Pediatric Type 1 Diabetes: Reducing Admission Rates for Diabetes Ketoacidosis

Interventions Across the Translational Research Spectrum: Addressing Disparities Among Racial and Ethnic Minoritized Youth with Type 1 Diabetes

Racial and ethnic minoritized youth with type 1 diabetes (T1D) are at elevated risk for health disparities. Few intervention studies have been conducted for these youth and evidence to support best practices to address their needs is lacking. Existing evidence supports the use of brief trials of diabetes technology with structured support from clinic staff, culturally tailored interventions such as language-congruent clinical care, and use of community health workers as promising directions to improve health outcomes. Clinicians and researchers should work collaboratively with community members to improve the quality of T1D intervention science for racial and ethnic minoritized youth.

Development of an "all-data-on-hand" deep learning model to predict hospitalization for diabetic ketoacidosis (DKA) in youth with type 1 diabetes (T1D)

BACKGROUND

While prior research has identified multiple risk factors for diabetic ketoacidosis (DKA), clinicians continue to lack clinic-ready models to predict dangerous and costly episodes of DKA. We asked whether we could apply deep learning, specifically use of a long short-term (LSTM) model, to accurately predict 180-day risk of DKA-related hospitalization for youth with type 1 diabetes (T1D).

OBJECTIVE

To describe the development of a LSTM model to predict 180-day risk of DKA-related hospitalization for youth with T1D.

METHODS

We used 17 consecutive calendar quarters of clinical data (01/10/2016-03/18/2020) for 1745 youth 8 to 18-years with T1D from a pediatric diabetes clinic network in the Midwestern US. Input data included demographics, discrete clinical observations (lab results, vital signs, anthropometric measures, diagnosis and procedure codes), medications, visit counts by type of encounter, number of historic DKA episodes, number of days since last DKA admission, patient-reported outcomes (answers to clinic intake questions), and data features derived from diabetes- and non-diabetes-related clinical notes via natural language processing (NLP). We trained the model using input data from quarters 1-7 (n=1377), validated using input from quarters 3-9 in a partial out-of-sample cohort (OOS-P; n=1505), and further validated in a full out-of-sample cohort (OOS-F; n=354) with input from quarters 10-15.

RESULTS

DKA admissions occurred at a rate of 5% per 180-days in both OOS cohorts. For the OOS-P and OOS-F cohorts, respectively: median age was 13.7 years (IQR=11.3,15.8) and 13.1 years (10.7,15.5); and HbA1c at enrollment was 8.6% (7.6,9.8) [70 (60,84) mmol/mol] and 8.1% (6.9,9.5) [65 (52,80) mmol/mol]; 14% and 13% had prior DKA admissions (post-T1D-diagnosis); and recall was 0.33 and 0.50 for the top-ranked 5% of youth with T1D. For lists rank-ordered by probability of hospitalization, precision increased from 0.33 to 0.56 to 1.0 for positions 1-80, 1-25, and 1-10 in the OOS-P cohort and from 0.50 to 0.60 to 0.80 for positions 1-18, 1-10, and 1-5 in the OOS-F cohort.

CONCLUSIONS

The proposed LSTM model for predicting 180-day DKA-related hospitalization is valid in the present sample. Future work should evaluate model validity in multiple populations and settings to account for health inequities that may be present in different segments of the population (e.g., racially and/or socioeconomically diverse cohorts). Rank-ordering youth by probability of DKA-related hospitalization will allow clinics to identify the most at-risk youth. The clinical implication of this is that clinics may then create and evaluate novel preventive interventions based on available resources.

