Comparing Mammography Visits in High and Low Repeat Rate Facilities

• Activity Number: 360
• Type: Contributed
• Date/Time: Tuesday, August 3, 2010 : 10:30 AM to 12:20 PM
• Sponsor: Biometrics Section
• Author(s): Jonathan D. Mahnken*+ and Jianghua (Wendy) He and Hung-Wen Yeh and Niaman Nazir and Linda L. Jianas and Kimberly K. Engelman
• Companies: The University of Kansas Medical Center and The University of Kansas Medical Center and The University of Kansas Medical Center and The University of Kansas Medical Center and The University of Kansas Medical Center and The University of Kansas Medical Center
• Keywords: Beta distribution ; Censored data ; Likelihood ratio test ; Proportion ; Survival analysis

Abstract

We collected data to compare the proportion of total mammography visit time spent conducting the exam between facilities with high rates of annual repeat visits to those with low rates. The primary hypothesis was that the means for these proportions were equal between facility types. Due to the limited range of the observed measures, we used the beta distribution to conduct our analysis; however, some of the observations were censored. Using the likelihood ratio test, we rejected the null hypothesis of equal means between low and high repeat rate facilities (p=0.0003). The mean proportion of visit time spent performing the exam was approximately 60% higher in the facilities with high repeat mammography rates (about 70% of the total visit time) compared to those with low repeat rates (about 45% of the total visit time).