The effect of diabetes on hospital readmissions

Multifactorial risk factors for hospital readmissions among patients with symptoms of advanced heart failure

AIMS

Economic burden of heart failure is attributed to hospital readmissions. Previous studies assessing risk factors for readmissions have focused on single group of risk factors, were limited to 30-day readmissions, or did not account for competing risk of mortality. This study investigates the biological, socio-economic, and behavioural risk factors predicting hospital readmissions while accounting for the competing risk of mortality.

METHODS AND RESULTS

In this prospective cohort study, we recruited 250 patients hospitalized with symptoms of advanced heart failure [New York Heart Association (NYHA) Class III and IV] between July 2017 and April 2019. We analysed their baseline survey data and their hospitalization records over the next 4.5 years (July 2017 to January 2022). We used a joint-frailty model to determine the multifactorial risk factors for all-cause and unplanned hospital readmissions and mortality. At the time of recruitment, patients' mean (SD) age was 66 (12) years, majority being male (72%) and NYHA class IV (68%) with reduced ejection fraction (72%). 87% of the patients had poor self-care behaviours, 51% had diabetes and 56% had weak grip strength. Within 2 years of a hospital admission, 74% of the patients had at least one readmission. Among all readmissions during follow-up, 68% were unplanned. Results from the multivariable regression analysis shows that the independent risk factors for hospital readmissions were biologic-weak grip strength [hazard ratio (95% CI): 1.59 (1.06, 2.13)], poor functional status [1.79 (0.98, 2.61)], diabetes [1.42 (0.97, 1.86)]; behavioural-poor self-care [1.66 (0.84, 2.49)], and socio-economic-preference for maximal life extension at high cost for those with high education [1.98 (1.17, 2.80)]. Risk factors for unplanned hospital readmissions were similar. A higher hospital readmission rate increased the risk of mortality [1.86 (1.23, 2.50)]. Other risk factors for mortality were biologic-weak grip strength [3.65 (0.57, 6.73)], diabetes [2.52 (0.62, 4.42)], socio-economic-lower education [2.45 (0.37, 4.53)], and being married [2.53 (0.37, 4.69)]. Having a private health insurance [0.40 (0.08, 0.73)] lowered the risk for mortality.

CONCLUSIONS

Risk factors for hospital readmissions and mortality are multifactorial. Many of these factors, such as weak grip strength, diabetes, poor self-care behaviours, are potentially modifiable and should be routinely assessed and managed in cardiac clinics and hospital admissions.

Interventions incorporating a multi-disciplinary team approach and a dedicated care team can help reduce preventable hospital readmissions of people with type 2 diabetes mellitus: A scoping review of current literature

AIMS

This review aimed to identify interventions that hospitals can implement to reduce preventable hospital readmissions of people with type 2 diabetes mellitus (T2DM).

METHODS

A scoping review framework was utilised to inform the overall process. The electronic databases Cumulative Index to Nursing and Allied Health Literature (CINAHL), Medline, the University of New England (UNE) library search engine and Google Scholar were utilised to search for relevant literature.

RESULTS

The results from this review demonstrate that interventions started at index admission for people diagnosed with T2DM can result in reductions in hospital readmissions. Common strategies which attributed to the success of interventions in reducing hospital readmissions of people with T2DM included a multidisciplinary team approach, a dedicated care team, certified diabetes educator appointments, basic survival skills education and influencing hospital protocol development and implementation.

CONCLUSION

This scoping review is an attempt at exploring and synthesising current research on interventions that hospitals can implement to reduce preventable hospital readmissions of people with T2DM.

A transition of care model from hospital to community for Hispanic/Latino adult patients with diabetes: design and rationale for a pilot study

BACKGROUND

The Hispanic/Latino population is disproportionately affected and has a higher risk of developing diabetes than their non-Hispanic White counterparts and worse diabetes-related outcomes. Diabetes continues to be an economic burden. This economic burden is partially due to the significantly higher rates of hospital readmission for individuals with diabetes. People with diabetes, particularly those who are members of racial/ethnic minority groups, are at higher risk for readmission and emergency department (ED) visits. Despite recommendations regarding transition of care, an optimal approach to the transition of care for ethnic/minority patients remains unclear.

METHODS

The study population includes self-identified Hispanic/Latino adults with diabetes. We have two aims: (1) designed and developed a transition of care model and (2) pilot test the newly developed transition of care model. For aim 1, semi-structures interviews conducted with patients and providers. For aim 2, patients admitted to the hospital enrolled to receive the newly designed transition of care model. For aim 1, patients and providers completed a short questionnaire. For aim 2, patients completed a set of questionnaires including demographic information, medical history, sociocultural, and social support. The primary outcome for aim 2 is emergency department visit within 30 days post-discharge. The secondary outcome is 30- days unplanned readmissions. Feasibility outcomes include the number of participants identified, number of patients enrolled, number of participants who completed all the questionnaires, number of participants with a 30-day follow-up call, and number of participants who completed the 30-day post-discharge questionnaire. Due to the COVID-19 pandemic, the study design was adapted to include the Plan-Do-Study-Act framework to adjust to the ongoing changes in transition of care due to the pandemic burden on the health care systems.

CONCLUSION

Transition of care for Hispanic/Latino patients with diabetes remains a major area of interest that requires further research. The pandemic required that we adapted the study to reflect the realities of health care systems during a time of crisis. The methods share in this manuscript can potentially help other investigators as they designed their studies.

TRIAL REGISTRATION

ClinicalTrials.gov identifier NCT04864639. 4/29/2021. https://clinicaltrials.gov/ct2/show/NCT04864639 .

A Systematic Analysis of Reviews Exploring the Scope, Validity, and Reporting of Patient-Reported Outcomes Measures of Medication Adherence in Type 2 Diabetes

INTRODUCTION

Non-adherence to medicines is estimated to cost billions to healthcare providers across the US and Europe each year. Addressing medication adherence (MA) can be challenging. Patient-reported outcome measures (PROMs) have been developed to collect self-reported data on MA, among other behaviours. Despite the myriad PROMs available and their widespread implementation in research, there is little commentary or standardization on the way they are reported, or their validity assessed. This review aims to provide a comprehensive analysis of systematic reviews (SRs) that report PROMs of MA with a focus on type 2 diabetes to explore PROM reporting and validity.

