PAUL L. MALONEY, Chief District Judge.

For the reasons that follow, we REVERSE the district court, vacate the ALJ's decision, and remand this case to the ALJ for further consideration. On remand, the ALJ must provide a much fuller opinion which specifies the weight given to each medical source's opinion — particularly treating psychologist Dr. Dana Watts, Ph.D., and treating physician Dr. Thomas Thysseril, M.D. — adequately explains the relative "weight" determinations, reflects a careful consideration of the medical evidence as a whole, and bases its residual functional capacity (RFC) assessment on such expressed consideration.

I.

Juliette Karger ("Karger") was born in 1982, was sixteen years old at the time of the alleged onset of disability in 1998, and was twenty-four years old at the time of the ALJ hearing in 2006. See Tr 53. Karger earned a high-school general equivalency degree ("GED"), Tr 65 and 89, but she has no past relevant work experience, Tr 20 and 56-57.

Karger was admitted to the hospital for psychiatric emergency on December 15, 1998 and remained there until December 23, 1998. See Tr 124-159. Dr. DeoGracias, M.D., diagnosed bipolar disorder-most recent episode of depression without psychotic features, "rule out" attention deficit disorder, and possible borderline personality disorder, and assigned a Global Assessment of Functioning ("GAF") score of only ten. See Tr 124-26. During this hospitalization, Dr. DeoGracias noted that Karger had increasingly been cutting herself and contemplating suicide, and she noted psychomotor retardation; anxious, sad and depressed mood; a slightly flat affect; a history of visual hallucinations; thoughts of suicide, and poor concentration and judgment. See Tr 124-26.

Over three years later, on January 8, 2002, Dana Watts, Ph.D., responded to a state disability questionnaire, opining that Karger had good or very good ability to make occupational adjustments, good ability to make performance adjustments, very good or good ability to make personal-social adjustments, and only a fair ability to persist in work activity. See Tr 164-65. Dr. Watts's January 2002 report opined that Karger had the intellectual ability to perform, but "emotional stress, conflict, high pressure situation, critical supervisor, and physically demanding job would interfere with her ability to perform job tasks." Tr 164-65. Dr. Watts assigned a GAF score of 39, which indicates some impairment in reality testing or communication, or major impairments in several areas such as work, school, family relations, judgment, thinking, or mood. See DSM at 32.

Four months later, on May 6, 2002, Karger entered a residential treatment program at Gould Farm and was diagnosed with bipolar disorder II and rapid cycling, and received a GAF score of 45, Tr 213-14, a score which indicates serious symptoms or serious impairment in social, occupational, or school functioning, see DSM at 32. During the Gould Farm residential program, Karger received care from psychiatrist Dr. Stuart Bartle, M-.D., and supervision from Tamsin Trelawny-Ross, Tr 213-14. Dr. Bartle conducted a psychiatric evaluation on May 9, 2002, noting that Karger had been hospitalized at age 16 for depression and had had obsessive-compulsive disorder symptoms before that; had OCD symptoms such as picking at her face, rumination, obsessive thoughts, non-repetitive checking, and worrying a great deal about something not being right; and had no experience with drugs except marijuana on one occasion. See Tr 210-11. Dr. Bartle diagnosed bipolar disorder II, rapid cycling, ADD, and OCD, and assigned GAF 45, see Tr 211.

While Karger was in the Gould Farm residential program, treating source Dr. Watts completed two questionnaires for the State, see Tr 175-182. Dr. Watts' May 2002 questionnaire stated that although Karger's mood and affect were normal at that time due to psychotropic medication, her "obsessive ideation and compulsive behavior increase when anxious and under stress", Tr 175. Karger again reported no drug or alcohol abuse, Tr 175. Dr. Watts diagnosed bipolar disorder, OCD, ADD, and circadian rhythm sleep disorder, Tr 176. Watts noted that Karger's attention and concentration diminished during bipolar episodes, which were currently controlled with medication, Tr 176. Watts further opined that Karger was unable to carry out age-appropriate tasks such as waking in the morning, driving a car, going to school, and budgeting and managing her finances, and that she "may experience increased anxiety and attention problems, especially if the work situation involves conflict." Tr 176 and 179-80. Watts noted that Karger's parents had to handle her financial affairs, give her a ride when she needed one, and help with transportation, shelter and independence skills, because she had periods when she was distracted and unfocused and had trouble completing tasks in a timely manner or at all, particularly when feeling anxious, Tr 180-81. Dr. Watts also found that Karger's social functioning had "drastic[ally]" decreased, leaving her with a decreased social network outside her family, Tr 178.

On July 28, 2002, Karger's residential treatment providers, Dr. Bartle and Ms. Trelawny-Ross, completed a questionnaire for the State disability agency. Noting that Karger had left college because of her symptoms and had been unable to work since then, Dr. Bartle and Ms. Trelawny-Ross stated that Karger would remain at Gould Farm for the foreseeable future because her OCD symptoms, depression, anxiety, and low self-esteem were interfering with her daily living activities, her ability to work, and her ability to complete college courses. She was unable to become and remain motivated, and was unable to complete tasks such as reading, studying, or caring for herself, in a timely manner. Tr 183-84. Dr. Bartle and Ms. Trelawny-Ross opined that Karger could sustain concentration and attention for extended periods "with support, e.g., while working in a supportive work program," Tr 184, which she was doing thirty hours per week, Tr 186. Bartle and Trelawny-Ross opined that Karger had good social skills but that her hypersensitivity to perceived criticism, which was caused by her depression and anxiety, inhibited her relationships. They believed that Karger could use public transportation if she had support in planning, becoming familiar with the transit system, and first using the system. Tr 184. Although Bartle and Trelawny-Ross found that the routine stress of changes or critical feedback had a "high impact" on Karger, they concluded that her prognosis was "fair/good", Tr 184-85. The Bartle/Trelawny-Ross questionnaire opined that Karger

can not hold a competitive job at this point. She needs more work around the depression, OCD, and anxiety. ADD would also play a part in her inability to work, as it would become more eviden[t] when her other symptoms are under control.

