News

“Mind boggling” variation in hospital prices is seen across US

BMJ 2015; 351 doi: https://doi.org/10.1136/bmj.h6851 (Published 16 December 2015) Cite this as: BMJ 2015;351:h6851
  1. Michael McCarthy
  1. 1Seattle

Largely owing to differences in hospital prices, total healthcare spending on beneficiaries covered by private insurance in the United States varies as much as threefold between regions, a large study of employer sponsored insurance claims has found, and prices for routine services often vary by a factor of more than 10.1

Healthcare spending was lowest in Honolulu, Hawaii, the researchers found, where only $1707.39 (£1140; €1560) was spent on each privately insured beneficiary in 2011. But spending was highest in Napa, California, at $5515.90 per beneficiary. Prices for routine services also varied widely between and even within regions. The price hospitals charged for routine magnetic resonance imaging (MRI) scans of lower limb joints, for instance, varied by a factor of six within the city of Philadelphia, Pennsylvania, and it varied by a factor of 12 across the nation, the study found.

“The fact that prices are so high and can vary so much for hospital treatments of the same costs and quality is simply mind boggling to foreign observers of the US healthcare system,” said John Van Reenen, professor of economics and director of the Centre for Economic Performance at the UK London School of Economics, one of the study’s authors. “This is surely one of the reasons why US healthcare absorbs a bigger share of GDP [gross domestic product] than every other large advanced country.”

In the study the researchers analyzed 92 billion insurance claims of 88 million people who had employer sponsored health insurance during 2007-11 from the three largest US health insurers: Aetna, Humana, and UnitedHealth. The database covered the claims of nearly 30% of US people with employer sponsored coverage and accounted for about 5% of total US healthcare spending, or 1% of its annual GDP.

Although US healthcare policy is often guided by studies of claims data from Medicare—the federal insurance plan for elderly and disabled people—the researchers found that Medicare spending did not reflect spending by private insurers. In fact, in many regions where Medicare spending is low, spending by private insurers is much higher.

For instance, La Crosse in Wisconsin, Rochester in Minnesota, and Grand Junction in Colorado have all been identified by policy makers as models for best practice, on the basis of their low Medicare spending.

But the study found that La Crosse, which had the lowest total spending on each Medicare beneficiary in the nation, had the 22nd highest spending on each privately insured beneficiary; Rochester, 14th lowest in Medicare spending, was 11th highest for privately insured beneficiaries; and Grand Junction, third lowest for Medicare spending, had the 43rd highest spending among privately insured beneficiaries.

The lead author, Zack Cooper, of Yale University in New Haven, Connecticut, said, “Virtually everything we know about health spending and most of the basis for federal health policy comes from the analysis of Medicare data. The rub is that Medicare only covers 16% of the population. The majority of individuals—60% of the US population—receive healthcare coverage from private insurers.”

The researchers found that price differences were primarily driven by individual hospitals’ market power. Hospitals in monopoly markets set prices that were over 15% higher than those in markets with four or more hospitals.

“While we cannot infer causality, these associations do suggest that, consistent with economic theory, hospital market concentration is strongly related to hospital prices,” the researchers wrote. “In terms of policy, our work suggests that vigorous antitrust enforcement is important and that hospital prices could be made more transparent.”

The data were provided by the Health Care Cost Institute, a non-partisan, non-profit organization based in Washington, DC that focuses on US health spending. The analysis was part of the Health Care Pricing Project, a research initiative funded by the Commonwealth Fund, the National Institute for Health Care Management Foundation, and the Economic and Social Research Council.

Notes

Cite this as: BMJ 2015;351:h6851

References

View Abstract

Sign in

Log in through your institution

Subscribe