MATERIALS AND METHODS

A literature search was conducted using the following databases: PubMed, EMBASE, CINAHL, Cochrane, and Web of Science. SRs reporting on PROMs related to MA behaviour in patients living with type 2 diabetes were included. Any SR published in English prior to December 2021 was included. Abstract and title screening were performed prior to full-text review by two independent researchers with discrepancies managed by a third. Protocols and SRs reporting on paediatric populations were excluded.

RESULTS

A total of 19 eligible SRs that included 241 unique PROM studies were captured from the initial 2074 records that were identified. Data were captured across a 30-year scope, with roughly half (47.4%, n=9/19) of the SRs published in the last 5 years. In total, 104 unique PROMs were identified. Inclusion of non-validated PROMs was identified in 63.2% (n=12/19) of the included SRs, and reporting issues were identified in 47.3% of studies (n=114/241). A lower journal impact factor was significantly associated with a higher prevalence of validity issues (r=0.44, p=0.04).

CONCLUSION

There are a broad range of available PROMs; however, they have been reported inconsistently in the literature, often lacking significant evidence with respect to validity criteria. Standardization of reporting and assessments of validity may help to address this.

The effect of home care on readmission and mortality rate in patients with diabetes who underwent general surgeries

BACKGROUND

More than one-half of people with diabetes need at least one surgery in their lifespan. Few studies have addressed how to manage the needs of these patients after discharge from the hospital. The present study is designed to determine the effect of home care on readmission of Type 2 diabetic patients who underwent surgical procedures.

MATERIALS AND METHODS

The present study was a randomized clinical trial. Sixty-nine patients with Type 2 diabetes undergoing surgery were assigned to the intervention and control groups via blocking order in the selected educational hospitals of Isfahan 2019. Home care was performed for 3 months with interprofessional team approach. Data collection tools were re-admission checklist. Data were entered in SPSS software version 23 and were analyzed by nonparametric tests.

RESULTS

The background characteristics in the intervention and control groups were not different. The frequency of readmission in the control and intervention groups from the time of discharge until 3 months later was 25.7% and 18.9%, respectively. Frequency of readmission in the intervention and control groups was not significant in 3 months from discharge, P > 0.05. The mortality rate was 11.4% and 0% in control and intervention groups, respectively, P < 0.05.

CONCLUSION

It can be argued that continued home care can decrease the rate of readmission and mortality rate in patients with Type 2 diabetes who will discharge from surgical wards.

The 30-days hospital readmission risk in diabetic patients: predictive modeling with machine learning classifiers

BACKGROUND AND OBJECTIVES

Diabetes mellitus is a major chronic disease that results in readmissions due to poor disease control. Here we established and compared machine learning (ML)-based readmission prediction methods to predict readmission risks of diabetic patients.

METHODS

The dataset analyzed in this study was acquired from the Health Facts Database, which includes over 100,000 records of diabetic patients from 1999 to 2008. The basic data distribution characteristics of this dataset were summarized and then analyzed. In this study, 30-days readmission was defined as a readmission period of less than 30 days. After data preprocessing and normalization, multiple risk factors in the dataset were examined for classifier training to predict the probability of readmission using ML models. Different ML classifiers such as random forest, Naive Bayes, and decision tree ensemble were adopted to improve the clinical efficiency of the classification. In this study, the Konstanz Information Miner platform was used to preprocess and model the data, and the performances of the different classifiers were compared.

RESULTS

A total of 100,244 records were included in the model construction after the data preprocessing and normalization. A total of 23 attributes, including race, sex, age, admission type, admission location, length of stay, and drug use, were finally identified as modeling risk factors. Comparison of the performance indexes of the three algorithms revealed that the RF model had the best performance with a higher area under receiver operating characteristic curve (AUC) than the other two algorithms, suggesting that its use is more suitable for making readmission predictions.

CONCLUSION

The factors influencing 30-days readmission predictions in diabetic patients, including number of inpatient admissions, age, diagnosis, number of emergencies, and sex, would help healthcare providers to identify patients who are at high risk of short-term readmission and reduce the probability of 30-days readmission. The RF algorithm with the highest AUC is more suitable for making 30-days readmission predictions and  deserves further validation in clinical trials.

Patient Understanding of Discharge Instructions for Home Diabetes Self-Management and Risk for Hospital Readmission and Emergency Department Visits

OBJECTIVE

The primary objective of this study was to examine the patient comprehension of diabetes self-management instructions provided at hospital discharge as an associated risk of readmission.

METHODS

Noncritically ill patients with diabetes completed patient comprehension questionnaires (PCQ) within 48 hours of discharge. PCQ scores were compared among patients with and without readmission or emergency department (ED) visits at 30 and 90 days. Glycemic measures 48 hours preceding discharge were investigated. Diabetes Early Readmission Risk Indicators (DERRIs) were calculated for each patient.

RESULTS

Of 128 patients who completed the PCQ, scores were similar among those with 30-day (n = 31) and 90-day (n = 54) readmission compared with no readmission (n = 72) (79.9 ± 14.4 vs 80.4 ± 15.6 vs 82.3 ± 16.4, respectively) or ED visits. Clarification of discharge information was provided for 47 patients. PCQ scores of 100% were achieved in 14% of those with and 86% without readmission at 30 days (P = .108). Of predischarge glycemic measures, glycemic variability was negatively associated with PCQ scores (P = .035). DERRIs were significantly higher among patients readmitted at 90 days but not 30 days.

CONCLUSION

These results demonstrate similar PCQ scores between patients with and those without readmission or ED visits despite the need for corrective information in many patients. Measures of glycemic variability were associated with PCQ scores but not readmission risk. This study validates DERRI as a predictor for readmission at 90 days.

The Association of Diabetes and Hyperglycemia on Inpatient Readmissions

OBJECTIVE

To evaluate the association between inpatient glycemic control and readmission in individuals with diabetes and hyperglycemia (DM/HG).

