Birth certificate data substantially misrepresent actual NICU admissions, including among most vulnerable

To assess the validity of NICU admission reporting on birth certificates in a large dataset.

We compared NICU admissions on birth certificates to actual NICU admissions for all potentially viable in-hospital deliveries in Pennsylvania and South Carolina from 2006 to 2018 (n = 2,470,917). We used data from linked birth certificate and hospital discharge records to identify actual NICU admissions and evaluated missed NICU admissions by gestational age, mother's race and ethnicity, and other factors.

38% of infants who were truly admitted to a NICU did not have a NICU admission on their birth certificates. Missed NICU admission occurred across all gestational ages. Birth certificates missed more than half of NICU admissions among term infants (37+ weeks), including among those with conditions considered to be severe.

NICU admission reporting on birth certificates grossly underestimates utilization of NICU care and can have serious implications for research, resource allocation, and policy.