OPINION
RONALD LEE GILMAN, Circuit Judge.
Charles Gayheart applied for Social Security disability insurance benefits (DIB) in December 2005 due to manifestations of anxiety, panic disorder, bipolar disorder, and depression. After an initial denial of his application and three separate hearings, an administrative law judge (ALJ) found that the limitations caused by Gayheart’s impairments did not preclude him from performing a significant number of jobs available in the national economy. The ALJ thus denied Gayheart’s application for DIB. Gayheart’s request for an administrative appeal was likewise denied, making the agency’s decision final. He then sought review in the federal district court pursuant to 42 U.S.C. § 405(g).
The Report and Recommendation issued by the assigned magistrate judge concluded that the ALJ’s decision was not supported by substantial evidence and that Gayheart should be awarded DIB. But the district court sustained the Commissioner’s objections to the magistrate judge’s Report and Recommendation and affirmed the ALJ’s decision. Gayheart then timely filed this appeal. For the reasons set forth below, we REVERSE the judgment of the district court and REMAND with instructions that the case be returned to the Social Security Administration for reconsideration.
I. BACKGROUND
A. Treatment history
After working as an assistant manager in an auto parts store for more than 20 years, Gayheart quit in September 2005 because he was suffering daily panic attacks in the workplace that lasted up to 15 minutes at a time. He had been taking prescription medication to treat depression, anxiety, and panic attacks since 2000. Sweating, shaking, dizziness, shortness of breath, and an accelerated heart rate accompanied the panic attacks, and Gayheart’s supervisor had to drive him home on several occasions after Gayheart suffered such attacks. Gayheart also regularly drank several “beers,” varying from as many as six to as few as two, every night since he was 22 years old.
Soon after quitting work, Gayheart sought treatment from Jackie Thompson, a psychological therapist. Gayheart reported daily panic attacks, headaches, nervousness, decreased sex drive, social anxiety, and sleeping difficulties to Thompson, who diagnosed him as having a generalized anxiety disorder and a panic disorder without agoraphobia (a fear of public places), with a differential diagnosis of bipolar disorder. Thompson assigned Gayheart a Global Assessment Functioning (GAF) score of 42, indicating that Gayheart’s overall physiological functioning reflected severe symptoms or serious impairment in social or occupational functioning.
See White v. Comm’r of Soc. See.,
572 F.3d 272, 276 (6th Cir.2009) (explaining the import of GAF scores); American Psychiatric Association Diagnostic and Statistical Manual of Mental Disorders, Text Revision 34 (4th ed. 2000) (noting that GAF scores between 41 and 50 indicate serious symptoms). Gayheart had seven therapy sessions with Thompson from October 2005 to February 2006, at which point
Thompson’s practice no longer accepted the medical insurance of Gayheart’s wife.
Thompson reported in January 2006 to the Social Security Administration’s Bureau of Disability Determination that Gayheart suffered from bipolar disorder and a generalized anxiety disorder, that his tolerance for stress was extremely low, and that his depression, mood swings, verbal aggression, frustration, and anxiety “interfere with [his] ability to function effectively.” Her report also described Gayheart’s poor concentration, slowness in processing information, and inability to be in public places or socialize in large crowds.
Gayheart first saw psychiatrist Alice Onady in mid-November 2005 and continued seeing her through the point of the ALJ’s final hearing in February 2009. Dr. Onady diagnosed Gayheart’s condition as a panic disorder, with bipolar II as a differential diagnosis. She also noted antisocial traits and assigned Gayheart a GAF score of 50, which is in the range of serious symptoms or impairments. Dr. Onady was aware of Gayheart’s beer consumption and instructed him not to drink alcohol with any of the antidepressant and anti-anxiety medication that she prescribed. Her diagnoses reflected clinical observations of depression, anxiety, isolation, and socially phobic behavior.
Like Thompson, Dr. Onady completed a report for the Bureau of Disability Determination. Her diagnoses in that report, dated May 2006, were panic disorder, bipolar disorder, and mixed personality disorder (paranoid). She noted Gayheart’s poor concentration and memory, his decreased cognitive function, and his anxious, nervous, and edgy affect. Also noted were his poor abilities to handle simple and routine tasks in a work setting, to interact socially, and to adapt to change. Six months later, Dr. Onady added diagnoses of pain disorder and “alcohol abuse in partial remission” in her interrogatory responses to the Social Security Administration’s Office of Hearings and Appeals. She rated Gayh-eart’s functional limitations as “marked” in the following three areas: (1) daily living, (2) social functioning, and (3) concentration, persistence, or pace. Dr. Onady also noted three episodes of decompensation (temporary exacerbations in the symptoms of an illness or disorder).
Dr. Onady’s treatment notes continued to document diagnoses of panic disorder and bipolar II throughout her treatment of Gayheart, and she regularly modified his prescriptions of psychotropic medications. From November 2007 to February 2008, she diagnosed alcohol abuse and dependency, but this diagnosis does not appear in the notes from April through October of 2008. She added major depression to her diagnoses starting in April 2008. Gayh-eart noted that his visits to Dr. Onady were limited to approximately ten per year because that was the maximum number of visits his wife’s insurance policy would cover.
