1/11/24 JAMA Diabetic ketoacidosis (DKA) and hyperosmolar hyperglycemic state are 2 forms of hyperglycemic crisis presenting as acute complications of DIABETES. These conditions represent a substantial source of morbidity and medical expenditure in the US, accounting for approximately 231 000 hospitalizations in 2019.1 Prior research reported a declining mortality rate in these conditions from 1985 to 2002.2 Given trends of worsening diabetes control and increasing diabetes-related complications beginning in the early 2010s, patterns of mortality due to hyperglycemic crises may have evolved over time.3,4 This study analyzed trends in US mortality attributed to both DKA and hyperosmolar hyperglycemic state from 1999 to 2022.
Methods
We examined CDC WONDER for multiple cause-of-death International Statistical Classification of Diseases and Related Health Problems, Tenth Revision (ICD-10) codes on death certificates associated with hyperglycemic crises from 1999 to 2022.5 Data from 2022 are provisional. The ICD-10 codes aligned with prior research.2 We examined mortality overall and by age group. We also used the Neighborhood Atlas Area Deprivation Index (ADI), a validated tool constructed from American Community Survey data, to rank all US counties by disadvantage and created 3 tertiles composed of equal numbers of counties for analysis. We selected age group and ADI as subgroup analyses given the potential role of these characteristics in accounting for changing trends in diabetes control.3 Overall and ADI analyses were age-adjusted using the direct method. Age-stratified analyses examined crude mortality within age groups.5
Trends from 1999 to 2019 were characterized using Joinpoint Regression Program, version 5.0.2 (National Cancer Institute). Changes in 2020, 2021, and 2022 were examined descriptively because of the COVID-19 pandemic. This program uses Monte Carlo Permutation, seeking to fit trends to the smallest number of statistically significant joinpoints. Stata, version 14.0 (StataCorp), was used for logistic regression. We defined significance as 2-sided P < .05. This study was determined to not constitute human participant research by the Johns Hopkins Medicine Institutional Review Board. Additional details including multiple cause-of-death certificate methods are provided in Supplement 1.
Results
Between 1999 and 2019, mortality due to hyperglycemic crises increased overall from 3306 deaths and an age-adjusted mortality rate (AAMR) of 1.211 per 100 000 population (95% CI, 1.170-1.253) in 1999 to 6051 deaths and an AAMR of 1.636 (95% CI, 1.594-1.679) in 2019 (Figure, A). Joinpoint analysis found decreasing AAMRs from 2001 to 2008 with annual percentage change (APC) of −2.8% (95% CI, −4.1% to −1.6%; P = .001), followed by increases from 2008 to 2015 (APC, 3.1%; 95% CI, 1.8%-4.3%; P < .001) and 2015 to 2019 (APC, 6.1%; 95% CI, 3.9%-8.3%; P < .001) (Table).
Age-specific mortality rates (ASMR) were highest in adults aged 65 years or older (4.736 per 100 000 in 2019; 95% CI, 4.552-4.919) and adults aged 45 to 64 years (2.726 in 2019; 95% CI, 2.613-2.838) (Figure, B). The ASMRs in adults aged 65 years or older decreased from 2001 to 2006 and stabilized from 2006 to 2014, followed by an increase from 2014 to 2019 (APC, 6.0%; 95% CI, 3.3%-8.7%; P < .001). High ADI counties experienced decreasing AAMRs from 1999 to 2010, with increasing AAMRs from 2010 to 2019 (APC, 5.6%; 95% CI, 4.3%-7.0%; P < .001) (Table). From logistic regression, the AAMR in high ADI counties (2.214 per 100 000; 95% CI, 2.025–2.402) was greater than low ADI counties (1.461; 95% CI, 1.414-1.509; P < .001) in 2019 (Figure, C).
Mortality increased following onset of the COVID-19 pandemic in 2020 and 2021 overall and among all subgroups and decreased in 2022. Mortality peaked in 2021 with 10 230 deaths (AAMR, 2.710 deaths per 100 000; 95% CI, 2.656-2.764) overall and decreased to 8707 deaths in 2022 (AAMR, 2.307; 95% CI, 2.258-2.357).
