588 - Sudden Unexpected Infant Death Diagnostic Codes in the First Hour of Life

Poster Number: 588
Publication Number: 588.241

Thomas Hegyi, Rutgers, Robert Wood Johnson Medical School, Englewood, NJ, United States; Barbara M. Ostfeld, Rutgers Robert Wood Johnson Medical School, New Brunswick, NJ, United States

Background: Following a decline in Sudden Unexpected Infant Death (SUID) in the 1990s, there has been little progress in the rate reduction. SUID is defined as deaths at less than one year of age by the ICD10 codes R95 (SIDS), R99 (ill-defined or unknown cause), or W75 (accidental suffocation and strangulation). The US SUID rate has remained stable at 0.94 and 0.91 per 1,000 live births for the year 2000 vs. 2018 (CDC WONDER). Contributions to the national rate stagnation have included the adverse social and health determinants, implicit bias, and unmet challenges to compliance with risk reduction guidelines. A less explored area is whether there are misapplications in the use of SUID diagnostic codes, particularly for deaths within the first hour of life.

Objective: We sought to examine the use of SUID coding as it is applied to deaths of infants in the first hour of life across gestational age (GA) groups. We hypothesized that misapplication in coding may spuriously elevate the SUID rate.

Design/Methods: Using the online linked birth/infant death files of CDC WONDER for the period from 2007-2018, we identified cases of SUID in the first hour of life, defined as those receiving the ICD10 codes R95, R99 and W75, and determined the proportion of infants < 23 weeks GA; 23-27 weeks (extremely preterm); 28-31 weeks (very preterm); 32-33 weeks (moderately preterm); 34-36 weeks (late preterm); >37 weeks (term).

Results: SUID in the first hour of life is rare, consisting of 0.6% of all SUID cases; 94.6% received an ICD10 code of R99, and 5.4% were coded as R95; 89.1% were hospital births. Of the 279 early deaths classified as SUID, gestational ages were available for 268 cases. Of these, 119 (44.4%) were cases of infants under 23 weeks GA. The gestational age distribution is shown in the Figure. The contribution to SUID deaths declined with increasing gestational age but rose in late-preterm and term infants.Conclusion(s): Most SUID cases in the first hour of life were infants less than 23 weeks GA and were classified as R99. We speculate that alternative coding, such as extreme prematurity, may better have characterized these deaths, which ranged from 17 to 22 weeks GA. Unlike deaths occurring after discharge and referred to the medical examiner system, deaths within the first hour are evaluated within the hospital. We therefore further speculate that there is a need to assess how causality is considered in those instances. Similarly, a more specific designation of death rather than SUID may be warranted for late-preterm and term infants dying within the first hour. Although death in the first hour represents only a small portion of SUID cases, the potential misapplication of coding may contribute to a spuriously sustained rate of SUID over decades.
Gestational Age Distribution of SUID in the First Hour of Life