Finally, Gayheart started treatment with Jennifer Fenske-Doyle, a psychological therapist in Dr. Onady’s medical-practice center, in September 2006 and continued treatment with her through October 2007. Fenske-Doyle’s initial diagnoses were bipolar II and panic disorder with agoraphobia. She assigned Gayheart a GAF score of 49. Her notes from 13 months of treating Gayheart provide considerable detail about the severity of the symptoms that Gayheart suffered and how they impacted his daily life. They reflect Gayheart’s description of the anxiety that he suffers from simply having to drive himself to and from his therapy sessions. The notes also reflect his avoidance of social situations, a loss of interest in hobbies, paranoid thinking, feelings of hope
lessness, difficulties concentrating, and passive thoughts of suicide.
At the end of her treatment sessions with Gayheart, Fenske-Doyle noted a recent increase in his alcohol consumption of upwards to six “beers” per day. Her last diagnoses were bipolar II and panic disorder with agoraphobia. She recorded that Gayheart’s current GAF score was 45 and that the highest it had been in the past year was 49. According to Fenske-Doyle’s final assessment, Gayheart’s severe anxiety and panic disorder had not improved over the course of treatment.
B. Procedural history
The Commissioner denied Gayheart’s DIB claim initially in March 2006 and again upon reconsideration five months later. Gayheart then requested a hearing. The initial hearing ■ was held in March 2008, and two supplemental hearings followed in February 2009. Testifying at the hearings were Gayheart, two vocational experts, and clinical psychologist Mary Buban, the latter three on behalf of the Commissioner. Prior to the hearings, the Bureau of Disability Determination referred Gayheart to Dr. Jerry Flexman and to Dr. David Chiappone for consultative psychological evaluations. A third psychologist, Dr. David Demuth, did not examine Gayheart, but instead reviewed the medical records on behalf of the Bureau.
1. Testimony
Gayheart’s testimony described the onset of his panic attacks in 2000 and how they forced him to stop working in September 2005 because they were occurring daily and lasting up to 15 minutes. The attacks persisted after he stopped working but at a diminished pace, occurring once or twice a week rather than daily. His anxiety prevents him from sleeping normally and causes his hands to shake constantly, making him unable to do tasks like gripping objects and buttoning clothes. He also described other physical ailments, including an embolism in his left eye that causes daily severe headaches, and knee pain that has persisted despite surgeries. According to Gayheart, the latter condition will likely require a whole knee replacement.
Gayheart also described the limitations caused by his anxiety and panic disorder. The magistrate judge’s Report and Recommendation summarizes this testimony as follows:
Plaintiff testified that he continued to experience panic attacks at least once or twice per week, which was significantly down from the daily attacks he was experiencing in September 2005. Tr. 512-13, 550-51. He stated that he is only able to drive two or three times per week — and only on short trips — because of his anxiety. Tr. 507-08. The act of being around people causes Plaintiff to shake and can be a trigger of panic attacks. Tr. 512-13, 564. Plaintiff is able to do a limited amount of household chores, including putting clothes in the washing machine and washing dishes, but is unable to do other chores such as making the bed, mopping, sweeping, or vacuuming. Tr. 523, 556. Plaintiff testified that he had not been- to the grocery store or any type of store by himself in over a year. Tr. 564. When Plaintiff goes out in public, it is generally with his wife. Tr. 556. Plaintiff testified that he and his wife do not go out to eat. Tr. 529. His wife does all of the shopping and cooking. Tr. 522-23. Plaintiffs interaction with family beyond his wife and children is limited to biweekly visits from his wife’s aunt and uncle who stop by his house. Tr. 529, 556, 560. Plaintiff testified that he occasionally interacts with a neighbor, but
has lost contact with all of his former Mends. Tr. 556. He stated that the only time he has left Ohio in the last three years had been to accompany his uncle to a car dealership in Indiana, but had an anxiety attack at the car dealership. Tr. 525.
The Report and Recommendation goes on to describe Gayheart’s testimony about his depression, medication, and alcohol consumption:
He acknowledged experiencing constant worrying, negative thoughts, and a lack of concentration. Tr. 514, 522, 528. He has taken prescribed medication to treat his depression, anxiety, and panic attacks since 2000. Tr. 552-53. He testified that the side effects from the psychotropic medication include drowsiness, decreased energy, and restlessness. Tr. 514-15, 517, 558.... He testified at the first hearing, on March 26, 2008, that he generally drinks two or three beers per day to help calm his nerves, and that he has been drinking two or three beers per day since he was twenty-two years old. Tr. 526. At the second hearing nearly eleven months later, on February 10, 2009, Plaintiff testified that he had cut his alcohol consumption down to having two or three beers only one time per week based upon the advice of his psychiatrist, Dr. Onady. Tr. 557. Plaintiff denied ever being arrested for any alcohol-related offense, and stated that he has never needed treatment for alcohol.
Id.
He denied any other form of substance abuse, but acknowledged that he smokes one pack of cigarettes per day. Tr. 557-58.
