Michael Pendergrass v. Commissioner Social Security

NOT PRECEDENTIAL UNITED STATES COURT OF APPEALS FOR THE THIRD CIRCUIT ____________ No. 24-2777 ____________ MICHAEL D. PENDERGRASS, Appellant v. COMMISSIONER SOCIAL SECURITY ____________ On Appeal from the United States District Court for the District of New Jersey (District Court No. 2:23-cv-02437) District Judge: Honorable Jamel K. Semper ____________ Submitted Pursuant to Third Circuit L.A.R. 34.1(a) on September 12, 2025 ____________ Before: HARDIMAN, KRAUSE, and CHUNG, Circuit Judges (Filed: March 4, 2026) ____________ OPINION * ____________ CHUNG, Circuit Judge.

Michael Pendergrass suffers from serious mental illness, which impairs his ability

* This disposition is not an opinion of the full Court and pursuant to I.O.P. 5.7 does not constitute binding precedent. to work. Pendergrass applied for a period of disability and disability insurance benefits under the Social Security Act (the “Act”), but the Commissioner of Social Security (the “Commissioner”) denied his claim, concluding that Pendergrass is not disabled within the meaning of the Act. Pendergrass appealed to the District Court, which affirmed the Commissioner’s decision. Because substantial evidence supports the Commissioner’s decision, we will affirm.

I. BACKGROUND 1 Pendergrass began to work in a “very stressful” customer service position in 2008.

Administrative Record (“A.R.”) 288–89. On September 27, 2017, he “had a[n] episode when [he] was at work” and was hospitalized for “a little over a week.” A.R. 39–40. He was diagnosed with major depressive disorder and post-traumatic stress disorder, prescribed medication, and placed in an intensive outpatient program (“IOP”). A.R. 40, 400–430. “[A]round Thanksgiving” in 2017, Pendergrass returned to work. A.R. 41.

Between November 2017 and March 2019, Pendergrass “avoided follow-up treatment.” A.R. 44, 457. Pendergrass stated that he “became unable to work because of [his] disabling condition on March 18, 2019.” A.R. 146. On March 22, 2019, Pendergrass was again placed in an IOP due to increased depression which was caused by, among other things, work stress. He was restarted on an earlier prescribed antidepressant that he had stopped taking. A.R. 457, 461. Pendergrass subsequently failed to attend many IOP sessions, and did not notify staff when he ran out of his

Because we write for the parties, we recite only the facts pertinent to our decision. antidepressant.

On May 6, 2019, Pendergrass was hospitalized again after having “another episode while [he] was at work.” A.R. 41, 317. Leading up to this hospitalization, Pendergrass was “battling with suicidal thoughts and … was really depressed and certain days … it was very difficult for [him] to even get out of bed or even wash [him]self … or take care of things in [his] house.” A.R. 43. After his four-day hospitalization, Pendergrass was again placed in an IOP. He has not returned to work since. App. 42.

On February 26, 2020, Pendergrass filed an application for a period of disability and disability insurance benefits under Title II and Part A of Title XVIII of the Social Security Act, 42 U.S.C. §§ 401–434, 1395c–1395i-6, alleging disability beginning March 18, 2019. The application was denied, and on February 22, 2021, Pendergrass requested a hearing before an Administrative Law Judge (“ALJ”). On March 7, 2022, the ALJ held a telephone hearing, where Pendergrass and a vocational expert testified. At the time of the hearing, Pendergrass attended therapy sessions “once, sometimes two, times a week,” saw a psychiatrist regularly, and took medication consistently. A.R. 42–43, 46–47.

Pendergrass still “spent a lot of time in bed,” but “[s]ometimes … ma[d]e some headway” and “sometimes … ha[d] relatively decent days” where he was “able to … just get up and move around … make it to get a shower and … actually sort of take care of [him]self.” A.R. 43–44. Otherwise, Pendergrass relied heavily on the assistance of family members. A.R. 44–45. He still occasionally struggled with suicidality, but he found that talking to his family members on the phone helped him when he was “feeling that downhearted.” A.R. 46.

