes and financially strain healthcare institutions, inviting penalties and discouraging potential patients. Enhancing inpatient care, care transitions, and case management is crucial to lowering readmissions. Our research underscores the role of care transition teams in reducing readmissions and financial stress in hospitals. By consistently applying transition strategies and focusing on high-quality care, we can improve patient outcomes and ensure hospital success in the long run. Methods This two-phase study investigated readmission rates and risk factors in a community hospital from May 2017 to November 2022. Phase 1 determined a baseline readmission rate and identified individual risk factors using logistic regression. In phase 2, a care transition team addressed these factors by providing post-discharge patient support through phone calls and assessing social determinants of health (SDOH). Readmission data from the intervention period was compared to baseline data using statistical tests. Data, including demographics, medical conditions, and comorbidities, were collected via electronic medical records and the International Classification of Diseases (ICD-10 codes). The study focused on patients aged 20-80 with readmissions within 30 days. Exclusions were made to minimize confounding effects from unmeasured comorbidities and ensure an accurate representation of factors affecting readmissions. Results In the study's initial phase, 74,153 patients participated, with an 18% mean readmission rate. Women accounted for 46% of readmissions, and the white population had the highest rate (49%). The 40-59 age group showed a higher readmission rate than other age groups, and certain health factors were identified as risk factors for 30-day readmission. In the subsequent phase, a care transition team intervened with high-risk groups using an SDOH questionnaire. They contacted 432 patients, resulting in a reduced overall readmission rate of 9%. The 60-79 age group and the Hispanic population experienced higher readmission rates, and the previously identified health factors remained significant risk factors. Conclusion This study emphasizes the crucial role of care transition teams in reducing hospital readmission rates and easing the financial strain on healthcare institutions. By identifying and addressing individual risk factors, the care transition team effectively lowered the overall readmission rate from 18% to 9%. Continually implementing transition strategies and prioritizing high-quality care focused on minimizing readmissions are essential for improving patient outcomes and long-term hospital success. Healthcare providers should consider utilizing care transition teams and social determinants of health assessments to better understand and manage risk factors and tailor post-discharge support for patients at higher risk of readmission.

A hybrid machine learning framework to improve prediction of all-cause rehospitalization among elderly patients in Hong Kong

BACKGROUND

Accurately estimating elderly patients' rehospitalisation risk benefits clinical decisions and service planning. However, research in rehospitalisation and repeated hospitalisation yielded only models with modest performance, and the model performance deteriorates rapidly as the prediction timeframe expands beyond 28 days and for older participants.

METHODS

A temporal zero-inflated Poisson (tZIP) regression model was developed and validated retrospectively and prospectively. The data of the electronic health records (EHRs) contain cohorts (aged 60+) in a major public hospital in Hong Kong. Two temporal offset functions accounted for the associations between exposure time and parameters corresponding to the zero-inflated logistic component and the Poisson distribution's expected count. tZIP was externally validated with a retrospective cohort's rehospitalisation events up to 12 months after the discharge date. Subsequently, tZIP was validated prospectively after piloting its implementation at the study hospital. Patients discharged within the pilot period were tagged, and the proposed model's prediction of their rehospitalisation was verified monthly. Using a hybrid machine learning (ML) approach, the tZIP-based risk estimator's marginal effect on 28-day rehospitalisation was further validated, competing with other factors representing different post-acute and clinical statuses.

RESULTS

The tZIP prediction of rehospitalisation from 28 days to 365 days was achieved at above 80% discrimination accuracy retrospectively and prospectively in two out-of-sample cohorts. With a large margin, it outperformed the Cox proportional and linear models built with the same predictors. The hybrid ML revealed that the risk estimator's contribution to 28-day rehospitalisation outweighed other features relevant to service utilisation and clinical status.

CONCLUSIONS

A novel rehospitalisation risk model was introduced, and its risk estimators, whose importance outweighed all other factors of diverse post-acute care and clinical conditions, were derived. The proposed approach relies on four easily accessible variables easily extracted from EHR. Thus, clinicians could visualise patients' rehospitalisation risk from 28 days to 365 days after discharge and screen high-risk older patients for follow-up care at the proper time.

Impact of Pharmacophobia and Pharmacophilia on Perception of Medication Use and Self-Medication Behaviors in Saudi Arabia

Objectives

Pharmacophilia and pharmacophobia have positive and negative effects on treatment outcomes. There exist lacune of information on these phenomena in middle east countries. Therefore the current study aimed to determine the pattern and predictors of Pharmacophilia and pharmacophobia in Saudi Arabia. The study further explored the impact of these phenomena on the perception of medication use and self-medication behaviours.

Experimental design

An online questionnaire-based cross-sectional study was conducted during March-May 2021. Saudi residents aged 15 years or older were eligible for this study. Individuals' attitude towards medication was assessed using the Drug attitude inventory classification questionnaire, and the general belief of medication was assessed using the Beliefs about Medicine Questionnaire.

Observation

One-fifth (n = 101; 20%) and two-third (n = 338; 65%) participants were categorized as pharmacophobic and pharmacophilic, respectively. The mean BMQ score was not statistically significant between the pharmacophobic and pharmacophilic groups (mean difference = 0.314; p-value = 0.176). Males were more likely pharmacophilic than females [AOR: 0.34 (95% CI 0.17-0.69)]. Both groups showed nearly a similar pattern of self-medication behaviours.

Conclusion

There exist a good number of pharmacophobic and pharmacophilic among the Saudi population. A detailed understanding of the attitudes towards treatment among these groups is essential for enhancing treatment outcomes.

Tasks as needs: reframing the paradigm of clinical natural language processing research for real-world decision support

Electronic medical records are increasingly used to store patient information in hospitals and other clinical settings. There has been a corresponding proliferation of clinical natural language processing (cNLP) systems aimed at using text data in these records to improve clinical decision-making, in comparison to manual clinician search and clinical judgment alone. However, these systems have delivered marginal practical utility and are rarely deployed into healthcare settings, leading to proposals for technical and structural improvements. In this paper, we argue that this reflects a violation of Friedman's "Fundamental Theorem of Biomedical Informatics," and that a deeper epistemological change must occur in the cNLP field, as a parallel step alongside any technical or structural improvements. We propose that researchers shift away from designing cNLP systems independent of clinical needs, in which cNLP tasks are ends in themselves-"tasks as decisions"-and toward systems that are directly guided by the needs of clinicians in realistic decision-making contexts-"tasks as needs." A case study example illustrates the potential benefits of developing cNLP systems that are designed to more directly support clinical needs.

Leveraging Large-Scale Electronic Health Records and Interpretable Machine Learning for Clinical Decision Making at the Emergency Department: Protocol for System Development and Validation

Background

There is a growing demand globally for emergency department (ED) services. An increase in ED visits has resulted in overcrowding and longer waiting times. The triage process plays a crucial role in assessing and stratifying patients’ risks and ensuring that the critically ill promptly receive appropriate priority and emergency treatment. A substantial amount of research has been conducted on the use of machine learning tools to construct triage and risk prediction models; however, the black box nature of these models has limited their clinical application and interpretation.

Objective

In this study, we plan to develop an innovative, dynamic, and interpretable System for Emergency Risk Triage (SERT) for risk stratification in the ED by leveraging large-scale electronic health records (EHRs) and machine learning.

