Some group practices are finding that with a little extra effort, and the help of specialized software, they can get paid a lot more by insurers. Just ask Bend (Ore.) Memorial Clinic. The 80-physician, three-site practice is recouping an average of more than $50,000 in increased reimbursements each month since adopting a new strategy.
In April 2006, the clinic implemented software that analyzes payments from insurance companies to identify claims underpaid or overpaid. The software also includes decision support and workflow functions to manage the process of appealing payment decisions.
The practice pays $9,000 a month for the Phynance software from Medical Present Value Inc., Austin, Texas, plus the cost of two existing employees who now dedicate a significant part of their time to being insurance analysts and working the disputed claims, says Shane Irving, service line director of business services and information systems at Bend Memorial.
Further, implementing and using the software is relatively easy, he adds. "It was probably the easiest implementation I've been through with external software."
The Phynance software extracts payment data from the clinic's practice management system and compares reimbursement against the terms of the appropriate insurer's contract with Bend Memorial.
A "variance report" flags potential underpayments, and decision support software gives the reasons a particular payment was flagged. For example, an insurer may have paid a claim using the incorrect relative value unit or conversion factor payment schedules established for a visit. A simple 10-minute office visit, for instance, might be 1.02 RVUs and billable at $81.60, while a slightly more complex 15-minute visit is 1.66 RVUs and billable at $132.60, Irving explains.
Another way a claim may be underpaid is if the payer bundled a series of services into one payment, although that payer's contract does not authorize bundling for these types of services, he adds.
To start using the software, the clinic sent copies of its insurance contracts and fee schedules to the vendor. The vendor scanned the documents into a data repository and worked with the clinic to resolve any questions or issues about contract terms.
The vendor also loaded Bend Memorial Clinic's claims payment data for the past year into the repository, a common practice because many insurers permit appeals on payments up to a year old.
As the clinic continues to use the software, it can tweak analysis of claims going to particular insurers based on their payment patterns. "The analysts will start seeing particular items flagged that don't appear consistent with the contract," Irving says. "So they can create specific filters for specific CPT codes to specific payers."
Validation The Key
Bend Memorial Clinic analyzes payments from insurers covering 80% of its business, including Medicare, which is the most accurate payer, Irving notes.
The clinic has not gotten much resistance from insurers. One wanted to see the practice's data "so they could see what we were seeing and could monitor their payment system," he adds.
But the low payer resistance, he believes, is because the practice took the time and attention to validate the system's accuracy.
The validation stage took the most work during the implementation phase. The practice used live data to determine if flagged payment variations were accurate and not false positives and to calibrate the modeling functions in the analysis software.
"We don't want to be appealing incorrect items because that's when you start to get a lot of resistance from payers," Irving says.
Training Bend Memorial Clinic's two analysts to use Phynance only took a few days, but it took a month for them to become confident with the application, Irving says.
And these analysts had substantial backgrounds for their new positions, he notes. One had experience working with insurers in the billing department and the other had good analytical skills and experience in data mining, extracting information and generating reports.
Other Tricks
Bend Memorial Clinic also checks for overpayments above a certain threshold, because it doesn't want insurers coming back to the practice and asking for refunds. Less than 1% of flagged claims are overpayments, usually resulting from an incorrect RVU loaded into a code, "so it stands out pretty quickly" and is easy to fix, Irving notes.