On April 8, 2022, the ALJ issued a decision, finding that Pendergrass was not disabled under sections 216(i) and 223(d) of the Social Security Act. Pendergrass sought review of the ALJ’s decision by the Appeals Council, which denied review. So the ALJ’s decision became the final decision of the Commissioner. A.R. 1. Then Pendergrass appealed to the District Court, and the District Court affirmed the denial of benefits.

Pendergrass timely appealed the District Court’s order.

II. DISCUSSION 2 We review the ALJ’s decision holistically to determine whether “the ALJ considered the appropriate factors in reaching the conclusion that” Pendergrass is not disabled under the Act. Jones v. Barnhart, 364 F.3d 501, 505 (3d Cir. 2004); see also Berry v. Schweiker, 675 F.2d 464, 469 (2d Cir. 1982) (even where the ALJ “fail[s] to explain his rejection of the claimed listed impairments, we [a]re able to look to other portions of the ALJ’s decision and to clearly credible evidence in finding that his determination was supported by substantial evidence”). We do not “require the ALJ to use particular language or adhere to a particular format in conducting his analysis.”

The District Court had jurisdiction over the Commissioner’s final decision denying Pendergrass’s benefits pursuant to 42 U.S.C. § 405(g). We have jurisdiction over this appeal from the final decision of the District Court pursuant to 28 U.S.C. § 1291. We review factual findings only to determine whether the administrative record contains substantial evidence supporting the findings. See 42 U.S.C. § 405(g).

Specifically, we look to the administrative record and ask whether the factual determinations are supported by “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.” Biestek v. Berryhill, 587 U.S. 97, 103 (2019) (internal quotations omitted). Determinations whether a claimant retains the capacity to work and whether there is work in the national economy that the claimant can perform are factual findings. Id. at 102. 4 Jones, 364 F.3d at 505; see also Diaz v. Comm’r of Social. Sec., 577 F.3d 500, 504 (3d Cir. 2009). All we require is “that there is sufficient development of the record and explanation of findings to permit meaningful review.” Jones, 364 F.3d at 505 (citing Burnett v. Comm’r of Social Sec. Admin., 220 F.3d 112, 20 (3d Cir. 2000)).

The Social Security Administration (“SSA”) has promulgated a five-step sequential analysis to evaluate a disability claim: First, the Commissioner considers whether the claimant is currently engaged in substantial gainful activity. If he is not, then the Commissioner considers in the second step whether the claimant has a “severe impairment” that significantly limits his physical or mental ability to perform basic work activities. If the claimant suffers a severe impairment, the third inquiry is whether, based on the medical evidence, the impairment meets the criteria of the impairment listed in the “listing of impairments,” which result in a presumption of disability, or whether the claimant retains the capacity to work. If the impairment does not meet the criteria for a listed impairment, then the Commissioner assesses in the fourth step whether, despite the severe impairment, the claimant has the residual functional capacity [(“RFC”)] to perform his past work. If the claimant cannot perform his past work, then the final step is to determine whether there is other work in the national economy that the claimant can perform.

Sykes v. Apfel, 228 F.3d 259, 262–63 (3d Cir. 2000) (internal citations omitted); see also 20 C.F.R. § 404.1520(a)(4). Pendergrass argues that the ALJ made several errors during steps three and five and as to his mental RFC. 3 A. The ALJ’s Step Three Finding was Supported by Substantial Evidence.

Pendergrass first argues that the ALJ’s step three decision was not supported by substantial evidence because he did not consider whether Pendergrass’s impairments in

The ALJ found that Pendergrass had satisfied the first, second, and fourth sequential steps. App. 9 (citing 20 CFR 404.1571 et seq.; 20 CFR 404.1520(c)), 15. combination rendered him per se disabled and because the ALJ provided insufficient reasoning to support his conclusion.