Dr. Mary Buban testified as a medical expert at the second and third hearings in February 2009. She did not examine or interview Gayheart, but she had reviewed all of the exhibits, including the reports and treatment notes from Gayheart’s treating doctors and therapists, as well as from the consultative doctors. After summarizing reports from Drs. Onady, Flexman, and Chiappone, she testified at the second hearing that Gayheart would meet the Social Security Administration’s Listing 12.06 (Anxiety Related Disorder) as provided in 20 C.F.R. Part 404, Subpart P, Appendix 1.
The ALJ then asked Dr. Buban whether the alcohol abuse documented in the record was material to a finding of disability. At first Dr. Buban testified that the alcohol abuse appears to have been material through February 21, 2008, the date of Dr. Onady’s most recent treatment notes that Dr. Buban had reviewed. (The exhibit containing medical records after that date was missing from the file that Dr. Buban initially reviewed, which necessitated the third hearing.) But when the ALJ asked the question again to clarify Dr. Buban’s response, she replied: “Well, according to this treatment record, there is not sufficient evidence that without the alcohol, the anxiety improved.... When alcohol is continuous throughout the record, it’s impossible for me to say that it, it would cease being. There had to be a period of the sobriety.”
When the ALJ asked a third time whether the continued use of alcohol “would be material throughout the record,” Dr. Buban replied affirmatively, noting that alcohol was of “significant concern” to Dr. Onady and Fenske-Doyle through February 2008. The ALJ then posed the question a fourth time, after Gayheart’s attorney attempted to clarify that a diagnosis of alcohol abuse does not necessarily mean that the abuse was material to the noted impairments. He asked: “[W]ould the claimant’s condition meet or equal the listing solely looking at anxiety, not considering alcohol?” Dr. Buban replied: “And, and what I stated is that
there is no period of sobriety to determine that just the condition of anxiety would meet a listing.”
Having reviewed the post-February 2008 records that were initially missing from her file review and a new letter from Dr. Onady, Dr. Buban testified again two weeks later. Whereas she had previously opined that Gayheart met Listing 12.06 for anxiety-related disorders, her new testimony was that the intensity of Gayheart’s panic attacks was not documented in the treatment records and that “[n]othing from the information as we have it” would meet or medically equal a listing. Her testimony does not reflect why she no longer believed that the medical evidence she had reviewed supported a finding that Listing 12.06 was met, and she did not acknowledge her previous opinion or indicate that the new exhibits contributed to her changed opinion.
Although Dr. Buban’s subsequent opinion did not find that a listing was met, she testified that Gayheart would nonetheless have the following restrictions if he were found to be not disabled: he should perform activities independently, have no contact with the public, have minimal contact with coworkers and supervisors, and have no production quotas or fast-paced work. She stated that “those would be the kind of restrictions that would be consistent with the treating notes.”
The ALJ then made one final attempt to clarify Dr. Buban’s opinion regarding the materiality of Gayheart’s alcohol abuse in the following exchange:
,Q: [Y]ou told us last time we talked with you, that — you told us several times, I think, that the — throughout the record the issues of the anxiety and the panic attacks and the, the alcohol abuse were pretty much intertwined, and you really couldn’t say that the — whether the alcohol would be material to any findings of disability.
A: Correct.
Q: If the claimant were in fact found disabled. Is that still what your opinion is today?
A: Yes, Sir.
Finally, two vocational experts testified, one at the first hearing and the other at the third hearing, but their testimony is not critical to the dispute in this ease. The magistrate judge’s Report and Recommendation summarized their testimony as follows:
Both [experts] were asked to consider a person of Plaintiffs age, education and work experience, who was limited in the manner ultimately found by the ALJ. [Tr. 535-42, 592-99.] The vocational experts each identified a number of jobs they believed Plaintiff would be able to perform, but both acknowledged that none of the jobs could be maintained by a person who was unable to perform activities within a schedule, maintain regular attendance, and be punctual within customary tolerances. Tr. 539, 598. Neither vocational expert addressed the issue of whether a person who experienced constant tremors and had limited use of his hands because of the tremors would be able to perform their suggested vocations. Tr. 535-42, 592-99.
2. Reports of consultative doctors
Gayheart was separately examined by two psychologists, Drs. Flexman and Chiappone, at the request of the Bureau of Disability Determination. Dr. Flexman saw Gayheart in February 2006. He diagnosed Gayheart with panic disorder without agoraphobia and a personality disorder (not otherwise specified). Dr. Flexman assigned Gayheart a GAF score of 55, indicating moderate symptoms or moderate
difficulty in social or occupational functioning.
See
American Psychiatric Association Diagnostic and Statistical Manual of Mental Disorders, Text Revision 34 (4th ed. 2000). Gayheart’s attitude was found to be depressed, his affect negative, and his eye contact poor during 30 percent of the evaluation, but his tone of voice was pleasant. His “response style” was “suboptimal,” causing Dr. Flexman to judge the reliability of the responses to be poor, suggesting “moderate malingering.”
Dr. Flexman concluded that Gayheart was moderately impaired in his ability to interact appropriately with the public and coworkers, and in his ability to respond appropriately to work pressures and changes in a work setting. All other limitations, such as concentration, sustaining attention, following simple instructions, and interacting with supervisors, were noted as slight.
