'Wired' Hospitals Post Lower Death, Complication Rates

But researchers note system chosen needs to match needs, culture of staff

HealthDay + More

By Amanda Gardner
HealthDay Reporter

TUESDAY, Jan. 27 (HealthDay News) -- The more "wired" a hospital is, the lower its rate of patient deaths and complications, a new study finds.

Automating hospital information systems also saves centers money, the researchers report.

Although there are many kinks to be worked out, said Devon M. Herrick, senior fellow at the National Center for Policy Analysis in Dallas, "I assume that over the course of the next few years, with or without the government's prodding, that we will begin to integrate this more and more, because it is a good idea, but I think there will be some growing pains about which system and how and whether it talks to neighboring hospitals and so on."

There will be not only growing pains, but often astronomical start-up costs as well.

There are high hopes that health information technology systems will help with the health-care reform advocated by President Barack Obama.

Yet there have been few comprehensive studies testing these beliefs, said the authors of a study in the Jan. 26 issue of the Archives of Internal Medicine.

Only about one-quarter of hospitals have some type of emergency medical records, and 5 percent have physician order entry, or "medical-record-lite," Herrick said.

Health information technologies include four main categories, the researchers stated: notes and records (case histories, admission histories, etc.), test results, order entry, and decision support (for example, information a doctor factors into a treatment decision).

"Every day there are more innovations [in medicine], more evidence-based guidelines. For a single physician to keep track of that is difficult," said study author Dr. Ruben Amarasingham, associate chief of medicine at Parkland Health & Hospital System and an assistant professor of medicine at the University of Texas Southwestern Medical Center at Dallas. "Computers and, in particular, electronic decision support, provides enormous adjunct service [to physicians and other health-care professionals] when taking care of patients."

Some of the components can act as a sort of spell check for doctors, for example, alerting physicians to changes in a patient's vitals, or noting a medication discrepancy. It can also improve communication among the various layers of staff now caring for any one patient.

The authors compared inpatient death rates, complications, length-of-stay and cost associated with greater and lesser levels of automation in 41 Texas hospitals.

The analysis involved more than 167,000 individuals over the age of 50 who were hospitalized between Dec. 1, 2005, and May 30, 2006.

The level of automation was measured by physician interactions with the system, using a tool that takes into account how well the physician is trained in the system, the usability of the system, and other factors.

A 10-point increase in the computerization of notes and records meant a 15 percent decrease in the death rate. This translated into a 1.4 percent mortality rate among those with the highest scoring on notes compared with a 1.9 percent rate among those with the lowest scores, or five fewer deaths per 1,000 patients.

Higher scores in the order entry category were associated with a 9 percent decrease in the risk of a heart attack and a 55 percent decrease in coronary artery bypass graft procedures.

Overall, higher scores in decision support equated with a 16 percent decrease in the rate of complications, while higher scores on test results, order entry and decision support were linked with lower costs.

There was no correlation between length of stay and technology score.

Of course, the success of a health-information technology system depends on so much more than the system itself.

These authors looked at the "socio-technical environment" of the automated systems. "This is an emerging view that would suggest that the best systems harmonize the relationships between it and the people who use it, i.e., health-care professionals' routines and culture or organization," Amarasingham explained. "It's very possible for organizations to decide, 'we're going to invest a lot of money in it' and not take these extra steps which are crucial to building an environment that supports the technology and in which the technology supports the existing people."