THCB welcomes back regular contributor Dr. Eric Novack, who has something to say about outcomes as well as some recent snide comments made about orthopedic surgeons by a certain other poster on the site. In addition to blogging for THCB in his (oh so rare) free time, Eric is also the host of The Eric Novack show, which airs every Sunday on KKNT 960 AM in Phoenix. You can find an archive of his recent shows here.
An Outcomes Primer
Many in medicine view those of us in orthopedics as the ‘dumb bone doctors’
(or, according to the IV, much worse than that). Much of this stems from the basic idea that fracture care, or broken bone treatment, seems very straight-forward. Oh, but wait…
So here is a brief sense of how difficult it can be to evaluate outcomes even in the ‘simple’ area of a broken wrist. And, how it can be absurd to make the surgeon completely responsible?
The first question we need to ask is, “what outcome are we measuring”? Are
we going to look at (a) has the bone healed? (b) how ‘good’ does the
xray look- i.e. how close to ‘perfect’ are the bones lined up? (c) how
is the patient’s function, and at what point after injury do you
measure- months, years?
THCB is big on functional outcomes, so let’s just say that we care about wrist function 1 year after injury. But what kind of function? Range of motion? Return to work? Return to sports?
I’ll
make it easy and say we’ll leave that to the patient and simply ask
about satisfaction with ability to return to pre-injury functioning.
Stick
with me- I know we are looking at the easy area of a broken bone. So,
we are trying to determine functional outcomes 1 year after a wrist
fracture.
Here is one way to look at the factors impacting the outcome:
1. Patient
factors – age, motivation to get better, willingness to listen to
medical advice and follow recommendations, nutritional status, other
medical conditions, previous injuries, secondary gain issues (workers’
comp, lawsuits), body’s response to injury (i.e. inflammatory response
to trauma)
2. Injury factors—severity of injury force (e.g. trip
over dog vs. 60mph motorcycle crash), location of fracture (e.g.
involving joint cartilage), degree of displacement (i.e. how ‘bad’ the
xray looks), associated soft tissue injuries, associated injuries
impacting treatment and rehab decisions
3. Surgeon
factors—appropriate decision making, surgical technical skill,
doctor-patient communication (discussing injury, options, risks, and
expectations)
Rhetorically (and not), I ask- how much of the outcome can the surgeon possibly control?
The
answer, of course, is only the ‘surgeon factors’, which I will claim
generally make up a relatively small piece of the total outcome pie.
So
I say again (and again)- until I can get some converts… the future of
quality improvement lies not in just trying to identify ‘best
practices’ that can be difficult to prove and identify and can change
every few years—but rather in identifying what are the WRONG approaches
for conditions (much easier to get agreement here), and emphasizing the
importance of communicating appropriate expectations to patients.
figured out that the way to save the $2 trillion healthcare industry – it’s for people to not get sick by getting doctors out of medicine. After spending the last few years following doctors and radiologists around, visiting cancer centers and spending time watching mice get poked and prodded, I’ve realized it is time to embed the expertise of doctors in silicon and software. Why have radiologists read mammograms to find 1 in 200 that have breast cancer? Today, a third of mammograms now have their second read done by computer, computer aided detection from companies like R2 and iCad, and for $29, much less than a radiologist, and perhaps more accurate. For me, that’s just a start. But I was astounded to learn that CT scans are on the same learning curve as PCs and iPods and cell phones. One slice per rotation moved to 4 slice, 16 slice, 64 slice and soon 256 slice CT scanners. Instead of film, the output is a high res color 3D model. Beats a blood pressure reading and cholesterol number, which is all that physicians can manage. They are flying blind.So I started running the numbers. State of the art scans are still close to $1000. Say 1% of adults have heart attacks every year. A stent procedure runs about $15,000 just for the stent, with the hospital stay and bandaids, you are in for closer to $20-30,000, let alone lost wages and productivity. Heart scans today are around $1000. So if you screen 100 people, it costs $100,000, certainly more than treating the 1 in 100 heart attack patient. So,…, Blue Cross won’t pay for scans. It is better for them if nature does their screening for them, you or I actually having a heart attack – ding, ding, ding, we found our 1 in 100.They probably still wouldn’t pay if the scans were $500. But they might at $200. And they certainly would pay at $100, because it would be cheaper to screen than to pay for care. Because it is on the silicon learning curve (down 30% every year, 50% every two years), it is pretty easy to see $100 scans within five years, probably less. Heart attacks and stroke may become a thing of the past.And cancer, the third member of the Big Three in healthcare spending? Structural CT scans will transition to molecular imaging to find cancer early. I can see biomarkers on antibody chips that can eventually sell for $1 or maybe even 10 cents can detect unique cancer proteins in blood and flag cancer early enough for much cheaper treatment, beating symptoms by five years.Doctors can’t do that. In the end, I believe that Silicon Valley will do to doctors what ATMs did to tellers.
0.40; P values <.001) for all pairwise comparisons between 
-blocker use at admission and discharge, aspirin use at admission and discharge, and angiotensin-converting enzyme inhibitor use, and weaker, but statistically significant, correlations between these medication measures and smoking cessation counseling and time to reperfusion therapy measures (correlation coefficients <0.40; P values <.001). Some process measures were significantly correlated with risk-standardized, 30-day mortality rates (P values <.001) but together explained only 6.0% of hospital-level variation in risk-standardized, 30-day mortality rates for patients with AMI.