Tr 186. Three days later, on August 1, 2002, Dr. Bartle reported that although Karger was generally doing well in treatment and was hoping to be admitted to an OCD program at McLean Hospital, her habit of picking at her face, which had receded somewhat, was now increasing slightly. See Tr 212.

The following week, on August 7, 2002, state reviewing psychologist M.E. Menken, Ph.D., completed a Psychiatric Review Technique Form which found that Karger met Listings 12.06a and 12.06b and had impairments under 12.02, 12.04 and 12.06 which resulted in marked limitations in maintaining social functioning and concentration, persistence or pace. See Tr 187 and 197.

On August 15, 2002, Karger's father completed a third-party disability questionnaire. After noting Karger was in a residential treatment program, Karger's father stated that she was able to complete housekeeping tasks, washed dishes 2-3 times per week, shopped weekly for personal items, watched about half an hour of television each week, visited her family in alternate months for 1-2 weeks at a time, enjoyed going to movies and restaurants, and played the piano about half an hour each day. On the other hand, Karger's father stated that she had difficulty waking without assistance, had difficulty concentrating and difficulty reading because her OCD constantly required her to reread, and suffered increased anxiety and discomfort with any changes in routine. See Tr 201-206.

Upon Karger's discharge from Gould Farm residential treatment program on December 11, 2002, Dr. Bartle reported that Karger's OCD symptom had waxed and waned over her seven months there as she underwent "many med changes"; her picking had increased on at least two occasions; and depression and tiredness remained a problem, Tr 209. Dr. Bartle's assistant, Ms. Trelawny-Ross, reported that, on the positive side of the ledger, Karger had made "significant progress" on managing her OCD symptoms and her sleep disorder during her stay at Gould Farm, Tr 207, the latter with the aid of multiple alarm clocks and a "bed shaker"; she worked well in teams in the supportive work setting, Tr 207; and she "did well overcoming many of her obsessions and finding ways to control her anxiety such that these interfered less with her work in the long run", Tr 208. On the negative side of the ledger, Ms. Trelawny-Ross observed that Karger's depression had worsened toward the end of her stay; her motivation to work and be generally involved in the community had declined; and her OCD, depression and physical complaints interfered with her abilities in numerous ways: obsessive rituals and checking prevented her from getting ready for work or completing tasks in a timely manner; she was struggling long-term to complete a workday, let alone a 30-hour work-week, in the face of her depression, sleep disorder, and frequent physical sickness; and she still experienced depression triggered by disappointment and characterized by isolation, decreased energy, increased sleep, decreased motivation to work or be social, increased self-absorption, and decreased ability to see another point of view or to realize how her actions affect others. See Tr 207-208.

On January 13, 2003, about a month after leaving Gould Farm, Karger was admitted to McLean Hospital's OCD Institute under the care of Dr. Darin Dougherty M.D., Perry Merlin M.S.W., and behavioral therapist Leslie Shapiro. See Tr 215-230. Dr. Dougherty initially diagnosed Karger with OCD and bipolar disorder and assigned a GAF score of 41, which indicates serious symptoms. See Tr 216 and 230. Two days later, on January 15, 2003, behavioral therapist Shapiro identified the following obsessions and compulsions that Karger needed to address:

— perfectionism/incompletion

— some contamination

— complexion, hair;

— herself or others losing her personal items

— riding in or driving a car

— being touched by bare feet

— hearing sounds such as silverware scratching a plate or hearing people with "dry mouths" talk

— fear of falling down the stairs

— fear of seeing moldy food or of eating food which is a few days old or past its expiration date

— making sure that she has packed everything for a trip and then checking the items during the trip

— picking her skin

— re-reading, writing with perfect penmanship, and concern with grammar

— checking electrical plugs

— ordering and arranging personal items in her room

— going down stairs one step at a time

— watching or looking at information about diseases

Tr 217-218. On January 23, 2003, ten days after Karger was admitted to the McLean OCD Institute, Dr. Dougherty noted that she was having difficulty and that there had been no change in her symptoms, Tr 228. On February 4, 2003, Dr. Dougherty noted that Karger was complying with her treatment plan and performing more difficult steps, Tr 224, and she was discharged to her parents' home on February 21, 2003 with an "improved" condition, Tr 215.

Karger began treatment with a county counseling center on April 4, 2003, and after the initial screening, therapist Alicia Thomas diagnosed bipolar disorder, depression, and OCD, but assigned a GAF score of 60, see Tr 239, at the high end of the 51-60 range which indicates moderate symptoms or moderate difficulty in social, occupational, or school functioning, see DSM at 32.

At the request of a State agency, John A. Comley, Psy.D., evaluated Karger on June 5, 2003, diagnosing ADHD — predominantly inattentive type, bipolar disorder I-rapid cycling type, OCD, and histrionic personality traits, and finding moderate limitation in eight categories of work-related functioning, marked limitation in five categories, and extreme limitation in five categories, calling her "psychologically disabled" and unemployable for 12 months or more, see Tr 243-45. Comley recommended continued psychiatric evaluation, psychological treatment, and vocational rehabilitation services, see Tr 244.

Also at the request of the state disability agency, Curt S. Ickes, Ph.D., evaluated Karger on October 29, 2003, diagnosing bipolar disorder, depression, OCD, and ADHD, and assigning a GAF score of 55, see Tr 274, the midpoint of the 51-60 range which indicates moderate symptoms or moderate difficulty in social, occupational, or school functioning, see DSM at 32. On the neutral or positive side of the ledger, Ickes found Karger to have a normal range of affect, full orientation, fair insight and judgment, estimated low-to-average intelligence, an intact memory, and no impairment in the ability to understand, remember, and follow instructions, see Tr 271-74. On the negative side of the ledger, Ickes found that Karger was moderately limited in her ability to relate well to others, including co-workers and supervisors; moderately impaired in her ability to maintain concentration, persistence and pace for the performance of simple tasks; and moderately impaired in the ability to withstand the normal day-to-day stress and pressures associated with work, see Tr 274.

