By Dr. Bradley Flansbaum

First Posted at The Hospitalist Leader on 3/24/2013

Bradley Flansbaum, Co-host of The Hospitalist Leader

Bradley Flansbaum, Co-host of The Hospitalist Leader

AHRQ released a helpful brief in 2013 assessing costs (not prices) for hospital stays.

Inpatient health care costs for Medicare, longer term, exhibit a downward growth rate–still ascending but slowing.  However, prices for employed individuals moved upward and at an accelerating rate (distinct from Medicare).  The number of admissions from both groups also decreased.

Medicare, a national government program, uses administrative pricing and calls the shots. No negotiations, and their methods for determining price reflect their assessment of input costs.  Commercial or employer-sponsored insurance–what we have at work–reflects a more ambiguous methodology.

Patients (or insurers) often pay far more than the actual costs of care (price discrimination). Think retail versus wholesale, and rate setting at the commercial level reflects a more market-oriented approach to contracting.  The stratospheric prices noted in the Time magazine Brill piece serve as an example of what the non-Medicare market resembles.

HCUP to the rescue.

The non-sequester threatened database uses a Federal-State-Industry partnership and assembles data from state organizations, hospital associations, private data organizations, and the Federal government creating a clearinghouse of patient-level health care data.

Essentially, the collective takes the family receipts from mom, dad, junior, and grandma and makes them more practicable.  Unfortunately, non-government data don’t grow on trees (proprietary) and as a result, HCUP serves a useful purpose.  We obtain a rare sliver of all payer intelligence.

Some excerpts:

1.  Costs incurred by payer related to number of stays.  More balanced than I expected, with a slight Medicare tilt (8%):

2.  Cost per stay broken out by payer.  Notice Medicaid, a group about to join the ACA’s health care marketplaces:

3. Costs by age.  Notice kids don’t cost, but curiously 85+ are less than average as well.  Not intuitive, but perhaps they get less intense inpatient rescue care or pass away sooner.

4.  Annual percentage change in cost by diagnosis.  Notice sepsis (? ICU) and arthritis (back pain). Only diagnosis to decrease?  CAD (clunky stat, ? less non-invasive). Click to enlarge:

Something to keep in mind.  When you hear about having skin in the game (bearing some or all of your health care costs to limit unnecessary resource use), look at the prices above and think about an average household salary.  Then think about the odds of a household member landing in the hospital.

Older age does focus the mind, but youthful adults grow up fast when out of pocket, coinsurance, and copay stubs enter the mailbox.  As providers, we are sometimes blind to the costs until its us or one of ours.  Stuff happens, and remember, what you see above reflects just the costs, not what you might pay–applying doubly so for age <65.

Hospital inpatient care:

  • Number of discharges: 36.1 million (US population: 315M)
  • Discharges per 10,000 population: 1,181.2
  • Average length of stay in days: 4.9


PS--”Economic statistics are like a bikini, what they reveal is important, what they conceal is vital”

—Professor Sir Frank Holmes, Victoria University, Wellington, New Zealand, 1967