UNPUBLISHED
UNITED STATES COURT OF APPEALS FOR THE FOURTH CIRCUIT
No. 21-2292
STEPHEN R., Plaintiff – Appellant,
v.
MARTIN J. O’MALLEY, Commissioner of Social Security Administration, Defendant – Appellee.
Appeal from the United States District Court for the District of South Carolina at Aiken. Bruce H. Hendricks, District Judge. (1:19-cv-03405-BHH)
Submitted: November 9, 2023 Decided: July 23, 2024
Before WILKINSON and BENJAMIN, Circuit Judges, and MOTZ, Senior Circuit Judge.
Vacated and remanded by unpublished per curiam opinion.
ON BRIEF: Dana W. Duncan, DUNCAN DISABILITY LAW, S.C., Nekoosa, Wisconsin, for Appellant. Brian O’Donnell, Regional Chief Counsel, Charles Kawas, Supervisory Attorney, David E. Somers, III, Special Assistant United States Attorney, Office of the General Counsel, SOCIAL SECURITY ADMINISTRATION, Philadelphia, Pennsylvania; Corey F. Ellis, United States Attorney, OFFICE OF THE UNITED STATES ATTORNEY, Columbia, South Carolina, for Appellee.
Unpublished opinions are not binding precedent in this circuit.
2
PER CURIAM:
Stephen Root appeals the district court’s order affirming the Social Security Administration’s denial of his application for disability benefits. Root served in the Marine Corps from 2000 to 2007, and was involved in combat in Iraq. Since his return, he has been diagnosed with anxiety, post-traumatic stress disorder (“PTSD”), affective disorder, personality disorder, and intermittent explosive disorder. Root stopped working in 2016 and applied for benefits in 2017, but an Administrative Law Judge (“ALJ”) denied his application after finding that the record did not substantiate the severity of his symptoms, and that his condition had significantly improved. For the following reasons, we vacate and remand for a new hearing.
I.
The Social Security Administration follows a five-step process to assess whether a claimant is disabled under the Social Security Act. 20 C.F.R. § 404.1520(a)(4). The first three steps address “(1) whether the claimant is working; (2) if not, whether [he] has a ‘severe impairment’; and (3) if [he] does, whether the impairment ‘meets or equals a listed impairment.’” Patterson v. Comm’r of Soc. Sec. Admin., 846 F.3d 656, 659 (4th Cir. 2017) (quoting 20 C.F.R. § 404.1520). If the claimant succeeds at the third step, the inquiry ends, and the ALJ must find that the claimant is disabled. Id.
“[I]f the claimant fails at step three, the ALJ ‘must determine the claimant’s residual functional capacity (RFC), which is the most the claimant can still do despite physical and mental limitations that affect [his] ability to work.’” Shinaberry v. Saul, 952 F.3d 113, 119
3
(4th Cir. 2020) (quoting Monroe v. Colvin, 826 F.3d 176, 179 (4th Cir. 2016)). The RFC guides the remainder of the inquiry. At step four, the ALJ evaluates whether the claimant’s impairments prevent him from performing “past relevant work.” Patterson, 846 F.3d at 659 (cleaned up). “If so, the ALJ proceeds to step five, where the burden shifts to the [agency] to prove, by a preponderance of the evidence, that the claimant can perform other work that exists in significant numbers in the national economy, considering the claimant’s RFC, age, education, and work experience.” Shinaberry, 952 F.3d at 119 (cleaned up).
Applying the five-step process, the ALJ first determined that Root was unemployed, that he has been diagnosed with several severe impairments, and that those impairments do not meet, or medically equal, impairments listed in 20 C.F.R. Part 404, Subpart P, Appendix 1. Accordingly, the ALJ performed a residual functional capacity assessment. Reviewing Root’s conditions, the ALJ found that the medical evidence did not support the “intensity, persistence, and limiting effects” of Root’s symptoms, and that Root had “made significant improvements following his voluntary inpatient treatment.” Based on this finding, the ALJ concluded that, as relevant:
[Root] can perform jobs where the worker is largely isolated from the general public, dealing with data and things rather than people. He can perform jobs where the work duties can be completed independently from coworkers;
however, physical isolation is not required. He can respond appropriately to reasonable and customary supervision.
