Managing Errors in Demographic/Clinical Information: Using Tracking Mechanisms
All insurance accounts fall into one of two groups: full resolution (at negotiated fees, where the credits only consist of cash, contract adjustment, and probable bad debt, if any patient balance cannot be collected) or full write-off (where the practice did not comply with a payor-based rule). Past articles illustrate how billing systems should organize information on billed results. One article¹ focuses on reports that can reveal lost revenues attributed to payor rules and on the importance of developing compliance-based coding. A second article² deals with reports that a billing company should be able to provide to track the dictation patterns of practice members, for exams that can be either complete or limited, to determine whether all members are consistent in articulating clinical findings between the simple and complex cases. There is a third component that a practice needs from its billing company or internal system. How does it manage exams that have missing or flawed demographic and/or clinical information that prevents the case from being billed? Ongoing Information Challenges Even though billing for radiology interpretations has been going on for more than 30 years, it can never be assumed that the new generations of radiologists know how to play the compliance game. Medical schools do not have the time or inclination to teach their students how to conduct their clinical affairs according to the rules set by the insurance companies. It is up to the administrative arms of the practice both to understand these rules and to put infrastructure in place to play by them. It has been stated many times that billing for the hospital-based component of a radiology practice involves different requirements than those of an office-based system. This has to do with reliance on the hospital for all clinical and demographic information and the volumes of new accounts per month. The radiology practice, in a hospital system, is the largest submitter of insurance claims in that system—by a factor of 10. A practice that does not account for the combination of volume and information needs will spend needless amounts on the billing process with, at best, mixed collection results. It is becoming an absolute, in these times, that billing systems must be able to interface with the hospital information system (HIS), both for admission demographic data and for clinical findings. The implication is that the system is highly dependent on the hospital for the quality of account data. There will be definite winners and losers in this market; hospitals all have differing systems and coverage populations. Data-field Editing Billing vendors focus on three fundamental operations on the front end of the billing process. The first is the classification of critical demographic and clinical field data from the hospital systems, resulting in daily file extractions via virtual private network (VPN) connection. The second is the performance of automated edits on the field data to isolate missing or erroneous information, and the third is the provision of enhanced tools, to certified coders, that enable them quickly to populate exception lists that document records requiring clinical feedback from the practice of origin. This operation uses RIS accession numbers to help the practice locate the source documents. Timing is a critical component in this process because the clock is moving toward a filing-date deadline. Most non-Medicare payors have become aggressive in requiring providers to file claims within a narrowing window after the date of service. This is especially troublesome in radiology because there are already sequenced time delays built into the department’s RIS; these delays are also built in when the hospital populates demographic files that the billing company can extract. An outline of problem fields can help in managing the important follow-up requirements (see Figure 1).
    I. Demographic problems
  1. Date-of-birth mismatch
  2. Patient-name mismatch
  3. Missing complete demographic record
  4. Missing patient record
  5. Missing responsible-party information
  6. Noncurrent patient information (more than 30 days old)
    II. Duplicate procedures
      III. Referring-physician problems
    1. Blank last name for referring physician
    2. Missing National Provider Identifier and/or referring-physician number
    3. Identical performing and referring physician
      IV. Procedure-related problems
    1. Unknown location of service
    2. Missing location of service
    3. Invalid units field
    4. Nonallowed provider–procedure combination
      V. Coding problems
    1. Absence of new procedure in client fee schedule
    2. Mismatched point of treatment (inpatient, outpatient, emergency department, or office) in procedure record
    3. Inactive procedure code
    4. Blank diagnosis code
    5. Unreadable diagnosis record
    6. Nonallowed procedure–diagnosis combination
    7. Procedure conflict
      VI. Other exceptions
    1. Blank point-of-treatment code
    2. Blank performing provider name
    3. New or unknown provider of service
    4. Invalid date of service
Figure 1. Problem fields, grouped by type. Once a batch of edited records has been produced, both the billing company and the practice will be tasked with locating missing information and updating the account files. An example of a system-generated report used by the account manager (Figure 2) illustrates the monitoring function.
Figure 2. Example report.
This summary report shows estimated dollar amounts that default to the highest possible fee per exam. The backup information is in the form of exception lists that are specific to the missing/flawed field information. The ideal circumstance, concerning demographic or referring-physician information, is that the billing vendor has established a password-based VPN connection to the HIS. All good systems have filters that enable the billing vendor to access only fields relevant to billing needs, keeping all remaining patient information confidential. Follow-up becomes a problem if the hospital is unwilling or unable to provide VPN access. This necessitates having a billing vendor work with an on-site employee of the practice who has been given permission to use a hospital workstation to seek the missing field data. The vendor has to rely on someone outside its control for timely response. Those tasked with managing the exceptions lists will have to monitor accession-number ages constantly to be certain that filing deadlines are not missed. A Delicate Balancing Act The greater challenge lies with the dictated finding. If it is not definitive concerning why the referring physician needs the information or what the clinical finding is, it will be necessary for the practice to dictate an addendum linked to the original accession number; this becomes part of the medical record. The billing vendor is policing the dictation patterns of practice members as a byproduct of the billing function. Certified coders must stay abreast of payor rules that seek justification for a provider claim. The combination of required diagnosis and procedure codes can sometimes differ, depending on the payor. Consider the perspectives of the two parties to the transaction. The radiologist is being asked, for what one hopes is a very small percentage of the annual caseload, to make revisions to his or her findings because a layperson has judged that the payor will reject a claim pertaining to this clinical service. Will egos interfere with this process? Both the billing vendor and the practice will have to adapt to this touchy circumstance. Perhaps the solution lies in having a practice member handle all modifications (assuming that this is practical), or at least act as the contact person for the vendor. The interaction is valuable. A practice should want to know if one or more of its members is jeopardizing its income by habitually dictating cases in a way that increases compliance rejections. The exception lists, combined with the information channeled through a designated member, offer an opportunity to change behavior patterns in a positive (and not ego-threatening) way. Radiology billing generates high volumes of new accounts that require the system marriage of registration data with clinical findings. There are no market statistics on the proportion of radiology cases that cannot be immediately billed due to missing demographic or clinical information. It is small, I hope, because there is high labor intensity involved in seeking out information to correct the errors. It would be wise to use a billing system that not only categorizes the nature of the errors, but also estimates their value to the practice, in charge dollars. A practice with a large population of ongoing accounts that cannot be activated immediately will find that it has to pay a good deal more than necessary for the billing function because the internal system or vendor must move off an automated process to handle information manually.