Impact of augmented intelligence (AI) on utilization of palliative care (PC) services in oncology.

2020 ◽  
Vol 38 (15_suppl) ◽  
pp. 12015-12015
Author(s):  
Ajeet Gajra ◽  
Marjorie Zettler ◽  
Jonathan Kish ◽  
Kelly Miller ◽  
John Frownfelter ◽  
...  
Keyword(s):  
Palliative Care ◽  
End Of Life ◽  
Machine Learning Techniques ◽  
Care Plan ◽  
Patient Specific ◽  
The Mean ◽  
Poor Outcomes ◽  
Oncology Practice ◽  
Practice Methods ◽  
The Impact

12015 Background: Timely integration of palliative care in the management of patients with advanced cancer is a quality benchmark in oncology. However, PC is often underutilized as evidenced by delays in identification of appropriate patients, in referrals to a PC service, and in enrollment to hospice. Jvion has developed a prescriptive analytics solution, the Machine, which combines AI algorithms with machine learning techniques and applies them to clinical and exogenous datasets to identify patients with a propensity for poor outcomes. The Machine was applied to risk for patients’ mortality within next 30 days, and recommended patient-specific, dynamic, and actionable insights. Use of the Machine requires no additional documentation within the electronic health record (EHR) and the insights generated can be integrated back in to any EHR to help inform the care plan. Herein, we report the results of a study evaluating the impact of AI-driven insights on PC utilization at a large community oncology practice. Methods: All patients were scored weekly using the Machine PC vector. The Machine risk stratified the patients and generated recommendations for the provider to consider as they developed a care plan. Patients identified as “at risk” by the Machine were assessed for a supportive care visit (PC referral) and then were referred as deemed clinically appropriate. The average monthly rates of PC consults and hospice referrals were calculated 5 months prior to and for 17 months after the launch of the Machine in the practice. Results: The oncology practice has 21 providers managing an average of 4329 unique patients per month (PPM). The mean rate of PC consults increased from 17.3 to 29.1 per 1000 PPM pre and post Machine deployment respectively (+168%). The mean monthly rate of hospice referrals increased by 8-fold from 0.2 to 1.6 per 1000 PPM pre and post deployment respectively. Eliminating the first 6 months of Machine deployment to account for user learning curve, the mean rates of monthly PC consults nearly doubled over baseline to 33.0, and hospice referrals rose 12-fold to 2.4 per 1000 patients in months 7-17 post Machine deployment. Conclusions: This oncology practice found deployment of this novel AI solution to be feasible and effective at generating actionable insights. These AI driven insights could be incorporated into workflow and improved the decision-making for whether and when a patient should be referred to PC and/or hospice services for end of life care. Further study is needed to confirm the value of AI for management of cancer patients at end of life.

Impact of augmented intelligence (AI) on identification and management of depression in oncology.

2020 ◽  
Vol 38 (15_suppl) ◽  
pp. e14059-e14059
Author(s):  
John Frownfelter ◽  
Sibel Blau ◽  
Marjorie Zettler ◽  
Kelly Miller ◽  
Jonathan Kish ◽  
...  
Keyword(s):  
Case Management ◽  
Machine Learning Techniques ◽  
Depression Screening ◽  
Care Plan ◽  
Patient Specific ◽  
Patients With Cancer ◽  
Depression Risk ◽  
Number Of Patients ◽  
Oncology Practice ◽  
The Impact