Methods

To achieve this objective, we will conduct a retrospective, single-center study based on a large, longitudinal data set obtained from the EHRs of the largest tertiary hospital in Singapore. Study outcomes include adverse events experienced by patients, such as the need for an intensive care unit and inpatient death. With preidentified candidate variables drawn from expert opinions and relevant literature, we will apply an interpretable machine learning–based AutoScore to develop 3 SERT scores. These 3 scores can be used at different times in the ED, that is, on arrival, during ED stay, and at admission. Furthermore, we will compare our novel SERT scores with established clinical scores and previously described black box machine learning models as baselines. Receiver operating characteristic analysis will be conducted on the testing cohorts for performance evaluation.

Results

The study is currently being conducted. The extracted data indicate approximately 1.8 million ED visits by over 810,000 unique patients. Modelling results are expected to be published in 2022.

Conclusions

The SERT scoring system proposed in this study will be unique and innovative because of its dynamic nature and modelling transparency. If successfully validated, our proposed solution will establish a standard for data processing and modelling by taking advantage of large-scale EHRs and interpretable machine learning tools.

International Registered Report Identifier (IRRID)

DERR1-10.2196/34201

Predictors of Rehospitalization and Mortality in Diabetes-Related Hospital Admissions

The risk factors of rehospitalization and death post-discharge in diabetes-related hospital admissions are not fully understood. To determine them, a population-based retrospective epidemiological survey was performed on diabetes-related admissions from the Polish national database. Logistic regression models were used, in which the dependent variables were rehospitalization due to diabetes complications and death within 90 days after the index hospitalization. In 2017, there were 74,248 hospitalizations related to diabetes. A total of 11.3% ended with readmission. Risk factors for rehospitalization were as follows: age < 35 years; male sex; prior hospitalization due to acute diabetic complications; weight loss; peripheral artery disease; iron deficiency anemia; kidney failure; alcohol abuse; heart failure; urgent, emergency, or weekend admission; length of hospitalization; and hospitalization in a teaching hospital with an endocrinology/diabetology unit. Furthermore, 7.3% of hospitalizations resulted in death within 90 days following discharge. Risk factors for death were as follows: age; neoplastic disease with/without metastases; weight loss; coagulopathy; alcohol abuse; acute diabetes complications; heart failure; kidney failure; iron deficiency anemia; peripheral artery disease; fluid, electrolytes, and acid–base balance disturbances; urgent or emergency and weekend admission; and length of hospitalization. We concluded that of all investigated factors, only hospitalization within an experienced specialist center may reduce the frequency of the assessed outcomes.

The Tasmanian Conception to Community (C2C) Study Database 2008-09 to 2013-14: Using linked health administrative data to address each piece in the puzzle

BACKGROUND

Tasmania, Australia has a small widely dispersed regional and rural population. The Conception to Community (C2C) Study Database was established as a research platform to inform service planning and policy development and improve health outcomes for Tasmanian mothers and children. The aims of this study were to establish by maternal socio-demographic characteristics: 1) the distribution of births in Tasmania; 2) hospital utilisation for children from birth to 5-years; and 3) the association between child and maternal emergency department (ED) presentation rates.

METHODS

Perinatal and public hospital ED and admitted patient data were linked for every child born in Tasmania between 2008-09 to 2013-14, and their mothers. Individualised rates of ED presentations and hospital admissions were calculated from birth to 5-years. Frequent presenters to ED were defined as having at least four presentations per annum. Ratios of ED presentation and hospital admission rates by sociodemographic characteristics (region (north, north-west, south), rurality, maternal age, and area socioeconomic disadvantage) were estimated using mixed-effects negative binomial models, with random intercepts for each child and family.

RESULTS

The C2C Database is comprised of records for 37,041 children and 27,532 mothers. One-in-ten Tasmanian babies lived in a remote area. The mean yearly rate of ED presentations per child varied by sex, age, region and rurality. Frequent presenters were more likely to reside in the north-west or north, in urban areas, have mothers under 20- years, be male, and live in more disadvantaged areas, with 2.3% of children frequent presenters in their first year of life. The odds of a child being a frequent presenter during their first-year was 6.1- times higher if the mother was a frequent presenter during this period.

CONCLUSION

Associations between maternal and child health service use and combined effects of regionality and rurality highlight opportunities for targeted intervention and service innovations.

The prevalence, characteristics, and risk factors of frequently readmitted patients to an internal medicine service

AIMS

To determine the prevalence, characteristics and risk factors associated with frequent readmissions to an internal medicine service at a tertiary public hospital.

METHOD

A retrospective observational study was conducted at an internal medicine service in a tertiary teaching hospital between 1st January 2010 and the 30th June 2016. Frequent readmission was defined as four or more readmissions within 12 months of discharge from the index admission. Demographic and clinical characteristics, and potential risk factors were evaluated.

RESULTS

50 515 patients were included, 1657 (3.3%) had frequent readmissions and were associated with nearly 2.5 times higher in 12-month mortality rates. They were older, had higher rates of Indigenous Australians (3.2%), more disadvantaged status (Index of Relative Socio-Economic Disadvantage decile of 5.3), and more comorbidities (mean Charlson comorbidity index 1.4) in comparison, to infrequent readmission group. The mean length of hospital stay during the index admission was 6 days for frequent readmission group (21.4% staying more than 7 days) with higher incidence of discharge against medical advice (2.0% higher). Intensive care unit admission rate was 6.6% for frequent readmission group compared to 3.9% for infrequent readmission group. Multivariate analysis showed mental disease and disorders, neoplastic, and alcohol/drug use and alcohol/drug induced organic mental disorders are associated with frequent readmission.

CONCLUSION

The risk factors associated with frequent readmission were older age, indigenous status, being socially disadvantaged, having higher comorbidities, and discharging against medical advice. Conditions that lead to frequent readmissions were mental disorders, alcohol/drug use and alcohol/drug induced organic mental disorders, and neoplastic disorders.

Quality of life after hospitalization predicts one-year readmission risk in a large Swiss cohort of medical in-patients

PURPOSE

Estimating the probability of readmission following hospitalization using prediction scores can be complex. Quality of life (QoL) may provide an easy and effective alternative.

METHODS

Secondary analysis of the prospective "TRIAGE" cohort. All medical in-patients admitted to a Swiss tertiary care institution (2016-2019) ≥18 years with a length of stay of ≥2 days (23,309 patients) were included. EQ-5D VAS, EQ-5D index, and Barthel index were assessed at a single telephone interview 30-day after admission. Patients lost to follow-up were excluded. Readmission was defined as a non-elective hospital stay at our institution >24 h within 1 year after discharge and assessed using area under the curve (AUC) analysis with adjustment for confounders.

RESULTS

12,842 patients (43% females, median age 68, IQR 55-78) were included. Unadjusted discrimination was modest at 0.59 (95% CI 0.56-0.62) for EQ-5D VAS. Partially adjusted discrimination (for gender) was identical. Additional adjustment for insurance, Charlson comorbidity index, length of stay, and native language increased the AUC to 0.66 (95% CI 0.63-0.69). Results were robust irrespective of time to event (12, 6 or 3 months). A cut-off in the unadjusted model of EQ-5D VAS of 55 could separate cases with a specificity of 80% and a sensitivity of 30%.

CONCLUSION

QoL at day 30 after admission can predict one-year readmission risk with similar precision as more intricate tools. It might help for identification of high-risk patients and the design of tailored prevention strategies.

Epidemiology of Frequent Visits to the Emergency Department at a Tertiary Care Hospital in Saudi Arabia: Rate, Visitors' Characteristics, and Associated Factors

Background

Emergency department overcrowding is becoming a challenge for the healthcare management system globally and locally. This study aimed to estimate the frequency of ED visits, describe the patients’ profile along with visit-related characteristics, and associated factors in a tertiary care hospital.

