Failure of information as an intervention to modify clinical management. A time-series trial in patients with acute chest pain.

OBJECTIVE

To test whether a low-intensity, nonintrusive intervention improved the efficiency of management of patients with acute chest pain.

DESIGN

Time-series trial with six 14-week cycles, each including a 5-week intervention period and a 5-week control period separated by 2-week "washout" periods.

SETTING

Urban teaching hospital.

PATIENTS

1921 patients aged 30 years or older with acute chest pain unexplained by local trauma or chest radiograph.

INTERVENTION

Risk estimates and triage recommendations were made available to physicians at the time of emergency department evaluation and, for hospitalized patients, on a daily basis before morning rounds. Flowsheets and stickers, but no direct human contact, were used to transmit this information.

MEASUREMENTS

Rates of admission to the hospital and coronary care unit, inpatient costs, and lengths of stay.

RESULTS

Rates of admission during intervention and control periods were similar in both the hospital (52% and 51%, respectively) and the coronary care unit (10% and 10%, respectively). Total lengths of stay in the hospital were similar (4.9 +/- 5.9 days and 4.9 +/- 5.7 days, respectively), as were average total costs ($7822 +/- $13,217 and $7955 +/- $13,400, respectively). No differences in management were detected for the subgroup of patients with low clinical risk for acute myocardial infarction.

CONCLUSIONS

The use of information alone--without direct human contact--did not affect management of patients with acute chest pain at this hospital. Although this low-intensity intervention might be more effective for other conditions and in other settings, our data support the use of other strategies to affect physician decision making.