Half of 30-Day Hospital Readmissions Among HIV-Infected Patients Are Potentially Preventable

Hospital Readmissions Among Persons With Human Immunodeficiency Virus in the United States and Canada, 2005-2018: A Collaboration of Cohort Studies

BACKGROUND

Hospital readmission trends for persons with human immunodeficiency virus (PWH) in North America in the context of policy changes, improved antiretroviral therapy (ART), and aging are not well-known. We examined readmissions during 2005-2018 among adult PWH in NA-ACCORD.

METHODS

Linear risk regression estimated calendar trends in 30-day readmissions, adjusted for demographics, CD4 count, AIDS history, virologic suppression (<400 copies/mL), and cohort.

RESULTS

We examined 20 189 hospitalizations among 8823 PWH (73% cisgender men, 38% White, 38% Black). PWH hospitalized in 2018 versus 2005 had higher median age (54 vs 44 years), CD4 count (469 vs 274 cells/μL), and virologic suppression (83% vs 49%). Unadjusted 30-day readmissions decreased from 20.1% (95% confidence interval [CI], 17.9%-22.3%) in 2005 to 16.3% (95% CI, 14.1%-18.5%) in 2018. Absolute annual trends were -0.34% (95% CI, -.48% to -.19%) in unadjusted and -0.19% (95% CI, -.35% to -.02%) in adjusted analyses. By index hospitalization reason, there were significant adjusted decreases only for cardiovascular and psychiatric hospitalizations. Readmission reason was most frequently in the same diagnostic category as the index hospitalization.

CONCLUSIONS

Readmissions decreased over 2005-2018 but remained higher than the general population's. Significant decreases after adjusting for CD4 count and virologic suppression suggest that factors alongside improved ART contributed to lower readmissions. Efforts are needed to further prevent readmissions in PWH.

Perceptions toward antiretroviral therapy and delayed ART initiation among people living with HIV in Changsha, China: mediating effects of treatment willingness

Introduction

Delayed antiretroviral therapy (ART) initiation is associated with poor HIV outcomes and a higher likelihood of HIV transmission.

Methods

This cross-sectional study assessed the proportion of delayed ART initiation which was defined as initiating ART after 30 days of HIV diagnosis, and evaluated the pathways influencing ART initiation among adult PLWH in Changsha, China who were diagnosed between 2014 and 2022.

Results

Of 518 participants, 37.8% delayed in initiating ART. Based on the theory of reasoned action (TRA), delayed initiation was indirectly associated with perceptions toward ART through the mediating pathway of patients' treatment willingness, with treatment willingness significantly being the full mediator.

Discussion

The findings may guide the development of interventions to improve timely uptake of ART in people who are newly diagnosed with HIV.

Interpretable Identification of Comorbidities Associated with Recurrent ED and Inpatient Visits

In the hospital setting, a small percentage of recurrent frequent patients contribute to a disproportional amount of healthcare resource utilization. Moreover, in many of these cases, patient outcomes can be greatly improved by reducing re-occurring visits, especially when they are associated with substance abuse, mental health, and medical factors that could be improved by social-behavioral interventions, outpatient or preventative care. Additionally, health care costs can be reduced significantly with fewer preventable recurrent visits. To address this, we developed a novel, interpretable framework that both identifies recurrent patients with high utilization and determines which comorbidities contribute most to their recurrent visits. Specifically, we present a novel algorithm, called the minimum similarity association rules (MSAR), which balances the confidence-support trade-off, to determine the conditions most associated with re-occurring Emergency department and inpatient visits. We validate MSAR on a large Electronic Health Record dataset, demonstrating the effectiveness and consistency in ability to find low-support comorbidities with high likelihood of being associated with recurrent visits, which is challenging for other algorithms such as XGBoost. Clinical relevance- In the era of value-based care and population health management, the proposal could be used for decision making to help reduce future recurrent admissions, improve patient outcomes and reduce the cost of healthcare.

A Multicomponent Intervention to Reduce Readmissions Among People With HIV

BACKGROUND

Hospital readmissions are common, costly, and potentially preventable, including among people with HIV (PWH). We present the results of an evaluation of a multicomponent intervention aimed at reducing 30-day readmissions among PWH.

METHODS

Demographic, socioeconomic, and clinical variables were collected from the electronic health records of PWH or those with cellulitis (control group) hospitalized at an urban safety-net hospital before and after (from September 2012 to December 2016) the implementation of a multidisciplinary HIV transitional care team. After October 2014, hospitalized PWH could receive a medical HIV consultation ± a transitional care nurse intervention. The primary outcome was readmission to any hospital within 30 days of discharge. Multivariate logistic regression and propensity score analyses were conducted to compare readmissions before and after intervention implementation in PWH and people with cellulitis.