METHODS

Two data sets were analyzed from fiscal years 2011 to 2013: hospital data using the International Classification of Diseases, Ninth Revision (ICD-9) codes for DM/HG and point of care (POC) glucose monitoring. The variables analyzed included gender, age, mean, minimum and maximum glucose, along with 4 measures of glycemic variability (GV), standard deviation, coefficient of variation, mean amplitude of glucose excursions, and average daily risk range.

RESULTS

Of 66 518 discharges in FY 2011-2013, 28.4% had DM/HG based on ICD-9 codes and 53% received POC monitoring. The overall readmission rate was 13.9%, although the rates for individuals with DM/HG were higher at 18.9% and 20.6% using ICD-9 codes and POC data, respectively. The readmitted group had higher mean glucose (169 ± 47 mg/dL vs 158 ± 46 mg/dL, P < .001). Individuals with severe hypoglycemia and hyperglycemia had the highest readmission rates. All 4 GV measures were consistent and higher in the readmitted group.

CONCLUSION

Individuals with DM/HG have higher 30-day readmission rates than those without. Those readmitted had higher mean glucose, more extreme glucose values, and higher GV. To our knowledge, this is the first report of multiple metrics of inpatient glycemic control, including GV, and their associations with readmission.

Laboratory and Pharmaceutical Data Associated With Hospital Readmission in Persons With Diabetic Foot Ulcers

Purpose: Diabetic foot ulcers (DFUs) are a leading cause of lower extremity amputations among persons with diabetes (PWD) and a common cause of hospitalizations. This study identified demographic characteristics, lab values, and comorbidities associated with 30-day and 90-day hospital readmission in persons with DFU.

Methods: A retrospective chart review at our institution examined 397 patients with type 2 diabetes admitted with DFU between January 2014 and December 2018. Variables were analyzed using descriptive statistics, t-tests, and logistic regressions.

Results: None of the studied demographic, laboratory (including Hemoglobin A1c) or comorbid diseases were associated with 30-day readmission in persons with DFU. Risk factors for 90-day readmission included discharge location to home with health care (OR: 2.62, 95% CI: 1.39, 4.95), anticoagulant use (OR: 2.36, 95% CI: 1.27, 4.39), and SQ insulin use (OR: 2.08, 95% CI: 1.20, 3.61).

Conclusions: None of the variables examined were associated with 30-day readmission; however, potential predictors for 90-day readmission included anticoagulation or insulin use and discharge home with healthcare services. Future studies should devise interventions to improve transition of care in patients with DFU to further assess the role of medications and home health care as a potential predictor of 90-day hospital readmission.

Mortality is not increased with Diabetes in hospitalised very old adults: a multi-site review

Background

Few data exist regarding hospital outcomes in people with diabetes aged beyond 75 years. This study aimed to explore the association of diabetes with hospital outcome in the very old patient.

Methods

A retrospective review was conducted of all presentations of patients aged 65 years or more admitted to three Sydney teaching hospitals over 6 years (2012–2018), exploring primarily the outcomes of in-hospital mortality, and secondarily the outcomes of length of stay, the development of hospital-acquired adverse events and unplanned re-admission to hospital within 28 days of discharge. Demographic and outcome data, the presence of diabetes and comorbidities were determined from ICD10 coding within the hospital’s electronic medical record. Logistic and negative binomial regression models were used to assess the association of diabetes with outcome.

Results

A total of 139,130 separations (mean age 80 years, range 65 to 107 years; 51% female) were included, with 49% having documented comorbidities and 26.1% a diagnosis of diabetes.

When compared to people without diabetes, diabetes was not associated with increased odds of mortality (OR: 0.89 SE (0.02), p < 0.001). Further, because of a significant interaction with age, diabetes was associated with decreased odds of mortality beyond 80 years of age. While people with diabetes overall had longer lengths of stay (10.2 days SD (13.4) v 9.4 days SD (12.3), p < 0.001), increasing age was associated with shorter lengths of stay in people aged more than 90 years. Diabetes was associated with increased odds of hospital-acquired adverse events (OR: 1.09 SE (0.02), p < 0.001) and but not 28-day re-admission (OR: 0.88 SE (0.18), p = 0.523).

Conclusion

Diabetes has not been shown to have a negative impact on mortality or length of stay in hospitalised very old adults from data derived from hospital administrative records. This may allow a more measured application of diabetic guidelines in the very old hospitalised patient.

Application of standardised effect sizes to hospital discharge outcomes for people with diabetes

BACKGROUND

Patients with diabetes are at an increased risk of readmission and mortality when discharged from hospital. Existing research identifies statistically significant risk factors that are thought to underpin these outcomes. Increasingly, these risk factors are being used to create risk prediction models, and target risk modifying interventions. These risk factors are typically reported in the literature accompanied by unstandardized effect sizes, which makes comparisons difficult. We demonstrate an assessment of variation between standardised effect sizes for such risk factors across care outcomes and patient cohorts. Such an approach will support development of more rigorous risk stratification tools and better targeting of intervention measures.

METHODS

Data was extracted from the electronic health record of a major tertiary referral centre, over a 3-year period, for all patients discharged from hospital with a concurrent diagnosis of diabetes mellitus. Risk factors selected for extraction were pre-specified according to a systematic review of the research literature. Standardised effect sizes were calculated for all statistically significant risk factors, and compared across patient cohorts and both readmission & mortality outcome measures.

RESULTS

Data was extracted for 46,357 distinct admissions patients, creating a large dataset of approximately 10,281,400 data points. The calculation of standardized effect size measures allowed direct comparison. Effect sizes were noted to be larger for mortality compared to readmission, as well as for being larger for surgical and type 1 diabetes cohorts of patients.

CONCLUSIONS

The calculation of standardised effect sizes is an important step in evaluating risk factors for healthcare events. This will improve our understanding of risk and support the development of more effective risk stratification tools to support patients to make better informed decisions at discharge from hospital.