e14059 Background: Depression is common in patients with cancer and is associated with worse cancer treatment outcomes. Depression is often underdiagnosed/treated as cancer clinicians are focused on the complex aspects of therapy and care coordination. AI has a potential application in the identification of patients at high risk for depression. Jvion has developed a prescriptive analytics solution (the Machine), which uses AI algorithms and machine learning techniques applied to combined clinical and exogenous datasets to identify patients with a propensity for poor clinical outcomes. The Machine was applied to depression risk (within next 6 months), and recommended patient-specific, dynamic, and actionable insights. While the Machine requires no additional documentation within the electronic health record (EHR) to generate its insights, those insights can be integrated back in to any EHR. Herein, we report the results of a pilot study evaluating the impact of AI-driven insights on depression screening and management at a single oncology practice. Methods: All patients were scored weekly using the Machine depression vector. The Machine risk-stratified the patients and generated recommendations for the provider to consider as they developed a care plan. Patients identified as “at risk” by the Machine were assessed for depression (PHQ-9) by the clinical team regardless of prior screening results. The rate per 1000 unique patients per month (PPM) of depression screenings, case management evaluations, and antidepressant prescriptions were calculated for the 5 months prior to and 17 months post deployment of the Machine in the practice. Results: The oncology practice has 21 providers managing an average of 4329 unique PPM. The mean rate of depression screenings increased from 6.0 per 1000 PPM pre- deployment to 16.2 per 1000 PPM post deployment (+271%). The downstream workflow outcomes of case management evaluations increased from 11.6 to 21.4 per 1000 PPM (+184%) and antidepressant prescriptions increased from 9.2 to 15.5 per 1000 PPM (+168%) pre and post-implementation respectively. The providers reported high satisfaction with the use of the AI solution in depression screening. Conclusions: This oncology practice found deployment of the Jvion AI solution to be feasible. The Machine-generated insights for depression risk were actionable, could be incorporated into workflow, and increased the number of patients identified. If confirmed in larger studies, AI-driven insights may improve the identification and management of depression in patients with cancer.

scholarly journals Improving End-of-Life Care: Palliative Care Embedded in an Oncology Clinic Specializing in Targeted and Immune-Based Therapies

2017 ◽  
Vol 13 (9) ◽  
pp. e729-e737 ◽  
Author(s):  
David J. Einstein ◽  
Susan DeSanto-Madeya ◽  
Matthew Gregas ◽  
Jessica Lynch ◽  
David F. McDermott ◽  
...  
Keyword(s):  
Palliative Care ◽  
Usual Care ◽  
End Of Life ◽  
Odds Ratio ◽  
Care Clinic ◽  
Usual Practice ◽  
Embedded Model ◽  
Enrollment Rates ◽  
Oncology Practice ◽  
The Impact

Purpose: Patients with advanced cancer benefit from early involvement of palliative care. The ideal method of palliative care integration remains to be determined, as does its effectiveness for patients treated with targeted and immune-based therapies. Materials and Methods: We studied the impact of an embedded palliative care team that saw patients in an academic oncology clinic specializing in targeted and immune-based therapies. Patients seen on a specific day accessed the embedded model, on the basis of automatic criteria; patients seen other days could be referred to a separate palliative care clinic (usual care). We abstracted data from the medical records of 114 patients who died during the 3 years after this model’s implementation. Results: Compared with usual care (n = 88), patients with access to the embedded model (n = 26) encountered palliative care as outpatients more often ( P = .003) and earlier (mean, 231 v 109 days before death; P < .001). Hospice enrollment rates were similar ( P = .303), but duration was doubled (mean, 57 v 25 days; P = .006), and enrollment > 7 days before death—a core Quality Oncology Practice Initiative metric—was higher in the embedded model (odds ratio, 5.60; P = .034). Place of death ( P = .505) and end-of-life chemotherapy (odds ratio, 0.361; P = .204) did not differ between the two arms. Conclusion: A model of embedded and automatically triggered palliative care among patients treated exclusively with targeted and immune-based therapies was associated with significant improvements in use and timing of palliative care and hospice, compared with usual practice.

Impact of integrated palliative care model on end-of-life (EOL) quality metrics for patients with kidney cancer (RCC) and melanoma (M).

