TORONTO — Computerized reminders that were flashed to primary care physicians as they checked and recorded their patients' blood pressures led to a small but significant improvement in the rate of prescribing drugs that followed hypertension-management guidelines.
But in this study, which randomized 14 general medicine clinics to either use or no use of the computer-generated reminders, automated prompts had no effect on the rate at which patients had their blood pressure controlled to target levels, according to Dr. LeRoi S. Hicks, an internal medicine physician at Brigham and Women's Hospital and Harvard Medical School, both in Boston.
It's possible that improved blood pressure control could be achieved by not only prompting physicians to use the right drugs, but also prompting them to use the right dosage or to add more drugs when needed, Dr. Hicks reported at the 14th World Congress on Heart Disease.
“We focused on which drugs were used, not on intensification of treatment. We may be better off focusing on intensification,” as well as on the reduction of racial disparities in prescribing patterns, Dr. Hicks said.
The study involved eight community-based and six hospital-based general medical clinics in the Boston area from July 2003 to February 2005. The physicians at seven of the clinics were randomized to treat patients for hypertension by their usual practice.
In the other seven participating clinics, when physicians measured their patients' blood pressures and then entered the readings in each patient's computerized record, they received a computer-generated reminder telling them which drugs to preferentially use to control blood pressure.
The study included 786 patients treated using the computer-generated messages, and 1,048 patients treated by usual care.
The drug recommendations were based on the sixth report of the Joint National Commission on the Prevention, Detection, Evaluation, and Treatment of High Blood Pressure (JNC 6), the authoritative guideline prepared and disseminated by the National Heart, Lung, and Blood Institute. JNC 6 was the prevailing guideline when the study began. (The subsequent edition, JNC 7, was released in December 2003; JNC 8 is expected in 2009.)
The two patient groups were similar by age; ethnic and racial profile; insurance coverage; and baseline level of blood pressure control. About 43% of patients in each group were at their goal blood pressure when the study began. Nearly 90% were also on regimens that were consistent with the JNC 6 recommendations at baseline.
During an average follow-up of about 1.5 years, the computer-generated prompts had essentially no effect on the extent of blood pressure control. Target pressures were reached by 45% of patients in the usual care group and by 48% in the intervention group, a difference that was not statistically significant, reported Dr. Hicks at the congress, sponsored by the International Academy of Cardiology.
However the computerized decision support system did lead to a small but significant rise in prescribing compliance with the JNC 6 guidelines when this was assessed 7 days after each patient's medical visit.
The computerized prompts were linked with a 32% increased rate of compliance, compared with the control patients, after adjustment for baseline differences in demographic and clinical parameters. But because most patients (nearly 90%) were in compliance at baseline, the absolute amount of increased compliance achieved by the intervention was modest, Dr. Hicks said.