Implementation of machine learning algorithms to create diabetic patient re-admission profiles

BACKGROUND

Machine learning is a branch of Artificial Intelligence that is concerned with the design and development of algorithms, and it enables today's computers to have the property of learning. Machine learning is gradually growing and becoming a critical approach in many domains such as health, education, and business.

METHODS

In this paper, we applied machine learning to the diabetes dataset with the aim of recognizing patterns and combinations of factors that characterizes or explain re-admission among diabetes patients. The classifiers used include Linear Discriminant Analysis, Random Forest, k-Nearest Neighbor, Naïve Bayes, J48 and Support vector machine.

RESULTS

Of the 100,000 cases, 78,363 were diabetic and over 47% were readmitted.Based on the classes that models produced, diabetic patients who are more likely to be readmitted are either women, or Caucasians, or outpatients, or those who undergo less rigorous lab procedures, treatment procedures, or those who receive less medication, and are thus discharged without proper improvements or administration of insulin despite having been tested positive for HbA1c.

CONCLUSION

Diabetic patients who do not undergo vigorous lab assessments, diagnosis, medications are more likely to be readmitted when discharged without improvements and without receiving insulin administration, especially if they are women, Caucasians, or both.

Racial and Ethnic Differences in 30-Day Hospital Readmissions Among US Adults With Diabetes

IMPORTANCE

Differences in readmission rates among racial and ethnic minorities have been reported, but data among people with diabetes are lacking despite the high burden of diabetes and its complications in these populations.

OBJECTIVES

To examine racial/ethnic differences in all-cause readmission among US adults with diabetes and categorize patient- and system-level factors associated with these differences.

DESIGN, SETTING, AND PARTICIPANTS

This retrospective cohort study includes 272 758 adult patients with diabetes, discharged alive from the hospital between January 1, 2009, and December 31, 2014, and stratified by race/ethnicity. An administrative claims data set of commercially insured and Medicare Advantage beneficiaries across the United States was used. Data analysis took place between October 2016 and February 2019.

MAIN OUTCOMES AND MEASURES

Unplanned all-cause readmission within 30 days of discharge and individual-, clinical-, economic-, index hospitalization-, and hospital-level risk factors for readmission.

RESULTS

A total of 467 324 index hospitalizations among 272 758 adults with diabetes (mean [SD] age, 67.7 [12.7]; 143 498 [52.6%] women) were examined. The rates of 30-day all-cause readmission were 10.2% (33 683 of 329 264) among white individuals, 12.2% (11 014 of 89 989) among black individuals, 10.9% (4151 of 38 137) among Hispanic individuals, and 9.9% (980 of 9934) among Asian individuals (P < .001). After adjustment for all factors, only black patients had a higher risk of readmission compared with white patients (odds ratio, 1.05; 95% CI, 1.02-1.08). This increased readmission risk among black patients was sequentially attenuated, but not entirely explained, by other demographic factors, comorbidities, income, reason for index hospitalization, or place of hospitalization. Compared with white patients, both black and Hispanic patients had the highest observed-to-expected (OE) readmission rate ratio when their income was low (annual household income <$40 000 among black patients: OE ratio, 1.11; 95% CI, 1.09-1.14; among Hispanic patients: OE ratio, 1.11; 95% CI, 1.07-1.16) and when they were hospitalized in nonprofit hospitals (black patients: OE ratio, 1.10; 95% CI, 1.08-1.12; among Hispanic patients: OE ratio, 1.08; 95% CI, 1.05-1.12), academic hospitals (black patients: OE ratio, 1.16; 95% CI, 1.13-1.20; Hispanic patients: OE ratio, 1.12; 95% CI, 1.06-1.19), or large hospitals (ie, with ≥400 beds; black patients: OE ratio, 1.11; 95% CI, 1.09-1.14; Hispanic patients: OE ratio, 1.09; 95% CI, 1.04-1.14).

CONCLUSIONS AND RELEVANCE

In this study, black patients with diabetes had a significantly higher risk of readmission than members of other racial/ethnic groups. This increased risk was most pronounced among lower-income patients hospitalized in nonprofit, academic, or large hospitals. These findings reinforce the importance of identifying and addressing the many reasons for persistent racial/ethnic differences in health care quality and outcomes.

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.

An individualized inpatient diabetes education and hospital transition program for poorly controlled hospitalized patients with diabetes

OBJECTIVE

To evaluate predictors of outcomes associated with an inpatient diabetes education and discharge support program for hospitalized patients with poorly controlled diabetes (glycated hemoglobin [HbA1c]>9%).

METHODS

Patients participated in individualized diabetes education conducted by a certified diabetes educator (CDE) that included an exploration of barriers and goal setting during hospitalization with telephone follow-up and communication with primary providers at discharge. Predictors of HbA1c reduction, successful follow-up, and readmission were analyzed.

RESULTS

There were 82 subjects, and 48% were insulin naïve. Patients with type 2 diabetes (T2D, n = 58) had a significant decrease in HbA1c at follow-up (-2.8%, P<.0001), while those with type 1 diabetes (T1D, n = 19) did not (+0.02%, P = .96). However, after adjustment for other factors, only increasing age, higher baseline HbA1c, earlier education, and initiation of basal insulin were significant predictors of reduction in HbA1c. Higher area level income and empowerment and earlier education were significant predictors of outpatient follow-up within 30 days. While 28% were admitted for severe hyperglycemia, only 1 patient was readmitted with severe hyperglycemia. Successful phone contact was 77% and 57% with and without the support of non-CDE assistants respectively, but all outcomes were similar.

CONCLUSION

The study suggests that an individualized inpatient diabetes education and transition program is associated with a significant reduction in HbA1c that is dependent on baseline HbA1c, older age, initiation of insulin, and earlier enrollment. Additional interventions are needed to ensure better continuity of care.

Retrospective study of inpatient diabetes management service, length of stay and 30-day readmission rate of patients with diabetes at a community hospital

Background: Hospitalized patients with diabetes are at risk of complications and longer length of stay (LOS). Inpatient Diabetes Management Services (IDMS) are known to be beneficial; however, their impact on patient care measures in community, non-teaching hospitals, is unknown.

