O'Connell v. Hartford Life And Accident Insurance Company

• Court: District Court, D. Massachusetts
• Decided: March 24, 2023
• Docket: 1:21-cv-10587
• Status: Unknown

Opinion

UNITED STATES DISTRICT COURT DISTRICT OF MASSACHUSETTS ____________________________________ ) DIANE O’CONNELL, ) ) Plaintiff, ) v. ) ) Civil Action No. 21-CV-10587-AK HARTFORD LIFE AND ACCIDENT ) INSURANCE CO., ) ) Defendant. ) )

MEMORANDUM AND ORDER ON CROSS-MOTIONS FOR JUDGMENT ON THE ADMINISTRATIVE RECORD

A. KELLEY, D.J. Plaintiff Diane O’Connell (“O’Connell”) has filed suit pursuant to the Employee Retirement Income Security Act of 1974 (“ERISA”), 29 U.S.C. § 1132(a)(1)(B), to recover long- term disability benefits terminated by Defendant Hartford Life and Accident Insurance Company (“Hartford”). [See Dkt. 1]. O’Connell and Hartford have filed cross-motions for judgment on the administrative record. [Dkts. 37, 40]. For the following reasons, O’Connell’s motion for judgment on the administrative record [Dkt. 40] is DENIED, Hartford’s motion for judgment on the administrative record [Dkt. 37] is likewise DENIED, and this case is REMANDED to Hartford for further proceedings consistent with this Order. I. BACKGROUND The Court has reviewed the vast administrative record and recites here only those facts necessary to understand what led to this action. [See Dkt. 44]. Further details relevant to the Court’s analysis will be discussed as needed. O’Connell participated in an employee welfare benefit plan (the “Plan”) sponsored by her former employer, PricewaterhouseCoopers, LLC (“PwC”). [Dkt. 1 at ¶¶ 4, 9, 13]. The Plan is governed by ERISA. [Dkt. 44 at 302]. Hartford is the Plan and claims administrator. PwC bears the responsibility of paying short-term disability (“STD”) benefits, while Hartford is responsible for paying long-term disability (“LTD”) benefits. [Id. at 834-46]. In relevant part, a Plan participant is “disabled” when, during the “elimination

period and for the next 60 months,” the employee is prevented by mental illness “from performing one or more of the essential duties of [her] occupation.” [Id. at 296-97]. The Plan defines “occupation” as “it is recognized in the general workplace,” not “the specific job [the employee] is performing for a specific employer at a specific location.” [Id. at 299]. A. Onset of Anxiety O’Connell claims she suffers from lifelong anxiety. [Id. at 629]. O’Connell worked at PwC as in-house counsel for approximately ten years before the pressure of her job began interfering with her ability to work. [Id. at 684]. O’Connell states that as early as 2014, she asked for a break from her responsibilities because “practice volume tripl[ed] and [the] company [did] not bring[] in any additional support.” [Id.]. She spoke with leadership about the issue,

and a new attorney was hired to assist. [Id. at 684-85]. However, per O’Connell, there were interpersonal issues with the new hire, and “her presence actually exacerbated the situation” to the point that “nothing seemed to work” and the “stress of [the] situation caused [O’Connell’s] anxiety to resurface.” [Id.]. In February 2018, O’Connell requested a transfer to another group within PwC, and she began in that role on May 1, 2018. According to O’Connell, her new position “heavily relied on financial diligence and working capital accounting knowledge which [she] did [not] have but [PwC] thought they could train [her],” though she ultimately received “little to no training or support.” [Id. at 630, 685]. This new position, along with other personal events, such as buying a new house requiring “significant renovation” and physical health concerns, impacted O’Connell’s mental health, causing her anxiety to resurface and impeding her ability to work at PwC. [Id. at 630]. On October 13, 2018, she was told that she was “not working out in this new position,” and on October 31, 2018, PwC put O’Connell on a performance plan. [Id. at 630, 685]. As a result, O’Connell “began to make even bigger

mistakes, panicking at every email and phone call,” and she reached the point where she felt she could no longer function at work. [Id. at 685-86]. O’Connell visited her primary care physician, Dr. Arlene Perkins (“Dr. Perkins”), on November 6, 2018. [Id. at 920]. Dr. Perkins noted that O’Connell was experiencing “some increased stress at work and new home with mortgage and possible job loss.” [Id.]. Dr. Perkins reported that O’Connell was “having . . . feeling[s] of dread” and could not function, think, or concentrate. [Id.]. Dr. Perkins diagnosed O’Connell with anxiety, referred her to psychology and psychiatry, and prescribed Xanax, which O’Connell chose not to take, opting instead for psychotherapy treatment. [Id. at 922, 959, 996, 1026]. B. Approval of Short-Term Disability Benefits

In or about December 2018, O’Connell applied for STD benefits as a result of mental illness, which Hartford granted from January 11, 2019, to February 10, 2019, and later extended through June 23, 2019. [Id. at 1002, 1010]. Hartford relied on conversations with and forms submitted by O’Connell’s therapist, Gary Karshmer (“Karshmer”), in reaching this decision. [See id. at 995-1001, 1007-1009, 1028-30]. Karshmer is a trained and certified psychoanalyst and a licensed social worker who works frequently with patients with generalized anxiety disorder (“GAD”). [Id. at 560]. Karshmer had treated O’Connell for fifteen years and diagnosed her with GAD. [Id. at 677, 1028-29]. C. Approval of Long-Term Disability Benefits On April 15, 2019, O’Connell applied for LTD benefits. [Id. at 840]. Hartford requested further medical evidence before approving O’Connell’s LTD benefits on June 28, 2019. [Id. at 218; see id. at 223, 227, 230, 233, 236]. Hartford’s approval letter explained that monthly

benefit payments would continue while O’Connell “meet[s] the policy definition of [d]isability,” and Hartford would “[p]eriodically . . . contact . . . [O’Connell] and [her] physician to obtain updates concerning [her] treatment and condition.” [Id. at 219-20]. Hartford also informed O’Connell that the Plan required her to apply for social security disability insurance (“SSDI”) if her disability was expected to last at least 12 months. [Id. at 220]. In the six months following O’Connell’s LTD benefits approval, Hartford sent O’Connell five requests for updated medical records and enclosed Attending Physician Statement (“APS”) forms and a Behavioral Functional Ability (“BFA”) form for Karshmer to complete. [Id. at 167, 171, 173, 181, 197]. Separately, Hartford wrote to Karshmer, asking him to provide “mental health treatment notes for the period from 11/01/2018 to present” and “all medical records . . .

pertinent to process [the] claim.” [Id. at 175, 183, 191, 199, 210, 214]. Harford requested similar records from Dr. Perkins four times. [Id. at 178, 187, 194, 207, 223]. O’Connell provided Hartford the notes from her November 2018 visit to Dr. Perkins and two APS forms from Karshmer, dated August 26, 2019, and October 16, 2019. [Id. at 917-18, 920-28, 939-40]. On the August 26th APS, Karshmer noted mild to moderate impairment of attention and concentration and moderate impairment of memory based on his observation and O’Connell’s self-reporting, though no testing was used for the mental health examination. [Id. at 917]. On the October 16th APS, Karshmer listed O’Connell’s self-reported symptoms as “general anxiety, irritability, restlessness, worry, mind going blank, [and] sleep disrupted,” and “muscle tension, mind going blank [and] freezing up, agitation, and not so verbally sharp while making connections in therapy sessions” as observed symptoms. [Id. at 939]. Karshmer noted mild to moderate impairment of attention, concentration, and memory, though he did not indicate how he assessed O’Connell, i.e., through observation, self-reporting, or testing. [Id.].