RESULTS

Overall, among PWH, 329 of the 2049 (16.1%) readmissions occurred before and 329 of the 2023 (16.3%) occurred after the transitional care team intervention. After including clinical and social predictors, the adjusted odds ratio of 30-day readmissions for postintervention for PWH was 0.81 (95% confidence interval: 0.66 to 0.99, P= 0.04), whereas little reduction was identified for those with cellulitis (adjusted odds ratio 0.91 (95% confidence interval: 0.81 to 1.02, P= 0.10). A dose-response effect was not observed for receipt of different HIV intervention components.

CONCLUSIONS

A multicomponent intervention reduced the adjusted risk of 30-day readmissions in PWH, although no dose-response effect was detected. Additional efforts are needed to reduce overall hospitalizations and readmissions among PWH including increasing HIV prevention, early diagnosis and engagement in care, and expanding the availability and spectrum of transitional care services.

Outcomes of people living with HIV after hospital discharge: a systematic review and meta-analysis

Background

The identification and appropriate management of people with advanced HIV disease is a key component in the HIV response. People with HIV who are hospitalised are at a higher risk of death, a risk that might persist after discharge. The aims of this study were to estimate the frequency of negative post-discharge outcomes, and to determine risk factors for such outcomes in people with HIV.

Methods

Using a broad search strategy combining terms for hospital discharge and HIV infection, we searched MEDLINE via PubMed and Embase from Jan 1, 2003 to Nov 30, 2021 to identify studies reporting outcomes among people with HIV following discharge from hospital. We estimated pooled proportions of readmissions and deaths after hospital discharge using random-effects models. We also did subgroup analyses by setting, region, duration of follow-up, and advanced HIV status at admission, and sensitivity analyses to assess heterogeneity.

Findings

We obtained data from 29 cohorts, which reported outcomes of people living with HIV after hospital discharge in 92 781 patients. The pooled proportion of patients readmitted to hospital after discharge was 18·8% (95% CI 15·3–22·3) and 14·1% (10·8–17·3) died post-discharge. In sensitivity analyses, no differences were identified in the proportion of patients who were readmitted or died when comparing studies published before 2016 with those published after 2016. Post-discharge mortality was higher in studies from Africa (23·1% [16·5–29·7]) compared with the USA (7·5% [4·4–10·6]). For studies that reported both post-discharge mortality and readmission, the pooled proportion of patients who had this composite adverse outcome was 31·7% (23·9–39·5). Heterogeneity was moderate, and largely explained by patient status and linkage to care. Reported risk factors for readmission included low CD4 cell count at admission, longer length of stay, discharge against medical advice, and not linking to care following discharge; inpatient treatment with antiretroviral therapy (ART) during hospitalisation was protective of post-discharge mortality.

Interpretation

More than a quarter of patients with HIV had an adverse outcome after hospital discharge with no evidence of improvement in the past 15 years. This systematic review highlights the importance of ensuring post-discharge referral and appropriate management, including ART, to reduce mortality and readmission to hospital among this group of high-risk patients.

Funding

Bill & Melinda Gates Foundation.

Translations

For the French and Spanish translations of the abstract see Supplementary Materials section.

CD4+ cell count and outcomes among HIV-infected compared with uninfected medical ICU survivors in a national cohort

BACKGROUND

People with HIV (PWH) with access to antiretroviral therapy (ART) experience excess morbidity and mortality compared with uninfected patients, particularly those with persistent viremia and without CD4+ cell recovery. We compared outcomes for medical intensive care unit (MICU) survivors with unsuppressed (>500 copies/ml) and suppressed (≤500 copies/ml) HIV-1 RNA and HIV-uninfected survivors, adjusting for CD4+ cell count.

SETTING

We studied 4537 PWH [unsuppressed = 38%; suppressed = 62%; 72% Veterans Affairs-based (VA) and 10 531 (64% VA) uninfected Veterans who survived MICU admission after entering the Veterans Aging Cohort Study (VACS) between fiscal years 2001 and 2015.

METHODS

Primary outcomes were all-cause 30-day and 6-month readmission and mortality, adjusted for demographics, CD4+ cell category (≥350 (reference); 200-349; 50-199; <50), comorbidity and prior healthcare utilization using proportional hazards models. We also adjusted for severity of illness using discharge VACS Index (VI) 2.0 among VA-based survivors.

