More Hospital Access Does Not Equal More Health

Posted on by

Nathan Petek writes,

I construct a hospital-level panel using American Hospital Association data from 1982-2010 that includes measures of the quantity of health care and indicators for hospital entry and exit. I combine this file with county-level measures of health, including mortality rates from 1982-2010 and self-reported health from 2002-2010, and detailed health care utilization and mortality data for all Medicare fee-for-service enrollees from 1999-2011.

He finds that having more hospitals in an area does increase the utilization of medical resources. It does not improve health outcomes. Like all studies, this one by itself is not convincing. But it does add to the huge pile of studies that go into the Hansonian medicine file.

Pointer from Tyler Cowen.

5 thoughts on “More Hospital Access Does Not Equal More Health”

  1. I think the conclusions expressed in the last two posts on healthcare just don’t make sense, especially when taken together.

    It seems odd to push back on the notion that having more than one hospital competitor in a given area is of any health outcome benefit. Why question the value of competition among hospitals, but push the idea of HSA’s as a way to introduce market forces into healthcare? If healthcare is mostly BS, then why would HSA’s help matters?

    • Probably when money is free they aren’t competing on price. So, they may be competing on facilities or unnecessary services. It is hard for a customer to opt out if it is free and the expert who does it is the expert that diagnoses it.

  2. I read about this issue in Overtreated.

    It’s very interesting, but one thing I wonder is how much benefit might simply be going unmeasured. A nicer waiting room, for example, might not show up in any measurable statistic.

    It’s a strong intuition that choice leads to competition and improvement. Why would that not be the case for hospitals?

    • No question, American medical facilities resemble luxury hotels, and you did not get that in the past, nor do you get that in other countries.

  3. Given US legislation passed in 2008, data from 2010-2015 is probably highly relevant given changes in patient usage patterns. I.e not going to DR at all, high deductibles, etc.

Comments are closed.