[andy.lippman_422]

No but I’ve always thought that would be a great idea. Like some kind of natural language parsing that would seek to estimate the complexity of a report. A CT AP with “Acute uncomplicated appendicitis.” should not be worth the same RVU as “Mixed disease response with decrease of blah blah and increase of blah blah”

[mario.mtz30_447]

Have thought similar.  Software without or with image analysis.  Could use many variables, modality, pt age, # priors, length of prior reports, time on list, etc etc and assign a real time value of work.  

[22002469]

Most reasonable progressive groups with “RVU” based pay do not use RVUs, but rather some internal version of effort based work value units. Long term it’s impossible to go strictly by RVUs in a subspecialized practice without a revolt. 
 
Obviously the goal is to make every case fairly compensated and therefore eliminate cherry picking. Of course this requires a granular assessment… the CT appy shouldn’t = cancer staging and 20 year old low back pain shouldn’t = 80 year old post op spine.
 
While that is the holy grail, most rads would settle for decreasing mammo/US and increasing XR to make things kind of resemble reality. 

[ruszja]

Quote from Radsoxfan

Obviously the goal is to make every case fairly compensated and therefore eliminate cherry picking. Of course this requires a granular assessment… the CT appy shouldn’t = cancer staging and 20 year old low back pain shouldn’t = 80 year old post op spine

Is anyone adjusting by complexity within a particular CPT ? Unless you go by a metric of ‘CPT + referrer’ and up the RVU for studies from certain surgeons and oncologists, I don’t see how this is gonna work in an automated way.

The only thing I have seen is where groups have decided to adjust RVUs by CPT and reduced the stuff that is overweight: screening mammo, head CT, MSK MR.

[22002469]

Quote from fw

Is anyone adjusting by complexity within a particular CPT ? Unless you go by a metric of ‘CPT + referrer’ and up the RVU for studies from certain surgeons and oncologists, I don’t see how this is gonna work in an automated way.

The only thing I have seen is where groups have decided to adjust RVUs by CPT and reduced the stuff that is overweight: screening mammo, head CT, MSK MR.

 
Not that I know of, I don’t think it’s really possible to automate to that level of granularity. It’s just that theoretically that’s what you would have to do to make each case properly valued.
 
Most places settle for some semi-rational RVU adjustment based on CPT code to bring things closer to reality, even if still far from perfect. 

[phillipsjada85_337]

Don’t think it would work well. I’ve known many a non productive rads who dictate huge reports for near normal ultrasounds. 
 
I think as long as people aren’t cherry picking, complexity variation evens out over time within a specific setting (ER vs inpatient vs outpatient imaging center). 
 
You could try measuring productivity by the type of shift each rad does, assuming there are fixed expectations for each type of shift (body inpatient, ER, outpatient msk, mammo slot, etc.)

[22002469]

Things only even out in small to medium sized groups with everyone rotating through the same shifts (and a lot of non-subspecialty reading). Then yeah, if you cut down on cherry picking, it doesn’t really matter. 
 
For larger sub specialized groups, things won’t ever even out without some sort of background adjustment.