2008
The Story Behind the Statistics
In the summer of 2005 Lance Roberts, a graduate research assistant in the Center for Health Policy and Research, was crunching the numbers for a public report on patient safety in Iowa hospitals when he noticed a disturbing trend.
“Clustered at the bottom of the chart were four categories of obstetric trauma, all of which had to do with maternal lacerations during vaginal deliveries,” Roberts says. “It’s unusual to have such a clear case of poor performance, persistent over several years, and I wondered what was going on.”
Benchmarking Performance
The public reports are published by the Iowa Healthcare Collaborative, a joint venture of the Iowa Medical Society and the Iowa Hospital Association. Administrative discharge data are available for all hospitals in Iowa, and the Center for Health Policy and Research, located in the Department of Health Management and Policy, processes the data using a set of quality indicator software tools developed by the Agency for Healthcare Research and Quality (AHRQ).
According to Marcia Ward, Ph.D., professor of health management and policy and director of the Center for Health Policy and Research, AHRQ’s patient safety indicators give hospitals a benchmark for comparison with other hospitals in Iowa as well as with the national average.
“For the first time administrators can glance at the indicators, which include quality indicators like hospital infections, postoperative sepsis, and obstetric trauma, and get a general idea of how their hospital compares with other hospitals in the state and nationwide,” Ward says.
According to the reports, available since 2002, Iowa hospitals scored better than the national average in rescue attempts, bed sores, and postoperative pulmonary embolism, and average on postoperative hip fracture and postoperative hemorrhage. But the four patient safety indicators related to obstetric trauma were at the bottom of the ratings for three consecutive years.
“However, the report ratings don’t give the whole story,” Ward says. “They point out areas of concern, but to understand what’s really going on, researchers need to dig deeper and look at the clinical picture behind the numbers.”
Risk Factors and Ratings
To investigate the clinical factors that might be contributing to obstetric trauma in Iowa, Ward and Roberts teamed up with John Ely, M.D., professor of family medicine. Using ICD-9 codes—insurance payment codes that identify a patient’s medical risk factors—the researchers were able to group together those factors that might make delivery difficult, including factors related to the both the mother (such as age and medical complications) and the baby (such as size and position).
“We categorized all the risk factors that seemed related to obstetric trauma, analyzed them statistically, and came up with a better idea of why Iowa appears to have higher rates of this type of trauma,” Ely says.
The researchers found that Iowa hospitals have a decreased use of forceps and episiotomies during vaginal deliveries, consistent with clinical practice guidelines, and also a lower rate of Cesarean deliveries. However, Iowa hospitals have an increased rate of long labor, late pregnancies (longer than 40 weeks), and obstructed labor. They suggested that in rural areas, where fewer Cesarean sections are performed, physicians may have to use forceps on an urgent basis in women who are having a difficult birth, a situation that can result in lacerations.
Roberts’ investigation of Iowa’s low ratings in obstetric trauma resulted in his first peer-reviewed publication in the Journal of American Medical Quality. Ward says the Iowa Healthcare Collaborative has had good feedback from hospitals about the public reports and that the reports are still evolving. “Future reports will allow Iowa hospitals to compare themselves not only with a national index but also with each other on a wider set of indicators, which will give them an even better idea of how they’re doing.