The statistical information portrayed in this article is drawn from an actual case. The payor names have been replaced with the system-assigned codes to mask the identity of the radiology practice and its location. The identity of the payors is immaterial to the concepts discussed.

The practice under review outsources its receivable processing to a regional billing company based in Richmond, Va. The billng company’s platform has powerful database features that enable the user to track all debits and credit transactions attributed to examinations performed in a month. A request was made for three files consisting of three separate months-of-service that included the identity of the original primary payor. Every type of cash and noncash debit and credit transaction was to be isolated by its coded description along with the dates that the transactions were posted. The monthly universes were worked by the billing company a minimum of 14 months to offer the practice a valid example of how well the payor populations were managed. It should be noted that it takes approximately 24 months to completely liquidate/resolve all examinations performed by a practice in a given month.

When a month’s charges are resolved to a zero balance, there are two final universes:

  • Examinations paid exactly at the fee negotiated by the practice and the payor.
  • Examinations paid at a level less than expected by the practice, with the balance written off as a bad debt/administrative adjustment.

During the period before complete resolution of all charges, there are four universes; the first two plus:

  • Examinations showing a debit balance,
  • Examinations showing a credit balance.
Table 1. Month 1 system totals. (Click the image for a larger version.)

Not all receivable systems have the capability to organize results by both month-of-service and primary payor, a point worth noting when practices make decisions about vendors or receivable systems they intend to purchase for internal use. Table 1 illustrates the entire population for the oldest of the 3 months.

MONTH ONE AUDIT

The data is organized in the order of highest cash payor. Month 1 consisted of 448 separate payors, of whom 19 accounted for 87% of charges and 85% of cash income. Self-Pay is specifically identified for reasons that will shortly be clear. The columns, left to right, from Fees to Balance are the reconciliation of each payor population: Fees, Cash, Contract Adjustment (Cont Adj), Charity, Bad Debt, All Other Adjustments (AO Adj) = Balance. The gross percentage (Gross%) is arrived at using the formula: Cash/Fees. The net percentage formula is Average Cash/(Fees (Cont Adj + Charity + AO Adj)). The net percentage is the definitive efficiency benchmark because we are removing the component of gross charges that the billing system cannot collect, based on agreements with payors. The difference between “1.00” and the net percentage is the net bad debt for the payor (gross bad debt is an irrelevant number when so much of the gross charges must be written off).

The first important feature is the linkage of the cash/noncash credits to the fees. It is critical that the system consolidate all cash, regardless of the source, according to primary payor. The primary payor determines maximum payment. However, deductible/co-pay requirements imply that the primary payor will not necessarily remit the entire fee. These amounts will come either directly from the patient or the patient’s secondary coverage. They must be included in the Cash figure in addition to the actual payments from the primary payor; otherwise, the analysis is not valid. Why? This model is to help clients understand how their receivable system is managing payor relationships, including the original claim submission(s) and follow-up when payment(s) are required of the patient/secondary insurance.

Consolidation of all cash income helps verify that the practice is receiving income consistent with the payor agreement. For example, payor 274 is the dominant payor to this practice. The file prepared by the vendor enables anyone to examine cash payments, by CPT code, for all 8,788 examinations. It is then possible to isolate examinations where payment(s) are at variance with the negotiated fee, something not possible if payments from all sources were not linked to the original CPT code.

You can see that the range of gross percentage for the top 19 payors is large; this is the nature of the marketplace. Generally, the more dominant the payor, the lower the gross percentage because of their buying power; the analysis does reveal that a few payors remit fees close to the practice charge. The net percentage shows a tighter range but also helps identify problem payors where bad debt exists; this offers an opportunity for the practice to examine problems with specific payors.

This linkage modeling identifies the true self-pay population and its impact on practice bad debt. No insurance information was obtained from the hospital admission system or subsequently from the patient. The Self Pay line in Table 1 reflects those examinations, after 14 months of processing, that could not be billed to a primary payor. The gross collection ratio on self-pay populations is always abysmal; it is consistently the largest source of bad debt. The size of the population has a direct impact on overall net percentage, per practice. This implies that there can never be a “standard net percentage” because each practice will have materially different self-pay levels.

This practice covers hospitals with large transient populations. While it was initially surprised at the size of its true self-pay population (charges, bad debt, and gross ratio), it helped the client understand why its overall net percentage differed from numbers they may have heard elsewhere. Practices are often told that they should expect to see net percentage in the ’90s. This might be true of populations where self-pay is less than 5% of total charges.