Objectives: To evaluate whether co-managing patients with diabetes by the IDMS team reduces LOS and 30-day readmission rate (30DR).

Methods: This retrospective quality improvement cohort study analyzed LOS and 30DR among patients with diabetes admitted to a community hospital. The IDMS medical team consisted of an endocrinologist, nurse practitioner, and diabetes educator. The comparison group consisted of hospitalized patients with diabetes under standard care of attending physicians (mostly internal medicine-trained hospitalists). The relationship between study groups and outcome variables was assessed using Generalized Estimating Equation models.

Results: 4,654 patients with diabetes (70.8 ± 0.2 years old) were admitted between January 2016 and May 2017. The IDMS team co-managed 18.3% of patients, mostly with higher severity of illness scores (p < 0.0001). Mean LOS in patients co-managed by the IDMS team decreased by 27%. Median LOS decreased over time in the IDMS group (p = 0.046), while no significant decrease was seen in the comparison group. Mean 30DR in patients co-managed by the IDMS decreased by 10.71%. Median 30DR decreased among patients co-managed by the IDMS (p = 0.048).

Conclusions: In a community hospital setting, LOS and 30DR significantly decreased in patients co-managed by a specialized diabetes team. These changes may be translated into considerable cost savings.

DETERMINING INSULIN DOSE AT THE TIME OF DISCHARGE IN A HIGH-RISK POPULATION: IS THERE ROOM FOR IMPROVEMENT?

OBJECTIVE

To evaluate the adequacy of the insulin dose prescribed at hospital discharge in a high-risk population and assess patient characteristics that influence insulin dose requirement in the immediate postdischarge period.

METHODS

This was a retrospective study conducted at Parkland Health System. We included all patients admitted to a medical floor who received an insulin prescription at discharge and had at least one follow-up visit within 6 months of discharge. All data were extracted by a detailed manual review of each electronic medical record.

RESULTS

At the postdischarge follow-up (N = 797, median 33 days from discharge), 60% of patients required an insulin dose adjustment; 47% of the patients required a dose decrease. Significant predictors of a decrease insulin requirement postdischarge included (multiple regression beta coefficient [95% confidence interval]): newly diagnosed diabetes, -12.7 (-17.7, -7.7); ketosis-prone diabetes, -8.4 (-15, -1.8); glycated hemoglobin A1c (HbA1c), <10% (86 mmol/mol) -7.0 (-11.4, -2.6); discharge insulin total daily dose, -5.3 (-9.3, -1.3); and metformin prescription, -4.9 (-8.4, -1.3).

CONCLUSION

An insulin dose adjustment (most commonly a decrease) was necessary shortly after discharge in more than half of our patients. A better model to estimate insulin dose at discharge is needed along with short-term follow-up after discharge for insulin titration. A pre-emptive insulin dose reduction at discharge should be considered for patients with newly diagnosed diabetes, ketosis-prone diabetes, metformin prescription, and those with HbA1c <10% at presentation.

ABBREVIATIONS

DKA = diabetic ketoacidosis; HbA1c = glycated hemoglobin A1c; KPDM = ketosis-prone diabetes mellitus; TDD = total daily dose.

Inpatient care for stiff person syndrome in the United States: a nationwide readmission study

Background

Stiff person syndrome (SPS) is a progressive neurological disorder characterized by axial muscle rigidity and involuntary spasms. Autoimmune and neoplastic diseases are associated with SPS. Our study objectives were to describe inpatient care for SPS in the United States and characterize 30-day readmissions.

Methods

We queried the 2014 Nationwide Readmission Database for hospitalizations where a diagnosis of SPS was recorded. For readmission analyses, we excluded encounters with missing length of stay, hospitalization deaths, and out-of-state and December discharges. National estimates of index hospitalizations and 30-day readmissions were computed using survey weighting methods. Unconditional logistic regression was used to examine associations between demographic, clinical, and hospital characteristics and readmission.

Results

There were 836 patients with a recorded diagnosis of SPS during a 2014 hospitalization. After exclusions, 703 patients remained, 9.4% of which were readmitted within 30 days. Frequent reasons for index hospitalization were SPS (27.8%) and diabetes with complications (5.1%). Similarly, readmissions were predominantly for diabetes complications (24.2%) and SPS. Most readmissions attributed to diabetes complications (87.5%) were to different hospitals. Female sex (OR, 3.29; CI: 1.22–8.87) and routine discharge (OR, 0.26; CI: 0.10–0.64) were associated with readmission, while routine discharge (OR, 0.18; CI: 0.04–0.89) and care at for-profit hospitals (OR, 10.87; CI: 2.03–58.25) were associated with readmission to a different hospital.

Conclusions

Readmissions in SPS may result from disease complications or comorbid conditions. Readmissions to different hospitals may reflect specialty care, gaps in discharge planning, or medical emergencies. Studies are required to determine if readmissions in SPS are preventable.

Overcoming missed opportunities in diabetes management to improve outcomes for hospitalized patients with diabetes

AIMS

The purpose of this study is to assess the impact of hospitalization on 6-12 month medication adjustment and glycemic control.

METHODS

We conducted a retrospective cohort study of hospitalized and non-hospitalized patients with diabetes of an internal medicine residency continuity clinic. Patients had baseline and outcome HbA1c taken 6-12 months apart. Multivariate linear regression was used to model predictors of HbA1c change from baseline to outcome. Multivariate logistic regression was used to model predictors of medication adjustment between baseline and outcome clinic visits.

RESULTS

Hospitalization was not a significant predictor of HbA1c change. Hospitalized patients with baseline HbA1c < 7% were more likely to have therapy adjusted (OR 3.05, p = .004), but this trend did not extend to adjustment in patients with baseline HbA1c ≥ 7% (OR 0.98, p = .249). A significant predictor of medication adjustment was having a specialized Chronic Care Model-based outpatient diabetic planned visit (DPV) (OR 1.63, p = .020). Depression was not a significant predictor for medication therapy change in well-controlled patients with diabetes, but was associated with a lower likelihood for medication adjustment in poorly-controlled patients with diabetes (OR 0.47, p = .004).

