PER CURIAM:
This lawsuit involves a claim for long-term disability benefits under a group plan (“the Plan”) governed by the Employee Retirement Income Security Act of 1974 (“ERISA”). Paul Mitchell (“Mitchell”) claims that Fortis Benefits (“Fortis”) wrongfully terminated his long-term disability benefits, thus abusing its discretion, in violation of various procedures under ERISA and North Carolina law. Fortis appeals from the district court’s denial of its motion for summary judgment, the court’s judgment in Mitchell’s favor on his claim for disability benefits, and the awarding of attorney’s fees to Mitchell. After careful review, we affirm in part, vacate in part, and remand for further proceedings in accordance with this opinion.
I.
Mitchell began working for Advanced Polymer, Inc. (“Advanced Polymer”)
as a sales representative in the textile chemical industry on September 1, 1997. Mitchell’s job involved a substantial amount of travel, , driving an average of five hours a day, for a total of approximately 1500 to 2000 miles per week. According to Mitchell, he began experiencing symptoms of back pain in June 1998.
In April 1999, Mitchell’s treating physician, Mark A. Goodson, M.D.
determined that Mitchell suffered from chronic lumbar discogenic disease primarily due to degenerative changes in his lower back. Dr. Goodson advised Mitchell against surgical intervention. He did recommend a regular course of care with a chiropractor, physical therapy, a low impact exercise regimen, and anti-inflammatory medication. He further advised Mitchell “that 1500 miles per week in an automobile is certainly contraindicated for his low back problem” and would likely “advance the progression of his disease.” J.A. 897. Dr. Goodson suggested that Mitchell obtain a “sit down job where he would be able to stand up and walk around approximately every 20 minutes.”
Id.
By November 1999, Mitchell’s condition had worsened. He complained to Dr. Goodson that his back pain increased with prolonged sitting or prolonged driving and that he found it difficult to assume an erect posture after a long drive. Once again, Dr. Goodson recommended that he discontinue the extensive travel. On November 16, 1999, Mitchell submitted a
long-term disability claim based on complaints of chronic back pain. By November 17, 1999 he was no longer reporting to work. Mitchell underwent further testing on November 22, 1999 and radiology reports revealed that he suffered from mild scoliosis of the lumbar spine with slight retrolisthesis of L2 and L3. Later that month, Mitchell requested a prescription for pain from Dr. Goodson and was prescribed Darvocet-N 100.
On December 6, 1999, Mitchell reported to Dr. Goodson that his pain, which he described as “a burning type pain with an occasional ‘numb’ type pain,” had improved since he had resumed physical therapy and quit working. Record on Appeal, Doc. Entry #22 at 542. On December 10, 1999, Mitchell saw T. Scott Ellison, M.D., an orthopedic specialist. Dr. Ellison obtained x-rays of Mitchell’s back, which revealed a degenerative collapsing lumbar scoliotic pattern, apex to the left. He noted that Mitchell’s disc spaces were fairly well preserved, with some narrowing of L4-5 with facet joint arthrosis. After reviewing Mitchell’s MRI scan from February 16, 1999, Dr. Ellison indicated that there were degenerative disc signal changes at multiple levels. Dr. Ellison advised Mitchell that his symptoms could be caused by the degenerative disc in his back, but that his lifestyle habits, namely extensive driving could be the cause as well.
Mitchell’s claim was referred to Fortis’ Clinical Services Department (“Clinical Services”) for review on March 6, 2000. Based on medical and physical-therapy notes from December 22, 1999 through February 23, 2000, Mitchell’s condition had improved — he showed increased range of movement and his area of pain had diminished. However, Dr. Goodson determined on February 22, 2000 that he “had suffered permanent partial impairment of his spine 5% relative to his lower thoracic and lumbar spine injuries requiring long-term pain management and likely long-term chiropractic manipulative care.” J.A. 900. Nevertheless, Clinical Services reviewed Mitchell’s medical records and concluded that no compelling medical information supported limitations from a primarily sedentary position that allowed for position changes. However, Clinical Services noted it “appears pain would prevent claimant from performing his occupation on a full-time basis.” Record on Appeal, Doc. Entry # 22 at 486.