Also at the request of the State disability agency, John Malinky, Ph.D., completed a Psychiatric Review Technique Form and a Mental RFC Assessment, see Tr 281-296. Malinky found that Karger was mildly restricted in her daily-living activities and moderately limited in a number of areas: maintaining social functioning; maintaining concentration, persistence or pace; understanding, remembering and carrying out detailed instructions; maintaining attention and concentration for extended periods; performing activities on a schedule; maintaining regular attendance and punctuality; completing a normal workday or workweek without interruption from psychologically-based symptoms; performing at a consistent pace without unreasonably frequent or unreasonably long rest periods; and accepting criticism and instructions from supervisors, see Tr 294-95. Malinky found Karger to be markedly limited in one area, the ability to interact appropriately with the general public, Tr 295. Ultimately, Malinky opined that Karger retained the RFC to "complete simple repetitive tasks that do not require prolonged periods of attention[,] in a low[-]stress environment with decreased contact with others." Tr 296.

At the request of the State disability agency, treating source Dana Watts, Ph. D., completed a Mental Assessment of Ability to Do Work-Related Activities, diagnosing OCD, bipolar disorder, and depression, and assigning her a GAF score of 35, see Tr 276-77 and DSM at 32.

On the positive side of the ledger, Dr. Watts found that Karger had unlimited or very good abilities to follow work rules, interact with supervisors, relate to coworkers, and maintain her personal appearance, and a good ability to use work judgment, and remember and carry out detailed but not complex job instructions and tasks, or maintain attention or concentration; although anxiety could interfere with those abilities, see Tr 276-78. Dr. Watts found that because of her chronic emotional problems, Karger had only a fair ability to behave in an emotionally stable manner, and poor to no ability to deal with work stresses, function independently, relate predictably in social situations, demonstrate reliability, or persist in work activity, see Tr 276-78. Watts opined that although Karger was "a very bright young woman", her "chronic emotional difficulties have had a severe impact on her ability to cope with age-appropriate demands" and "have interfered with her ability to master independent living skills", Tr 276.

Significantly, Karger has not alleged that she personally interacted with, or was ever examined by, Watts after 2002. Thus, absent contrary evidence from Karger, it appears that Watts's December 2003 opinion was based on Watts's interactions and notes from at least a full year earlier. Therefore, Watts was no longer Karger's treating psychologist by the time of Watt's third opinion in December 2003. Cf. Hash v. Comm'r of Soc. Sec, 309 Fed. Appx. 981, 985-87 (6th Cir. 2009) (affirming denial of DIB and upholding ALJ's decision to reject Dr. Koford's April 14, 2005 RFC questionnaire, implying that he was no longer a treating source because "[a]t the time of completion [of the questionnaire], Dr. Koford had not seen Hash for over three months").

Because Watts was not Karger's treating source in December 2003, the ALJ would not have been under any special obligation to defer to that opinion or to explain why he elected not to defer to it (as he was obligated to do with respect to Watts's earlier opinions, rendered while she was still a treating source for Karger). Moreover, even a treating physician's opinion has limited probative value where his treatment of the claimant stopped long before he rendered the opinion in question. Cf. Swain v. Comm'r of Soc. Sec 379 Fed. Appx. 512, 516-17 (6th Cir. 2010) ("[A] treating physician's opinion is `minimally probative' when the physician began treatment after the expiration of the claimant's insured status.") (quoting Siterlet v. HHS, 823 F.2d 918, 920 (6th Cir. 1987) ("Dr. Modzinski first saw him on November 23, 1983, almost eight months after expiration of his insured status. Appellant suffers from one or more degenerative disorders. Thus, Dr. Modzinski's report is minimally probative of his condition prior to March 31, 1983.")).

For purposes of medication management, psychiatrist Thomas Thysseril, M.D., of the Oak Tree Behavioral Health practice, evaluated Karger on May 5, 2005, diagnosing bipolar disorder I-most recent episode hypomanic, OCD, cannabis abuse, and personality disorder NOS with borderline features, and assigning a GAF score of 60, see Tr 419-20. It appears that Dr. Thysseril examined Karger on at least ten occasions within a nearly one-year period in 2005-2006: the original May 5, 2005 visit, then May 16, June 15, July 6 and 27, September 26, October 19, and November 30, 2005, and finally on February 13 and April 12, 2006. After every examination, Dr. Thysseril continued the same diagnoses, while lowering and raising her GAF within the range of 50-70. See Tr 405-424. GAF scores in the range of 61-70 are intended to indicate some mild symptoms or some difficulty in social, occupational, or school functioning, see DSM at 32. Dr. Thysseril advised Karger to stop using marijuana on July 27, 2005, and in February 2006 he noted that she had been abstaining from marijuana use for three months, see Tr 411-12 and Tr 421-22. Other than reducing Karger's Lamictal prescription and adding Abilify in June 2005, the portions of the record identified by the parties do not disclose other medication changes by Dr. Thysseril.

Appellant Karger emphasizes her hearing testimony about the alleged limitations imposed by her obsessive-compulsive disorder. Karger points to her testimony that she has difficulty grooming because she gets "stuck in front of the mirror" if she gets nervous, cannot wash dishes because she is "revolted" by the idea and afraid of contamination, and bathed or showered only about two days per week. See Karger's Br at 22 (citing Tr 448, 449-50 and 469). She recalls her explanation to the ALJ as to why she felt that she could not bathe or shower most days:

Karger: Like I cannot get myself to turn the faucet on. I'll just sit on the side of the tub and think okay, I've got to turn on the water, but I just don't feel like I can do it so —

ALJ: Okay. That's okay. Why can't you do it? Is it a physical reason or is it a mental reason?

Karger: It's a mental reason. It's a totally —

ALJ: Okay.

Karger: — mental reason and then even when I turn it on I have a hard time getting myself to go in the water you know —

ALJ: Okay.

Karger: — to get wet. It just seems like it's a — such a change from being dry. I don't know if that makes any sense.

Tr 447 to 448. Karger acknowledges that she smoked marijuana regularly for about six months until stopping in November 2005, resuming in March 2006 and ending in April 2006. See Karger's Br at 22 (citing Tr 460 and 479).

