Lead/Presenter: Jonathan Shaw, COIN - Palo Alto
All Authors: Shaw JG (VA Palo Alto)
Schmitt SK (VA Palo Alto)
Frayne SM (VA Palo Alto)
Shaw KA (Stanford University)
Danielsen B (Health Information Solutions)
Kimerling R (VA Palo Alto)
Joyce VR (VA Palo Alto)
Asch SM (VA Palo Alto)
Phibbs CS (VA Palo Alto)
Objectives:
Since 2000 VA has served as maternity care payer for a small but growing fraction of Veterans. Veterans in VA often have higher medical and mental health disease burden. We examined whether women using VA as the payer for maternity care have higher obstetric risk, and also explored Veteran vs non-Veteran obstetric risk profiles.
Methods:
We identified all in-hospital deliveries by VA-enrolled Veterans using linked California birth data (2000-2012); payer type came from VA purchased care and insurance codes from California data. We compared the most common pregnancy outcomes (cesarean, preterm birth, gestational diabetes, preeclampsia, and newborn NICU admission) between two groups: (1) enrolled Veterans for whom the VA was the payer vs. enrolled Veterans using non-VA payers, and 2) enrolled Veterans (regardless of payer) vs. all other women delivering in California.
Results:
Of 21,196 VA-enrolled Veterans with births, VA was the payer for 6% (1,257). Veterans relying on VA-coverage were slightly younger (median age 28 vs. 29); race/ethnicity did not differ across payer. Outcomes were similar for Veterans who did vs. did not use VA coverage, with no significant difference in incidence of cesarean delivery (33% vs 32%, p = 0.3), preterm birth (13.2% vs 13.3%, p = 0.96), gestational diabetes (7.5% vs 6.8%, p = 0.4) or preeclampsia (4.0% vs 3.4%, p = 0.2). Despite similar maternal outcomes, 7.2% of newborns of Veterans using VA coverage received neonatal intensive care unit (NICU) care vs 5.2% delivered under non-VA coverage (relative risk 1.38, p = 0.002). Comparing all-payer Veteran births to the remaining 6.5 million California births, Veteran mothers were slightly older (median 29 vs 28), and more likely to be Black. There were few differences in maternal outcomes, with slightly less preterm birth (12.4% vs 12.8%, p = 0.04) and slightly more cesarean delivery (31.2% vs. 29.4%, p < 0.001). NICU care was again a notable exception--increased among Veterans' offspring, 5% vs 4% (p < 0.001).
Implications:
VA does not cover an overtly higher risk obstetric population, but Veterans' newborns are more likely to use the NICU, particularly if the birth was VA-covered.
Impacts:
As the VA disproportionately serves Veterans with high mental health needs, these findings suggest a need for more nuanced examination of newborn outcomes with focus on potential maternal-fetal exposures (e.g. psychotropics and drugs of abuse).