Assessing whether the claimant’s “impairment … result[s] in a presumption of disability, or whether the claimant retains the capacity to work” requires consideration of Appendix 1 to Subpart B of Part 404. Sykes, 228 F.3d at 262–63; 20 C.F.R. § 404.1520(a)(4)(iii), (d). For the here relevant category of “Mental Disorders,” Appendix 1 sets forth in Paragraphs A, B, and C, the requirements that the claimant’s disorder must satisfy. 4 20 C.F.R. § 404, Subpt. P., App’x. 1 § 12.00(A)(2). Under Appendix 1, a presumption of disability is only triggered for claimants like Pendergrass 5 when the claimant satisfies criteria of both paragraphs A and B, or both A and C. Id. at § 12.00(A)(2). In addition, “if there is more than one impairment, the [ALJ] will consider whether the combination of impairments are equivalent to any listed impairment.” Burnett v. Comm’r of Soc. Sec. Admin., 220 F.3d 112, 119 (3d Cir. 2000).

See also 20 C.F.R. § 404.1526(a)(3).

Paragraph A “includes the medical criteria that must be present in [the claimant’s] medical evidence.” 20 C.F.R. § 404, Subpt. P., App’x. 1 § 12.00(A)(2)(a). Paragraph B “provides the functional criteria [ALJ’s] assess, in conjunction with a rating scale … to evaluate how [the claimant]’s mental disorder limits [his] functioning.” Id. at § 12.00(A)(2)(b). And Paragraph C “provides the criteria [ALJ’s] use to evaluate ‘serious and persistent mental disorders.’” Id. at § 12.00(A)(2)(c).

That is, for claimants diagnosed with 1) major depressive disorder, which fails into the category of “depressive, bipolar and related disorders,” see 20 C.F.R. § 404, Subpt.

P., App’x 1 § 12.00(B)(3)(b); and 2) post[-]traumatic stress disorder, which falls into the category of “trauma- and stressor-related disorders,” see 20 C.F.R. § 404, Subpt. P., App’x 1 § 12.00(B)(11)(b).

1. The ALJ’s Step 3, Paragraph B Analysis The ALJ began his analysis with Paragraph B, which provides four “functional criteria[ 6] … in conjunction with” a “rating scale” to assess how the claimant’s “mental disorder limits [his] functioning.” 20 C.F.R. § 404, Subpt. P., App’x. 1 § 12.00(A)(2)(b).

Under the rating scale, “marked” and “extreme” are the two highest levels of impairment, id. at § 12.00(F)(2)(a)–(e), and to satisfy Paragraph B, the claimant’s disorder “must result in extreme limitation of one, or marked limitation of two, Paragraph B areas of mental functioning.” Id. at § 12.00(F)(2).

Here, the ALJ determined that Pendergrass’s disorders “moderate[ly] limited” him in all four broad areas of mental functioning under Paragraph B. 7 App. 10–11. For each area, the ALJ noted Pendergrass’s many difficulties and explicitly considered the cumulative effect of both of Pendergrass’s disorders. Pendergrass argues that the ALJ failed to explain why he concluded Pendergrass’s limitations were moderate, despite Pendergrass’s treating psychiatrist’s assessment that the limitations were “marked.”

This argument fails as Pendergrass confines his reading of the ALJ’s opinion to a single section therein. But we may “look to other portions of the ALJ’s decision,” to determine whether the ALJ’s conclusions are supported by substantial evidence. Berry,

These include the claimant’s abilities to (1) “[u]nderstand, remember, or apply information;” (2) “[i]nteract with others;” (3) “[c]oncentrate, persist, or maintain pace;” and (4) “[a]dapt or manage oneself.” App’x 1 at § 12.00(E)(1)–(4).

A “moderate limitation” occurs when the claimant’s “functioning in th[e particular] area independently, appropriately, effectively, and on a sustained basis is fair.”