Karger assails the opinion of psychologist Dr. Jeffrey Madden, who testified as a medical expert. Karger's argument implies that the ALJ should not have credited Madden's opinion where it conflicted with treating psychologist Watts's opinion. Specifically, Madden testified that Karger suffered from OCD, bipolar disorder, and ADD but her conditions did not meet or equal a listed impairment and she would have only moderate impairments. Karger points out that Madden did not provide a specific RFC assessment and "did not cite any basis for his opinion other than a reference to a one-time consultative examination." Karger Br at 23 (citing Tr 480-82). Karger stresses that on cross-examination, Madden testified that treating psychologist Watts's opinion would not be inconsistent with Karger's psychological diagnoses, and he did not dispute Watts's findings. Karger Br at 23 (citing Tr 485). Ultimately, of course, Karger contends that the ALJ should have accepted the opinions of treating psychologist Watts, as the vocational expert ("VE") testified that if Watts's opinion (Tr 276-78) were accepted, Karger would be incapable of gainful employment, see Tr 492-93.

II.

On May 30, 2003, Karger filed the instant application for Social Security disability benefits ("DIB"), alleging a disability onset date of November 1, 2000; the agency denied the application on December 4, 2003 and denied reconsideration on March 31, 2004. See Tr 32-43 and 52-57. On April 29, 2004, Karger requested a hearing, which was held before ALJ John H. Metz on June 12, 2006. See Tr 427-494. Karger was represented by counsel (attorney Bradley J. Davis) and participated at the hearing, along with medical expert ("ME") Jeffrey Madden, Ph.D., and VE Bruce Growick. In his December 28, 2006 decision denying DIB, the ALJ found that Karger had five severe impairments — bipolar disorder, obsessive-compulsive disorder, attention deficit hyperactivity disorder, personality disorder, and degenerative disease of the lumbar spine — but determined that they did not render her totally disabled during the relevant period. Tr 16. The ALJ made the following findings of fact and conclusions of law.

2. The claimant has the following impairments that reduce her ability to perform basic work-related functions: bipolar disorder, obsessive-compulsive disorder, attention deficit hyperactivity disorder, personality disorder, and degenerative disease of her lumbar spine.

3. The claimant does not have an impairment or combination of impairments listed in, or medically equal to one listed in, 20 CFR Part 404, Subpart P, Appendix 1.

4. The claimant's subjective complaints are disproportionate to and not supported by the objective and substantial evidence in the record to the extent they suggest that she is disabled.

5. The claimant has the residual functional capacity to do the following: (1) lift and carry 50 pounds occasionally and 25 pounds frequently; (2) stand and/or walk about six hours in a workday; (3) sit for about six hours in a workday; (4) push and pull within normal limits; (5) kneel and crawl within normal limits; (6) frequently balance; (7) occasionally stoop, crouch, and climb; and (8) manipulate with her hands and fingers within normal limits. She is able to complete simple repetitive tasks that do not require prolonged periods of attention[,] in a low[-] stress environment with decreased contact with others.

* * *

9. Based on a capacity for the full range of light or medium work, and the claimant's age, education, and work experience, 20 CFR 416.969, Rules 202.20 and 203.28, Tables No. 2 and 3, respectively, Appendix 2, Subpart P, Regulations No. 4 would direct a conclusion of "not disabled."

10. Applying the above-cited vocational rules as a framework for decision-making, the claimant could have performed a significant number of jobs in the regional and national economies.

11. The claimant was not under a disability, as defined in the Social Security Act, at any time through the date of this decision.

Karger filed a request for review, which the Appeals Council denied on May 23, 2008. See Tr 5-9 and 425-26. Accordingly, the ALJ's decision became the final decision of the agency. See Howard v. Comm'r of Soc. Sec, 276 F.3d 235, 237 (6th Cir. 2002) (citing 20 C.F.R. § 416.1481).

Karger filed a complaint in the United States District Court for the Northern District of Ohio, contending that the ALJ committed reversible error by failing to comply with the treating-source rule and by improperly analyzing the medical opinion evidence of record. See N.D. Ohio No. 5:08-cv-1713, Doc 13 at 1. Specifically, Karger urged the district court to find error in the ALJ's failure to either give controlling weight to treating psychologist Watts's opinion or explain why he did not accord it such weight; his wrongful rejection of the opinion of non-treating psychiatrist Dr. John Comely, Psy.D.; and his failure to even mention the disability-favorable opinions of non-treating sources M.E. Menken, Ph.D., and Dr. Stuart Bartle, M.D. See id. at 14-19. The Commissioner's response brief before the district court conceded that the ALJ did not comply with the regulations, but reasoned that because "he met the goal of the regulations . . . [his] errors were harmless." Id. Doc 15 at 9-10 (citing Wilson v. Comm'r of Soc Sec, 378 F.3d 541, 544 (6th Cir. 2004) (discussing 20 C.F.R. § 416.927(d), which requires the agency to "always give good reasons in our notice of determination of decision for the weight we give [the claimant's] treating source's opinion.")).

III.

The parties presented expert opinion testimony or reports to the ALJ from at least the following nine medical professionals:

On July 31, 2009, the Honorable Benita Y. Pearson, United States Magistrate Judge, issued a Report and Recommendation (R R) recommending that the ALJ's decision be vacated and remanded for further consideration pursuant to sentence four of 42 U.S.C. § 405(g). Judge Pearson recommended holding that the ALJ violated the treating-source regulation with regard to Karger's treating psychologist, Dr. Watts:

Neither party contests that Dr. Watts is Karger's treating physician. The Commissioner conceded that the ALJ omitted Dr. Watts' opinion from his decision, in (presumably) any meaningful way. The Court noted two occasions in his written decision where the ALJ mentioned the work of Dr. Watts. Neither mention satisfies the reason-giving requirement of the treating source rule. The record before the court contains Dr. Watts' three assessments of Karger's abilities dated January 8, 2002, May 9, 2002, and December 27, 2003. The ALJ's entire "analysis" regarding Dr. Watts' December 2003 opinion is: "In December 2003, Dr. Watts commented that the claimant's psychological problems had interfered with her ability to master independent living skills, but Dr. Ickes['] independent evaluation findings, as just summarized, document that the claimant does have satisfactory independent living skills."

