Appeal from the United States District Court for the Western District of Wisconsin. No. 3:23-cv-00063 — James D. Peterson , Chief Judge .

ARGUED NOVEMBER 13, 2025 — DECIDED JULY 6, 2026

Before EASTERBROOK, LEE, and MALDONADO, Circuit Judges .

MALDONADO, Circuit Judge . Dimitrios Liapis suffers from bipolar disorder, knee problems, a bad back, a left-foot and ankle injury, and an ear injury. In July 2020, Liapis applied for disability benefits , alleging that he had been unable to work since December 2019. Wisconsin’s disability agency sent Liapis to be evaluated by several consultative examiners, one of Page 2whom, Dr. Mark Pushkash, Ph.D., observed that Liapis's “concentration” and “persistence” abilities were “markedly impaired.' Still, an Administrative Law Judge (' ALJ') found Liapis was not disabled, determining, among other things, that Dr. Pushkash’s medical opinion was “generally unpersuasive.' Liapis unsuccessfully challenged the ALJ's treatment of Dr. Pushkash's medical opinion before the district court, and now he turns to this court. While the ALJ’s analysis contained several errors, these errors were harmless because they made no difference to the ALJ's ultimate disability conclusion . Therefore, we affirm .

As a result of his back, ankle, and knee injuries, Liapis suffers from chronic pain, walks with a cane, and has constrained mobility. Specifically, Liapis experiences pain when walking, sitting, or standing for extended periods. Liapis also has been prescribed several strong psychotropic medications to manage his bipolar disorder and anxiety, but he struggles with medication management. Liapis's providers have, at times, recommended hospitalization to stabilize his mental health conditions and recalibrate medication intake.

In July 2020, Liapis filed applications for disability benefits. The Wisconsin state disability agency referred Liapis to several consultative examiners who opined fairly uniformly that Liapis has significant physical and mental impairments but is able to work in certain contexts.

For example, one doctor found that Liapis “has chronic pain that interferes with his ability to concentrate” and “needs breaks, which would interfere with his pace,” but that he still “would be able to carry out 2-3 step instructions on a con- Page 3sistent basis provided he has access to routine breaks.” As a result, the doctor found Liapis “capable of routine, unskilled work.” A second doctor, a clinical psychologist, found that Liapis’s “ability to complete a normal workday and workweek without interruptions from psychologically based symptoms and to perform at a consistent pace without an unreasonable number and length of rest periods' was only 'moderately limited.” A third doctor found that Liapis had “sustained concentration and persistence limitations,” but that he was “able to sustain the mental demands associated with carrying out simple tasks over the course of routine workday/workweek within acceptable attention, persistence, [and] pace tolerances.” Finally, a fourth doctor opined that Liapis could work at the sedentary level, occasionally lift ten pounds, frequently lift less than ten pounds, and sit for about six hours in a normal workday.

Most importantly for the purposes of this appeal, Dr. Pushkash, another clinical psychologist, evaluated Liapis and reported that his “ability to concentrate and persist on tasks in a work environment would be markedly impaired because of the interfering effects of chronic pain and anxiety/depression.” Still, Dr. Pushkash concluded that “[i]n all[,] results of the cognitive measures were unremarkable.' As Dr. Pushkash observed:

This man has the intellectual capabilities to comprehend, recall, and follow through on instructions. … It is felt that he would be able to appropriately relate to supervisors and coworkers in a work environment as long as he takes his psychotropic medication; however, without it he would have difficulty maintaining control

Page 4over his anger. … It appears that the primary impairment for this man is due to his medical conditions.

After the Wisconsin disability agency denied Liapis's claim, both initially and on reconsideration, Liapis requested and received a hearing before an ALJ. Applying the five -step inquiry required by 20 C.F.R. § 404.1520, the ALJ denied Liapis's claim. In relevant part, the ALJ found that Dr. Pushkash's opinion was 'generally unpersuasive' because Dr. Pushkash “evaluated the claimant on only one occasion” and inappropriately opined on Liapis’s chronic pain, an area outside of his expertise. The ALJ also found that despite Dr. Pushkash’s opinion to the contrary, “[t]he overall records support good mental status examination findings and good symptom control with relatively conservative treatment.” The Appeals Council declined review of the ALJ’s decision, rendering that decision 'the Administration's final word .” See Poole v. Kijakazi , 28 F.4th 792, 794 (7th Cir. 2022). Liapis then sought review in the district court, focusing exclusively on the ALJ’s determination that Dr. Pushkash’s opinion was unpersuasive. The district court affirmed the ALJ's decision.

