Article Text

  1. A. K. Maniatis1,2,
  2. S. H. Goehrig1,
  3. A. Rewers1,
  4. P. Walravens1,
  5. G. J. Klingensmith1
  1. 1The Barbara Davis Center for Childhood Diabetes
  2. 2The Children's Hospital of Denver, University of Colorado Health Sciences Center, Denver
  3. 3University of Maryland


Objective This study analyzed pediatric subjects at the time of initial diagnosis of type 1 diabetes (T1DM) with two objectives: 1) to determine the incidence and severity of diabetic ketoacidosis (DKA), and 2) to stratify these subjects according to insurance status.

Methods The subject population included all children less than 18 years who presented with new onset T1DM from January 2002 to December 2003 and were subsequently followed at the Barbara Davis Center. The insurance status was collected on all subjects, and the initial pH was collected on subjects who presented to an emergency department and/or were admitted to the hospital.

Results 359 patients presented with new onset T1DM. Forty-three (12.0%) of these children had no insurance. One hundred two (28.4%) subjects presented with DKA. When stratified by insurance status, 26 (60.5%) of the 43 uninsured subjects presented with DKA compared to only 76 (24.1%) of the 316 insured subjects, p≤0.001, OR=1.5. Further stratification based on pH severity revealed that uninsured subjects tended to present with more severe DKA than subjects with insurance, p=0.09 (borderline significant chi-square test for trend). The proportions among the most life-threatening DKA (most severe pH ≤ 6.90) included 7 (26.9%) of 26 uninsured subjects compared to 7 (9.2%) of 76 insured subjects, p=0.04.

Conclusion At the time of initial diagnosis, patients without insurance were more likely to present with DKA than subjects with insurance. Furthermore, when the uninsured subjects presented with DKA, the condition tended to be more severe and life-threatening, as indicated by a lower initial pH. A potential explanation is that subjects without insurance may delay seeking timely medical care and therefore present more critically ill, whereas subjects with insurance may have their T1DM diagnosed earlier.

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.