Dr. Chiappone examined Gayheart in June 2008. He diagnosed Gayheart with a pain disorder due both to psychological factors and to Gayheart’s general medical condition (depression), with panic disorder without agoraphobia, and with borderline intellectual functioning. Dr. Chiappone assigned Gayheart a GAF score of 45 based upon the latter’s symptoms, but noted a functional level of 51 because Gayheart could “do some basic tasks on a limited basis in the confines of home.” Gayheart appeared to be depressed, anxious, and nervous, and “[h]e did not appear to be malingering.” Dr. Chiappone noticed a tremor and that Gayheart had his head in his hands a lot. Gayheart said during the evaluation that he suffered anxiety attacks weekly that were “very severe and feel like a heart attack.”
Dr. Chiappone opined that Gayheart was moderately impaired in his ability to remember one- and two-step job instructions and was “at least moderately impaired in his ability to maintain concentration and attention.” Likewise, the doctor noted moderate impairment in Gayheart’s ability to persist with a task or to tolerate stress. He also noted marked impairment in Gayheart’s ability to relate to coworkers, supervisors, and the public, to respond to usual work situations and changes in a routine work setting, and to understand, remember, and carry out complex instructions.
The Bureau of Disability Determination also requested that Dr. David Demuth review the medical evidence in the case. Dr. Demuth performed this review not long after Dr. Flexman’s February 2006 examination. He reported that Gayheart exhibited no generalized persistent anxiety, depressive syndrome, or substance-abuse disorder, and that Gayheart required no restrictions on daily living activities. Gayheart was not found to be significantly limited in his ability to understand, remember, and carry out detailed instructions, to work in coordination with or proximity to others without being distracted, to respond to changes in the work setting, or to travel in unfamiliar places or use public transportation. Moderate restrictions were noted in social functioning.
3. Dr. Onady’s letter
Between the second and third hearings, Dr. Onady wrote a letter at the request of Gayheart’s counsel. She explained that Gayheart “still suffers from frequent panic attacks, irritability, has a low frustration level, and extreme social anxiety, even with medications.” In her opinion, Gayh-eart was “not capable of being dependable and reliable to the extent that he would be able to work on a normal basis.” She acknowledged that “his alcohol consumption does not help his depression,” but noted that alcohol was “not the root of the problem, and the severity of his symptoms
as well as the frequency would still persist if he were no longer drinking alcohol.”
4. The ALJ’s decision and the appeal to the district court
After reviewing the evidence in the record and having heard the testimony at the three hearings, the ALJ denied Gayheart’s claim for DIB. The ALJ concluded that Gayheart suffered severe impairments, but that the combination of those impairments did not meet or medically equal any of the listed impairments in 20 C.F.R. Part 404, Subpart P, Appendix 1, nor precluded Gayheart from being able to perform jobs that exist in significant numbers in the national economy. Gayheart’s requested review of the ALJ’s decision by the Appeals Council was denied, making the Commissioner’s denial of DIB final. He then sought review of that decision in the federal district court as allowed by 42 U.S.C. § 405(g).
The assigned magistrate judge issued a Report and Recommendation concluding that the ALJ’s decision was not based on substantial evidence because it failed to accord proper weight to the opinions of Gayheart’s treating psychiatrist and therapists. He farther concluded that the “proof of disability is overwhelming” and that the case should be remanded to the Commissioner for an immediate award of benefits rather than for reconsideration.
But the district court sustained the Commissioner’s objections to the Report and Recommendation, finding that the ALJ gave adequate reasons for discounting the opinions of Gayheart’s treating psychiatrist and therapists and that the ALJ applied the correct legal criteria in reaching a decision that is supported by substantial evidence. The court therefore affirmed the Commissioner’s denial of DIB and ordered the case terminated. This timely appeal followed.
II. ANALYSIS
A. Standard of review
We review de novo a district court’s decision regarding Social Security disability benefits.
Cole v. Astrue,
661 F.3d 931, 937 (6th Cir.2011). “However, that review is limited to determining whether the Commissioner’s decision is supported by substantial evidence and was made pursuant to proper legal standards.”
Id.
(internal quotation marks omitted). Substantial evidence is “such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.”
Heston v. Comm’r of Soc. Sec.,
245 F.3d 528, 534 (6th Cir.2001) (internal quotation marks omitted). A reviewing court will affirm the Commissioner’s decision if it is based on substantial evidence, even if substantial evidence would also have supported the opposite conclusion.
Colvin v. Barnhart,
475 F.3d 727, 730 (6th Cir.2007). But “[a]n ALJ’s failure to follow agency rules and regulations denotes a lack of substantial evidence, even where the conclusion of the ALJ may be justified based upon the record.”
Cole,
661 F.3d at 937 (internal quotation marks omitted).
B. Weighing medical opinions
The Commissioner’s regulations provide a five-step process for the evaluation of disabilities.
See
20 C.F.R. § 404.1520. This court has summarized the five steps as follows:
1. If claimant is doing substantial gainful activity, he is not disabled.
2. If claimant is not doing substantial gainful activity, his impairment must be severe before he can be found to be disabled.
3. If claimant is not doing substantial gainful activity and is suffering from a severe impairment that has lasted or is
expected to last for a continuous period of at least twelve months, and his impairment meets or equals a listed impairment, claimant is presumed disabled without further inquiry.
