The recurring narrative among health reformers is that hospitals that provide more care raise health costs, but don’t necessarily improve quality. This has lead to a backlash against so-called “aggressive” hospitals and doctors, with upcoming financial penalties to match. But the situation, as always, appears to be more nuanced than that.
found no difference in the rate of complications for aggressive and nonaggressive hospitals. But when they looked at all the patients who had complications and examined their outcomes, the researchers found that regardless of the urgency of their operations, those patients who were cared for at more aggressive hospitals were significantly more likely to survive their complications than those who had their operations at less aggressive hospitals.
In addition, the investigators found that characteristics associated with intensity of care treated surgical complications better:
… a hospital’s failure or success in treating surgical complications correlated consistently with factors that also characterized intensity of care — general expenditures, intensive care unit use and the total days of hospitalization — they found that benefits of this more aggressive care extended well beyond the time of the operation.
I constantly remind readers of this blog that more medicine isn’t necessarily better. The counter-intuitive findings from the Dartmouth Atlas study have been instructive in convincing patients that they are, in many cases, overtreated.
But we must be careful to ensure the pendulum doesn’t swing too far. Admonishing hospitals and doctors simply because they are one end of the care spectrum may unfairly punish some who benefit patients. It’s not black and white.
There will be exceptions showing that aggressive — and, yes, more expensive — care is better. Highlighting these cases will help balance the narrative, and hopefully help the public understand the nuance that’s needed when discussing solutions to fixing our health system.
*This blog post was originally published at KevinMD.com*