Reducing Diabetic Ketoacidosis Readmissions with a Hospital-School-Based Improvement Partnership

Diabetic ketoacidosis (DKA) is the leading cause of morbidity and mortality in pediatric type 1 diabetes mellitus (T1D). Baseline data showed 139 of 182 DKA readmissions (76.4%) were due to missed basal insulin dosing. The team used quality improvement tools to implement a process change around basal insulin. The project utilized insulin degludec and school-based nurses when missed basal insulin was noted as a main driver for readmission. The DKA readmission rate averaged 5.25 per month from January 2017 to April 2019. The rate decreased to 3.64 per month during the intervention from May 2019 to March 2020, a 31% reduction over 11 months. This standardized approach for patients with T1D readmitted with DKA, using a school-based intervention and insulin degludec, reduced the number of DKA readmissions. This method is safe and effective for lowering DKA readmissions due to missed basal insulin in areas with reliable school nursing.

Rising Readmission Rates After Diabetic Ketoacidosis Hospitalization Among Adults With Type 1 Diabetes Throughout a Decade in the United States

Research on longitudinal trends in readmission rates after diabetic ketoacidosis (DKA) is lacking. This retrospective study was aimed at identifying trends in readmissions after hospitalization for DKA, as well as trends in outcomes after readmission, over time among adults with type 1 diabetes in the United States. Findings indicate that the DKA readmission rate increased from 53 to 73 events per 100,000 between 2010 to 2018, and low-income and uninsured patients had higher odds of readmission. There was no significant change in mortality after readmission over time. Improved access to care and affordable management options may play a crucial role in preventing readmissions.

Exploring HbA1c variation between Australian diabetes centres: The impact of centre-level and patient-level factors

Background

Increasing global diabetes incidence has profound implications for health systems and for people living with diabetes. Guidelines have established clinical targets but there may be variation in clinical outcomes including HbA1c, based on location and practice size. Investigating this variation may help identify factors amenable to systemic improvement interventions. The aims of this study were to identify centre-specific and patient-specific factors associated with variation in HbA1c levels and to determine how these associations contribute to variation in performance across diabetes centres.

Methods

This cross-sectional study analysed data for 5,872 people with type 1 (n = 1,729) or type 2 (n = 4,143) diabetes mellitus collected through the Australian National Diabetes Audit (ANDA). A linear mixed-effects model examined centre-level and patient-level factors associated with variation in HbA1c levels.

Results

Mean age was: 43±17 years (type 1), 64±13 (type 2); median disease duration: 18 years (10,29) (type 1), 12 years (6,20) (type 2); female: 52% (type 1), 45% (type 2). For people with type 1 diabetes, volume of patients was associated with increases in HbA1c (p = 0.019). For people with type 2 diabetes, type of centre was associated with reduction in HbA1c (p <0.001), but location and patient volume were not. Associated patient-level factors associated with increases in HbA1c included past hyperglycaemic emergencies (type 1 and type 2, p<0.001) and Aboriginal and Torres Strait Islander status (type 2, p<0.001). Being a non-smoker was associated with reductions in HbA1c (type 1 and type 2, p<0.001).

Conclusions

Centre-level and patient-level factors were associated with variation in HbA1c, but patient-level factors had greater impact. Interventions targeting patient-level factors conducted at a centre level including sick-day management, smoking cessation programs and culturally appropriate diabetes education for and Aboriginal and Torres Strait Islander peoples may be more important for improving glycaemic control than targeting factors related to the Centre itself.

An Automated Risk Index for Diabetic Ketoacidosis in Pediatric Patients With Type 1 Diabetes: The RI-DKA

Identifying patients at high risk for diabetic ketoacidosis (DKA) is crucial for informing efforts at preventive intervention. This study sought to develop and validate an electronic medical record (EMR)-based tool for predicting DKA risk in pediatric patients with type 1 diabetes. Based on analysis of data from 1,864 patients with type 1 diabetes, three factors emerged as significant predictors of DKA: most recent A1C, type of health insurance (public vs. private), and prior DKA. A prediction model was developed based on these factors and tested to identify and categorize patients at low, moderate, and high risk for experiencing DKA within the next year. This work demonstrates that risk for DKA can be predicted using a simple model that can be automatically derived from variables in the EMR.

