Maternal outcomes associated with a statewide severe hypertension obstetric quality improvement initiative

OBJECTIVE An obstetric severe hypertension bundle was implemented through a statewide quality improvement initiative in NY State. The objective of this study was to evaluate trends in adverse maternal outcomes associated with hypertensive disorders of pregnancy (HDP) associated with bundle implementation. STUDY DESIGN Delivery hospitalizations in the 2007-2022 NY State Inpatient Database (SID) were analyzed for this repeated cross-sectional analysis. The NY SID includes all acute care inpatient discharge data for NY State. First, HDP diagnoses among all delivery hospitalizations were trended. Then, among deliveries with HDP, the following outcomes were trended: (i) transfusion, (ii) non-transfusion severe maternal morbidity (SMM), (iii) disseminated intravascular coagulation (DIC), and (iv) stroke. Trends analyses were performed with joinpoint regression to determine the average annual percent change (AAPC) with the exception of stroke. Because of small numerators, stroke risk during two 8-year periods (2007-2014 and 2015-2022) was compared with the chi square test. RESULTS Among 3,579,336 delivery hospitalizations, HDP increased from 6.1% in 2007 to 15.6% in 2022. In joinpoint analysis, transfusion among deliveries with HDP increased from 2007 to 2012 but then decreased from 2012 to 2016 (AAPC -4.9%, 95% CI -10.9%, -0.2%) before increasing after 2016. SMM increased from 2007 to 2014 before decreasing from 2014 to 2017 (AAPC -7.9%, 95% CI -11.5%, -1.4%), before rising again from 2017 to 2022. DIC increased from 2007 to 2013, decreased from 2013 to 2017 (AAPC -13.2%, 95% CI -19.4%, -7.9%), and increased non-significantly from 2017 on. From 2007 to 2014, 44 cases of HDP-associated stroke occurred compared to 19 from 2015 to 2022 (3.5 vs 1.0 per 10,000, p<0.01). CONCLUSION The initiation of hypertension bundle was associated with decreased risk for a range of adverse outcomes among deliveries with HDP including stroke. Decreases in risk continued for approximately 3 to 4 years after initiation of the program