Palliative Care
Conditions
Keywords
machine learning, predictive model, palliative care
Brief summary
A machine learning algorithm will be used to accurately identify patients in certain primary care units who may benefit from palliative care consults.
Detailed description
A machine learning algorithm will be used to accurately identify patients in certain primary care units who may benefit from palliative care consults. These patients will be presented weekly to a palliative care specialist in a custom user interface. The palliative care specialist will reach out to primary care teams if she determines that the patient would benefit from palliative care. If the primary care provider agrees, he/she would write a palliative care consult order for the patient. The goal is to reduce the time to palliative care for these patients, who may not have been identified as quickly without the algorithm.
Interventions
OTHERPalliative care contacts primary care
Palliative care specialist reaches out to primary care to recommend a palliative care consult. If the primary care provider agrees, he/she will write an order for a palliative care consult.
Sponsors
Mayo Clinic
Study design
Allocation
RANDOMIZED
Intervention model
CROSSOVER
Primary purpose
SCREENING
Masking
NONE
Intervention model description
Step-wedge design with 7 wedges: the first wedge has all primary care teams in the standard of care arm; every six weeks one or two care teams switch to the intervention arm.
Eligibility
Sex/Gender
ALL
Age
18 Years to No maximum
Healthy volunteers
No
Inclusion criteria
* Adult patient assigned to a primary care unit from July 2020 to June 2021. * Weekly the palliative care specialists will select patients by looking at patients in sorted order starting with the highest score and proceeding down the list and evaluating each patient for
Exclusion criteria
.
Design outcomes
Primary
| Measure | Time frame | Description |
|---|---|---|
| Timely identification for need of palliative care | Through study completion, an average of 1 year | Time to electronic record of consult by the palliative care team in the outpatient setting |
Secondary
| Measure | Time frame | Description |
|---|---|---|
| Number of advanced care planning notes documented in the EHR | Through study completion, an average of 1 year | Number of advanced care planning notes documented in the EHR on both arms |
| Number of billing codes for palliative care | Through study completion, an average of 1 year | Number of ICD-10 billing codes for palliative care on both arms |
| Number of palliative care consults | Through study completion, an average of 1 year | Number of palliative care consults that occurred on intervention and standard of care arms |
| Percent of patients who are eligible for ECH based palliative care | Through study completion, an average of 1 year | Percent of patients who are eligible for employee/community health (ECH) based palliative care compared to the Palliative Care Clinic. |
| Percent agreement between Palliative Care and Primary Care and average time between Primary Care Contact and Response | Through study completion, an average of 1 year | Agreement statistics (percent agreement and Kappa statistics) between Palliative Care and Primary Care and descriptive statistics (mean, etc.) on time between primary care contact and response. |
| Positive predictive value of screened patients | Through study completion, an average of 1 year | Percentage of screened patients that received palliative care consults |
Countries
United States
Outcome results
None listed