RESULTS

In adjusted models, CD4+ categories <350 cells/μl were associated with increased risk for both outcomes up to 6 months, and risk increased with lower CD4+ categories (e.g. 6-month mortality CD4+ 200-349 hazard ratio [HR] = 1.35 [1.12-1.63]; CD4+ <50 HR = 2.14 [1.72-2.66]); unsuppressed status was not associated with outcomes. After adjusting for VI in models stratified by HIV, VI quintiles were strongly associated with both outcomes at both time points.

CONCLUSION

PWH who survive MICU admissions are at increased risk for worse outcomes compared with uninfected, especially those without CD4+ cell recovery. Severity of illness at discharge is the strongest predictor for outcomes regardless of HIV status. Strategies including intensive case management for HIV-specific and general organ dysfunction may improve outcomes for MICU survivors.

A Retrospective Analysis on Risk Factors for 30-day Readmission Rates in Patients Living With HIV and Severe Major Depression Disorder

Background

Major depression disorder (MDD) is the most common psychiatric comorbidity in patients living with HIV (PLWHIV). The prevalence rate of MDD is higher in PLWHIV in comparison to the general population. In our study, we focus specifically on the 30-day readmission rate of PLWHIV and severe major depression.

Methods

The Health Care Agency (HCA) databank was used to conduct a retrospective study on PLWHIV and severe MDD. Keywords such as HIV, severe MDD, CD4, viral load were used to identify the data. 30-day readmission rate is studied in PLWHIV and severe MDD (N=143). Variables such as age, sex, gender, adherence to antiretroviral medications, cluster of differentiation 4 (CD4), and viral load were studied in this population. Diagnostic and Statistical Manual of Mental Disorders, 5th Edition (DSM-5) criteria were used to diagnose severe MDD in PLWHIV. An antiretroviral therapy (ART) medication list was used to analyze adherence in this population group. Geographical locations were identified using urbanization codes.

Results

Logistic regression analysis for the 30-day readmission rate in PLWHIV was found to be higher in the older age group (p<0.01). Caucasian population (p<0.01) and rural areas (p<0.01), ART non-adherence (p<0.05), and severe major depression were also found to be significant in this population (p<0.01).

Conclusion

As more patients live longer with HIV/AIDS, it gives rise to illnesses such as anxiety, depression, and cognitive impairment. Thus, it is important to identify severe depression in PLWHIV since it can have an impact on rates of hospitalization, morbidity/mortality, and the financial burden, specifically within 30-days of discharge.

Hospitalization Rates and Outcomes Among Persons Living With Human Immunodeficiency Virus in the Southeastern United States, 1996-2016

BACKGROUND

Antiretroviral therapy (ART) advances, aging, and comorbidities impact hospitalizations in human immunodeficiency virus (HIV)-positive populations. We examined temporal trends and patient characteristics associated with hospitalization rates and outcomes.

METHODS

Among patients in the University of North Carolina Center for AIDS Research HIV Clinical Cohort receiving care during 1996-2016, we estimated annual hospitalization rates, time to inpatient mortality or live discharge, and 30-day readmission risk using bivariable Poisson, Fine-Gray, and log-binomial regression models.

RESULTS

The 4323 included patients (29% women, 60% African American) contributed 30 007 person-years. Overall, the hospitalization rate per 100 person-years was 34.3 (95% confidence interval [CI], 32.4-36.4) with a mean annual change of -3% (95% CI, -4% to -2%). Patients who were black (vs white), older, had HIV RNA >400 copies/mL, or had CD4 count <200 cells/μL had higher hospitalization rates (all P < .05). Thirty-day readmission risk was 18.9% (95% CI, 17.7%-20.2%), stable over time (P > .05 for both 2010-2016 and 2003-2009 vs 1996-2002), and higher among black patients, those with detectable HIV RNA, and those with lower CD4 cell counts (all P < .05). Higher inpatient mortality was associated with older age and lower CD4 cell count (both P < .05).

CONCLUSIONS

Hospitalization rates decreased from 1996 to 2016, but high readmissions persisted. Older patients, those of minority race/ethnicity, and those with uncontrolled HIV experienced higher rates and worse hospitalization outcomes. These findings underscore the importance of early ART and care engagement, particularly at hospital discharge.

COVID-19: What Is Next for Portugal?

Highly infectious with the possibility of causing severe respiratory complications, the novel COVID-19 began stretching health systems beyond their capacity all over the world and pushing them to breaking points. Giving the devastating effects caused by this infection, unprecedented measures have to be adopted in order to mitigate its impacts on the health system. This perspective aims to review the epidemic of COVID-19 in Portugal, possible areas of improvement, and potential interventions that can help to mitigate the effect of COVID-19 on the Portuguese health system.