In addition to the APS forms, Karshmer submitted a BFA on October 16, 2019. [Id. at 941-43]. He selected moderate to marked limitations in “directing, controlling or planning activities of others”; “performing repetitive or short-cycled work”; and “influencing people in their opinions, attitudes, and judgments.” [Id. at 941 (capitalization omitted)]. Karshmer selected marked impairments in “performing under stress”; “attaining precise set limits, tolerances, and standards”; “working under specific instructions”; and “reliability consistency.” [Id. at 942-43 (capitalization omitted)]. D. Initial Termination of LTD Benefits and Dr. Dale Panzer’s Review On December 12, 2019, Hartford terminated O’Connell’s LTD benefits after determining O’Connell no longer met the Plan’s definition of “disability.” [Id. at 153]. In the termination

letter, Hartford briefly referenced one of Karshmer’s APS forms but largely based its decision on an independent medical review by Dr. Dale Panzer (“Dr. Panzer”). [Id. at 154]. Hartford concluded that “[t]he weight of the medical evidence in O’Connell’s claim file does not support psychiatric functional impairment,” because “Dr. Panzer opined that the clinical data does not support any psychiatric functional impairment due to mental illness for the current timeframe.” [Id. at 156]. Dr. Panzer, a board-certified psychiatrist, had been recommended by a third-party vendor. [Dkt. 38 at 11]. Around December 3, 2019, Dr. Panzer reviewed Karshmer’s APS forms and Dr. Perkins’ notes, in addition to speaking with both of them for approximately 20 minutes each [Dkt. 44 at 957, 961, 962], but he did not speak with or examine O’Connell. [Dkt. 42 at 10]. Dr. Panzer concluded that O’Connell “has supported diagnoses of unspecified anxiety disorder and unspecified depressive disorder,” but he “did not find restrictions or limitations were clinically supported for the current timeframe due to psychiatric or psychological

condition.” [Dkt. 44 at 957]. Dr. Panzer explained that [w]hile the claimant reported difficulty with focus in her work environment in the past, there was no mental status examination testing on file to support cognitive impairment. While the claimant reported focus difficulty in her place of work due to harassment and unfair treatment by her employer, there was no information in the record about her employer’s treatment of her nor was there mental status examination testing that would support the claimant’s reported concentration difficulty. Likewise, while the claimant reported overall impairment in performing work-related activity, there was no indication in the records that the claimant had difficulty with day-to-day activities such as cooking, cleaning or grocery shopping.

[Id. at 959].

Dr. Panzer also noted that there “was no indication [O’Connell] had been referred to a higher level of care or had actually seen a psychiatrist based upon the available clinical record,” nor was she “taking medication for her condition.” [Id.]. Dr. Panzer concluded that O’Connell’s “treatment was inconsistent for what is expected for an ongoing psychiatric condition that might be functionally impairing.” [Id.]. Dr. Panzer acknowledged that “[t]here was no specific information about [O’Connell’s] workplace environment on file.” [Id. at 959]. Neither Dr. Perkins nor Karshmer responded to Dr. Panzer’s report. [See id. at 159, 160]. E. O’Connell’s Appeal and Harford’s Final Termination of LTD Benefits O’Connell appealed Hartford’s termination of her LTD benefits on May 15, 2020. [Id. at 606-22]. As part of that appeal, she submitted a personal statement, a letter from Karshmer, and reports from Dr. Michael Kahn and Michael Raia, a vocational consultant. [Id. at 370-72, 376- 378, 629-632, 641-51]. Hartford requested a medical expert from a third-party vendor to review the entire record, excluding Dr. Panzer’s report and including O’Connell’s new submissions. [Id. at 571-77]. The third-party vendor selected Dr. Heather Joppich, a board-certified psychologist, to review O’Connell’s file. [Id.]. 1. O’Connell’s Statement

O’Connell submitted a four-page statement, dated April 28, 2020, recounting the events that led to her leaving work, the impact of her anxiety, and the treatment she had received. [Id. at 629-632]. O’Connell explained feeling “paralyzed” and “drained,” having “difficulty focusing,” and panicking “every time an email [came] in or the phone [rang].” [Id. at 631]. O’Connell wrote, “I realize that intellectually I am more than capable of doing my job, but it’s the emotional element that I am still working on.” [Id. at 631]. She explained that she was “exploring what this means and whether [she] can resume [her] duties if she learn[s] how to deal with situations differently, or if [she] should adjust [her] work environment into something less demanding.” [Id.]. 2. Karshmer’s Letter

Karshmer submitted a letter describing O’Connell’s condition and providing an update on her treatment on April 30, 2020.1 [Id. at 370]. He explained that “[g]iven the nature of [O’Connell’s] occupation, [she] would not have days to recover from an anxiety-provoking episode, which could include a meeting, a telephone call, [or a] deadline,” and that O’Connell’s treatment was working well for her because she “does NOT have a structured work schedule and performance demands and requirements.” [Id. at 371].

1 The parties’ dispute the characterization of Karshmer’s letter; Hartford claims it is an “advocacy letter,” while O’Connell labels it a “treatment summary.” [Dkt. 38 at 13; Dkt. 53 at 5]. 3. Dr. Michael Khan’s Medical Evaluation O’Connell arranged for Dr. Michael Khan (“Dr. Khan”), a staff psychiatrist and Director of Medical Student Education at Beth Israel Deaconess Medical Center, to conduct a psychiatric evaluation of O’Connell. [Id. at 676]. Dr. Kahn met with O’Connell for ninety minutes on

March 31, 2020, in addition to reviewing her medical records. [Id.]. He submitted a report dated April 30, 2020. [Id.]. Dr. Khan noted that “[t]he main stressor to [O’Connell] was her new job, which she felt she was not suited for. [Id.]. She constantly felt overwhelmed and even confused by the demands of her new position.” [Id. at 677]. She “dealt with it” until “she felt she could no longer take the pressure and resulting anxiety, and left the job.” [Id.]. Dr. Kahn’s mental status examination revealed a “casually dressed and groomed woman . . . friendly . . . somewhat over-detailed at times, but without any evidence for disorder of thought form or content.” [Id. at 677]. There was “no evidence of delusions, hallucinations, hopelessness, or suicidality,” and Dr. Kahn reported that O’Connell was “alert and fully oriented, her fund of knowledge is good, she recited the months of the year backwards quickly and

without error, performed simple calculations easily, and remembered 2/3 words after 3 minutes, and got the third with a prompt.” [Id. at 377]. Still, Dr. Kahn found that O’Connell’s job was stressful and, given her trauma history, she was “particularly vulnerable to stressors over which she feels she has no control” but had “nevertheless found ways to cope with this stress for three years, until changing positions within the company.” [Id. at 378]. He concluded that while O’Connell was generally non-depressed and functional, that was a result of her not being in an overwhelming work environment, and this would not be the case if she were to return to her prior job. He concurred with Karshmer’s reporting of O’Connell’s limitations and concluded that the “likelihood of her symptoms returning as of 12/12/19 would be very high, and the limitations to work function have been substantiated.” [Id. at 678]. Dr. Kahn explained that O’Connell was “vulnerable to being in situations where the demands made on her [are] not subject to her control.” [Id.]. 4. Michael Raia’s Vocational Assessment