Equity in Medical Care for People Living With Diabetes

Disparities and inequities exist for individuals with diabetes in marginalized communities of color, especially among people with low socioeconomic status. Although these barriers are apparent, only a few care models have been designed for and examined in racially and ethnically diverse individuals. This article reviews models that have been developed and examined in a variety of different populations and focuses on how to implement elements from these programs in clinical practice. Health equity-promoting ideas and approaches that can be applied throughout the life span (children to seniors) are also included. As diabetes health care providers, researchers, educators, policymakers, and advocates, we must now combine our efforts and focus on historically excluded populations to bridge the gap to essential diabetes care.

Intensive sick day rules to prevent recurrent diabetic ketoacidosis- An intervention that exemplifies health disparities

OBJECTIVE

Diabetic ketoacidosis (DKA) is the leading cause of morbidity and mortality in children with type 1 diabetes (T1D). In established T1D patients, DKA is frequently a result of insulin omission or inadequate insulin administration during illness or stress. Ethnic minorities and patients with lower socioeconomic status are affected disproportionately. We hypothesized that implementation of intensive sick day rules with frequent reinforcement would reduce hospitalizations secondary to DKA in T1D youth irrespective of their demographics.

METHODS

Intensive sick day rules were implemented beginning January 2016. All T1D patients seen in the pediatric endocrinology clinic or hospital between January 1^st 2015 through December 31^st 2017 were included for chart review. Categorical variables were analyzed with Chi-square test. For the continuous variables, t test was used. Episodes of DKA per 100 patients were compared using the trends test over the three-year period. Patients who had DKA in 2015 were analyzed as a subgroup.

RESULTS

The frequency of DKA episodes per 100 patient years for 2015 was 19.1, for 2016 was 15.2 and was 12.4 for 2017. This decrease was statistically significant (p=0.006). The decline was also statistically significant for the subgroup of patients who developed DKA in 2015 and followed longitudinally. The decline was not uniform across all patient groups and DKA episodes remained associated with African- American race, Medicaid insurance status and higher HbA1c throughout the years.

CONCLUSION

Implementation of intensive sick day rules led to a decrease in total number of DKA admissions in our population with T1D youth. However, this intervention did not reduce the health disparity in this population and African-Americans on Medicaid insurance continued to form the disproportionate majority of admissions with DKA. This study highlights the need for further research into interventions that can improve outcomes across racial and socio-economic barriers.

Risk Factors, Trends, and Preventive Measures for 30-Day Unplanned Diabetic Ketoacidosis Readmissions in the Pediatric Population

Background There has been a steady rise in types 1 and 2 diabetes mellitus among the youth in the USA from 2001 to 2017. Diabetic ketoacidosis (DKA) is a common and preventable presentation of both types of diabetes mellitus. According to the Centers for Disease Control and Prevention's (CDC) United States Diabetes Surveillance System, during 2004-2019 an increase in DKA hospitalization rates by 59.4% was noted, with people aged less than 45 years having the highest rates. Readmissions reflect the quality of disease management, which is integrally tied to care coordination and communication with the patient and their families. This study analyzes the trends and risk factors contributing to 30-day unplanned DKA readmissions in the pediatric age group and looks into possible preventive measures to decrease them. Methods A retrospective study was performed using the National Readmission Database (NRD) from January 1, 2017, to December 1, 2017. Pediatric patients aged 18 years and younger with the primary diagnosis of DKA were included using the International Classification of Diseases, Tenth Revision, Clinical Modification (ICD-10-CM) code E10.10. All statistical analysis was performed using IBM SPSS Statistics for Windows, version 1.0.0.1327 (IBM Corp., Armonk, NY, USA). Pearson's chi-square test was used for categorical variables and Mann-Whitney U test was used for continuous variables. To independently determine the predictors of readmission within each clinical variable, multiple logistic regressions with values presented as odds ratios (OR) with 95% confidence intervals (CI) were performed. Results A weighted total of 19,519 DKA-related pediatric index admissions were identified from the 2017 NRD. Of these pediatric patients, 831 (4.3%) had 30-day DKA readmission. The median age of a child for readmission was 16 years with an interquartile range of 0 to 18 years. A sharp rise in 30-day DKA readmissions was noted for ages 16 years and over. Females in the 0-25th percentile median household income category, with Medicaid covered, large metropolitan areas with at least 1 million residents, and metropolitan teaching hospitals were found to have a statistically significant higher percentage of readmissions. The mean length of stay for those who had a DKA readmission was 2.06 days, with a standard deviation of 1.84 days. The mean hospital charges for those who had a DKA readmission were $ 20,339.70. The 30-day DKA readmission odds were seen to be increased for female patients, Medicaid-insured patients, admissions at metropolitan non-teaching hospitals, and children from 0-25th percentile median household income category. Conclusion There has not been much of a change in the trend and risk factors contributing to the 30-day unplanned DKA readmissions over the years despite the steady rise in cases of diabetes mellitus. The length of stay for those who did not get readmitted within 30 days was longer than for those who did. This could reflect more comprehensive care and discharge planning that may have prevented them from readmission. Diabetes mellitus is a chronic disease that demands a team effort from the patient, family, healthcare personnel, insurance companies, and lawmakers. There is scope for a lot of improvement with the way our patients are being managed, and a more holistic approach needs to be devised.

