Originally posted July 13, 2013 on Pulse
By Ulfat Shaikh
It is likely that by now many of you have read your way through Steven Brill’s tome in Time Magazine, Bitter Pill: Why Medical Bills Are Killing Us. This 36-page article is apparently the longest article ever published by a single author by Time.
Brill quotes Johns Hopkins health economist Gerard Anderson – “All the prices are too damn high.” He highlights the nefarious chargemaster (a hospital’s listing of charges), which may give a indication of medical costs, but is far from what the hospital is actually reimbursed by federal or private insurers. Perhaps this statement needs to be fine-tuned to, “all the prices are just too damn bizarre.”
Ever since then, health care costs have been in the media limelight.
The Centers for Medicare and Medicaid Services recently released billing data from over 3,000 hospitals in the US for 100 common treatments and procedures. These procedures included joint replacements, cardiac stents and gallbladder removal. These data show widely varying prices that hospitals charge for the same procedure, not just regionally but also within hospitals in a single city.
Part of these differences may be due to patients being sicker or older in a certain geographic area, or a hospital treating a disproportionally larger number of sicker patients with medical complications. The goal of publishing this information is to make it more accessible and transparent. However, it is far from clear what the utility of this information is, especially given the fact that many hospitals set these charges based not on actual costs or payments but based on what insurers cover.
So if these charges are in fact so distorted where does one go from here. As quality guru W. Edwards Deming once said, “If I could reduce my message to just a few words, I’d say it all has to do with reducing variation.” For example the price of treating a breathing problem that requires a ventilator could vary by $200,000 in Los Angeles depending on which hospital you look up. Even if these charges are distorted, the fact that they vary so extremely offers an insight into the degree of practice variation, even within a single city.
The Centers for Medicare and Medicaid’s online billing data can only take you so far since it does not cover procedures such as delivering a baby. A New York Times article published a week ago highlights the experiences of a pregnant woman who set out to determine the comparative costs of a normal delivery at two hospitals in New Jersey.
To quote Deming once again, “Every system is perfectly designed to achieve the results it gets.” Unless payment models are aligned to ensure that value is rewarded over volume, billing data our health care system generates are perfectly designed to result in confusing and bizarre side-effects that could keep you up at night.
–Ulfat Shaikh, MD, MPH, MS is director of health care quality at the University of California Davis School of Medicine. She blogs about health care quality improvement at Pulse.