At O’Connell’s request, Michael Raia (“Raia”) provided a vocational assessment report based on a review of O’Connell’s medical records. [Id. at 641-651]. Raia, a vocational consultant and rehabilitation counselor, reviewed the medical information and Hartford’s policy terms to determine whether O’Connell was able to perform the duties of an attorney, but he did not meet with O’Connell. Raia concluded that O’Connell’s condition presents many severe limitations in her function . . . [which] all combine and compound to preclude Ms. O’Connell from being able to meet the mental requirements of her occupation as an [a]ttorney. It is my opinion that she is incapable of performing the material and substantial duties of her regular occupation and she is incapable of engaging in any full or part time employment of any kind at this time.

[Id. at 651]. Based on the records and giving more weight to O’Connell’s treating providers than Hartford’s consultants, Raia found that O’Connell’s condition would prevent her from performing the functions of her occupation as an attorney.2 [Id. at 644-45, 650-51]. 5. Dr. Heather Joppich’s Independent Medical Review In or around June 2020, Hartford, through a third-party vendor, had Dr. Heather Joppich (“Dr. Joppich”), a board-certified psychologist, review O’Connell’s medical records.3 [Id. at 572]. Dr. Joppich summarized the medical evidence and concluded that the record did not

2 Raia evaluated the requirements of O’Connell’s occupation as an attorney using the Dictionary of Occupational Titles (“DOT”) and O*NET, both of which evaluate occupations as they are performed in the national economy and are not specific to O’Connell’s job at a particular place. [See Dkt. 44 at 696-699, 700-01, 704].

3 Dr. Joppich considered Dr. Perkins’ November 2018 clinical note; Karshmer’s APS forms; summaries of other communications with Karshmer; Karshmer’s BFA form; O’Connell’s statement; Dr. Kahn’s evaluation; Raia’s vocational assessment report; and an undated job description of a director of financial due diligence at PwC that Hartford requested. [Id. at 572]. Dr. Panzer’s report was not included. support limitations that would preclude O’Connell’s occupational functioning from December 13, 2019, through the present. [Id. at 575]. While Dr. Joppich acknowledged self-reported impairments in attention, concentration, and memory, she found “there was no objective documentation to support how the impairments were measured” or to show that O’Connell “was

unable to perform her essential job duties.” [Id.]. Dr. Joppich noted that O’Connell’s treatment plan, which consisted of once-per-week psychotherapy sessions, did “not reflect . . . severe enough symptoms to support restricted work,” and the lack of escalation in treatment, which had been consistent since 2018, did not support “the necessity for [O’Connell] to remain off work.” [Id. at 576]. As to O’Connell’s occupation, Dr. Joppich wrote that O’Connell worked in the “deals risk management” group at a “law firm” and listed the general responsibilities as indicated in the job description provided by Hartford and obtained from PwC. [Id. at 572]. In response to the question about whether O’Connell was “globally impaired from occupational functioning due to her psychiatric medical conditions,” including “the ability to perform in a high stress

environment and work a consistent and reliable schedule,” Dr. Joppich opined that O’Connell “does not appear to be globally impaired from her occupational functioning due to her psychiatric condition.” [Id. at 575]. She determined that while O’Connell worked in a high stress environment, “there was no objective documentation to support how impairments were measured” and it was unclear “if the impairments were judged solely on the claimant’s report or if testing was performed.” [Id.]. Because “[t]here was no measurable data provided . . . to support that the claimant was unable to perform her essential job duties” and O’Connell “was able to recite the months of the year backwards, quickly, without error, and perform simple calculations easily,” it appeared that O’Connell “was functional enough to return to work.” [Id. at 575]. Beyond this, Dr. Joppich did not discuss O’Connell’s symptoms with respect to her essential duties. Karshmer submitted a response to Dr. Joppich on July 1, 2020. [See id. at 560-65]. In that response, he explained his credentials; his diagnosis of O’Connell’s generalized anxiety

disorder; why he believed psychological testing would not be helpful in O’Connell’s case; and how O’Connell’s symptoms interfered with her ability to work. [Id.]. Karshmer acknowledged that he diagnosed O’Connell over several sessions utilizing diagnostic interviews, including the “Structured Clinical Interview for DSM Disorders.” [Id. at 562]. He stated that “[t]esting, unlike standardized assessment tools, is not generally as effective in diagnosing or treating Generalized Anxiety Disorder.” [Id.]. Karshmer claimed that “[d]iagnostic assessments were utilized to arrive at the appropriate diagnosis,” but “[p]sychological testing will not shed light on [O’Connell’s] symptoms and treatment modalities, unlike for example, an individual suspected to be suffering from a cognitive disorder.” [Id. at 563-4]. Karshmer also described O’Connell’s treatment, which included weekly sessions and telephone calls as needed. [Id. at 560]. He urged

Dr. Joppich to reconsider her conclusion that the absence of testing was grounds for dismissal of O’Connell’s claim. [Id. at 564]. Dr. Joppich responded to Karshmer’s letter on July 14, 2020, with an addendum to her report. [Id. at 548-50]. She summarized Karshmer’s letter and concluded again that “the objective documentation provided does not support a significant psychological condition that significantly inhibits [O’Connell’s] return to work, at this time.” [Id. at 550]. In particular, Dr. Joppich found that there still “was no measurable data provided to gauge the claimant’s impairments as of 12/13/2019 to the present, to support that the claimant was unable to perform her essential job duties.” [Id.]. While Karshmer told Dr. Joppich he had conducted “structural clinical interview[s] for DMS disorders,” Dr. Joppich noted there was “no clear documentation to support when” those interviewed were completed, nor was there “specific documentation of these interviews on multiple dates,” and there was no documentation of when it was determined during any such interview that the claimant experienced limitations on work activity. Dr.

