HIS Insider Online

August 8, 2005
EHRs a Pipe Dream in Short Run, Research Shows

The way things are going, just 29 small American hospitals (<300 beds) will have an electronic health record five years from now, according to a revealing study just published in the Annals of Internal Medicine. Add in larger facilities (<300 beds) with EHR, and the number grows to 70, or roughly 1.4% of existing hospitals nationwide.

Granted, numbers in “The Costs of a National Health Information Network” are estimates based on “the existing status of IT within the health system.” And its authors hail from the Center for Information Technology, a research group founded by Partners HealthCare System (10 hospitals, 2,500 beds) in Boston, where CIO John Glaser not only is a strong supporter of White House HIS czar David Brailer but also heads the MA-Share RHIO. 

Other study headliners include Janet Marchibroda of eHealth Initiative and Dr. William Yasnoff, who works for HHS, both heavily invested in the Brailer initiative.

…& Costs Would be Astronomical Under Current Scenarios

The study is based on specific functionality-based technology: EHRs, computerized physician order-entry, electronic prescribing, online viewing of test results, patient-provider e-mail, electronic claims processing and eligibility verification.

“Most of this [EHR vision] is fiction,” says Dan Morreale, CIO at Jacobi Medical Center (2 hospitals, 823 beds) in the Bronx, N.Y. “There are no standards for interoperability which are universally accepted; hence I don’t think it will happen within 5 years. The cost issue is not real since the mandate in unfunded. And without an interoperability design you cannot know how much it will cost. I’ll take this seriously when agreed standards are in place and the financial incentive is there to make a change.”

One thing study authors and Brailer detractors seem to agree on, though, is the vast gap between current HIT reality and 2014 EHR/interoperability goals set by President Bush a year ago.

Getting from “here” to “there” in the next five years requires a capital investment of $156 billion plus $48 billion in annual operating costs. In the study’s scenario, hospitals will wind up paying about two-thirds of that – the cost of adding electronic functionalities such as EHR and CPOE - while physician practices will fund interoperability.

The $156 billion “seems like an awfully high number,” says Dr. Kerry Stratford, a St. George, Utah, family doctor who’s also vice chairman of a RHIO called the Utah Health Information Network. His then 8-doctor practice (now nine), bought eClinicalWorks for $40,000 (excluding hardware). So he says study results may be based on more expensive products, noting his group got bids from Next Gen and Allscripts,  “which cost three to four times more.”

“You don’t have to pay that much to provide quality care and do what the government wants,” he adds. Stratford also questions how study authors parcel cost responsibility, arguing that if 80% of EHR/interoperability benefits come from clinical decision-making (as the study states), docs should shoulder a smaller part of the costs.

“It’s a lot of money,” concedes study chief Dr. Rainu Kaushal of Brigham and Women’s Hospital (716 beds) in Boston. Actual costs could be less, partly because study results are based on existing Stark limits. There are nine HIT bills before Congress, but the study modeled how much it would cost the nation to add several IT functions and adopt interoperability based on the Santa Barbara (Calif.) County Care Data Exchange, a peer-to-peer network and among the earliest RHIOs, she says.

But the Price Tag Depends on Structure: Access or Exchange?

Since RHIO structures vary, and it’s still too early to predict which RHIO structure could be dominant, numbers could trend down. “We also didn’t model decreases in software and hardware expected as time goes by,” instead sticking to present price tags, Kaushal adds.

Structure is an issue to David Chabner, IS director at Merle West Medical Center (261 beds) in Klamath Falls, Ore., who is “part of the camp that believes ‘access of data’ is all we need, as opposed to ‘exchange of data.’ The latter will take a very long time, since it is dependent on standard data sets and fields that have yet to be defined, and would probably take government mandate to implement.” 

Other study predictions: in five years, 152 hospitals will have test-results viewing; 91 will have CPOE; 198 will have electronic claims processing; and 123 will have electronic eligibility verification.

An important caveat: the $156 billion represents just 2% of annual healthcare spending for five years, Kaushal says. And a big part of the $48 billion in annual operating/maintenance costs will come from pay-for-performance initiatives that favor interoperable providers. “Some payors are doing [P4P] with providers using EHR,” she adds. In this scenario, the government’s role is “providing upfront financial incentives to stimulate the industry and private sector to invest in these systems.”

Numbers and dollars aside, the study proves that “interoperability standards and policies are needed….the sooner the better, and less costly,” says Rick Warren, CIO at Foote Health Systems (517 beds) in Jackson, Mich.

“Certification of functionality, including standards adherence in vendor products, will decrease waste and rework of failed implementations and false starts,” Warren adds. “Regardless of the costs, whatever it turns out to be, the benefits in quality improvement, reduction of inappropriate tests, better patient safety, better follow-up care, better communication among providers and with patients, and greater productivity will certainly outweigh the costs and save millions of lives over time.”