Dr. Ana Lourenco and her colleagues who worked in both hospitals and private practices noticed something odd one day when informally comparing notes: they seemed to be calling back more patients at the former than the latter.

Sure enough, this hunch was substantiated when Lourenco, radiologist at the Rhode Island Hospital and the School of Medicine at Brown University, and study co-authors decided to follow it up with hard data. The recall rate for hospitals was 8.6 percent while the private practice rate was only 6.9 percent.

The study authors analyzed data on over 74,000 mammograms performed in private practices and hospitals, and reviewed by five radiologists between May 2008 and September 2011.
Several factors may account for this difference in recall rates. For one, many more hospital patients had undergone previous biopsies and surgery than patients at the private practice.

Furthermore, hospital patients were lower in average age than private practice patients, a factor associated with higher recall rates.

Interestingly enough, this marked disparity persisted even on an individual level with each radiologist — meaning the same radiologist working in different sites had significantly variable recall rates due to factors out of their control.

"This gives more credence to the variability — it held up even when you broke it down by each of the five radiologists who read at both sites," said Lourenco.

Recall rates are one of the most commonly used metrics to measure quality in mammography, but this study shows these rates may vary because of factors that have nothing to do with the quality of the radiologist's interpretation.

"Cancer detection rate one would argue would be a better metric, but it's more difficult to track down," said Lourenco.

More Industry Headlines