Point-of-care Clinical Decision Support Reduces CT Scan Rates

We know that adding clinical decision support at the point of care improves knowledge translation and increases the use of good medical evidence.  What’s hard is measuring how much of an improvement there is.

During rounds yesterday, an emergency physician did a quick chart audit to check positive rates from CTPAs for suspected pulmonary embolism.  What he found was surprising.

Physicians who decided to order a CT Scan while using point-of-care clinical decision support had a diagnostic positive rate of 25%; those who didn’t use clinical decision support had a diagnostic positive rate of 8%.

This suggests that for pulmonary embolisms, point-of-care clinical decision support reduced unnecessary testing by over 300%!

In light of the above, this site has decided to add the 2-level Well’s score, PERC score and a reminder to use an age adjusted range for D-Dimer to its “Chest Pain – Non-Cardiac” and Respiratory Distress – No Fever” charts.

Now, this wasn’t a rigorous study and the sample size was small and not controlled (9 months of orders resulting in 150 scans), so the results should be viewed conservatively.  But as a mental exercise, what could these numbers mean?

Scenario #1: A Medium-sized ED

Imagine you are a medium sized emergency department without clinical decision support.  Every year, your emergency physicians order 200 CT Scans for possible pulmonary embolisms.

Out of these 200 CT Scans, 8% are positive.  This means there are 16 patients with pulmonary embolisms that are diagnosed through CT.

Now imagine you could increase your positive rate to 25%, meaning your physicians are less likely to order an unnecessary CT Scan (and we know that the sensitivity of this decision making is 100%).  With a 25% positive rate, only 64 CT Scans are required to correctly identify those 16 pulmonary embolisms.

Our medium sized emergency department just eliminated 136 unnecessary CT Scans!  At $1,000 a scan, our hospital just saved $136,000!

Scenario #2: A Small ED

For a small hospital, reducing unnecessary CT Scans for pulmonary embolisms would take savings a step further.  Sure, a small hospital might only order 60 CT Scans per year looking for possible pulmonary embolisms, but each of these patients must be transferred to a larger facility.

Using the same logic as above, this emergency department is correct in transferring 5 of these patients; the rest of the patients did not have pulmonary embolisms and their transfer was precautionary only.

By adding clinical decision support, this small hospital need only transfer 19 patients to catch those same 5 pulmonary embolisms. This means they need to transfer 41 fewer patients every year.

Implications for Your Hospital

Again, these numbers are not the result of a rigorous study.  Rather, they were a point of interest from a chart audit presented at weekly rounds.  Every hospital will have different diagnostic positive rates for pulmonary embolisms, and not all will experience 300% reductions in rates by adding clinical decision support at the point of care.

Instead of debating whether these findings are statistically significant or representative of all hospitals, consider three questions:

  1. What is your hospital’s diagnostic positive rate for pulmonary embolisms?
  2. At this rate, how many unnecessary CT Scans are ordered every year?
  3. How big of a reduction in unnecessary CT Scans would have an impact on your hospital’s bottom line or CT Scan wait times?


Comments are closed.