On appeal, “[w]e review the district judge’s decision de novo , without deference. But we apply a deferential standard to the ALJ’s decision. We will reverse an ALJ’s decision denying benefits only if it is not supported by substantial evidence or if it is the result of an error of law.” Gedatus v. Saul , 994 F.3d 893, 900 (7th Cir. 2021) (citing 42 U.S.C. § 405(g); Lopez v. Barnhart , 336 F.3d 535, 539 (7th Cir. 2003)). “Although we will not reweigh the evidence or substitute our own judgment for that of the ALJ, we will examine the ALJ’s decision to determine Page 5whether it reflects a logical bridge from the evidence to the conclusions[.]” Moore v. Colvin , 743 F.3d 1118, 1121 (7th Cir. 2014) (citations omitted). But '[a]n ALJ need not address every piece or category of evidence identified by a claimant, fully summarize the record, or cite support for every proposition or chain of reasoning.' Warnell v. O'Malley , 97 F.4th 1050, 1053 (7th Cir. 2024).

Liapis offers several reasons why we should find that the ALJ erred in finding Dr. Pushkash’s medical opinion unpersuasive. To start, he contends that the ALJ did not comply with 20 C.F.R. § 404.1520c, which, he asserts, required the ALJ to articulate how persuasive he found Dr. Pushkash’s medical opinion by discussing Dr. Pushkash’s relationship with Liapis as well as Dr. Pushkash’s familiarity with the other evidence in the claim. Liapis also argues that the ALJ erred by discrediting Dr. Pushkash’s medical opinion based on his findings that (1) Dr. Pushkash provided only a “one-time” evaluation; (2) Dr. Pushkash 'confabulate[d]' Liapis's 'mental limitations with his alleged physical issues, for which Dr. Pushkash ha[d] no expertise'; and (3) '[t]he overall records support good mental status examination findings and good symptom control with relatively conservative treatment[.]” While it is not dispositive of the outcome here, we agree with Liapis that the ALJ's analysis of Dr. Pushkash's medical opinion was flawed in four main ways.

First, the ALJ erred in his analysis of Dr. Pushkash’s opinion because he did not explain his reasoning as to the two factors most critical to determining an opinion’s persuasiveness: supportability and consistency. As background, the regula- Page 6tion governing the Social Security Administration's ('the Agency”) consideration of medical opinions, 20 C.F.R. § 404.1520c, sets forth five factors to determine an opinion’s persuasiveness: (1) supportability, (2) consistency, (3) the medical source’s relationship with the claimant, (4) whether the medical source has an advanced specialization, and (5) other factors, such as the medical source's familiarity with other evidence in the claim. The regulation provides, in key part, that 'the most important factors' the Agency should consider are the first two: 'supportability … and consistency.' Id. § 404.1520c(b)(2). But while the Agency must explain those two factors in its analysis, it “may, but [is] not required to, explain how [it] considered' the third through fi fth factors. Id.

Liapis argues that the regulation requires “some consideration' of the third through fifth factors because supportability and consistency necessarily hinge on information contained in those other factors. The regulation's defin itions of 'supportability' and 'consistency' do not mention the third through fifth factors, though. See id. § 404.1520c(c)(1)–(2). And while Liapis points us to some cases requiring consideration of those factors, all involved claims filed before the Agency updated the regulatory scheme as described above. See Albert v. Kijakazi , 34 F.4th 611, 614 (7th Cir. 2022) (noting the effects of the updated framework). Consistent with the plain text of § 404.1520c(b)(2), therefore, an ALJ is not required to reference the third, fourth, and fifth factors outlined in § 404.1520c(c) when articulating his reasons for finding a medical opinion unpersuasive.

Still, for different reasons than those identified by Liapis , the ALJ erred in his application of the § 404.1520c(c) factors.

Page 7Here, the ALJ never stated that Dr. Pushkash’s opinion was not supported by relevant, objective medical evidence (supportability), and he never stated that Dr. Pushkash’s opinion was inconsistent with the other evidence (consistency), as required by § 404.1520c(b)(2). Instead, the ALJ focused on the criteria that do not require explicit articulation, finding that Dr. Pushkash’s medical opinion was not reliable because he “evaluated [Liapis] on only one occasion” (relationship with the claimant), and 'some of the limitations' identified by Dr. Pushkash “confabulate [Liapis’s] mental limitations with his alleged physical issues, for which Dr. Pushkash has no expertise” (specialization).

In short, the ALJ got the analysis under § 404.1520c backwards. While he was supposed to “explain how [he] considered the supportability and consistency factors” for Dr. Pushkash’s opinion, he provided no such explanation, and instead, provided explanation for the discretionary factors in the third through fifth paragraphs. This was clear legal error.

