LYNN ADELMAN District Judge

DECISION AND ORDER

In deciding whether applicants for social security disability benefits can, despite their impairments, still perform jobs in the economy, Administrative Law Judges ("ALJ's") often rely on testimony from a vocational expert ("VE"). The VE's job is to provide an impartial assessment of the types of occupations in which the applicant can work and the availability of positions in such occupations. Weatherbee v. Astrue, 649 F.3d 565, 569 (7^th Cir. 2011). Often, the VE will provide no more than a bottom line - identifying certain jobs the applicant can do and the number of such jobs within the applicant's region or state; if unchallenged, such testimony may suffice to deny a claim. See, e.g., Barrett v. Barnhart, 355 F.3d 1065, 1067 (7^th Cir. 2004); Donahue v. Barnhart, 279 F.3d 441, 446 (7^th Cir. 2002).

However, the VE must make the data and reasoning underlying his opinion available on demand, Overman v. Astrue, 546 F.3d 456, 464 (7^th Cir. 2008), and if the applicant challenges the foundations of the VE's conclusions at the hearing the ALJ must determine whether the purported expert's conclusions are reliable, McKinnie v. Barnhart, 368 F.3d 907, 911 (7^th Cir. 2004). Moreover, the ALJ must solicit a reasonable explanation for any VE testimony in conflict with the standardized occupational information contained in the Dictionary of Occupational Titles ("DOT") and its companion publication, the Selected Characteristics of Occupations Defined in the Revised Dictionary of Occupational Titles ("SCO"), SSR 00-4p, 2000 WL 1898704, at *1.

In the present case, plaintiff Diane Marotti Roxbury raised serious doubts about the reliability of the VE's testimony, pointing out conflicts between his testimony and the DOT, questioning his data and methods, and submitting her own contrary vocational report. The ALJ brushed these criticisms aside, modifying his assessment of plaintiff's work capacity to account for the VE's flawed testimony, noting that plaintiff's counsel was permitted to cross-examine the VE regarding his methodology, and then choosing to credit the testimony from the VE over plaintiff's expert given the inexact nature of vocational science.

Plaintiff now seeks judicial review of the ALJ's decision. Even under the less stringent standards that apply to expert testimony at administrative proceedings, and even in light of the deferential standard the court applies in reviewing an ALJ's decision, the matter must be remanded.

I. BACKGROUND

Plaintiff's odyssey through the social security system began in 2004, when she applied for benefits based on a variety of conditions, including degenerative disc disease, fibromyalgia, arthritis, depression, autoimmune disease, and migraine headaches, with an alleged disability onset date of November 2, 2002. (Tr. at 70, 73, 635.) Denied initially (Tr. at 630) and on reconsideration (Tr. at 627), she requested a hearing, and on January 19, 2007, she appeared with counsel before ALJ Jean Kingrey in Medford, Oregon (Tr. at 643). The ALJ took testimony, then continued the hearing in order to obtain a neuropsychological evaluation (Tr. at 670), which Thomas Shields, Ph.D. conducted on March 19, 2007 (Tr. at 545). Based on his testing and clinical interview, Dr. Shields concluded that in addition to depression and anxiety plaintiff suffered from a learning disorder, and that given her psychological impairments she would have moderate to severe impairment in her capacity to sustain concentration, persistence, and pace over the course of a full-time work week. He set her "GAF" at 50-55. (Tr. at 552.)

At the continued hearing, the ALJ asked the VE to identify jobs for a person limited to light work, with no over the shoulder work, no detailed tasks or instructions, no close interaction with the general public or coworkers, and only occasional handling and fingering. (Tr. at 681-82.) The VE testified that such a person could not perform plaintiff's past work as a tanning salon owner/operator and retail salesclerk, as those jobs involved close interaction with the public, or as a bookkeeper, which required the performance of detailed tasks. The VE testified that the person could do other jobs, however, such as rental clerk, office helper, and routing clerk. (Tr. at 683.)

