Monday, January 5, 2009

Pharma Takes Baby Steps in Health IT, learns CP/M

CP/M was the predecessor and template Bill Gates used (bought, in fact) for DOS, of course. CP/M is primitive, over three decades old, decidely "old fashioned", and to just be learning it today would decidedly make an organization appear a laughing stock.

In "Pfizer Project Looks at Side Effects", Wall Street Journal, Jan. 2, 2009 (subscription required), we learn that:

Drug maker Pfizer Inc. is joining with two Boston hospitals to test whether computerized patient records can be used in helping federal regulators detect dangerous drug side effects. [VERY innovative of them. Perhaps this George Jetson-ish idea came from their CEO, former head lawyer for McDonald's? - ed.]

Massachusetts General and Brigham and Women's hospitals are encouraging 30 of their doctors to report serious side effects to the Food and Drug Administration by making the reports part of their normal routine filling out electronic patient charts.

... "The public-health impact is much quicker recognition of problems with medicines," said Jeffrey A. Linder, a Brigham and Women's internist who is leading the three-month pilot. The two hospitals belong to the Partners HealthCare system.


Indeed. This article might better have been titled "An Experiment With The Obvious."

We also learn that:

Martin P. Solomon, another Brigham and Women's internist, said he had submitted only a half dozen reports in 32 years of seeing patients because the [written - ed.] reports took so long to fill out and send. Since the study began Dec. 9, Dr. Solomon estimates he has filed at least a dozen.

Removing inconvenience is simply good sociotechnical engineering for ultra-busy doctors. Workflow 101 for simpletons.

Sociotechnical refers to the interrelatedness of social and technical aspects of an organization. Sociotechnical theory therefore is about joint optimization, with a shared emphasis on achievement of both excellence in technical performance and quality in people's work lives.

Now, thirty doctors is rather a minority of doctors in these venerable institutions. The amount of data to be entered is rather straightforward:




The cost of this experiment is far less than the cost of one glossy drug ad:

Pfizer, which has agreed to pay the hospitals $226,806 for the project, has talked with a few other pharmaceutical companies beyond Lilly about expanding the Boston approach, said Michael A. Ibara, a drug-safety official at the New York company. Pfizer has given an additional $117,000 to support groups that developed the technology.

Even at these low figures, excuse me? "Pfizer has given an additional $117,000 to support groups that developed the technology?" What technology, exactly? The same technology I use to send reports of my Mac OS X application crashes to Apple, perhaps?

The question really is, why do we need a funded experiment with 30 doctors to do what EHR experts have been proposing as a possibility for decades, namely, improved postmarketing surveillance? All that's really happening is an increase in convenience away from using paper forms and mailing them. I presented this issue a decade ago as a simple side benefit of EHR, for example, at an Ethics & Public Policy Center conference on healthcare in Washington. It has been brought up over the years by many others. I brought up the possibility of linking to academic centers with EHR's at Merck as well to gather postmarket data, to no avail.

So, why is pharma taking the baby step of soliciting adverse events reports electronically through an incredibly miniscule, cheap study?

This initiative's lilliputian scope is laughable.

It's decades behind the times and the capabilities of modern information systems. It's as if pharma has decided to learn CP/M in 2009. This raises several questions:

  • Why did this "experiment" not happen long ago?
  • Why not go with a much stronger effort at collecting this data, which then opens up excellent opportunities for further research - as I pointed out in my 2003 paper "A Medical Informatics Grand Challenge: the EMR and Post-Marketing Drug Surveillance" (pdf, link)? I'm not sure what you can do in that regard with data from a mere 30 doctors.
  • Why not spend pharma-sized money on the project to allow the above to occur? After all, billions of dollars are spent annually on "gifts" to doctors and DTC advertising. Direct to patient (consumer) marketing expenditures of Pharma – est. $30 billion / year. (The entire NIH budget not long ago - est. $28 billion/yr.)
  • CRO's are moving to electronic CRF's (case report forms). What's the hangup on use of EHR's to report drug AE's in patients not involved in a clinical trial?

The best answer I can propose: perhaps such capabilities will provide information that pharmas would rather have become apparent in a much longer timeframe. Time is, after all, money.

However, such an initiative can provide great P.R. on just how leading edge pharma is becoming in leveraging the EHR. The Wall Street Journal article looks like something Pfizer's CEO would be proud to show grandma - and its marketing people to brag about to the public. I can't wait for the TV infomercial.

CP/M is, after all, a great operating system to tinker with when experimenting with computers to learn if they can help run an international business enterprise.

-- SS