More Hospital IT Linked to Lower Mortality, Lower Costs

Greater automation of a hospital information system appears to be associated with substantial reductions in mortality, complications, and costs for many patients, according to a report.

In a study of 72 general urban hospitals across Texas, a random sample of physicians who provided inpatient care rated the degree to which the hospitals had effectively computerized four information technology (IT) domains: medical notes and records, test results, entry of medical orders, and support for clinical decision making.

These results were correlated with patient outcomes at the hospitals, said Dr. Ruben Amarasingham of the University of Texas Southwestern Medical Center, Dallas, and his associates.

The more than 7,400 physicians who participated in the study practiced internal medicine (including 9 subspecialties), general surgery (including 10 subspecialties), or family practice. Patients older than age 50 years who were being treated for four clinical conditions—myocardial infarction, heart failure, coronary artery bypass grafting, or pneumonia—were assessed.

The study statistically controlled for the likelihood that medical centers with more IT tend to have more resources available and better performance on quality-of-care measures to start with, the researchers noted.

They found “impressive” associations between several information technologies and beneficial outcomes.

For example, hospitals that scored well for computerized entry of medical orders showed a 9% reduction in MI mortality and a 55% reduction in CABG mortality. Those that scored high for computerized support for medical decision making—such as easy online access to treatment guidelines—showed a 21% decrease in complications.

Hospitals that scored high for automation of medical notes and records showed a 15% decrease in all-cause fatalities. “This would suggest that for every 1,000 patients, 5 fewer patients die at hospitals with the highest notes and records scores,” Dr. Amarasingham and his colleagues said (Arch. Intern. Med. 2009;169:108-14).

Hospitals with effective IT also had substantially lower costs.

Length of stay was the only outcome that did not show any relation with effective IT. Given that length of stay already is extremely low because of payer scrutiny of hospital stays, “this measure may already be so low as to be resistant to the efficiencies introduced by IT,” the investigators noted.

In an editorial comment accompanying this report, Dr. David W. Bates of Brigham and Women's Hospital, Boston, termed this study a “landmark” because it assessed physicians' everyday use of IT across numerous hospitals with a diverse array of patient populations.

“Hospital IT is expensive, and there have been serious doubts about the extent to which it will actually be beneficial,” Dr. Bates noted (Arch. Intern. Med. 2009;169:105-7).

At a time when many hospitals are losing money, most “have been nervous about making large investments in technology that is difficult to implement, creates major issues with change management, carries a substantial risk of failure, and has uncertain benefits,” Dr. Bates said.

The findings of this study demonstrate that the negative effects of implementing hospital IT do not “overwhelm or wash out the positive ones, as some have suggested,” Dr. Bates added.

Dr. Amarasingham is also affiliated with the center for knowledge translation and clinical innovation at Parkland Health and Hospital System, Dallas.