Joppich concluded that there was “no quantifiable objective documentation of the claimant’s symptoms” to support that the “symptoms preclude work function.” [Id.]. She requested Karshmer describe O’Connell’s functional status as of December 12, 2019, and currently; her expected return to work date and plan; specific examination findings supporting O’Connell’s inability to return to work; whether O’Connell completed psychological evaluation tools, such as Beck’s Anxiety Inventory (“BAI”) or Generalized Anxiety Disorder 7-Item (“GAD-7”); and the rationale for why testing was not necessary in O’Connell’s case. [Id. at 552]. On July 15, 2020, Karshmer responded to Dr. Joppich by providing an undated BAI and GAD-7 results dated July 14, 2020. [Id. at 424-27]. The BAI revealed that O’Connell had moderate to severe anxiety symptoms. [Id. at 425]. On the GAD-7, O’Connell selected “nearly

every day” for all seven questions regarding how she often experienced anxiety symptoms, such as feeling nervous, worrying too much, and feeling as if something awful could occur. [Id. at 426]. O’Connell also indicated that those problems made it “extremely difficult” to “work, take care of things at home, or get along with other people.” [Id.]. In an addendum dated July 29, 2020, Dr. Joppich responded to Karshmer’s submissions, finding them insufficient to support functional limitations. [Id. at 396]. Despite the high anxiety level indicated on GAD-7 and BAI testing, Dr. Joppich explained that there was no documentation of such tools on multiple dates, which would “would be necessary to track the severity of the claimant’s symptoms, and provide quantifiable documentation of the claimant’s condition.” [Id.]. Therefore, Karshmer’s response did not change her prior conclusions. [Id.] Karshmer submitted another APS on August 13, 2020. [Id. at 365-66]. The APS stated that O’Connell had moderate to severe impairments in attention, concentration, and memory,

however, it is unclear how this was determined, i.e., through self-reporting, observation, or assessment. [Id. at 365]. He also indicated that O’Connell’s status “regressed.” [Id.]. On September 10, 2020, Dr. Joppich included another addendum to her report after considering Karshmer’s final APS, in addition to previous submissions, such as Dr. Kahn’s report, the GAD- 7, and the BAI. [Id. at 357]. The additional documentation did not alter Dr. Joppich’s prior assessment. [Id. at 359]. Because there was still “no objective evidence” to support O’Connell’s inability to work and “the documentation appears to indicate that the claimant will return to work on 09/01/2020,” Dr. Joppich determined there was no evidence warranting restrictions and limitations regarding work-related activities. [Id. at 359]. 6. Hartford’s Final Termination Letter

On October 16, 2020, Hartford issued a final termination letter to O’Connell, affirming its prior decision. [Id. 123-27]. Hartford’s letter largely summarizes Dr. Joppich’s report and addenda, noting that Dr. Joppich concluded that the employee does not appear to be globally impaired from her occupational functioning due to her psychiatric condition as of December 13, 2019 through present as there was a lack of objective documentation or rationale as to show the impairments in attention, concentration and memory were determined. Specifically, [t]here was no measurable data provided to gauge the claimant’s impairments . . . there is a lack of clear documentation to support that the employee’s symptoms are severe enough to warrant restricted work.

[Id. at 125]. Hartford’s letter also briefly recounts Karshmer’s responses to Dr. Joppich’s addenda and cursorily references the job description it received from PwC. [Id. at 124-26]. Hartford found that the “objective documentation provided does not support a significant psychological condition that significantly inhibits the employee’s return to work,” and there were “no limitations to prevent the employee from performing her . . . [o]ccupation” as defined by the Plan. [Id. at 126]. Hartford clarified that it was “not disputing” O’Connell’s complaints,

however, “disability determinations are not solely based upon a diagnosis or certification of disability. The physical testing and evidence must be able to substantiate the employee’s condition was of such a degree of severity that it caused a functional impairment. Based on the totality of the medical information in the claim file, no restrictions are supported to impact occupational functioning.” [Id.]. On April 8, 2021, O’Connell filed this action challenging Hartford’s decision to terminate LTD benefits under the Plan. [See Dkt. 1]. II. LEGAL STANDARD The standard of review for actions challenging benefit eligibility determinations largely depends on whether “the benefit plan gives the administrator or fiduciary discretionary authority to determine eligibility for benefits or to construe the terms of the plan.” Firestone Tire &

Rubber Co. v. Bruch, 489 U.S. 101, 115 (1989). The parties do not dispute that the arbitrary and capricious, or abuse of discretion, standard applies here, as the benefit plan underlying this action gives the claims administrator discretionary authority to determine eligibility for benefits and to construe the terms of the plan. See Leahy v. Raytheon Co., 315 F.3d 11, 15 (1st Cir. 2002). Such review requires the Court to “determine whether the claims administrator’s decision is arbitrary and capricious or, looked at from another angle, whether that decision is reasonable and supported by substantial evidence on the record as a whole.” McDonough v. Aetna Life Ins. Co., 783 F.3d 374, 379 (1st Cir. 2015). “Substantial evidence” is “evidence reasonably sufficient to support a conclusion,” Arruda v. Zurich Am. Ins. Co., 951 F.3d 12, 21 (1st Cir. 2020) (citation omitted), and “a conclusion can still be supported by substantial evidence if contrary evidence exists,” Ovist v. Unum Life Ins. Co., 14 F.4th 106, 117 (1st Cir. 2021). A court deciding a motion for judgment on the administrative record under the arbitrary and capricious does not draw inferences in favor of the nonmoving party, and doubts are typically resolved in favor of

the claims administrator. Liston v. Unum Corp. Officer Severance Plan, 330 F.3d 19, 24 (1st Cir. 2003). Ultimately, the Court “must uphold” the insurer’s determination unless it was “unreasonable in light of the information available” to it. Ovist, 14 F.4th at 117 (citation omitted). Under any standard of review, the claimant bears the burden of proving disability within the meaning of her plan. See, e.g., Orndof v. Paul Revere Life Ins. Co., 404 F.3d 510, 518-19 (1st Cir. 2005). III. DISCUSSION

The parties raise many arguments in support of their cross-motions for judgment on the administrative record. O’Connell argues that Hartford’s structural conflict of interest must be considered when evaluating whether Hartford’s review was arbitrary and capricious. [Dkt. 42 at 15]. Hartford counters that the so-called structural conflict of interest is not an important factor because the evidence weighs heavily in its favor. [Dkt. 38 at 29]. O’Connell also claims that Hartford did not give appropriate consideration to O’Connell’s treating physicians and did not properly explain why it disagreed with the medical evidence offered. [Dkt. 42 at 20-24]. Hartford responds that its independent medical reviewers considered all of the evidence in the record and fully explained their opinions. [Dkt. 48 at 11-15]. O’Connell further argues that Hartford abused its discretion by failing to evaluate O’Connell’s limitations against the demands of her occupation. [Dkt. 42 at 17-20]. Hartford counters that it considered all of the medical evidence submitted, but that evidence did not include proof of any functional limitation. [Dkt. 48 at 5-10]. A. Conflict of Interest O’Connell argues that Hartford’s role as both the claims administrator and the payor of

benefits under the Plan creates a conflict of interest, which affects the Court’s review of whether Hartford’s decision was an abuse of discretion. [Dkt. 42 at 15]. Hartford maintains that the structural conflict of interest is not relevant in cases, such as this one, where the evidence is not closely balanced and where the defendant takes steps to reduce the impact of any bias. [Dkt. 38 at 29]. The Supreme Court has acknowledged that when “a plan administrator both evaluates claims for benefits and pays benefits claims,” it creates a structural conflict of interest. See Metro. Life Ins. Co. v. Glenn, 554 U.S. 105, 112 (2008). A conflict of interest, therefore, exists when a claims administrator or fiduciary both reviews the claim for benefits and pays the benefits should the claim be approved. See id.; see also Denmark v. Liberty Life Assur. Co., 566