A Multi-Intervention Campaign Lowers Pediatric and Young Adult Diabetic Ketoacidosis Hospitalizations in a Canadian Province

OBJECTIVES

The Newfoundland and Labrador diabetic ketoacidosis Project (NLdkaP) is a multi-intervention, province-wide project aimed at lowering rates of diabetic ketoacidosis (DKA) within the pediatric and young adult populations.

METHODS

The NLdkaP interventions were first selected, developed and implemented. We then conducted a retrospective study of hospitalization data over three 2-year periods: pre-, during and post-NLdkaP. Data included demographic factors, DKA hospitalizations and length of hospital stay.

RESULTS

There were 412 DKA hospitalizations over the study period. Before the NLdkaP, the provincial hospitalization rate of DKA for patients <25 years of age was 55.61 per 100,000. During the NLdkaP, the rate dropped to 38.48 per 100,000 (p<0.001). After the NLdkaP, the rate rose to 54.53 per 100,000 (p<0.001). Hospitalization rates were highest for females (p<0.001) and for those in the 19- to 24-year age group (p<0.001).

CONCLUSIONS

The NLdkaP was associated with decreased rates of DKA hospitalizations, but the rates remained relatively stable in both the pre- and postintervention periods. Although the approach and resources developed in the NLdkaP appear effective, continuous preventive efforts are needed to sustain reductions in DKA hospitalizations.

Risk factors for hospitalization in youth with type 1 diabetes: Development and validation of a multivariable prediction model

OBJECTIVE

To develop a multivariable prediction model to identify patients with type 1 diabetes at increased risk of hospitalization for diabetic ketoacidosis or hyperglycemia with ketosis in the 12 months following assessment.

METHODS

Retrospective review of clinical data from patients with type 1 diabetes less than 17 years old at a large academic children's hospital (5732 patient years, 652 admissions). Data from the previous 12 months were assessed on October 15, 2015, 2016, 2017, and 2018, and used to predict hospitalization in the following 12 months using generalized estimating equations. Variables that were significant predictors of hospitalization in univariate analyses were entered into a multivariable model. 2014 to 2016 data were used as a training dataset, and 2017 to 2019 data for validation. Discrimination of the model was assessed with receiver operator characteristic curves.

RESULTS

Admission in the preceding year, hemoglobin (Hb)A1c, non-commercial insurance, female sex, and non-White race were all individual predictors of hospitalization, but age, duration of diabetes and number of office visits in the preceding year were not. In multivariable analysis with threshold P < .0033, admissions in the previous 12 months, HbA1c, and non-commercial insurance remained as significant predictors. The model identified a subset of ~8% of the patients with a collective 42% risk of hospitalization, thus increased 5-fold compared with the 8% risk of hospitalization in the remaining 93% of patients. Similar results were obtained with the validation dataset.

CONCLUSION

Our multivariable prediction model identified patients at increased risk of admission in the 12 months following assessment.