Methods

A retrospective cohort study was conducted at a tertiary care hospital. The study included patients age 14 years and above visiting the main emergency department in year 2013. Data were extracted from electronic medical records by a qualified data extraction team. Statistical analyses were performed, including the odds ratio and 95% confidence interval for the factors associated with highly frequent (≥14 visits) ED visits using logistic regression models.

Results

There were 150,727 visits to the emergency department within a year. The number of frequent visitors was 7696 (9.38%), with 42,226 visits (28.01% of total ED visits). Highly frequent visitors totaled 249 (0.30%), with 5173 visits (3.43% of total ED visits). The frequent visitors’ average age was 42.55 (SD 20.14), and 48.99 (SD 21.33) for the highly frequent visitors’ group. More than half of the emergency visitors were females. The most common complaints among the highly frequent visitors were Gastrointestinal (21.34%), followed by Respiratory (13.47%), Orthopedic (12.57%), and Cardiovascular (12.43%). Multivariate analysis indicated that age, history of diabetes, history of cardiac diseases, insurance status, and nationality were significant predicators of highly frequent visits to the hospital emergency.

Conclusion

Frequent and highly frequent visitors to emergency departments represent a significant proportion of adult patients presenting to ED. Their visits constitute almost one-third of total ED visits. Several factors associated with highly frequent ED visits have been identified. This study provides local empirical evidence to develop improvement policy and actions related to chronic issue of frequent and highly frequent visitation to hospital ED.

Factors Influencing Patients Using Long-Term Care Service of Discharge Planning by Andersen Behavioral Model: A Hospital-Based Cross-Sectional Study in Eastern Taiwan

Taiwan has been an aged society since March 2018, and the elderly population suffer from multiple comorbidities and long duration of disability. Therefore, the service of discharge planning of long-term care 2.0 is an important stage before patients go back to the community. Strengthening the sensitivity when identifying predisabled patients is a principal development of discharge planning. In the current study, we analyzed the characteristics and predictive factors of patients who used the service of long-term care 2.0 from the perspective of discharge planning. In this retrospective study, we included patients who received the discharge planning service in a hospital located in southern Hualien during November 2017 to October 2018. The data were collected and classified as predisposing factors, enabling factors, and need factors according to the analysis architecture of the Andersen Behavioral Model. There were 280 valid patients included in this current study; age, medical accessibility, possession of a disability card, and cerebrovascular diseases, cardiovascular diseases, and diabetes mellitus were the vital factors which influenced the coherence and cohesion between discharge planning and the service of long-term care 2.0. Among them, the most influencing factor was age. We hope that the current study will make policymakers in hospitals pay attention to the usage of the discharge planning service to link long-term care 2.0 and effectively promote the usage of long-term care 2.0.

Frequent avoidable admissions amongst Aboriginal and non-Aboriginal people with chronic conditions in New South Wales, Australia: a historical cohort study

BACKGROUND

Aboriginal and Torres Strait Islander people have high rates of avoidable hospital admissions for chronic conditions, however little is known about the frequency of avoidable admissions for this population. This study examined trends in avoidable admissions among Aboriginal and non-Aboriginal people with chronic conditions in New South Wales (NSW), Australia.

METHODS

A historical cohort analysis using de-identified linked administrative data of Aboriginal patients and an equal number of randomly sampled non-Aboriginal patients between 2005/06 to 2013/14. Eligible patients were admitted to a NSW public hospital and who had one or more of the following ambulatory care sensitive chronic conditions as a principal diagnosis: diabetic complications, asthma, angina, hypertension, congestive heart failure and/or chronic obstructive pulmonary disease. The primary outcomes were the number of avoidable admissions for an individual in each financial year, and whether an individual had three or more admissions compared with one to two avoidable admissions in each financial year. Poisson and logistic regression models and a test for differences in yearly trends were used to assess the frequency of avoidable admissions over time, adjusting for sociodemographic variables and restricted to those aged ≤75 years.

RESULTS

Once eligibility criteria had been applied, there were 27,467 avoidable admissions corresponding to 19,025 patients between 2005/06 to 2013/14 (71.2% Aboriginal; 28.8% non-Aboriginal). Aboriginal patients were 15% more likely than non-Aboriginal patients to have a higher number of avoidable admissions per financial year (IRR = 1.15; 95% CI: 1.11, 1.20). Aboriginal patients were almost twice as likely as non-Aboriginal patients to experience three or more avoidable admissions per financial year (OR = 1.90; 95% CI = 1.60, 2.26). There were no significant differences between Aboriginal and non-Aboriginal people in yearly trends for either the number of avoidable admissions, or whether or not an individual experienced three or more avoidable admissions per financial year (p = 0.859; 0.860 respectively).

CONCLUSION

Aboriginal people were significantly more likely to experience frequent avoidable admissions over a nine-year period compared to non-Aboriginal people. These high rates reflect the need for further research into which interventions are able to successfully reduce avoidable admissions among Aboriginal people, and the importance of culturally appropriate community health care.

Avoiding unnecessary hospitalisation for patients with chronic conditions: a systematic review of implementation determinants for hospital avoidance programmes

Background

Studies of clinical effectiveness have demonstrated the many benefits of programmes that avoid unnecessary hospitalisations. Therefore, it is imperative to examine the factors influencing implementation of these programmes to ensure these benefits are realised across different healthcare contexts and settings. Numerous factors may act as determinants of implementation success or failure (facilitators and barriers), by either obstructing or enabling changes in healthcare delivery. Understanding the relationships between these determinants is needed to design and tailor strategies that integrate effective programmes into routine practice. Our aims were to describe the implementation determinants for hospital avoidance programmes for people with chronic conditions and the relationships between these determinants.

Methods

An electronic search of four databases was conducted from inception to October 2019, supplemented by snowballing for additional articles. Data were extracted using a structured data extraction tool and risk of bias assessed using the Hawker Tool. Thematic synthesis was undertaken to identify determinants of implementation success or failure for hospital avoidance programmes for people with chronic conditions, which were categorised according to the Consolidated Framework for Implementation Research (CFIR). The relationships between these determinants were also mapped.

Results

The initial search returned 3537 articles after duplicates were removed. After title and abstract screening, 123 articles underwent full-text review. Thirteen articles (14 studies) met the inclusion criteria. Thematic synthesis yielded 23 determinants of implementation across the five CFIR domains. ‘Availability of resources’, ‘compatibility and fit’, and ‘engagement of interprofessional team’ emerged as the most prominent determinants across the included studies. The most interconnected implementation determinants were the ‘compatibility and fit’ of interventions and ‘leadership influence’ factors.

Conclusions

Evidence is emerging for how chronic condition hospital avoidance programmes can be successfully implemented and scaled across different settings and contexts. This review provides a summary of key implementation determinants and their relationships. We propose a hypothesised causal loop diagram to represent the relationship between determinants within a complex adaptive system.

Trial registration

PROSPERO 162812 

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.