F.3d 1, 7 (1st Cir. 2009). However, the mere existence of that conflict does not change the standard of review, which continues to be deferential, see Wright v. R.R. Donnelley & Sons Grp. Benefits Plan, 402 F.3d 67, 75 (1st Cir. 2005), and instead is “one factor among many that a reviewing judge must take into account,” Glenn, 554 U.S. at 116. A conflict is “more important . . . where circumstances suggest a higher likelihood that it affected the benefits decision,” but “the conflict prove[s] less important (perhaps to the vanishing point) where the administrator has taken active steps to reduce potential bias and to promote accuracy.” Petrone v. Long Term Disability Income Plan, 935 F. Supp. 2d 278, 288 (D. Mass. 2013) (alterations in original) (quoting Glenn, 554 U.S. at 117). Courts may give no weight at all to such conflict if the plan administrator has implemented appropriate procedures to help remove the potential for bias and promote accuracy. Tebo v. Sedgwick Claims Mgmt. Servs. Inc., 848 F. Supp. 2d 39, 51-52 (D. Mass. 2012). Courts may consider a conflict of interest as a “tie-breaker” if all relevant factors weigh equally. Id.

The record shows that Hartford retained third-party reviewers at every stage to evaluate O’Connell’s claim, including the initial application and the appeal, and O’Connell “has provided no evidence that [Hartford] retained its reviewers specifically because they have a record of denying claims.” Cusson v. Liberty Life Assur. Co., 592 F.3d 215, 227 (1st Cir. 2010), abrogated on other grounds by Montanile v. Bd. of Trs. of the Nat’l Elevator Indus. Health Ben. Plan, 577 U.S. 136, 141 & n.2 (2016). Hartford also referred O’Connell’s appeal to its Appeals Unit, a division separate from that which made the initial decision to terminate O’Connell’s LTD benefits. Courts have recognized that such measures are “active steps” that reduce the risk of bias and promote accuracy and are sufficient to limit the weight accorded to the conflict. See Petrone, 935 F. Supp. 2d at 288; Tebo, 848 F. Supp. 2d at 52; see also Estrella v. Hartford Life &

Accident Ins. Co., No. 09-11824-RWZ, 2011 WL 4007679, at *4 (D. Mass. Sep. 6, 2011); Salvador v. Liberty Life Assur. Co., C.A. No. 12-CV-30196-MAP, 2014 WL3749206, at *8 (D. Mass. July 28, 2014). Such is the case here, and the Court “will therefore apply the abuse-of- discretion standard of review, giving weight to the structural conflict if all other relevant considerations weigh evenly, and overturn defendant’s decision only if ‘it is unreasonable, or arbitrary and capricious.’” Tebo, 848 F. Supp. 2d at 54 (citation omitted).4

4 O’Connell also argues that Hartford’s reminders that O’Connell should apply for social security disability income (“SSDI”) created a financial conflict of interest, because the Plan required O’Connell to repay some of the LTD benefits upon approval of her SSDI application. [Dkt. 42 at 24-25]. This argument is unpersuasive. Employers “have large leeway to design disability and other welfare plans as they see fit,” and claims administrators have a “duty . . . to see that the plan is maintained pursuant to [that] written instrument.” Heimeshoff v. Hartford Life & B. Hartford’s Consideration of the Evidence The fundamental issue in this case is whether Hartford’s decision was supported by substantial evidence. O’Connell argues that Hartford did not give appropriate consideration to O’Connell’s treating physicians and failed to explain why it disagreed with the medical evidence

she offered. [Dkt. 42 at 20-24, 26-30]. Hartford disagrees. [See Dkt. 48 at 11-15]. 1. Weight Given to Physicians and Reviewers and Hartford’s Explanation for Disregarding Certain Evidence

O’Connell stresses that Hartford failed to give proper consideration to the opinions of her treating provider and instead relied exclusively on its own consultants. According to O’Connell, Hartford was required to give greater weight to Karshmer’s evaluations and Dr. Kahn’s report given the nature of O’Connell’s illness. [Dkt 42 at 29]. While a claims administrator “may not arbitrarily refuse to credit a claimant’s reliable evidence, including the opinions of a treating physician,” Black & Decker Disability Plan v. Nord, 538 U.S. 822, 834 (2003), the “opinion of the claimant’s treating physician . . . is not entitled to special deference,” Orndorf, 404 F.3d at 517 (citation omitted). Still, in cases where the disability assessment depends on largely subjective, self-reported symptoms, “those who have had in-person exposure—whether treating physician or not—have access to information unavailable to non-examining doctors . . . [and] the impressions of examining doctors sensibly may be given more weight than those who looked only at paper records.” Gross v. Sun Life Assu. Co., 880 F.3d 1, 14 (1st Cir. 2018) (emphasis added). In other words, a claims administrator can give a treating provider’s opinion greater

Acc. Ins. Co., 571 U.S. 99, 108 (2013) (citation and internal quotations omitted). The Plan here provides that claimants “must apply for Social Security disability benefits when the length of [their] [d]isability meets the minimum duration required for such benefits.” [Dkt. 44 at 293]. The Plan further defines social security disability benefits as “other income benefits” that offset LTD benefits. [Dkt. 44 at 289, 298]. Hartford’s SSDI reminders to O’Connell and her agreement to offset any social security award in calculating LTD benefits are express terms of the Plan, and Hartford merely fulfilled its “duty . . . to see that the plan is ‘maintained pursuant to [that] written instrument.’” Heimeshoff, 571 U.S. at 108. weight but is not required to do so. To the extent that O’Connell claims that Hartford must credit the opinion of her treating therapist or physician, the Court disagrees. See Morales-Alejandro v. Med. Card Sys., Inc., 486 F.3d 693, 700 (1st Cir.2007) (“[A] plan administrator is not obligated to accept or even to give particular weight to the opinion of a claimant’s treating physician.”).

Hartford was certainly entitled to disagree with Karshmer and Dr. Kahn’s conclusions. Karshmer’s evidence was contradictory at times. For example, Karshmer said O’Connell “generally is okay cognitively though may have some concentration difficulty” [Dkt. 44 at 961], and one of Karshmer’s reports noted an increase in anxiety symptoms but also included a plan to return to work [id. at 917-18]. It would not be unreasonable to conclude that such notes undermine Karshmer’s position that O’Connell’s anxiety was disabling to the point that she could not practice law in the general workplace. The fact that O’Connell’s treatment plan— once-per-week psychotherapy sessions—did not escalate when her anxiety began to interfere with her ability to work could further support that conclusion. Moreover, Hartford could not evaluate whether Karshmer’s conclusions regarding O’Connell’s limitations corresponded to her

actual treatment, because Karshmer refused to provide his treatment notes. Cf. Price v. Disability, RMS, No. 06-CV-10251-GAO, 2008 WL 763255, at *18 (D. Mass. Mar. 21, 2008) (“While the record includes statements from Dr. Tomb and Mr. Kennedy describing Dr. Price’s symptoms in general terms and offering opinions that Dr. Price was not able to work during the period at issue, what is lacking is record support for those opinions. The actual treatment records that are in the record are at best inconsistent in significant ways with the opinions offered during the claim evaluation process, and in some instances the treatment records—especially those of Dr. Clemente and Dr. Tomb—seem rather to contradict those opinions.”). Dr. Kahn’s report was also internally inconsistent at times. Dr. Kahn noted that O’Connell was “without evidence of disorder of thought form or content,” was “alert and fully oriented,” and had a “good” “fund of knowledge.” [Dkt. 44 at 377]. Dr. Kahn noted some “low-level anxiety which [did] not seem to interfere with [O’Connell’s] activities of daily living” and wrote that O’Connell was “exploring options for returning to work as an attorney” before concluding that “an overwhelming work environment,” such as “her prior job,” would result in the return of symptoms and limitations.5