Retention in Care and Viral Load Improvement After Discharge Among Hospitalized Out-of-Care People With HIV Infection: A Post Hoc Analysis of a Randomized Controlled Trial

BACKGROUND

Understanding factors influencing retention in care (RIC) and viral load improvement (VLI) in people with HIV (PWH) who are out of care and hospitalized will assist in intervention development for this vulnerable population.

METHODS

The study was a post hoc analysis of prospectively collected data. Hospitalized participants were enrolled if they were newly diagnosed with HIV during the hospitalization or out of HIV care. Participants completed surveys at baseline and 6 months postenrollment and laboratory studies of viral load (VL). Outcomes were RIC (2 completed visits, 1 within 30 days of discharge) and VLI (VL <400 or at least a 1-log10 decrease) 6 months after discharge. Univariate and multivariate regression analyses were conducted examining the contributions of predisposing, enabling, and need factors to outcomes.

RESULTS

The study cohort included 417 participants enrolled between 2010 and 2013. The population was 73% male, 67% non-Hispanic black, 19% Hispanic, and 70% uninsured. Sixty-five percent had a baseline CD4 <200 cells/mm3, 79% had a VL >400 copies/mL or missing, and the population was generally poor with low educational attainment. After discharge from the hospital, 60% did not meet the definition for RIC, and 49% did not have VLI. Modifiable factors associated with the outcomes include drug use (including marijuana alone and other drugs), life instability (eg, housing, employment, and life chaos), and using avoidance coping strategies in coping with HIV.

CONCLUSIONS

Hospitalized out-of-care PWH in the United States are at high risk of poor re-engagement in care after discharge. Interventions for this population should focus on improving socioeconomic stability and coping with HIV and reducing drug use.

Patient and Provider Perspectives on 30-Day Readmissions, Preventability, and Strategies for Improving Transitions of Care for Patients with HIV at a Safety Net Hospital

Thirty-day hospital readmissions, a key quality metric, are common among people living with HIV. We assessed perceived causes of 30-day readmissions, factors associated with preventability, and strategies to reduce preventable readmissions and improve continuity of care for HIV-positive individuals. Patient, provider, and staff perspectives toward 30-day readmissions were evaluated in semistructured interviews (n = 86) conducted in triads (HIV-positive patient, medical provider, and case manager) recruited from an inpatient safety net hospital. Iterative analysis included both deductive and inductive themes. Key findings include the following: (1) The 30-day metric should be adjusted for safety net institutions and patients with AIDS; (2) Participants disagreed about preventability, especially regarding patient-level factors; (3) Various stakeholders proposed readmission reduction strategies that spanned the inpatient to outpatient care continuum. Based on these diverse perspectives, we outline multiple interventions, from teach-back patient education to postdischarge home visits, which could substantially decrease hospital readmissions in this underserved population.

Preventability of unplanned readmissions within 30 days of discharge. A cross-sectional, single-center study

Objectives

To identify the preventability, determinants and causes of unplanned hospital readmissions within 30 days of discharge using a multidisciplinary approach and including patients’ perspectives.

Design

A prospective cross-sectional single-center study.

Setting

Urban teaching hospital in Amsterdam, the Netherlands.

Participants

430 patients were included. Inclusion criteria were: age ≥ 18 years, discharged from one of seven participating clinical departments and an unplanned readmission within 30 days.

Methods

Residents from the participating departments individually assessed whether the readmission was caused by healthcare, the preventability and possible causes of readmissions using a tool. Thereafter, the preventability of the cases was discussed in a multidisciplinary meeting with residents of all participating departments and clinical pharmacists. The primary outcome was the proportion of readmissions that were potentially preventable. Secondary outcomes were the determinants for a readmission, causes for preventable readmissions, the change in the final decision on preventability after the multidisciplinary meeting and the value of patient interviews in assessing preventability. Differences in characteristics of potentially preventable readmissions (PPRs) and non-PPRs were analyzed using multivariable logistic regression.

Results

Of 430 readmissions, 56 (13%) were assessed as PPRs. Age was significantly associated with a PPR (adjusted OR: 2.42; 95%, CI 1.23–4.74; p = 0.01). The main causes for PPRs were diagnostic (30%), medication (27%) and management problems (27%). During the multidisciplinary meeting, the final decision on preventability changed in 11% of the cases. When a patient interview was available, it was used as a source of information to assess preventability in 26% of readmissions. In 7% of cases, the patient interview was mentioned as the most important source.

