Association of Maternal Comorbidity With Severe Maternal Morbidity: A Cohort Study of California Mothers Delivering Between 1997 and 2014

CA 2020
Annals of Internal Medicine
Scientific/Peer-reviewed article

Background:
Rates of maternal mortality and severe maternal morbidity (SMM) are higher in the United States than in other high-resource countries and are increasing further.
Objective:
To examine the association of maternal comorbid conditions, age, body mass index, and previous cesarean birth with occurrence of SMM.
Design:
Population-based cohort study using linked delivery hospitalization discharge data and vital records.
Setting:
California, 1997 to 2014.
Patients:
All 9 179 472 mothers delivering in California during 1997 to 2014.
Measurements:
SMM rate, total and without transfusion-only cases; 2019 maternal comorbidity index.
Results:
Total SMM increased by 160% during this time, and SMM excluding transfusion-only cases increased by 53%. Medical comorbid conditions were associated with an increasing portion of SMM occurrences. Medical comorbid conditions increased over the study period by 111%, and obstetric comorbid conditions increased by 30% to 40%. Identified medical comorbid conditions had high relative risks ranging from 1.3 to 14.3 for total SMM and even higher relative risks for nontransfusion SMM (to 32.4). The obstetric comorbidity index that is most often used may be undervaluing the degree of association with SMM.
Limitations:
Hospital discharge diagnosis files and birth certificate records can have misclassifications and may not include all relevant clinical data or social determinants. The period for analysis ended in 2014 to avoid the transition to the International Classification of Diseases, 10th Revision, Clinical Modification, and therefore missed more recent years.
Conclusion:
Obstetric and, particularly, medical comorbid conditions are increasing among women who develop SMM. The maternal comorbidity index is a promising tool for patient risk assessment and case-mix adjustment, but refinement of factor weights may be indicated.

risk appropriate care; data quality and surveillance improvement