Applying this RFC, the ALJ found that Root could not return to his previous positions, including work as a marksmanship instructor and trucking dispatcher, among other roles. Relying on a vocational expert’s testimony, the ALJ nevertheless concluded that there are hundreds of thousands of jobs Root could perform in the national economy.
4
II.
“In social security proceedings, a court of appeals applies the same standard of review as does the district court.” Brown v. Comm’r of Soc. Sec. Admin., 873 F.3d 251, 267 (4th Cir. 2017). That is, we “must uphold the determination when an ALJ has applied correct legal standards and the ALJ’s factual findings are supported by substantial evidence.” Id. (cleaned up). “Substantial evidence is that which a reasonable mind might accept as adequate to support a conclusion.” Pearson v. Colvin, 810 F.3d 204, 207 (4th Cir. 2015) (cleaned up). “In reviewing for substantial evidence, we do not undertake to reweigh conflicting evidence, make credibility determinations, or substitute our judgment for that of the ALJ.” Hancock v. Astrue, 667 F.3d 470, 472 (4th Cir. 2012) (cleaned up).
Root challenges several aspects of the ALJ’s decision, arguing sequentially: (1) that the ALJ failed to address conclusions by a vocational expert; (2) that the ALJ cherrypicked the record, disregarding relevant evidence that substantiated the severity of his symptoms; and (3) that the ALJ did not properly analyze the medical reports. We reject the first argument, but find the remaining two meritorious.
A.
Root first argues that the ALJ failed to explain why his RFC determination did not account for two limitations discussed with a vocational expert. As “a necessary predicate to engaging in substantial evidence review,” the record must include the basis for the ALJ’s decision, requiring “a discussion of which evidence the ALJ found credible and why, and specific application of the pertinent legal requirements to the record evidence.” Radford
5
v. Colvin, 734 F.3d 288, 295 (4th Cir. 2013). “[T]he RFC ‘assessment must include a narrative discussion describing how the evidence supports each conclusion, citing specific medical facts (e.g., laboratory findings) and nonmedical evidence (e.g., daily activities, observations).’” Monroe, 826 F.3d at 189 (quoting Mascio v. Colvin, 780 F.3d 632, 636 (4th Cir. 2015)). “To pass muster, ALJs must ‘build an accurate and logical bridge’ from the evidence to their conclusions.” Arakas v. Comm’r, Soc. Sec. Admin., 983 F.3d 83, 95 (4th Cir. 2020) (quoting Monroe, 826 F.3d at 189).
Root argues that the ALJ failed to modify his RFC analysis to address a vocational expert’s conclusion that no work exists in the national economy for an individual who cannot tolerate reasonable supervision, and who requires isolation from his coworkers.1 But the expert rendered this opinion in response to the ALJ’s questions about a hypothetical claimant — he did not conclude that these limitations accurately reflect Root’s symptoms. Absent a finding that Root cannot tolerate supervision or requires isolation from coworkers, the ALJ was not required to address how such restrictions would affect his employability.
B.
Root next argues the ALJ cherrypicked the record and ignored evidence supporting a greater disability finding. “An ALJ has the obligation to consider all relevant medical evidence and cannot simply cherrypick facts that support a finding of nondisability while
1
Root also argues that the ALJ failed to address his need for “occasional” absences from work. He argues that the “the term ‘occasional’” indicates he will be absent on as many as one third of his scheduled workdays. We are not persuaded. Drs. Jackson and Clanton opined that “[t]he claimant can attend work regularly, needing to miss no more than an occasional day of work due to mental illness.” That assessment does not imply the level of absenteeism Root suggests.
6
ignoring evidence that points to a disability finding.” Lewis v. Berryhill, 858 F.3d 858, 869 (4th Cir. 2017) (quoting Denton v. Astrue, 596 F.3d 419, 425 (7th Cir. 2010)).
The ALJ found that the record does not substantiate Root’s claims regarding the “intensity, persistence, and limiting effects” of his conditions, and that his symptoms have significantly improved since his hospitalization in February and March 2017. Root takes issue with these findings, arguing that the ALJ ignored significant evidence of his symptoms before and after his outpatient treatment. We agree.