Supportive Care for Superutilizers of a Managed Care Organization

Background: Ohana Health Plan, Inc., (OHP) is one of the first managed care organizations offering supportive care services targeted to superutilizers. Bristol Hospice Hawaii, LLC, partnered with OHP to provide interdisciplinary supportive care services to home-bound OHP members. Objectives: The purpose of this study was to measure symptom relief, satisfaction, resource utilization, and cost savings associated with supportive care. Design: Prospective study. Setting: Over 12 months, 27 superutilizer members residing in the community were referred by OHP, 21 members were enrolled into supportive care. Measurements: Data were collected upon admission and repeatedly thereafter using the Edmonton Symptom Assessment Scale (ESAS) and the Missoula-Vitas Quality of Life Index (MVQOLI). The Family Satisfaction with Advanced Cancer Care (FAMCARE) Scale was administered at discharge. Emergency department (ED) visits and hospital utilization were tracked. Results: Median age was 63 years; more than half had cardiac diagnoses. Majority of members were Hawaiian and other Pacific Islander. Median length of stay in supportive care was 90 days. Five (23%) members enrolled in hospice following supportive care. Symptom improvement occurred in pain (p < 0.0001), anxiety (p = 0.0052), and shortness of breath (p = 0.0447). This model has shown a 79.5% reduction of ED visits per thousand members and a 75% reduction of hospitalizations per thousand. Overall net savings was 36%. Discussions and documentation of end-of-life wishes increased from 23% to 85%. Conclusion: Supportive care is highly effective in reducing costs associated with superutilizers. Our experience demonstrates the effectiveness of supportive care approaches in this population through improved care and lower health care costs overall.

Persistent frequent emergency department users with chronic conditions: A population-based cohort study

BACKGROUND

Frequent emergency department users are patients cumulating at least four visits per year. Few studies have focused on persistent frequent users, who maintain their frequent user status for multiple consecutive years. This study targets an adult population with chronic conditions, and its aims are: 1) to estimate the prevalence of persistent frequent ED use; 2) to identify factors associated with persistent frequent ED use (frequent use for three consecutive years) and compare their importance with those associated with occasional frequent ED use (frequent use during the year following the index date); and 3) to compare characteristics of "persistent frequent users" to "occasional frequent users" and to "users other than persistent frequent users".

METHODS

This is a retrospective cohort study using Quebec administrative databases. All adult patients who visited the emergency department in 2012, diagnosed with chronic conditions, and living in non-remote areas were included. Patients who died in the three years following their index date were excluded. The main outcome was persistent frequent use (≥4 visits per year during three consecutive years). Potential predictors included sociodemographic characteristics, physical and mental comorbidities, and prior healthcare utilization. Odds ratios were computed using multivariable logistic regression.

RESULTS

Out of 297,182 patients who visited ED at least once in 2012, 3,357 (1.10%) were persistent frequent users. Their main characteristics included poor socioeconomic status, mental and physical comorbidity, and substance abuse. Those characteristics were also present for occasional frequent users, although with higher percentages for the persistent user group. The number of previous visits to the emergency department was the most important factor in the regression model. The occasional frequent users' attrition rate was higher between the first and second year of follow-up than between the second and third year.

CONCLUSIONS

Persistent frequent users are a subpopulation of frequent users with whom they share characteristics, such as physical and mental comorbidities, though the former are poorer and younger. More research is needed in order to better understand what factors can contribute to persistent frequent use.

Statistical tools used for analyses of frequent users of emergency department: a scoping review

OBJECTIVE

Frequent users represent a small proportion of emergency department users, but they account for a disproportionately large number of visits. Their use of emergency departments is often considered suboptimal. It would be more efficient to identify and treat those patients earlier in their health problem trajectory. It is therefore essential to describe their characteristics and to predict their emergency department use. In order to do so, adequate statistical tools are needed. The objective of this study was to determine the statistical tools used in identifying variables associated with frequent use or predicting the risk of becoming a frequent user.

METHODS

We performed a scoping review following an established 5-stage methodological framework. We searched PubMed, Scopus and CINAHL databases in February 2019 using search strategies defined with the help of an information specialist. Out of 4534 potential abstracts, we selected 114 articles based on defined criteria and presented in a content analysis.

RESULTS

We identified four classes of statistical tools. Regression models were found to be the most common practice, followed by hypothesis testing. The logistic regression was found to be the most used statistical tool, followed by χ2 test and t-test of associations between variables. Other tools were marginally used.

CONCLUSIONS

This scoping review lists common statistical tools used for analysing frequent users in emergency departments. It highlights the fact that some are well established while others are much less so. More research is needed to apply appropriate techniques to health data or to diversify statistical point of views.

Retrospective evaluation of healthcare utilisation and mortality of two post-discharge care programmes in Singapore

OBJECTIVE

To evaluate the impact on healthcare utilisation frequencies and charges, and mortality of a programme for frequent hospital utilisers and a programme for patients requiring high acuity post-discharge care as part of an integrated healthcare model.

DESIGN

A retrospective quasi-experimental study without randomisation where patients who received post-discharge care interventions were matched 1:1 with unenrolled patients as controls.

SETTING

The National University Health System (NUHS) Regional Health System (RHS), which was one of six RHS in Singapore, implemented the NUHS RHS Integrated Interventions and Care Extension (NICE) programme for frequent hospital utilisers and the NUHS Transitional Care Programme (NUHS TCP) for high acuity post-discharge care. The programmes were supported by the Ministry of Health in Singapore, which is a city-state nation located in Southeast Asia with a 5.6 million population.

PARTICIPANTS

Linked healthcare administrative data, for the time period of January 2013 to December 2016, were extracted for patients enrolled in NICE (n=554) or NUHS TCP (n=270) from June 2014 to December 2015, and control patients.

INTERVENTIONS

For both programmes, teams conducted follow-up home visits and phone calls to monitor and manage patients' post-discharge.

PRIMARY OUTCOME MEASURES

One-year pre- and post-enrolment healthcare utilisation frequencies and charges of all-cause inpatient admissions, emergency admissions, emergency department attendances, specialist outpatient clinic (SOC) attendances, total inpatient length of stay and mortality rates were compared.

RESULTS

Patients in NICE had lower mortality rate, but higher all-cause inpatient admission, emergency admission and emergency department attendance charges. Patients in NUHS TCP did not have lower mortality rate, but had higher emergency admission and SOC attendance charges.

CONCLUSIONS

Both NICE and NUHS TCP had no improvements in 1 year healthcare utilisation across various setting and metrics. Singular interventions might not be as impactful in effecting utilisation without an overhauling transformation and restructuring of the hospital and healthcare system.

How much emergency department use by vulnerable populations is potentially preventable?: A period prevalence study of linked public hospital data in South Australia

OBJECTIVES

To quantify emergency department (ED) presentations by individuals within vulnerable populations compared with other adults and the extent to which these are potentially preventable.

DESIGN

Period prevalence study from 2005-2006 to 2010-2011.

SETTING

Person-linked, ED administrative records for public hospitals in South Australia.

PARTICIPANTS

Adults aged 20 or more in South Australia's metropolitan area presenting to ED and categorised as Refugee and Asylum Seeker Countries of birth (RASC); Aboriginal; those aged 75 years or more; or All others.

MAIN OUTCOME MEASURES

Unadjusted rates of ambulatory care sensitive condition (ACSC), general practitioner (GP)-type presentations and associated direct ED costs among mutually exclusive groups of individuals.

RESULTS

Disparity between RASC and All others was greatest for GP-type presentations (423.7 and 240.1 persons per 1000 population, respectively) with excess costs of $A106 573 (95% CI $A98 775 to $A114 371) per 1000 population. Aboriginal had highest acute ACSC presenter rates (125.8 against 51.6 per 1000 population) with twice the risk of multiple presentations and $A108 701 (95% CI $A374 to $A123 029) per 1000 excess costs. Those aged 75 or more had highest chronic ACSC presenter rates (119.7vs21.1 per 1000), threefold risk of further presentations (incidence rate ratio 3.20, 95% CI 3.14 to 3.26) and excess cost of $A385 (95% CI $A178 160 to $A184 609) per 1000 population.