2015 ◽  
Vol 33 (29_suppl) ◽  
pp. 137-137
Author(s):  
Jessica A. Lynch ◽  
Susan DeSanto-Madeya ◽  
Jessica A. Zerillo ◽  
Matt Gregas ◽  
David F. McDermott ◽  
...  
Keyword(s):  
Palliative Care ◽  
End Of Life ◽  
Symptom Assessment ◽  
Integrated Model ◽  
Optimal Timing ◽  
Early Palliative Care ◽  
Location Of Death ◽  
Edmonton Symptom Assessment Scale ◽  
Oncology Practice ◽  
The Impact

137 Background: Early palliative care (PC) improves quality of life (QOL) and enhances end-of-life (EOL) care, but the optimal timing and most effective model for integrating PC into oncologic care is uncertain. To understand the impact of an integrated model with PC providers embedded with oncologists vs. usual care (UC) with referral at the discretion of the same oncologists, we examined the timing and delivery of PC and Quality Oncology Practice Initiative (QPOI) EOL metrics among patients with RCC and M in a single clinic. We hypothesized that integrated PC would result in more referrals, earlier contact with PC and better QOPI EOL metrics compared with UC. Methods: In a retrospective cohort study of patients with RCC and M in the Beth Israel Deaconess Biologics Clinic who expired between 10/1/12 and 12/31/14, we compared patients seen 2 days/week, when referral to PC was discretionary, with a third day when PC providers shared the clinic for real-time consultations. Patients were identified as meeting PC eligibility if they had recurrent, metastatic disease and were on active treatment or had a symptom severity of 7+ on Edmonton Symptom Assessment Scale (ESAS). Two oncologists saw all patients, regardless of day. Results: Seventy-six patients expired, 19 in the Integrated PC model and 57 with UC. Patients were similar with respect to diagnosis and demographics except for smoking. The integrated model substantially improved timing and location of PC. In the integrated PC model, 85% were seen by PC compared with 45% in UC (P = 0.002). All patients in the integrated model began PC as an outpatient compared with 36% in UC (P < 0.001). The mean number of days from first PC contact to death was 28 (SD = 54) for UC and 118 (SD = 120) with integrated PC (P < 0.001). The location of death did not differ significantly between models, occurring outside the hospital with hospice among 71% of patients in the integrated model and 53% in UC (P = 0.25). Results were similar in relative risk models adjusted for smoking. Conclusions: A practice model that integrated PC with oncologic care was associated with more PC referrals, earlier contact, and a nonsignificant trend toward fewer deaths in hospital and ICU.

scholarly journals Impact of Augmented Intelligence on Utilization of Palliative Care Services in a Real-World Oncology Setting

2021 ◽  
pp. OP.21.00179
Author(s):  
Ajeet Gajra ◽  
Marjorie E. Zettler ◽  
Kelly A. Miller ◽  
John G. Frownfelter ◽  
John Showalter ◽  
...  
Keyword(s):  
Palliative Care ◽  
Mortality Risk ◽  
Real World ◽  
Short Term ◽  
Short Term Mortality ◽  
Medium Risk ◽  
Palliative Care Services ◽  
The Mean ◽  
Oncology Practice ◽  
The Impact

PURPOSE: For patients with advanced cancer, timely referral to palliative care (PC) services can ensure that end-of-life care aligns with their preferences and goals. Overestimation of life expectancy may result in underutilization of PC services, counterproductive treatment measures, and reduced quality of life for patients. We assessed the impact of a commercially available augmented intelligence (AI) tool to predict 30-day mortality risk on PC service utilization in a real-world setting. METHODS: Patients within a large hematology-oncology practice were scored weekly between June 2018 and October 2019 with an AI tool to generate insights into short-term mortality risk. Patients identified by the tool as being at high or medium risk were assessed for a supportive care visit and further referred as appropriate. Average monthly rates of PC and hospice referrals were calculated 5 months predeployment and 17 months postdeployment of the tool in the practice. RESULTS: The mean rate of PC consults increased from 17.3 to 29.1 per 1,000 patients per month (PPM) pre- and postdeployment, whereas the mean rate of hospice referrals increased from 0.2 to 1.6 per 1,000 PPM. Eliminating the first 6 months following deployment to account for user learning curve, the mean rate of PC consults nearly doubled over baseline to 33.0 and hospice referrals increased 12-fold to 2.4 PPM. CONCLUSION: Deployment of an AI tool at a hematology-oncology practice was found to be feasible for identifying patients at high or medium risk for short-term mortality. Insights generated by the tool drove clinical practice changes, resulting in significant increases in PC and hospice referrals.