On September 28, 2007, ALJ Kingrey issued an unfavorable decision. The ALJ found that plaintiff had not worked since November 2, 2002, the alleged onset date, and that she suffered from the severe impairments of depression/anxiety, borderline IQ/learning disorder, degenerative disc disease of the lumbar spine, fibromyalgia, and right shoulder adhesive capsulitis and possible arthritis, none of which qualified as conclusively disabling. (Tr. at 32-34.) The ALJ then determined that plaintiff retained the residual functional capacity ("RFC") to perform light work, with no over the shoulder work and no more than occasional fingering and handling. The ALJ found that plaintiff was further limited to simple routine tasks with no close interaction with the general public or co-workers (Tr. at 34), limitations the ALJ believed accounted for Dr. Shields's findings (Tr. at 41).

In making this determination, the ALJ discounted a January 2007 report from Dr. George Johnston, plaintiff's treating pain specialist, who opined that plaintiff could not sustain even sedentary work since September 2004. The ALJ noted that plaintiff alleged subjectively disabling symptoms since 2002, and Dr. Johnston failed to explain why he selected the September 2004 date. The ALJ further noted that, despite the statement in Dr. Johnston's notes that narcotics were ineffective and their prescription controversial in the treatment of fibromyalgia, the doctor continued to prescribe such medications for plaintiff. (Tr. at 40.) The ALJ also found Dr. Johnston's opinion inconsistent with plaintiff's daily activities, which included some work activity after September 2004, suggesting some functional ability to perform work. (Tr. at 40-41.) Finally, the ALJ noted that the conclusion that plaintiff was unable to work was an opinion on an issue reserved to the Commissioner, which required vocational expertise outside the purview of the treating physician. (Tr. at 41.)

The ALJ also found plaintiff's subjective claims not fully credible, noting reports of various somatic complaints, drug-seeking behaviors, and exaggerated pain behavior in the treatment records. (Tr. at 41.) The ALJ also noted inconsistencies between some of plaintiff's contentions and the record, such as her claim that she fired one of her doctors because he provided inadequate care, while the doctor's records indicated that he terminated her as a patient based on her non-compliance. (Tr. at 41-42.) The record contained evidence of other conflicts with providers, with plaintiff alleging mistakes in her care while the providers noted problems in plaintiff's compliance. The ALJ further noted plaintiff's activities, including gardening, playing with her dog, and traveling for fun, which were inconsistent with her complaints of disabling symptoms. (Tr. at 42.)

Based on this RFC, the ALJ determined that plaintiff could not perform any of her past jobs, but in light of her age, education, and work experience, she could perform other jobs as identified by the VE, including rental clerk, office helper, and clerical dispatcher. (Tr. at 42-43.) The ALJ accordingly found plaintiff not disabled. (Tr. at 44.)

Plaintiff brought an action for judicial review in the District of Oregon, raising various claims of error, including a conflict between the VE's testimony and the DOT, improper rejection of her testimony, and improper rejection of reports from the state agency consultants and Dr. Johnston. She argued for a judicial award of benefits. The Commissioner agreed that the ALJ erred by accepting VE testimony in conflict with the DOT and by failing to provide reasons for rejecting the consultants' reports, but argued that the proper remedy was a remand for further proceedings. (Tr. at 753.)

In a March 23, 2010 decision, the district court found that the ALJ provided sufficient reasons for rejecting Dr. Johnston's report and for finding plaintiff not fully credible. (Tr. at 754- 55.) However, the court agreed with the parties that the ALJ erred by relying on VE testimony in conflict with the DOT; specifically, the ALJ found plaintiff limited to occasional fingering and handling, while all three of the jobs listed by the VE required frequent handling and fingering. (Tr. at 755.) The court also agreed with the parties that the ALJ failed to provide reasons for rejecting the state agency doctors' reports, which limited her to sedentary work. (Tr. at 755-56.) The court found the record insufficient to support an immediate award of benefits and thus remanded for further proceedings and findings. These findings were to include, inter alia, further consideration of the state agency experts' opinions that plaintiff was limited to sedentary work, reevaluation of plaintiff's RFC, and composition of a hypothetical consistent with the record for the VE's consideration. (Tr. at 756.)

