NONPRECEDENTIAL DISPOSITION
To be cited only in accordance with Fed. R. App. P. 32.1 Before DIANE P. WOOD, Chief Judge MICHAEL S. KANNE, Circuit Judge AMY J. ST. EVE, Circuit Judge Appeal from the United States District Court for the Southern District of Indiana, Terre Haute Division. No. 2:17-cv-00235-MJD-WTL Mark J. Dinsmore, Magistrate Judge.

ORDER

Carrie Paul suffers from several mental impairments. She applied for supplemental security income claiming disability based on a seizure disorder, anxiety, and polysubstance dependence, among other things. An administrative law judge concluded that she had moderate mental difficulties, but that she could work with certain limitations. The district court upheld that decision. She now contends that the ALJ improperly rejected the opinion of an agency psychologist and failed adequately to account for her mental impairments. We agree that the substantial evidence does not support the ALJ's findings, and so we remand this case to the agency for further proceedings.

I

Paul was 39 years old when she filed her application for benefits, alleging an onset date of February 1, 2013. She identified several ailments that prevented her from working: a spinal-fusion surgery in 1983 with lumbar spondylolisthesis, scoliosis, schizoaffective disorder, seizure disorder, anxiety, and polysubstance dependence. Because Paul contests only the ALJ's conclusions regarding her mental conditions, we limit our analysis to those conditions.

In connection with her application, Paul had a psychological consultative examination in April 2013 with Dr. Leah Powell, who concluded that Paul's difficulties managing her behavior hindered her ability to work. Dr. Powell opined that "[s]he is impulsive, has difficulty making good decisions, and has difficulty controlling her own behavior" and that "[t]hese symptoms are contraindicated with work related activities." Dr. Powell diagnosed schizoaffective disorder-bipolar type, generalized anxiety disorder, and alcohol abuse. Dr. Powell assigned a "Global Assessment of Function," or GAF, of 51, indicating moderate symptoms or moderate difficulty in social or occupational functioning. AM. PSYCHIATRIC ASS'N, DIAGNOSTIC AND STATISTICAL MANUAL OF MENTAL DISORDERS 32 (4th ed. 1994). She also observed that Paul's mood was "tense," and that she was "compromised" in managing her mood, anger, and her responsibilities in interpersonal relationships. During the exam, Paul discussed her depressed mood, irritability, difficulty concentrating and remembering, and sleep disturbances. She stated that she began treatment for her depression 20 years ago and experiences bipolar symptoms. Paul also said that she experienced daily panic attacks, chronic crying episodes, and visual hallucinations, delusions, and suicidal ideations. Though Paul showed "no evidence of distractibility," Dr. Powell also reported Paul's "difficulty with immediate memory and in her attending skills" and that her social judgment was compromised.

Agency psychologist Dr. Joseph Pressner reviewed Dr. Powell's report and treatment records in July 2013. He noted Paul's difficulties maintaining concentration, persistence, or pace, but he concluded that she could work. He identified four areas in his mental residual-capacity assessment in which Paul has moderate limitations: the ability to (1) understand and remember detailed instructions; (2) respond to detailed instructions; (3) maintain attention and concentration for long periods; and (4) perform activities within a schedule and maintain regular attendance. Yet he reported that despite her limitations in immediate memory and sustained attention, Paul "is cognitively capable of performing simple, repetitive work-like tasks on a sustained basis at a reasonable pace without special considerations."

After a referral from probation following a DUI charge, Paul began counseling in June 2014 with a behavioral health therapist. The therapist diagnosed polysubstance dependence and noted the possibility of bipolar disorder based on Paul's history of manic behavior and depressive symptoms. She further recommended that Paul attend intensive outpatient therapy for substance abuse three times a week. The therapist observed that Paul's mood was depressed and anxious but that her memory, orientation, and intellect were within normal limits. Paul is "often impulsive," the therapist noted, and fails to think through consequences easily. At her first session, Paul stated that she cannot sleep because of her anxiety and feels sad daily, is constantly fatigued, and struggles getting out of bed, concentrating, and remaining focused. Paul lacks friends, the therapist said, because she does not trust others. The therapist also assessed a GAF of 60, demonstrating moderate difficulty in social or occupational functioning. AM. PSYCHIATRIC ASS'N, supra. Although Paul did not attend therapy regularly, the therapist observed that she was "highly distractible" and had difficulty maintaining focus.

