Wake Forest initiative hastens ID of patients mislabeled 'self-pay'
Wake Forest initiative hastens ID of patients mislabeled 'self-pay'
Six Sigma project leads to real-time audits, enhanced communication
A Six Sigma project at Wake Forest University Baptist Medical Center in Winston-Salem, NC, has significantly quickened the identification of patients who initially are listed as self-pay but found to have insurance after being admitted.
The effort began because financial counselors working on obtaining information and making payment arrangements for self-pay accounts were finding that many of those patients actually had insurance, explains Margaret Currie-Coyoy, Medicaid program specialist.
"The focus of these financial counselors, who refer accounts to myself as a Medicaid [specialist] and other individuals who work with state agencies, is on collections," she notes. "We wanted that focus to be on true self-pay patients, so we decided to see if we could reduce the number of accounts they were receiving for which insurance could have been gotten sooner."
Because they put an extra focus on collections, Currie-Coyoy adds, these financial counselors are called resource recovery specialists.
Verification and quality services (VAQS) staff members, meanwhile, were noticing that these accounts — typically identified a day or so after admission as having insurance — were having to be appealed because of the miscommunication, she says.
"Medicaid of Virginia, for example, requires a 24-hour notice for admission," adds Keith Weatherman, CAM, MHA, associate director, patient financial services. "If it looked like a patient came through the emergency department [ED] and was admitted, and [staff] had not picked up on Medicaid of Virginia [as the insurer], they would deny payment."
The Wake Forest medical center gets a large number of referrals from other facilities, he notes, and some of those patients may go through the ED before being taken directly to a bed, but are not actually registered in the ED. Many of those patients, Weatherman adds, do bypass admissions.
The project team — led by Currie-Coyoy as part of the requirements for obtaining her Six Sigma green belt certification — first looked at the number of inpatient accounts miscoded as "personal pay" (the "defects," using Six Sigma terminology) and the admission source, she explains. "Since the greater number of defects occurred in the ED — approximately 50% — the project was scoped to focus on the ED registration process."
The team then defined the project's primary metric as the number of inpatient accounts miscoded as personal pay in the ED, Currie-Coyoy says. (See graphic.)
 |
"We worked with data from June 2005 through August 2005, which indicated that of the 683 inpatients registered as personal pay in the ED during that period, 185, or 27%, actually had insurance," she adds.
Meanwhile, the VAQS staff had to appeal $40,000 in charges denied by insurance companies because of missed pre-notification, she says.
Weatherman points out that the problem being addressed involved "very sick patients coming through the ED" and was more a procedural issue rather than people not doing what they should be doing.
"Six Sigma is about constantly seeking the cause [of a problem], with the assumption that people are trying to do their job the best they can, and seeing if anything can be done to improve the situation," Currie-Coyoy says. "Of course, people will not always have their insurance card with them and family members may arrive after the patient has left the ED [and been admitted].
"We analyzed that 27% with the Six Sigma tools and tried to reduce the number of defects as much as possible, but allowed for circumstances outside our control," she adds. "We just wanted to reduce it. Sometimes it takes repeated attempts."
To come up with the project's target, Currie-Coyoy explains, the team looked at the 27% average of misidentified self-pay patients, and then at the one week during that period when it was the lowest (19%).
"We take the difference between the baseline and the best-case data, and try to improve the baseline by at least 70%," she says. "So we at least wanted to go from 27% down to 21%. The team also wanted to ensure that no ED inpatient accounts miscoded as personal pay resulted in denial for missed pre-notification, Currie-Coyoy adds.
Potential improvements ID'd
Through the use of Six Sigma's DMAIC (define, measure, analyze, improve, control) tools, she says, the team recognized three potential improvements for the ED registration process regarding inpatient accounts miscoded as personal pay: the need to standardize registration procedures for patients who present to the ED without insurance cards, the need for real-time auditing, and the need for a communication log for ED registrars and nurses.
Members estimated that improvement would result in a saving of labor hours for the financial counseling group, Currie-Coyoy says, as well as reducing denials.
"We looked at causes as to why accounts were left as 'self-pay,' such as trauma patients and those with high acuity, and at something getting missed because of high volume during certain shifts," she notes.
The team surveyed admitting representatives regarding what happens if a patient comes in without an insurance card or if pieces of information are missing, she says, and did a closer study with three of the employees.
"We realized that about half the clerks thought they were correct in leaving the designation as 'self-pay' (rather than noting that the insurance information was unverified or incomplete)," Currie-Coyoy says. "We verified with them that we want to leave an 'insurance shell' even if all the information isn't there."