Conclusion and implications

13% of readmissions were potentially preventable with diagnostic, medication or management problems being main causes. A multidisciplinary review approach and including the patient’s perspective could contribute to a better understanding of the complexity of readmissions and possible improvements.

Risk of Readmission Among HIV Patients in Public Portuguese Hospitals: Longitudinal Multilevel Population-Based Study

Background: Thirty-day hospital readmission is receiving growing attention as an indicator of the quality of hospital care. Understanding factors associated with 30-day hospital readmission among HIV patients in Portugal is essential given the high burden cost of HIV hospitalizations in Portugal, a country suffering from financial constrains for almost 10 years. Objectives: We aimed to estimate the 30-day hospital readmission rates among HIV patients in Portugal and to identify its determinants using population-based data for Portuguese public hospitals. Study Design: A multilevel longitudinal population-based study. Methods: Between January 2009 and December 2014, a total of 37,134 registered discharges in the Portuguese National Health Service (NHS) facilities with HIV/AIDS as a main or secondary cause of admission were analyzed. Logistic regression was used to compare 30-day hospital readmission categories by computing odds ratio (OR) and corresponding 95% confidence intervals (95% CIs). A normal random effects model was used to determine unmeasured factors specific to each hospital. Results: A total of 4914 (13.2%, 95% CI: 12.9%-13.6%) hospitalizations had a subsequent 30-day readmission. Hospitalizations that included exit against medical opinion (OR = 1.18, 95% CI: 1.01-1.39), scheduled admissions (OR = 1.71, 95% CI: 1.58-1.85), and tuberculosis infection (OR = 1.20, 95% CI: 1.05-1.38) exhibited a higher risk of hospitalizations with subsequent 30-day readmission. In contrast, hospitalizations that included females (OR = 0.87, 95% CI: 0.81-0.94), a transfer to another facility (OR = 0.78, 95% CI: 0.67-0.91), and having a responsible financial institution (OR = 0.63, 95% CI: 0.55-0.72) exhibited a lower risk of hospitalizations with subsequent 30-day readmission. Hospitalizations associated with higher number of diagnosis, older ages, or hospitalizations during the economic crisis showed an increasing trend of 30-day readmission, whereas an opposite trend was observed for hospitalizations with higher number of procedures. Significant differences exist between hospital quality, adjusting for other factors. Conclusion: This study analyzes the indicators of 30-day hospital readmission among HIV patients in Portugal and provides useful information for enlightening policymakers and health care providers for developing health policies that can reduce costs associated with HIV hospitalizations.

Risk factors for potentially preventable hospital readmissions among persons living with human immunodeficiency virus infection

People living with HIV (PLWH) have a significant risk for experiencing a 30-day readmission; many of which may be potentially preventable readmissions (PPR). The objective of this study was to evaluate 30-day readmission rates for PLWH and identify risk factors for PPR. This was a single center retrospective study. Patients were included if they were ≥18 years of age, had a diagnosis of HIV, and were admitted to University of New Mexico Hospitals between 1 January 2010 and 31 December 2014 and readmitted within 30-days of the index admission. Preventability of readmission was defined using previously published criteria. Of the 908 identified admissions for PLWH during 2010-2014, 162 (17.8%) were 30-day readmissions. A total of 60 patient readmissions met study inclusion criteria, of which 55% were determined to be PPR. Multivariate logistic regression analysis revealed that being discharged on ≥10 medications (OR 3.92, 95% CI 1.181-13.043) and having an appointment scheduled upon discharge (OR 3.59, 95% CI 1.057-12.212) were significantly associated a PPR. These results further highlight the vulnerability of this patient population and help to identify risk factors for PPR. Targeted transitions of care interventions that address polypharmacy may help to reduce PPR among PLWH.

Clinical and Sociobehavioral Prediction Model of 30-Day Hospital Readmissions Among People With HIV and Substance Use Disorder: Beyond Electronic Health Record Data

BACKGROUND

Under the Affordable Care Act, hospitals receive reduced reimbursements for excessive 30-day readmissions. However, the Centers for Medicare and Medicaid Services does not consider social and behavioral variables in expected readmission rate calculations, which may unfairly penalize systems caring for socially disadvantaged patients, including patients with HIV.

SETTING

Randomized controlled trial of patient navigation with or without financial incentives in HIV-positive substance users recruited from the inpatient setting at 11 US hospitals.