Second, the ALJ also erred in dismissing Dr. Pushkash’s medical opinion as unpersuasive based on Dr. Pushkash's one-time evaluation of Liapis. The ALJ's decision does not 'reflect[ ] an adequate logical bridge from the evidence to the conclusion[],” because the ALJ found the opinions of the other four doctors—each of whom also evaluated Liapis just once— to be persuasive. Gedatus , 994 F.3d at 900; see Myles v. Astrue , 582 F.3d 672, 678 (7th Cir. 2009) (per curiam) (' An ALJ may not selectively consider medical reports[.]'). This basis for finding Dr. Pushkash's opinion unpersuasive was therefore error.

Third, the ALJ erred further by dismissing Dr. Pushkash’s medical opinion as unpersuasive based on Dr. Pushkash Page 8opining on 'the interfering effects of chronic pain ,” a “physical issue[], for which Dr. Pushkash[, a psychologist,] has no expertise.” But the other clinical psychologist who evaluated Liapis similarly concluded that Liapis 'is struggling with chronic pain, which interferes with his ability to adapt,” and the ALJ found that opinion persuasive. This inconsistent treatment of the two opinions is illogical. See Lothridge v. Saul , 984 F.3d 1227, 1234 (7th Cir. 2021) (“[A]n internally inconsistent opinion by an ALJ is likely to fail to build a logical bridge between the evidence and the result.”).

Further, our precedent is clear that physical pain can be the source of a mental limitation. See Simila v. Astrue , 573 F.3d 503, 522 (7th Cir. 2009) (the claimant's 'moderate difficulties with concentration, persistence, and pace stemmed from his chronic pain syndrome and somatoform disorder. … These impairments are rooted in [the claimant's] allegations of pain.”); Carradine v. Barnhart , 360 F.3d 751, 754, 756 (7th Cir. 2004) (observing that “[p]ain is always subjective in the sense of being experienced in the brain” and citing approvingly the medical opinion of a clinical psychologist who reported that the claimant’s “attention and concentration are impaired by her focus on pain”). The ALJ’s determination that a psychologist cannot persuasively opine on the psychological effects of physical symptoms, such as pain, was therefore in error.

Fourth, the ALJ erred by finding Dr. Pushkash's medical opinion unpersuasive based on the ALJ's view that '[t]he overall records support good mental status examination findings and good symptom control with relatively conservative treatment[.]” The ALJ again provided no “logical bridge” between the evidence and his conclusion, Gedatus , 994 F.3d at 900, as he included no analysis of Liapis’s treatments or symp- Page 9toms. The ALJ also ignored evidence in the record suggesting poor symptom control and aggressive psychiatric treatment, which would support Dr. Pushkash’s opinion. See Scrogham v. Colvin , 765 F.3d 685, 701 (7th Cir. 2014) (fi nding that ALJ committed “logical error [] ' by 'misinterpreting the significance of [the claimant’s] extensive treatment”).

The record reflects that for years, Liapis has been prescribed a cocktail of strong psychotropic and pain-relieving medications such as amitriptyline, cyclobenzaprine, lithium, oxycodone, pregabalin, quetiapine, baclofen, gabapentin, hydroxyzine HCL, and trazodone. And, as Liapis notes, it is not “obvious what the next rung of the treatment ladder would be. Lithium and Seroquel [a brand name of quetiapine] are already two of the most aggressive psychoactive medications available.' Despite the many medications Liapis was prescribed, he continued to tell providers that he was feeling anxious, 'deal[ing] with racing thoughts,' and considered his medications 'ineffective.' On one occasion, after Liapis reported that his symptoms included 'feeling confused, or ‘brain-dead,’” he tested positive for lithium toxicity, caused by excessive intake of his lithium medication. And as the other clinical psychologist who evaluated Liapis observed, Liapis 'does not seem to have many coping skills outside of medication.” This evidence suggests that Liapis did not have “good symptom control.”

Further, to reduce his ankle pain, Liapis underwent two major ankle surgeries in a year, including an ankle fusion surgery wherein the bones in Liapis's left ankle were permanently joined using rods and screws. It would be unreasonable to consider the long-term combination of powerful medications, as well as the surgical insertion of parts to perma- Page 10nently fuse a joint, to be 'conservative treatment[s].' See Scrogham , 765 F.3d at 701 (“Instead of showing that [the claimant’s] limitations were not as severe as he alleged, evidence that he was willing to undergo risky surgery and take powerful pain medication—and that physicians were willing to prescribe this course of treatment— reflects that [the claimant's] symptoms caused him real problems.”).

