Reconciling Dependent Coverage

Life happens. Children grow up and get married, couples get divorced, a spouse or child dies. Sometimes employees are so busy or preoccupied that they forget to update their employers about important changes to their family-member status. Likewise, a surprising number of employees either aren’t aware of or don’t remember the details about dependent eligibility. This can result in wasteful spending associated with ongoing coverage of individuals who are no longer eligible for benefits under an employer’s medical plan. A growing number of higher education institutions are looking to remedy that situation through dependent audits of their medical plans. In a nutshell, employers essentially survey employees to confirm the status of spouses and other family members, often requesting documentation to verify employee records. While the process can initially raise concerns among employees over issues of privacy or trust, a well-executed approach that explains the rationale and alleviates anxieties about sharing information can significantly reduce benefits costs for an institution. This article highlights the approach taken by two institutions—Princeton University and Washington University in St. Louis—and what each saved, and learned, in the process. Princeton: High-Touch Flexibility By Lianne Sullivan-Crowley and Alison Nelson
	Lianne Sullivan-Crowley
	Alison Nelson
	Tom Lauman

As has been the case for many institutions in the past several years, significant losses to endowment values and the prospect of a sustained economic downturn prompted Princeton University leaders to dig deep to uncover additional savings from operating expenses campuswide. In January 2009 we began holding university town hall meetings at which senior leadership discussed budgetary issues and addressed faculty and staff concerns about recession-related impacts. With employee input, the university generated hundreds of ideas for cost-saving measures to reduce our overall operating budget by $170 million over the next two years while minimizing the impact to our human capital.

In that context, every $1 million counts. Within the employee benefits arena, the idea of a medical plan audit to ensure the eligibility of employee dependents was well received from the standpoint that it could help to save real university dollars that would not have to be cut elsewhere. We rolled out Princeton’s inaugural audit in July 2009.

How We Did It

We first researched what other institutions and companies have done in this area. While we found that many have outsourced data collection to a third party, we felt strongly that we should conduct the process internally, in large part because of our belief in high-touch customer service. We also wanted to take direct charge of alleviating employee anxiety about the process, meeting directly with faculty and staff members who had specific concerns.

Since we had never done this before, and because we wanted to make the process as painless as possible, we took the audit directly to employees, visiting individual departments and providing sign-up sheets so that employees could schedule a quick 10-minute meeting with a benefits administrator. We were also careful to establish clear rules for the process and to identify specific documentation that would be required, including marriage certificates and birth certificates of children, and redacted tax returns indicating dependent status.

Because each passing year brings changes to the status of employee dependents—whether because of a change in marital status or the aging of a son or daughter—in addition to required documentation, we asked each employee to fill out a form indicating which family members were currently still eligible for coverage on the plan. Once we heard back from all employees we handed over the census data to the insurance company to do a full refresh of our entire employee population.

Results

For our 6,000 benefits-eligible employees, we have almost a one-to-one correlation of dependents, totaling about 6,000 family members who also receive coverage through our plan. As a result of our 2009 audit, we identified 264 individuals whom we needed to remove from the plan. Based on our calculations of what we spend per employee, conservative estimates suggest that that this measure alone will save the university more than $500,000 in dependent coverage costs annually. We now also have a process in place for all new hires to provide information about dependent status up front, and we encourage all employees to let us know of changes to their status throughout the year, such as when they get married, or within 30 days of when a divorce is finalized.

Advice to Others

1. Offer an amnesty period. In a few rare cases, employees may know they are in violation of dependent eligibility. Allow employees a chance to come clean with that information without reprimand.
2. Do all you can up front to let employees know how the information they provide will be used. In our experience, many employees were compliant without reluctance, since a fair number had gone through something similar in their previous private-sector jobs. While the large majority of our employee population (about 80 percent) understood the reasoning behind this new process, we did receive significant pushback from a small segment. Some employees were offended, saying they felt the process signaled a lack of trust on the part of the institution since they had never before had to provide this kind of personal information. Be ready to do a lot of handholding the first time through the process to assure employees that the institution is not interested in their household earnings or personal financial information. The purpose of a redacted tax return is simply to verify the number of dependents and marital status they have claimed to compare to the family status information contained within their employment records.
3. Be flexible and patient. We learned at the beginning of the process to be flexible regarding what alternate documentation we would accept. For instance, in some cases, employees who had been married for decades could no longer locate their marriage certificates. In those situations, we provided information to help them obtain new copies of these documents and agreed to accept scanned or faxed copies of the originals. In a few cases we even allowed a wedding photo album to serve as proof of marital status. As long as employees are making a good faith effort to comply and to obtain documentation, give them the benefit of the doubt. In our case, flexibility also included extending our deadline—from our original goal of September 30 to the end of October, and ultimately until December 31. Especially the first time through, employees may need ample time to comply.
4. Underscore the rationale for the audit. From the beginning, our communication about why we were doing this stressed the value of such a measure for employees. Highlight the effort as a good business practice that ultimately keeps the costs of benefits down while maintaining quality benefits for all who are eligible and helping to preserve jobs by eliminating cuts to other parts of the institution budget.

Lianne Sullivan-Crowley is vice president for human resources, and Alison Nelson is benefits director, Princeton University, New Jersey. E-mail: lsulliva@princeton.edu; anelson@princeton.edu.