[Id. at 376, 378]. However, in its termination letters, Hartford merely summarized the various medical reports, focusing primarily on its independent medical reviewers, and did not explain how inconsistencies in the evidence provided by O’Connell affected its decision to terminate O’Connell’s LTD benefits. And neither Dr. Panzer nor Dr. Joppich, Hartford’s independent medical reviewers, engaged meaningfully with those contradictions in their statements, instead emphasizing the lack of “objective” evidence supporting Karshmer and Dr. Kahn’s conclusions. Under ERISA, the denial or termination of LTD benefits must: (1) provide adequate notice in writing to any participant or beneficiary whose claim for benefits under the plan has been denied, setting forth the specific reasons for such denial, written in a manner calculated to be understood by the participant, and (2) afford a reasonable opportunity to any participant whose claim for benefits has been denied for a full and fair review by the appropriate named fiduciary of the decision denying the claim.

29 U.S.C. § 1133. In other words, Hartford was required to explain why it dismissed Karshmer and Dr. Khan’s opinions and credited the reports of the independent medical reviewers. See Petrone, 935 F. Supp. 2d at 294 (“Absent some compelling explanation for this stark inconsistency with all of the other evidence of record, it is unreasonable to use Dr. Saris’s single report to discredit all of the other consistent evidence in support of Ms. Petrone’s disability

5 Dr. Kahn appears to have evaluated O’Connell’s limitations against her specific job “for a specific employer or at a specific location,” and not “as it is recognized in the general workplace,” which is the standard articulated in the Plan. [See id. at 299]. claim.”). The failure to do so can be an abuse of discretion requiring remand of the claim to Hartford even if the evidence itself may support the claims administrator’s decision. Hartford’s termination letters rely on Dr. Panzer and Dr. Joppich’s reports, which dismissed Karshmer’s submissions because they lacked “objective” evidence. The Court focuses

on the final termination letter, which primarily summarizes Dr. Joppich’s report and addenda and Karshmer’s responses thereto. [Dkt. 44 at 125-26; see id. at 395-96, 357-59, 548-50]. Hartford’s final termination letter provides little explanation for why Hartford reached the decision it did, i.e., why it credited Dr. Joppich’s report above all others. For example, the letter fails to meaningfully address inconsistencies in the evidence, O’Connell’s high anxiety test results, Karshmer’s explanation of how GAD requires different treatment and diagnosis, and Dr. Kahn’s conclusion that the “limitations to work function have been substantiated,” instead merely recapping Dr. Joppich’s findings. [Id. at 378]; see Doe v. Unum Life Ins. Co., 35 F. Supp. 3d 182, 192 (D. Mass. 2014) (noting that cursorily referencing or ignoring evidence in favor of finding a functional disability while “engaging with only evidence which supports [the

plan administrator’s] conclusion is not meaningful review” (citation and internal quotation marks omitted)); Ampe v. Prudential Ins. Co. of Am., No. 17-CV-11119-RGS, 2018 WL 5045184, at *5 (D. Mass. Oct. 17, 2018) (“What concerns the court in this case is the appearance that Prudential gave conclusive weight to Dr. Fiano’s opinions based on her file review without giving any substantive consideration to the records and opinions of Ampe’s treating physician . . . and the doctors who had examined Ampe to evaluate his neuropsychological symptoms . . . .”). If Hartford found Karshmer’s opinions unreliable or inconsistent for any reason, it needed to articulate that conclusion to O’Connell. All Hartford states is that it did “not disput[e] the employee’s complaints,” but the decision to terminate O’Connell’s benefits was “appropriate” because there were “no limitations to prevent the employee from performing” her occupation as defined by the Plan. [Dkt. 44 at 126]. In support of this statement, Hartford writes only that the “physical testing and evidence must be able to substantiate the employee’s condition was of such a degree of severity that it caused a functional impairment.” [Id.]. This

“conclusory basis for rejecting [Karshmer’s] opinions does not satisfy [Hartford’s] obligation under 29 U.S.C. § 1133 to give ‘specific reasons’ for its denial.” Cowern v. Prudential Ins. Co., 130 F. Supp. 3d 443, 467 (D. Mass. 2015). Moreover, if Dr. Joppich’s review of the evidence and explanation of her conclusion was not explained or well-reasoned, Hartford’s decision-making—given its heavy reliance on Dr. Joppich’s report—may be found to be arbitrary and capricious. Upon a close examination, it appears that Dr. Joppich’s engagement with Karshmer’s submissions was somewhat superficial. For example, in response to Dr. Joppich’s conclusion regarding the lack of “objective documentation to support how the impairments were measured,” and whether they “were judged solely on the claimant’s report or if testing was performed” [Dkt. 44 at 575], Karshmer provided

a detailed explanation of his diagnosis of O’Connell [id. at 560-65]. He said that he “diagnosed Ms. O’Connell over several sessions utilizing diagnostic interviews, including the Structured Clinical Interview for DSM Disorders.” [Id. at 562]. He also explained how testing is “generally not as effective in diagnosing and treating GAD.” [Id.]. Dr. Joppich dismissed Karshmer’s response because there was “no clear documentation to support when these ‘structural clinical interview[s] for DMS disorders’ were completed” and emphasized the lack of “objective documentation of claimant’s symptoms,” noting the absence of “specific psychological evaluation tools, such as Beck’s Anxiety Inventory or GAD-7” tests to “provide quantifiable consistent documentation of the severity of the claimant’s symptoms.” [Id. 44 at 550]. When Karshmer then provided the results of BAI and GAD-7 tests, per Dr. Joppich’s comment, Dr. Joppich again affirmed her finding of no limitations because there was no

documentation of such testing “on multiple dates,” despite acknowledging the high anxiety levels these tests revealed. [Id. at 396.]. Dr. Joppich also dismissed Karshmer’s concerns that “early testing would [have] cause[d] the claimant unnecessary anxiety and negatively impact[ed] the developing treatment relationship,” stating that there was “no documentation to support why testing could not be completed after the first few visits, and after a claimant/provider relationship ha[d] been established.” [Id.]. Such a dismissal is short-sighted; Karshmer clearly explained why, in his professional opinion as O’Connell’s treating therapist, testing could not be completed early on. See Doe, 35 F. Supp. 3d at 192 (explaining that the claims administrator “did not fully and reasonably consider” the evidence of the claimant’s disability when it “paid little or no attention to [the claimant’s] well-documented history in their reports” and its reviewer “offered

