If you are, or have, a provider group where coding and billing occurs external to the facility, this article is for you. I have seen many variations of transmission reconciliation and have come up with my version of how it should work. It is not perfect, but it is logical, and it is always where I start when something does not make sense.
How do you know if you have received everything you were supposed to receive from a facility for your external coding and billing?
Short answer: If you do not have solid reconciliation processes, then you won't know.
Reconciliation is critical if your processes are external to the facility system since you are getting passed your data via programming. It does not matter how well you program your data transmissions, there will still be gaps. Also, your programming, or ideal of what you should receive in your data transmissions, has specific dependencies.
Considerations for Data Transmissions:
The Facility ADT - Is the registration system separate from the medical record system?
The Facility EMR/EHR - Is the medical record system separate from the registration system?
The Version of the System(s) - What is the system's capabilities for the current version?
The Historical Experience - How much experience do the facility and RCM vendor teams have related to the specific feed/feed types and transmissions?
The Type of Unit(s) - Can specific unit(s) or providers be isolated within the ADT/EHR?
The Outlier Care - Is there any care rendered that would fall outside traditional unit programming?
Once you have a baseline knowledge for the facility and RCM vendor side, you can easily target the best processes for your integration and build out reconciliation. The ideal place to start is determining what your source of truth will be for all transmissions. In my experience, the ideal data type to use is the Census Log.
Census Log - Traditionally a daily report that is run for a specific unit. Note that if the care being rendered is not isolated to a specific facility unit(s), it is difficult to ensure you are programming your report as the source of truth.
Lets assume that you have a unit (like an emergency department/ED) that can be isolated easily. We will request a daily facility census log report for any patients that were registered in the ED. This report is our baseline. If a patient was an ED registered patient, then they should be on that list.
Next we are going to need to use this report against something. There are two components we will measure against. The first is the ADT feed, or the patient demographics and insurance. For this example we are going to assume we have a ADT feed. We should get a set of this data for each patient that is on the Census Log.
Reconciliation Check Point 1
If we had 100 patients on our census log, but only received the ADT for 90 unique patients, then we have a gap.
- Are there non-ED patients being pulled into the Census Log?
- Is the ADT feed sending more patients than those registered into the ED?
- How are the Census Log and ADT programming factoring patient types like transfers from another facility, or direct admits?
- Was there an ADT transmission failure, or middleware programing issue that prevented the data from being loaded?
Once we have vetted out the possible source(s), then we can move on. Lets assume that we found the feed data of our 10 missing patients due to a middleware processing issue. Next we need to receive the documentation from the instance of care. So we have 100 patient we are looking for, based on our Census Log report. Then we received the ADT data so we created the patient shell. Now we can await the documentation for the specific date(s) of service. This is where it may get tricky, depending on the documentation transmission type of PDF, ORU, or MDM. Each comes with its own complexities. Keeping it simple, we will say PDF.
Reconciliation Check Point 2
In this case, what would happen if we received 80 patient records out of the anticipated 100?
- Did the Middleware stop any records?
- Did the facility have a downtime?
- Does the facility have programming holds for incomplete vs complete records?
- Does the facility have programming for Non-billable accounts based on the unit care?
This step again takes some discovery to determine why records were not received. Often I find it to be either non-billables, or if an entire day then a downtime at the facility. Various combinations are also possible. Once all billable encounters are determined and accounted for, the last step can occur. This is where we ensure all components of the record have been received. Did we get all signatures, documentation elements, orders, nursing notes, etc? Or were some missing or incomplete?
Reconciliation Check Point 3
Our final assumption will be that after we received the records, we pulled off 5 charts as non-billables for various reasons. Our final total coming to 95 billable encounters. What if on 10 of these charts documentation was missing or incomplete?
- Was the documentation gap due to a timing issue (record transmitted on day 3, but provider completed account on day 4)?
- What if there was a physician and a mid-level provider both caring for the patient, the account was signed off but the mid-level added something, therefore making the provider's signature invalid and needing re-sign off?
- Did the provider put the exact time of critical care in minutes, or the exact length of a laceration?
There are many deficiencies that can occur, and are slightly different depending on the unit. These deficiencies will need to be sent back via a notification process to the provider to correct. Then those 10 records can be coded with the other 85 and get over to the billing team.
That is it! Just repeat that process on every date of service, for every provider, for every unit, for those 3 checks until you are reconciled for that day. As long as there are services being rendered, there will be reconciliation needed. Over time you can enhance your process based on what you know about the facility, its programming, your middleware, how providers behave on certain shifts, or patient types. The longer you have the process in place, the cleaner it becomes. Also, with the various connection types and interfaces between the facility health system and the coding and billing team, the more processes and variances you will learn and be able to solve for ahead of time.