CONCLUSIONS

Vulnerable groups had excess ED presentations for a range of issues potentially better addressed through primary and community healthcare. The observed differences suggest inequities in the uptake of effective primary and community care and represent a source of excess cost to the public hospital system.

A multifaceted primary care practice-based intervention to reduce ED visits and hospitalization for complex medical patients: A mixed methods study

BACKGROUND

The management of complex, multi-morbid patients is challenging for solo primary care providers (PCPs) with limited access to resources. The primary objective of the intervention was to reduce the overall rate of Emergency Department (ED) visits among patients in participating practices.

METHODS AND FINDINGS

An interrupted time series design and qualitative interviews were used to evaluate a multifaceted intervention, SCOPE (Seamless Care Optimizing the Patient Experience), offered to solo PCPs whose patients were frequent users of the ED. The intervention featured a navigation hub (nurse, homecare coordinator) to link PCPs with hospital and community resources, a general internist on-call to provide phone advice or urgent assessments, and access to patient results on-line. Continuous quality improvement (QI) strategies were employed to optimize each component of the intervention. The primary outcome was the relative pre-post intervention change in ED visit rate for patients of participating practices compared with that for a propensity-matched control group of physicians over the contemporaneous period. Themes were identified from semi-structured interviews on PCP's experiences and influential factors in their engagement. Twenty-nine physicians agreed to participate and were provided access to the intervention over an 18-month time period. There were a total of 1,525 intervention contacts over the 18-months (average: 50.6±60.8 per PCP). Both intervention and control groups experienced a trend towards lower rates of ED use by their patients over the study time period. The pre-post difference in trend for the intervention group compared to the controls was not significant at 1.4% per year (RR = 1.014; p = 0.59). Several themes were identified from qualitative interviews including: PCPs felt better supported in the care of their patients; they experienced a greater sense of community, and; they were better able to provide shared primary-specialty care.

CONCLUSIONS

This multifaceted intervention to support solo PCPs in the management of their complex patients did not result in a reduced rate of ED visits compared to controls, likely related to variable uptake among PCPs. It did however result in more comprehensive and coordinated care for their patients. Future directions will focus on increasing uptake by improving ease of use, increasing the range of services offered and expanding to a larger number of PCPs.

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.

Analysing repeated hospital readmissions using data mining techniques

Few studies have examined how to identify future readmission of patients with a large number of repeat emergency department (ED) visits. We explore 30-day readmission risk prediction using Microsoft's AZURE machine learning software and compare five classification methods: Logistic Regression, Boosted Decision Trees (BDTs), Support Vector Machine (SVM), Bayes Point Machine (BPM), and Two-Class Neural Network (TCNN). We predict the last readmission visit of frequent ED patients extracted from the electronic health records of their 8455 penultimate visits. The methods show differential improvement, with the BDT indicating marginally better AUC (area under the ROC curve) than logistic regression and BPM, followed by the TCNN and SVM. A comparison of BDT and Logistic Regression results for correct and incorrect classification highlights the similarities and differences in the significant predictors identified by each method. Future research may incorporate time-varying covariates to identify other longitudinal factors that can lead to readmission risk reduction.

Identifying Drivers of Multiple Readmissions After Pulmonary Lobectomy

BACKGROUND

Readmissions after pulmonary lobectomy for lung cancer are important markers of healthcare quality for surgeons and hospitals. The implications on resources and quality are magnified when examining patients who require multiple readmissions within the perioperative period.

METHODS

The Nationwide Readmission Database between 2013 and 2014 was investigated for patients with a primary diagnosis of lung cancer who underwent pulmonary lobectomy. Using adjusted hierarchical regression models, demographic and clinical factors during the index hospitalization were investigated for associations with single and multiple readmissions during the 90-day postoperative period. First and second readmissions during this period were compared for the primary diagnosis at the time of readmission using Clinical Classification Software codes.

RESULTS

Of the 41,576 lobectomies during the study period 7,030 patients (16.9%) were readmitted. Among this group 1,554 patients (3.7%) had at least two readmissions. After adjustment for other factors, postoperative arrhythmia (odds ratio [OR], 1.51; 95% confidence interval [CI], 1.25-1.83; p < 0.0001), postoperative infection (OR, 1.55; 95% CI, 1.11-2.17; p = 0.01), and postoperative sepsis (OR, 1.70; 95% CI, 1.08-2.67; p = 0.02) during the index hospitalization were associated with an increased risk of at least two readmissions. The most frequent Clinical Classification Software diagnosis for first readmissions was "postoperative complications" (892, 12.7%) and for second readmissions was heart disease (173, 11.2%).

CONCLUSIONS

Approximately one-fifth of patients readmitted after pulmonary lobectomy would go on to be readmitted two or more times within 90 days. Although first readmissions were most likely to present with postoperative infection or complication, second readmissions were most likely to present with heart disease.

Care Complexity Individual Factors Associated With Hospital Readmission: A Retrospective Cohort Study

PURPOSE

To determine the frequency of care complexity individual factors documented in the nursing assessment and to identify the risk factors associated with hospital readmission within 30 days of hospital discharge.

DESIGN

Observational analysis of a retrospective cohort at a 700-bed university hospital in Barcelona, Spain. A total of 16,925 adult patient admissions to a ward or intermediate care units were evaluated from January to December 2016. Most patients were admitted due to cardiocirculatory and respiratory disorders (29.3%), musculoskeletal and nervous system disorders (21.8%), digestive and hepatobiliary conditions (17.9%), and kidney or urinary disorders (11.2%).

METHODS

Readmission was defined as rehospitalization for any reason within 30 days of discharge. Patients who required hospital readmission were compared with those who did not. The individual factors of care complexity included five domains (developmental, mental-cognitive, psycho-emotional, sociocultural, and comorbidity or complications) and were reviewed using the electronic nursing assessment records. Multivariate logistic analysis was performed to determine factors associated with readmission.

FINDINGS

A total of 1,052 patients (6.4%) were readmitted within 30 days of hospital discharge. Care complexity individual factors from the comorbidity or complications domain were found to be the most frequently e-charted (88.3%). Care complexity individual factors from developmental (33.2%), psycho-emotional (13.2%), mental-cognitive (7.2%), and sociocultural (0.7%) domains were less frequently documented. Independent factors associated with hospital readmission were old age (≥75 years), duration of first hospitalization, admission to a nonsurgical ward, major chronic disease, hemodynamic instability, immunosuppression, and relative weight of diagnosis-related group.

CONCLUSIONS

A substantial number of patients required readmission within 30 days after discharge. The most frequent care complexity individual factors recorded in the nursing assessment at index admission were related to comorbidity or complications, developmental, and psycho-emotional domains. Strategies related to transition of care that include clinical characteristics and comorbidity or complications factors should be a priority at hospital discharge and after leaving hospital, but other factors related to developmental and psycho-emotional domains could have an important effect on the use of healthcare resources.

CLINICAL RELEVANCE

Nurses should identify patients with comorbidity or complications, developmental, and psycho-emotional complexity factors during the index admission in order to be able to implement an effective discharge process of care.

Risk factors and direct medical cost of early versus late unplanned readmissions among diabetes patients at a tertiary hospital in Singapore

OBJECTIVE

To examine the risk factors and direct medical costs associated with early (≤30 days) versus late (31-180 days) unplanned readmissions among patients with type 2 diabetes in Singapore.