App’x 1 at § 12.00(F)(2)(c). 675 F.2d at 469. Reading the ALJ’s conclusions at step three in conjunction with the ALJ’s later analysis of Pendergrass’s impairments, it is clear that the ALJ gave ample “indication” of how he considered the “significant probative evidence.” Burnett, 220 F.3d at 121 (quotation marks omitted). Specifically, the ALJ discussed reports of Pendergrass’s providers dating back to 2019, as well as Pendergrass’s own account of the impact his disorders had on his functioning as told to the ALJ at his hearing and to his own treatment providers. Moreover, the assessment of marked limitations by Pendergrass’s treating psychiatrist appears in a check-the box form entitled to little if any weight. See Mason v. Shalala, 994 F.2d 1058, 1065 (3d Cir. 1993) (“Form reports in which a physician’s obligation is only to check a box or fill in a blank are weak evidence at best.”). And while the ALJ could have reached the same conclusion as Pendergrass’s treating psychiatrist, he was not required to do so. Zirnsak v. Colvin, 777 F.3d 607, 614 (3d Cir. 2014). Viewing the decision as a whole, “there is sufficient development of the record and explanation of findings” to support the ALJ’s decision that Pendergrass was only moderately impaired in each of the four areas of function laid out in Paragraph B.

Jones, 364 F.3d at 505.

2. The ALJ’s Paragraph C Analysis Because Pendergrass had not met the requirements of Paragraph B, the ALJ then assessed whether Pendergrass met the criteria under Paragraph C. See 20 C.F.R. § 404, Subpt. P., App’x 1 at § 12.00(A)(2) (The ALJ must find that the disorder satisfies the criteria of both paragraphs A and B, or both A and C). Paragraph C provides that a mental disorder is “‘serious and persistent’ when there is a medically documented history of the existence of the mental disorder in the listing category over a period of at least 2 years.” Id. at § 12.00(G)(2)(a). Under Paragraph C, the claimant must also satisfy the criteria in Paragraphs C1 and C2, which query the claimant’s reliance on treatment, therapy, or other support to “diminish [his] symptoms,” and whether the claimant has “achieved only marginal adjustment” and struggles “to adapt to changes … or to demands.” Id. at § 12.00(G)(2)(b)–(c). Paragraph C1 also authorizes the ALJ to “consider periods of inconsistent treatment or lack of compliance with treatment that may result from [the claimant’s] mental disorder.” Id. at § 12.00(G)(2)(b).

As with the ALJ’s Paragraph B analysis, Pendergrass argues that the ALJ provided no explanation for his finding that Pendergrass did not satisfy the Paragraph C criteria and thus, his determination was not supported by substantial evidence. Again, Pendergrass limits his reading of the ALJ’s decision to a single section, rather than reading the entirety of the decision. See Jones, 364 F.3d at 505. As noted above, the ALJ undertook a detailed assessment of the administrative record in a later portion of the opinion. App. 11. That analysis includes discussion of Pendergrass’s “periods of inconsistent treatment or lack of compliance with treatment,” 20 C.F.R. § 404, Subpt. P., App’x 1 at § 12.00(G)(2)(b), as well as the lack of evidence that “despite [Pendergrass’s] diminished symptoms and signs, [Pendergrass] ha[s] achieved only marginal adjustment,” id. at § 12.00(G)(2)(c). Considering the ALJ’s opinion holistically, his Paragraph C finding is supported by substantial evidence. App. 11.

3. The ALJ’s Assessment of the Paragraph A Criteria Having found that Paragraph B and Paragraph C were not satisfied, the ALJ declined to conduct the Paragraph A analysis, which considers “the medical criteria that must be present in [the claimant’s] medical evidence.” 20 C.F.R. § 404, Subpt. P., App’x at § 12.00(A)(2)(a). Pendergrass argues that the ALJ was required to conduct a Paragraph A analysis. We disagree and see no reason why an ALJ must conduct analysis that would have no impact on the outcome at step 3. 8 20 C.F.R. § 404, Subpt. P., App’x at § 12.00(A)(2) (for finding of disability at step 3, claimant must satisfy the requirements of either Paragraphs B and C, even where Paragraph A is satisfied).

