Socioeconomic deprivation results in high rates of diabetic ketoacidosis at type 1 diabetes diagnosis in England: A multicentre observational study

Abstract

The global incidence of type 1 diabetes continues to rise, yet a considerable proportion of children continue to present in diabetic ketoacidosis (DKA) at diagnosis despite improved recognition, education, and management strategies.^{1, 2} DKA remains a significant contributor to morbidity, long-term glycaemic instability, and elevated healthcare costs.^{3, 4}

Although international estimates of DKA incidence at diagnosis vary considerably, ranging from 13% to 80% in those under 20 years,² the most recent UK data from the National Paediatric Diabetes Audit (NPDA) reported a 23.3% incidence of DKA at type 1 diabetes diagnosis in 2023.⁵ The reasons for this geographical variation are multifactorial and include delayed diagnosis due to non-specific symptoms, particularly in younger children, as well as disparities related to socioeconomic deprivation and ethnicity.^1-3 However, UK-based data exploring these associations in large, diverse cohorts are limited.

We conducted a retrospective, multicentre observational study to investigate the sociodemographic factors associated with DKA at the time of type 1 diabetes diagnosis in children and adolescents. Data were collected from nine paediatric diabetes units across five English regions: West Midlands, East Midlands, Yorkshire and Humber, London and the South East, and East of England. All individuals younger than 18 years diagnosed with stage 3 type 1 diabetes between January 2023 and August 2024 were included. Key variables included ethnicity (self-identified into 5 NHS-defined categories), socioeconomic status (determined using the 2019 Index of Multiple Deprivation [IMD] quintiles based on postcode), and interpreter requirement.

DKA severity was defined using the British Society for Paediatric Endocrinology and Diabetes (BSPED) criteria.⁶ Of 539 eligible patients, 499 had complete data for full analysis (see Table 1). Overall, 45.5% presented in DKA, a rate substantially higher than national figures. The cohort was notably diverse (38.3% non-white) and socioeconomically deprived (54.9% from IMD quintiles 1 and 2). There was regional variation in deprivation (lowest seen in the East of England: 2.6% in IMD1 vs. the West Midlands: 51.2%).

This study is the largest contemporary analysis of DKA presentation at type 1 diabetes diagnosis in England, which identified socioeconomic deprivation as the most prominent risk factor for this acute presentation. While ethnicity, language, and family history did not significantly influence outcomes in our cohort, the persistent high rates of DKA, particularly among deprived communities, point to a broader public health issue. Efforts to reduce DKA incidence must focus on improving public and professional awareness, especially in high-risk populations.

We advocate for long-term, multi-stage awareness campaigns targeting deprived areas, alongside consideration of targeted islet autoantibody screening to support earlier diagnosis and education to prevent DKA. Addressing socioeconomic inequalities and improving health literacy are essential steps to reverse the growing trend of DKA at type 1 diabetes diagnosis in the UK.

No funding has been received for this work. RPD is supported by the NIHR (Ref NIHR304587).

RPD, MLM, FC, and RHW have received honoraria from Sanofi (participation in the advisory board for teplizumab). RHW has received consultancy fees from Sanofi (review of HCP and patient materials for teplizumab), conference attendance support from Sanofi, and speaker fees from Insulet; TR has received honoraria (participation in advisory boards in screening for type 1 diabetes) and lecture fees from Sanofi. FC has received speaker fees from Insulet. Other authors declared no conflict of interest.

Institutional review board approval for clinical data review was obtained from Birmingham Women's and Children's (BWC) NHS Foundation Trust (reference: CARMS-31593). All data collected from other sites were part of anonymised National Paediatric Audit data collection and submission, and therefore, specific additional ethical approval was not required.