one conclusory paragraph to each review question”). While the Court may not “second guess the reviewing physician’s judgments about how much weight should be afford to particular pieces of medical evidence,” it can examine and find fault with the reviewer’s decision “to either cursorily reference that evidence or ignore it altogether.” Id. Dr. Joppich’s report and addenda—which form the basis of Hartford’s final termination of O’Connell’s benefits—did not provide a compelling explanation for why Karshmer’s conclusions carried no weight or were inconsistent with other evidence in O’Connell’s record. 2. Requests for Objective Evidence Hartford’s independent medical reviewers repeatedly emphasized the lack of objective evidence as justification for the termination for O’Connell’s LTD benefits, as did Hartford’s final termination letter. [Dkt. 44 at 126, 359, 550, 575]. While it is “impermissible to require

objective evidence to support claims based on medical conditions that do not lend themselves to objective verification,” Desrosiers v. Hartford Life & Accident Ins. Co., 515 F.3d 87, 93 (1st Cir. 2008), an “administrator may require objective evidence that the illness rendered the claimant unable to work,” Tebo, 848 F. Supp. 2d at 61 (citing Boardman v. Prudential Ins. Co., 337 F.3d 9, 16 n.5 (1st Cir. 2003)). The reasonableness of a request for “objective” medical evidence can be particularly difficult to evaluate in cases such as this one, where the disability, by its very nature, relies largely on the reports of the claimant for diagnosis and evaluation. While the demand for objective evidence alone may not render a claims administrator’s decision unreasonable, the administrator’s explanation for that requirement and its consideration of the provided objective evidence may.

As an initial matter, this Court disagrees that there is no objective evidence in support of O’Connell’s claim. Karshmer’s APS and BFA forms indicated observed—not merely self- reported—symptoms of GAD, and the BAI and GAD-7 tests submitted by Karshmer also serve as objective evidence. Hartford may not ignore Karshmer’s opinions simply because they recount O’Connell’s self-reported symptoms. Love v. Nat’l City Corp. Welfare Benefits Plan, 574 F.3d 392, 397-98 (7th Cir. 2009) (“[P]lan administrators . . . may not simply ignore . . . medical conclusions or dismiss . . . conclusions without explanation.”). Beyond a cursory reference to Dr. Joppich’s comment that there was no evidence of BAI and GAD-7 tests on “multiple dates,” Hartford failed to explain why it disregarded those reports as “objective evidence.” [See Dkt. 44 at 126]. Moreover, while an insurer may request objective evidence of an individual’s functional limitations, the First Circuit has recognized that “laboratory tests or similar diagnostic

procedures will not always be necessary to substantiate a claim of disability, as certain disabling conditions are not susceptible to such objective evaluations.” Brigham v. Sun Life, 317 F.3d 72, 84 (1st Cir. 2003); see Desrosiers, 515 F.3d at 93 (noting that it may be “impermissible to require objective evidence to support claims based on medical conditions that do not lend themselves to objective verification”). Karshmer, O’Connell’s treating therapist of over a decade, acknowledged the difficulty associated with quantifying anxiety, the nature of O’Connell’s mental health illness, and how her illness did not lend itself to the type of testing Hartford requested. [See Dkt. 44 at 560-65]. He also repeatedly explained that O’Connell opted for treatment that included therapy, not medication. [See id. at 561-62]. However, Hartford failed to engage with Karshmer’s conclusions, observations, and test results in its termination letter.

Even if Hartford were to ultimately find that O’Connell’s symptoms were inconsistent with or unsupported by other evidence in the record, Hartford had to consider O’Connell’s subjective complaints about the impact of her anxiety on her ability to do her job, including her impaired concentration, memory, and attention, regardless of whether there was a dearth of “objective” medical evidence. See Colby v. UnumProvident, 328 F. Supp. 2d 186, 191 (D. Mass. 2004) (“Unum asserts that it does not need to consider Dr. Selhub’s opinions concerning Colby’s restrictions and limitations because they were not supported by objective medical evidence. This court disagrees. Unum must consider those restrictions and limitations even if they were not supported by objective medical evidence.”); cf. Kramerer v. Unum Life Ins. Co., 334 F. Supp. 3d 411, 424 (D. Mass. 2018) (“[W]hile subjective reports of pain are difficult to prove, they should be accorded some weight on the ledger.”). It did not do so, and such failure is procedurally unreasonable. 3. Dr. Kahn and Raia’s Reports

Hartford engaged with neither Dr. Khan’s findings nor Raia’s vocational report in its final termination letter, both of which supported O’Connell’s claim that her anxiety resulted in functional limitations. Hartford argues that it does not have a “discrete burden to specifically discuss every piece of evidence in the record.” [Dkt 48 at 16]. While this is true, see, e.g., Orndof, 783 F.3d at 927 (noting that “the denial letter need not detail every bit of information in the record”), Hartford’s final determination letter needed to provide O’Connell with “(i) a discussion of the decision, including an explanation of the basis for disagreeing with or not the following: (A) [t]he views presented by the claimant to the plan of health care professionals treating the claimant and vocational professionals who evaluated the claimant.” 29 C.F.R. § 2560.503-1(j)(6). Such a discussion is largely missing from the final termination letter

Hartford issued to O’Connell. The final termination letter focuses on Dr. Joppich’s report. While Dr. Joppich does address some of the medical evidence, she only briefly engages with Dr. Kahn’s report6 and does not address Raia’s vocational report at all, except for listing it as part of the evidence she reviewed. In its motion papers, Hartford argues that Dr. Kahn’s opinions regarding O’Connell’s

6 While Dr. Joppich summarizes Dr. Kahn’s report, she does not explain how that report fails to support functional limitations, despite observing that Dr. Kahn “noted that the claimant strains to maintain focus and to concentrate without going blank . . . [and] that the claimant’s constant worrying and self-questioning[] impaired her ability to function effectively in a work environment,” other than to state that O’Connell was “able to recite the months of the year backwards, quickly, without error, and perform simple calculations easily.” [Dkt. 44 at 573-75]. This brief recitation of Dr. Kahn’s findings is not “meaningful review.” Doe v. Unum Life Ins. Co., 35 F. Supp. 3d 182, 192 (D. Mass. 2014). functional impairments were inconsistent with his mental status examination. [Dkt. 48 at 7, 9- 10]. Hartford also claims that it did not discuss Raia’s report in the final termination letter because it is not required to explain why it discredits every piece of evidence, and because it found the report was unworthy of much weight. [Id. at 16 & n.9]. However, Hartford articulated

none of this in final termination letter—Hartford fails to mention Dr. Kahn and Raia’s reports entirely (beyond listing them as documents included in its review)—and a “plan administrator may not simply ignore medical or vocational evidence which contradicts its conclusion.” Doe, 35 F. Supp. 3d at 190. Hartford’s failure to engage with Dr. Khan and Raia’s reports supports the finding that Hartford’s termination of O’Connell’s benefits was, at least, procedurally unreasonable.7 See Petrone, 935 F. Supp. 2d at 293 (“[T]he administrator cannot simply ignore contrary evidence, or engage with only that evidence which supports his conclusion.”). C. Hartford’s Evaluation of O’Connell’s Occupation O’Connell argues that Hartford failed to conduct a vocational review of her claim, in violation of ERISA and its own policy terms, and failed to evaluate whether O’Connell’s mental

health limitations prevented her from performing the duties of her occupation. “[M]edical evidence is only part of the equation” when evaluating a claim for LTD benefits, and a benefits determination is not “‘reasoned’ where the [claims] administrator sidesteps the central inquiry . . . [of] whether the claimant [is] . . . able to perform the material duties of [her] own occupation.” McDonough, 783 F.3d at 380-81 (citation omitted). The administrator must also take “the obligatory step of assessing whether and to what extent (if at all) the appellant’s