Discussion
Mortality due to DKA and hyperosmolar hyperglycemic state increased from 1999 to 2019 with disparities by age and area deprivation. The etiology of this upward trend in mortality is likely multifactorial, and interpretation of 2020 to 2022 data is complicated by the COVID-19 pandemic. Inadequate preventive care, disproportionately affecting disadvantaged populations and exacerbated by the pandemic, likely contributes to these findings. Recent practice trends seeking to avoid hypoglycemia by relaxing glycemic targets may have led to an increased risk of hyperglycemia.3,4 Sodium-glucose cotransporter 2 inhibitors, first approved in 2013, are associated with increased risk of DKA and may also play a role in these results.6
This analysis is limited by a reliance on death certificates, which may be miscoded. These findings suggest a need for further investigation to understand the causes behind increasing deaths due to hyperglycemic crisis.
Section Editors: Jody W. Zylke, MD, Deputy Editor; Karen Lasser, MD, and Kristin Walter, MD, Senior Editors.
Article Information
Accepted for Publication: November 30, 2023.
Published Online: January 11, 2024. doi:10.1001/jama.2023.26174
Corresponding Author: Matthew A. Crane, BS, Johns Hopkins University School of Medicine, Edward D. Miller Research Building, 733 N Broadway, Ste 137, Baltimore, MD 21205 (crane@jhu.edu).
Matthew A. Crane, BS1; Audrey Lam, BS2; Emil Ekanayake, BS2; et alYazan Y. Alshawkani, MD1; Colleen Christmas, MD3; Alison Gemmill, PhD4; John A. Romley, PhD2
Author Contributions: Mr Crane had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Crane, Lam, Christmas.
Acquisition, analysis, or interpretation of data: Crane, Lam, Ekanayake, Alshawkani, Gemmill, Romley.
Drafting of the manuscript: Crane, Ekanayake.
Critical review of the manuscript for important intellectual content: All authors.
Statistical analysis: Crane, Romley.
Administrative, technical, or material support: Lam, Ekanayake.
Supervision: Alshawkani, Christmas.
Conflict of Interest Disclosures: None reported.
Disclaimer: This article reflects the views of the authors and should not be construed to represent the views or policies of the US Food and Drug Administration, the Department of Health and Human Services, or the US government.
Data Sharing Statement: See Supplement 2.
References
1.
Centers for Disease Control and Prevention. National diabetes statistics report. Updated November 14, 2023. Accessed November 16, 2023. https://www.cdc.gov/diabetes/data/statistics-report/index.html
2.
Wang J, Williams DE, Narayan KM, Geiss LS. Declining death rates from hyperglycemic crisis among adults with diabetes, US, 1985-2002. Diabetes Care. 2006;29(9):2018-2022. doi:10.2337/dc06-0311PubMedGoogle ScholarCrossref
3.
Gregg EW, Hora I, Benoit SR. Resurgence in diabetes-related complications. JAMA. 2019;321(19):1867-1868. doi:10.1001/jama.2019.3471
ArticlePubMedGoogle ScholarCrossref
4.
Fang M, Wang D, Coresh J, Selvin E. Trends in diabetes treatment and control in US adults, 1999-2018. N Engl J Med. 2021;384(23):2219-2228. doi:10.1056/NEJMsa2032271PubMedGoogle ScholarCrossref
5.
National Center for Health Statistics. Mortality data on CDC WONDER. Centers for Disease Control and Prevention. Updated September 8, 2023. Accessed October 23, 2023. https://wonder.cdc.gov/mcd.html
6.
Fralick M, Schneeweiss S, Patorno E. Risk of diabetic ketoacidosis after initiation of an SGLT2 inhibitor. N Engl J Med. 2017;376(23):2300-2302. doi:10.1056/NEJMc1701990PubMedGoogle ScholarCrossref