Hospital antenatal admissions for threatened preterm labor: how long should we be "observing"?

We sought to describe temporal trends in hospital admissions for threatened preterm labor (TPTL) and to examine hospital admission duration among women delivered or discharged undelivered.

We carried out a cohort study on all TPTL admissions among pregnancies with a live singleton fetus and intact membranes between 1999 and 2015 using the United States' Healthcare Cost and Utilization Project-Nationwide Inpatient Sample. ICD-9 codes were used to identify women with TPTL. Duration of antenatal admission length of stay was calculated in days following admission to hospital until delivery ("Delivery Admission") or undelivered discharge ("Observation Admission"). Analyses included evaluating trends of birth admissions over total admissions, identifying predictors of delivery using logistic regression, and measuring risk for delivery with increasing duration of antepartum hospitalization.

Of 15,335,288 pregnancy admissions, 1,089,987 admissions were for TPTL, with 61.8% being 'Delivery Admissions". During the 16-year study period, overall rates of TPTL admissions declined with a rising proportion of admissions being "Delivery Admissions". "Delivery Admissions" were more common among patients who were older, non-Caucasian, obese, or who had placental abruption. "Observation Admissions" were more common among admissions for antepartum hemorrhage or antepartum spotting. Among all "Delivery Admissions" for TPTL, 89% had delivered within 2 days, 7% delivered within 3-6 days, and 5% delivered beyond 6 days.

Overall admissions for TPTL declined over the study period with increasing proportions being "Delivery Admissions". Protocols taking into consideration declining risk of preterm birth among patients undelivered after 2 days may be helpful in reducing unnecessary prolonged observation admissions.