Objective

To compare the effects of a decision aid versus usual care on patient knowledge of personal risk for acute coronary syndrome, choice of admission to the hospital for observation, and further cardiac testing in patients with possible acute coronary syndrome.

Study Design

Design Element	Description
Design	Randomized controlled trial
Population	898 patients who visited the emergency room (ER) for chest pain and had no prior history of heart attack
Interventions/ Comparators	A decision aid containing personalized risk assessment for acute coronary syndrome in the next 45 days, information regarding follow-up cardiac care, and information about additional testing options Standard discussion with clinician about results of diagnostic tests (usual care)
Outcomes	Primary: patient knowledge about short-term risk for acute coronary syndrome and options for care Secondary: patient engagement in the hospital-admission decision, proportion of patients admitted for cardiac testing, proportion of patients who had cardiac testing within 30 days of ER visit, and number of major adverse cardiac events 45 days after ER visit
Timeframe	Immediately following ER visit for primary outcome

This multicenter pragmatic parallel randomized controlled trial tested a decision aid that included personalized information about patients’ chest pain diagnosis, options for additional tests, and information on patients’ individual risk for acute coronary syndrome in the next 45 days. After reviewing the decision aid, patients engaged in a shared decision making conversation with attending ER clinicians, including doctors, nurses, and physician assistants. The primary study outcome was patients’ knowledge of their risk for acute coronary syndrome and options for future care, measured by patient answers to knowledge questions after ER visits. Secondary outcomes included patient involvement in the hospital-admission decision, proportion of patients who decided on admission for further cardiac testing, and 30-day rate of major adverse cardiac events.

The study included 898 adult patients who visited an ER for chest pain in six hospitals across the United States. The study included patients whose preliminary tests in the ER indicated that they had not had a myocardial infarction. In this study, 60% of patients were white, 34% were black, and 60% were female. The mean patient age was 50, and 30% had a high school education or less.

The research team randomized patients to one of two groups. Patients in the intervention group read and discussed the information in the decision aid with attending ER physicians during ER visits. Patients in the control group received usual care without the decision aid. Researchers collected postvisit survey data immediately after the intervention in the ER. Whenever possible, the research team made visual and audio recordings of the doctor-patient discussions to assess clinicians’ efforts to engage patients in the decision-making process. The research team contacted study patients 45 days after their ER visit to determine whether they experienced a major adverse cardiac event. For patients who the researchers could not reach, the researchers verified mortality status using a national database. The research team compared patient-level outcomes between the intervention and control groups, adjusting for intrasite and intraclinician correlations.

Results

Compared with patients in the control group, patients in the intervention group

• Had significantly greater knowledge about their risk of acute coronary syndrome
• Were more engaged in the decision-making process
• Were less likely to decide with their clinician to be admitted into the ER observation unit for cardiac testing (37% versus 52%)
• Were less likely to undergo cardiac stress testing within 30 days of their ER visit (38% versus 46%)

There were no deaths from cardiac or unknown causes in either group. One participant in the intervention group had a major adverse cardiac event following hospital discharge; the data safety monitoring board classified this event as unrelated to the study intervention. No participants in the control group had major adverse cardiac events.

Limitations

The research team was unable to contact 70 patients for follow-up assessment of secondary outcomes; however, the team verified 45-day mortality using an independent database and found no differences between treatment groups. The investigators were unable to obtain video recordings in 40% of encounters because of patient refusal and problems with recording equipment; however, the sample size was sufficient to examine the extent to which physicians engaged patients in shared decision making. Finally, the study included only patients who had chest pain. The research team didn’t design the decision aid for patients at risk for heart problems who come to the ER because of other symptoms.

Conclusions and Relevance

For patients visiting the ER for low-risk chest pain, the decision aid was an effective way to increase knowledge before deciding about hospital admission for observation or further cardiac testing. The decision aid was also effective in increasing patient engagement in the decision-making process.

Future Research Needs

Further research studies could examine how best to use the decision aid in the ER and whether larger studies confirm the findings from this study.