Dr. Gawande, Please Open the Can of Worms
Dr. Gawande, Please Open the Can of Worms
Congratulations, Dr. Atul Gawande. On July 9, 2018, you will start leading the Amazon (NASDAQ: AMZN), Berkshire Hathaway (NYSE: BRK.A, BRK.B) and JPMorgan Chase & Co. (NYSE: JPM) partnership “to address healthcare for their U.S. employees, with the aim of improving employee satisfaction and reducing costs.” What follows is a polite request:
Please, open the can of worms
When announcing the new partnership, Warren Buffett, Chairman and CEO of Berkshire Hathaway, said, “The ballooning costs of healthcare act as a hungry tapeworm on the American economy.” I have bad news—it’s not just one tapeworm. Patients and non-patients are different. Start with patients – that’s where the costs are. If the 1.1 million U.S. employees of Amazon, Berkshire Hathaway, and JPMorgan Chase are like a typical subset of Americans, about half have annual healthcare costs of less than $300. So half of these employees are essentially non-patients. Nationally, according to the Peterson-Kaiser Health System Tracker, the top 5% of people in health spending have costs averaging $51,000 annually; the top 1% have costs averaging $112,000 annually. Instead of thinking of employee averages, bifurcating patients and non-patients should help ABC pursue different strategies for different categories of employees.
Spending variations seem obvious. You, Dr. Gawande, have been writing about the issue for years. Treating a rare disease is more expensive than treating a sinus infection. But that’s just one “tapeworm.” What is well known (but often unsaid) should be tackled directly by ABC: The U.S. healthcare system pays for incorrect diagnoses and treatments; it pays for treatments evidence shows don’t work; it rarely differentiates payment between mediocre and superb health practitioners.
We need to stop paying for what doesn’t work
The system tolerates (and pays dearly for) incorrect diagnoses and incorrect treatment. ABC shouldn’t. Errors in diagnosis occur about 10 to 15% of the time, though errors vary widely by disease or condition and specialty area of the physician. A study of second opinions provided at The Mayo Clinic in Minnesota found that in only 12% of cases was the referral diagnosis the same as the final diagnosis. In other words, more than 85% of the time the first diagnosis was wrong, or needed to be refined.
Medical errors, including incorrect diagnoses, treatments, and mistakes, kill between 44,000 and 250,000 people a year (which would make it the third leading cause of death). That is a big range, because medical errors statistics are controversial. Since death certificates don’t list cause of death as “medical error,” researchers have to estimate. When researchers at Johns Hopkins Medicine reviewed 25 years of U.S. malpractice claim payouts, they found “diagnostic errors — not surgical mistakes or medication overdoses — accounted for the largest fraction of claims, the most severe patient harm, and the highest total of penalty payouts.” Health systems analysts can quibble about the numbers, but ABC should strive to lower medical errors. Study lead, David E. Newman-Toker, M.D., Ph.D., associate professor of neurology at the Johns Hopkins University School of Medicine, says, “Experts have often downplayed the scope of diagnostic errors not because they were unaware of the problem, but ‘because they were afraid to open up a can of worms they couldn’t close.’”
To best serve employees who are patients, ABC needs to open up this can of worms. ABC should collect, verify, and use data to determine how employees can be better served by the healthcare system. Which doctors have the lowest error rates? Where is care delivered with the best outcomes for a particular disease or condition, or for a particular type of patient? Continuing to provide employer-sponsored health insurance without taking additional steps to help patients and their families figure out where they can get the best care is something akin to employer malpractice. This approach also perpetuates the idea that there is little difference in quality or outcomes at different health facilities, which is demonstrably false.
For over 20 years, the Dartmouth Atlas Project has been documenting the stark variations in medical use across the U.S. For example, the Atlas shows discretionary surgery, such as knee or hip replacement or back surgery, varies significantly by region, in part because “physician opinion can vary widely as to when the treatment is necessary, and which patients are appropriate.” Diagnosis and treatment differences across practitioners have varying causes. Grand Rounds, for example, was started in 2011 in recognition of the gap between first and final diagnoses and treatment recommendations. As co-founder Owen Tripp (who was a candidate for this job…) explained in an interview with Bruce Rogers for Forbes.com, he was shocked by the realization it takes about 17 years for a proven technology or approach in an academic setting to “make its way to the majority of doctors who practice in the field,” while in the software world it only takes about 18 months between discovery and adoption. Connecting patients to information about which healthcare providers perform best is an easy step for ABC to take (if they haven’t already). If the data exists, ABC should use it.
Open up the can, differentiate the tapeworms by type, and make decisions that will benefit employees accordingly. We can’t starve the tapeworms unless we start to open up these cans. We are all looking forward to you taking the first crack.