Paul's claim for disability benefits was denied initially and upon reconsideration, and so she proceeded to a hearing before an administrative law judge. Paul first testified about her daily activities, recounting that her husband and teenage children vacuum and do the laundry because those activities cause her pain. She primarily cooks microwaved meals for her family because she cannot stand at the stove. She then testified about her mental-health limitations, particularly her difficulties with interpersonal relationships and maintaining focus. She said that she does not have any friends and her social interactions are limited to her family because being around others causes her stress. She also struggles to concentrate and gets off task easily. On days when she experiences bipolar episodes, she does not get up at all and remains in her room for days at a time without leaving or showering.

A vocational expert ("VE") testified that Paul had the capacity to work if the work adhered to specific mental-health limitations. The ALJ asked the expert to consider a hypothetical claimant with Paul's vocational background and education who had the mental capacity to perform "simple, routine, repetitive tasks" that do not require directing others, abstract thought, or planning; and would involve only "simple, work related decisions, and routine workplace changes." The ALJ added that the claimant would require work that could be performed at a "flexible pace," meaning that it is "free of production rate pace where there are no tandem tasks or teamwork or one production step that's dependent upon the prior step." And the job must entail only occasional interaction with the public, coworkers, or supervisors. The expert testified that, with these limitations and the relevant physical limitations, the person could work as a housekeeping cleaner, marker, routing clerk, microfilm document preparer, ampule sealer, and para-mutual ticket checker.

Appling the familiar five-step analysis, see 20 C.F.R. § 416.920(a), the ALJ concluded that Paul was not disabled. At Step 1, the ALJ determined that Paul had not engaged in substantial gainful employment since her onset date. At Step 2, the ALJ found that Paul's remote spinal fusion with lumbar spondylolisthesis, scoliosis, schizoaffective disorder, seizure disorder, anxiety, and polysubstance disorder were severe impairments. At Step 3, he acknowledged that she had moderate difficulties in social functioning and concentration, persistence or pace because of her compromised memory, attention, thought processes, and judgment. But he found that she did not meet any of the agency's listings for a presumptively disabling impairment. Between Steps 3 and 4, the ALJ then assigned a residual functional capacity ("RFC") consistent with that of the hypothetical claimant he described to the VE. The ALJ attributed some of Paul's symptoms to her medically determinable impairments, but found she overstated the intensity, persistence, and limiting effects of the symptoms. And the ALJ decided that the opinion evidence of the agency's psychologist, Dr. Powell, warranted "little weight," because it was a one-time examination, not specific in terms of functioning, and was inconsistent with Paul's daily activities and treatment records. Finally, at Step 5, the ALJ determined that Paul could not perform any past relevant work but could find other work in the national economy.

After the Appeals Council denied review, Paul sought judicial review. The parties agreed to have the case adjudicated by a magistrate judge, see 28 U.S.C. § 636(c), and he in turn upheld the ALJ's decision. The judge concluded that the ALJ reasonably discounted Dr. Powell's opinion because it was vague and conclusive, and that the ALJ's RFC assessment appropriately incorporated Paul's mental-health limitations, given Dr. Pressner's opinion that Paul could perform simple tasks at a "reasonable pace without special considerations."

II

A. Dr. Powell

Paul first contends that the ALJ improperly evaluated the opinion of the agency's examining psychologist, Dr. Powell. She challenges each of the ALJ's reasons for discounting Dr. Powell's opinion (i.e., that it was a one-time examination, was not specific in terms of functioning, and was inconsistent with Paul's daily activities and treatment records). If fully credited, Dr. Powell's opinion—that Paul's mood disorder, impulsivity, and difficulties in interpersonal relationships are "contraindicated with work related activities"—suggests that Paul required a more restrictive RFC.

The ALJ's decision to afford Dr. Powell's opinion "little weight" is perplexing because, although an ALJ may discount the opinion of the agency's examining physician when contrary evidence exists, see Beardsley v. Colvin, 758 F.3d 834, 839 (7th Cir. 2014), Dr. Powell's examining notes are consistent with Paul's treatment records. Both Paul's therapist and Dr. Powell reported that Paul had a depressed mood, experienced sleep disturbances because of anxiety, had difficulty with interpersonal relationships, concentration, and focus, and was impulsive, failing to think through the consequences. Both providers assigned GAF scores that indicated moderate difficulty in social or occupational functioning.

An ALJ may not discount the opinion of an examining physician without a valid explanation, see Beardsley, 758 F.3d at 839, and here the ALJ failed to support his decision. First, the ALJ wrongly discounted Dr. Powell's opinion for being based on a one-time examination when he gave "great weight" to non-examining physician Dr. Pressner, whose opinion was based on a one-time review of Paul's mental-health records (including a review of Dr. Powell's own examining notes). Further, the agency's regulations instruct that the medical opinion of an examining source, such as Dr. Powell, generally receives more weight than the medical opinion of a non-examining source, such as Dr. Pressner. See 20 C.F.R. § 416.927(c)(1). A conclusion to the contrary requires a "good explanation for this unusual step," Beardsley, 758 F.3d at 839, which the ALJ did not supply here.

