Transforming Data into Knowledge: How Sarah Cannon Analyzes Risk-Adjusted and Unadjusted Survival Outcome Trends and Projections
By Therese Dodd, BA, MBA, RN, CPHQ, FNAHQ
Quality and Accreditation/Regulatory Compliance Specialist
Transplant & Cellular Therapy Network
1100 Dr. Martin L. King, Jr. Blvd, Suite 800, Nashville, TN 37203
The centers of Sarah Cannon Transplant & Cellular Therapy Network (SCTCTN) are not alone in their need to understand clinical outcomes, the definitive outcome being patient survival. It falls to the centers to transform an abundance of data into knowledge for performance improvement.
The Center for International Blood & Marrow Transplant Research (CIBMTR®) Transplant Center-Specific Survival Report provides survival data to eligible hematopoietic cell therapy (HCT) centers for first allogeneic transplants for a three-year period to at the end of each calendar year. The report displays a “–1” in the performance status column for centers whose actual survival is lower than the 95% confidence limits for predicted survival, a “1” if performing above the confidence limit, and a “0” if performing as predicted. The SCTCTN centers recognized that reading the report is just a first step. One challenge to understanding the survival results is that these reports only display the information from preceding years as a table (see example in Figure 1). Supplying information to centers in this manner may provide them a false sense of confidence if they do not trend the specific outcomes results. In the figure 1 example, the “-1” assigned in the 2015 report may have been an unpleasant surprise if the center wasn’t rigorously analyzing its clinical outcomes internally.
The SCTCTN centers have created box graphs, similar to the Excel template available on the FACT website in the Clinical Outcomes Resource Center, to track the clinical outcomes from the CIBMTR® Transplant Center-Specific Survival Report . As can be seen in Figure 2, there was a decline in actual % 1 year survival for transplants performed during 2010-12, dropping from greater than the % predicted survival in the previous report to lower than % predicted survival and approaching the lower limit of the 95% confidence interval.
With the 2011-13 report, the 95% confidence interval increased, which indicates lower-risk patients were transplanted. Since the report is risk-adjusted, there was an expectation that patient survival would increase; however, the % actual survival was worse than the lower limit of the 95% confidence interval. Fortunately, the center had already started root cause analysis and identified several opportunities for improvement including, but not limited to, increasing data abstraction and entry accuracy, strengthening survivorship programs, and expanding caregiver support services. The center was able to improve its clinical outcomes by the time of the subsequent report.
In addition to tracking data from the CIBMTR® Transplant Center-Specific Survival Report, the SCTCTN have overlaid a trend line to the box graphs (see black dotted line in Figure 2) of actual % survival using data from the CIBMTR® enhanced data back to centers (eDBtC). Based on experience, and not surprisingly, SCTCTN centers determined that the eDBtC % actual survival for first allogeneic transplants tracked closely to the CIBMTR® report data year over year. Because of this, the centers can determine an estimate that is prognostic for the actual % survival in the upcoming report by finding the % survival at 12 months using the CIBMTR® eDBtC “Survival Probability – HCT” functionality (see Figure 3).
In addition to annual analysis of the CIBMTR® Transplant Center-Specific Survival Report, the SCTCTN centers monitor actual autologous and allogeneic hematopoietic cell transplant (HCT) % 30-day, 100-day, and 1-year survival and treatment-related mortality in smaller timeframe cohorts on a quarterly basis and more concurrently. Tracking these data (as well as many other quality and outcome metrics) supports rigorous quality management programs that improve the experience for SCTCTN patients.