New study on decision making for infants with complex life threatening illnesses

Sharron Docherty understands that medical care for a critically ill infant involves much more than treatments designed to bring about a cure.

Treating and caring for an infant born with a life threatening condition involves parents, doctors, nurses and social workers caught up in a web of complex decision making, usually made in tight timeframes and under incredible stress. The questions are life and death: Should another surgery or transplant be attempted? Will another treatment cause the child extreme discomfort? Should the focus shift from trying to cure to providing comfort to an infant who most likely won't survive?

Docherty, an associate professor in the Duke University School of Nursing, wants to understand that decision making process in an effort to improve communications and, ultimately, care during events which are almost always life changing to the families involved.

"It's a complex phenomenon with many moving pieces and the whole is greater than the sum of the parts," said Docherty, who leads a study on decision making for infants with complex life threatening illnesses with colleague Debra Brandon, also a faculty member in the Duke School of Nursing. 'We want to understand the decision-making trajectory for everyone involved in these cases, describe it, and search for patterns in a scientific way."

An understanding of the decision making process, including how and when parents make decisions, how parents acquire and use information and how they are influenced by doctors, surgeons, nurses and social workers will help hospitals and caregivers develop appropriate strategies for working with families facing crisis situations, according to Docherty. In addition, being able to visualize data to see decision-making trajectories can uncover patterns and reveal turning points or even warning markers of an infant's deteriorating condition.

For their five-year study of the decision making process Docherty and Brandon will follow 40 cases at Duke University Medical Center for one year, conducting in-depth interviews with parents and caregivers during and after treatments and aggregating medical data on the infants from the duration of their hospital stays. All the case studies involve infants with one of three complex, life-threatening conditions: extreme prematurity (babies born before 26 weeks gestation); inborn errors of metabolism that require a stem cell transplant from bone marrow or blood; or complex congenital heart disease.

A Data Viz Challenge
Last fall, the researchers were awarded funding from the National Institute of Nursing Research at the National Institutes of Health to work with RENCI to develop a new and intuitive way to visualize and analyze their complex, disparate data that would reveal relationships among data points and undercover patterns and trends.

Since then, Xunlei Wu, a senior visualization researcher at RENCI's Duke University Engagement Center, has joined the research team. The project presented a new challenge for Wu, who needed to find a way to integrate and present textual data from interviews as well as data captured by monitors and medical records.

"My job is to organize the data so that it can be presented along a graph over time that shows the trajectory of the illness and the trajectory of the decision making process," said Wu. "We need to represent many different events and many different decisions. It is very different from the types of scientific data I have visualized in the past. It's a challenge, but we think it will pay off with a visualization system that can be generalized for use by other researchers."

Initially, Wu is mapping data from single cases—decisions made by parents and healthcare providers, surgeries and other treatments performed, when parents and providers began to consider withdrawal of treatment, etc.—onto the trajectory of the illness, each of which ends with either the infant's death or long-term release from the hospital. Wu will superimpose other attributes derived from interviews along the graph, such as parents' and caregivers' levels of hope, perceptions of the infant's discomfort and parental regrets.

Later, Wu will combine the individual graphs so that the researchers can view all their case studies and discover commonalities and patterns.

"What has made this project work with RENCI is the constant collaboration and iteration and learning that takes place," said Docherty. "That allows us to create something new that fits our needs, but also should be able to help other scientists who do case study research."

She admitted that interviewing parents about the life and death struggles of their infants can be difficult, but added that everyone involved in the research—including parents—understands that others could benefit from what the study reveals.

"The least we can do is to tell their stories in a compelling way, present the data and learn from it."