By ignoring these significant medical interventions, and improperly construing the record evidence, the ALJ erred in its review of Dr. Pushkash's opinion. See Myles , 582 F.3d at 678 (cautioning against selective consideration of medical evidence); Goins v. Colvin , 764 F.3d 677, 680 (7th Cir. 2014) (ALJs may not make assessments about the severity of a condition that only a trained physician could make).

Though we acknowledge the many errors in the ALJ's analysis of Dr. Pushkash’s medical opinion, we conclude that ultimately, these errors were harmless because there is no basis in the record to suggest that Liapis had more than a single impairment in any broad area of mental functioning.

“[T]he harmless error standard applies to judicial review of administrative decisions,” Butler v. Kijakazi , 4 F.4th 498, 504 (7th Cir. 2021), and “we will not remand a case to the ALJ for further specification where we are convinced that the ALJ will reach the same result,” McKinzey v. Astrue , 641 F.3d 884, 892 (7th Cir. 2011). “In assessing whether an error is harmless, we examine the record to determine whether we can 'predict with great confidence what the result of remand will be.'' Butler , 4 F.4th at 504 (quoting McKinzey , 641 F.3d at 892).

Page 11Here, even if the ALJ found Dr. Pushkash’s medical opinion persuasive, “we are convinced that the ALJ will reach the same result.” See McKinzey , 641 F.3d at 892. First, Liapis challenges only the ALJ's treatment of Dr. Pushkash's medical opinion, leaving the rest of the ALJ’s opinion—which largely faithfully reports the medical evidence in the record—intact and not up for our review.

Second, under the Agency's five -step process for finding disability, 20 C.F.R. § 404.1520, an ALJ cannot find a claimant disabled based solely on one “marked” limitation in an area of mental function. See 20 C.F.R. Part 404, Subpart P, Appendix 1, § 12.00A(2)(b) . ALJs instead must evaluate 'the effects of [the claimant's] mental disorder' on a 'five -point rating scale consisting of none, mild, moderate, marked, and extreme limitation,” by weighing the claimant’s ability to (1) understand, remember, or apply information; (2) interact with others; (3) concentrate, persist or maintain pace; and (4) adapt or manage oneself. 20 C.F.R. Part 404, Subpart P, Appendix 1, § 12.00E, F(2). The ALJ must find that the claimant has a “marked” limitation in two of these four areas of mental functioning, or an 'extreme' limitation in one, in order to fi nd the claimant disabled. See 20 C.F.R. Part 404, Subpart P, Appendix 1, § 12.00A(2)(b) .

Here, the ALJ found that Liapis had moderate limitations in the second through fourth areas of functioning (interacting, concentrating, and adapting), and a mild limitation in the first (understanding). Ultimately, the ALJ concluded that “[b]ecause the claimant’s mental impairments do not cause at least two ‘marked’ limitations or one ‘extreme’ limitation,” Liapis was not disabled.

Page 12The ALJ's findings as to every factor but the third (concentrating) are in line with Dr. Pushkash’s conclusions. Dr. Pushkash found that Liapis 'has the intellectual capabilities to comprehend, recall, and follow through on instructions,” and “he would be able to appropriately relate to supervisors and coworkers in a work environment as long as he takes his psychotropic medication.” Thus, even if the ALJ found Dr. Pushkash’s medical opinion persuasive, it would not change the ALJ’s conclusions as to Liapis’s ability to understand information, interact with others, or manage himself.

Nor would it change the ALJ’s conclusion as to the third area of mental functioning because the four other doctors all agreed that Liapis was largely unimpaired with respect to his ability to “concentrate, persist or maintain pace,” and therefore, was capable of routine, unskilled work. Dr. Pushkash’s opinion that Liapis was markedly impaired would therefore be so outweighed as to not change the ALJ’s conclusions. And even if the ALJ adopted Dr. Pushkash’s conclusion that Liapis had a marked limitation in 'concentrating, persisting or maintaining pace,' despite the other doctors' conclusions, that single limitation would be insufficient to warrant a finding of disability. See 20 C.F.R. Part 404, Subpart P, Appendix 1, § 12.00A(2)(b) . Instead, Liapis would still need a second marked limitation, or an extreme limitation, to be found disabled, and neither Liapis nor Dr. Pushkash have suggested another area of mental functioning where Liapis might have such a limitation.

As a result, because 'we can predict with great confidence [t]hat the result on remand will be” the same, the ALJ’s illogic in its evaluation of Dr. Pushkash’s medical opinion was harmless error. See McKinzey , 641 F.3d at 892.

Page 13Accordingly, we AFFIRM the judgment of the district court.