On March 24, 2000, Clinical Services conducted a phone interview of Mitchell. Its report indicates that Fortis was aware that Mitchell’s therapy was being phased out and that he was able to drive for 10-30 minutes at a time, but experienced significant pain when driving 30-45 minutes. Mitchell informed Clinical Services that he had discussed the possibility of another sedentary job with his employer and that he did not know whether Advanced Polymer would be willing to decrease the number of miles he had to travel. Clinical Services concluded that the medical information supported a finding that Mitchell’s physical limitations prevented him from returning to work in his own occupation.
On March 28, 2000, Mitchell received notification of Fortis’ decision to grant his claim for long-term disability. Fortis considered November 17,1999 as his disability onset date, commencing benefits on February 17, 2000 (following the three months expiration of his qualifying period, as described in the Plan). On May 11, 2000, Mitchell’s case manager from Concentra Managed Care Services, Inc. (“Concentra”),
Cordeila Bortner, accompanied him
to his appointment with Dr. Goodson.
Ms. Bortner asked about the feasibility of Mitchell returning to work with a decreased route, given his improvement. Dr. Goodson prescribed a Functional Capacity Evaluation (“FCE”) to determine what level of activity Mitchell could tolerate. The following day, Fortis deferred all activity pending clarification of issues surrounding the worker’s compensation claim filed by Mitchell, the recent medical records noting improvement, and their bearing on Mitchell’s disability.
In May 2000, Mitchell participated in the FCE. Both Dr. Good-son, claimant’s treating physician, and Dr. Craig S. Heligman, from Clinical Services, reviewed the results and generally found that Mitchell would be able to do light work.
Although, Mitchell’s low back pain continued to improve, he reported to Dr. Goodson that he had developed pain over his left hip and buttocks. On January 19, 2001 Mitchell was diagnosed with moderate degenerative joint disease of the hip. Dr. Goodson noted that he remained active and was not taking Darvocet, but remained on Neurontin. Dr. Goodson advised his patient to continue chiropractic care, his home exercise program, and the Neuron-tin. Mitchell continued to see Dr. Goodson on a regular basis.
In January 2001, a nurse with Clinical Services reviewed Mitchell’s file and found that based upon the “available medical records” and multi-discipline reviews, including the peer review performed by Dr. Heligman, his condition had improved such that he had the physical capacity to perform light work demands and could return to his own occupation. Record on Appeal, Doc. Entry # 22 at 275-76. In the latter part of February 2001, Mitchell reported to Dr. Goodson that “his pain had significantly decreased since he quit working, that he shifts positions frequently, that he never sits for more than 20 minutes at a time, that he never drives for more than 60 minutes at one time without having an extended rest, ... and that his hips were bothersome.” J.A. 908.
On March 9, 2001 Fortis advised Mitchell that he no longer met the requirements for long-term disability benefits under the Plan and that benefits would cease immediately. Fortis contended in its letter denying further benefits that Mitchell was able to perform at least light-duty work with recommendations to allow for position changes every 25-30 minutes. Fortis stated that, “[p]er a discussion with Kuni Nakamura, of Advanced Polymer, [Mitchell] was allowed to take rest breaks at will.” Record on Appeal, Doc. Entry # 22 at 263. The letter also stated that Mitchell’s “employment as a Salesman is defined as a LIGHT physical occupation by both the Department of Occupational Titles and the Department of Labor Standards, under the occupation of Network Control Operator.”
Id.
at 264. Fortis’ letter concluded by stating that Mitchell’s records did not indicate any medical conditions that would physically limit him from performing his own occupation.
Mitchell appealed Fortis’ denial of further benefits, and on July 11, 2001, Fortis arranged for an independent medical examination before Robert Saltzman, M.D., an orthopaedic surgeon. Based on his review of the medical records and his own
examination, Dr. Saltzman concluded that Mitchell was capable of medium duty work, 8 hours per day in a 40 hour work week. The only limitation placed on Mitchell was for him to alternate between sitting and standing as he felt necessary.
On August 29, 2001, Fortis notified Mitchell that it had completed review of his first appeal from the claim denial and concluded that the decision to deny the claim was appropriate. The letter referred to the updated medical records, the result of the independent medical examination, and a labor market survey. According to Fortis, the medical records revealed no abnormal neurological findings, no muscle weakness or atrophy, and an ability to work at a light-duty level, if not higher. The labor market study, according to Fortis, identified sales positions — classified as light duty, in Mitchell’s geographical area that could accommodate his need to alternate positions, which Fortis asserted confirmed Mitchell’s ability to perform the material duties of his occupation. Therefore, Fortis affirmed the decision to deny further benefits. Fortis concluded its letter by notifying Mitchell that he had one further level of appeal prior to exhausting his administrative reviews.