The Commissioner excuses the ALJ's lapse by explaining that "[t]he ALJ failed to address Dr. Watts' opinion in his decision. But the ALJ asked Dr. Madden to address Dr. Watts's opinion at the hearing." Th[is] . . . is incorrect. * * * Karger's counsel, not the ALJ, asked Dr. Madden to address Dr. Watts's opinion. The relevant portion of the exchange between Karger's counsel and Dr. Madden is as follows:

Q [Attorney]: Okay. Did you have a chance to review the 22F [Watts's December 2003 opinion], some notes from the treating source Donna Lots [phonetic] [for Dana Watts] a psychologist?

A [Dr. Madden]: Yes, I did.

Q. You've testified, obviously a little bit inconsistent with her opinion, but is her opinion that the Claimant has a poor or no ability in a number of areas[,] work areas? Is her opinion reasonable or is she just off base with it?

* * *

A. Obviously — I wouldn't say that what she's describing would be inconsistent with somebody that has a diagnosis that the claimant has.

* * *

The record clearly demonstrates that the ALJ did not ask Dr. Madden to address Dr. Watts' opinion at the hearing. Furthermore, even if the ALJ had asked Dr. Madden about Dr. Watts' opinion at the hearing, the ALJ neglected to mention in his written decision that Dr. Madden's opinion conflicted with Dr. Watts' opinion and to analyze or spell out for the reader how he resolved the conflicting medical opinion evidence.

This dialogue glaringly illustrates that Dr. Madden was unable to convincingly and confidently discredit Dr. Watts' opinion. "The testimony of [Dr. Madden] cannot provide a sufficient basis for rejecting the opinions of plaintiffs treating physicians since `the opinion of a nonexamining physician is entitled to little weight if it is contrary to the opinion of the claimant's treating physician.'" Shelman v. Heckler, 821 F.2d 316, 321 (6th Cir. 1987). . . . Therefore, Dr. Madden's contrary opinion should only be afforded little weight.

N.D. Ohio No. 5:08-cv-1713, Doc 16 at 10-11 (other citations omitted). The Magistrate further recommended holding that the ALJ's violation of the treating-source rule could not qualify as harmless error on the premise that his opinion nonetheless met the goal of the rule. See R R at 12-13.

Having concluded that the ALJ's failure to discuss Watts's opinions was not harmless error under Wilson, the Magistrate next recommended that his failure to mention state reviewing psychologist Menken's opinion was likewise not harmless error. See R R at 14-15.

Next, the Magistrate reasoned that the ALJ's rejection of Psy.D. Comley's opinion and his failure to even mention the opinions of Dr. Stuart Bartle and Tamsin Trelawny-Ross further established that the ALJ's opinion failed to reflect adequate, careful consideration of the record evidence. The Magistrate wrote as follows on this score:

The Court is "not unsympathetic to the plight of an ALJ confronted, as in this case, with the opinions of [at least nine] different psychological medical sources, in addition to multiple opinions relating to alleged physical impairments." Bowen, 478 F[.3d] at 750.[T]he Court is cognizant that an ALJ can consider all the medical opinion evidence without directly addressing in his written decision every piece of evidence submitted by a party, but his factual findings as a whole must show that he implicitly resolved such conflicts. See . . . Loral Defense Systems-Akron v. NLRB, 200 F.3d 436, 453 (6th Cir. 1999). . . . Nevertheless, while a "deficiency in opinion-writing is not a sufficient reason to set aside an ALJ's finding where the deficiency [has] no practical effect on the outcome of the case," inaccuracies, incomplete analysis and unresolved conflicts of evidence can serve as a basis for remand.

[T]he ALJ's errors and omissions were not harmless. Even if the ALJ had carefully considered the entire record in contemplating his decision and asked probative questions at the hearing, his written decision failed to articulate that careful consideration and the answers to the hearing questions. In sum, the ALJ's opinion does not represent the entire record. Given that, according to the Court's review of the entire record, at least four medical source opinions found Karger disabled because of her psychological impairments and at least four did not, the ALJ must carefully articulate his resolution of the conflict with contextual awareness of the record evidence. See, e.g., Kester v. Astrue, No. 3:07cv00423, 2009 WL 275438 at *9 (S.D.Ohio Feb.3, 2009) (stating that an ALJ cannot "pick and choose" only the evidence that supports his position) (internal citations omitted).

R R at 15-16 (n. 10, underlining other citations omitted). Finally, the Magistrate concluded that because an assessment of RFC "is based upon all of the relevant medical and other evidence," the ALJ's failure to properly evaluate medical opinion evidence made it necessary to perform a new RFC assessment after such evaluation. Id. at 17 (quoting 20 C.F.R. § 416.945(a)(3)). "After a new RFC determination has been made," she stated, "the ALJ may continue the sequential analysis, including posing an appropriate hypothetical to the Vocational Expert to determine whether Karger has the ability to perform any jobs in the national economy." R R at 17.

The agency filed an objection on August 12, 2009. The gist can be gleaned from the first and last paragraphs:

The Commissioner is concerned that the magistrate judge has recommended remand based on an unpublished district court case which relies solely on out-of-circuit law, where 6th Circuit precedent suggests a different outcome. [E]ven if the standard were as the magistrate judge says, [h]e has not considered the possibility of harmless error, where five physician opinions support the ALJ's decision. [T]he magistrate judge also did not consider Plaintiffs extensive activities.

* * *

In sum, although the ALJ did not expressly mention the opinions of treating psychologist Dr. Watts or reviewing psychologist Dr. Menken, the ALJ reasonably relied on five contrary opinions, as well as Karger's activities, to find Karger was not disabled. This clearly constitutes substantial evidence. * * *

N.D. Ohio No. 5:08-cv-1713, Doc 17 at 1-3 (other nn. and cites omitted). Karger filed a response to the objections on August 13, 2009, seeking the adoption of the R R. Doc 19 at 1-2. Karger filed a timely notice of appeal in October 2009, seeking the adoption of the R R.

