Health-claims processors routinely cost employers thousands of dollars by letting overpayments slip by, asserts a recent report by a benefits consulting firm.
In one case, a programming glitch on a claim vendor’s computer automatically lowered the co-payment for employees of a large self-insured company from about $10 to $5, according to MaryAnne Watson, a claims-auditing specialist at The Segal Co. Over the course of about a year, says Watson, paying an extra $5 for every office visit cost the employer $225,000.
What happened next was a rarity: The vendor owned up to its mistake and reimbursed the employer. Much more often, third-party claims administrators will ask the company, “Do you want us to go back and ask your members to pay $5 more now?” says Watson. “Most employers aren’t going to go there,” she adds. Likewise, doctors routinely hang on to overpayments until the employer or the administrator asks for a refund, Segal also found. It’s very hard for self-insured corporations to retrieve excessive payments from employees and doctors — and very likely that any vendor errors will cost those employers dearly.
Unfortunately, those mishaps themselves are very likely, according the Segal report, which examined about 225 employee health claims generated by about 50 of its self-funded employer clients.
A major cause of benefit-cost leakage, the consultancy found, is that third-party health-plan administrators fail to meet industry standards for a large chunk of medical claims. “In 50 percent of our health-plan audits, the carriers [that is, insurers and other firms that provide claims-handling services] are not at the level they should be in handling claims,” said Segal vice president Anthony Rienzi.
The vendors fall short of health-industry strictures for claims-processing timeliness, overall processing accuracy, and financial accuracy, according to Segal. For example, Watson says that to meet the timeliness standard, administrators must process between 90 and 95 percent of claims within 10 business days or 14 calendar days. In addition, under Department of Labor rules they must process all claims within 30 days, and they must adhere to state strictures, which tend to be stiffer than DoL regulations. Administrator tardiness can lead to employer losses, says Watson, since they rarely can collect on an overpayment if a claim lingers beyond the state statutory limit, typically one year.
Third-party claims handlers are also expected to meet standards of 95 percent overall processing accuracy and 99 percent financial accuracy (that is, 99 percent correct in terms of the dollar amount claimed). Despite widespread inaccuracies, vendors have improved somewhat lately; according to Segal, 35 percent of vendors failed the overall processing standard in 2004, while 24 percent failed a year later.
Administrators can also cost employers money if they fail to properly cross-check claims against employee history. In one case studied by Segal, the beneficiary was an employee’s ex-wife, whose coverage ceased when the couple’s divorce became effective. Because the employee was late in informing his employer about the divorce, the employer was tardy in delivering a termination-of-coverage notice to the claims handler, which until then didn’t recognize that the patient had incurred hospital bills after her coverage was terminated.
When the claims handler did receive the notice from the employer, it correctly backdated the patient’s termination date, but it still failed to check if any claims had come in afterward — and the company ended up footing an extra $210,000 worth of hospital bills.
The employer’s delay in delivering the termination-of coverage notice makes it unlikely that it will get its money back, says Watson. But even though the employer may have been equally at fault, she adds, an administrator who receives a retroactive termination notice should “look back and see if claims were paid when they shouldn’t have been.”
When checking up on vendors, advises Watson, employers would do well to look closely at the kinds of data that administrators present as proof of their performance. The data is typically based on the vendor’s performance for a group of employers, not an individual client, which might present an inaccurate picture, she observes.