4. If claimant’s impairment does not prevent him from doing his past relevant work, he is not disabled.
5. Even if claimant’s impairment does prevent him from doing his past relevant work, if other work exists in the national economy that accommodates his residual functional capacity and vocational factors (age, education, skills, etc.), he is not disabled.
Walters v. Comm’r of Soc. Sec.,
127 F.3d 525, 529 (6th Cir.1997).
At step one of this process, the Commissioner does not dispute that Gayheart has had no “substantial gainful activity” since he ceased working in September 2005 on account of his alleged disability. Nor does the Commissioner dispute at step two that Gayheart has the following severe impairments: headaches, an anxiety disorder and depression, borderline intellectual functioning, and alcohol abuse.
Rather, this case turns on whether the ALJ appropriately deferred to the opinions of Drs. Buban and Flexman over the opinion of Dr. Onady in: (1) determining at step three that the combination of Gayheart’s impairments do not meet or medically equal the criteria of an impairment listed in 20 C.F.R. Part 404, Subpart P, Appendix 1; and (2) determining the limitations on Gayheart’s residual functioning capacity (RFC) for use in steps four and five. Had the ALJ found at step three that an impairment listing was met, Gayheart would have been deemed disabled.
See
20 C.F.R. § 404.1520(a)(4)(iii). Incorporating additional limitations into the RFC, as Dr. Onady’s opinion called for, would likewise have resulted in a finding that Gayheart was disabled under step five because he would be deemed unable to work.
See id.
§ 404.1520(a)(4)(v).
2.
Standards for weighing medical opinions
“The Commissioner has elected to impose certain standards on the treatment of medical source evidence.”
Cole,
661 F.3d at 937. These standards, set forth in administrative regulations, describe (1) the various types of evidence that the Commissioner will consider, 20 C.F.R. § 404.1512; (2) who can provide evidence to establish an impairment, 20 C.F.R. § 404.1513; and (3) how that evidence will be evaluated, 20 C.F.R. § 404.1520b. Such evidence may contain medical opinions, which “are statements from physicians and psychologists ... that reflect judgments about the nature and severity of [a claimant’s] impairment(s), including [] symptoms, diagnosis and prognosis,” physical and mental restrictions, and what the claimant can still do despite his or her impairments. 20 C.F.R. § 404.1527(a)(2). Medical opinions are to be weighed by the process set forth in 20 C.F.R. § 404.1527(c).
As a general matter, an opinion from a medical source who has examined a claimant is given more weight than that from a source who has not performed an examination (a “nonexamining source”),
id.
§ 404.1502, 404.1527(c)(1), and an opinion from a medical source who regularly treats the claimant (a “treating source”) is afforded more weight than that from a source who has examined the claimant but does not have an ongoing treatment relationship (a “nontreating source”),
id.
§ 404.1502, 404.1527(c)(2). In other words, “[t]he regulations provide progressively more rigorous tests for weighing opinions as the ties between the source of the opinion and the individual become weaker.” Soc. Sec. Rul. No. 96-6p, 1996 WL 374180, at *2 (Soc. Sec. Admin. July 2,1996).
The source of the opinion therefore dictates the process by which the Commissioner accords it weight. Treating-source opinions must be given “controlling weight” if two conditions are met: (1) the opinion “is well-supported by medically acceptable clinical and laboratory diagnostic techniques”; and (2) the opinion “is not inconsistent with the other substantial evidence in [the] case record.” 20 C.F.R. § 404.1527(c)(2). If the Commissioner does not give a treating-source opinion controlling weight, then the opinion is weighed based on the length, frequency, nature, and extent of the treatment relationship, id, as well as the treating source’s area of specialty and the degree to which the opinion is consistent with the record as a whole and is supported by relevant evidence,
id.
§ 404.1527(c)(2)-(6).
The Commissioner is required to provide “good reasons” for discounting the weight given to a treating-source opinion.
Id.
§ 404.1527(c)(2). These reasons must be “supported by the evidence in the case record, and must be sufficiently specific to make clear to any subsequent reviewers the weight the adjudicator gave to the treating source’s medical opinion and the reasons for that weight.” Soc. Sec. Rul. No. 96-2p, 1996 WL 374188, at *5 (Soc. Sec. Admin. July 2, 1996). This procedural requirement “ensures that the ALJ applies the treating physician rule and permits meaningful review of the ALJ’s application of the rule.”
Wilson v. Comm’r of Soc. Sec.,
378 F.3d 541, 544 (6th Cir.2004).
On the other hand, opinions from non-treating and nonexamining sources are never assessed for “controlling weight.” The Commissioner instead weighs these opinions based on the examining relationship (or lack thereof), specialization, consistency, and supportability, but only if a treating-source opinion is not deemed controlling. 20 C.F.R. § 404.1527(c). Other factors “which tend to support or contradict the opinion” may be considered in assessing any type of medical opinion.
Id.
§ 404.1527(c)(6).
