GHS Borrows a Tool from the Warriors to Improve Health and Save Lives

By Anne Wolf

Stephen Curry of the Golden State Warriors takes a shot.

Video technology that helped the Golden State Warriors win 73 games this season and get to the NBA finals for two years running is helping save the lives of mothers and their babies in Bihar, India.

Studiocode™, a video coding software that enables viewers to mark the action on video clips in order to quickly view specific points for analysis and data collection, was developed from Sportscode, a similar technology that professional and amateur athletic teams, including the Warriors, use to analyze their players' and competitors' performance.

The software is key to research Hilary Spindler, investigator at UCSF Global Health Sciences, is conducting in collaboration with the NGO PRONTO International, to monitor and improve obstetric simulation training for midwives and nurses in Bihar, India. With a population of 104 million and approximately 2.8 million births per year, Bihar is one of the most densely populated states in India, and the world, and has one of the highest rates of maternal mortality in India.

The Warriors and 28 other NBA teams use Sportscode™ to maintain team stats, analyze the plays and strategy of their opponents, and enhance the team's or individual players' performance. For example, says Will Callanen, senior account manager for Studiocode, the Warriors might mark and look at every clip of the team's season when Stephen Curry drives to the basket against the Cleveland Cavaliers. By watching every such clip, they can analyze the Cavaliers' tendencies: who's guarding Curry and where they come from, which can help Curry prepare his game plan.

Similarly, Spindler and her colleagues are using Studiocode to measure and analyze key indicators in teamwork and communication and clinical birth practices used in simulation. By coding thousands of birth simulation videos, Spindler is trying to show that the training improves teamwork and clinical practices related to normal vaginal deliveries, management of maternal postpartum hemorrhage, and neonatal resuscitation.

Spindler is part of a team working with GHS's Dilys Walker, MD, professor of obstetrics, gynecology and reproductive health and principal investigator for the project in Bihar.

The Bihar project aims to reduce maternal and infant deaths during labor and delivery through birth simulation training at 320 primary health facilities and 56 district hospitals. Approximately 300,000 maternal deaths and 2.9 million neonatal deaths occur annually worldwide. The state of Bihar in India has an unacceptably high, and probably grossly underestimated, maternal mortality rate with an estimated 208 deaths per 100,000 live births.

The project, funded by the Bill and Melinda Gates Foundation, is a partnership with PRONTO International, UCSF Global Health Sciences and Care India.

Walker and her team have trained 120 nurse-midwife mentors who, in turn, have trained and mentored nurses in primary health clinics and district hospitals across the state. The birth simulations are videotaped, then coded by a team of analysts in Bihar using Studiocode at specific points in the scenario to assess both clinical and team behavior. The videos are used to provide rapid, evidence-based feedback to the mentors and mentees on improvements and weaknesses in communication and clinical skills and techniques.

Spindler then uses the marked video clips to measure the effectiveness of the training over time. About a quarter of the simulations to date have been coded and analyzed to determine whether the training is improving neonatal resuscitation practices. The research focuses on neonatal resuscitation because more than a quarter of infant deaths in the first seven days of life are due to birth asphyxia, a preventable cause of death.

Spindler says the analysis shows there is no drop off in quality of care as the simulation scenarios become more difficult.

"As simulation difficulty increased there was no change in the rate of completion of basic tasks of newborn care and neonatal resuscitation," Spindler said. "This remained true even with only 1-2 health workers in the simulation."

In addition, as simulation difficulty increased, response times were faster for specific clinical practices such as placing the newborn in a warmer after delivery and initiating positive pressure ventilation to help the newborn breathe. This may indicate that the small teams are becoming more efficient and effective in times of extreme stress.

This early analysis enables the project team to fine-tune the program to improve its effectiveness and then continue to use the Studiocode software to further research the training's effectiveness.

Walker, who is also co-founder of PRONTO International, previously has conducted smaller projects using the birth simulation training and videotaped the trainings primarily for use in debriefing.

"But the Bihar project is so big; we could not manage the implementation or see if quality is being maintained without the ability Studiocode provides to analyze the videos from the field," she said.