B. The ALJ’s Step Five Analysis Is Supported by Substantial Evidence.

Under the five-step framework, if it is determined that a claimant does not have a listed impairment at step 3, then the claimant’s RFC (residual functional capacity) must be assessed to determine whether he can “perform his past work” at step 4, and if not, whether “there is other work in the national economy that the claimant can perform” at step 5. Sykes, 228 F.3d at 262–63 (internal citations omitted). An RFC assessment determines “the most [the claimant] can still do despite [his] limitations” and is “based on all the relevant evidence in [the claimant]’s case record.” 20 C.F.R. § 404.1545(a)(1).

When considering medical evidence, “the ALJ is free to accept some medical evidence and reject other evidence, provided that he provides an explanation for discrediting the rejected evidence,” Zirnsak, 777 F.3d at 614.

Further, because the Paragraph A analysis was unnecessary, the ALJ’s mistaken identification of listings 12.03 (schizophrenia spectrum and other psychotic disorders) and 12.04 (depressive, bipolar and related disorders) rather than 12.04 and 12.15 (trauma- and stressor-related disorders) was harmless. 5 U.S.C. § 706.

The ALJ found that Pendergrass’s RFC included, among other things, “occasional interaction with co-workers and supervisors able to understand; … remember and carry out simple instructions[.]” App. 11. Pendergrass contends that the ALJ’s RFC was improper because the phrase “occasional interaction with co-workers and supervisors able to understand” is “incomprehensible.” Op. Br. 31. This argument would be more persuasive had the ALJ simply set forth the RFC and failed to discuss evidence in the record. However, as we have stated above, it is clear that the ALJ considered all of the evidence and carefully weighed it as a whole. Moreover, the phrase, “able to understand” seems to be nothing more than a typographical mistake meant to be set off by commas, and precede “remember and carry out simple instructions.” Id. This reading is consistent with the ALJ’s request at the March 7, 2022 hearing for the vocational expert to assume an individual who, among other things, “would be limited to understanding, remembering, and carrying out simply instructions.” A.R. 50.

After determining Pendergrass’s RFC, the ALJ concluded at step 5 that there was “work that exists in significant numbers in the national economy” that Pendergrass could perform as “the evidence supports a finding that the claimant could perform the simple task of unskilled work in a low contact environment.” A.R. 22, 24. Pendergrass again asserts that the ALJ did not adequately explain his step five finding and again critiques only a snippet of the ALJ’s decision in isolation. But the ALJ’s determination that Pendergrass is capable of simple, low contact work was predicated on a five-page, evidence-based narrative, none of which Pendergrass discusses or successfully undermines.

Pendergrass also argues that the ALJ improperly disregarded the conclusions of Pendergrass’s treating psychiatrist, Dr. Lozovatsky. The ALJ thoroughly analyzed the objective medical evidence in the record, though, including key evidence of “[p]sychiatric signs,” a form of “objective medical evidence,” which is required to “show[]” that the claimant “ha[s] a medical impairment(s) which could reasonably be expected to produce the pain or other symptoms alleged.” 20 C.F.R. § 404.1529(a); id. § 404.1502(f); id. § 404.1502(g). Surveying years of medical records, the ALJ observed that the treatment notes of Pendergrass’s providers indicated Pendergrass’s continued depression symptoms, but otherwise “revealed no significant psychiatric findings.” 9 App. 13–14. Moreover, the ALJ explained that not only was Dr. Lozovatsky’s RFC assessment contradicted by the opinions of Pendergrass’s other providers, but it also conflicted with Dr. Lozovatsky’s “own treating notes that indicated that the claimant repeatedly reported doing ‘ok’ and the relatively benign mental status examinations.”