Mitchell appealed the second denial of his claim through his attorney, who argued that his client was incapable of performing the material duties of his prior job. On December 5, 2001, in its response to Mitchell’s final appeal, Fortis notified him that his disability claim must be based on an inability to perform the material duties of his occupation and not his specific job with his former employer. The letter further explained that the FCE and independent medical examination confirmed Mitchell’s ability to work at least at the light duty level. Based on this, Fortis upheld its decision to deny further benefits and advised Mitchell that he had exhausted all administrative reviews. Mitchell responded by filing this lawsuit.
II.
Generally, we review a district court’s order granting summary judgment
de novo. Buzzard v. Holland,
367 F.3d 263, 268 (4th Cir.2004). However, when a plan grants discretionary authority to the decision maker, the deferential abuse of discretion standard of review applies to our review of the eligibility decision.
Id.
(citing
Firestone Tire & Rubber Co. v. Bruch,
489 U.S. 101, 115, 109 S.Ct. 948, 103 L.Ed.2d 80 (1989)). Under this deferential standard of review, a fiduciary’s “discretionary decision will not be disturbed if reasonable, even if the court itself would have reached a different conclusion.”
Smith v. Cont’l Cas. Co.,
369 F.3d 412, 417 (4th Cir.2004) (quoting
Booth v. Wal-Mart, Inc. Assocs. Health & Welfare Plan,
201 F.3d 335, 341 (4th Cir.2000)).
In this case, the Plan contains the following language explicitly granting discretionary authority to Fortis:
we have the sole discretionary authority to determine eligibility for participation or benefits and to interpret the terms of the Policy. All determinations and interpretations made by us are conclusive and binding on all parties.
J.A. 121. The Plan defines “we” and “us” as Fortis Benefits Insurance Company.
Id.
However, the deference owed to a plan fiduciary’s decision is tempered when the fiduciary is operating under a conflict of interest.
Smith,
369 F.3d at 417. When the decision to deny benefits impacts the fiduciary’s financial interests, as in the case of a plan insurer, that conflict is weighed by the court as a factor in deciding whether there was an abuse of discretion.
Bernstein v. CapitalCare Inc.,
70
F.3d 783, 787 (1995). In other words, if a plan administrator acts as both the fiduciary making claim decisions and the insurer paying claims, an inherent conflict of interest exists. Accordingly, deference to the plan administrator will be lessened, but only “to the degree necessary to neutralize any untoward influence resulting from the conflict.”
Doe v. Group Hospitalization and Med. Serv.,
3 F.3d 80, 87 (4th Cir.1993). Under no circumstances may the court deviate entirely from the abuse of discretion standard.
Ellis v. Metro. Life Ins. Co.,
126 F.3d 228, 233 (4th Cir.1997). Instead, the “more incentive for the administrator or fiduciary to benefit itself by a certain interpretation of benefit eligibility or other plan terms, the more objectively reasonable the administrator or fiduciary’s decision must be and the more substantial the evidence must be to support it.”
Id.
In this case it is undisputed that a conflict exists, because Fortis functions as both the Plan Administrator and the insurer, thus deference to Fortis must be tempered.
III.
Fortis contends that the district court erred: (1) by reviewing Fortis’ claim decision
de novo
rather than determining if the decision to deny the claim was reasonable; (2) by disregarding the language in the policy and Fourth Circuit precedent when considering Mitchell’s specific job duties rather than the “material duties” of his “regular occupation,” when concluding that he was disabled; (3) as a matter of law, by accepting the inconsistent opinion of Mitchell’s treating physician over other substantial evidence, including the FCE, a peer review, and an independent medical examination, which it argues supported the denial of benefits; (4) by ignoring the language in the policy and refusing to give Fortis a credit for an overpayment, as well as calculating benefits to include amounts that are not allowed under the policy; and (5) by granting attorney’s fees to Mitchell, because he should not have prevailed on his claim.
A.
Fortis contends that the district court misapplied the abuse of discretion standard discussed
supra.
Fortis alleges that the district court substituted its own judgment for that of Fortis, the Plan Administrator. Further, Fortis asserts that the court gave little to no deference to Fortis’ conclusions based on the court’s perception that Fortis’ conduct constituted “bad faith.” Fortis disputes the court’s assertion that “failure to provide Plaintiff with any notice that termination of his disability benefits was being considered suggests bad faith.”
