Keep Payors Honest With the Practice Receivable System

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon
It is not a completely fair analogy, but the differing agendas of payors and providers place receivable processing in something of a battle zone. The payors seek to limit payments to no more than absolutely necessary to cover the contractual benefits of their beneficiaries. They put up a gauntlet of compliance rules to ensure that the providers justify their claims for services. The mission of the insurance company is not only to cover its beneficiaries, but to maintain positive cash flows as an organization. Naturally, providers seek to maximize income for services rendered. Nothing pleases payors more, it seems, than rejecting claims for lack of compliance, because it slows down cash outflows. They may eventually pay the claims, but at a later time. They will gladly delay their initial response until absolutely necessary and hope that the provider system has sufficient inefficiencies that corrected claims are seriously delayed, or never even resubmitted. Payors use a variety of tactics to delay cash outflow, and providers are well advised to monitor the payment records of individual payors using several basic accounting reports. Practices can never assume that insurance companies pay the correct fees 100% of the time. Insurance companies negotiate with providers to pay a fee for each authorized CPT® code-based procedure over a contract term. Many commercial payors prefer a fixed fee; others will use a schedule that changes, depending upon a market benchmark. One of the most commonly used is the annually published Medicare rates. These are unique to either a state or a region within a state and generally change each January. Payors with adjustable-fee contracts are betting on declining Medicare rates to reduce their costs. Multiple Product Lines One reason for payment errors is the multiple product lines that health care insurers offer to companies, where the fees are based upon the type of contract. The lowest fees are generally linked to HMO-like product lines that limit provider options. These sometimes carry a small patient copayment requirement that is administratively bothersome because it is costly to bill a small patient balance. The next level of product line will be the PPO, which expands patient choice and generally pays higher fees to providers. The third type of product, becoming increasingly rare, is an indemnity-like programs that pays a percentage of fees, rather than using a fixed schedule, and may even allow full-balance billing. The patient has full discretion over provider choice. These programs are the most expensive to the employer purchasing coverage for employees. The challenge for radiology billing systems that rely on hospitals for demographic data is that the type of payor product may not be distinguishable at the time of claim submission. Therefore, a practice will often send all payor claims to a single address and will not know what the product coverage is until response. This has the added effect of making it difficult to assign a primary payor code, which is essential to the monitoring process described here. Many systems have touted a feature that checks the payor allowance at the time of cash posting, but none of them deliver on this promise. This feature requires, to be useful, the correct primary payor designation and the correct fee schedule for a date of service. The fee-schedule data require maintenance of multiple file libraries because date-based changes will require the system first to check the date of service and then to cross-reference the correct schedule. On average, a radiology billing system will be filing claims with more than 250 different payors per month; 20 of those are likely to account for over 90% of a practice’s revenue. It is possible to construct a payor schedule independently if it is based on a specific Medicare year; the formula will be a multiple of the Medicare fee. As will be shown later, it is easy to verify the formula. Fewer and fewer payors are using an internally produced fee schedule not based on Medicare; often, it is one they are unwilling to share with providers. This leaves the provider with the option of acquiring the fees by monitoring the remittance advice that accompanies payments. Mounting a Defense Medicare is the simplest payor model for which we have access to data.
This is a Medicare population for December 2007. The database extraction was performed nine months later, meaning that all but the most troublesome problems have been resolved. Column E shows the published Medicare professional fees for this carrier (all carrier fees differ due to economic adjustment factors). Column G shows actual payments, measured against column H, which shows perfect resolution. Three of the top 20 exams show income greater than expectations. This might be due to one or more accounts with refunds, because 20% of the allowance comes directly from the patient or the patient’s secondary payor. The top 20 shows slippage of $3,400 against perfect resolution. The remaining 186 diagnostic procedures have approximately the same dollar variance, at $3,500 (the surgical codes must be kept separate because there are special discounting rules for second, third, and subsequent procedures that materially reduce the aggregate gross ratios). The fees may be large, but reimbursement is dramatically lower. Note the relationships between this practice’s fee schedule and the stand-alone Medicare fees. They are almost a match (some Medicare fees actually are greater than the practice fee), yet, when the discounting rules are applied, the income is barely 22% of the stand-alone fees. This payor was purposely chosen not only because of ready access to the published fee schedule, but because of the complexity of tracking the full resolution of the charges. This next table helps put this in perspective.
