Inpatient rehabilitation facilities (IRF) and acute rehabilitation are coming under increased scrutiny in the era of pay for performance and bundled care. While there is some evidence that IRF may be superior for many patents who need post acute care (PAC) in an inpatient setting, pressure has been made to focus on skilled nursing facility (SNF) based subacute rehabilitation settings based only on cost assessments. However, new data mining and assessment tools can allow for assessment of the efficacy of outcomes of IRF and SNF based rehabilitation. These assessments can combine and compare outcomes based on medical data set (MDS) and uniform data set data (UDS) to allow for a more appropriate assessment of the efficacy of patents based on complex profiles of their social, medical and functional profiles. Using this data, there can be 3 achievable goals: 1) assessment of PAC partner institutions to compare outcomes with peers (goal attainment, readmissions, complications, return to community) 2) assessment of appropriate type of PAC setting (IRF vs SNF) for best match of possible outcomes with patient type 3) assessment of potential areas for reduction of overuse of PAC days to decrease over utilization of inpatient PAC and allow for better initiation of the appropriate post discharge care in the community. Knowledge of the strength of these predictions can help to place rehabilitation medicine at the center of PAC planning in hospital networks and systems.

Learning Objectives

  • Understand the need for IRF and SNF rehabilitation physiatrists as well as acute care consulting physiatrists to learn the different strengths and weaknesses of these types of facilities.
  • Learn how to utilize big data from the MDS and UDS data sets to help to evaluate the relative merits of PAC provider settings (IRF vs SNF) as well as between providers in a given setting.
  • Learn how to utilize big data to assess the relative risk/benefit for an individual patient when choosing among multiple PAC providers.