Improving end-of-life care: Palliative care embedded in an outpatient oncology clinic.

2017 ◽  
Vol 35 (8_suppl) ◽  
pp. 77-77
Author(s):  
David Johnson Einstein ◽  
Susan DeSanto-Madeya ◽  
Matt Gregas ◽  
Jessica A. Lynch ◽  
David F. McDermott ◽  
...  
Keyword(s):  
Palliative Care ◽  
Usual Care ◽  
End Of Life ◽  
Care Clinic ◽  
Usual Practice ◽  
Referral Criteria ◽  
Palliative Care Teams ◽  
Embedded Model ◽  
Oncology Practice ◽  
The Impact

77 Background: Patients with advanced cancer benefit from early involvement of palliative care. Nonetheless, the ideal method of palliative care integration remains to be determined. Prior studies proposed automatic referral criteria and embedding palliative care teams within specialty clinics. Methods: We studied the impact of an embedded palliative care team that saw patients in an academic oncology clinic based on automatic referral criteria. Patients seen in this clinic on a specific day had access to the “embedded” model, whereas patients seen on two other days could access a separate palliative care clinic upon oncologist referral (usual care). We abstracted data from the medical records of 118 patients who were cared for in this oncology clinic and died during the 3 years following implementation of the embedded model. Results: Compared with those with access to usual care (n = 88), patients with access to the embedded model (n = 30) encountered palliative care as outpatients more often (p < 0.001) and twice as long before death (mean 223 versus 106 days, p = 0.001). Hospice enrollment rates were similar (p = 0.717) but duration was twice as long (mean 53.5 versus 25.3 days, p = 0.03), and enrollment greater than 7 days before death—a core Quality Oncology Practice Initiative metric—was significantly higher in the embedded model (OR 5.60, p = 0.034). Place of death (p = 0.505) and end-of-life chemotherapy (OR 0.361, p = 0.204) did not differ significantly. Conclusions: A model of embedded palliative care with automatic referral criteria, compared with usual practice, was associated with significant improvements in utilization and timing of palliative care and hospice.

Cannabis and End-of-Life Care: A Snapshot of Hospice Planning and Experiences Among Illinois Medical Cannabis Patients With A Terminal Diagnosis

2021 ◽  
pp. 104990912110186
Author(s):  
James Alton Croker ◽  
Julie Bobitt ◽  
Sara Sanders ◽  
Kanika Arora ◽  
Keith Mueller ◽  
...  
Keyword(s):  
Palliative Care ◽  
End Of Life ◽  
Fast Track ◽  
Hospice Care ◽  
Cannabis Use ◽  
Care Plan ◽  
Medical Cannabis ◽  
Cross Sectional Survey ◽  
Cross Sectional ◽  
Eol Care

Introduction: Between 2013 and 2019, Illinois limited cannabis access to certified patients enrolled in the Illinois Medical Cannabis Program (IMCP). In 2016, the state instituted a fast-track pathway for terminal patients. The benefits of medicinal cannabis (MC) have clear implications for patients near end-of-life (EOL). However, little is known about how terminal patients engage medical cannabis relative to supportive care. Methods: Anonymous cross-sectional survey data were collected from 342 terminal patients who were already enrolled in ( n = 19) or planning to enroll ( n = 323) in hospice for EOL care. Logistic regression models compare patients in the sample on hospice planning vs. hospice enrollment, use of palliative care vs. hospice care, and use standard care vs non-hospice palliative care. Results: In our sample, cancer patients ( OR = 0.21 (0.11), p < .01), and those who used the fast-track application into the IMCP ( OR = 0.11 (0.06), p < .001) were less likely to be enrolled in hospice. Compared to patients in palliative care, hospice patients were less likely to report cancer as their qualifying condition ( OR = 0.16 (0.11), p < .01), or entered the IMCP via the fast-track ( OR = 0.23 (0.15), p < .05). Discussion: Given low hospice enrollment in a fairly large EOL sample, cannabis use may operate as an alternative to supportive forms of care like hospice and palliation. Clinicians should initiate conversations about cannabis use with their patients while also engaging EOL Care planning discussions as an essential part of the general care plan.