The Medicare rules state that a participating provider must accept the Medicare fee schedule as payment in full. The patient is responsible for an annual deductible plus 20% of the allowed fee. Many patients choose to have a secondary payor responsible for out-of-pocket costs. Sometimes these copayments are small and expensive to pursue, especially if the patient has no secondary coverage. Medicare generally contracts with one of the larger commercial payors in the state to administer its claims. If a patient happens to be covered by that commercial payor, there is an automatic coordination of benefits in which both the Medicare payment and copayment are processed in tandem, eliminating the need to file the secondary claim. Otherwise, the billing vendor has to wait for the Medicare response and then submit the claim for the balance; this complicates processing. In Table 2, naturally, Medicare has paid most of the December 2007 claims. By this time, most of the patient deductibles have been met by prior services and the payments shown from the other payors will principally be the 20% copayment balances. Self-payment, even though it is a dominant source of the copayment, is segregated below the top commercial payors. A number of state Medicaid payors are in this mix. Medicaid often does not pay the 20% copayment, but this is not recorded as a bad debt. More than 240 payors other than Medicare contributed to the revenue stream. Some of these payors paid less than $4 to complete adjudication of the procedure. This leads to another practical point. Billing vendors lose money on the tail-end completion of account resolution because the dollar balances are so low. Multiple statements or claims caused by delayed responses place a lot of pressure on the billing system because it can get to a point where it costs more to pursue the balance than the balance itself. The column at the far right shows the proportion of income from Medicare/all other payors. Medicare has to be below 80% because a few accounts still have annual deductibles to meet. At bottom is a set of resolution values using the posted credits to the system. The first is the liquidation rate achieved, close to the Table 1 value. The second is the simulation of the fee schedule. If a balance is removed as bad debt, it means that the practice was entitled to the income. Therefore, adding it to the cash approximates maximum income due. The number is a bit higher than the perfect universe in Table 1. This model is intended to highlight the challenge of achieving perfection in this complex environment. Much of this challenge is intentionally created by the payors. Complexity can work to their advantage just by wearing down a provider too much to chase the minor balances. The secondary payors may have a subset of rules that allows them to avoid even the copayments; the examples cited here were the state Medicaid programs. Big-picture Modeling Table 3 provides a broader snapshot of the practice’s top 20 payors, while the first two tables focused specifically on the Medicare population for the purpose of explaining the mechanics of payor-based resolution. Aggregating income by primary payor enables the billing vendor to produce the following report.
Table 3 is a receivable reconciliation by primary payor for the December 2007 month of service, with the cash and noncash credits posted through mid-September 2008. The model was first sorted by highest exam volume, and then the top 20 were reorganized by highest cash amount. The self-pay accounts and total for the remaining more than 300 payors are separate line items. Column E represents the difference between the practice fee and the maximum payor allowance. This must be credited off the account as a condition of the payor contract. Column F shows noncash credits, where the circumstances of the claim were not in compliance with payor rules. These may have been appealed after rejection, then judged noncollectible (the patient cannot be billed). The column G courtesy credits are there for numerous reasons. Some services are rendered under fixed monthly contracts, regardless of volume, where the posted exams are kept as a record only (then credited off later). Others may represent elective courtesies performed by physicians supplied with information about the patients’ circumstances. In column H, bad-debt credits per major payor are an indication that the contract relationship requires the patient to pay some part of an agreed-upon fee and, after numerous statements from the billing vendor, the patient has declined to respond. Generally, the vendor removes the account with a bad-debt credit entry and forwards the patient’s demographic details to a separate collection agency. Note the dollar amount of bad-debt credits attributed to self-pay accounts. These are accounts that have no insurance of any kind. These accounts, universally, will constitute over 60% of a practice’s bad debt. The majority of them will originate in the emergency department. The formulas for columns J and K are footnoted. Combining actual cash with bad-debt writeoffs (column J) offers a picture of the true cash value of