DISCUSSION

Our study supports previous research in that hospitalization may be seen as a "missed opportunity" to intensify treatment when indicated. Based on our findings, hospitalized patients may benefit from enhanced focus on outpatient follow-up. A next step for research is to assess efficacy of scheduling a DPV proximate to discharge for HbA1c reduction when diabetes is poorly controlled.

Correction to: Hospital Readmission of Patients with Diabetes

This article was originally published with errors that were introduced during the editing process. The corrected version of this article appears below.

Practical Recommendations for Transitioning Patients with Type 2 Diabetes from Hospital to Home

PURPOSE OF REVIEW

The purpose of this review is to provide practical evidence-based recommendations for transitioning hospitalized patients with type 2 diabetes (T2DM) to home.

RECENT FINDINGS

Hospitalized patients who have newly diagnosed or poorly controlled T2DM require initiation or intensification of their outpatient diabetes regimen. Potential barriers to medication access and continuity of care should be identified early in the hospitalization. Throughout hospitalization, patients should receive diabetes education focused on basic survival skills and tailored to learning needs. Patients should leave the hospital with personalized discharge instructions that include a list of all medications and follow-up appointments with both the outpatient diabetes provider and a diabetes educator whenever possible. An approach to transitioning patients with T2DM from hospital to home that focuses on optimizing the patient's discharge diabetes regimen, anticipating patients' needs during the immediate post-discharge period, providing survival skills education, and ensuring continuation of diabetes care and education following hospital discharge has the potential to improve glycemic control and reduce emergency department visits and hospital readmissions.

The effect of diabetes case management and Diabetes Resource Nurse program on readmissions of patients with diabetes mellitus

AIMS

Patients with diabetes have higher readmission rates than those without diabetes, yet limited data on efforts to reduce their readmissions are available. We describe a novel model of inpatient diabetes care, expanding the role of diabetes educators to include case management, and establishment of a Diabetes Resource Nurse program, aimed at increasing the knowledge of staff nurses, and evaluate the impact of this program on readmission rates.

METHODS

We performed retrospective analysis of 30-day readmission rates of patients with diabetes before (July 2010-December 2011), and after (January 2012-June 2013) starting the implementation of this tiered inpatient diabetes care delivery model.

RESULTS

We analyzed 34,472 discharged patient records from the 18-month pre-intervention period, and 32,046 records from the 18-month post-intervention period. The overall 30-day readmission rate for patients with diabetes decreased significantly from 20.1% (pre) to 17.6% (post) intervention (p < 0.0001). Patients seen by diabetes educators had the lowest 30-day readmission rates (∼15% during the whole study), a rate approaching the overall hospital readmission rates in those without diabetes in our institution.

CONCLUSION

The Diabetes Resource Nurse program is effective in decreasing readmission rates. Patients seen by the diabetes educators have the lowest rates of readmission.

Risk factors associated with 30-day readmission and length of stay in patients with type 2 diabetes

AIMS

Patients with type 2 diabetes mellitus (type 2 DM) are at greater risk of poor hospital outcomes. The purpose of this study was to determine the impact of type 2 DM on 30-day hospital readmission and length of stay (LOS).

METHODS

We studied all inpatient admissions in Pennsylvania during 2011 using data from the Pennsylvania Health Care Cost Containment Council. Outcomes included 30-day readmission and inpatient LOS. We estimated the impact of type 2 DM on readmission and LOS, and identified risk factors for readmission and prolonged LOS.

RESULTS

Among inpatient admissions, patients with diabetes were more likely to be readmitted (AOR=1.17, P<0.001) and have longer LOS (0.19days, P<0.001) compared to patients without diabetes. Among those with diabetes, several factors were associated with readmission, including demographics, source of admission, and comorbidities. Patients with diabetes were more likely to be readmitted for infectious complications (9.4% vs. 7.7%), heart failure (6.0% vs. 3.1%), and chest pain/MI (5.5% vs. 3.3%) than patients without diabetes.

CONCLUSIONS

Diabetes is associated with risk of 30-day readmission and LOS, and several patient-specific factors are associated with outcomes for patients with diabetes. Future studies may target risk factors to develop strategies to reduce readmissions and LOS.

Risk of hospitalization in patients with diabetes mellitus who have solid-organ malignancy

Background: To determine the relationship between diabetes mellitus (DM) and hospitalization risk in patients with solid-organ malignancies, hospitalized patients with a new solid-organ malignancy and DM were retrospectively analyzed. Results: The presence of DM conferred a 72% greater chance (odds ratio [OR]: 1.72, 95% CI: 1.46–2.04; p < 0.01) of requiring any hospitalization and increased the chances of having multiple admissions by 84% (OR: 1.84, 95% CI: 1.53–2.21; p < 0.01). Additionally, the presence of DM increased the duration of hospital stay by 0.57 days (p < 0.01). Conclusion: The presence of DM in patients with solid-organ malignancies increases the risk of any hospitalization, multiple hospitalizations and length of hospital stay.

A framework for parameter estimation and model selection in kernel deep stacking networks

BACKGROUND AND OBJECTIVES

Kernel deep stacking networks (KDSNs) are a novel method for supervised learning in biomedical research. Belonging to the class of deep learning techniques, KDSNs are based on artificial neural network architectures that involve multiple nonlinear transformations of the input data. Unlike traditional artificial neural networks, KDSNs do not rely on backpropagation algorithms but on an efficient fitting procedure that is based on a series of kernel ridge regression models with closed-form solutions. Although being computationally advantageous, KDSN modeling remains a challenging task, as it requires the specification of a large number of tuning parameters.

METHODS AND MATERIAL

We propose a new data-driven framework for parameter estimation, hyperparameter tuning, and model selection in KDSNs. The proposed methodology is based on a combination of model-based optimization and hill climbing approaches that do not require the pre-specification of any of the KDSN tuning parameters. We demonstrate the performance of KDSNs by analyzing three medical data sets on hospital readmission of diabetes patients, coronary artery disease, and hospital costs.