METHODS

Risk factors and associated costs among diabetes patients were investigated using electronic medical records from a local tertiary care hospital from 2010 to 2012. Multivariable logistic regression was used to identify risk factors associated with early and late unplanned readmissions while a generalized linear model was used to estimate the direct medical cost. Sensitivity analysis was also performed.

RESULTS

A total of 1729 diabetes patients had unplanned readmissions within 180 days of an index discharge. Length of index stay (a marker of acute illness burden) was one of the risk factors associated with early unplanned readmission while patient behavior-related factors, like diabetes-related medication adherence, were associated with late unplanned readmission. Adjusted mean cost of index admission was higher among patients with unplanned readmission. Sensitivity analysis yielded similar results.

CONCLUSIONS

Existing routinely captured data can be used to develop prediction models that flag high risk patients during their index admission, potentially helping to support clinical decisions and prevent such readmissions.

Causes of Hospital Admissions in Domus: A Randomized Controlled Trial of Specialized Palliative Cancer Care at Home

CONTEXT

Avoidable hospital admissions are important negative indicators of quality of end-of-life care. Specialized palliative care (SPC) may support patients remaining at home.

OBJECTIVES

Therefore, the purpose of this study was to investigate if SPC at home could prevent hospital admissions in patients with incurable cancer.

METHODS

These are secondary results of Domus: a randomized controlled trial of accelerated transition to SPC with psychological intervention at home (Clinicaltrials.gov: NCT01885637). Participants were patients with incurable cancer and limited antineoplastic treatment options and their caregivers. They were included from the Department of Oncology, Rigshospitalet, Denmark, between 2013 and 2016. The control group received usual care. Outcomes were hospital admissions, causes thereof, and patient and caregiver perceptions of place of care (home, hospital, etc.) at baseline, four weeks, eight weeks, and six months.

RESULTS

During the study, 340 patients were randomized and 322 were included in modified intention-to-treat analyses. Overall, there were no significant differences in hospital admissions between the groups. The intervention group had more admissions triggered by worsened general health (22% vs. 16%, P = 0.0436) or unmanageable home situation (8% vs. 4%, P = 0.0119). After diagnostics, admissions were more often caused by clinical symptoms of cancer without progression in the intervention group (11% vs. 7%, P = 0.0493). The two groups did not differ significantly in overall potentially avoidable admissions. Both groups felt mostly safe about their place of care.

CONCLUSION

The intervention did not prevent hospital admissions. Likely, any intervention effects were outweighed by increased identification of problems in the intervention group leading to hospital admissions. Overall, patients and caregivers felt safe in their current place of care.

The effect of a gerontology nurse specialist for high needs older people in the community on healthcare utilisation: a controlled before-after study

BACKGROUND

Nurse-led models of comprehensive geriatric assessment and care coordination can improve health management as well as reduce hospitalisations for high risk community dwelling older people. This study investigated the effect on healthcare utilisation of systematic case finding to identify high risk older people in the community with a subsequent comprehensive assessment and care coordination intervention by a Gerontology Nurse Specialist based in primary care.

METHODS

This was a controlled before-after study design located within primary healthcare practices in Auckland, New Zealand. An intervention model was initiated within two primary healthcare practices and involved a screening tool to identify high risk older people with succeeding gerontology nurse specialist assessment and care coordination. The comparison group included older people who received usual care at three comparable primary healthcare practices. The primary outcome measure was acute hospital admissions. Secondary outcomes included hospital re-admissions, length of stay, emergency department presentations, residential care admissions, and community contacts.

RESULTS

A total of 579 older people were posted the screening tool in the intervention group, with 517 completed screens (89% response rate) formulating the intervention group. A total of 101 older people were identified as high risk from these screens (20%). The comparison group comprised 883 older people. Comparing the intervention and comparison group, no statistical differences were found for hospital admissions, emergency department presentations, hospital re-admissions, length of stay, or residential care admission. Community physiotherapy showed a statistically significant increase for the intervention compared to the comparison group (p = 0.03). Non-significant findings revealed decreased risk of entering residential care and fewer frequent hospital re-admissions for the intervention group when compared with the comparison group.

CONCLUSIONS

This specialist nurse-led intervention involving comprehensive assessment and care coordination care did not appear superior to usual care, however, there is benefit to exploring a more robust randomised controlled trial design.

TRIAL REGISTRATION

Retrospectively registered on 18/09/2017 with the Australian New Zealand Clinical Trials Registry (ANZCTR). Registration number ACTRN12617001332314.

Analysing repeated hospital readmissions using data mining techniques

Few studies have examined how to identify future readmission of patients with a large number of repeat emergency department (ED) visits. We explore 30-day readmission risk prediction using Microsoft's AZURE machine learning software and compare five classification methods: Logistic Regression, Boosted Decision Trees (BDTs), Support Vector Machine (SVM), Bayes Point Machine (BPM), and Two-Class Neural Network (TCNN). We predict the last readmission visit of frequent ED patients extracted from the electronic health records of their 8455 penultimate visits. The methods show differential improvement, with the BDT indicating marginally better AUC (area under the ROC curve) than logistic regression and BPM, followed by the TCNN and SVM. A comparison of BDT and Logistic Regression results for correct and incorrect classification highlights the similarities and differences in the significant predictors identified by each method. Future research may incorporate time-varying covariates to identify other longitudinal factors that can lead to readmission risk reduction.

Joint modelling of potentially avoidable hospitalisation for five diseases accounting for spatiotemporal effects: A case study in New South Wales, Australia

Background

Three variant formulations of a spatiotemporal shared component model are proposed that allow examination of changes in shared underlying factors over time.

Methods

Models are evaluated within the context of a case study examining hospitalisation rates for five chronic diseases for residents of a regional area in New South Wales: type II diabetes mellitus (DMII), chronic obstructive pulmonary disease (COPD), coronary arterial disease (CAD), hypertension (HT) and congestive heart failure (CHF) between 2001–2006. These represent ambulatory care sensitive (ACS) conditions, often used as a proxy for avoidable hospitalisations. Using a selected model, the effects of socio-economic status (SES) as a shared component are estimated and temporal patterns in the influence of the residual shared spatial component are examined.

Results

Choice of model depends upon the application. In the featured application, a model allowing for changing influence of the shared spatial component over time was found to have the best fit and was selected for further analyses. Hospitalisation rates were found to be increasing for COPD and DMII, decreasing for CHF and stable for CAD and HT. SES was substantively associated with hospitalisation rates, with differing degrees of influence for each disease. In general, most of the spatial variation in hospitalisation rates was explained by disease-specific spatial components, followed by the residual shared spatial component.

Conclusion

Appropriate selection of a joint disease model allows for the examination of temporal patterns of disease outcomes and shared underlying spatial factors, and distinction between different shared spatial factors.

Clinical characteristics of chronic bronchitic, emphysematous and ACOS phenotypes in COPD patients with frequent exacerbations

Purpose

Chronic bronchitis (CB), emphysematous (EM) and asthma–chronic obstructive pulmonary disease (COPD) overlap syndrome (ACOS) phenotypes in COPD are well recognized. This study aimed to investigate distinguishing characteristics of these phenotypes in COPD patients with frequent exacerbations (FE).

Patients and methods

A retrospective study was carried out. COPD patients with acute exacerbations were consecutively reviewed from November 2015 to October 2016. Patients were divided into FE and infrequent exacerbations (iFE) subgroups.