METHODS

External validation of an existing 30-day readmission prediction model, using variables available in the electronic health record (EHR-only model), in a new multicenter cohort of HIV-positive substance users was assessed by C-statistic and Hosmer-Lemeshow testing. A second model evaluated sociobehavioral factors in improving the prediction model (EHR-plus model) using multivariable regression and C-statistic with cross-validation.

RESULTS

The mean age of the cohort was 44.1 years, and participants were predominantly males (67.4%), non-white (88.0%), and poor (62.8%, <$20,000/year). Overall, 17.5% individuals had a hospital readmission within 30 days of initial hospital discharge. The EHR-only model resulted in a C-statistic of 0.65 (95% confidence interval: 0.60 to 0.70). Inclusion of additional sociobehavioral variables, food insecurity and readiness for substance use treatment, in the EHR-plus model resulted in a C-statistic of 0.74 (0.71 after cross-validation, 95% confidence interval: 0.64 to 0.77).

CONCLUSIONS

Incorporation of detailed social and behavioral variables substantially improved the performance of a 30-day readmission prediction model for hospitalized HIV-positive substance users. Our findings highlight the importance of social determinants in readmission risk and the need to ask about, adjust for, and address them.

Towards a patient journey perspective on causes of unplanned readmissions using a classification framework: results of a systematic review with narrative synthesis

BACKGROUND

Several literature reviews have been published focusing on the prevalence and/or preventability of hospital readmissions. To our knowledge, none focused on the different causes which have been used to evaluate the preventability of readmissions. Insight into the range of causes is crucial to understand the complex nature of readmissions. We conducted a systematic review to: (1) evaluate the range of causes of unplanned readmissions in a patient journey, and (2) present a cause classification framework that can support future readmission studies.

METHODS

A literature search was conducted in PUBMED and EMBASE using "readmission" and "avoidability" or "preventability" as key terms. Studies that specified causes of unplanned readmissions were included. The causes were classified into eight preliminary root causes: Technical, Organization (integrated care), Organization (hospital department level), Human (care provider), Human (informal caregiver), Patient (self-management), Patient (disease), and Other. The root causes were based on expert opinions and the root cause analysis tool of PRISMA (Prevention and Recovery Information System for Monitoring and Analysis). The range of different causes were analyzed using Microsoft Excel.

RESULTS

Forty-five studies that reported 381 causes of readmissions were included. All studies reported causes related to organization of care at the hospital department level. These causes were often reported as preventable. Twenty-two studies included causes related to patient's self-management and 19 studies reported causes related to patient's disease. Studies differed in which causes were seen as preventable or unpreventable. None reported causes related to technical failures and causes due to integrated care issues were reported in 18 studies.

CONCLUSIONS

This review showed that causes for readmissions were mainly evaluated from a hospital perspective. However, causes beyond the scope of the hospital can also play a major role in unplanned readmissions. Opinions regarding preventability seem to depend on contextual factors of the readmission. This study presents a cause classification framework that could help future readmission studies to gain insight into a broad range of causes for readmissions in a patient journey.

Analysis and prediction of unplanned intensive care unit readmission using recurrent neural networks with long short-term memory

BACKGROUND

Unplanned readmission of a hospitalized patient is an indicator of patients' exposure to risk and an avoidable waste of medical resources. In addition to hospital readmission, intensive care unit (ICU) readmission brings further financial risk, along with morbidity and mortality risks. Identification of high-risk patients who are likely to be readmitted can provide significant benefits for both patients and medical providers. The emergence of machine learning solutions to detect hidden patterns in complex, multi-dimensional datasets provides unparalleled opportunities for developing an efficient discharge decision-making support system for physicians and ICU specialists.

METHODS AND FINDINGS

We used supervised machine learning approaches for ICU readmission prediction. We used machine learning methods on comprehensive, longitudinal clinical data from the MIMIC-III to predict the ICU readmission of patients within 30 days of their discharge. We incorporate multiple types of features including chart events, demographic, and ICD-9 embeddings. We have utilized recent machine learning techniques such as Recurrent Neural Networks (RNN) with Long Short-Term Memory (LSTM), by this we have been able to incorporate the multivariate features of EHRs and capture sudden fluctuations in chart event features (e.g. glucose and heart rate). We show that our LSTM-based solution can better capture high volatility and unstable status in ICU patients, an important factor in ICU readmission. Our machine learning models identify ICU readmissions at a higher sensitivity rate of 0.742 (95% CI, 0.718-0.766) and an improved Area Under the Curve of 0.791 (95% CI, 0.782-0.800) compared with traditional methods. We perform in-depth deep learning performance analysis, as well as the analysis of each feature contribution to the predictive model.