RESULTS

Our numerical studies show that the run-time of the proposed KDSN methodology is significantly shorter than the respective run-time of grid search strategies for hyperparameter tuning. They also show that KDSN modeling is competitive in terms of prediction accuracy with other state-of-the-art techniques for statistical learning.

CONCLUSIONS

KDSNs are a computationally efficient approximation of backpropagation-based artificial neural network techniques. Application of the proposed methodology results in a fast tuning procedure that generates KDSN fits having a similar prediction accuracy as other techniques in the field of deep learning.

Incidence and long-term outcomes of adult patients with diabetic ketoacidosis admitted to intensive care: A retrospective cohort study

AIMS

Diabetic ketoacidosis is a life-threatening but avoidable complication of diabetes mellitus often managed in intensive care units. The risk of emergency hospital readmission in patients surviving an intensive care unit episode of diabetic ketoacidosis is unknown. We aimed to report the cumulative incidence of emergency hospital readmission and costs in all patients surviving an intensive care unit episode of diabetic ketoacidosis in Scotland.

METHODS

We used a national six-year cohort of survivors of first diabetic ketoacidosis admissions to Scottish intensive care units (1 January 2005-31 December 2010) identified in the Scottish Intensive Care Society Audit Group registry linked to acute hospital and death records (follow-up censored 31 December 2010). Diabetic ketoacidosis-related emergency readmissions were identified using International Classification of Disease-10 codes.

RESULTS

During the study period, 386 patients were admitted to intensive care units in Scotland with diabetic ketoacidosis (admission rate 1.5/100,000 Scottish population). Median age was 44 (IQR 29-56); 51% male; 55% required no organ support on admission. Mortality after intensive care unit admission was 8% at 30 days, 18% at one year, and 35% at five years. A total of 349 patients survived their first intensive care unit diabetic ketoacidosis admission [mean (SD) age 42.5 (18.1) years; 50.4% women; 46.1% required ≥1 organ support]. Following hospital discharge, cumulative incidence of 90-day, one-year, and five-year diabetic ketoacidosis readmission (all-cause readmission) was 13.8% (31.8%), 29.7% (58.9%) and 46.4% (82.6%).

DISCUSSION

Diabetic ketoacidosis in patients requiring intensive care unit admission is associated with high risk of long-term mortality and high hospital costs. An understanding of the precipitating causes of diabetic ketoacidosis in patients admitted to intensive care units may allow patients who are at high risk to be targeted, potentially reducing future morbidity and the substantial burden that diabetic ketoacidosis currently places on the healthcare system.

Risk factors for recurrent admissions with diabetic ketoacidosis: a case-control observational study

AIM

To perform a detailed analysis of patients with recurrent diabetic ketoacidosis admissions in order to establish risk factors for readmission.

METHODS

The medical records of all adults and young people (> 15 years) with Type 1 diabetes admitted to Auckland City Hospital over a 15-year period from 1997 to 2011 with a primary diagnosis of ketoacidosis were analysed. Patients readmitted with ketoacidosis within 5 years of their index admission were identified and compared with patients without ketoacidosis readmission who were matched for age, gender, ethnicity and duration of diabetes.

RESULTS

A total of 268 patients accounted for a total of 412 admissions. In all, 58 patients had more than one admission for diabetic ketoacidosis during this period. Of these, 40 patients readmitted with diabetic ketoacidosis were compared with matched control subjects (n = 40) who had only one admission for diabetic ketoacidosis. The mean ± sd age of the cohort was 31 ± 12 years. The readmission group had more severe diabetic ketoacidosis and poorer glycaemic control. Alcohol abuse was commonly noted in both groups, with insulin dose omission being the main contributor to the development of ketoacidosis. Both groups had high rates of clinic non-attendance. There were no other differences noted between the groups.

CONCLUSION

When patients with recurrent diabetic ketoacidosis were matched for age, duration of diabetes, gender and ethnicity with patients who had only one admission for diabetic ketoacidosis, few differences were noted. This makes designing intervention strategies to reduce readmission with diabetic ketoacidosis difficult.

C-Peptide Level in Fasting Plasma and Pooled Urine Predicts HbA1c after Hospitalization in Patients with Type 2 Diabetes Mellitus

In this study, we investigate how measures of insulin secretion and other clinical information affect long-term glycemic control in patients with type 2 diabetes mellitus. Between October 2012 and June 2014, we monitored 202 diabetes patients who were admitted to the hospital of Asahi Life Foundation for glycemic control, as well as for training and education in diabetes management. We measured glycated hemoglobin (HbA1c) six months after discharge to assess disease management. In univariate analysis, fasting plasma C-peptide immunoreactivity (F-CPR) and pooled urine CPR (U-CPR) were significantly associated with HbA1c, in contrast to ΔCPR and C-peptide index (CPI). This association was strongly independent of most other patient variables. In exploratory factor analysis, five underlying factors, namely insulin resistance, aging, sex differences, insulin secretion, and glycemic control, represented patient characteristics. In particular, insulin secretion and resistance strongly influenced F-CPR, while insulin secretion affected U-CPR. In conclusion, the data indicate that among patients with type 2 diabetes mellitus, F-CPR and U-CPR may predict improved glycemic control six months after hospitalization.

Incidence, Risk Factors, and Causes for Thirty-Day Unplanned Readmissions Following Primary Lower-Extremity Amputation in Patients with Diabetes

BACKGROUND

The Centers for Medicare & Medicaid Services targeted thirty-day readmissions as a quality-of-care measure. Hospitals can be penalized on unplanned readmissions. Given the frequency of amputation in diabetic patients and our changing health-care system, the purpose of this study was to determine the incidence, risk factors, and causes for unplanned thirty-day readmissions following primary lower-extremity amputation in diabetic patients.

METHODS

Patients with a diagnosis of diabetes undergoing primary lower-extremity amputation between 2002 and 2013 were retrospectively identified in a single-center patient database. Chart review determined patient factors including comorbidities, hemoglobin A1c level, amputation level, and demographic characteristics. Patients were divided into groups with and without unplanned readmission within thirty days postoperatively. Univariate and multivariate logistic regression analyses were used to compare cohorts and to identify variables associated with readmission.