Etiologies, trends, and predictors of 30-day pediatric readmissions after hospitalizations for diabetic ketoacidosis in the United States

OBJECTIVE

To determine the causes, predictors, and trends of 30-day readmissions following hospitalizations for pediatric diabetic ketoacidosis (DKA) in the United States (US) from 2010 to 2014.

RESEARCH DESIGN AND METHODS

We used International Classification of Diseases, ninth revision, Clinical Modification codes to identify children with DKA aged 2 to 18 years from the National Readmission Database in the US. Patients who had readmission within 30 days after an index admission for DKA were included in the study. We combined similar diagnoses into clinically important categories to determine the cause of readmission. The primary outcome was all-cause 30-day (AC30) readmissions. Categorical and continuous variables were analyzed using chi-square or student's t-test or Wilcoxon rank sum tests respectively. We performed multivariable logistic regression to identify predictors of 30-day readmission.

RESULTS

From 2010 through 2014, a weighted total of 87 815 index DKA-related pediatric hospitalizations were identified of which, 4055 patients (4.6%) had AC30 readmissions and this remained unchanged during the study period. Of all the readmissions, 69% were attributed to DKA. In multivariable regression analysis, the odds of AC30 readmission and 30-day readmission attributed to DKA alone were increased for females, adolescents, patients with depression and psychosis, and discharge against medical advice, while private insurance, the highest income quartile, and admission at teaching hospitals were associated with lower odds of AC30 readmission and 30-day readmission attributed to DKA only.

CONCLUSION

We identified several factors associated with readmission after hospitalization for DKA. Addressing these factors such as depression may help lower readmissions after an admission for DKA.

Health Care Utilization and Spending for Children With Mental Health Conditions in Medicaid

OBJECTIVE

To examine how characteristics vary between children with any mental health (MH) diagnosis who have typical spending and the highest spending; to identify independent predictors of highest spending; and to examine drivers of spending groups.

METHODS

This retrospective analysis utilized 2016 Medicaid claims from 11 states and included 775,945 children ages 3 to 17 years with any MH diagnosis and at least 11 months of continuous coverage. We compared demographic characteristics and Medicaid expenditures based on total health care spending: the top 1% (highest-spending) and remaining 99% (typical-spending). We used chi-squared tests to compare the 2 groups and adjusted logistic regression to identify independent predictors of being in the top 1% highest-spending group.

RESULTS

Children with MH conditions accounted for 55% of Medicaid spending among 3- to 17-year olds. Patients in the highest-spending group were more likely to be older, have multiple MH conditions, and have complex chronic physical health conditions (P <.001). The highest-spending group had $164,003 per-member-per-year (PMPY) in total health care spending, compared to $6097 PMPY in the typical-spending group. Ambulatory MH services contributed the largest proportion (40%) of expenditures ($2455 PMPY) in the typical-spending group; general health hospitalizations contributed the largest proportion (36%) of expenditures ($58,363 PMPY) in the highest-spending group.

CONCLUSIONS

Among children with MH conditions, mental and physical health comorbidities were common and spending for general health care outpaced spending for MH care. Future research and quality initiatives should focus on integrating MH and physical health care services and investigate whether current spending on MH services supports high-quality MH care.

Diabetic ketoacidosis

Diabetic ketoacidosis (DKA) is the most common acute hyperglycaemic emergency in people with diabetes mellitus. A diagnosis of DKA is confirmed when all of the three criteria are present - 'D', either elevated blood glucose levels or a family history of diabetes mellitus; 'K', the presence of high urinary or blood ketoacids; and 'A', a high anion gap metabolic acidosis. Early diagnosis and management are paramount to improve patient outcomes. The mainstays of treatment include restoration of circulating volume, insulin therapy, electrolyte replacement and treatment of any underlying precipitating event. Without optimal treatment, DKA remains a condition with appreciable, although largely preventable, morbidity and mortality. In this Primer, we discuss the epidemiology, pathogenesis, risk factors and diagnosis of DKA and provide practical recommendations for the management of DKA in adults and children.