7 O’Connell also argues that Hartford’s lack of discussion of her childhood trauma as the origin of her anxiety is an abuse of discretion. [Dkt. 53 at 8]. The Court disagrees. The exact nature of O’Connell’s childhood trauma and any anxiety associated with that trauma is not in the record. Similarly, O’Connell’s argument regarding Hartford’s failure to consider her risk of relapse is without merit. [Dkt. 42 at 25]. The case she relies on, Colby v. Union Sec. Ins. Co. & Mgmt. Co., 705 F.3d 58 (1st Cir. 2013), is about substance abuse, not anxiety, and is distinguishable from the present case. impairments compromised [her] ability to carry out the material duties of [her] own occupation as normally performed in the [general workplace].” Id. at 380. Dr. Panzer and Hartford’s initial termination letter make no mention of O’Connell’s occupational duties. [See Dkt. 44 at 154-56]. Hartford’s final termination makes a cursory reference to obtaining O’Connell’s job description

to verify the essential duties of her occupation but goes no further. [Id. at 123-27]. Dr. Joppich, whose report Hartford discusses in detail in its final termination letter, lists O’Connell’s general job responsibilities8 and describes a “high stress and high-pressure environment,” but otherwise states only that “the claimant does not appear to be globally impaired from her occupational functioning due to her psychiatric condition,” a conclusion she reached in large part because “there was no objective documentation to support how impairments were measured.” [Id. at 572-73, 575]. Such brief references do not “articulate the contours of the [‘occupation’] standard” articulated in the Plan and “apply that standard in a meaningful way.”9 McDonough, 783 F.3d at 382. This failure “renders [Hartford’s] benefits termination decision arbitrary and capricious.” Id.; see Elliott v. Metro. Life Ins. Co., 473 F.3d 613, 619-20 (6th Cir. 2006).

Hartford argues that without any functional limitations, there is no vocational analysis to perform. [See Dkt. 48 at 5-10, 17]. Such a position here is short-sighted. A claims administrator “may not dismiss evidence merely because it is subjective, but must meaningfully

8 Dr. Joppich initially wrote that O’Connell worked as an attorney at a law firm. [Dkt. 44 at 575]. While O’Connell makes much of this misstatement in her briefing [Dkt. 42 at 11, 18], it does not affect the Court’s findings. Dr. Joppich corrected the mistake shortly after O’Connell raised the error during the appeal in response to Dr. Joppich’s report [Dkt. 44 at 557, 549], and a discussion of O’Connell’s occupation is largely absent from Dr. Joppich’s analysis anyway. [See id. at 572-73].

9 The Plan requires an evaluation of a claimant’s “occupation” as “it is recognized in the general workplace,” not as “the specific job [the employee] is performing for a specific employer at a specific location.” [Dkt. 44 at 299]. It is entirely possible that O’Connell, even with her claimed limitations, is able to perform the “essential duties” of her occupation as an attorney, and more specifically, as in-house counsel, as it is recognized in the “general workplace,” outside the confines of PwC, where she experienced issues with her colleagues and leadership that placed added stress on her and, by her own admission, “increased [her] stress levels,” “caused [her] anxiety to resurface,” and led to feelings of “incompeten[ce]” and “burn[] out.” [Id. at 684-85]. As O’Connell herself states, she is “intellectually . . . more than capable of doing [her] job, but it’s the emotional element that [she is] still working on.” [Id. at 686]. address why reported symptoms [are] either false or exaggerated or do not impede a claimant’s ability to work.” Ampe, 2018 WL 5045184, at *6 (citing Miles v. Principal Life Ins. Co., 720 F.3d 472, 487 (2d Cir. 2013)). This is obviously a fine balance. But other courts have found that a claims administrator’s failure to discuss the claimant’s symptoms—even if they do not

necessarily lead to functional limitations—in light of the claimant’s occupational requirements renders that decision unreasonable. See Winters v. Liberty Life Assurance Co., No. 20-CV- 11937-MLW, 2022 WL 6170588, at *17, 19 (D. Mass. Oct. 7, 2022) (noting that the medical review cited in the administrator’s decision concluded that the claimant’s “anxiety symptoms are not impairing” but “did not take the next step of comparing [the claimant’s] cognitive abilities to the job duties identified by the vocational expert”); Ampe, 2018 WL 5045184, at *5. Those courts have deemed it “insufficient” for the administrator to “rely on blanket statements from the doctors that restrictions and limitations are not supported by the record without addressing the duties [the defendant] was called upon to perform.” Winters, 2022 WL 6170588, at *19; see also Ampe, 2018 WL 5045184, at *4-5 (finding that the claims administrator’s denial, which stated

that the “information available does not provide any valid or reliable evidence supportive of any level of cognitive dysfunction, nor any level of impairment from a physical or psychological perspective,” was unreasonable). Indeed, one of those courts dismissed an argument similar to Hartford’s argument here as “circular logic.” See Ampe, 2018 WL 5045184, at *6 n.9. Neither the independent consultants’ reports nor Hartford’s communications with O’Connell sufficiently addressed O’Connell’s claim with respect to her occupation, as an attorney, in the general workplace. Hartford’s failure to measure O’Connell’s illness against the demands of her occupation renders its termination of her LTD benefits unreasonable. IV. CONCLUSION Where a claims administrator abuses its discretion, the Court may either award benefits to the claimant or remand the decision to the administrator. See Buffonge v. Prudential Ins. Co., 426 F.3d 20, 31 (1st Cir. 2005) (noting that courts have “considerable discretion” in fashioning

an appropriate remedy). While the Court finds that Hartford abused its discretion by failing to adequately explain its decision to terminate O’Connell’s benefits based upon the medical record, it does not find that the determination was necessarily wrong. It is O’Connell’s burden to prove that she is disabled within the meaning of the long-term disability plan. See Orndof, 404 F.3d at 518-19. O’Connell has not met this burden. Where, as here, the record does not compel the conclusion that O’Connell is “clearly entitled” to benefits, remand is the appropriate remedy. Buffonage, 426 F.3d at 31. On remand, Hartford must, at minimum, provide a thorough explanation of its rationale for terminating benefits and assess O’Connell’s limitations against the demands of her occupation as an attorney. For the foregoing reasons, Plaintiff’s motion for judgment on the administrative record [Dkt. 40] is DENIED, Defendant’s motion for judgment

on the administrative record [Dkt. 37] is likewise DENIED, and the case is REMANDED to Hartford for further proceedings consistent with this Order. SO ORDERED. Dated: March 24, 2023 /s/ Angel Kelley Hon. Angel Kelley United States District Judge