the charges originally submitted to a payor because the bad debt implies that the practice was entitled to the cash, but the patient failed to respond. This formula also helps make another important point about the realities of collecting medical service charges. Note the last ratio in the column: 38.17%. Even if the vendor’s collection activities were perfect, it would only be possible to receive $0.38 of every billed dollar because of the contractual relationships between payors and providers. The ratio in column K ignores the charges in column C. It is a ratio that can be used at any time after the charges are first pursued because it only includes resolved credits, and is a good measure of how well the vendor has managed the accounts for each payor. Once the balance in column I reaches zero, the column K formula is identical to D divided by C. Note how little can be derived from patients with no insurance: 5.29%. The ratio for the 301 other payors is better than for the top 20. The reason is market share. The big vendors have muscle due to the number of patients they cover, and they use it to negotiate the best rates with providers. The minor payors have less clout and cannot negotiate as successfully. Thus their fees/costs are higher. Columns L and M offer perspective on the influence that the top 20 payors have on this practice’s income. Medicare dominates both exam volume and income, but this practice has successfully negotiated favorable contracts with some of the dominant commercial payors. For example, even though CIGNA covers a fraction of Medicare’s volume, it still provides almost half as much income due to its fee schedule of 2.5 times Medicare. In fact, this payor’s fee schedule is a multiple of Medicare’s fee schedule. The top 20 payors constitute 85% of exam volume and account for 92% of monthly income. Table 3 is an excellent report that should be reproducible on demand by a billing vendor. Its importance is in providing a quick analysis of how well major payors’ accounts are being managed. In this instance, the charges with a December 2007 date of service are all but resolved, with only $20,000 of account balances remaining of the original $1.87 million billed. If reports for each month of service were produced routinely, you would be likely to find that the first 5 to 10 payors are always the same (in order of importance). The remainder will often shift positions. Production of these models on a monthly basis allows a practice to scrutinize the column J and K ratios to make certain that payments are consistent with fee-schedule relationships. Random Auditing The tables presented are big-picture views of receivable management of payor relationships, but they cannot entirely replace account audits. It is unreasonable to assume that a vendor will evaluate every payor response to a claim at the point of adjudication. This medium-sized practice, alone, generates more than 12,000 new accounts per month. It is not possible to perform the tracking necessary to guarantee perfect resolution of every exam at the time of payment unless the client wishes to pay a much higher processing fee to the vendor. It is, however, reasonable to expect random audits of exam populations that test the integrity of payor responses. The powerful database features of today’s systems make this possible. Consider the following extraction queries by an examiner with access to this information: cash income greater than zero 0 (Table 3, column D); contract adjustment greater than zero (Table 3, column E); bad-debt credit equal to zero (Table 3, column H); and balance equal to (Table 3, column I). This would produce a file of all exams in the December 2007 population that only resolved to zero because of cash and contract adjustment. This would be done by CPT code, and would include patient account identity to enable an examiner to research anomalies. An anomaly is defined as any cash per CPT code that differs from the published Medicare fee. The client can be given a report card on perfect matches, with specific explanations on those that were not perfect. This process is not limited to Medicare. If the fee schedule of a major commercial carrier is a multiple of the Medicare rates, it is possible for the examiner to build a library, by CPT code, to match against that payor’s database. In reviewing Table 3, we have assumed that the cash value is an accurate representation of the fee-schedule relationship. It may not be, however, and the only way to find out for sure is through a random audit. Without doing a random audit, you are not going to be certain that the ratios that you see in Table 3 are a true reflection of your payor relationships. Any robust receivable system can readily produce such a report. The models used here come from an actual vendor database. Note that it pertains to a month for which the accounts have been worked for 10 billing cycles. It is best to choose a month-of-service range that enables the vendor to resolve at least 80% of the exams (generally, four billing cycles). This audit can be conducted at least once per year, where each year’s date range varies from the prior year.