Hospital or birth facility leadership appoints a standing Severe Maternal Morbidity Committee. This may require new bylaws.

Committee membership is multidisciplinary and reflects the professional make-up of clinicians and staff who provide or support maternity services institutionally. Example members are: obstetricians, family physicians, certified nurse–midwives, and advanced-practice nurses; anesthesia personnel; registered nurses providing antepartum, intrapartum, or postpartum care; and members of the hospital Quality Improvement team and administration. A public member or patient advocate could be considered. Ad-hoc members representing other expertise can be invited as deemed necessary. If there are learners such as residents or fellows, they should be represented as well.

The Committee has a chairperson, an individual responsible for minutes, and an individual responsible for data management.

At a minimum, the Committee will review all pregnant or postpartum women receiving 4 or more units of blood or admitted to an ICU. These criteria may be expanded as needed by an individual center.

For each case of severe maternal morbidity, a debriefing with involved care providers, which does not replace the standardized review, is suggested and ideally occurs proximate to the severe maternal morbidity. Information obtained from the debriefing can be retained for the standardized review process. There are several debrief tools available (https://www.cmqcc.org/resources/1533/download and http://www.med.unc.edu/ticker/toolkit/teamwork/brief-debrief-form). Another example is a debrief tool developed by C. Lee and D. Goffman, who gave permission for its inclusion in this article (Montefiore Medical Center/Albert Einstein College of Medicine, Bronx, NY) (Fig. 1).

The severe maternal morbidity review should be conducted at each facility, if possible.

Centers with a low volume of deliveries or obstetric providers may opt to partner with centers within their perinatal region, or outsource their reviews to a center with sufficient staff and providers to conduct the reviews.

Leaders from regional perinatal centers, state maternal and child health departments, state medical and nursing societies, and ACOG districts could identify experts who will be available to assist local committees.

Chart abstraction for patients whose care is being reviewed will be done by an individual trained in the abstraction and review process.

Data abstraction and review should follow a specific format to facilitate consistent collection of meaningful information that can be adequately evaluated. One method of achieving this is to use a specific form for data abstraction and review. A good example of such a form is the Severe Maternal Morbidity Abstraction and Assessment Form (www.safehealthcareforeverywoman.org). This form is consistent with the root cause analysis and action framework as it applies to obstetric sentinel events.¹⁴ Other tools such as the fishbone diagram for root cause analysis or sentinel event could also be used. The goals of the abstraction are to capture specific data as well as the essence of the severe morbidity, including a narrative with a detailed time line of the pertinent events. To aid the abstractor in identifying key issues, a sample list of disease-specific questions associated with severe maternal morbidity are included (Appendix). These types of questions can be used for any morbidity and are meant only as a guide.

The abstracted data should be presented to the Severe Maternal Morbidity Review Committee for their review and assessment.

The assessment portion of the Severe Maternal Morbidity Abstraction and Assessment Form is completed based on the results of the Committee’s review of the case. Each review should conclude with an assessment of whether there were opportunities to improve outcome. If identified, these opportunities are enumerated, and specific recommendations for potential alterations in outcome should be suggested to the appropriate responsible institutional person or department.

Each institution should have agreed-upon mechanisms in place to implement any recommendations and to evaluate the effects of the suggested changes.

Reviews are not conducted as peer review, which addresses credentialing and formal discipline issues, but rather as expert review focused broadly on improving systems and care provision. If a case has peer-review issues, then these should have been identified and handled in the existing institutional peer-review process. Expert review may be anonymous and has no authority to discipline individuals, but rather looks for opportunities to reduce preventable morbidity and mortality. State laws often protect expert review from discovery, but this must be determined on a state-by-state basis. The culture and tone of the review process must be nonjudgmental, and the final assessment, with respect to potential improvements in care, rendered as dispassionately as possible. One example of such an approach is the “Just Culture,” whose key principles include educating caregivers about risk, holding them responsible for following best practices, creating a safe haven around reporting, and recognizing what can and cannot be controlled.¹⁵

Data, including outcome data, are trended. The identification of potentially improvable measures as the primary goal of these reviews is to recognize systems, provider, and patient factors that are amenable to alterations that could improve outcomes in future cases. Using a consistent data form, such as the Severe Maternal Abstraction and Assessment Form, will facilitate acquisition of analyzable data.

Each facility will determine the best method to disseminate important recommendations for improvement in outcomes locally and more broadly.

Data from internal facility reviews could be aggregated at regional levels (ie, perinatal regional, state, and ACOG districts, and AWHONN sections and chapters). De-identified aggregate data reviewed at regional and national levels could help to identify trends and, more importantly, opportunities for improvement in the delivery of obstetric health care. ACOG districts may accumulate reports and share trend and action data with others.

Consideration could be given to utilizing Patient Safety Organizations in this national effort to consolidate and disseminate appropriate findings. Patient Safety Organizations were developed by the Patient Safety Improvement Act of 2005 (www.pso.ahrq.gov) and provide privilege and confidentiality to clinicians and health care organizations to collect and analyze data to improve quality by reducing risks associated with patient care.

Depending on the severity of the event, reviews should occur soon after the event and include obtaining information from the debriefing. The more severe the event, the more proximate the review should be to the event. Hospitals that wish to review a broader list of morbidities or that have a large number of cases may choose to have the Severe Maternal Morbidity Review Committee meet regularly.

The Severe Maternal Morbidity Review Committee should be sanctioned by the hospital and protected from discovery. State codes and relevant statutes must be reviewed to determine if protection or authority exists for maternal morbidity review. State ACOG sections and hospital associations may be helpful in identifying statutes and lobbying state legislative bodies for appropriate changes if needed.

All Committee members sign affidavits of confidentially.