Blood Glucose Monitoring Before and After Type 1 Diabetes Clinic Visits

Objective

To determine patterns of blood glucose monitoring in children and adolescents with type 1 diabetes (T1D) before and after routine T1D clinic visits.

Methods

Blood glucose monitoring data were downloaded at four consecutive routine clinic visits from children and adolescents aged 5-18 years. Linear mixed models were used to analyze patterns of blood glucose monitoring in patients who had at least 28 days of data stored in their blood glucose monitors.

Results

In general, the frequency of blood glucose monitoring decreased across visits, and younger children engaged in more frequent blood glucose monitoring. Blood glucose monitoring increased before the T1D clinic visits in younger children, but not in adolescents. It declined after the visit regardless of age.

Conclusions

Members of the T1D care team need to consider that a T1D clinic visit may prompt an increase in blood glucose monitoring when making treatment changes and recommendations. Tailored interventions are needed to maintain that higher level of adherence across time.

A Quality Improvement Initiative to Reduce Hospitalizations for Low-risk Diabetic Ketoacidosis

BACKGROUND AND OBJECTIVES

Children with established type 1 diabetes (T1D) who present to the emergency department (ED) with mild diabetic ketoacidosis (DKA) are often hospitalized, although outpatient management may be appropriate. Our aim was to reduce hospitalization rates for children with established T1D presenting to our ED with mild DKA who were considered low risk for progression of illness.

METHODS

We conducted a quality improvement initiative between January 1, 2012, and December 31, 2018 among children and young adults ≤21 years of age with established T1D presenting to our tertiary care ED with low-risk DKA. Children transferred to our institution were excluded. DKA severity was classified as low, medium, or high risk on the basis of laboratory and clinical criteria. Our quality improvement initiative consisted of development and implementation of an evidence-based treatment guideline after review by a multidisciplinary team. Our primary outcome was hospitalization rate, and our balancing measure was 3-day ED revisits. Statistical process control methods were used to evaluate outcome changes.

RESULTS

We identified 165 patients presenting with low-risk DKA. The baseline preimplementation hospitalization rate was 74% (95% confidence interval 64%-82%), and after implementation, this decreased to 55% (95% confidence interval 42%-67%) (-19%; P = .011). The postimplementation hospitalization rate revealed special cause variation. One patient in the postimplementation period returned to the ED within 3 days but did not have DKA and was not hospitalized.

CONCLUSIONS

Hospitalization rates for children and young adults presenting to the ED with low-risk DKA can be safely reduced without an increase in ED revisits.

Recurrent diabetic ketoacidosis and a brief history of brittle diabetes research: contemporary and past evidence in diabetic ketoacidosis research including mortality, mental health and prevention

Pharmacological, technological and educational approaches have advanced the treatment of Type 1 diabetes in the last four decades and yet diabetic ketoacidosis (DKA) continues to be a leading cause of admission in Type 1 diabetes. This article begins by reviewing the contemporary epidemiological evidence in DKA. It highlights a rise in DKA episodes in the last two decades, with DKA continuing to be the leading cause of death in young people with Type 1 diabetes, and that DKA episodes are a marker for subsequent all-cause mortality. It also summarizes the limited evidence base for DKA prevention and associations with psychopathology. To emphasize the importance of this group with high-risk Type 1 diabetes and the degree to which they have been overlooked in the past two decades, the article summarizes the research literature of recurrent DKA during 1976-1991 when it was extensively investigated as part of the phenomenon of 'brittle diabetes'. This period saw numerous basic science studies investigating the pathophysiology of recurrent DKA. Subsequently, research centres published their experiences of brittle diabetes research participants manipulating their treatment under research conditions. Unfortunately, the driver for this behaviour and whether it was indicative of other people with ketoacidosis was not pursued. In summary, we suggest there has been a stasis in the approach to recurrent DKA prevention, which is likely linked to historical cases of mass sabotage of brittle diabetes research. Further investigation is required to clarify possible psychological characteristics that increase the risk of DKA and thereby targets for DKA prevention.