J.A. 938.
The district court stated that it had “questions about Defendant’s own financial interest as the insure[r] and administrator [which] require this Court to hold Defendant to a higher standard in terms of its objective medical evidence and give less deference to Defendant’s discretionary acts.” J.A. 938-39. Specifically, the district court discussed the standard of review of a Plan Administrator’s decision when it functions as both the insurer and the Plan Administrator. The court also noted that:
In determining the reasonableness of a fiduciary’s discretionary decision, a court applying the modified abuse of
discretion standard may consider, but is not limited to, the following factors:
(1) the language of the plan; (2) the purposes and goals of the plan; (3) the
adequacy
of the materials considered to make the decision and the degree to which they support it; (4) whether the fiduciary’s interpretation was consistent with other provisions in the plan and with earlier interpretations of the plan; (5) whether the decision making process was
reasoned and principled;
(6) whether the decision was consistent with the procedural and substantive requirements of ERISA; (7) any external standard relevant to the exercise of discretion; and (8) the fiduciary’s motives and any
conflict of interest
it may have.
Booth v. Walr-Mart Stores, Inc.,
201 F.3d 335, 342-343 (4th Cir.2000) (emphasis added).
J.A. 917.
Using the above criteria, the district court reviewed Fortis’ decision for reason-ability — determining whether the decisions and interpretations made by Fortis were supported by the evidence. For example, the district court found that it was reasonable, pursuant to the Plan, for Fortis to review Mitchell’s eligibility on a month-to-month basis.
Therefore, Fortis was not bound to its original determination that Mitchell was eligible for long-term disability.
A thorough reading of the district court’s opinion demonstrates that it reviewed Fortis’ decisions by looking at its interpretations and finding them reasonable when Fortis’ interpretations were supported by other language in the Plan or was consistent with other outside information upon which they purported to rely.
Therefore, we find that the district court did use the proper modified abuse of discretion standard of review in this case.
B.
To qualify for long-term disability benefits, Mitchell must satisfy the Plan’s Occupation Test and demonstrate that:
during the first 36 months of a
period of disability
(including the
qualifying
period), an
injury,
sickness, or pregnancy requires that you be under the
regular care and attendance
of a
doctor,
and prevents you from performing at least one of the
material duties
of your regular occupation.
Record on Appeal,
Doc.
Entry # 22 at 10. Therefore, to cease paying disability benefits, after initially finding Mitchell eligible, there must be evidence in the record to support Fortis’ determination that Mitchell could now perform all the “material duties” of his “regular occupation.”
We first consider the Plan’s language. The Plan does not include a definition of “your regular occupation.” However, the district court agreed with Fortis that “your regular occupation” did not mean Mitchell’s specific job with Advanced Polymer, but rather referred to a position more specific to Mitchell than “gainful occupation”, which is the language used in the Plan to test when a claimant qualifies for being disabled for more than three years.
J.A. 919. We also consider the Second Circuit’s opinion in
Kinstler v. First Reliance Standard Life Ins. Co.,
181 F.3d 243 (2nd Cir.1999), which interpreted “regular occupation” as “a position of the same general character as the insured’s previous job, requiring similar skills and training, and involving
comparable duties.” Id.
at 252 (emphasis added). The court explained that the “term is defined more narrowly than any means for making a living, but it is not limited to the insured’s particular job.”
Id.
In other words, a reasonable description of a claimant’s regular occupation must also take into account the specific nature of the claimant’s prior employment.
On appeal, Fortis argues that extensive driving is not a material duty of Mitchell’s regular occupation and asserts that the district court substituted its own judgment for the labor market survey provided by Fortis.
The Plan defines “material duties” as:
[T]he sets of tasks or skills required generally by employers from those engaged in a particular occupation. One material duty of your regular occupation is the ability to work for an employer on a full-time basis as defined in the policy.
Record on Appeal, Doc. Entry # 22 at 11.
Fortis claims that because it provided a survey that identified “numerous” sales positions in Mitchell’s geographical area that were suitable to Mitchell’s background, which did not require “extensive travel,” there was thus, substantial evidence that traveling approximately 1500 miles per week was not a material duty of Mitchell’s occupation. Appellant’s Br. at 33.
