In 15 years of working in women’s health, I’ve come to know that labor and delivery (L&D) clinicians and providers are some of the most committed caregivers in existence, but increasingly the most hamstrung by market forces and inadequate systems. We must lean on emerging technologies to augment care and fill dangerous gaps that are all too human.
It is not the obvious, dramatic saves that show these gaps in care. A rapidly deteriorating obstetric emergency is not difficult to recognize. In most cases, it does not go unnoticed – teams mobilize quickly, moving with speed and purpose to protect both mother and baby.
But that’s not how most adverse outcomes in childbirth develop.
More often, risk builds gradually, often unnoticed. Subtle, evolving fetal heart rate changes such as decreasing variability, increasing baseline, and increasing frequency, depth, and duration of decelerations – in isolation not concerning, but in aggregate a threat to the fetus. These warning signs are easy to overlook in the moment, but accumulate until the pattern becomes unmistakable and threatening. By then, the window to intervene may have narrowed or closed. It is in recognizing these small but compounding warning signs – signs which humans are not well equipped to track over time – where technology must do the heavy lifting.
Designing devices to counter a worsening trajectory in maternal care
The underlying dynamics of maternal care are in a critical state. Labor and delivery units are closing across the country, severely diminishing access to care. Experienced nurses are being replaced by less tenured staff and attrition among veteran providers is on the rise. Patients are older and present with more comorbidities. Longstanding disparities continue to put women of color at significantly higher risk.
These pressures are converging in real time at the bedside — and many of the tools meant to support care have not kept pace. Technologies that generate fragmented data or isolated signals introduce ambiguity and could increase risk for clinicians. This in turn forces them to synthesize information continuously across patients and to be hyper-vigilant to incremental changes happening slowly over time. In already strained environments, this can increase cognitive load and amplify variability rather than reduce it.
This is where certain well designed technology — AI is specifically promising here – must bear the brunt of processing data to highlight meaningful patterns over time and prioritize patients who require attention. Systems designed for this reduce cognitive load and support more consistent decision-making, reducing risk. The distinction is not between using technology or not, but between tools that create noise in the moment and those that provide situational awareness, context and clarity. Improving maternal safety will come from better data synthesis and orchestration. From tools that help clinicians recognize patterns earlier, communicate and act with confidence, regardless of geography or level of care.
Combating unintentional variability in care
There is much concern about the ‘black box’ of AI persisting bias, especially that shown towards women of color in peripartum care. But AI models born of intentional supervised learning may in fact offer an opportunity to combat that bias by specifically omitting any racial or ethnic factors in training. Well developed and carefully validated systems, trained on intentionally curated data, can help shift the focus toward objective clinical patterns, supporting earlier and more uniform recognition of risk across diverse patient populations. Human decision-making will always be influenced by experience, fatigue, or implicit bias, whereas these models apply consistent criteria across patients.
AI-enabled consistency can improve adherence to clinical standards by reducing variability in interpretation. Tasks like assessing fetal heart rate patterns are inherently subjective. AI can help prevent differences in interpretation that lead to inconsistent care.
By applying the same criteria across patients, algorithmic tools can help create a more uniform understanding of patient state. That consistency makes it easier to trigger protocols at the right time – supporting more reliable, system-wide adherence to standards of care. Ultimately, these tools improve the speed and consistency with which entire health systems recognize, communicate and act on clinical warning signs.
Bridging the experience gap
AI will help address the growing experience gap within labor and delivery units. Less experienced nurses have had fewer opportunities to build the pattern recognition and confidence that comes with repetition. In high-pressure settings – particularly during overnight shifts or in lower-volume facilities – this can make it more difficult to consistently recognize early signs of risk.
Thoughtfully designed decision-support tools can help bridge that gap. By providing a consistent assessment of evolving clinical patterns, these tools reinforce situational awareness, validate initial impressions, and support timely escalation when needed. The goal is not to replace clinical judgment, but to extend it, helping to ensure that patients receive the same level of vigilance and consistency regardless of staffing mix or setting.
This benefit is even more pronounced in resource-limited settings globally, where experienced clinicians may be scarce. In these environments, the ability to consistently interpret evolving clinical patterns can be the difference between early intervention and a poor outcome, giving decision-support tools an outsized impact in healthcare delivery.
Scaling care virtually
These tools also effectively extend the reach of an increasingly limited pool of experienced clinical talent across an entire health system. By using algorithmic stratification to identify the patients most in need of attention across large populations, health systems can better direct limited expertise to where it matters most – including settings where specialty support may not be readily available.
In many organizations, this capability is deployed through virtual care models. Centralized teams monitoring patients across multiple labor and delivery units, focusing on those with persistent or evolving risk. In this role, these clinicians serve as a second set of eyes – supporting bedside teams, reinforcing situational awareness, and helping ensure more consistent documentation.
When implemented thoughtfully, these approaches can improve consistency in care delivery and reduce variability across sites – key factors in strengthening quality and mitigating risk at the system level.
Improving maternal safety will not come from adding more data or more devices. It will come from building systems that help clinicians recognize risk earlier and communicate and act with confidence in the moment.
Technology has an urgent role to play in improving maternal care, but it must be engineered to provide clarity, not just more or different data and it must be flexible and easy to deploy, even in the most bare bones environment. When that occurs, technology will reduce variation in care, support clinical judgment, and fit seamlessly into care delivery. In already strained environments, the wrong tools just add noise and increase burden. The right tools bring clarity – and will become standards of care. In maternal health and beyond, health systems can’t afford to be out of step here.
Photo: damircudic, Getty Images
Matthew Sappern has served as Perigen’s Chief Executive Officer since 2012, leading the company through a period of significant growth, product expansion, and global impact. Under his leadership, the organization has scaled adoption across major health systems, advanced a portfolio of FDA‑cleared clinical software, and established partnerships focused on improving safety and equity in childbirth worldwide.
Matthew brings nearly 30 years of experience building and operating regulated healthcare technology businesses. Prior to his current role, he held senior leadership positions at Allscripts, Eclipsys, and WebMD, with responsibilities spanning product development, commercialization, and operations. He holds a Bachelor of Arts from Emory University, serves on numerous boards, and is an Executive in Residence at the Yale University Venture Lab.
This post appears through the MedCity Influencers program. Anyone can publish their perspective on business and innovation in healthcare on MedCity News through MedCity Influencers. Click here to find out how.
