The Impact of a Protocol on Equitable Labor and Delivery Substance Use Screening: A Retrospective Cohort Study

The Society for Maternal-Fetal Medicine (SMFM), the American College of Obstetricians and Gynecologists (ACOG), and the American Society of Addiction Medicine (ASAM) jointly recommend universal substance use screening in pregnancy with a validated tool. Biologic testing with urine drug screening (UDS) is not recommended unless testing would change medical management. Despite recommendations, racial and socioeconomic differences in substance use screening and biological testing exist. We evaluated the impact of an obstetric substance use screening protocol on validated screening, UDS indication documentation, pre-UDS patient assent, and UDS disparities across race, ethnicity, and payor.

This retrospective cohort study used electronic health record data before (July 1, 2020-June 9, 2021) and after (June 10, 2021-May 31, 2022) implementation of an obstetric substance use screening protocol at a tertiary care center. Documented, validated screening, UDS indication, and patient assent were assessed among UDS encounters pre- and postintervention. Association of race, ethnicity, and payor with UDS was assessed using generalized estimating equations models pre- and postintervention. Categorical variables were compared using chi-square and Fisher's exact tests.

UDS occurred less frequently postintervention than preintervention (1.9% [124/6,412 encounters] vs. 7.8% [441/5,658 encounters], p < 0.001). Adjusting for payor, Black individuals had higher odds of UDS (adjusted odds ratio [aOR] = 2.11, 95% confidence interval [CI]: 1.64-2.70) than White individuals preintervention; however, there were no differences (aOR = 1.23, 95% CI: 0.81-1.87) postintervention. Privately insured individuals had lower odds of UDS preintervention (OR = 0.23, 95% CI: 0.18-0.30) and postintervention (OR = 0.08, 95% CI: 0.04-0.15) than government-insured individuals. Among those with UDS, none (0/441) underwent validated screening preintervention versus 36.3% (45/124) postintervention. Documented UDS indication occurred in 27.9% (123/441) preintervention versus 55.7% (69/124) postintervention (p < 0.001). Documented assent occurred in 0.5% (2/441) of encounters preintervention versus 47.6% (59/124) postintervention (p < 0.001).

Implementation of an obstetric substance use screening protocol was associated with decreasing UDS and increasing validated screening, documented UDS indication, and patient assent. Racial disparities in UDS were reduced postintervention; however, suboptimal screening remained, with persistent disparities by payor.

· Substance use screening in pregnancy with a validated tool is performed suboptimally.. · There are known disparities in biologic testing for substance use by race and socioeconomic status.. · We evaluated an obstetric substance use screening protocol pre- and postintervention.. · The protocol was associated with validated tool use, patient assent, and screening documentation.. · Racial disparities in substance use screening were reduced postintervention..