2. The ALJ failed to weigh the medical opinions according to 20 C.F.R. § 404.1527
i. Dr. Onady’s opinions and treatment notes
The ALJ accorded “little weight” to Dr. Onady’s opinions and instead relied on the medical opinions of Dr. Buban, Dr. Flexman, and, to a lesser degree, Dr. Chiappone. Dr. Onady’s opinions, according to the ALJ, fail both prongs of the test for controlling weight — they “are not well-supported by any objective findings” and are “inconsistent with other credible evidence.” (Given that the ALJ correctly noted the controlling-weight standard earlier in his decision, we will assume that he meant “other substantial evidence” rather than “other credible evidence.”
See
20 C.F.R. § 404.1527(c)(2).) But the ALJ did not provide “good reasons” for why Dr. Onady’s opinions fail to meet either prong of this test.
To be sure, the ALJ discusses the frequency and nature of Dr. Onady’s treatment relationship with Gayheart, as well as alleged internal inconsistencies between the doctor’s opinions and portions of her reports. But these factors are properly applied only after the ALJ has determined that a treating-source opinion will not be given controlling weight.
See
20 C.F.R. § 404.1527(c)(2) (listing seven specific factors to be applied when a treating-source opinion is not given controlling weight, including the general consistency of the opinion with the record as a whole). The ALJ also concludes that Dr. Onady’s opinions “seem[ ] to have minimized the impact of the claimant’s alcohol abuse.” Putting
aside for the moment whether the ALJ had a proper basis for his conclusion regarding such impact, the analysis does not explain to which aspect of the controlling-weight test this critique is relevant.
The failure to provide “good reasons” for not giving Dr. Onady’s opinions controlling weight hinders a meaningful review of whether the ALJ properly applied the treating-physician rule that is at the heart of this regulation.
See Wilson,
378 F.3d at 544. For example, the conclusion that Dr. Onady’s opinions “are not well-supported by any objective findings” is ambiguous. One cannot determine whether the purported problem is that the opinions rely on findings that are not objective (i.e., that are not the result of medically acceptable clinical and laboratory diagnostic techniques,
see
20 C.F.R. § 404.1527(c)(2)), or that the findings are sufficiently objective but do not support the content of the opinions.
Similarly, the ALJ does not identify the substantial evidence that is purportedly inconsistent with Dr. Onady’s opinions. Surely the conflicting substantial evidence must consist of more than the medical opinions of the nontreating and nonexam-ining doctors. Otherwise the treating-physician rule would have no practical force because the treating source’s opinion would have controlling weight only when the other sources agreed with that opinion. Such a rule would turn on its head the regulation’s presumption of giving greater weight to treating sources because the weight of such sources would hinge on their consistency with nontreating, nonex-amining sources.
As noted above, the ALJ provided a modicum of reasoning that is relevant to how Dr. Onady’s opinions should be weighed
after
determining that they were not controlling, but even this reasoning fails to justify giving those opinions “little weight.” The ALJ highlighted portions of Dr. Onady’s clinic notes and reports to suggest that her “gloomy” and “pessimistic” opinions were “inconsistent with her other ratings of [Gayheart’s] functional capacity.” For example, the ALJ refers to clinic notes reflecting that Gayheart was looking forward to being outside and planned to go to a store to buy a new lawnmower blade. The record, according to the ALJ, is clear that Gayheart’s “alleged anxiety has not prevented him from leaving home, driving, keeping medical appointments, visiting friends and neighbors, shopping with his wife, and attending three hearings.” Although not clearly stated, the apparent implication is that these activities are inconsistent with the social and daily living restrictions noted in Dr. Onady’s opinions.
But the ALJ does not contend, and the record does not suggest, that Gayheart could do any of these activities on
a sustained basis,
which is how the functional limitations of mental impairments are to be assessed.
See
20 C.F.R. § 404.1520a(c)(2); 20 C.F.R. Part 404, Subpart P, Appendix 1, at 12.00 (“Social functioning refers to your capacity to interact independently, appropriately, effectively, and on a sustained basis with other individuals.”). Gayheart’s ability to visit his aunt and uncle, who live on his street, and to receive occasional visits from his neighbor does not undermine Dr. Onady’s opinion that Gayheart’s ability to interact independently and appropriately with others on a sustained basis is markedly impaired. The same is true of his ability to accompany his wife on grocery-shopping trips once per month. These activities would be relevant if they suggested that Gayheart could do something on a sustained basis that is inconsistent with Dr. Onady’s opinions. But they do not.
Furthermore, many of these examples are either taken out of context or are offset by other examples in the record. Although it is accurate to say that Gayheart can drive, for instance, the record also shows that driving triggers his anxiety and that he thus relies on his wife to do most of the driving. Gayheart is likewise able to travel, but the record indicates that he generally avoids travel and that the one out-of-state trip he took in the three years prior to his testimony resulted in a panic attack. Nothing in the record suggests that he has left the house independently and on a sustained basis. Gayheart testified, in fact, that he had not gone to a store by himself in more than a year. We therefore conclude that the ALJ’s focus on isolated pieces of the record is an insufficient basis for giving Dr. Onady’s opinions little weight under 20 C.F.R. § 404.1527(c).
ii.
