Patients who were in the intervention arm were 67% less likely to go to the emergency department in the last days of their life.
At the VA Palo Alto Health Care System, a new approach to advance care planning significantly improved end-of-care outcomes for patients while reducing overall net costs. Manali Patel, MD, MPH, MS, associate professor of oncology at Stanford Medicine, presented on the success of the intervention during the 2023 ASCO Quality Care Symposium.
The 10-year analysis showed that patients who were in the intervention arm were 67% less likely to go to the emergency department in the last days of their life (OR, 0.33; 95% CI, 0.18-0.63), and 72% less likely to go to the hospital (OR, 0.28; 95% CI, 0.15-0.54). At the same time, they were significantly more likely to receive palliative care (OR, 1.86, 95% CI, 1.03-3.33), and hospice care (OR, 2.36, 95% CI, 1.28-4.36). As she noted, for this effect to be so profound at the 10-year mark for a 6-month intervention was surprising.
“We saw a sustainable impact on end-of-life [care],” Patel said. “[These were] some robust effects 10 years out from a 6-month intervention.”
The study also showed a potential survival impact. Patients in the intervention arm who survived for at least 2 years appeared to live longer than those in the usual care arm (HR, 0.75; 95% CI, 0.56-0.95; P = .04).
“People that lived for more than 2 years, perhaps lived longer,” she said. “This is really an effect that we did not expect to see.”
Patel’s team designed a 3-pronged model that was implemented across a 6-month period. The key change was lay health workers to engage with patients about advance care planning before their visit. It became the standard—Patel likened it to how one must express if they want to be an organ donor when they get their driver’s license; their patients had to express their advance care wishes ahead of their first oncology visit.
The second prong of their approach was tailored messaging. As Patel explained, most patients are not intuitively going to understand what an advance care planning conversation entails or its potential importance. The helped patients see advance care planning as a living, breathing document, subject to change.
“We provided tailored messaging, helping patients really take time to formulate their goals and to also educate patients that this is not a ‘one-time shot,’ ” she said. “We explained that when there are certain trigger points, hospitalizations, and expected side effects, that this is an opportunity to reengage in this living conversation.”
Then, finally, their team provided tailored support to help guide patients through completing the documents, because it can be cumbersome for many individuals.
The team used VA funding to randomly assign veterans with newly diagnosed stage III or stage IV disease, or with recurrent cancer, to receive either usual care or usual care plus the intervention.
The primary analysis looked at the rate of goals of care conversations documented at 6 months and 15 months of follow-up.
Of note, 213 patients were included in the primary analysis, 108 received usual care, and 105 received the intervention. Very few veterans were lost to follow-up or withdrew from the study. All were included in the intent-to-treat analysis.
Primary trial results showed that by the 6-month and 15-month follow-up dates, the intervention had 3 times more documented advanced care planning conversations than the control group, utilized hospice services more than 2 times than control, were 3 times more likely to have had a positive patient experience, and were 95% less likely to require acute care at the end of life. Lastly, in the overall cohort, there was a 20% net reduction in total health care costs because the implementation costs were so low.
This overall analysis looked at patients who had died, assessing their overall survival and quality of life at end of life. Key end points in this analysis included healthcare use in the last 30 days of life, and total healthcare costs during the last 12 months of life.
As Patel noted in the beginning of her presentation, 7 out of 10 patients are concerned about the potential comfort or pain they may experience at the end of life, as well as the costs surrounding it, yet only 3 out of 10 patients have an end-of-life conversation. Moreover, physicians are only accurate in predicting end-of-life preferences 33% of the time.
Her team found that better conversations about goals and preferences occurred with patients in the waiting room than they did in the visiting room with their clinical teams. During the presentation, Patel shared a video in which some of the patients and healthcare workers involved in this intervention spoke about the significance.
“They get someone who is actually helping them through their care,” said Labiba Shere, a Lay Health Worker involved with the VA. “With me, they’re actually discussing things, what they’re feeling, what they are going through, what they want to do, what they understand is their prognosis.”
“I just decided not to go through treatment,” said 47-year-old Rafael Arias, a former Army police officer who has recurrent sarcoma and lost his leg to the disease. “Right now, were just trying to take advantage of the fact that I’m still here and trying to do some things while I can.”
When asked why he decided not to have chemotherapy, Arias responded by saying that it made him sick. “Within a week, we had to stop it,” he said. “So, as difficult as it was, I decided not to go through with it.”
He explained that the treatment was not offering him quality of life and preventing him from spending his last days how he wanted. “My wife and I have plans for doing some things before my departure,” he said.
Based on these robust findings, this model has been spreading to other clinics across the country, and many of the clinical sites outside of the VA have helped to highlight this as a critical service. Many of Patel’s team have advocated in their state for policy change. Now, as of last year, the team has community health workers that are reimbursed.
They have since launched a 24-site randomized cluster randomized trial funded by PCORI, the goal of which is determine if this approach is as effective compared with more passive education approaches for patients.
Patel ML, Agrawal M, Blayney DW, et al. Engagement of patients with advanced cancer (EPAC) randomized clinical trial: Long-term effects on survival and healthcare use. Presented at: ASCO Quality Care Symposium. October 27-28, 2023; Boston, MA.