scholarly journals 77 National Audit of Care At the End of Life (NACEL)—Round Two Findings

2021 ◽  
Vol 50 (Supplement_1) ◽  
pp. i12-i42
Author(s):  
D Hibbert
Keyword(s):  
Palliative Care ◽  
End Of Life ◽  
Care Service ◽  
Palliative Care Service ◽  
Care Plan ◽  
Specialist Palliative Care ◽  
Experience Of Care ◽  
Outcomes Of Care ◽  
Specialist Palliative Care Service ◽  
The Right

Abstract   NACEL is a national comparative audit of the quality and outcomes of care experienced by the dying person and those important to them during the final admission in acute and community hospitals in England and Wales. Mental health inpatient providers participated in the first round but excluded from the second round. NACEL round two, undertaken during 2019/20, comprised: Data was collected between June and October 2019. 175 trusts in England and 8 Welsh organisations took part in at least one element of NACEL (97% of eligible organisations). Key findings include Recognising the possibility of imminent death: The possibility that the patient may die was documented in 88% of cases. The median time from recognition of dying to death was 41 hours (36 hours in the first round). Individual plan of care: 71% of patients, where it had been recognised that the patient was dying (Category 1 deaths), had an individualised end of life care plan. Of the patients who did not have an individualised plan of care, in 45% of these cases, the time from recognition of dying to death was more than 24 hours. Families’ and others’ experience of care: 80% of Quality Survey respondents rated the quality of care delivered to the patient as outstanding/excellent/good and 75% rated the care provided to families/others as outstanding/excellent/good. However, one-fifth of responses reported that the families’/others’ needs were not asked about. Individual plan of care: 80% of Quality Survey respondents believed that hospital was the “right” place to die; however, 20% reported there was a lack of peace and privacy. Workforce Most hospitals (99%) have access to a specialist palliative care service. 36% of hospitals have a face-to-face specialist palliative care service (doctor and/or nurse) available 8 hours a day, 7 days a week. NACEL round three will start in 2021.

scholarly journals Video Intervention to Increase Knowledge About and Interest in Palliative Care Among MSW Students

2020 ◽  
Vol 4 (Supplement_1) ◽  
pp. 70-70
Author(s):  
Cathy Berkman
Keyword(s):  
Palliative Care ◽  
Social Work ◽  
End Of Life ◽  
Social Workers ◽  
End Of Life Care ◽  
Social Work Students ◽  
Life Care ◽  
Video Intervention ◽  
Msw Students ◽  
The Mean

Abstract As the population ages and more people live longer with chronic and life-limiting illnesses, more healthcare professionals with palliative care skills are needed. Social workers are part of the palliative care team, but there is little, if any, content on palliative and end-of-life care in MSW programs. A 24-minute video featuring nine palliative and hospice social workers was produced with two goals: 1) increase knowledge of social work students about palliative and end-of-life care; and 2) interest social work students in a career in palliative social work. MSW students from three schools, in NY and Alabama, viewed the video. After viewing the video, 94 students participated in the mixed methods study, completing the brief, anonymous, online survey. The mean level of understanding about what palliative social workers do, rated from 1 (no understanding) to 5 (very good understanding), was 2.96 (SD=.99) before viewing the video and 4.31 (SD=.61) after, for an increase of 1.35 points (95% CI=1.14, 1.55) (p&lt;.001). The mean level of interest in a career in palliative care social work and working with seriously ill persons and their family members, rated from 1 (Not at all interested) to 5 (Extremely interested), was 2.52 (SD=.99) before viewing the video and 3.45 SD=.80) after, for an increase of .91 points (95% CI=.79, 1.07) (p&lt;.001). Qualitative data supporting the quantitative findings will be presented. This study suggests that a video intervention may be an effective tool to increase knowledge and interest in palliative and end-of-life care among social work students.