The district judge declined to adopt the R R, issuing an order stating, in its entirety: In particular, this Court finds merit in Defendant's position that, although the AM did not expressly mention the opinions of [treating psychologist] Dr. Watts or [examining physician] Dr. Menken, the ALJ reasonably relied on five contrary opinions, in addition to Plaintiffs activities, in finding Plaintiff not disabled. Because the Court finds that the ALJ's determination is supported by substantial evidence, Defendant's Objections to the [R R] are SUSTAINED and the decision of the ALJ is AFFIRMED. This case is TERMINATED.

Id. Doc 19 at 1-2. Karger filed a timely notice of appeal in October 2009.

IV.

The district court had jurisdiction under 42 U.S.C. § 1383(c)(3), which provides: The final determination of the Commissioner of Social Security after a hearing under paragraph (2) [on an application for supplemental security income for the aged, blind or disabled] shall be subject to judicial review as provided in section 405(g) of this title to the same extent as the Commissioner's final determinations under section 405 of this title [an application for old age, survivors or disability insurance benefits].

In turn, 42 U.S.C. § 405(g) provides in pertinent part, "Any individual, after any final decision of the Commissioner of Social Security made after a hearing to which he was a party . . . may obtain a review of such decision by a civil action . . . in the district court of the United States in the district in which the plaintiff resides . . ." We have appellate jurisdiction under 28 U.S.C. § 1291.

V.

The district court's determination that substantial evidence supports the ALJ's decision is a legal conclusion that we review de novo. See Kyle v. Comm'r of Soc. Sec. 609 F.3d 847, 854 (6th Cir. 2010) (citing, inter alia, White v. Comm'r of Soc. Sec, 572 F.3d 272, 281 (6th Cir. 2009)).

The district court was obligated to accord great deference to the Commissioner's decision, as are we. The courts may not try the case de novo, nor resolve conflicts in evidence, nor decide questions of credibility. See Jordan v. Comm'r of Soc. Sec, 548 F.3d 417, 422 (6th Cir.2Q08) (citation omitted). "Under 42 U.S.C. § 405(g), the ALJ's findings are conclusive as long as they are supported by substantial evidence." Foster v. Halter, 279 F.3d 348, 353 (6th Cir. 2001) (citation omitted). "Substantial evidence" means "more than a mere scintilla. It means such relevant evidence as a reasonable mind might accept as adequate to support a conclusion." Id. at 353 (citing Kirk v. HHS, 667 F.2d 524, 535 (6th Cir. 1981)). Notably, substantial evidence is less than a preponderance of the evidence. See Kyle, 609 F.3d at 854 (citation omitted).

If the Commissioner's determination is supported by substantial evidence, it must stand regardless of whether a court might have resolved the disputed issues differently in the first instance. See Kyle v. Comm'r of Soc. Sec, 609 F.3d 847, 854-55 (6th Cir. 2010) (citing Lindsley v. Comm'r of Soc Sec. 560 F.3d 601, 604-4)5 (6th Cir. 2009), reh'g reh'g en banc denied (6th Cir. July 14, 2009)); Kinsella v. Sckweiker, 708 F.2d 1058, 1059 (6th Cir. 1983). The Commissioner's determinations are not subject to reversal merely because there is substantial evidence that would have supported the opposite conclusion. See Longworth, 502 F.3d at 595 (citation omitted).

VI.

The Commissioner does not contest that Watts and Thysseril were treating physicians, at least for some period of time. Watts and Thysseril's status as treating sources is legally significant, because

the Commissioner's regulations require that if the opinion of the claimant's treating physician is "well-supported by medically acceptable clinical and laboratory diagnostic techniques" and [is] "not inconsistent with the other substantial evidence in the case record," it must be given "controlling weight."

Tilley v. Comm'r of Soc. Sec, 394 Fed. Appx. 216, 222 (6th Cir. 2010) (per curiam) (Guy, Moore, Griffin) (quoting Hensley v. Astrue, 573 F.3d 263, 266 (6th Cir. 2009)). If a treating source's opinion is not well supported by such techniques or is inconsistent with the other substantial record evidence, however, it is not entitled to controlling weight. Tilley, 394 Fed.Appx. at 222-23 (citing 20 C.F.R. § 404.1527(d)(2)).

"Even if the ALJ does not give controlling weight to a treating physician's opinion, he must still consider how much weight to give it. . . ." Friend v. Comm'r of Soc. Sec, 375 Fed.Appx. 543, 550 (6th Cir. 2010). In doing so, the ALJ must take into account the length of the treatment relationship, the frequency of examination, the extent of the source's knowledge of the impairments, the amount of relevant evidence supporting the source's opinion, the extent to which the opinion is consistent with the record as a whole, whether or not the physician is a specialist in the relevant field, and any other relevant factors tending to support or contradict the opinion. See Friend, 375 Fed.Appx. at 550 (citing 20 C.F.R. § 404.1527(d)(2) through (6). The ALJ's decision on how much weight to accord a treating source's opinion must be accompanied by "good reasons" that are "`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.'") Friend, 375 Fed.Appx. at 550 (quoting SSR 96-2p, POLICY INTERPRETATION RULING TITLES II AND XVI: GIVING CONTROLLING WEIGHT TO TREATING SOURCE MEDICAL OPINIONS, 1996 WL 374188, *5 (July 2, 1996)). This "good reason" requirement serves both to ensure adequacy of review and to enable the claimant to understand the disposition of his case. See Rogers v. Comm'r of Soc, 486 F.3d 234, 242 (6th Cir. 2007).

The ALJ's decision mentioned one of Dr. Watts's opinions cursorily, and he did not mention the other two Watts opinions not at all, let alone explain why he did or did not credit those opinions instead of conflicting opinions from non-treating sources. These omissions are especially harmful given the VE's testimony that the limitations found by Watts — i.e., Karger's alleged "poor or no ability" to deal with work stressors and function independently, "poor or no ability" to relate predictably in social situations, "poor or no ability" to be reliable and to persist in work activity — if accurate, would prevent Karger from working (Tr 276-77). See Hall v. Comm'r of Soc Sec, 148 Fed.Appx. 456, 462 (6th Cir. 2005) (ALJ's failure adequately to explain rejection of treating physician Caudill's opinion was not harmless error; "Dr. Caudill was most familiar with the pain Hall experienced because of his back. The ALJ could not fairly decide to not address this key opinion of Dr. Caudill. Its importance is confirmed by the vocational expert . . ., who stated that if the lifting/sitting/standing restriction were added to other findings by the ALJ, Hall could not find work."). Contrast Tilley, 394 Fed. Appx. at 222-23 ("[T]he ALJ's decision in the instant case fully described the reasoning for discounting [treating physician] Dr. Srivastava's opinion. The ALJ thoroughly reviewed Dr. Srivastava's notes and found that his physical exams of Tilley were `essentially normal,' and that the evidence contained `little or no objective findings showing any significant change in [Tilley's] medical condition since he was released by his surgeon in April 2005.").