Opinions and treatment notes of Jackie Thompson and Jennifer Fenske-Doyle
The ALJ also gave “little weight” to the opinion of Gayheart’s first therapist, Jackie Thompson, and made no mention at all of Gayheart’s second therapist, Jennifer Fenske-Doyle, who saw Gayheart on a regular basis for 13 months. Although these therapists do not qualify as “acceptable medical sources” under the regulations, an ALJ must consider all relevant evidence in the case record. Soc. Sec. Rul. No. 06-03p, 2006 WL 2329939, at *4 (Soc.Sec.Admin. Aug. 9, 2006). The factors set forth in 20 C.F.R. § 404.1527, which under the regulation apply only to medical opinions from acceptable medical sources, nevertheless “represent basic principles that apply to the consideration of all opinions from medical sources ... who have seen the individual in their professional capacity.”
Id.
The ALJ mentioned three primary reasons for according Thompson’s opinion little weight. First, Thompson listed “unable to go to public places” as a “significant restriction of daily activities.” The ALJ notes that in fact Gayheart “has been able to go out into public places, notwithstanding allegations about social anxiety and panic.” But similar to his weighing of Dr. Onady’s opinions, the ALJ does not cite record evidence showing that Gayheart is able to go out into public places on a sustained basis. In other words, the fact that Gayheart can sometimes go into public places (i.e., his once-a-month shopping trips with his wife) does not contradict Thompson’s opinion that Gayheart’s general inability to be in public places significantly restricts what he can do on a day-to-day basis.
The ALJ’s second and third reasons are more well-founded. His second reason is that Thompson’s opinion relies on Gayheart’s subjective claims rather than on detailed clinical data, which goes to the supportability of the opinion.
See
20 C.F.R. § 404.1527(c)(3). The third reason is that Thompson saw Gayheart for only five months. Length of treatment is also a relevant factor.
See id.
§ 404.1527(e)(2)(i). These reasons sufficiently support the little weight that the ALJ gave Thompson’s opinion, particularly given the secondary importance of that opinion relative to Dr. Onady’s opinions.
In contrast, the ALJ’s decision does not discuss or even acknowledge the extensive treatment notes from Fenske-Doyle. The district court found that the ALJ “presumably concluded Fenske-Doyle also lacked the medical training to properly address Gayheart’s reported problems” and that her “findings were represented by Dr. Onady” because she was a therapist at Dr. Onady’s medical-
practice center. But nothing in the ALJ’s decision supports the court’s rationalizations of Fenske-Doyle’s complete absence from the ALJ’s evaluation of the record. Her treatment notes were relevant evidence,
see
20 C.F.R. § 404.1512, and lent significant support to the opinions of Dr. Onady. Because the ALJ discounted Dr. Onady’s opinions largely due to their alleged lack of consistency with the record as a whole, some explanation should have been given for ignoring this large portion of the record.
iii.
Opinions of Dr. Buban, Dr. Flexman, and Dr. Demuth
The foregoing analysis of the ALJ’s explanation for giving Dr. Onady’s opinions little weight reflects the rigorous scrutiny he applied to those opinions. His failure to apply the same level of scrutiny to the opinions of the consultative doctors on which he relied, let alone the greater scrutiny of such sources called for by 20 C.F.R. § 404.1527, further demonstrates that his assessment of Dr. Onady’s opinions failed to abide by the Commissioner’s regulations and therefore calls into question the ALJ’s analysis.
See Cole v.
As
true,
661 F.3d 931, 937 (6th Cir.2011) (“An ALJ’s failure to follow agency rules and regulations denotes a lack of substantial evidence.” (internal quotation marks omitted)).
The ALJ’s decision provides no indication that he applied the factors set out in § 404.1527(c) — supportability, consistency, specialization- — when weighing the consultative doctors’ opinions, other than to note Dr. Buban’s “well-reasoned analysis of the evidence” and the consistency of her conclusions with the other consultative doctors’ assessments. Although the ALJ was quite critical of the alleged inconsistencies between Dr. Onady’s opinions and other record evidence, his decision does not acknowledge equivalent inconsistencies in the opinions of the consultative doctors. A more rigorous scrutiny of the treating-source opinion than the nontreating and nonexamining opinions is precisely the inverse of the analysis that the regulation requires.
See
20 C.F.R. § 404.1527(c); Soc. Sec. Rul. No. 96-6p, 1996 WL 374180, at *2 (Soc. Sec. Admin. July 2,1996).
The starkest example of the ALJ’s failure to scrutinize the opinions of the consultative doctors is reflected in his assessment of Dr. Buban’s testimony. No mention is made of the doctor’s unequivocal testimony at the second hearing that, based on her file review, Gayheart met Listing 12.06 for anxiety disorders. Instead, the ALJ discussed her testimony only from the third hearing that “there was insufficient evidence to show that a mental listing was met or equaled.” The ALJ appears to have either overlooked or accepted this stark change in opinion regarding one of the fundamental issues in the ease without any explanation for that change.
Similarly, Dr. Flexman assigned Gayh-eart a GAF score significantly higher than that given by any other professional who examined Gayheart either before or after Dr. Flexman’s examination. And Dr. De-muth’s opinion that Gayheart had no restrictions on daily living activities and no indications of depressive syndrome conflicts with overwhelming record evidence to the contrary. Yet the ALJ made no mention of these inconsistencies, which appear to be just as relevant, if not more so, than the inconsistencies discussed with regard to Dr. Onady’s opinions.
