A study of 3,000 U.S. hospitals in the process of changing to digital records found no significant cost benefits. (iStock)I recently flew into Charleston, S.C., to see how the Medical University of South Carolina was handling its electronic health record systems. I went Deep South because MUSC has had the Oacis medical record system in place since 1993 — it was the first hospital system in the world to buy one.

I went there to check out the system because all the hospitals in Montreal, along with Sunnybrook Hospital in Toronto and the Children's Hospital of Eastern Ontario in Ottawa (not to mention almost the entire state of South Australia) either have or are about to use an Oacis system supplied by Canada's Telus Health Solutions.

I especially wanted to see the effectiveness of an EHR system after it had been used for 16 years in a locale that has a million patient visits a year.

At first glance, everything seemed somewhere between coolisimo and highly efficient.

Larry Afrin, who is both a hematologist and director of information technology for MUSC's Office of Graduate Medical Education, demonstrated how he could comb through 16 years' worth of medical records and uncover a previously unforeseen negative interaction between a standard antibiotic and the suppressed immune system of patients undergoing chemotherapy.

He then showed me, with a couple of keystrokes, how you can retrieve an electronic version of a chest X-ray, or indeed any image, at will. And finally, he gleefully described how the moment Oacis was put in place he had torn up stacks and stacks of the three-by-five-inch cards on which he previously wrote all his patients' medical particulars to refer to when visiting their bedsides.

Following the money

With these successes in mind, I began asking people at MUSC a very different kind of question. Had the purchase of electronic health records saved or made money for the hospital?

I asked first because cost-saving is the lure many EHR vendors claim or imply when selling their products.

Novarad Corp. of Utah, for example, crows on its website about one of its products: "Because it takes care of all of your information and imaging needs, you'll start saving tons of money before you even plug it in." It then directs purchasers towards an online return-on-investment calculator to figure out their future savings.

Companies are not alone in seeing the EHR systems turning into operational gold for hospitals and clinics.

Canada Health Infoway, the federally funded agency partnering with provinces and territories to create a cross-country electronic health record system, commissioned a private consulting firm to monetize the values of computerized health imaging — what is known as PACS, or picture archiving and communication systems.

Their estimate: It could save our medical system up to $1 billion a year in existing or projected expenses.

Another study CHI quotes says EHRs could save the country as a whole $6 billion to $7 billion. In the U.S., Rand Corp. estimated in 2005 that a digitized American medical system could save that country $81 billion annually.

But what is happening on the ground in South Carolina is a lot less convincing.

After 16 years of experience with electronic health records, and after spending $23 million a year on its clinical computer systems, MUSC still doesn't know if it is saving or making money.

"We have some studies underway to document ROI — that is, the return on our investment — but at this stage we feel we can't say it has been positive," says Frank Clark, chief information officer for the MUSC.

What about all those money-saving claims from companies?

"Vapourware," Clark says in his soft southern accent.

Looking closer to home

Well, you might say, that's just one American hospital complex's experience. And we all know how wasteful things are in U.S. health care.

Maybe things are going to be different in this country. And that is when I became fascinated with an account of PACS's introduction in western Newfoundland, which recently appeared in an online version of the Journal of Digital Imaging.

The Western Newfoundland Health Authority encompasses about 80,000 people and includes Corner Brook and Stephenvillle. It was studied because until 2005 it was virgin territory for PACS. No electronic imaging capabilities were in place whatsoever.

Two years later, all imaging is electronic. This allowed Don MacDonald, a senior analyst who has been conducting ehealth benefit analyses for seven years with the Newfoundland Centre for Health Information, to do an exhaustive before-and-after study.

'I fundamentally disagree with all of those studies that say there are billions of dollars to be saved. It is just not going to materialize.'—Sam Marafioti, Sunnybrook Hospital

What his study found — please sit down — was that rather than saving anything, the inflation-adjusted cost per examination of computerized X-rays, MRIs and CT scans has gone up from $9.50 to $11.80.

The paper argued that because of PACS's high price, the cost differential will continue over the estimated six-year lifespan of the $4.1-million system.

Then I chatted with MacDonald, asking what happened to the film librarians and darkroom people whose layoffs theoretically should save the hospitals $230,000 a year. "In each and every case, I was told they were all retrained. There were no layoffs," he tells me.

And indeed, when he asked the same question across Newfoundland he was told the same thing occurred in all other hospitals.

I heard a similar story in Charleston. Thus, even apparent savings from staff cuts seem dubious.

Measuring return

Now, that is not to say there are no benefits. MacDonald points out that the ease of sending images back and forth means that the people in small towns who are X-rayed don't have to travel long distances to reach someone who can interpret results for them.

Simply having the system makes it easier to attract radiologists to the province. They can read the records anywhere and don't have to be in smaller centre. That's good for the patients and that's good for the medical system at large. But as to the hospitals' economic well-being … there's a question mark.

Canada Health Infoway estimates it will take up to $8 billion more to implement EHRs in their entirety across Canada. (Canadian Press)MacDonald says the PACS return-on-investment economics might be better in the largest urban hospitals, but at present hardware and software prices he thinks it is unlikely "there will be any financial return on investment for the majority of PACS implemented in Canada."

He is not alone in his doubts that money is going to be made or saved. Sam Marafioti, chief technology officer at Sunnybrook Hospital, which has been in the forefront of Canadian EHR record implementation, radiates cost/benefit doubt during an interview:

"People who have the best intention in marketing the benefits of EHR, they oversell it. And I fundamentally disagree with all of those studies that say there are billions of dollars to be saved. It is just not going to materialize."

A recent study of 3,000 U.S. hospitals in the process of changing to digital records found no significant cost benefits. But Canada Health Infoway strongly resists such negative conclusions.

"Well, there may be better outcomes. Maybe cost avoidance, versus saving money is how to put it," Dennis Goikas, chief technology officer for CHI, tells me. "Health care is not always driven by hard costs. Sometimes it is driven by the softer outcomes as well."

The disconnect between who pays (an already strapped medical system) and who benefits (the patient) is the electronic medical record's deep and troubling contradiction.

In other industries, the computerization of records was equally and obviously good for both companies' bottom lines and their clients. In my working life I watched newspapers abandon typewriters, film cameras and their own printing presses as computerization grew.

While readers got articles that had been more extensively rewritten and edited, media firms could lay off proofreaders, compositors, dark room attendants and press operators — and these staff reductions saved the companies money. The same kind of thing happened with banks, architects and so many other sectors.

But in hospitals, the group that must foot the bill for going digital may see no improvement to its bottom line. It may even see costs go up.

And this is a fiscal perversity that leads me to an equally perverse recommendation. The question is not, as it is often stated: 'Why are hospitals such Luddites when it comes to creating digital records?'

Rather, I think we should praise them for being cautious about radically exchanging one system with another whose financial benefits to them are somewhere between murky and non-existent. I think we should laud them for wanting to be sound fiscal managers.

And given the fact that Canada Health Infoway estimates it will take up to $8 billion more to implement EHRs in their entirety across Canada, I think we should continue to encourage them to make future changes cautiously.

When people are cautious about adopting a technology that requires them to spend more money than they will earn back, I don't call them Luddites. I call them Smartites.