The maternal ward is known as the happiest department in every hospital, a place where new life is greeted with tender and focused attention. But what about the patient who has just ushered this new life into the world—the mother? In America, her chance of death or severe harm is jarringly high. A ProPublica/NPR investigation noted that 700 to 900 women annually die in the U.S. from pregnancy or childbirth-related causes. The investigation further found that another 65,000 come close to dying as a result of these causes.
This is practically unheard of in other developed countries. At first glance, it’s difficult to understand why these deaths are happening in the U.S. We have the world’s best-trained clinicians, working in hospitals equipped with sophisticated instruments and monitoring machines. Indeed, other countries flock to our healthcare technology conferences to learn the latest innovations in care.
Yet, a closer examination of maternal mortality in America uncovers some surprising realities.
It turns out we don’t regularly staff enough nurses in our nation’s labor and delivery units, in part because of a national nursing shortage and a lengthy training period for nurses to be considered tenured in this specialty. Those that are present experience persistent “alarm fatigue” from the ceaseless onslaught of blips and beeps that emanate from various monitoring devices. Or they work with devices upwards of 20 years old, and that require manual data interpretation. Add to these factors a patient population with ever more complex co-morbidities, and our high maternal mortality rates are a mystery no longer. Neither are the solutions.
An AI-driven approach for labor and delivery
To save expectant mothers, hospitals should improve the technology and processes upon which our nation’s labor and delivery nurses rely. The most foundational change would be to heed the recommendation of the National Partnership for Maternal Safety to institute maternal-fetal Early Warning Systems (EWS). EWS in obstetrics have been associated with improved compliance related to treatment with prompt use of medications and protocols known to prevent maternal complications and death associated with the most common causes or maternal morbidity and mortality.
Current safeguards can span from laminated card checklists that nurses use to assess each patient at prescribed time periods or EHR modules that use a linear, rules-based approach to automate analysis of nursing documentation. It should be pointed out, however, that these tools won’t be nearly as effective as those that use artificial intelligence (AI) to identify patterns not immediately obvious to humans.
In the healthcare arena, specifically childbirth, AI-powered systems can be taught to scan real-time data such as fetal and maternal heart rate signals, contractions and vital signs. They can then compare any concerning trends to past outcomes and then alert nurses of situations that require their immediate attention, instead of overwhelming them with meaningless alarms. Such alerts can break through the hectic fog of inpatient care in even the most short-staffed wards to inform a nurse of imminent risk.
And clearly, advance warning is a crucial step in reducing what are largely preventable deaths. Here are just some of the compelling stats, also compiled by the Partnership, that make the case for instituting advance or early warnings:
- Between 1998 and 2009, post-partum mortality increased by 66 percent;
- 93 percent of hemorrhagic deaths were potentially preventable; a common factor was lack of attention to the warning signs; and
- 60 percent of severe hypertension-related deaths were potentially preventable; A common factor was the failure to recognize certain complications that cause hypertension.
As further proof that these early warning systems work, a published study of more than 78,000 births involving one baby, conducted at healthcare organization MedStar, found its patient safety program, which included automated early warning software, resulted in significantly lower unanticipated NICU admissions. There were also fewer newborn resuscitations, controlled cesarean rates, and a reduction in the proportion of malpractice complaints and awards coming from the obstetrics unit.
Here is what is surprising. Most of the respondents, all labor and delivery department leaders, are aware of research on the impact of an EWS, yet only 25 percent have implemented one, according to a recent survey. Moreover, a full 50 percent of respondents aren’t even considering it yet. Equally concerning, the vast majority of these systems rely on manual intervention rather than automation. As an industry, this can’t wait. Expectant mothers and their families can’t wait.
We know that technology and knowledge exist to create an effective, automated EWS, and it should be part of every L&D department. Organizational, industry and especially government leaders must amplify this clarion call for maternal safety. In a primary example, CMS should require hospitals to report on unanticipated NICU admissions and other quality measures that indicate how quickly clinicians are responding to life-threatening conditions that require attention. Ratings and reimbursement would subsequently tie to the outcomes. Put such quality measures in place and we will soon see numbers trend in the right direction again.
The key word is “soon.” Too many families are being shattered and children left motherless due to not recognizing early signs of distress. The sooner we act, the more of these lives we can save.
Photo: asiseeit, Getty Images