Fortis’ attempt to use a labor market survey to demonstrate that driving or traveling is not a material duty of Mitchell’s regular occupation is insufficient. By using limiting criteria applicable to Mitchell (light-duty, no lifting over twenty pounds, and the ability to change positions every thirty minutes), Fortis guaranteed that it would find at least a few jobs that Mitchell could possibly engage in with his physical limitations.
Thus, the labor market survey does not support Fortis’ determination that driving or extensive travel was not a general skill required by employers in Mitchell’s occupation.
Consequently, we consider objective information to determine if Fortis’ determination was reasonable. The United States Department of Labor in its Dictionary of Occupational Titles (“DOT”) defines a Chemical Sales Representative as “Light Work — Exerting up to 20 pounds of force occasionally, and/or up to 10 pounds of force frequently, and/or a negligible amount of force constantly (Constantly: activity or condition exists 2/3 or more of the time) to move objects. Physical de
mand requirements are in excess of those for Sedentary Work.” Dictionary of Occupational Titles 262.357-010. The Department of Labor, in its Occupational Outlook Handbook, states that “Sales representatives spend much of their time traveling to and visiting with prospective buyers and current clients.” Bureau of Labor Statistics, U.S. Department of Labor Occupational Handbook 412 (2004-05). The Department of Labor continues, stating—
After the sale, representatives may make followup visits to ensure that the equipment is functioning properly and may even help train customers employees to operate and maintain new equipment.
Working Conditions
Some sales representatives have large territories and travel considerably. A sales region may cover several States, so they may be away from home for several days or weeks at a time. Others work near their “home base” and travel mostly by automobile. Due to the nature of the work and the amount of travel, sales representatives may work more than 40 hours per week.
Although the hours are long and often irregular, most sales representatives have the freedom to determine their own schedule. Sales representatives often are on their feet for long periods and may carry heavy sample products, which necessitates some physical stamina.
Id.
at 413.
The Department of Labor’s in-depth explanation of the travel required by a sales representative supports the district court’s finding that driving was a material duty of Mitchell’s regular occupation. The DOT’S description of a Chemical Sales Representative requiring the exertion of at least “a negligible amount of force constantly” — like the force necessary to drive a car — further supports the district court’s finding that driving was a material duty. Thus, we affirm the district court’s finding and conclude that it was unreasonable and an abuse of discretion for Fort is to find that extensive travel was not a material duty of Mitchell’s regular occupation.
C.
We now review Fortis’ determination that Mitchell was able to perform the material duties of his regular occupation, including extensive travel. Our review, like the district court’s, is limited to evidence that was in the administrative file when the Plan Administrator rendered its decision to terminate Mitchell’s benefits.
Bernstein v. CapitalCare, Inc.,
70 F.3d 783, 788 (4th Cir.1995).
The information Fortis had prior to the March 9, 2001 decision to cease disability benefits to Mitchell was: (1) the FCE, which concluded that Mitchell was unable to sit still for any length of time and suffered from a burning sensation in his lower back when sitting and typing; (2) Dr. Heligman’s review of the FCE and his opinion that the “claimant has the capacity for light work and would be able to perform his occupation of sales representative,” Record on Appeal, Doc. Entry # 22 at 333; (3) Dr. Heligman’s recommendation that Mitchell limit his lifting to between 10 and 25 pounds, and that he change his position frequently “every 30 minutes or so,”
Id.;
and (4) Dr. Goodson’s review of the FCE and his finding that Mitchell was capable of light duty work with restricting his lifting to 20 pounds, changing positions every 20-30 minutes, and sitting no more than six hours in a 24 hour period, and if driving is involved— “limit driving to a max. of 2 hours per 24 hours with a max. of 10 hours per. week.
While driving he should stop every 30 nain. and walk around the car.”
Id.
at 248.
In addition, the record indicates that in January 2001, Dr. Goodson scheduled Mitchell for x-rays to investigate his complaints of hip pain. Mitchell was diagnosed with moderate left hip degenerative joint disease. In February 2001, after the June 2000 FCE, but before the March 9, 2001 denial of benefits, Dr. Goodson opined that Mitchell would not be able to “maintain gainful employment without significant increase in pain symptoms making him totally non-functional.”
Id.
at 255.
The medical evidence does not reasonably support Fortis’ finding that Mitchell was capable of doing all material duties of his occupation. It was unreasonable for Fortis to find that Mitchell could drive extensively every day if he could only drive a maximum of 30 minutes at a time with a significant rest afterward. It would be almost impossible, even given an extended work day, for Mitchell to timely attend his appointments. Further, the facts support the district court’s finding that “the record demonstrates that there was never an official RTW [“return to work”] proposal made or rejected.”