To be sure, a properly balanced analysis might allow the Commissioner to ultimately defer more to the opinions of consultative doctors than to those of treating physicians. Soc. Sec. Rul. No. 96-6p, 1996 WL 374180, at *3 (“In appropriate circum
stances, opinions from ... psychological consultants ... may be entitled to greater weight than the opinions of treating or examining sources.”)- But the regulations do not allow the application of greater scrutiny to a treating-source opinion as a means to justify giving such an opinion little weight. Indeed, they call for just the opposite. The alleged inconsistencies in Dr. Onady’s opinions, therefore, cannot constitute “good reasons” for affording them little weight when more flagrant inconsistencies go unquestioned in the medical opinions to which the ALJ deferred.
In the end, a proper analysis of the record might not support giving controlling weight to the opinions of Dr. Onady. But this circuit “has made clear that [it] do[es] not hesitate to remand when the Commissioner has not provided good reasons for the weight given to a treating physician’s opinion.”
Cole,
661 F.3d at 939 (internal quotation marks omitted). And even if Dr. Onady’s opinions do not warrant controlling weight, they still must be weighed as the regulations prescribe, with no greater scrutiny being applied to her opinions than to those of the nontreating and nonexamining sources. Moreover, the failure to provide “good reasons” was not a harmless error in the present case because the Commissioner cannot show that, despite his failure to comply with the terms of 20 C.F.R. § 404.1527(c)(2), he has otherwise met the regulation’s goal.
See Wilson v. Comm’r of Soc. Sec.,
378 F.3d 541, 547 (6th Cir.2004).
C. Gayheart’s alcohol abuse
The preceding analysis provides a sufficient basis for remanding this case to the Commissioner. But we will briefly address Gayheart’s argument that the ALJ erred in assessing the materiality of his alcohol abuse so as to avoid an improper evaluation of this issue on remand.
“An individual shall not be considered to be disabled ... if alcoholism or drug addiction would ... be a contributing factor material to the Commissioner’s determination that the individual is disabled.” 42 U.S.C. § 423(d)(2)(C). The Commissioner’s regulation explains that “[t]he key factor” in determining whether drug or alcohol abuse is material in a given case is whether the claimant would still be disabled if he or she stopped using drugs or alcohol. 20 C.F.R. § 404.1535. Substance abuse is not considered until the Commissioner first makes a finding that a claimant is disabled.
See id.
(“If we find that you are disabled ... we must determine whether your drug addiction or alcoholism is a contributing factor material to the determination of disability.”).
The ALJ did not conclude that alcohol abuse was material to Gayheart’s disability. Rather, he never found Gayheart to be disabled at all, thus precluding the need to assess the materiality of alcohol abuse. But alcohol abuse was a factor in the ALJ’s decisionmaking process, particularly in his weighing of Dr. Onady’s opinions.
The ALJ first discussed Gayheart’s alcohol abuse in his findings of severe mental impairments. He concluded that “it is evident from treatment notes that the claimant’s excessive consumption of alcohol has coincided with his emotional problems and very likely has been an aggravating factor.” Alcohol abuse is then listed as one of Gayheart’s severe mental impairments.
The ALJ next discussed alcohol abuse in the context of explaining the “little weight” that he gave to the opinions of Dr. Onady, who he found had “minimized the impact of the claimant’s alcohol abuse.” He cited Dr. Buban’s observation that Gayheart’s alcohol abuse and “emotional problems” were closely intertwined. Even considering the effects of Gayheart’s alcohol abuse
and other severe mental impairments together, the ALJ concluded that Gayheart “appears capable of carrying out at least low stress work with limited interactions and concentration as described in the statement of his RFC.” This assessment conforms to 20 C.F.R. § 404.1535 because the ALJ made clear that he found Gayh-eart not disabled without first attempting to separate limitations solely attributable to alcohol abuse. But the ALJ proceeded to hypothetically determine the materiality of the alcohol abuse by writing the following:
However, if it is argued that [Gayheart] is incapable of doing even these limited kinds of duties, such incapacity could only be attributed to the effect of his excessive drinking. But Public Law 104-120 clearly precludes entitlement based on alcohol or drug addictions as a material factor to the determination of disability.
To the degree that this hypothetical finding was offered to justify according Dr. Onady’s opinions little weight, such a finding is unpersuasive because alcohol abuse is not a factor to be considered in determining the weight to be given to a treating-source opinion.
See
20 C.F.R. § 404.1527(c). The finding is also unpersuasive because it is not supported by the present record. Dr. Onady wrote in a letter that alcohol was “not the root of the problem, and the severity of [Gayheart’s] symptoms as well as the frequency would still persist if he were no longer drinking alcohol.” And even Dr. Buban conceded that she “really couldn’t say” whether the alcohol abuse would be material to any finding of disability. In sum, if Gayheart is found on remand to be disabled, the materiality of his alcohol abuse needs to be reevaluated with proper attention to the record.
III. CONCLUSION
For all of the reasons set forth above, we REVERSE the judgment of the district court and REMAND with instructions that the case be returned to the Social Security Administration for reconsideration.