“COPD is projected to rise to the third leading cause of mortality by 2030,” says Dana Tripp, BA, MD-Candidate. Rebecca N. Hutchinson, MD, MPH adds, “Unfortunately, although COPD is a terminal condition, many patients do not have the necessary conversations with their physicians to prepare for end of life. The inherent prognostic uncertainty, due to the condition’s unpredictable trajectory, is one possible explanation behind the lack of advance care planning (ACP) for this population. We currently do not have an easy and accurate prognostic assessment to facilitate improving ACP for patients with COPD.”
For a study published in the Journal of General Internal Medicine, Dr. Hutchinson, Tripp, and colleagues sought to assess how well the surprise question (SQ), a prognostic tool, predicts mortality and prompts ACP for patients admitted with an acute exacerbation of COPD. Studies in patients with other diseases—including cancer, kidney disease, and heart failure—have found the SQ to effectively predict mortality. “The SQ asks clinicians, ‘Would you be surprised if this patient died within the next year (or 30 days)?’” explains Tripp. “An answer of, ‘No, I would not be surprised,’ is considered test positive (SQ+) and ‘Yes, I would be surprised,’ is SQ-.”
Patients With SQ+ at Least Three Times More Likely to Receive ACP
The study team set out to determine 1) the accuracy of the SQ in predicting 1-year and 30-day mortality and 2) if a “no” answer to the SQ effectively prompted clinicians to complete ACP. “We defined ACP broadly, to include completion of advance directives/physician orders for life-sustaining treatment documents, having documented goal of care conversations, or consulting specialty palliative care,” Tripp adds. “We performed a chart review to determine if patients had a documented goal of care conversation. Mortality data was obtained from the EMR, online obituary searches, and HealthInfoNet.”
According to Tripp, there was no significant association between 30-day SQ results and 30-day mortality. However, the 1-year SQ performed better: “Patients with a 1-year SQ+ had more than two times greater odds of dying within 1 year than those with SQ-,” says Tripp. “We believe these results demonstrate that the 1-year SQ may serve as an effective component of prognostication for patients with COPD.” After multivariable adjustment, the 30-day SQ was not associated with increases in ACP. “One-year SQ+ patients had more than three times greater odds of receiving ACP than SQ- patients in unadjusted models and 2.63 times greater odds after multivariable adjustment (Figure),” Tripp adds.
Consider the SQ in Managing Patients With COPD
Dr. Hutchinson cautions, however, that showing an association between SQ answer and receipt of ACP does not demonstrate causality. “In other words, we do not know if the answer to the SQ effectively triggered the clinician to strive for a goals of care conversation or if there was another factor that resulted in the clinician engaging the patient in ACP,” she says. “We also do not know if the type of clinician answering the SQ (eg, resident vs attending, years of practice, etc.) changes the accuracy. We anticipate that future research will provide a better understanding of whether or not the SQ influences care and if the person answering the SQ changes the accuracy in the COPD population.”
In the meantime, the study team is hopeful that clinicians who treat patients with COPD will consider using the SQ in their practice. “Increasing ACP in this population will help to ensure that patients with COPD receive care aligned with their preferences,” Dr. Hutchinson says. “This may help to decrease the currently high proportion of patients with COPD who receive mechanical ventilation and other invasive treatments close to the end of life.”