Looking at data and just presenting it in a scientific session is not enough, said Woojin Kim MD, University of Pennsylvania, speaking at the Radiological Society of North America (RSNA) annual meeting in Chicago this week. “We need to convert that data, get insights from it, and be able to act on it. If you cannot get to the action phase, it’s not working well.”

Kim opened his presentation on quality metrics with a quote from Teri Yates: ”Developing quality metrics will be a crucial task for radiology groups - now is the time to start proactively tracking quality metrics”. Kim noted that not a week goes by without hearing how radiology’s payment model will change from fee for service to value-based - the American College of Radiology’s Imaging 3.0 proposition. “We are at the tip of the iceberg”, he said, as radiology moves from interpretation-focused to outcome-focused measures.

The first step in quality metrics is to define organisational goals and desired outcomes. Then radiologists need to ask the right questions about the processes needed to measure success.

Kim observed that as healthcare IT is about 15-20 years behind the curve, radiology needs to learn from how business measures quality. Business looks at defect ratios, customer satisfaction and cost indices. In radiology defect ratio relates to quality safety and efficiency. He commended the article by Boland et al. (JACR 2014) on delivery of appropriateness, quality, safety, efficiency and patient satisfaction, which puts data mining and business intelligence at the centre of the imaging value chain.

Measuring customer satisfaction may show good figures, but where’s the insight? Kim showed a detailed spreadsheet showing customer satisfaction in percentages, all scoring over 90%. But where’s the insight? You cannot really act on it. Sentiment analysis is a good tool to measure. People are used to reviewing with travel sites, for example. What do people think about their radiology service? Kim has an alert on Penn Radiology. Based on a single review of a patient’s experience of the US service, they were able to act and make scheduling improvements.  

Cost indices are also important to consider. For example, by analysing CT biopsy recommendations on inpatients, one study reduced length of stay by three days. As the average cost of a day’s stay in hospital average $2000 that translates into potential savings of $360m per year.

Kim’s parting advice? Start with the question, not the data. Define your own quality metrics, and take an agile approach, which emphasises creating small deliverables quickly.

Claire Pillar
Managing Editor, HealthManagement



Latest Articles

Radiology, Quality, RSNA 2014, Quality measure, #RSNA14 Looking at data and just presenting it in a scientific session is not enough, said Woojin Kim MD, University of Pennsylvania, speaking at the Radiological...