App. 14–15. In other words, the ALJ permissibly disregarded Dr. Lozovatsky’s opinion, as “he provide[d] an explanation for discrediting the rejected evidence.” See Zirnsak, 777 F.3d at 614.

Finally, Pendergrass argues that the ALJ’s finding that Pendergrass could not

For example, the ALJ noted that “an August 2020 mental status consultative examination failed to reveal any disabling clinical findings” because “[t]he examiner noted that the claimant was pleasant, cooperative, and responded appropriately; his hygiene and grooming were adequate; he did not report any psychotic features or suicidal/homicidal ideations; he was fully oriented; he was able to repeat three words immediately and after five minutes of interference; and he was able to complete a simple in the head arithmetic problem and serial sevens, but unable to spell the word world backwards.” App. 13. handle direct contact with the public was inconsistent with the ALJ’s additional finding that he could handle occasional interaction with his coworkers. We perceive no inconsistency. Pendergrass testified that his “very stressful” customer service job, in which he interacted constantly with the general public, was a primary cause of his two “episode[s]” requiring psychiatric hospitalization. A.R. 40–41, 255, 289. However, the record evidence also reflected that Pendergrass was still able to engage in limited communications with others.

In sum, Pendergrass has presented us with no reason to conclude substantial evidence does not support the ALJ’s RFC finding or finding at step five.

III. CONCLUSION The ALJ’s decision was supported by substantial evidence, and the District Court was therefore right to affirm. We will affirm as well.

KRAUSE, Circuit Judge, dissenting.

No doubt, substantial evidence review is a deferential standard, see Maj. Op. 4 (citing Jones v. Barnhart, 364 F.3d 501, 505 (3d Cir. 2004)), but it is not toothless. And reviewing for substantial evidence here, I respectfully dissent because the decision of the Administrative Law Judge (ALJ) in this case reflects a problem we see too often in benefits determinations: the failure to account for the unique features of mental illness in assessing medical opinions and claimants’ testimony.

With mental illness, unlike with most physical disabilities, a claimant’s “symptoms [may] wax and wane in the course of treatment,” Garrison v. Colvin, 759 F.3d 995, 1017 (9th Cir. 2014), and because a mentally ill claimant “will have better days and worse days, . . . a snapshot of any single moment says little about his overall condition,” Meuser v. Colvin, 838 F.3d 905, 912 (7th Cir. 2016) (citation modified).

In short, the mere fact “[t]hat a person who suffers from severe panic attacks, anxiety, and depression makes some improvement does not mean that the person’s impairments no longer seriously affect her ability to function in a workplace.” Holohan v. Massanari, 246 F.3d 1195, 1205 (9th Cir. 2001). And that is why, when we are reviewing an ALJ’s denial of benefits in this context for substantial evidence, we must ensure that the ALJ has given “a sufficient rationale in crediting certain evidence and discrediting other evidence,” Drumgold v. Comm’r of Soc. Sec., 144 F.4th 596, 605 (4th Cir. 2025) (citation modified), and has addressed whether and how the particular statements and evidence from medical sources on which the ALJ relies are (1) consistent with other record evidence, 20 C.F.R. §§ 404.1520c(b), (c)(2), and (2) supported by “objective medical evidence and . . . explanations,” id. § 404.1520c(c)(1). When an ALJ has not done so and, instead, cherry-picks evidence of intermittent improvement to the exclusion of, and without reconciling it with, other medical and record evidence of persistent and debilitating mental illness, we should, if not reverse, at least vacate and remand for the ALJ to conduct that analysis. See, e.g., Adorno v. Shalala, 40 F.3d 43, 48 (3d Cir. 1994); Meuser, 838 F.3d at 912-13.