Nor did the ALJ's decision even mention Dr. Thysseril, let alone explain whether he was considered a treating physician and, if so, why the ALJ did or did not credit his opinion generally, how he found Thysseril's opinion to conflict with Watts's opinions, and most of all, why the ALJ (possibly) considered Thysseril's opinion to be more reliable than Watts's opinions to the extent that they conflicted. The lack of discussion of Thysseril's opinions is no minor matter, because his findings were dramatically different from the findings of the other, earlier treating source, Watts.

For example, Thysseril found that during his treatment of her from May 2005 to April 2006, Karger had generally moderate symptoms and limitations, GAF ratings in the mild to moderate range, normal speech, logical and goal-directed thoughts, good orientation and memory, good cognition, and good impulse control, insight, and judgment, see AR 405 — 424, all of which arguably corroborates the testimony of medical expert Menken (AR 481-82), examining source Ickes (AR 272-74), and state-agency reviewing psychologists Malinky and Marlow (AR 291 and 296), and tends to support the conclusion that Karger was not disabled. Cf. Blakley v. Comm'r of Soc. Sec, 581 F.3d 399, 408 (6th Cir. 2009) (vacating denial of benefits and remanding to ALJ) (where psychiatrist Raza started out as a consulting source only, the relationship expanded to include treatment at the Cumberland River Care Center; "Though the ALJ summarizes Blakley's time at Cumberland . . . the ALJ does not explain in her decision whether she weighed Dr. Raza as an expert, a treater, or both. Here again, the ALJ failed to account for Dr. Raza's opinions as a treating physician in disregard of 20 C.F.R. § 404.1527."). Contrast Carrelli v. Comm'r of Soc. Sec, 390 Fed.Appx. 429, 436-38 (6th Cir. 2010) (ALJ was not required to give controlling weight to opinion of treating physicians Wakham and Naylor, because ALJ's opinion explained that he gave "considerable weight" to a non-treating physician Summers's contrary assessment because it took account of the claimant's left shoulder and hip pain but also was consistent with benign CT scans, EMGs, MRIs and nerve-conduction studies and with the claimant's self-reported and observed daily activities, and an expert in Social Security disability evaluation agreed with the non-treating source).

Naturally, that means that the ALJ did not explain the nature and extent of any conflict between treating source Watts and treating source Thysseril and several non-treating sources. Less still did the ALJ explain how he resolved any such conflict. Contrast Tilley, 394 Fed.Appx. at 222-23 ("Furthermore, the ALJ determined that the opinion of Dr. Lovell, Tilley's neuro-surgeon, was more entitled to more weight as a specialist and specific treater of Tilley's back problems, in accordance with 20 C.F.R. § 416.927(d)(5).").

In Wilson v. HHS, we acknowledged that the regulatory requirement to give good reasons for not deferring to a treating source's opinion "does not require conformity at all times." Coldiron, 391 Fed. Appx. at 439 (citing Wilson v. Comm'r of Soc. Sec, 378 F.3d 541, 547 (6th Cir. 2004)). Violation of the requirement constitutes harmless error if the ALJ has "met the goals of the procedural requirement — to ensure adequacy of review and to permit the claimant to understand the disposition of his case — even though he failed to comply with the regulation's terms." Id. (citing Wilson, 378 F.3d at 547). "An ALJ may accomplish the goals of this procedural requirement by indirectly attacking the supportability of the treating physician's opinion or its consistency with other evidence in the record." Id. (citing, inter alia, Hall v. Comm'r of Soc. Sec, 148 Fed.Appx. 456, 464-65 (6th Cir. 2005)). "Notably, courts look to the ALJ's decision itself, and not other evidence in the record, for support." Id. In other words, the requisite evidence and analysis must appear in the ALJ's decision, not simply be present in the rest of the record.

While an ALJ is not required to discuss every piece of medical opinion evidence, the ALJ here did not nearly discuss enough of that evidence to enable us to determine whether substantial evidence supports the determination that Karger's impairments do not render her disabled. For example, the ALJ's only mention of an opinion by treating psychologist Watts merely noted that "[i]n December 2003, Dr. Watts commented that the claimant's psychological problems had interfered with her ability to master independent living skills . . ., but Dr. Ickes['] independent evaluation findings, as just summarized, document that the claimant does have satisfactory independent living skills." Tr. 17. This is insufficient under our precedents. "[I]t is not enough to dismiss a treating physician's opinion as `incompatible' with other evidence of record; there must be some effort to identify the specific discrepancies and to explain why it is the treating physician's conclusion that gets the short end of the stick." Friend, 375 Fed.Appx. at 552; see, e.g., id. at 551 ("The ALJ's rationale for discounting that opinion [of primary-care treating physician Angerman] was expressed simply as `the testimony of [Medical Expert] Dr. Nusbaum, which would allow the claimant to stand/walk for one hour [at a] time to a total of six hours in an eight-hour workday, is more consistent with the objective clinical findings,' and `there is no basis for Dr. Angerman's conclusion that the claimant can stand/walk for only one hour in a day.' This is not `sufficiently specific' to meet the requirements of the rule on its face, inasmuch as it neither identifies the `objective clinical findings' at issue nor discusses their inconsistency with Dr. Angerman's opinion.").