CONCLUSION

Our manuscript highlights the ability of machine learning models to improve our ICU decision-making accuracy and is a real-world example of precision medicine in hospitals. These data-driven solutions hold the potential for substantial clinical impact by augmenting clinical decision-making for physicians and ICU specialists. We anticipate that machine learning models will improve patient counseling, hospital administration, allocation of healthcare resources and ultimately individualized clinical care.

How do studies assess the preventability of readmissions? A systematic review with narrative synthesis

Background

A large number of articles examined the preventability rate of readmissions, but comparison and interpretability of these preventability rates is complicated due to the large heterogeneity of methods that were used.

To compare (the implications of) the different methods used to assess the preventability of readmissions by means of medical record review.

Methods

A literature search was conducted in PUBMED and EMBASE using “readmission” and “avoidability” or “preventability” as key terms. A consensus-based narrative data synthesis was performed to compare and discuss the different methods.

Results

Abstracts of 2504 unique citations were screened resulting in 48 full text articles which were included in the final analysis. Synthesis led to the identification of a set of important variables on which the studies differed considerably (type of readmissions, sources of information, definition of preventability, cause classification and reviewer process). In 69% of the studies the cause classification and preventability assessment were integrated; meaning specific causes were predefined as preventable or not preventable. The reviewers were most often medical specialist (67%), and 27% of the studies added interview as a source of information.

Conclusion

A consensus-based standardised approach to assess preventability of readmission is warranted to reduce the unwanted bias in preventability rates. Patient-related and integrated care related factors are potentially underreported in readmission studies.

Electronic supplementary material

The online version of this article (10.1186/s12874-019-0766-0) contains supplementary material, which is available to authorized users.

Impact of Patient-Provider Attachment on Hospital Readmissions Among People Living With HIV: A Population-Based Study

BACKGROUND

Hospital readmission 30 days after discharge is associated with adverse health outcomes, and people living with HIV (PLWH) experience elevated rates of hospital readmission. Although continuity of care with a health care provider is associated with lower rates of 30-day readmission among the general population, little is known about this relationship among PLWH. The objective of this study is to examine whether engaging with the same provider, defined as patient-provider attachment, is associated with 30-day readmission for this population.

SETTING

Data were derived from the Seek and Treat for Optimal Prevention of HIV in British Columbia cohort.

METHODS

Using generalized estimating equation with a logit link function, we examined the association between patient-provider attachment and 30-day hospital readmission. We determined whether readmission was due to all cause or to a similar cause as the index admission.

RESULTS

Seven thousand thirteen PLWH were hospitalized during the study period. Nine hundred twenty-one (13.1%) were readmitted to hospital for all cause and 564 (8.0%) for the similar cause as the index admission. Patient-provider attachment was negatively associated with 30-day readmission for all causes (adjusted odds ratio = 0.85, confidence interval = 0.83 to 0.86). A second multivariable model indicated that patient-provider attachment was also negatively associated with 30-day readmission for a similar cause (adjusted odds ratio = 0.86, confidence interval = 0.84 to 0.88).

CONCLUSIONS

Our results indicate that a higher proportion of patient-provider attachment was negatively associated with 30-day hospital readmission among PLWH. Our study findings support the adoption of interventions that seek to build patient-provider relationships to optimize outcomes for PLWH and enhance health care sustainability.

Claims data-driven modeling of hospital time-to-readmission risk with latent heterogeneity

Hospital readmission risk modeling is of great interest to both hospital administrators and health care policy makers, for reducing preventable readmission and advancing care service quality. To accommodate the needs of both stakeholders, a readmission risk model is preferable if it (i) exhibits superior prediction performance; (ii) identifies risk factors to help target the most at-risk individuals; and (iii) constructs composite metrics to evaluate multiple hospitals, hospital networks, and geographic regions. Existing work mainly addressed the first two features and it is challenging to address the third one because available medical data are fragmented across hospitals. To simultaneously address all three features, this paper proposes readmission risk models with incorporation of latent heterogeneity, and takes advantage of administrative claims data, which is less fragmented and involves larger patient cohorts. Different levels of latent heterogeneity are considered to quantify the effects of unobserved factors, provide composite measures for performance evaluation at various aggregate levels, and compensate less informative claims data. To demonstrate the prediction performances of the proposed models, a real case study is considered on a state-wide heart failure patient cohort. A systematic comparison study is then carried out to evaluate the performances of 49 risk models and their variants.

