Clinical Audits: Obstetrics
Some Obstetrics Performance & Outcomes Challenges
Maternal and baby care present significant challenges in both performance and outcomes. Clinical issues range from prenatal care, especially for patients that present to deliver without it or high-risk mothers abandoned at the last minute by their paraprofessional care giver, through delivery and post-partum care. And care challenges cross the professions of obstetricians and family practice physicians, anesthesiologists, pediatricians and perinatologists.
Variance in Cesarean rates is a major challenge in American health care with extraordinary complexity. States are now instituting initiatives to control healthcare spending primarily based on C-section rates, while risk management and insurers are concerned about the indicated section that was not performed or performed more than the ACOG standard requiring the intervention within 30 minutes after a C-section is called. In the United States, about 17% of births are by C-section, while much less in Europe but without worse outcomes. But clearly there are medical and financial benefits to a lower C-section rate if outcomes are not adversely affected. There are also powerful patient satisfaction and preferences that influence physician procedure selection. And primigravida (first delivery) and VBAC deliveries also factor into the variance complexity equation.
In a study published in AMIA Symposium Proceedings in 2003, machine learning tools were used to evaluate the C-section rates of different physician practices. The rate of a large hospital population of expectant mothers was 16.8%. However, the 17 physician group C-section rates varied widely from 13% to 23%. The question whether this variance was due to intrinsic “high-risk” patient sub-populations or differences in physician practice patterns, and the study concludes it was the latter.
The remarkable study published in Health Affairs (August 2014) by Laurent Glance, MD and his associates used the AHRQ database to identify low and high-performing hospitals based on rates of major obstetrical complications among US hospitals. Of the approximately 4 million women who give birth each year, nearly 13% experience one or more major complications. Using multivariate logistic regressions models to examine variance in obstetrical complication outcomes, they identified a five-fold difference among American hospitals. Women undergoing cesarean delivery at low-performing hospitals had nearly 5 times the rate of major complications (20.93%) compared to high-performing hospitals (4.37%).
Dr. Glance and his associates recommend development of comprehensive quality metrics for obstetrical care focused on improving obstetrical outcomes. Organized medicine responded: “The American Congress of Obstetricians and Gynecologists (ACOG) and the American Society of Anesthesiologists (ASA) agree that an important way to improve care is by increasing our understanding of the causes of complications and other negative outcomes… ACOG and ASA partnered to launch the Maternal Quality Improvement Program … to provide hospitals with the information they need to understand and improve their outcomes. The program will provide a framework for reporting performance and outcomes measures for obstetrical care. Although the article by Glance and colleagues provides important insight into obstetrical variations, the limitations in the survey demonstrate the need for clinical data to inform quality improvement efforts.”
At the request of ACOG and major university medical and law schools, Quture’s predecessor company co-sponsored with them the first Medico-Legal Dialogues in Obstetrics held in San Diego and New York. Quture’s expertise and experience in improving outcomes through it understanding of not only measuring outcomes but the root causes to achieve optimal clinical obstetrical outcomes using audit analytics now supported by big data and machine learning technology is unparalleled.
At a major hospital in Florida, one of our many external clinical obstetrical audits determined that the cesarean section rates for all methods and circumstances of deliveries were consistent with the national averages. Maternal and fetal complication rates were less than the expected national means.
Clinical performance problem identified with analytics
One significant pattern of variance existed, however. Although the Ob-Gyn department rates were within expected ranges, one physician had a much higher cesarean section rate than other members of the department. If the rates of the other physicians had not been significantly lower, the departmental rate would have been significantly higher. Furthermore, the maternal and fetal complication rates were significantly higher for this single outlier.
Pattern analytics demonstrated that cesarean sections for this physician, who lived farther than permitted by medical staff by-laws from the hospital, occurred just prior to dinner hour in almost 80% of the cases. Physician and patient convenience further demonstrated a dramatically high incidence of correlation between poor infant outcomes with clinically unindicated incidence of induction of labor with Pitocin.
Clinical performance problem solved with prescriptive knowledge solution
We instituted we instituted a rigorous, prospective clinical protocol that proscribed precise indications for induction of labor and conversion from vaginal and VBAC deliveries to C-sections. Without disciplinary action, these prospective patient safety protocols resulted in reduced C-section rates, especially those not indicated for provider convenience, and significantly improved infant outcomes. Since this hospital was part of a self-insured system trust for malpractice, we continued to monitor potential litigation and determined that the reduced C-section rate did not result in any spike in tort claims; indeed, they diminished from reduced complications.
We are not only an analytics company; we have extensive experience in conducting clinical audits in obstetrics to identify patterns and determine root causes. Where possible, we have successfully changed behavior with prospective clinical patient safety and quality management CLINICAL KNOWLEDGE SOLUTIONS.