Claims errors: the bane of your bottom line

Mistakes will happen; the trick is to minimize them.

By Kevin J. Corcoran, COE, CPC, CPMA, FNAO

As all agree, reimbursement is a complex process, so it is no surprise that errors occur. In the vast majority of cases, an error does not result from perfidy or fraud. The Office of Inspector General (OIG) of the Department of Health and Human Services (DHHS) publishes an annual report on the error rate for Part B Medicare. Figure 1 is a compilation of those reports for the period 1996 through 2014.

Figure 1. Improper payments

The allied programs, Medicare Advantage and Medicaid, have somewhat lower error rates than Part B Medicare. In 2014, they were 9% and 7%, respectively.¹ OIG’s reports also contain detailed information about specialties. While the overall, all-provider error rate in 2014 was 12.1%, the error rate for ophthalmology was 3.5% and for optometry it was 5.6% — much lower than the average.

OIG also looked closely at cataract surgery and found an error rate of 4.1%.² This rate is not much different from the overall rate for ophthalmology and still lower than average.

However, when OIG looked at office visits,³ specifically those reported using evaluation and management (E/M) codes (992xx), it found a much bigger problem. Forty-two percent of all E/M services were incorrectly coded. Twenty-six percent of all E/M services had been upcoded, yielding 14% in overpayments; 15% of E/M services had been downcoded, yielding 6% in underpayments; and 2% had other coding errors, yielding a 2% error rate.

OIG inspectors also sought to assess the quality of the chart documentation for E/M services. They found 19% of E/M services lacked sufficient documentation in the medical record to support the claim for reimbursement. Seven percent were undocumented, and 12% were sparsely documented with inadequate chart notes.

The OIG has not published comparable data for eye codes (920xx). For this article, we looked at chart audits performed in the last few years by Corcoran Consulting Group and isolated the eye codes. The sample was large enough to be representative (4,185 encounters). Forty percent of all office visits reported using eye codes that were incorrectly coded. Twenty-two percent had been upcoded, yielding a 6% overpayment; 7% were downcoded, yielding a 2% underpayment; and 11% had a variety of other coding errors, yielding a 7% overpayment.

Corcoran Consulting Group chart audits

Errors occur for various reasons, and the impact in terms of dollars also varies as we saw with E/M coding.

During the past 25 years that our group has performed chart audits for clients, we have kept a running total of the results. We have found 65% of all claims are without error; 23% represent some kind of overpayment; 9% represent an underpayment; 3% have some other error but do not have an economic impact on reimbursement. While there are overpayments and underpayments in virtually all our chart audits, the net economic impact in dollars is a 4% overpayment — the same amount that OIG found for ophthalmology.

ASCs that concentrate on ophthalmic procedures tend to do very well with fewer errors than clinics because claims are simpler. Optical dispensaries that provide postcataract eyeglasses to Medicare beneficiaries tend to do poorly because claims are much more complicated and require a significant number of supportive documents.

Currently, an eye exam performed by an ophthalmologist on a Medicare beneficiary is associated with an average of one covered diagnostic test. We find that about 35% of all diagnostic tests do not support a claim for reimbursement due to poor charting of the interpretation and report; an inappropriate indication for the test; or too frequent testing. Other than errors associated with eye exams, errors with diagnostic tests are the next most common problem.

In our experience, there are many fewer errors associated with surgical services. When we find problems, they fall into two categories: The chart notes do not provide an adequate rationale for the surgery, or the claim contains gratuitous procedure codes. The medical records for unsupported surgical procedures provide an ambiguous or incomplete reason for the surgery. For example, if cataract surgery is planned, the chart note might not describe any hindrance to normal activities of daily living. A claim for reimbursement with too many procedure codes has problems with unbundling when the National Correct Coding Initiative edits are not followed. For example, if ptosis repair and blepharoplasty are performed on the same eyelid, only one procedure is reimbursed because the other procedure is incidental and bundled with the first.⁴

Physicians at the plate

Some practices do better than others on our chart reviews. Table 1 provides a breakdown into quintiles: “A” for the top 20%; “B” for the next 20%; and so on.

Based on these results, we conclude that most of our clients are doing a fair or good job. Rarely do we find an outstanding performance or a very poor performance. We appreciate that there may be a bias in these results because someone mismanaging his reimbursement might not want it confirmed or pointed out in great detail.

Practice patterns

While chart reviews focus on individual encounters, procedures and dates of service, it is also helpful to review practice patterns. In 2012, CMS looked at coding trends for E/M services (Figure 2).⁵

Figure 2. Coding trends of Medicare E/M services

Over a 10-year period, E/M codes for established patient office visits (99211-99215) have shifted. The data show code creep with lower-paying codes declining while higher-paying codes have increased. The trend is due in part to the implementation of EMRs, which have dramatically increased the amount of information in the chart with a corresponding inflation of CPT codes.

During our chart audits, we also look at practice patterns for eye exams, diagnostic tests, laboratory tests, surgical services and modifiers on claims. The comparison with CMS data is useful because it helps to determine whether the sample of charts is representative of the practice, and it points to services that are not part of the sample because they are not performed. From this information, it is a short hop to identifying underused services. Most recently, we looked at complex cataract surgery (66982) and found that 60% of ophthalmic surgeons (9,107 in CMS’ dataset) report this procedure in fewer than 5% of cataract surgeries while the average rate is 10% — they might be missing a good opportunity.

In the last few years, payers have sent threatening letters to ophthalmologists and optometrists based on claims databases alleging overutilization of services compared to their peers. Too often, physicians have no idea where they stand with respect to practice patterns. Too high? Too low? Just right?

Many good reasons exist as to why some doctors are not “average” or align well with the benchmarks. Possible outliers include subspecialists, doctors who practice in communities with significant public health issues, those with special clinical research interests, areas of the country with aggressive tort lawyers and regional differences in practice patterns.

What to do?

In 2000, DHHS published OIG Compliance Program for Individual and Small Group Physician Practices.⁶ Within this highly recommended but voluntary plan, OIG suggests:

• Periodically reviewing charts

• Monitoring utilization patterns

• Remediating errors

• Training physicians and staff

In our experience, there are no perfect practices when it comes to reimbursement — there are simply too many ways for errors to occur. The goal is to minimize their number and financial impact as much as possible. Aim for continuous improvement, not perfection. OM

REFERENCES

About the Author
	Kevin Corcoran COE, CPC, CPMA, FNAO, is president of Corcoran Consulting Group in San Bernadino, Calif.