One tenet of Six Sigma is to reduce variation, she points out, which led to another facet of the project. In some cases, Currie-Coyoy says, family members arrive in the ED after a trauma patient or one with high acuity has been taken to the nursing unit, and the employee at the information desk refers them to the patient family coordinator, who is part of the nursing staff.
"Generally, the ED [registrar], if she had a registration she was unable to complete, might tell the coordinator, 'If you see this family member, please refer [him or her] to me so I can get more information,'" she says. "So there was [already] a verbal communication in place."
The team, however, recommended creating a log book, with carbon copies, in which the ED registrar could write a memo to the coordinator. In addition to the patient's name and registration number, the memo would include a message noting that the registration was not complete and asking that family members who arrived be directed to a particular registrar to provide assistance.
"Then when the family members arrive, [the coordinator] knows exactly which person to send them to," she says.
In addition to formalizing the verbal communication that was already happening, the new procedure — which provides documentation — gives more impetus for the patient family coordinator to get involved, she adds. "We had the head of nursing sit in on a couple of meetings, and she fully agreed and is backing the effort."
Another team recommendation was to have some real-time audits performed by the ED supervisor during each shift, Currie-Coyoy says, "to see if [registrars] are getting the insurance information, creating the shell, and making sure to get all the demographic information."
While some fairly aggressive auditing was done during the pilot project, which ran from February to April 2006 the effort has since lessened but is still ongoing, she adds.
"The team also recommended that the [registrar] put a note on the account when something is not able to be obtained," she explains. "If, for example, the patient stated, 'I have insurance,' but can't remember the name of the insurer or anything else, the [registrar] could document that."
After spending a couple of months analyzing data, discussing ideas, and getting input from ED registrars, the team began instituting changes, she says. "Immediately, in that month, the defect rate, as an average number, went from 27% to 19%. We continued our pilot, brought in one [recommendation] at a time, and officially ended the pilot in April."
Over that three-month period, the average defect rate was 19.8%, which was still below the target (21%), she adds. "We had our best month ever in May, when the average rate was 16.3%. That brought the average since the changes down to 19%."
In addition, the hospital has not had to appeal any accounts that were denied because they were left personal pay since September 2005, Currie-Coyoy notes. "That means we haven't had any lost or delayed revenue [due to that issue]."
Much of the improvement was because of the enhanced cooperation between ED registrars and the financial counseling group, she says. "It was the whole notion of having the team together and talking — with just that, we started making positive changes. It's really about the communication."
While the project has yet to register any gains under the technical definition of "hard savings," there are now three financial counselors working with self-pay patients where before there were four, she adds. "The fourth position is now in limbo. If we are able to free it up, we can give that person other responsibilities."
Not counting that potential elimination of a position, the project resulted in $3,100 in saved labor costs, Currie-Coyoy explains. The reduction from 27% to 19% represents "soft savings," she adds, because as of yet, nothing can be cut from the budget as a result.
Verification system added
In light of the frequency with which patients present without an insurance card or without complete information, she adds, the final team recommendation had to do with implementing a real-time insurance benefit verification system.
The process of bringing that idea to reality was well under way by late June, Weatherman notes, and the new system was expected to be up and running by early August.
The product is "like a template that sits on top of the registration pathway and helps guide the registrar through the process," he says.
Using triggers such as name, date of birth and Social Security number, Weatherman adds, it will be able to verify coverage by Medicaid and a number of other third-party payers without the registrar having to leave the registration screen to pull up an on-line verification web site or call and enter the eligibility information by phone.
The hospital can put some of its own rules into the system, which likely will be able to "pop up and ask questions, like TurboTax," he notes.
The verification system also can check the address given by the patient against a national database of addresses to see if there is a match, Weatherman points out. "We can save a lot of time for people who are currently verifying insurance manually and those worrying about returned mail," he says.
Once the system is in place, Weatherman notes, "we will be looking at [full-time equivalent] savings."
(Editor's note: Margaret Currie-Coyoy can be reached at [email protected]. Keith Weatherman can be reached at [email protected].)
A Six Sigma project at Wake Forest University Baptist Medical Center in Winston-Salem, NC, has significantly quickened the identification of patients who initially are listed as self-pay but found to have insurance after being admitted.Subscribe Now for Access
You have reached your article limit for the month. We hope you found our articles both enjoyable and insightful. For information on new subscriptions, product trials, alternative billing arrangements or group and site discounts please call 800-688-2421. We look forward to having you as a long-term member of the Relias Media community.