In the News: Feb 13–20
Analytics Enhance Care Coordination, Reduce ED Usage
A recent Healthcare Finance article highlights how the Physicians of Southwest Washington (PSW) are taking an analytical approach to reducing ED utilization by focusing on social determinants of health. PSW identified gaps across its network and tracked ED utilization, along with care management referrals, using analytic insights. PSW also implemented a program where patients discharged from the ED were contacted within 48 business hours to determine whether they needed help with transitioning their care. Additionally, during these outreach efforts, patients were informed of alternative facilities such as urgent care and walk-in clinics.
Through its newly implemented outreach program and data-driven analysis of care management, PSW’s ED utilization decreased by 8% within a year. This work has been especially relevant as PSW transitioned to an accountable care organization (ACO) model at the beginning of 2017. In an ACO, it is highly important that providers treat patients as a whole and understand their movement across the care continuum.
As health care delivery shifts to value-based care, it will be financially rewarding to incorporate population health initiatives into decision making. As health systems seek to scale value-based initiatives, they will need to reassess organizational structure to optimize care delivery. To learn more about management models Sg2 has identified to help organizations meet the needs of value-based care adoption, please read the Sg2 report Organizational Structures to Advance Value-Based Care.
Machine Learning Can Inform Clinical Decisions
A recent study in Nature Biotechnology suggests machine learning can identify complex patterns in brain activity and help inform clinical decision making. Major depression affects about 7% of adults in the US, but symptoms vary by individual. Currently, there are some evidence-based options to treat depression, but, often, finding a treatment that works best for an individual requires trial and error. Researchers have discovered a neural signature that predicts whether patients with depression can benefit from a commonly prescribed antidepressant medication.
As part of the study, participants with depression were placed in 2 groups for 8 weeks: an antidepressant group or a placebo group. Researchers applied a new machine learning algorithm, specialized for reading electroencephalogram (EEG) data, to the participants’ pretreatment EEG data to see if the machine learning technique could identify a model that predicted depressive symptoms after treatment. The algorithm was able to reliably predict individual patient responses to the antidepressant medication based on a specific type of brain signal. Researchers also found that the algorithm predicted improvement for participants showing partial responses to at least one antidepressant compared to others who hadn’t responded to 2 or more medications, in line with clinical outcomes.
Artificial intelligence (AI)–based solutions have moved from being a novelty to producing novel solutions to complex health care challenges. These solutions are only scratching the surface of the value they will deliver over time, but they have positioned AI to disrupt nearly every aspect of the clinical enterprise. AI is not only already being utilized, but it is also improving clinical care across service lines. Please read the Sg2 Expert Insight AI is Here to Stay: Adopting Artificial Intelligence to Service Line Strategy (Part 1) to learn more about how AI is disrupting the industry and how it has been applied in different clinical settings.
Data Measure Value and Deliver Whole Person Care
A recent Healthcare Finance article describes how the Institute for Human Caring at Providence St Joseph Health is using data to deliver what they call “whole person care.”
Metrics around advance care planning, goals of care, shared decision making, symptom management, palliative care and hospice care were the primary focus of the institute, as these areas require clear understanding of—and acting in accordance with—patients’ priorities and wishes. These integrated measures are centrally managed and used to align treatment with a patient’s goals and care plan. Monitoring these metrics allows the Institute for Human Caring to show the value of care being received by the patient.
As the shift toward value continues, using analytics to demonstrate value is becoming increasingly more important. The Sg2 Analytics platform is an online set of analytics tools designed to provide critical information health care organizations need to grow and improve their business. To learn more about Sg2 Analytics, please read the Sg2 Analytics Getting Started Guide.
Tags: ACO, AI, analytics, behavioral health, clinical decision making, clinical outcomes, data, ED utilization, machine learning, metrics, social determinants of health, value-based care, whole person care