On April 4, 2010, the Appeals Council vacated the ALJ's decision and remanded the case to an ALJ for further proceedings consistent with the order of the court. (Tr. at 762.) Those proceedings were delayed, however, by plaintiff's appeal of the district court's remand decision. On May 6, 2011, the Ninth Circuit affirmed the district court's decision to remand rather an immediately awarding benefits. Roxbury v. Commissioner, Social Sec. Admin., 431 Fed. Appx. 597 (9^th Cir. 2011).

In the interim, plaintiff moved from Oregon, first to Hawaii and then to Wisconsin, and the Social Security Administration transferred her case accordingly. (Tr. at 779-82.) Ultimately, the case was set for a hearing in Milwaukee before ALJ Timothy Malloy on April 26, 2012. (Tr. at 793, 1194.)

Prior to the hearing, plaintiff's counsel requested the issuance of a subpoena duces tecum requiring the VE to bring with him documents upon which he may rely in forming his opinion (Tr. at 839), but the ALJ denied that request as "unnecessary" (Tr. at 712), noting that counsel could question the VE about where his statistical or job information came from but that the VE was not being called as a statistician or census taker (Tr. at 1199).

At the hearing, plaintiff's counsel amended the onset date to April 12, 2004, explaining that plaintiff had been diagnosed with a number of problems prior to that date but at that point it became obvious that her low back problems were significant, and her doctors started doing work-ups regarding her low back, including x-rays, MRI's, and epidural injections. (Tr. at 1201-02.)

Plaintiff testified that she was fifty-one years old, a high school graduate with some additional classes in business management. (Tr. at 1202.) She reported having trouble academically in school but denied special education classes. (Tr. at 1226.) She indicated that she last worked in 2006, doing part-time bookkeeping in Oregon. (Tr. at 1203.) That job ended because she took too many sick days. (Tr. at 1224.) Prior to that, she owned and managed a tanning salon in California for four to five years. She was forced to sell that business due to her failing health. She then moved to Oregon to live with her mother, where she briefly worked in phone and retail sales. (Tr. at 1218-19.) In 2010, she moved to Hawaii with her new husband. (Tr. at 1204.) The marriage ended due to his abuse, and he later took his life. (Tr. at 1211-12.) In 2011, she moved to Wisconsin. (Tr. at 1205.)

Plaintiff indicated that on a good day, she did some gardening, a little walking with her dog, or cooking. (Tr. at 1205.) She did light housework, such as dusting, cleaning, and folding laundry; her boyfriend did the heavier chores. She denied other recreational activities. (Tr. at 1206.) She had no income and relied on her boyfriend for support. She also lacked health insurance, receiving treatment at a free clinic. (Tr. at 1207.) Plaintiff testified that she took Paxil for depression, Trazodone for sleep, and Oxycodone and Morphine for pain. (Tr. at 1208-09.)

Plaintiff testified that she had trouble sitting or standing for long periods of time, maintaining concentration, and using her hands. (Tr. at 1209.) She stated that her finger joints were swollen, and she could not wear rings and had trouble holding things such as a pen. (Tr. at 1232-33.) She also complained of fibromyalgia flare-ups, where it hurt just to wear undergarments. (Tr. at 1209-10.) She also had a frozen right shoulder, which made it difficult to reach overhead. (Tr. at 1210, 1227-28.) Plaintiff testified that she dressed slowly, with help from her boyfriend. (Tr. at 1213.) She indicated that the furthest she walked with her dog was six blocks. She did no lifting around the house. Her ability to sit varied. (Tr. at 1215.) She could stand for about ½ hour. (Tr. at 1216.)

Plaintiff testified that on a bad day she got out of bed only to use the bathroom. (Tr. at 1220.) She related experiencing such days at least once per month dating back to 2002 or 2003. (Tr. at 1221.) She currently experienced bad days, which put her in bed for 24 hours or more, two to three times per month. (Tr. at 1222.) Cold, damp weather triggered bad days, as did overdoing it physically. (Tr. at 1223.)