First, the ALJ overlooked evidence of Root’s symptoms prior to his hospitalization.
The ALJ recounted that Root “functioned well for years” after his discharge from the Marines in 2007, and attributed Root’s downturn to an extramarital affair that he began sometime in late 2015. These conclusions overlook important evidence that Root struggled with severe symptoms both during his service and following his discharge. For example, Dr. Jachna reported that Root attempted suicide multiple times during his military service; that in 2010, he “held a gun to his head [and] threatened to kill his wife and himself”; and that in 2016, he nearly threw his girlfriend off a balcony during an explosive episode. And Dr. Wentworth noted that Root “has been suffering with PTSD . . . for several years,” recounted that “[h]e has made four serious, highly lethal attempts to commit suicide,” and attributed his self-destructive behavior — including the affair — to his PTSD:
It would seem on the surface that it is [the affair] that is the cause of his downward spiral. However, it is important to view this from a different, more complete perspective. Stephen has now lost his job, his children, and his standing with family, friends, and the community. It is not unusual for a combat veteran with PTSD to believe that he does not deserve to live. After Stephen attempted suicide four times and was unsuccessful, he stated he decided to stop trying to kill himself. However, another way to exact extreme
7
punishment on oneself, short of suicide, is to destroy one’s life. This is the basis of Stephen’s behavior over the last year. . . . It was a way to destroy his life, one that didn’t require his physical death.
These reports cut against the ALJ’s evaluation of Root’s symptoms, and his finding that Root “functioned well for years” following his military service. But they do not appear anywhere in the ALJ’s decision. In fact, the ALJ nowhere mentions that Root has made multiple suicide attempts, or that he has entertained homicidal thoughts.
The ALJ also omitted recent reports suggesting Root’s symptoms have returned.
See Thomas v. Berryhill, 916 F.3d 307, 312 (4th Cir. 2019) (finding reversible error where ALJ declined to discuss relevant reports by mental health specialists). In his decision, the ALJ found “that the claimant made significant improvements following his voluntary inpatient treatment, and there are very few visits for mental health treatment after his discharge . . . in March of 2017.” To support this finding, the ALJ noted that Root reported reduced symptoms in follow-up visits during May and September 2017. He also recounted that by May 2018, Root was having no hallucinations and fewer nightmares — and that Root had remarried, and was visiting state parks with his wife.
But Root’s medical history did not end in May 2018, and neither did the record before the ALJ. In November 2018 and February 2019, Root saw a significant downturn — according to medical reports, he began to experience insomnia, nightmares, increased irritability, disturbing flashbacks, and severe depression, with a PHQ-2 I9 score of 23. He also admitted that he was having generalized homicidal thoughts, although he was not likely to act on them. And he informed his doctors that his medication was no longer effectively resolving his symptoms. This medical evidence at least undermines the ALJ’s
8
conclusion that Root’s inpatient treatment was a turning point. But it does not appear anywhere in the ALJ’s decision.
In a similar vein, the ALJ’s description of Root’s testimony at his June 2019 hearing was selective. As evidence that Root’s intermittent explosive disorder has subsided, the ALJ recounted that Root attends Freemason meetings and Bible study; that he watches television all day despite his concentration issues; and that he has learned breathing techniques to manage his panic attacks. But Root also testified that his breathing techniques are not always effective; that television shows can trigger his PTSD; that he has stopped attending Bible study; and that he has been unable to tolerate crowded spaces for more than “45 minutes to an hour” without feeling overwhelmed. The ALJ brushed past these negative signs entirely, “highlighting [Root’s] good moments and bypassing the bad.” Shelley C. v. Comm’r of Soc. Sec. Admin., 61 F.4th 341, 362 (4th Cir. 2023); see, e.g., Arakas, 983 F.3d at 102 (“The ALJ selectively cited evidence concerning tasks which [Arakas] was capable of performing and improperly disregarded her qualifying statements.” (cleaned up)).