The Populations block immediately below the model helps illustrate this better. It shows collection results for those examinations where a legitimate primary payor was identified and billed. This population accounts for 92% of gross charges and 97% of cash. Conversely, the self-pay examinations account for 8% of charges, 3% of cash, 83% of bad debt write-off, and 62% of the 14-month-old unresolved Balance (most of which will probably be written off as Bad Debt). There are also some different columns in this block that help explain collection ratio mathematics. Net Pop shows the actual paying population after stripping away the amounts that the vendor is not allowed to collect by either payor contract or practice direction (charity). Now the Self-Pay population accounts for 20% of the collectible value. The last column Net*Net% is the multiplication of the Net Pop times the Net%. It shows you why this specific practice has an aggregate net collection ratio that cannot approach a 90+% expectation. The net ratio (Net%) on the payor population has reached 89.78% after 14 months of processing and contributes toward 71.79% of the aggregate net ratio. Even if the payor population was a perfect 100%, the resulting aggregate net ratio would be 79.96 (Insurance Net Pop) + 2.10 (Self-Pay Net*Net%) = 82%.

The last block begins the process of explaining the four universes prior to Month 1 charges reaching zero. The Perfect universe consists of those examinations where payments equal the negotiated fee with the payor. A caveat is necessary before we go forward. This analysis was a “pre-audit” evaluation. It seeks to identify problem areas requiring in-depth analysis of receivable system operations. The file records show how the vendor posted cash and noncash credits to the system. None of the 16,737 examinations were examined to the extent of the backup information from the payor or testing of physical cash payments. Given the number of examinations, a statistically significant testing population would take a large amount of staff time. All practices have limits on what they are willing to pay for detailed auditing. This modeling is an alternative, taking advantage of the powerful database features of this system.

The only noncash credit is the write-off of the difference between practice fee and payor allowance. The sum of cash and adjustments equaled the fee, leaving no balance. The gross ratio is 33.70%; net ratio has to be 100%. The gross ratio helps to speculate on a “perfect collection month.” The prior block shows you the gross charges attributed to the primary payor population: 2,193,953. The extension of this charge base by 33.70% is 739,000. This implies that: 739,000 687,000 = 52,000 was lost to other circumstances. As you will see shortly, the world is never perfect especially when dealing with more than 400 payors.

The All Other Zero universe is records that also are a zero balance, but some of the reason has to do with charity or bad debt credits. This implies that some of the income legitimately due the practice was to be paid by the patient, and they failed to respond. Every receivable system should invoice the patient for legitimate balances after insurance payment. Some balances are small and it is not cost-effective to continuously sent statements. These are removed from the active receivable system after one or two statements and not sent to collection. Other balances are significant and go through a standard sequence of statements and dunning letters. If the patient does not respond after this sequence, the account is credited off this system and sent to a collection agency. Finally, some of these zero balances are the write-off of delinquent self-pay accounts that are also sent to a collection agency.

The last two universes, Debit and Credit, will eventually disappear, consisting of debit and credit balances. We will now illustrate the models that detail the payor results for the four universes. This will help you understand the challenges of collecting income attributed to such a large volume of examinations.

THE FOUR UNIVERSES

Table 2. Perfect population. (Click the image for a larger version.)

There are 370 payors in the Perfect Population (see Table 2). The self-pay contract adjustment probably should have been a charity write-off; the wrong noncash credit code was posted. This model offers a good illustration of the large differences in the amount the major primary payors will pay for interpretations. The largest payor reimburses the lowest fees. Some of this also has to do with issues pertaining to additional rules about discounting of surgical code sequences; this population has a disproportionate amount of interventional caseload. Organizing the payor population this way helps the client understand why their collection ratio differs from another practice with a different payor mix. The block at bottom organizes the totals for all payors and the self-pay population. The gross ratio on the self-pay examinations in this universe is really 100% because those records with the incorrect noncash code belong in the next universe.

Table 3. All Other Zero balance population. (Click the image for a larger version.)

This universe of 107 payors (see Table 3) helps illustrate why the world is not perfect. The balances are at zero, no longer being pursued by the billing company. The Self-Pay universe shows a large amount of bad debt and the majority of charity credit (notice no contract adjustment here. There is also bad debt attributed to the self-pay balances due from patients with insurance. This is why it is impossible to achieve perfect collections within the population that has insurance coverage; payors shift some of the health care cost to the patient in the form of annual deductible and/or co-payment requirements. The Populations block illustrates the drain on both the gross and net collection ratios by the self-pay population. The Net Pop column shows the actual collectible population after removal of the contract adjustments and charity. When the collection ratios are applied to the restructured population, it shows the poor net collection results on this segment of the Month 1 population.