In addition, the ALJ did not discuss the opinions of two non-treating sources, examining psychologist Dr. Bartle and reviewing-source ME Dr. Menken, so their opinions (and the support for their opinions) were not analyzed at all, let alone for their tendency to support or undermine either of the treating sources' opinions. This is potentially significant, because Dr. Bartle assigned Karger a GAF score of 45 upon her admission to Gould Farm in May 2002, see Tr 231, and he opined that Karger could not function outside her home without extensive support and could muster sustained concentration and attention only in a supportive work program, see Tr 183-86. The complete omission of any discussion of ME Dr. Menken's opinion is no minor matter, either, because Menken opined in August 2002 that Karger suffered "marked" limitations in maintaining social functioning and concentration, persistence and pace, and that her impairments satisfied Listings 12.06a and 12.06b. This suggests that the ALJ did not consider the opinions of Bartle and Menken. As to ME Menken, that would violate the regulation providing that "all evidence from non-examining sources" is considered to be opinion evidence, 20 C.F.R. § 416.927(f), and that while ALJs are not bound by the findings of State agency or other "program" physicians and psychologists, such sources are "highly qualified [professionals] who are also experts in Social Security disability evaluation" and ALJs " must consider [their]findings . . . as opinion evidence. . . ." 20 C.F.R. § 416.927(f)(2)(i). The omission of any mention of ME Menken's opinion violates the regulatory requirement that

[u]nless the treating source's opinion is given controlling weight, the administrative law judge must explain in the decision the weight given to the opinions of a State agency medical or psychological consultant or other program physician or psychologist, as the administrative law judge must do for any opinions from treating sources, nontreating sources, and other nonexamining sources who do not work for us.

20 C.F.R. § 416.927(f)(2)(ii).

For instance, it might be that the ALJ carefully considered all the opinions which his opinion neglects to discuss — 1999-2002 treater Watts, 2005-2006 treater Thysseril, and non-treaters Bartle and Menken — and discounted Watts's opinions in favor of some or all of those three others in part because they were issued much later than Watts' opinion and thus rested on a fuller, more up-to-date medical record and a longer-term perspective on Karger's condition. Cf. Blakley v. Comm'r of Soc. Sec, 581 F.3d 399, 409 (6th Cir. 2009) (an ALJ may credit the opinion of a State agency medical consultant over that of a treating or examining source "when the `. . . consultant's opinion is based on a review of a complete case record that . . . provides more detailed and comprehensive information than what was available to the individual's treating source."`) (quoting SSR 96-6p, POLICY INTERPRETATION RULING TITLES II AND XVI: CONSIDERATION OF ADMINISTRATIVE FINDINGS OF FACT, BY STATE AGENCY MEDICAL AND PSYCHOLOGICAL CONSULTANTS AND OTHER PROGRAM PHYSICIANS AND PSYCHOLOGISTS AT THE ADMINISTRATIVE LAW JUDGE AND APPEALS COUNCIL LEVELS OF ADMINISTRATIVE REVIEW: MEDICAL EQUIVALENCE, 1996 WL 374180, *3 (July 2, 1996)); see, e.g., Fisk v. Astrue, 253 Fed.Appx. 580, 584-86 (6th Cir. 2007) (vacating non-disability determination and remanding to ALJ, where ALJ's decision to discount the opinion of treating physician Agarwal in favor of State agency physicians' opinions failed to take into account the fact that Agarwal's opinion was based on a year of additional test results which were not available to the consulting sources when they rendered their opinions). But the ALJ's opinion leaves us no way of knowing or even reasonably inferring that any such thought process took place. The harmless-error doctrine cannot be stretched far enough to excuse the ALJ's failure to meaningfully indicate, even indirectly, how much weight he accorded the two treating sources (Watts and Thysseril) vis-a-vis each other, vis-a-vis the numerous non-treating sources, and why. It is not this Court's role, or even the district court's role, to scour the record for evidence and expert reasoning which the ALJ might have relied on and which could support a finding of no-disability if the ALJ actually considered it.

In short, the ALJ's opinion did not serve the twin goals of the regulatory "good reasons" requirement — ensuring adequate appellate review and enabling the claimant to understand the disposition of her case — and the case must be remanded to the ALJ without regard to the likelihood that the ALJ will reach the same conclusion on remand. See Friend, 375 Fed.Appx. at 551 ("We will reverse and remand a denial of benefits, even though `substantial evidence otherwise supports the decision of the Commissioner,' when the ALJ fails to give good reasons for discounting the opinion of the claimant's treating physician.") (citing Wilson, 378 F.3d at 543-46 (a reviewing court "cannot excuse the denial of a mandatory procedural protection simply because . . . there is sufficient evidence in the record of the ALJ to discount the treating source's opinion and, thus, a different outcome on remand is unlikely.")). This Court has even said that an ALJ's failure to identify the reasons for discounting treating-source opinions (if any were discounted) and to explain "precisely how those reasons affected the weight accorded the opinion denotes a lack of substantial evidence, even where the conclusion of the ALJ may be justified based upon the record." Rogers, 486 F.3d at 243 (emphasis added).

We need not delve further into Karger's contention that the ALJ erred in his assessment of the medical opinions and evidence which were meaningfully discussed in his opinion. "In light of the above review, and the resulting need for remand of this case, further analysis of Plaintiffs remaining contentions is unwarranted." Neal v. Comm'r of Soc. Sec, No. l:08-cv-512, 2009 WL 3010848, *14 (S.D.Ohio Sept.21, 2009).

Finally, on this record, it would be rash to simply award benefits, as Karger urges. Karger cites Mowery v. Heckler, 771 F.2d 966, 973 (6th Cir. 1985) (citing Lashley v. HHS, 708 F.2d 1048, 1053-54 (6th Cir. 1983)), which held that a court may directly award benefits, rather than remanding for proper consideration by the ALJ, where the record is complete and the Commissioner's decision is clearly erroneous, proof of disability is strong and evidence to the contrary is lacking, or the claimant has presented overwhelming evidence of disability. On this record, none of these conditions is satisfied. Therefore, in accordance with Faucher [v. HHS, 17 F.3d 171, 176 (6th Cir. 1994)], the Court will remand the matter for further proceedings.

VII.

For the foregoing reasons, we REVERSE the district court, VACATE the denial of Karger's application, and REMAND to the ALJ pursuant to 42 U.S.C. § 405(g) Sentence Four for further consideration and a new opinion consistent with this opinion and the regulations.