To be sure, “there is no rigid requirement that the ALJ specifically refer to every piece of evidence in his decision.” Reid v. Comm’r of Soc. Sec. Admin., 769 F.3d 861, 865 (4th Cir. 2014) (quoting Dyer v. Barnhart, 395 F.3d 1206, 1211 (11th Cir. 2005)). And, of course, we may not “reweigh conflicting evidence, make credibility determinations, or substitute our judgment for that of the ALJ.” Hancock, 667 F.3d at 472 (cleaned up). But when the ALJ’s decision suggests that he selectively examined the record and ignored countervailing evidence, Lewis, 858 F.3d at 869, or that he mischaracterized the facts to
9
buttress his conclusions, Arakas, 983 F.3d at 102, that decision is not supported by substantial evidence.
We conclude that the ALJ’s analysis of the record in this case was impermissibly selective. The ALJ concluded that Root “functioned well for years” after his discharge — without mentioning his violent tendencies and “highly lethal attempts to commit suicide” both during and after his military service. The ALJ also found that Root’s conditions improved “significantly” following his hospitalization — without acknowledging more recent reports suggesting that Root’s medication is ineffective, and his symptoms may have returned. We express no opinion as to how the ALJ must weigh this evidence on remand. But as these significant omissions affect the ALJ’s evaluation of the “intensity, persistence, and limiting effects” of Root’s symptoms, the ALJ’s assessment of Root’s RFC is not supported by substantial evidence.
C.
Third, Root argues that the ALJ did not correctly evaluate several medical reports.
We agree, and we briefly note two deficiencies in the ALJ’s analysis of the medical evidence on the record.
First, the ALJ did not correctly apply 20 C.F.R. § 404.1520c to reports by state psychologists Drs. Jackson and Clanton, which informed his evaluation of Root’s RFC. That regulation requires an ALJ to “explain how [he] considered the supportability and consistency factors for a medical source’s medical opinions.” 20 C.F.R. § 404.1520c(b)(2). Although the ALJ invoked these factors, his analysis was threadbare and lacked citations. He merely noted that “[t]hese doctors’ opinions were grounded in the evidence contained
10
in the record at the time,” and that they “are consistent with and supported by other medical evidence in the file and the entire record as a whole.” Nowhere did he reference or describe to what evidence he was referring. “Put simply, the ALJ’s lack of explanation requires remand.” Patterson, 846 F.3d at 663 (cleaned up).
Second, the ALJ provided invalid reasons for discounting the medical reports of Drs. Wentworth, McCann, and Fitzsimmons. His most pertinent reason for doing so was that “there was evidence of significant improvement following the claimant’s inpatient treatment, and the level of severity endorsed by these providers is not supported by the longitudinal evidence of record.” As discussed above, that conclusion is not supported by substantial evidence. But the ALJ also noted (1) that these reports were procured as part of Root’s earlier applications for Veterans Affairs and Postal Service disability benefits; and (2) that they failed to provide a function-by-function evaluation of Root’s abilities and limitations. Neither is a valid reason to discredit these reports. An ALJ must “consider all of the supporting evidence underlying” a decision from another government agency, 20 C.F.R. § 404.1504, and while medical opinions must provide an assessment of the claimant’s limitations and capacities, 20 C.F.R. § 404.1513(a)(2), nothing in the applicable regulations requires medical sources to conduct that assessment on a function-by-function basis. 2
We also note that the ALJ presented his RFC before making a finding as to Root’s
2
mental and physical limitations. See Thomas, 916 F.3d at 312 (“Stating a claimant’s RFC before conducting a function-by-function analysis is an error — even though, on its own, it does not necessarily require remand.”). The ALJ made several relevant findings at the third step of his analysis, where he assessed the severity of Root’s mental impairments (Continued)
11
III.
Because the ALJ analyzed the record selectively, and failed to evaluate essential medical opinions consistent with 20 C.F.R. § 404.1520c, the ALJ’s decision is not supported by substantial evidence. Pursuant to 42 U.S.C. § 405(g), we vacate and remand for a rehearing on Root’s application. We dispense with oral argument because the facts and legal contentions are adequately presented in the materials before this court and argument would not aid the decisional process.
VACATED AND REMANDED
under the “paragraph B criteria.” But as the ALJ himself acknowledged, “[t]he mental residual functional capacity assessment used at steps 4 and 5 of the sequential evaluation process requires a more detailed assessment.” On remand, the ALJ is directed to conduct such an assessment prior to any RFC finding.