The Normalization of Deviance and Maternal Health Outcomes

There are a few terms that are used throughout healthcare quality improvement: Cognitive bias, confirmation bias, and normalization of deviance (NoD). What is normalization of deviance and what impact does it have on maternal health outcomes?

Posted under: Maternal Health, Other, Quality of Care

Healthcare workers, including providers, nurses, respiratory and occupational therapists, and so many others, enter healthcare to make a difference for patients, and their communities, and (let’s face it), create a sustainable livelihood and career.

At the end of the day, there is the desire to support individuals and communities from illness to wellness, and more recently, to support prevention efforts to avoid illness in the first place. The issues surrounding maternal health have accelerated in the past decade, particularly with the launch of the 2017 NPR/ProPublica Lost Mothers Series. Some would say that it was this work of journalism that catapulted maternal health to the discussion it is today. Not only did the journalists show the stark data on maternal deaths and mortality, but they showed the faces of those women who died and told their stories, which created (and quite frankly forced) the discussion of maternal health into the public domain. In addition, these stories also began to illuminate the stark inequities that exist, including the rates of death of Black and Brown women when compared to white women (3-4 times, and even today that data is dismissed).

Since then, a great deal of emphasis has evolved into discussions of maternal morbidity and mortality.

There are a few terms that are used throughout healthcare quality improvement: Cognitive bias, confirmation bias, and normalization of deviance (NoD) (there are others, but the focus here will be on NoD).

Cognitive bias was first described in the 1970s, as human behaviors that simplify situations that can seem complicated or uncertain. Confirmation bias is the process of only accepting data that is consistent with preconceived ideas, opinions, or thoughts. Normalization of deviance is the acceptance, over time, of processes and solutions that are broken or defective, and essentially, “normalize” poor outcomes and/or performance.

Normalization of Deviance is a term that originated with the Shuttle Challenger disaster in 1986. Sociologist Diane Vaughn described the process as “people within the organization became so accustomed to a deviation that they don’t consider it as deviant, despite the fact that they far exceed their own rules for elementary safety.” When Doing Wrong Feels So Right: Normalization of Deviance - PubMed (

So, what is the Normalization of Deviance:
  • Gradual reduction of safety standards (to a new normal) after an absence of negative outcomes, which reinforces the “Band-Aids” of workarounds, shortcuts, and other changes to processes when sustainable solutions are not available. Normalization of Deviance: Concept Analysis - PubMed (
  • Has the potential to exist in EVERY healthcare environment if left unchecked.
  • Requires just the right combination of exhaustion and frustration when processes are broken, which lead to workarounds, shortcuts, and “MacGyver” type fixes.
What Normalization of Deviance is NOT:
  • An individual issue. This type of behavior and reaction is a systemic issue and takes time to fester and grow. While each individual is responsible for the care of a patient or community, the “snowball effect” of individual responses creates a more global (and more dangerous) situation.
So, how does the Normalization of Deviance connect to maternal safety?

The latest Centers for Disease Control Report of Maternal Mortality Review Committees outlined the most recent maternal death data. Of the maternal deaths, 13% occurred on the day of delivery. While many programs focus on postpartum care in the community, between 6 weeks through 1 year postpartum (and beyond), we cannot overlook the importance of elevating care during the immediate labor/delivery/postpartum period which includes teams that rely on one another for the provision of high-quality care.

Dr. Christine Morton in 2014 described the issue surrounding the normalization of deviance in maternal health as the intersection of assuming childbirth to be risk-free and inherently risky, and clinicians not understanding practice patterns and maternal outcomes data tracking (Morton, C. (2014). The problem of increasing maternal morbidity: Integrating normality and risk in maternity care in the United States. Birth, 41(2)). While data capture and evaluation have improved since 2014, many of these same issues still exist in 2023.

States that are currently working through the AIM Safe Reduction of Primary Cesarean Birth Bundle know the normalization of deviance far too well. If your cesarean section rate is above 50%, you have probably heard the following statements:

“Our patients are sicker.”
“Our patients are older.”
“Our patients have more comorbidities.”
“Our patients _______________
(you fill in the blank).”

Don’t get me wrong…there are very specific indications why a cesarean section should occur. Cesarean birth can save the life of the mother, baby, or both.

But when cesarean section becomes the normal route of delivery and not the exception, fostering and supporting an intended vaginal birth can almost seem like a “workaround.”

Think about the care processes you have “normalized” that are anything but normal:
  • Vital signs in delivering women (“She has a higher heart rate because she is pregnant, no reason for worry”).
  • Setting alarm parameters just a “bit higher” than normal.
  • Dismissing data outcomes because you know your “patients are sicker than that” (and they might be…but the data you enter and validate is the data that comes out).
If you are not comfortable with the term Normalization of Deviance, that is okay. But it is a term that your unit and hospital should embrace and explore what “normalizations” your teams have created…they are there, but it takes courage, curiosity, and transparency to find them and to create a true culture of patient AND staff safety.

One of the primary reasons hospitals reach out to NPIC is to benchmark themselves against similar size hospitals, birth volumes, payers, and acuities. Your data is your data, and easy enough to track (potential confirmation bias). When you compare your data to another group of facilities much like yours, the difference may be minimal or may be significant, and variations in practice patterns may emerge.

Transparency and a true culture of safety that openly supports the identification of shortcuts, workarounds, errors, and near-misses is the only culture that will stall and eliminate the Normalization of Deviance.

The National Perinatal Information Center (NPIC) provides maternal and newborn data reporting and analytics for hospitals across the US. With the most robust and longest-tenured linked mother/baby database in the nation, NPIC has partnered with hospitals to offer better insights into their data and practice patterns. Those conversations have evolved into understanding how data comparison and benchmarking can support internal quality improvement programs and reduce the normalization of deviance in outcomes. If you would like information on building a more resilient maternal and newborn quality improvement program, contact NPIC at
Share this article:
< Back to NPIC blog home