Second, the ALJ's finding that Dr. Powell's opinion "is vague and not specific in terms of functioning" is unsupported by the evidence. Again, the ALJ did not substantiate this finding. And Dr. Powell's opinion was sufficiently specific: the six-page narrative report recounted Paul's performance on her mental-status exam and included the doctor's behavioral observations of Paul; included observations that Paul's mood was tense, that she had symptoms associated with bipolar disorder like "mood swings, depressed mood, irritability, racing thoughts, and difficulty concentrating and remembering"; and contained explicit findings that Paul had difficulty with her immediate memory and in her attending skills, had a compromised mood due to her schizoaffective disorder, and that "work related activities" were "contraindicated" with her difficulty managing her behavior and her responsibilities in interpersonal relationships. In any case, if the ALJ believed that Dr. Powell's opinion was deficient, the ALJ should have sought additional clarification from Dr. Powell before discounting it outright. See 20 C.F.R. § 416.919p; Simila v. Astrue, 573 F.3d 503, 516 (7th Cir. 2009).

We are also unimpressed with the final two reasons the ALJ gave for rejecting Dr. Powell's opinion—that the opinion deserved little weight because "the claimant was not consistent with attending her mental health therapy, and she continues to be able to take care of a household, including her two children." Regarding the former, the ALJ does not explain why Paul's inconsistent attendance at her therapy sessions provides a reasonable basis to discount the examining physician's opinion. Moreover, ALJs assessing mental illness and bipolar disorder must consider possible alternative explanations before racing to conclusions about noncompliance with medical directives. See Jelinek v. Astrue, 662 F.3d 805, 814 (7th Cir. 2011) (stating that bipolar disorder is "by nature episodic and admits to regular fluctuations even under proper treatment" so ALJs must consider alternative explanations before determining that non-compliance supports an adverse credibility inference). And Paul's ability to maintain her household, by performing simple chores, does not indicate a capacity to work full-time. See Roddy v. Astrue, 705 F.3d 631, 639 (7th Cir. 2013). Paul's family, including her two teenage children, do the household chores such as laundry and vacuuming. And even though Paul sometimes cooks using the microwave, she testified that during bad periods she does not leave her room for consecutive days.

B. Limitations in concentration, persistence, or pace

In her second argument, Paul contends that the ALJ erred in his RFC assessment when he failed to account for his own findings of moderate limitations in concentration, persistence, or pace. This flawed RFC, she adds, tainted the hypothetical question that he posed to the VE. To the extent that the RFC refers to "simple, routine, repetitive tasks," Paul says that these limitations are insufficient to encompass her difficulties maintaining her focus and concentration. Further, she asserts that the reference in the RFC to "flexible pace"—as tasks that exclude "production rate pace" or "tandem tasks or teamwork where one production step is dependent on the prior step"—also does not adequately address her difficulties, because it fails to specify the particular pace at which she can work.

Paul's challenge to the RFC is well-taken. Though an RFC assessment need not recite the precise phrase "concentration, persistence, or pace," any alternative phrasing must clearly exclude those tasks that someone with the claimant's limitations could not perform. See Moreno v. Berryhill, 882 F.3d 722, 739 (7th Cir. 2018), as amended on reh'g (Apr. 13, 2018); O'Connor-Spinner v. Astrue, 627 F.3d 614, 619 (7th Cir. 2010). Paul's ability to learn routine, unskilled tasks does not address whether she can also maintain the concentration and focus needed to sustain her performance of that task for an extended period. See Moreno, 882 F.3d at 730. The RFC and the hypothetical question here exclude complex tasks that require higher-level thinking but do not acknowledge Paul's moderate limitations with following a schedule and sticking to a given task. And the ALJ's reference to "flexible pace" is insufficient to account for Paul's difficulties maintaining focus and performing activities within a schedule, because the reference excludes only production-pace employment. See Varga v. Colvin, 794 F.3d 809, 815 (7th Cir. 2015). Without more, the VE cannot determine whether someone with Paul's limitations could maintain the proposed pace or what the proposed pace even is. See id.

Thus, because the ALJ failed to provide sufficient reasons to discount Dr. Powell's opinion, and because the hypothetical questions to the vocational expert did not capture Paul's difficulties with attention and concentration, we VACATE the district court's judgment and REMAND this case to the Social Security Administration for further proceedings.