Decision aid for low-risk chest pain

A trial at six EDs in the U.S. compared usual care or a structured risk assessment.


Decision aid may help with low-risk chest pain in the ED

Patients with low-risk chest pain who are being considered for observation admission and additional testing may benefit from a decision aid designed to increase knowledge of risk while decreasing decisional conflict, according to a recent study.

The pragmatic parallel randomized controlled trial, conducted at six EDs in the U.S., compared the effects of either providing usual care or conducting a structured risk assessment using a quantitative pretest probability web tool and corresponding decision aid, “Chest Pain Choice,” which featured validated 45-day risk estimates for acute coronary syndrome (ACS). Results were published on Dec. 5, 2016, by The BMJ.

A total of 361 clinicians (emergency physicians, nurse practitioners, and physician assistants) participated in the study. Eligible patients presented to the ED reporting chest pain and were being considered for admission to the observation unit for cardiac stress testing or coronary computed tomography angiography (CCTA). In total, 451 were randomly assigned to receive the decision aid, and 447 received usual care. Most patients were women (59.5%), and the mean age was 50.3 years.

The primary outcome was patient knowledge about ACS risk, which was assessed by immediate post-visit survey. Secondary outcomes included the degree of uncertainty patients experienced and patient trust in their clinician. Researchers assessed safety by determining whether patients experienced a potentially avoidable major adverse cardiac event up to 45 days after discharge.

Patients who were given the decision aid answered more questions correctly on the knowledge survey, getting 4.2 questions correct out of 8 compared to 3.6 in the usual care group (mean difference 0.66; 95% CI, 0.46 to 0.86). Patients in the decision aid arm also reported significantly less decisional conflict (43.5 on a decision conflict scale vs. 46.4; mean difference, −2.9; 95% CI; −4.8 to −0.90). Use of the decision aid, which took an average of one more minute of clinician time, did not significantly impact patients' trust in their physician.

A significantly lower proportion of patients in the decision aid arm decided, with their clinician, to be admitted to the observation unit for cardiac stress testing or CCTA (37.3% vs. 52.1%; absolute difference 14.8%; 95% CI, 1.1% to 13.9%), and a significantly lower proportion underwent cardiac stress testing within 30 days (38.1% vs. 45.6%; difference, 7.5%; 95% CI, 1.1% to 13.9%). No major adverse cardiac events were found related to the intervention, but due to lack of power, a large-scale implementation trial is needed to definitely assess safety, the study authors said.

They noted limitations of the trial, such as how the decision aid cannot be used in patients with non-chest pain syndromes and that more accurate methods to estimate patient risk are likely to become available. They recommended that clinicians consider using the decision aid, which is available online, in appropriate patients.

The study “supports the view that giving patients more information about risk is not to be feared,” according to an accompanying editorial.