A new way to prepare doctors for difficult conversations
ĚěĂŔ´ŤĂ˝ researchers have developed SOPHIE, a virtual âpatientâ that trains doctors to communicate better.
Editorâs note: This post was originally published on July 15, 2021. It has been updated and republished to include a video about using SOPHIE for dentist-patient communication training.Â
As many as 68 percent of late-stage cancer patients leave their doctorâs offices either , âor both. These misunderstandings can hinder the ability of patients and their families to make realistic decisions about whether to continue aggressive treatments or instead turn to palliative care.
To address the problem, computer scientists, palliative care specialists, and practicing oncologists are perfecting SOPHIE (Standardized Online Patient for Healthcare Interaction Education)âan online virtual âpatientâ that helps physicians practice how to communicate effectively with late-stage cancer patients about their disease.
Effective communication in this context often means demonstrating empathy and understanding of the complex emotions that patients are experiencing.
âDuring difficult conversations about facing the potential of oneâs own death, patients are frightened and donât know how to ask the right questions, and clinicians may oversimplify, omit, or sugar-coat information, or feel too pressed for time to address patientsâ emotions,â says , an associate professor of at ĚěĂŔ´ŤĂ˝âs . The COVID-19 pandemic made effective communication even more difficult by causing increasing reliance on virtual rather than in-person interactions between patients and physicians, adds Hoque.
Hoque is a pioneer in the area of research that shows it is possible for people to learn and improve their social and interpersonal skills by interacting with automated systems. For example, he helped develop LISSA (Live Interactive Social Skills Assistant)âa tool thatâs been used effectively by people on the autism spectrumâin collaboration with professor of computer science Len Schubert.

Practicing end-of-life conversations
SOPHIE is possible because of a body of nearly 400 conversations that were recorded between late-stage cancer patients and their oncologists, and initially analyzed by palliative care expert and his collaborators at . Epsteinâs focuses on how to improve communication between clinicians, patients, and their loved ones.
A in IEEE Transactions on Affective Computing describes how lead author Mohammad Rafayet Ali, a postdoctoral researcher, and PhD student Taylan Sen, both members of Hoqueâs , created algorithms that could be applied to the transcripts of these conversations, so the researchers could develop metrics to assess a physicianâs ability to communicate clearly with patients.
The metrics focus specifically on the extent that physicians engage in lecturingâdelivering a lot of information without giving the patient a chance to ask questions or to respondâor, on the plus side, how well they employ âsentiment trajectoryââusing positive words and phrases in ways associated with increased patient understanding.
âThe online format can provide training at a fraction of the cost and be made available to physicians worldwide.â
The researchers used a physician communication training protocol, developed by Epstein and his collaborators, to script SOPHIEâs practice sessions, and enlisted nine practicing oncologists at the University to try the system.
âWe received a lot of feedback from the doctors on how natural the conversation was, the kinds of questions SOPHIE should ask, and the concerns a typical patient might raise,â Ali says.
âIn our prior research, physicians practiced with trained actors to learn how to bring up difficult issues, such as prognosis and end-of-life care,â Epstein says. âBut actors can be costly. The online format can provide training at a fraction of the cost and be made available to physicians worldwide. And Ehsan Hoqueâs group is on the cutting edge.â
The researchers are now fine-tuning SOPHIE. âOur hope is that once we have demonstrated its efficacy through a randomized controlled experiment, first-year students at the Medical Center can practice with it,â Hoque says.
The demo has been designed to be available through a computer browser, allowing physicians from anywhere around the world to practice their skills live and receive feedback on their speaking rate, turn-taking, type of questions asked and sentiment trajectory.
SOPHIE could help raise awareness of palliative care as an option. âIn many parts of the world, there is no palliative care. People only focus on how to treat the cancer,â Ali says.
SOPHIE says

A sample conversation with SOPHIE works something like this:
SOPHIE appears as a real-lifeâalbeit AI-derivedâhuman face on the user’s screen.
She introduces herself and mentions that she has lung cancer. Then SOPHIE raises the topic of her sleep pattern at night and asks if she needs to change her pain medication, allowing the physician to assess her perception. She states that her current pain medication, Lortab, is not working anymore.
As the doctor responds, SOPHIE begins collecting data so that she can provide the doctor with an end-of-conversation assessment resembling the image above.
SOPHIE then turns attention to her test results, giving the physician a chance to engage SOPHIE in more difficult topics, before asking more specifically about her prognosis.
SOPHIE next asks about other options and expresses some additional fears and concerns, providing the physician with opportunities to respond empathically.
Finally, she follows up by discussing whether chemotherapy remains an option, whether she should focus on comfort care, what the side effects of chemotherapy are, and how to break the news to her family, allowing the physician to summarize information and suggest a strategy.
Close proximity fosters collaboration
The SOPHIE project is a shining example of the research collaborations that are possible as a result of the close proximity of the ĚěĂŔ´ŤĂ˝ Medical Center to the rest of the University. The River Campus, for example, is just a five-minute walk away.
The project had its origins seven years ago when Hoque, who had just joined the University, spoke to a digital media studies class on the River Campus. âAn undergraduate student in the class was employed by Ron Epsteinâs lab,â Hoque says. âShe talked to Ron and suggested we should talk. We hit it off right away.â
Hoque and his lab members were impressed by Epsteinâs work on improving communication between terminally ill patients and their physicians. They were especially excited by the research possibilities presented by the patient-physician transcripts that Epstein and his collaborators had begun analyzing.
âThey did all the hard work in collecting high-stakes conversations of cancer patients,â Hoque says. âWe were fortunate to be at the right place at the right time to leverage the data and their expertise.
âThis wasnât part of Rafayetâs and Tayâs thesis trajectories at all, and we didnât have any funding to do the work. But we knew that we needed to find a way to do this exciting work.â
Other collaborators coauthors of the paper include Thomas Carroll, associate professor of medicine; Lenhart Schubert, professor of computer science; Benjamin Kane (advised by Schubert); and Shagun Bose, an undergraduate researcher in the Hoque Lab.