The Impact of Implementing an Obstetric Co-Morbidity Scoring System during Delivery Admissions on Maternal Morbidity
PresenterFlash Talk Presenter
Adina Kern-Goldberger, MD MPH is a third-year maternal-fetal medicine (MFM) fellow at the Hospital of the University of Pennsylvania and a second-year student in the MSCE program. She is interested in obstetric healthcare delivery science research and in optimizing safety, quality, access, and equity to reduce maternal morbidity. Her current research focuses on comorbidity-based risk stratification in obstetric health services, labor floor quality and safety (especially alarm fatigue), and the utilization of telemedicine to improve access to MFM care. She is also pursuing additional training in quality improvement. She looks forward to leveraging her training in public health, epidemiology, and QI as a clinician-investigator commited to addressing the challenge of maternal morbidity.
Background: Severe maternal morbidity in the U.S. has increased significantly over the past decade during delivery admissions. A novel strategy to improve risk prediction and early identification of maternal morbidity is to utilize a comorbidity-based tool that stratifies risk for morbidity based on demographic features and clinical conditions. Comorbidity indices have been broadly utilized with success across many medical disciplines to predict morbidity, and an obstetric comorbidity index (OB-CMI) has been developed and validated for clinical morbidity risk in obstetric patients.
Objective: The goal of this project is to implement the OB-CMI as standard of care on the labor and delivery unit at the Hospital of the University of Pennsylvania (HUP) for all patients over a one-year period and then to assess the impact on the incidence of a maternal morbidity composite variable compared to a one-year pre-implementation time period.
Methods: This is a prospective hybrid implementation-effectiveness study examining maternal morbidity before and after the implementation of the OB-CMI as standard of care on the HUP labor and delivery unit. The OB-CMI will be implemented during the one-year period from January 1, 2021 – December 31, 2021 and compared to the one-year period from January 1, 2020 – December 31, 2020. A user-friendly OB-CMI online calculator was created and made available to all obstetric providers. The OB-CMI was calculated for all patients on admission and documented in the medical record. An OB-CMI ≥ 6 prompted an additional structured interdisciplinary “morbidity reduction huddle” between the nursing and medical teams at admission and when the patient was transferred to the postpartum floor to review the patient’s care. The primary maternal morbidity outcome composite was defined by as ≥ one of the following during the delivery admission: endometritis, postpartum hemorrhage (defined as estimated blood loss > 1L), blood product transfusion, venous thromboembolism, hysterectomy, ICU admission, length of postpartum stay ≥ 5 days, or 30-day readmission. Secondary outcomes included alternate maternal morbidity definitions: (1) The Joint Commission definition [blood transfusion ≥ 4 units or ICU admission] and (2) CDC definition of severe maternal morbidity [SMM] based on ICD-10 codes (acute myocardial infarction, aneurysm, acute renal failure, adult respiratory distress syndrome, amniotic fluid embolism, cardiac arrest/ventricular fibrillation, conversion of cardiac rhythm, disseminated intravascular coagulation, eclampsia, heart failure/arrest during surgery or procedure, puerperal cerebrovascular disorders, pulmonary edema/acute heart failure, severe anesthesia complications, sepsis, shock, sickle cell disease with crisis, air and thrombotic embolism, blood product transfusions, hysterectomy, temporary tracheostomy, ventilation). The individual components of the primary outcome composite were also evaluated. Demographic and clinical characteristics as well as patient outcomes were extracted from the medical record and verified with chart review. Secondary implementation outcomes including intervention fidelity, acceptability, and perceived impact were assesed as well. Fidelity was evaluated via chart review to assess for documentation of the OB-CMI score and the huddle, as appropriate. Acceptability and perceived impact of the OB-CMI intervention were assessed with a survey of labor and delivery clinicians at the study midpoint, August 2021. As the OB-CMI score is already known to predict severe maternal morbidity as an outcome, the primary analysis assessed whether any differences in outcomes between 2019 and 2021 are attributable to differences in OB-CMI scores themselves or to the implementation of the clinical OB-CMI intervention. Raw comparisons of demographics, co-morbidity index scores, and outcomes between the exposed and unexposed groups were conducted with chi-square, Student’s t-test, and Wilcoxon rank-sum tests, as appropriate. A multivariable logistic regression model evaluating cohort year [as a proxy for intervention], OB-CMI score, and the interaction between cohort year and OB-CMI score was utilized to compare the dichotomous primary outcome. Time-series analysis of trends in the primary outcome from 2019 - 2021 was also employed to determine whether changes were due to the intervention or to secular trends. The baseline rate of the composite severe maternal morbidity outcome in the HUP patient population is approximately 15% based on prior studies. Given approximately 4,000 deliveries per year at HUP, the study has 80% power to detect a 2.3% absolute reduction in maternal morbidity. This correlates with a reduction in approximately 100 cases of severe morbidity per year.
Results: A total of 4,150 patients were included in the pre-implementation cohort (2020) and 4,047 patients in the post-impementation cohort (2021). There were no significant differences in baseline demographics, obstetric characteristics, co-morbidities, median OB-CMI score, or percent of patients with an elevated OB-CMI score ≥ 6 between the two groups. Raw comparison between the pre and post-implementation periods demonstrated a higher rate of the primary outcome in the 2021 cohort (15.8%) compared to the 2020 cohort (13.8%) [p = 0.01]. Adjusted analysis of the impact of OB-CMI score implementation on the incidence of the primary outcome composite demonstrated no significant difference (OR 1.18, 95% CI 0.96 - 1.45, p = 0.12). This was also true for all secondary outcomes, other than endometritis, which was shown to increase in 2021 relative to 2020 (OR 2.40, 95% CI .19 - 4.85, p = 0.02). Multiple sensitivity analyses were conducted, including by OB-CMI score bracket and excluding patients with the highest risk co-morbidity conditions, and there were no changes in the overall study findings. A further analysis combining 2019 and 2020 data as a single pre-intervention period, to obviate any abberancies in 2020 outcomes based on the COVID-19 pandemic, also demonstarted no differences in the odds of the primary outcome compared to 2021. Time series analysis of trends of the primary outcome composite over time, from 2019 - 2021, using negative binomial and piecewise linear regression models did not demonstrate significant trends over this period. In terms of implementation metrics, OB-CMI documentation fidelity reached saturation at 65% and huddle documentation fidelity at 80%, by the study midpoint. Mean acceptability, appropriateness, and feasibility scores ranged between 4 and 5 (out of 5) and were statistically higher for attending physicians compared to residents (p < 0.01).
Conclusion: Implementation of an OB-CMI scoring system for delivery admissions did not reduce the rate of the primary maternal morbidity outcome composite. More active interventions to improve intrapartum and postpartum maternal morbidity are required, and may be informed by OB-CMI scores, in order to optimize maternal outcomes. Implementation of an intervention of this nature is feasible and acceptable to obstetric clinicians.
Keywordsmaternal morbidity, implementation, co-morbidity
To understand health and disease today, we need new thinking and novel science —the kind we create when multiple disciplines work together from the ground up. That is why this department has put forward a bold vision in population-health science: a single academic home for biostatistics, epidemiology and informatics.