Today’s Managing Health Care Costs Number is 88 million
 |
| Source |
Most information on variation in cost of care has been derived from Medicare data. That’s good – because Medicare beneficiaries are older and sicker and have a lot of claims, and it’s good because the Centers for Medicare and Medicaid Services (CMS) makes these deidentified claims available to researchers. It’s bad because CMS dictates prices –and so total cost of care is highly influenced by utilization and past Congressional actions that distorted prices. It’s not influenced at all by the prices charged by hospitals and physicians to privately insured individuals, where there is no national fee schedule in place.
We’ve known for a long time that Medicare cost of care and commercial health plan cost of care are not always proportional. Communities with a single hospital get the (often low) Medicare price, but they have the leverage to extract much higher prices from commercial insurers who cannot sell insurance without that hospital’s participation.
The New York Times offered up a blockbuster report this morning on research showing that for those insured by private (employer sponsored) health plans, costs per unit are much higher than Medicare, and it’s these prices that are driving health care costs.
The Health Care Cost Institute, a collaboration of Aetna, United Health Care, and Humana, has made data available for care delivered to over 88 million Americans (almost 3 billion claims) from 2007-2011. Researchers used this data to show that Medicare facility fees were dramatically lower than costs paid by the commercial plans – generally over 40% less. They also showed huge variations from hospital to hospital in the same market. Costs of knee replacements in Atlanta and knee or hip MRIs in Columbus Ohio were over six times higher at the most expensive facility compared to the least expensive.
The Dartmouth Atlas, and seminal articles like Atul Gawande’s Cost Conundrum, showed us that there is substantial variation in total cost of care in Medicare – largely due to utilization differences and some minor price differences. The Institute of Medicine noted in 2011 that some of the variation in these analyses could be explained by severity of illness. The current Health Care Cost Institute research demonstrates that we cannot extrapolate from the Medicare population to the commercial population. High prices underlie our high total cost of health care for the population insured by employer- sponsored health insurance. We must address prices paid by employer-sponsored health plans to achieve affordable health care.
 |
| Source |
You can see that there is no relationship between high or low Medicare and private insurance.
One interesting journalistic element of this NYTimes reporting: The web version detects where the reader is - and presents data for the applicable geography. On my home computer, I got Boston data (showing very large disparities in cost of total knee replacement), while on my work computer, I got Arlington VA data (based on my company's network connection). There is little facility fee price difference in Arlington. I find this personalization helpful - and not creepy -but it is a reminder that the web sites we visit know where we are!