The ALJ then turned to the VE, Joseph Entwhistle, asking a series of four hypothetical questions. The first limited the person to light work, occasional lifting above the right shoulder, and occasional bilateral fingering and handling. The VE testified that such a person could perform plaintiff's past manager job, but not the sales clerk and bookkeeping jobs as they required frequent use of the hands. (Tr. at 1237.) The second question reduced the exertional level to sedentary. The VE responded that the manager position would be eliminated. (Tr. at 1237-38.) However, the person could perform other jobs, such as general office clerk, receptionist and information clerk, and video surveillance monitor. (Tr. at 1238, 1242.) The third question modified the first (light) hypothetical, adding mental limitations to unskilled work involving simple, routine, and repetitive tasks, and allowing the person to be off task up to 5% of the day in addition to breaks. (Tr. at 1238.) The VE said this would eliminate the manager position, but other jobs were available at the light level, including parking lot attendant (DOT # 375.587-010), with 779 jobs in Wisconsin; lobby attendant (DOT # 358.677-014), with 512 jobs; and interviewer (DOT # 205.367-054), with 1620 jobs). Finally, the fourth question reduced the exertional level to sedentary. The VE testified that the person could work as a general office clerk, credit checker, and information clerk. If the person would be absent more than two days per month or off task more than 10% of the day all jobs would be eliminated. (Tr. at 1240, 1243-44).

The VE stated that his testimony was consistent with the DOT. (Tr. at 1240.) He indicated that he used as sources the Bureau of Labor Statistics' "Occupational Employment Statistics" and U.S. Publishing Co.'s "Occupational Employment Quarterly," both government publications, along with his experience and background. (Tr. at 1241, 1246.) However, on cross-examination plaintiff established that the Occupational Employment Quarterly was produced by a private company using government data. (Tr. at 1246-47.) The VE further admitted that these sources did not report jobs by DOT title, requiring him to use a "proprietary" methodology "to provide clusters of jobs that have the same requirements." (Tr. at 1249-50.)

After the hearing, plaintiff submitted a report from David Traver, M.S., J.D., an attorney holding a masters in vocational rehabilitation, critiquing the VE's testimony. (Tr. at 1055-56.) Specifically, Traver noted that in response to hypothetical question three (light work, occasional lifting above the right shoulder, occasional bilateral fingering/handling, simple routine repetitive work off task 5% of the time), the VE identified three jobs: parking lot attendant (DOT # 375.587-010), lobby attendant (DOT # 358.677-014), and interviewer (DOT # 205.367-054). (Tr. at 1057.) However, according to the DOT, the first job was actually called "parking enforcement officer," a government service job, and it required frequent handling and fingering according to the SCO. The lobby attendant job was actually called "locker room attendant," and it too required frequent use of the hands. The third job, known primarily as "survey worker," also required frequent handling and fingering. (Tr. at 1058.) Traver further noted that the interviewer/survey worker job required a "Reasoning level" of 3, defined by the DOT as: "Apply common sense, understand and carry out instructions furnished in written, oral or diagrammatic form. Deal with problems involving several concrete variables in or from standardized situations." (Tr. at 1059.) Traver opined that a limitation to simple, routine, repetitive work would be incompatible with Reasoning level 3; rather, it compared to Reasoning level 1, defined by the DOT as: "Apply common sense, understand and carry out simple one or two step instructions. Deal with standardized situations with occasional or no variables . . . in or from the situations encountered on the job." (Tr. at 1059.) Finally, Traver questioned the VE's reliance on U.S. Publishing Co.'s Occupational Employment Quarterly and the Bureau of Labor Statistics' Occupational Employment Statistics. Specifically, Traver noted that the former arbitrarily distributed jobs equally by DOT codes within a single census code, while the latter did not report jobs by DOT code and did not permit one to reliably determine the number of jobs by DOT code. (Tr. at 1061-62.)

On August 9, 2012, ALJ Malloy issued an unfavorable decision. (Tr. at 707.) He noted that prior to the hearing plaintiff's counsel sought a subpoena duces tecum requiring the VE to bring with him documents relevant to his testimony, which request was denied as "unnecessary." (Tr. at 712.) ALJ Malloy then incorporated by reference ALJ's Kingery's analysis of the medical records and plaintiff's credibility for the period she addressed. ALJ Malloy noted that the district court found no fault with ALJ's Kingery's analysis of the medical evidence but rather remanded based on the inconsistency between the VE's testimony and the DOT and ALJ's Kingery's failure to discuss the state agency doctor's physical RFC assessment. (Tr. at 715.) ALJ Malloy directed his focus to the records since April 2007. (Tr. at 716.)

