The Perfect is the Enemy of the Good
Before I started working for Remedy Partners as a Transitional Care Specialist assisting a Model 2 Acute Care Hospital Episode Initiator (EI) in identifying, engaging, and tracking BPCI patients during their episodes of care, one of the nurses preparing me for the job told me that “if you accomplish nothing else… get them to pick up the phone.” Her point was that if we are not able to speak to our patients after discharge then it will be difficult, if not impossible, to influence the outcome of that patient’s recovery. This was some of the best advice that I received but it assumed that we had already identified the patient as a BPCI patient.
Patient identification is the first and often most troublesome aspect of the BPCI program. I often borrow a term from my days in the military to describe patient ID as a “moving target”. This is because to be a participating patient in the BPCI program one must be admitted to an EI, have a Diagnosis Related Group (DRG) billed to Medicare that the EI selected for their program, and not meet one of a few exclusionary criteria set by Medicare.
The diagnosis portion of this equation is the elusive variable and, unfortunately for those of us seeking them out but fortunately for patients, they will not linger in a medical facility just because we have not been able to successfully identify them as a BPCI patient. I have experienced instances where a patient was not identified until months after their 90-day episode had completed. Simply put, patient identification is a task that will create immense frustration for perfectionists.
The easiest way to identify BPCI patients is when those patients are coming to a medical facility for pre-scheduled procedures. This patient population can be identified and engaged before their episodes are ever initiated. Unfortunately, only a small portion of patients fall into the pre-scheduled bucket and if your program does not include orthopedic or cardiac surgery bundles then the opportunity for identification will likely begin with the admission to the EI.
Using a Working or Possible DRG
Identifying patients using a Working or Possible DRG assigned by a team of Clinical Documentation Specialists is a sometimes timely and often accurate method of identifying BPCI patients. A high performing CDI team can assign a Working DRG within a day or two of the admission and this method will allow identification of the majority of BPCI patients while they are still admitted to the EI.
This method of identification does lead to identifying patients that will end up “falling out” due to mismatching Working and Final DRGs. One of the big pitfalls of this method is a lack of weekend coverage. Patients admitted late in the week or during the weekend are sometimes discharged before there is a chance to review the chart and code a DRG.
A BPCI-dedicated CDI Nurse or a BPCI-experienced nurse performing a chart review can be the best option for identifying BPCI patients but the patient volume of a BPCI program may not be enough to justify the time and resources it takes to produce a Working or Possible DRG. This is where a software solution can be utilized to predict the DRG.
Using a Predicted DRG Engine
A Predicted DRG Engine, such as the one available in Remedy Partners’ software platform, can help BPCI programs to overcome limitations imposed by a lack of resources. This software absorbs certain information and uses that information to predict the most likely DRG for that patient’s episode of care. When it comes to evaluating information, humans will always trump technology (in my mind at least) but a Predicted DRG Engine can be a valuable addition to the patient ID process.
The best way to succeed when searching for BPCI patients is to avoid getting distracted by perfection. There will always be a portion of patients that are either not identified in a timely manner or misidentified, and it would behoove a program to accept it and establish procedures for dealing with those instances instead of fighting this fact.
Another maxim of identifying patients in the BPCI program is that it is always better to over-identify rather than under-identify. Your program may or may not see a financial reward associated with the work done to improve the care delivered to patients that were incorrectly identified but the patient will benefit. Additionally, if that patient comes back for subsequent episodes, the patient and the program will benefit from the work done during the prior admission.
The BPCI program, at its core, is about doing the right thing for the patient and trusting that the financial savings will follow. Until bundled payments are the standard across the entire Medicare fee-for-service patient population, patient identification will always be troublesome, more or less so depending on the capabilities of your Episode Initiator and/or your Awardee Convener. It takes a certain personality to avoid feeling the need to pull your hair out when searching for BPCI patients. Set up a process as best you can with the resources at your disposal and understand the limitations. Trust your process, dig in and press on!
Daniel Maloney is the Director of Performance Improvement of Remedy Partners.