Incidence of Diabetic Ketoacidosis at Diagnosis of Type 1 Diabetes in Colorado Youth, 1998-2012 | Acid Base, Electrolytes, Fluids | JAMA | JAMA Network
[Skip to Navigation]
Sign In
Figure.  Temporal Trends in Diabetic Ketoacidosis (DKA) at Time of Type 1 Diabetes Diagnosis in Colorado Youth by Type of Health Insurance, 2007-2012
Temporal Trends in Diabetic Ketoacidosis (DKA) at Time of Type 1 Diabetes Diagnosis in Colorado Youth by Type of Health Insurance, 2007-2012
Table.  Multiple Logistic Regression Analysis of Factors Associated With Diabetic Ketoacidosis (DKA) at Time of Type 1 Diabetes (T1D) Diagnosis
Multiple Logistic Regression Analysis of Factors Associated With Diabetic Ketoacidosis (DKA) at Time of Type 1 Diabetes (T1D) Diagnosis
Research Letter
April 21, 2015

Incidence of Diabetic Ketoacidosis at Diagnosis of Type 1 Diabetes in Colorado Youth, 1998-2012

Author Affiliations
  • 1Department of Pediatrics, University of Colorado School of Medicine, Aurora
  • 2Barbara Davis Center for Childhood Diabetes, University of Colorado School of Medicine, Denver
JAMA. 2015;313(15):1570-1572. doi:10.1001/jama.2015.1414

Diabetic ketoacidosis (DKA) at time of type 1 diabetes (T1D) diagnosis is life threatening and has detrimental long-term effects. It may reflect delayed access to health care, lower quality of care, or income inequality.1 In Scandinavian countries, incidence of DKA at diagnosis has decreased below 20%.2

Little is known about long-term trends in the United States. We examined the temporal trends in DKA at T1D diagnosis between 1998 and 2012 in Colorado and factors associated with DKA.


Between 1998 and 2012, youth diagnosed with T1D before age 18 years at any medical facility were included if a Colorado resident and followed up at the Barbara Davis Center for Diabetes in Denver, which serves more than 80% of youth with diabetes in Colorado with no change in referral patterns over time. Standard criteria3 were used to define DKA. Data were extracted from medical records. Information was insufficient to determine DKA status in 2.8% of patients.

The change in incidence over time was analyzed by fitting a linear term for calendar year in logistic regression models. Logistic regression analysis was used to identify factors associated with DKA. The insurance × year interaction was analyzed for a period with the most rapid increase in incidence (2007-2012). Statistical analysis was performed using SAS version 9.2 (SAS Institute Inc). P < .05 was considered significant (2-sided). The University of Colorado multiple institutional review board approved this study and granted waiver of informed consent.


Diabetic ketoacidosis was present in 1339 of 3439 youth (38.9%) at T1D diagnosis. Youth with DKA had a median age of 9.4 years (interquartile range, 5.6-12.6 years), 53.8% were male, and 75.7% were white.

The proportions with DKA increased significantly, especially after 2007 (29.9% in 1998; 35.0% in 2007; 46.2% in 2012; P < .001 for trend). The only characteristic that changed over time was insurance, with those covered by public insurance increasing from 17.1% in 2007 to 37.5% in 2012 (P < .001).

In multiple logistic regression models (Table), calendar year, age, race/ethnicity, insurance status, and family history of T1D were associated with DKA; however, no significant interactions were detected. Younger age (odds ratio [OR], 1.80; 95% CI, 1.41-2.31) and African American race (OR, 1.80; 95% CI, 1.24-2.62) were associated with higher risk, whereas private insurance (OR, 0.37; 95% CI, 0.29-0.49) and history of T1D in a first-degree relative (OR, 0.35; 95% CI, 0.26-0.48) were associated with lower risk. During 2007-2012, private insurance was associated with an increase in DKA by 2.5% per year, whereas public insurance was associated with a decrease of 1.3% per year (P = .02 for interaction) (Figure).


The incidence of DKA in youth at diagnosis of T1D in Colorado between 1998 and 2012 increased by 55% (from 29.9% to 46.2%), suggesting a growing number of youth may experience delays in diagnosis and treatment. Diabetic ketoacidosis is preventable because most youth present with classic symptoms. This incidence is consistent with incidences in countries with poor access to health care and low community and physician awareness of diabetes, and is much higher than incidences reported in Canada or the United Kingdom.4

Some of the factors associated with DKA at diagnosis are potentially modifiable. For example, the association with family history suggests the importance of awareness of diabetic symptoms. However, economic factors are more difficult to modify. Increasing incidence of DKA correlated temporally with an increase in Colorado child poverty prevalence from 10% in 2000 to 18% in 2012.5 The recent increase of DKA incidence among youth with private insurance may be related to proliferation of high-deductible health plans.

The large size and extended period of surveillance are strengths of the study. Limitations include exclusion of approximately 20% of youth with T1D and results from a single state may not be generalizable.

To our knowledge, this is the only report of increasing incidence of DKA in the developed world. Further research on the reasons for the increase and interventions to decrease the incidence are warranted.

Section Editor: Jody W. Zylke, MD, Deputy Editor.
Back to top
Article Information

Corresponding Author: Arleta Rewers, MD, PhD, Department of Pediatrics, University of Colorado School of Medicine, 13123 E 16th Ave, Aurora, CO 80045 (

Author Contributions: Drs A. Rewers and M. Rewers had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: A. Rewers, M. Rewers.

Acquisition, analysis, or interpretation of data: All authors.

Drafting of the manuscript: A. Rewers, M. Rewers.

Critical revision of the manuscript for important intellectual content: All authors.

Statistical analysis: Dong.

Obtained funding: A. Rewers, M. Rewers.

Administrative, technical, or material support: A. Rewers, M. Rewers.

Study supervision: A. Rewers, Slover, Klingensmith, M. Rewers.

Conflict of Interest Disclosures: The authors have completed and submitted the ICMJE Form for Disclosure of Potential Conflicts of Interest. Dr A. Rewers reported receiving personal fees from the Diabetes Technology Society and Roche Diagnostics International. Dr Klingensmith reported serving on a safety monitoring board for Novo Nordisk. No other disclosures were reported.

Funding/Support: This study was supported by grant P30 DK57156 from the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK) and by funding from the Children’s Diabetes Foundation in Denver.

Role of the Funder/Sponsor: The NIDDK and the Children’s Diabetes Foundation in Denver had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

Limenis  E, Shulman  R, Daneman  D.  Is the frequency of ketoacidosis at onset of type 1 diabetes a child health indicator that is related to income inequality?  Diabetes Care. 2012;35(2):e5.PubMedGoogle ScholarCrossref
Hekkala  A, Knip  M, Veijola  R.  Ketoacidosis at diagnosis of type 1 diabetes in children in northern Finland: temporal changes over 20 years.  Diabetes Care. 2007;30(4):861-866.PubMedGoogle ScholarCrossref
Wolfsdorf  J, Craig  ME, Daneman  D,  et al.  Diabetic ketoacidosis in children and adolescents with diabetes.  Pediatr Diabetes. 2009;10(suppl 12):118-133.PubMedGoogle ScholarCrossref
Usher-Smith  JA, Thompson  M, Ercole  A, Walter  FM.  Variation between countries in the frequency of diabetic ketoacidosis at first presentation of type 1 diabetes in children: a systematic review.  Diabetologia. 2012;55(11):2878-2894.PubMedGoogle ScholarCrossref
Annie E. Casey Foundation. Kids count data center. Accessibility verified March 5, 2015.