WUSTL: Signed Affidavits

By Tom Lauman

Within the employee benefits arena in particular, we routinely look for ways to contain costs or at least slow budget increases from year to year. About four years ago we decided to conduct an informal audit of our dependent tuition benefit. This is a very generous benefit we offer with an average payout of about $14,000 in tax-free assistance to eligible employee family members. At that time about 1,000 of our 12,000 benefits-eligible employees were participating in the program. A simple sample of this group revealed that about 10 percent of the total who were signed up for the benefit couldn’t provide documents to support eligibility. This sent up a red flag that we needed to take a closer look at enforcing our eligibility requirements. Since that time, we started requiring documentation from all employees who apply for the university’s child tuition benefit, including birth certificates and general tax documents showing status of spouse and dependents.

That also got us thinking bigger picture. Because we spend up to four times on dependent health care compared to what we spend on the institution’s child tuition benefit, we decided that an audit of our medical plan was in order.

How We Did It

Prior to our open enrollment period in fall 2007, we conducted our first random audit of dependents enrolled in our health-care program. We compiled a sample group of about 350 employees using a computer-generated list that selected every 14th name from an alphabetical listing of faculty and staff. This approach provided a good cross-section of employee demographics and employment scenarios. From this group of employees, we requested documentation based on the scenario that each fell under according to our records, including records of children or stepchildren, spouses, domestic partners, and so forth. This resulted in some significant pushback from a number of employees who felt singled out or who were upset by the process. Even some senior members of administration and senior faculty voiced consternation that they had to provide a marriage certificate or tax documentation. However, the results of this first audit revealed that about 5 percent of employees were carrying a spouse or child on their plan who were no longer eligible. Cancellation of coverage of ineligible dependents revealed by this random sample resulted in a savings to the institution of about $50,000.

In fall 2008, we performed another random sample of about the same number of employees, using the same computer-generated method for arriving at a new list. This time, however, instead of requiring proof of documentation from selected employees we created an affidavit that each was required to sign. The affidavit clearly defined terms of eligibility for dependents and noted that the university reserved the right to request documents from employees at a later date if deemed necessary to confirm dependent eligibility. Our legal counsel also suggested that we add language clearly stating that cancellation of coverage and/or potential termination of employment would be a possibility if an employee was found to have provided false information. Under this scenario we again identified about 5 percent of cases where individuals were no longer eligible for coverage. However, we experienced much less pushback and opposition with the signed affidavit approach.

For 2009 we decided to conduct a full audit of employee dependents. Weeks ahead we notified employees via e-mail to ensure that every benefits-eligible employee was aware of the process. We also posted information in the university newspaper and on our Intranet site and made presentations to senior management teams and our faculty senate to emphasize the importance of getting all employees to comply. Since much of our employee benefits information is communicated via e-mail (about 90 percent of our employees have a university e-mail account), we decided to conduct the audit electronically. This entailed working with our IT staff to develop a process to distribute and collect signed affidavits online. Confirmations of receipt of affidavits were generated electronically, and e-mail reminders were sent to employees who failed to respond by the deadline. (For a few stragglers, we had to remind them of the possibility that their coverage would be cancelled if we did not hear from them.) For the remainder of employees without e-mail, we collected paper copies of their signed affidavits. The process, which we began in September, was completed by November (we extended the deadline twice to accommodate stragglers).

In light of our specific wording in the affidavit regarding the possibility of requesting support documentation, we decided to conduct another random audit in April 2010 to verify information provided in connection with the Fall 2009 audit. We requested documents from 350 employees based on their family situations, making certain to eliminate employees who had been selected for either of the two previous random audits. We plan to continue conducting both the full and random audits each year.

Results

For the approximately 6,200 employees receiving dependent coverage, we identified 227 children and 69 spouses or domestic partners who were no longer eligible for coverage under our plan. This has resulted in a savings of upwards of $800,000 to the institution. In addition to future plans for conducting an audit each fall, we now also require a signed affidavit of all new employees upon hiring. By normalizing the process and making it mandatory for all employees who receive dependent benefits, we no longer receive any pushback from employees. We also have the full support of senior management, who appreciate the fiduciary responsibility we have assumed in weeding out wasteful spending.

Advice to Others

1. Centralize administration of the process. Plan to have one person in charge who is responsible for tracking receipt of forms and ensuring that confirmations and reminders get sent to employees.
2. Automate the process. Especially for larger institutions that must collect information from many employees across the university, develop a system that is as quick and painless as possible for employees and for those responsible for data collection.
3. Anticipate employee resistance. Don’t be surprised if employees provide significant pushback or express anger about the process. While it can be hard not to take comments personally, swallow your dismay and instead plan to spend ample time during the first several cycles explaining the need to streamline costs so that the institution can spend its benefits dollars wisely.
4. Find the teachable moment. Finally, use the process as a way to educate employees about their benefits and eligibility requirements for family members. You may be surprised how many employees are unclear about the terms of eligibility or even unaware of who is listed as a dependent on their plan.

Tom Lauman is director of benefits, Washington University in St. Louis, Missouri. E-mail: tomlauman@wustl.edu.