Article Text

  1. T. M. Bird1,
  2. J. M. Tilford1,
  3. M. A. Cleves1,2,
  4. C. A. Hobbs1,2,
  5. J. M. Robbins1,2
  1. 1Department of Pediatrics
  2. 2Department of Arkansas Center for Birth Defects Research and Prevention, University of Arkansas for Medical Sciences, Arkansas Children's Hospital


Background Birth defect surveillance in the U.S. is a state-based initiative. State surveillance systems vary greatly in intensity of case ascertainment, inclusion of prenatally diagnosed cases, and age of eligibility of cases. Eleven states have no surveillance system at all. For these reasons, it is difficult to generate national rates of birth defects and to compare rates across states.

Purpose To determine the value of a national hospital discharge database as a complement to state birth defect surveillance systems.

Methods The 2000 Kids' Inpatient Database (KID) is a nationally representative, pediatric specific database which includes an 80% sample of all newborn hospital discharge records from every short-term, community, and pediatric specialty hospital in each of 27 participating states. State surveillance systems are classified by intensity of surveillance as active (on-site ascertainment of cases by abstractors), passive (reliance on hospital discharge reports), or mixed. Rates of 44 ICD-9 codable birth defects reported by the state surveillance systems are compared to state-specific rates generated by the KID. Also, rates reported by state systems are compared to rates generated by the KID after stratification by intensity of state-based surveillance. Rate ratios (RR) were calculated by dividing the rate of birth defects calculated from the KID by the rate of birth defects observed by the state surveillance systems.

Results Overall RR across birth defects and surveillance systems was 1.04. For active systems the RR was 0.92, for mixed systems the RR was 1.02, and for passive systems the RR was 1.18. Eight of 44 birth defects had a RR of less than 0.75. Many of these eight disorders commonly result in fetal death or do not present until after 10 days of life. Ten of the 44 birth defects have RR above 1.25. Many of these defects are either overdiagnosed at birth or minor defects not captured by many surveillance systems. Of the remaining 25 birth defects with RR between 0.76 and 1.24, 12 have RR between 0.90 and 1.10.

Conclusions While the KID lacks confirmatory diagnostic information, it captures as many birth defects as mixed surveillance systems and more than passive systems. The KID may be the only available source of defect rates for some states. In contrast to the variability of state-based systems, the KID employs a single methodology to derive national birth defect rates and allows direct state to state rate comparisons.

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.