A Case–Control Study Evaluating the Impact of Dedicated Palliative Care Training on Critical Care Interventions at the end of Life

2021 ◽  
pp. 082585972110374
Author(s):  
Jee Y. You ◽  
Lie D. Ligasaputri ◽  
Adarsh Katamreddy ◽  
Kiran Para ◽  
Elizabeth Kavanagh ◽  
...  
Keyword(s):  
Palliative Care ◽  
Critical Care ◽  
High Risk ◽  
End Of Life ◽  
Poor Prognosis ◽  
Medical Interventions ◽  
Risk Of Death ◽  
Before And After ◽  
The Difference ◽  
The Impact

Many patients admitted to intensive care units (ICUs) are at high risk of dying. We hypothesize that focused training sessions for ICU providers by palliative care (PC) certified experts will decrease aggressive medical interventions at the end of life. We designed and implemented a 6-session PC training program in communication skills and goals of care (GOC) meetings for ICU teams, including house staff, critical care fellows, and attendings. We then reviewed charts of ICU patients treated before and after the intervention. Forty-nine of 177 (28%) and 63 of 173 (38%) patients were identified to be at high risk of death in the pre- and postintervention periods, respectively, and were included based on the study criteria. Inpatient mortality (45% vs 33%; P = .24) and need for mechanical ventilation (59% vs 44%, P = .13) were slightly higher in the preintervention population, but the difference was not statistically significant. The proportion of patients in whom the decision not to initiate renal replacement therapy was made because of poor prognosis was significantly higher in the postintervention population (14% vs 67%, P = .05). There was a nonstatistically significant trend toward earlier GOC discussions (median time from ICU admission to GOC 4 vs 3 days) and fewer critical care interventions such as tracheostomies (17% vs 4%, P = .19). Our study demonstrates that directed PC training of ICU teams has a potential to reduce end of life critical care interventions in patients with a poor prognosis.

scholarly journals End-Of-Life Discussions Reduce The Utilization of Life-Sustaining Treatments During The Last Three Months of Life in Cancer Patients : A Nationwide Population-Based Cohort Study

2021 ◽  
Author(s):  
Shang-Yih Chan ◽  
Yun-Ju Lai ◽  
Yu-Yen Hsin Chen ◽  
Shuo-Ju Chiang ◽  
Yi-Fan Tsai ◽  
...  
Keyword(s):  
Cohort Study ◽  
Cardiopulmonary Resuscitation ◽  
End Of Life ◽  
Cancer Patients ◽  
Medical Records ◽  
Population Based ◽  
Lower Likelihood ◽  
The Mean ◽  
The Impact ◽  
To Receive

Abstract Purpose Studies to examine the impact of end-of-life (EOL) discussions on the utilization of life-sustaining treatments near death were limited and had inconsistent findings. This nationwide population-based cohort study determined the impact of EOL discussions on the utilization of life-sustaining treatments in the last three months of life in Taiwanese cancer patients. Methods This cohort study included adult cancer patients from 2012–2018, which were confirmed by pathohistological reports. Life-sustaining treatments during the last three months of life included cardiopulmonary resuscitation, intubation, and defibrillation. EOL discussions in cancer patients were confirmed by their medical records. Association of EOL discussions with utilization of life-sustaining treatments were assessed using multiple logistic regression. Results Of 381,207 patients, the mean age was 70.5 years and 19.4% of the subjects utilized life-sustaining treatments during the last three months of life. After adjusting for other covariates, those who underwent EOL discussions were less likely to receive life-sustaining treatments during the last three months of life compared to those who did not (Adjusted odds ratio [AOR]: 0.82; 95% confidence interval [CI]: 0.80–0.84). Considering the type of treatments, EOL discussions correlated with a lower likelihood of receiving cardiopulmonary resuscitation (AOR = 0.43, 95% CI: 0.41–0.45), endotracheal intubation (AOR = 0.87, 95%CI: 0.85–0.89), and defibrillation (AOR = 0.52, 95%CI: 0.48–0.57). Conclusion EOL discussions correlated with a lower utilization of life-sustaining treatments during the last three months of life among cancer patients. Our study supports the importance of providing these discussions to cancer patients to better align care with preferences during the EOL treatment.

Sign in / Sign up

Export Citation Format

Share Document