That disposition is warranted here, in my view, because the ALJ erred in his treatment of four critical pieces of evidence on which he relied to deny Pendergrass benefits: (1) treatment notes from Pendergrass’s therapist at True Care Mental Health, (2) medical sources from Pendergrass’s treating psychiatrist, (3) Pendergrass’s testimony, and (4) two non-treating experts’ evaluations.

I. Therapy Notes from True Care Mental Health A comparison of the ALJ’s description of Pendergrass’s therapy treatment notes from True Care Mental Health in June 2021 and the notes themselves reveals that the ALJ overlooked or omitted critical evidence that Pendergrass’s limitations resulting from his mental illness were severe. Citing only the therapist’s comments that supported the determination that Pendergrass’s limitations were moderate, including that Pendergrass reported feeling “okay” and “presented with a flat affect,” A.R. 21, 471, the ALJ ignored the surrounding comments reflecting that Pendergrass “has had a minimal response to treatment”; “reports that his depressive-related symptoms . . . are . . . unchanged”; and “continues to exhibit symptoms of an emotional disorder that interfere with day to day functioning and requires continued treatment and is in need of medication management,” A.R. 471-72.

This was error. An ALJ must “mention and explain” medical record evidence that contravenes his determination, justifying “his conciliations and rejections.” Burnett v. Comm’r of Soc. Sec. Admin., 220 F.3d 112, 122 (3d Cir. 2000); see also Kohler v. Astrue, 546 F.3d 260, 268 (2d Cir. 2008) (holding that the ALJ committed error by “overlook[ing] or mischaracteriz[ing] relevant evidence, often to [the claimant’s] disadvantage”). And the ALJ’s failure to do so with the therapy treatment notes was particularly significant because they are the most recent medical source in the record, and the comments the ALJ failed to discuss suggest that Pendergrass’s mental illness and resulting limitations on his ability to function in a workplace are more severe and persistent than are represented in the ALJ’s opinion, which classifies Pendergrass’s limitations as “moderate,” A.R. 18, 22, and highlights “improve[ment]” he made with treatment, A.R. 21.

II. Treating Psychiatrist’s Medical Opinions The ALJ also failed to adequately explain his discounting of Dr. Lozovatsky’s Residual Functioning Capacity (RFC) assessment completed in December 2020.

Dr. Lozovatsky had been Pendergrass’s treating psychiatrist for over a year and observed in the RFC that Pendergrass had “marked” limitations that prevented him from “work[ing] on a regular and sustained basis.” A.R. 463-66. But the ALJ deemed the RFC to be “inconsistent” with other evidence and therefore “not persuasive.” A.R.

22-23. That summary rejection of the RFC does not withstand substantial evidence review for two reasons.

First, it does not address the supportability of the RFC, 20 C.F.R. § 404.1520c(c)(1), because it fails to engage with the supporting evidence Dr. Lozovatsky gave in the document, including his notes detailing Pendergrass’s condition when working: Despite following his prescribed treatment . . . his ability to concentrate was significantly impaired by his illness, which caused him to make frequent mistakes in his work . . . [he experienced] feelings of worthlessness, breakdowns and thoughts of hurting himself. The result was that [Pendergrass] could only stay on task about 50% and often would have to leave work after many of these episodes.

A.R. 466; cf. Drumgold, 144 F.4th at 606 (noting that claimant’s mental-health counselor gave no explanation “from which her logic could even be inferred” for why she characterized the claimant’s limitations as marked).

Second, the ALJ’s explanation reveals a fundamental misconception about mental illness. Namely, the ALJ presumed that Dr. Lozovatsky’s holistic assessment in the RFC that Pendergrass’s mental illness imposed “marked” limitations, A.R. 464-66, was out-of-step with the doctor’s description of Pendergrass’s affect after certain appointments—including that on certain days, Pendergrass appeared “well,” A.R. 385, or was “fairly groomed,” A.R. 373. But when assessing someone like Pendergrass, who suffers from severe mental illness, an ALJ cannot rest his decision on a “snapshot of any single moment,” because mentally ill patients “will have better days and worse days.”

