Auto-Adjudication (AA) Rate -- Metric Definition

This section outlines a definition for the claims Auto-Adjudication (AA) rate.

The AA Rate is a core measure of claims process efficiency. The AA Rate measures the percent of processed claims that are processed without a human touch (auto-adjudicated). Claims operations that process more claims automatically, tend to have a lower administrative cost per claim.

Key Terms. There are a number of terms which make it easier to critical to the AA Rate definition.

1. Total Auto-Adjudicated Claim Count. The numerator of our AA Rate metric is the Total Auto-Adjudicated Claim Count -- the count of claims that were processed without a human touch. In general, a touch, is an adjudication hold or pend. Specifically, a claim that is 'touched' is a claim that recieved a hold or a pend during adjudication which required a human to 'work/develop/fix' the claim. However, what constitutes a 'human' touch can be complicated. Paper claims, for example, are touched during data entry. EDI claims are sometimes 'touched' during front-end provider and member matching processes which occur prior to adjudication. Many times these front end touches are not recorded. Holds and pends are almost always recorded. For that reason, we recommend using an AA Rate definition which defines a touch as a hold or a pend.
2. Total Claim Count. The denominator of our AA Rate metric is the Total Claim Count.
3. Exclusions. Common AA Rate exclusions include adjusted claims.

Frequency. The AA Rate is not a daily operational metric. In general, a payer should monitor the AA Rate each week. A comprehensive AA Rate picture includes the the weekly AA Rate for the whole plan, the AA Rates for each line of business, the top 25 providers, the top 25 self-funded employers.

Operational Uses. The AA Rate is not a metric used to manage daily production (as noted above in the frequency section). It is used on a periodic basis (weekly/monthly/quarterly) to manage claims process efficiency and as a key driver of administrative cost per claim.

1. AA Rate for the Plan. The AA Rate is a measure of efficiency. At the plan level, the AA Rate helps in a number of ways. First it can be used to check staffing levels. Divide the total manual claims by the average claims per day (for an examiner/associate/rep) and you get a staff number -- albiet a rough one. Second it helps with activity costing. In rough terms a claim hold costs approximately $3.00 which, multiplied by total manual claims, provides a rough manual processing cost estimate.
2. AA Rate for a Line of Business(LOB). Certain lines of business may be characterized by hold/pend code patterns that are important to track. For example, for HMO lines of business that require authorizations/referrals it can be helpful to track the AA Rate and top 10 hold/pend reasons to make sure that members and providers are correctly submitting and using authorizations/referrals. Other lines, such as Medicare Secondary, which require a EOB are benefit by tracking denials.
3. AA Rate for a Provider. The AA Rate is an provider metric. A very common operational use of the denial rate is to report it for individual providers, particular large hospital systems -- which tend to be more aggressive about managing claims payment. It also often helps to report the top 5 or 10 denial reasons with the denial rate -- which helps to answer the question of what exactly is being denied for a particular provider.
4. AA Rate for an ASO/Self-Funded Employer. Tracking the AA Rate and hold/pend reasons for self-funded employers is helpful in managing their expectations around administrative costs. It also enables them to manage member related issues (missing auths, eligibility, other benefits...)

Issues, Notes and Challenges. As a metric the AA Rate also has a number of notes and issues.

1. Header versus Line Holds/Pends The manner in which claims systems 'stop' claims -- that is apply a hold or a pend -- varies by system. Today, most platforms allow claims to be held at the header level and/or at the individual line level. Once the system applies a hold/pend the claim is put into a queue for manual review and processing.
2. Input Process Edits. At the front end of the claims process there are sub-processes to input paper claims and load EDI claims. Both of these set of claims pass through processes that often involve 'touches'. On the EDI side, there are steps to match providers and members that sometimes involve manual intervention. On the paper side, there are often small changes made to claims during input. In general, few claims are tagged in a manner that enables back end tracking of these 'touches'. Put another way, given 100 random claims, it would be difficult to determine how many of them had front end edits that involved a 'touch'. As a result, it is generally easier to ignore Input Process Edits when in the AA Rate definition.
3. 'Blockers'. Based on our experience and surveys, approximately 1/3 of manual claims are claims that are 'blocked' -- that is claims that are held/pended for no reason other than there MIGHT be a problem. It is a very common adjudication practice to force claims into manual review becuase they might have problems. This is very seldom beneficial. A first step in any AA Rate improvement effort is to review all the holds/pends to determine which ones are blockers and further determine how they can be removed.
4. Duplicate Lines. Duplicate claim lines affect an AA Rate. Depending upon the duplicate logic applied, potential dupes may require manual processing -- driving the AA Rate down. On the other hand, dupe logic which 'auto' denies a significant portion of potential dupes might drive the AA Rate up. In a nutshell, it is OK to leave duplicate claims in the AA Rate but keep an eye on them.
5. Adjusted Lines. For many payers, claims adjustments are a world onto themselves. Generally speaking between 2 and 5 percent of claims are adjusted. Because of the exception related (and often strange) things that happen to adjustments it is better to exclude them from the AA Rate definion.
6. Information Requests. Claims are often held while additional information is requested from a provider. For simplicity's sake, any claims that are held, for any reason, should be considered manual claims.

Attributions.

1. Sources. This metric was developed during the fifteen years of association with health plans.
2. Update History. This definition was last updated January 17th, 2007.