Table 4. Debit balance population. (Click the image for a larger version.)

There are 121 payors in the Debit Balances universe (see Table 4). The balances for the primary payors are likely being pursued from the patients, but they will also include accounts being administratively challenged by the payors (precertifications, coding combinations) and auto/industrial accidents that sometimes take years to resolve. The large block of self-pay balance probably should have been sent to collection months before this analysis (a topic of discussion at the client presentation). Delays hurt the ability of collections agencies to secure payments or track down patients who move. It is likely that, after 14 months of processing, the majority of these balances will be lost to bad debt or administrative write-off. Note the reorganized Net Pop to once again see the impact of this large self-pay population on collection ratios.

Table 5. Credit balance population. (Click the image for a larger version.)

The Credit Balance universe (see Table 5) might be considered irrelevant because it is so small, but it is a critical part of system oversight. A practice does not want to get a false sense of collection performance if it is holding on to income that does not belong to it. Past analyses of other systems have uncovered unreasonably large credit balances that exposed practices to legal/financial problems. Credit balances are not always caused by duplicate payments. They can also be from erroneous entry of duplicate noncash credits. You can get a sense of what type of credit balance you have by looking at the $FEE/$CASH columns. $CASH exceeding $FEE is a good indicator.

Table 6. Comparison of benchmarks: Month 1 versus Month 2. (Click the image for a larger version.)

The value of compilation by month-of-service is examination of continuity. The billing company worked Month 1 and Month 2 approximately the same number of billing cycles (14). Table 6 shows first how consistent (or inconsistent) the structure of the payor population is revealed in the percent of charges (% CRGS) column. There will naturally be shifting, but the same 19 payors comprise slightly more than 86% of gross charges. The self-pay population in Month 2 is unfortunately larger (negative trend), and its gross percentage is slightly worse. However, overall gross percentage and net percentage are better in Month 2 because of combinations of larger populations of payors with higher reimbursements or improvements in the collection ratios of some payors. The last three columns give you the percentage changes in the ratio benchmarks in Month 2. Month 2 required billings to 469 different primary payors. Large changes in the gross percentage might prompt further review of that payor universe. However, large shifts in the examination mix within a payor universe can also cause the collection ratios to change dramatically, especially when it involves interventional cases. These are generally reimbursed poorly, in relation to gross fees.

Table 7. Comparison of Perfect universe benchmarks: Month 1 versus Month 2. (Click the image for a larger version.)

The Perfect universe, where reimbursement is at the payor fee schedule, also offers useful benchmarking. The net ratio comparison, always 100%, is replaced with average cash per examination. This isolates anomalies that might prompt a more in-depth review, even if just for information purposes. For example, payor 7917 shows a large change in both Gross% and $Cash. However, the number of examinations in this universe is so small that it is probably attributed to examination mix, where the successfully resolved examinations were at higher fees. Large shifts in the top five primary payors might more appropriately prompt in-depth review.

FINAL OBSERVATIONS

Clearly, a close examination of billing practices and results can yield valuable information.

  • Practices need to challenge their billing vendors or internal systems to provide routine payor-based summaries as part of the end-of-month reporting. These reports should be recompiled to link the cash/non-cash credits back to the original charges generated from the clinical work in a month.
  • The value of the payor-based reporting is contingent upon the ability to consolidate all cash and noncash credits by original primary payor, regardless of multiple sources of payments per CPT code.
  • Most of the current reports we have seen use generic groupings of payors (HMO, Managed Care, Commercial, etc) to list collection results. This article illustrates that only 15 to 20 payors account for over 80% of income. They should be specifically isolated in the monthly reporting rather than buried in a generic grouping.
  • It is critical that noncash credits be distinguished between legitimate contract adjustments, charity write-off, and bad debts (bad addresses, delinquent accounts sent to a collection agency, small balance write-offs) to facilitate accurate calculation of net collection/net bad debt ratios.
  • Linkage of all cash back to each CPT code enables a practice to do cost-effective random testing of payor responses to claims. This enables a practice to verify that a payor is remitting amounts consistent with agreements between provider and payor.
  • Payor-based compilation by month-of-service is a far more cost-effective way to examine system efficiency than labor-intensive audits.

Pay attention to the true Self-Pay population. It answers many questions concerning Net Collection ratios.

James Kieffer is president, Proforma Financial Group Inc. He welcomes comments and feedback: (603) 598-2944