How PointClickCare is using predictive modelling to reduce risk among Skilled Nursing Facility (SNF) Residents
The Challenge:
CMS introduced value-based care programs, rewarding high-quality outcomes.
The Demand:
Post-acute settings need tools to identify high-risk patients and predict adverse outcomes.
A study by PointClickCare and Schmidt College analyzed SNF EHR data during short-stays to predict risk of re-hospitalizations.
A Shift in Predictive Focus
Conventional solutions for risk prediction typically only cover 30 days.
Single 30-days prediction window
100 days of SNF care prediction window
Providing caregivers with the right information, at the right time enhances:
surveillance | assessments | intervention
Improving the resident’s clinical trajectory and reducing the chance of an adverse event.
Impact of Comprehensive Data
Our sample population of 5,642,474 residents,
totaling 289,433,542 patient days revealed:
EHR data enables prediction among short-stay residents beginning on day one where the greatest risk of re-hospitalization exists.
Incorporating data sources beyond demographics and diagnoses boosts ability to predict risk.
Tailored predictive models using SNF-specific data outperform general models designed for other care settings.
The Value of SNF-Specific Models
Predictive models that provide insights on a per-resident basis enable the daily decision making required among these high-needs populations.
Read the full study