Id.
at 923. Finally, the medical evidence is not in conflict, both Dr. Heligman and Dr. Goodson limited Mitchell’s ability to sit or drive to 30 minutes or less at a time. Therefore, we find that Mitchell was unable to perform a material duty of his occupation, and that even giving Fortis the appropriate deference it was unreasonable for Fortis to find otherwise on March 9, 2001.
After the March 9, 2001 denial, Mitchell appealed and Fortis gathered more medical evidence, presumably to re-evaluate the decision denying Mitchell long-term disability benefits. In August 2001, Dr. Saltzman conducted a physical examination, reviewed “all available medical records,” and found in his independent medical evaluation that the “objective findings are not consistent with the patient’s subjective symptoms. The objective findings are minimal (i.e. his scoliosis compared to a subjective inability and there is no muscle weakness nor atrophy).” J.A. 141. Dr. Saltzman opined that Mitchell was capable of performing “medium category work level for an 8-hour day, 40-hour week, with the ability to sit and stand, alternating as he feels necessary.”
Id.
Dr. Saltzman was aware that Mitchell’s prior position as a sales representative involved driving and frequent sitting. Dr. Saltzman also estimated Mitchell’s level of ability as “there were no tasks that I think the claimant would be unable to perform, other than lifting more than 40-lb on more-than-occasional basis.”
Id.
Dr. Saltzman indicates that he reviewed Mitchell’s medical records from 1998 until February 2001, however, it is also plain that Dr. Saltzman had incomplete medical information, for example “no x-rays were made available for review.” J.A. 140. Nevertheless, Dr. Saltzman concludes that his objective findings were inconsistent with Mitchell’s subjective complaints, Dr. Saltzman did “recommendf ], however, that updated MRI’s of the thoracic and lumbar spine be obtained and x-rays, plain films of the thoracic and lumbar spine be obtained. It may be advantageous to get EMGs of the (L) lower extremity to detect and denervation to the muscle since clinically none are noted.” J.A. 248.
In contrast, both Dr. Heligman and Dr. Goodson opined that Mitchell had some limitations in his abilities and necessitated the ability to change positions every 20-80 minutes. In addition, Dr. Ellison recognized that Mitchell’s extensive driving may have caused (or exacerbated) the degenerative disc in his back. Further, the FCE supported the findings and recommendations made by Dr. Ellison, Dr. Goodson, and Dr. Heligman. The only evidence in the record that clearly supports Fortis’ determination is Dr. Saltzman’s opinion. With an incomplete record and without any further tests, Dr. Saltzman’s opinion— that Mitchell was able to perform all tasks except lifting in excess of 40 pounds — does not outweigh the medical opinions of Dr. Goodson, Dr. Ellison, and Dr. Heligman who limited Mitchell’s abilities to sit and drive to 80 minutes. Thus, we find that Dr. Saltzman’s opinion cannot alone provide enough reasonable support for Fortis’ decision, even according Fortis tempered deference. Therefore, we affirm the district court’s findings on this issue.
D.
Next, Fortis contends that the district court erred when the court failed to offset for the Social Security benefits that Mitchell received, when it added to the benefit calculation money received by Mitchell as a “bonus”, and when the district court granted Mitchell benefits beyond what was remaining in the first thirty-six months.
First, despite Fortis’ assertion to the contrary, the court did offset the benefits payment by the Social Security benefit Mitchell received. The district court awarded Mitchell $108,241.00 for the total amount of past-due disability benefits owed under the terms of the Plan.
The district court determined that $3,820.00 was the initial monthly benefit for the period of May 2001 until May 2002. In May 2002 Mitchell began receiving Social Security disability benefits, thus the district court reduced the monthly benefit owed by Fortis to Mitchell to $2,590 per month for the period of May 2002 until February 16, 2004.
Second, Fortis does not cite to any language in the district court’s opinion to support its assertion that a “bonus” was included in the court’s calculations of Mitchell’s benefits. Further, the record shows that Fortis determined in an internal memo, dated August 22, 2001, that $4,830 was commissions and should be added to Mitchell’s income. Fortis originally calculated the “adjusted pre-db [“pre-disability”] earnings $69,000/12 = $5,750 + 214.80 (2,577.63/12) = $5,964.80 — release underpayment to claimant.” Record on Appeal, Doc. Entry # 22 at 145. The district court calculated Mitchell’s monthly disability benefits to be considerably lower, as described above. Thus, the record does not support Fortis’ contention that the district court included a “bonus” in its calculation of benefits.
