UPMC pilots machine learning, telehealth to inform about patient transfers

Every year, thousands of patients are referred between UPMC hospitals for high-precision, complex medical care.

To ensure patients are fully informed prior to these transfers, the Pittsburgh healthcare system is currently running a new nursing process supported by a machine learning tool.

While it is sometimes necessary to transfer patients for more specialized care, such transfers can have unintended consequences – such as removing the patient from their family and other support systems, a particularly difficult decision for patients who are near death and may not be spending want their last days in the hospital. It is important for clinicians to discuss such decisions with patients to ensure they understand the severity of their illness and to base the next steps on the patient's wishes.

To ensure these conversations take place, researchers at the UPMC and the University of Pittsburgh School of Medicine developed a machine learning algorithm that predicts the mortality of patients who may be transferred to another hospital for higher levels of care. Patients at highest risk are flagged for more in-depth discussions about their care goals.

The researchers published a study validating the algorithm known as SafeNET [Safe Non-Elective Emergent Transfers] in the journal PLOS One earlier this month.

The SafeNET algorithm evaluates 14 variables, including age and vital signs, to assess a patient's risk of death.
When a patient is at high risk, two processes are initiated: a three-way conversation between an emergency doctor, an intensive care doctor in the possible transfer facility and a palliative clinic, as well as palliative medicine for telemedicine services between the patient and family members, in order to Discuss expectations and options for next steps.

Dr. Daniel Hall, medical director for high risk populations and outcomes at the UPMC Wolff Center and author of the study, emphasized that the algorithm does not make decisions about patient care. It is intended to trigger a "pause" in which doctors and patients speak at length before deciding whether to make a transfer.

The SafeNET algorithm is currently being tested in three EDs at UPMC. Since November, the algorithm has marked 11 patients with the highest risk of death. After talking to the palliative care team, four of the patients ultimately decided to continue intensive care and seven decided not to be transferred.

The seven patients "decided that all things considered – their care goals, their personal values, which is important to them – to stay with them," said Dr. Karl Bezak, Medical Director for Palliative Care at UPMC Presbyterian and Montefiore Hospitals. Rather than having higher visual acuity care further from home, some of these patients opted for options like the home hospice.
said dahalli.

By admin

Leave a Reply

Your email address will not be published. Required fields are marked *