Statistical Testing
 
 

Statistical Testing

Statistical Testing

We used a widely-accepted statistical formula1 to calculate the 95% confidence level required to perform a one-proportion, two-tailed hypothesis test to determine if our hospitals’ scores for each quality measure were statistically the same or statistically different from the U.S. average.

  • A score that was significantly higher than the U.S. average received a "Green" designation.
  • A score that was significantly lower than the U.S. average received a "Red" designation.
  • A score that was not significantly different from the U.S. average received a "Beige" designation.

We obtained the data used in our calculations from the Centers for Medicare & Medicaid Services (CMS) Web site. We downloaded the scores and the denominators for each CMS process of care measure for each BayCare hospital. The denominator is the total number of patients who met the criteria for inclusion in the measure calculation. From there, we calculated the numerator (the number of patients who actually received the recommended care) for each measure. Then, we entered the information into the statistical formula to apply tests of statistical significance.



1. Breyfogle, F.W. (1999). Implementing Six Sigma - Smarter Solutions Using Statistical Methods. New York, N.Y.: John Wiley & Sons, Inc.