Meuser, 838 F.3d at 912 (citation modified); see also Ryan v. Comm’r of

Soc. Sec., 528 F.3d 1194, 1200-01 (9th Cir. 2008) (concluding that remarks that a claimant’s anxiety and depression were “improving” did not “undermine” her mental health diagnoses or the doctor’s report finding her condition was severe). Accounting for that reality, Dr. Lozovatsky’s observations of Pendergrass’s appearance or his self- reporting on a given day do not inherently contradict the doctor’s conclusion that Pendergrass suffered “marked” limitations.

III. Pendergrass’s Testimony The ALJ also appears, without adequate explanation, to have extrapolated from Pendergrass’s testimony that he had “some decent days,” A.R. 20, and was “able to live alone and drive a car,” A.R. 18, the supposition that Pendergrass could undertake full- time work. But especially where mental illness is involved, “[d]isability does not mean that a claimant must vegetate in a dark room excluded from all forms of human and social activity.” Smith v. Califano, 637 F.2d 968, 971 (3d Cir. 1981). Nor does living independently translate into an ability to sustain a full-time job, because “the work environment is completely different from [the] home or a mental health clinic.” Morales v. Apfel, 225 F.3d 310, 319 (3d Cir. 2000). To the contrary, making “some improvement” with severe mental illness “does not mean that the person’s impairments no longer seriously affect her ability to function in a workplace.”

Holohan, 246 F.3d at 1205.

Where, as here, a claimant reports experiencing depression so severe that he would stay in his house for days at a time, unable to care for himself without help from relatives—limitations utterly incompatible with full-time employment—an ALJ must adequately explain how a claimant’s ability to occasionally complete some of the basic activities of daily living shows that he could sustain participation in the workforce.

IV. Consultative Experts’ Reports Lastly, the ALJ relied upon the reports of consultative experts Dr. Yalkowsky, a psychologist who described the severity of Pendergrass’s “mental health difficulties” as “moderate,” A.R. 389, and Dr. Smighelschi, a primary care physician who noted that Pendergrass could “perform [daily household activities] without help,” A.R. 392.

But, even as the ALJ acknowledged, neither physician “assess[ed] any vocational limitations” for Pendergrass, A.R. 22, and the ALJ failed to adequately explain how the reports nevertheless meant that Pendergrass could sustain full-time employment.

Cf. Brown v. Astrue, 649 F.3d 193, 196-97 (3d Cir. 2011) (noting that the ALJ adequately explained why she relied on the medical opinion of a non-treating doctor who did assess the claimant’s ability to work); Jones v. Sullivan, 954 F.2d 125, 129 (3d Cir. 1991) (concluding that the ALJ did not commit error in weighing heavily reports by two consultative experts who found the claimant had no “condition that would preclude gainful employment”).

Nor did the ALJ appear to recognize the significance of the fact that both reports were completed before, and thus, were not informed by Dr. Lozovatsky’s RFC. On similar facts, the Ninth Circuit concluded that a consultative expert’s report that classified an impairment as moderate, according to the ALJ, but which did not “opine about [the claimant’s] ability to work,” did not undermine the treating physician’s report characterizing the claimant’s condition as more severe. Regennitter v. Comm’r Soc. Sec., 166 F.3d 1294, 1299 (9th Cir. 1999).

In sum, the ALJ here failed to engage with these inconsistencies—or, at a minimum, failed to explain how he reconciled them—so that we could engage in meaningful appellate review to conclude his decision was supported by substantial evidence.

* * * Assessing the effect of mental illness on a claimant’s ability to work requires special care and attention, and an assessment of the relevant evidence that accounts for its unique features. The ALJ’s analysis fell short of these expectations, precluding meaningful review. We should vacate and remand for the ALJ to properly analyze the evidence in this case. As my colleagues decline to do so, I respectfully dissent.