Well-timed prompts directed at oncologists and patients with advanced cancer can encourage discussions about care goals — though they may not be enough to spur those conversations in most cases, a new clinical trial finds.

The study, of over 1000 patients with poor-prognosis cancers, found that simple “nudges” to clinicians and patients boosted the likelihood that conversations around care goals would take place. Even so, only about one third of patients ultimately had documented discussions.

“That is a really important takeaway because it just suggests that there is more work to be done in this field,” co-lead investigator Cody E. Cotner, MD, of Harvard Medical School in Boston, told Medscape Medical News.

These important conversations may actually occur more often than they are recorded, Cotner pointed out. But the problem with that, he said, is that undocumented discussions can’t reliably guide care when patients become seriously ill and end up in the hospital.

“They’ll meet a provider they’ve never met before in their entire life, sometimes they’ll be unable to advocate for themselves, and they’ll receive really aggressive care that may not be in line with their ultimate goals,” Cotner said.

Prompting More Discussions

For patients with advanced cancer, discussions about care preferences can ease anxiety, improve quality of life, and ensure that they don’t receive treatments out of step with their values. Yet research shows that those conversations often don’t happen, and patients commonly end up receiving care that is at odds with their goals.

The new findings, published in the June issue of the Journal of the National Comprehensive Cancer Network, come from the PATH-SIC trial. It evaluated the impact of simple clinician and patient nudges designed to support more communication.

The clinician nudge consisted of an email reminder sent before an oncology visit encouraging discussion of goals and preferences for care, while the patient nudge included a mailed letter and questionnaire designed to prepare patients for those discussions.

The trial enrolled 1051 adults (median age, 65 years) with poor-prognosis cancers who were starting treatments associated with an expected survival of less than 1 year. Patients were randomly assigned to the clinician-directed nudge, the patient-directed nudge, both interventions, or usual care.

Overall, Cotner’s team found that the combined clinician- and patient-directed nudges led to the best outcomes.

Within 60 days of randomization, 17.3% of patients in the combined group had a documented serious illness conversation in their electronic medical records (EMRs) vs 10.7% in the control group.

The clinician-only nudge group achieved a similar rate of documented conversations but did not significantly outperform the control group (16.7% vs 10.7%).

In adjusted analyses, patients in the combined-nudge and clinician-nudge groups had 79% and 70% higher odds of a documented conversation than the control group, respectively (P = .02 for both).

The investigators also used natural language processing to identify care discussions documented in free-text clinical notes. When those discussions were included, conversation rates rose to 32.5% in the combined-nudge group vs 22.6% in the control group (P = .01).

The researchers found that natural language processing identified an additional 134 conversations in clinical notes — an 116% increase over structured advance care planning documentation alone. That suggests many important discussions are not being recorded in designated advance care planning fields.

“It’s really critical that they be documented in an easy-to-find place in the electronic medical record,” Cotner said.

That point was echoed by Robert M. Arnold, MD, a professor of geriatrics and palliative medicine at the Icahn School of Medicine at Mount Sinai in New York City.

“If it’s not documented, then it’s very unlikely to be usable by the next clinician,” said Arnold, who was not involved in the study.

“The goal is to make sure that downstream people are using those conversations,” he told Medscape Medical News, adding that future work should focus on whether such interventions improve care delivery and help ensure that treatments align with patients’ values and wishes.

Toward Better Documentation, Better Care

Arnold also pointed to a potentially concerning finding from an exploratory patient survey. Among the 52 patients who completed questionnaires before and after the intervention, those in the combined-nudge group showed greater declines in “heard and understood” scores than patients in the other study groups.

The authors cautioned that this analysis was based on a small number of patients, but said future research should examine whether such interventions affect patients’ perceptions of clinician communication.

Arnold pointed to possible explanations for the finding — conversations may have been poor-quality or patients may have received inadequate support afterward — but agreed that more study is needed.

For now, the current findings offer a “realistic, scalable step toward bringing serious illness conversations into oncology care earlier and more reliably,” said Elise Carey, MD, a palliative care specialist at Mayo Clinic Comprehensive Cancer Center in Rochester, Minnesota, who wasn’t part of the study.

In a statement, Carey noted that the benefit was largely driven by the clinician nudge, suggesting that “small, well-timed supports can help clinicians make space for these essential conversations even in a busy clinical practice.”

Timing and precision are a critical element, Cotner stressed. “Burnout is real and alert fatigue undermines the goal of these nudges,” he said. “When they feel like noise, they get ignored.”

Looking ahead, Cotner expressed optimism that AI could be helpful when it comes to improving documentation.

“We’re really excited about how AI is going to impact this area,” he said. “You could envision a future where the ambient scribe listens to something that sounds like a serious illness conversation and subsequently takes that and documents it in the EMR for the clinician.”

To Cotner, the broader push toward personalized, precision medicine must include palliative care, as well — with the goal of “delivering the right care to the right patient at the right time.”

The research was supported by Dana-Farber Cancer Institute. Neither the study authors nor Arnold had any relevant disclosures.