Results

A total of 142 eligible COPD subjects were reviewed. In the CB phenotype subgroup, age, body mass index, forced expiratory volume in 1 second (FEV₁) % predicted, COPD assessment test (CAT), modified Medical Research Council breathlessness measurement (mMRC) dyspnea scale, emphysema scores and arterial carbon dioxide pressure (PaCO₂) were significantly different in subjects with FE when compared to those in subjects with iFE of CB. In the EM phenotype subgroup, age, CAT, mMRC scores and history of COPD were different in subjects with FE when compared to those in CB subjects with iFE. Multivariate analysis indicated that FEV₁% predicted (odds ratio [OR] =0.90, P=0.04) and PaCO₂ (OR =1.22, P=0.02) were independent risk factors for FE in COPD with CB phenotype, and CAT (OR =2.601, P=0.001) was the independent risk factor for FE in COPD with EM phenotype. No significant differences in characteristics were observed in ACOS phenotype subgroups with FE or iFE.

Conclusion

In CB or EM phenotypes, COPD patients with FE present several differential clinical characteristics compared to patients with iFE, while the characteristics of ACOS phenotype in patients with FE need more investigation.

Evaluation of the Effectiveness of a Phone Based Care Coordination Pilot on Hospital Utilisation and Costs for Patients With Chest Pain

BACKGROUND

A small percentage of the population represents a disproportionate number of attendances at emergency departments (ED). "Frequent presenters" to ED with chest pain do not always fit into established pathways for acute myocardial events. With accelerated "rule out" protocols, patients are often discharged from the ED after short lengths of stay. This research will evaluate the effectiveness of a phone based care-coordination pilot designed to meet the needs of patients attending ED with cardiac and non-cardiac chest pain.

METHODS

A longitudinal, single-arm interventional study with retrospectively recruited control group. Ninety-five patients were enrolled as the intervention group; 97 patients were retrospectively identified as controls. These patients had re-presented with chest pain within 6 months of a cardiac event, or attended hospital within 12 months two or more times with chest pain and/or complex needs. Intervention group patients were holistically assessed then phone-coached to support self-management of chest pain over 6 months. Following descriptive and univariate analysis, multivariate analysis was conducted to adjust for noted differences between the intervention and control groups.

RESULTS

Thirty-day representation to ED was significantly less for the intervention group (14.1%) compared to controls (27.7%). After adjusting for baseline differences, intervention patients were more than two-fold less likely to re-present compared to controls (OR=0.42, 95%CI: 0.19-0.96). After adjustment for baseline differences, the savings in subsequent inpatient costs was $1588 per person, as a result of intervention, patients were less likely to have inpatient readmissions (16.3%) compared to controls (20.2%), although this was not statistically significant (p=0.588).

CONCLUSION

A phone based care-coordination pilot with targeted interventions has the potential to reduce ED presentations and hospital readmissions among patients representing with chest pain.

The impact of walk-in centres and GP co-operatives on emergency department presentations: A systematic review of the literature

BACKGROUND

Internationally, non-urgent presentations are increasing the pressure on Emergency Department (ED) staff and resources. This systematic review aims to identify the impact of alternative emergency care pathways on ED presentations - specifically GP cooperatives and walk-in clinics.

METHODS

Based on a structured PICO enquiry with either walk-in clinic or GP cooperative as the intervention, a search was made for peer-reviewed publications in English, between 2000 and 2014. Medline plus, OVID, PubMed, and Google Scholar were searched. The Critical Appraisal Skills Program (CASP) guidelines were used to assess study quality and data was extracted using an adapted JBI Qualitative Assessment and Review Instrument (QARI). Subsequent reporting followed the PRISMA guideline.

RESULTS

Eleven high quality quantitative studies met the inclusion criteria. Walk-in clinics do have the potential to reduce non-urgent emergency department presentations, however evidence of this effect is low. GP cooperatives offer an alternative care stream for patients presenting to the ED and do significantly reduce local ED attendances. Community members need to be made aware of these options in order to make informed treatment choices.

CONCLUSION

GP cooperatives in particular do have the potential to reduce ED workload. Further research is required to uncover recent trends and patient outcomes for walk-in clinics and GP cooperatives.

Factors associated with high-utilization in a safety net setting

Background

Patients with frequent hospital readmissions, or high-utilizer patients (HUPs), are a major driver of rising healthcare costs in the United States. This group has a significant burden of medical illness, but less is known about whether or how social determinants of health may drive their increased healthcare use and poor health outcomes. Our study aimed to define the population of HUPs at a large, safety-net hospital system, to understand how these patients differ from patients who are not HUPs, and to analyze how their demographic, medical, and social factors contribute to their healthcare use and mortality rates.

Methods

For this case-control study, data were collected via retrospective chart review. We included 247 patients admitted three or more times in a single calendar year between 2011 and 2013 and 247 controls with one or two admissions in a single calendar year matched for age, sex, and year of high-utilization. We used multivariable logistic regression models to understand which demographic, clinical, and social factors were associated with HUP status, and if HUP status was independently associated with mortality.

Results

The factors that contributed significant odds of being a HUP included having Medicaid (OR 3.34, 95% CI 1.50, 7.44) or Medicare (OR 3.39, 95% CI 1.50, 7.67), having a history of recreational drug use (OR 2.44, 95% 1.36, 4.38), and being homeless (OR 3.73, 95% CI 1.69, 8.23) The mortality rate among HUPs was 22.6% compared to 8.9% among controls (p < 0.0001).

Conclusions

These data show that social factors are related to high-utilization in this population. Future efforts to understand and improve the health of this population need to incorporate non-clinical patient factors.

Electronic supplementary material

The online version of this article (doi:10.1186/s12913-017-2209-0) contains supplementary material, which is available to authorized users.

Predicting AKI in emergency admissions: an external validation study of the acute kidney injury prediction score (APS)

OBJECTIVES

Hospital-acquired acute kidney injury (HA-AKI) is associated with a high risk of mortality. Prediction models or rules may identify those most at risk of HA-AKI. This study externally validated one of the few clinical prediction rules (CPRs) derived in a general medicine cohort using clinical information and data from an acute hospitals electronic system on admission: the acute kidney injury prediction score (APS).

DESIGN, SETTING AND PARTICIPANTS

External validation in a single UK non-specialist acute hospital (2013-2015, 12 554 episodes); four cohorts: adult medical and general surgical populations, with and without a known preadmission baseline serum creatinine (SCr).

METHODS

Performance assessed by discrimination using area under the receiver operating characteristic curves (AUCROC) and calibration.

RESULTS

HA-AKI incidence within 7 days (kidney disease: improving global outcomes (KDIGO) change in SCr) was 8.1% (n=409) of medical patients with known baseline SCr, 6.6% (n=141) in those without a baseline, 4.9% (n=204) in surgical patients with baseline and 4% (n=49) in those without. Across the four cohorts AUCROC were: medical with known baseline 0.65 (95% CIs 0.62 to 0.67) and no baseline 0.71 (0.67 to 0.75), surgical with baseline 0.66 (0.62 to 0.70) and no baseline 0.68 (0.58 to 0.75). For calibration, in medicine and surgical cohorts with baseline SCr, Hosmer-Lemeshow p values were non-significant, suggesting acceptable calibration. In the medical cohort, at a cut-off of five points on the APS to predict HA-AKI, positive predictive value was 16% (13-18%) and negative predictive value 94% (93-94%). Of medical patients with HA-AKI, those with an APS ≥5 had a significantly increased risk of death (28% vs 18%, OR 1.8 (95% CI 1.1 to 2.9), p=0.015).