Finally, Fortis is correct in its contention that the district court granted Mitchell “Reinstatement of disability benefits
under the Plan, in the amount of $2,590 per month, from February 17, 2004, through the present,” J.A. 1040, in the amount of $129,795.93 which is in direct conflict with the Plan. According to Fortis’ Plan language, “after 36 months of
disability,
an injury, sickness, or pregnancy prevents you from performing at least one of the
material duties
of each
gainful occupation
for which your education, training, and experience qualifies you.” Record on Appeal, Doc. Entry #22 at 10. The district court did not make a finding that Mitchell satisfied the Plan’s requirement that he be unable to perform a material duty in his “gainful occupation”, making him eligible for disability benefits after the initial thirty-six months. In fact, the district court specifically recognized the difference between qualifying for disability for the first thirty-six months and qualifying later after the initial thirty-six months. Therefore, we reverse the district court’s grant of $129,795.93 for disability benefits after February 17, 2003.
E.
Fortis argues that because Mitchell should not have prevailed in his claim, the district court’s grant of attorney’s fees was in error. We review the district court’s award for abuse of discretion.
Metropolitan Life Ins. Co. v. Pettit,
164 F.3d 857, 865 (4th Cir.1998). In awarding attorney’s fees, a district court should use the five factors articulated in
Quesinberry v. Life Ins. Co. of North Am.,
987 F.2d 1017, 1029 (4th Cir.1993) (en banc), as a guide, keeping in mind the remedial purposes of ERISA. The five factors are: (1) degree of the opposing parties’ culpability or bad faith, (2) the ability of opposing parties to pay fees, (3) whether the fee award would deter others similarly situated, (4) whether the parties requesting fees sought to benefit other claimants or to resolve a significant ERISA-related legal question, and (5) the relative merits of the parties’ positions.
Id.
Although the district court did not go into great detail about the basis for granting Mitchell attorney’s fees we can surmise that the court found that Fortis’ actions suggested bad faith and that the evidence Fortis purported to rely on in making its claims decision was not substantial and supportive of its decision to terminate Mitchell’s disability insurance benefits as of March 2001. Accordingly, granting of attorney’s fees would likely deter Fortis from making future hasty and unsupported decisions. Therefore, we find that attorney’s fees were properly granted to Mitchell.
Fortis also disputes the district court’s calculation of $120,000 for attorney’s fees. Mitchell’s counsel asserts that he spent approximately 542.10 hours on this case at a rate of $275 an hour and requested $149,077.50 in attorney’s fees. A court may award, in its discretion, reasonable attorneys’ fees and costs to a prevailing plaintiff in a ERISA action.
See
29 U.S.C. § 1132(g)(1) (2005). Thus, we review the amount of the award for abuse of discretion.
See Johnson v. Hugo’s Skateway,
974 F.2d 1408, 1418 (4th Cir.1992).
In calculating an award of attorneys’ fees, a court should usually “determine! ] a ‘lodestar’ figure by multiplying the number of reasonable hours expended times a reasonable rate.”
Daly v. Hill,
790 F.2d 1071, 1077 (4th Cir.1986). The district court should generally be guided by the particular factors articulated in
Barber v. Kimbrell’s, Inc.,
577 F.2d 216, 226 n. 28 (4th Cir.1978), when deciding what constitutes a “reasonable” number of hours and rate.
See also Brodziak v. Runyon,
145 F.3d 194, 196 (4th Cir.1998). Because we
have only affirmed the district court’s finding that Fortis wrongly denied Mitchell benefits during the initial thirty-six months period and the district court has given us little information on how it determined the amount of fees, we find that the district court must re-calculate the amount of the award of attorney’s fees using the above cases as guidance.
IV.
Based on the foregoing, we affirm the district court’s finding that Fortis abused its discretion when it terminated Mitchell’s long-term disability benefits during the initial thirty-six months of disability and its finding that Mitchell was eligible to receive attorney’s fees in this case. We vacate the district court’s decision that Mitchell was eligible to receive long-term disability benefits after the initial thirty-six months period and we remand for recalculation of the amount of attorney’s fees in accordance with this opinion.
AFFIRMED IN PART, VACATED IN PART, AND REMANDED.