RESULTS

Overall, forty-six (10.5%) of 439 diabetic patients undergoing primary lower-extremity amputation had an unplanned thirty-day readmission. The top reason for readmission was a major surgical event requiring reoperation (37.0%), followed by medical events (28.3%) and minor surgical events (28.3%). In the univariate analysis, discharge on antibiotics (p = 0.002), smoking (p = 0.003), chronic kidney disease (p = 0.002), peripheral vascular disease (p = 0.002), and higher Charlson Comorbidity Index (p = 0.001) were each associated with readmission. In the multivariate analysis, diagnosis of gangrene (odds ratio [OR], 2.95 [95% confidence interval (95% CI), 1.37 to 6.35]), discharge on antibiotics (OR, 4.48 [95% CI, 1.71 to 11.74]), smoking (OR, 3.22 [95% CI, 1.40 to 7.36]), chronic kidney disease (OR, 2.82 [95% CI, 1.30 to 6.15]), and peripheral vascular disease (OR, 2.47 [95% CI, 1.08 to 5.67]) were independently associated with readmission.

CONCLUSIONS

Thirty-day readmission rates following primary lower-extremity amputation in patients with diabetes were high at >10%. Both medical and surgical complications, many of which were unavoidable, contributed to readmission. Quality-reporting metrics should include these risk factors to avoid undeservedly penalizing surgeons and hospitals caring for this patient population.

LEVEL OF EVIDENCE

Prognostic Level III. See Instructions for Authors for a complete description of levels of evidence.

Type 2 Diabetes Mellitus Patients' Healthcare Costs Related to Inpatient Hospitalizations: A Retrospective Administrative Claims Database Study

INTRODUCTION

Diabetes-related healthcare costs are increasing in the United States, with inpatient hospitalization the largest component of medical expenditures. The aims of this study were to characterize hospitalized type 2 diabetes mellitus (T2DM) patients, understand the relationship between hospitalization and healthcare costs, and explore treatment modification after inpatient hospitalization.

METHODS

A retrospective cohort analysis of Humana Medicare Advantage and commercial members with T2DM was conducted. T2DM members were identified and assigned to three groups: (1) inpatient hospitalization (IPH) without a 30-day readmit (IPH group); (2) IPH with a 30-day readmission (IPH readmission group); and, (3) matched non-IPH group. Demographics, clinical characteristics, comorbidities and healthcare costs were measured based on enrollment data and claims. Descriptive statistics were used and the relationship between IPH and costs was assessed using generalized linear models.

RESULTS

A total of 15,555 IPH patients, 1757 IPH readmission patients, and 17,312 matched non-IPH patients were included in the study. The IPH readmission group had the highest adjusted mean all-cause total costs ($76,806), followed by the IPH group ($42,011), and the non-IPH group ($9624). A similar trend was observed for adjusted all-cause mean medical and pharmacy costs. DM-related total healthcare costs were highest for the IPH readmission group ($13,714), followed by the IPH group ($7477), and non-IPH group ($1620). While overall therapy modification (discontinuation, addition, switch) was low, T2DM patients with an IPH (with or without a readmission) had greater rates of therapy modification relative to the non-IPH patients.

CONCLUSION

Adjusted all-cause and DM-related total costs were greatest for IPH readmission patients. Rates of treatment modification within 10 days of discharge after IPH were generally low. Identifying T2DM patients at high risk of readmission and employing methods to decrease that risk during the index hospitalization could have a significant impact on health system costs.

FUNDING

Novo Nordisk.

Hospital readmission of patients with diabetes

Hospital readmission is a high-priority health care quality measure and target for cost reduction. Despite broad interest in readmission, relatively little research has focused on patients with diabetes. The burden of diabetes among hospitalized patients, however, is substantial, growing, and costly, and readmissions contribute a significant portion of this burden. Reducing readmission rates of diabetic patients has the potential to greatly reduce health care costs while simultaneously improving care. Risk factors for readmission in this population include lower socioeconomic status, racial/ethnic minority, comorbidity burden, public insurance, emergent or urgent admission, and a history of recent prior hospitalization. Hospitalized patients with diabetes may be at higher risk of readmission than those without diabetes. Potential ways to reduce readmission risk are inpatient education, specialty care, better discharge instructions, coordination of care, and post-discharge support. More studies are needed to test the effect of these interventions on the readmission rates of patients with diabetes.

Effectiveness of an educational intervention on the management of type 2 diabetic patients hospitalized in Internal Medicine: results from the FADOI-DIAMOND study

Appropriate management of hyperglycemia is crucial for patients with type 2 diabetes. Aim of the FADOI-DIAMOND study was to evaluate real-world management of type 2 diabetic patients hospitalized in Internal Medicine wards (IMW) and the effects of a standardized educational intervention for IMW staff. DIAMOND has been carried out in 53 Italian IMW, with two cross-sectional surveys interspersed with an educational program (PRE phase and POST phase). In PRE phase, each center reviewed the charts of the last 30 hospitalized patients with known type 2 diabetes. An educational program was conducted in each center by means of the "outreach visit," a face-to-face meeting between IMW staff and a trained external expert. Six months after, each center repeated the data collection (POST phase), specular to the PRE. A total of 3,167 patients were enrolled (1,588 PRE and 1,579 POST). From PRE phase to POST, patients with registered anthropometric data (54.1 vs. 74.9 %, p < 0.001) and in-hospital/recent measurement of glycated hemoglobin (48.2 vs. 61.4 %, p < 0.005) increased significantly. After educational program, more patients received insulin during hospitalization (68.3 vs. 63.6 %, p = 0.005). A more relevant variation in glycemia during hospitalization was observed in POST phase than PRE (-22.2 vs. -15.5 mg/dL, p < 0.001), without differences as for occurrence of hypoglycemia (12.3 vs. 11.9 %). A one-shot educational intervention led to persistent improvement in the management of hospitalized patients with type 2 diabetes and to significant better glycemic control. Further studies might evaluate the effectiveness of a more aggressive educational program, on both management and outcomes.