CONCLUSIONS

On external validation the APS on admission shows moderate discrimination and acceptable calibration to predict HA-AKI and may be useful as a severity marker when HA-AKI occurs. Harnessing linked data from primary care may be one way to achieve more accurate risk prediction.

Integrating the Principles of Evidence Based Medicine and Evidence Based Public Health: Impact on the Quality of Patient Care and Hospital Readmission Rates in Jordan

INTRODUCTION

Hospital readmissions impose not only an extra burden on health care systems but impact patient health outcomes. Identifying modifiable behavioural risk factors that are possible causes of potentially avoidable readmissions can lower readmission rates and healthcare costs.

METHODS

Using the core principles of evidence based medicine and public health, the purpose of this study was to develop a heuristic guide that could identify what behavioural risk factors influence hospital readmissions through adopting various methods of analysis including regression models, t-tests, data mining, and logistic regression. This study was a retrospective cohort review of internal medicine patients admitted between December 1, 2012 and December 31, 2013 at King Abdullah University Hospital, in Jordan.

RESULTS

29% of all hospitalized patients were readmitted during the study period. Among all readmissions, 44% were identified as potentially avoidable. Behavioural factors including smoking, unclear follow-up and discharge planning, and being non-compliant with treatment regimen as well as discharge against medical advice were all associated with increased risk of avoidable readmissions.

CONCLUSION

Implementing evidence based health programs that focus on modifiable behavioural risk factors for both patients and clinicians would yield a higher response in terms of reducing potentially avoidable readmissions, and could reduce direct medical costs.

Potentially Preventable 30-Day Hospital Readmissions at a Children's Hospital

BACKGROUND AND OBJECTIVES

Hospital readmission rates are increasingly used to assess quality. Little is known, however, about potential preventability of readmissions among children. Our objective was to evaluate potential preventability of 30-day readmissions using medical record review and interviews.

METHODS

A cross-sectional study in 305 children (<18 years old) readmitted within 30 days to a freestanding children's hospital between December 2012 and February 2013. Interviews (N = 1459) were conducted with parents/guardians, patients (if ≥13 years old), inpatient clinicians, and primary care providers. Reviewers evaluated medical records, interview summaries, and transcripts, and then rated potential preventability. Multivariate regression analysis was used to identify factors associated with potentially preventable readmission. Adjusted event curves were generated to model days to readmission.

RESULTS

Of readmissions, 29.5% were potentially preventable. Potentially preventable readmissions occurred sooner after discharge than non-potentially preventable readmissions (5 vs 9 median days; P < .001). The odds of a readmission being potentially preventable were greatest when the index admission and readmission were causally related (adjusted odds ratio [AOR]: 2.6; 95% confidence interval [CI]: 1.0-6.8) and when hospital (AOR: 16.3; 95% CI: 5.9-44.8) or patient (AOR: 7.1; 95% CI: 2.5-20.5) factors were identified. Interviews provided new information about the readmission in 31.2% of cases.

CONCLUSIONS

Nearly 30% of 30-day readmissions to a children's hospital may be potentially preventable. Hospital and patient factors are associated with potential preventability and may provide targets for quality improvement efforts. Interviews contribute important information and should be considered when evaluating readmissions.

Hospital Readmission Through the Emergency Department

BACKGROUND

Hospital readmission places a high burden on both health care systems and patients. Most readmissions are thought to be related to the quality of the health care system.

OBJECTIVES

The aim of this study was to examine the causes and rates of early readmission in emergency department in a Tehran hospital.

PATIENTS AND METHODS

A cross-sectional investigation was performed to study readmission of inpatients at a large academic hospital in Tehran, Iran. Patients admitted to hospital from July 1, 2014 to December 30, 2014 via the emergency department were enrolled. Descriptive statistics were used to summarize the distribution demographics in the sample. Data was analyzed by chi2 test using SPSS 20 software.

RESULTS

The main cause of readmission was complications related to surgical procedures (31.0%). Discharge from hospital based on patient request at the patient's own risk was a risk factor for emergency readmission in 8.5%, a very small number were readmitted after complete treatment (0.6%). The only direct complication of treatment was infection (17%).

CONCLUSIONS

Postoperative complications increase the probability of patients returning to hospital. Physicians, nurses, etc., should focus on these specific patient populations to minimize the risk of postoperative complications. Future studies should assess the relative connections of various types of patient information (e.g., social and psychosocial factors) to readmission risk prediction by comparing the performance of models with and without this information in a specific population.

Parent and Physician Perceptions Regarding Preventability of Pediatric Readmissions

OBJECTIVE

To assess the causes and preventability of pediatric readmissions from the perspectives of parents and their physicians to guide future interventions.

METHODS

Parent interview, physician survey, and medical record review were completed for children who were readmitted to a pediatric hospitalist service within 30 days of an index admission. Questions were asked about Health Belief Model constructs (perceived severity, susceptibility or preventability of admission, and perceived barriers), discharge readiness, and follow-up plans. Parent and physician perceptions about reasons for readmissions were examined, and responses to open-ended questions were coded.

RESULTS

60 parent-physician pairs completed the study. The mean age of the patients was 6.43 (SD 6.42) years; 45% (n=27) had a chronic disease, and 47% (n=28) of patients were readmitted with the same or similar condition as in the previous hospitalization. At readmission, parents were more likely than physicians to think that the condition was serious (parent 98%, physician 76%; P<.001) and that the readmission could have been prevented (parent 59%, physician 36%; P=.04). Most parents (63%) and physicians (65%) thought it was likely that the child may have future hospitalizations. Opportunities to prevent readmission included need for parent education, improving medication access and adherence, and need for coordination of follow-up care.

CONCLUSIONS

Many parents and physicians thought the readmission was preventable, and the majority of both thought that the patient was susceptible to another hospitalization. Parents and physicians suggest opportunities to improve care processes during hospitalization and in services provided after discharge to reduce readmissions.

Characteristics and temporal pattern of the readmissions of patients with multiple hospital admissions in the medical departments of a general hospital

INTRODUCTION

Patients with multiple hospital admissions represent a small percentage of total hospitalizations but result in a considerable proportion of the healthcare expenditure. There are no studies that have analyzed their long-term clinical evolution.

OBJECTIVES

To study the characteristics, temporal patterns of readmissions and clinical evolution of patients with multiple hospital admission in the long term.

METHODS

A retrospective study was conducted of all hospital admissions in the medical area of the Hospital of Lugo (Spain) between January 1, 2000 and December 31, 2012, based on data from the center's minimum basic data set.

RESULTS

A total of 139,249 hospital admissions for 62,515 patients were studied. Six hospital admissions were recorded for 6.4% of the patients. The overall mortality rate was 16% (9982 patients). The readmissions rate steadily increased with each new admission, from 48% after the first event to 74.6% after the fifth. The rate of hospital readmission before 30days increased from 18.3% in the second admission to 36.3% in the sixth. The number of chronic diseases increased from 3.1 (SD, 2) in the first hospital admission up to 4.9 (2.8) in the sixth. The Department of Internal Medicine treated a third of all hospital admissions. In the sixth hospitalization, conditions associated with admission in Internal Medicine were CIRS score, age, heart failure, COPD, dementia, diabetes, atrial fibrillation and anemia.

CONCLUSIONS

Patients with multiple hospital admissions are complex patients whose temporal pattern of readmissions changes with time, such that each hospital admission constitutes a factor facilitating the next.