Thirty-day Readmission Rates in an HIV-infected Cohort From Rio de Janeiro, Brazil

BACKGROUND

The 30-day readmission rate is an indicator of the quality of hospital care and transition to the outpatient setting. Recent studies suggest HIV infection might increase the risk of readmission although estimates of 30-day readmission rates are unavailable among HIV-infected individuals living in middle/low-income settings. Additionally, factors that may increase readmission risk in HIV-infected populations are poorly understood.

METHODS

Thirty-day readmission rates were estimated for HIV-infected adults from the Instituto Nacional de Infectologia Evandro Chagas/Fiocruz cohort in Rio de Janeiro, Brazil, from January 2007 to December 2013. Cox regression models were used to evaluate factors associated with the risk of 30-day readmission.

RESULTS

Between January 2007 and December 2013, 3991 patients were followed and 1861 hospitalizations were observed. The estimated 30-day readmission rate was 14% (95% confidence interval: 12.3 to 15.9). Attending a medical visit within 30 days after discharge (adjusted hazard ratio [aHR] = 0.73, P = 0.048) and being hospitalized in more recent calendar years (aHR = 0.89, P = 0.002) reduced the risk of 30-day readmission. In contrast, low CD4 counts (51-200 cells/mm³: aHR = 1.70, P = 0.024 and ≤ 50 cells/mm³: aHR = 2.05, P = 0.003), time since HIV infection diagnosis ≥10 years (aHR = 1.58, P = 0.058), and leaving hospital against medical advice (aHR = 2.67, P = 0.004) increased the risk of 30-day readmission.

CONCLUSIONS

Patients with advanced HIV/AIDS are most at risk of readmission and should be targeted with prevention strategies to reduce this risk. Efforts to reduce discharge against medical advice and to promote early postdischarge medical visit would likely reduce 30-day readmission rates in our population.

Results and Implications of Routine HIV Testing in the Inpatient Setting: A Descriptive Analysis

Policy changes and scientific advances have guided new methods of diagnosing and managing HIV that reduce mortality, morbidity, and transmission. In a high HIV prevalence urban setting, a hospital initiative was implemented to routinely perform HIV testing and provide linkage to care for those with positive results and for individuals with a prior diagnosis of HIV. Maryland's unique all-payer model presents an opportunity to implement population health initiatives in health systems. The rationale, methodology, results and lessons learned from this approach will be discussed. Providers and nurses offered routine HIV screening and activated a Linkage to Care Navigator (LCN) for all HIV positive patients. The LCN provided referrals to HIV care and supportive services. In 22 months, 28 persons were newly diagnosed with HIV. Eighty-two percent (n = 23) were linked to outpatient care; 28.6% (8) were readmitted within 30 days for an inpatient stay. Of 517 patients previously diagnosed with HIV, 27.7% (n = 143) were not engaged in outpatient HIV care. Nearly 50% of those (n = 71) were relinked to care. Of 143 patients with a previous diagnosis who were considered out of care at the time of inpatient admission, 16 (11.2%) were readmitted as an inpatient within 30 days. Routinizing HIV testing and linkage to care in an inpatient setting identifies new and previously diagnosed HIV infected individuals who are not in care. This process has potential to identify HIV earlier, lower community viral load, and decrease transmission of HIV.

Readmissions in HIV-Infected Inpatients: A Large Cohort Analysis

OBJECTIVE

Hospital readmissions impose considerable physical and psychological hardships on patients and represent a high, but possibly preventable, cost for insurers and hospitals alike. The objective of this study was to identify patient characteristics associated with 30-day readmission among persons living with HIV/AIDS (PLWH) using a statewide administrative database and to characterize the movement of patients between facilities.

DESIGN

Retrospective cohort analysis of HIV-infected individuals in New York State using a comprehensive, all-payer database.

SETTING

All hospitals in New York State.

PARTICIPANTS

HIV-infected adults admitted to a medical service in 2012. PLWH identified using International Classification of Disease (ICD)-9 diagnosis codes 042 and V08.

RESULTS

Of 23,544 index hospitalizations, 21.8% (5121) resulted in readmission. Multivariable predictors of readmission included insurance status, housing instability, psychoses, multiple comorbid chronic conditions, substance use, and past inpatient and emergency department visits. Over 30% of readmissions occurred at a different facility than that of the initial hospitalization.

CONCLUSION

A number of patient characteristics were independently associated with hospital readmission within 30 days. Behavioral health disorders and comorbid conditions may be the strongest predictors of readmission in PLWH. Readmissions, especially those in urban areas, often result in fragmented care which may compromise the quality of care and result in harmful discontinuity of medical treatment.