The ALJ first discussed the consultative neuro-psychological examination report from Dr. Shields, which noted IQ scores in the upper 70's to low 80's and academic testing in the first to sixth grade level, which the ALJ found odd given plaintiff's high school diploma and work history as a bookkeeper and business owner. The ALJ further noted that Dr. Shields discussed medical records, a subject beyond his expertise, and that he seemed to accept plaintiff's statements at face value. (Tr. at 716.)

The ALJ then reviewed the more recent consultative exams performed in Wisconsin. Specifically, in September 2011, plaintiff was seen for a consultative physical exam by Dr. Daryl Melzer. Dr. Melzer indicated that plaintiff's history and findings were consistent with fibromyalgia. He found no active rheumatoid arthritis although she likely did have degenerative arthritis in some of her joints. That same month, plaintiff underwent a mental consultative exam with Dr. Pushkash, who diagnosed dysthymic disorder with a GAF of 55. He opined that plaintiff had the ability to understand, remember, and carry out instructions, but that her ability to concentrate and persist on tasks may be mildly to moderately compromised due to anxiety and depression. He noted no problems interacting with others. (Tr. at 721.)

The ALJ noted that plaintiff alleged a multitude of complaints affecting numerous body systems, which, if considered valid, would be disabling. (Tr. at 723.) However, the ALJ found the medical evidence nowhere near as impressive as the subjective complaints. (Tr. at 724.) The ALJ accepted that pain can very subjective, and that plaintiff had been prescribed narcotics, suggesting a definite element of pain. However, the ALJ noted that the objective medical testing was less than impressive, questions had been raised as to the appropriateness of narcotic drugs for fibromyalgia, and some records suggested dependency issues and drug seeking behavior. (Tr. at 724.) The ALJ further noted the inconsistent references to rheumatoid arthritis and the absence of any evidence to confirm lupus. Physical exams generally revealed good grip strength and hand mobility. As to her back and neck complaints, physical exams typically showed good range of motion; scans showed only mild degenerative changes, with no evidence of disc herniation or impingement of the nerve roots to suggest radiculopathy; and neurological exams showed intact strength, reflexes, and sensation. He further stated: "[W]hile on occasion straight leg raising has been reduced, it was because of subjective complaints of low back pain and not because of any neurological involvement." (Tr. at 724.) Plaintiff never had surgery or referral to a specialist for her spine; rather, she opted for pain clinic treatments, receiving narcotics and the occasional injection. (Tr. at 724.)

Regarding the Oregon state agency consultant's determination of sedentary work, the ALJ found this a "premature assessment based on limited records then available." (Tr. at 724.) The ALJ noted a more recent Wisconsin state agency determination of light work. (Tr. at 725.)

The ALJ acknowledged plaintiff's argument that the VE, in responding to a hypothetical for occasional handling and fingering, identified jobs that would involve frequent handling and fingering. "To make this a moot point, the undersigned finds that [plaintiff] would be capable of frequent, but not constant fingering or handling." (Tr. at 725.) The ALJ based this on "the fact that there is no rheumatoid arthritis and whatever degenerative changes [plaintiff] may have, they would be mild." (Tr. at 725.) The ALJ cited a November 2006 EMG test, which showed minimal right median sensory neuropathy and borderline right ulnar sensory neuropathy. (Tr. at 725.)

Regarding plaintiff's mental limitations, ALJ Malloy noted the finding of no severe impairment by the Oregon and Wisconsin consultants. (Tr. at 725.) The ALJ found the consultative report from Dr. Shields "questionable" given that his testing would put plaintiff at close to being illiterate and cognitively in the borderline range, despite the fact that she was never in special education classes, owned her own successful business, and worked for many years as a bookkeeper. (Tr. at 726.) The ALJ noted that while plaintiff may have some degree of depression given her current economic circumstances, she would have the mental capacity for both unskilled and low level semiskilled tasks. (Tr. at 726.)

The ALJ noted plaintiff's argument that the jobs identified by the VE had a Reasoning level of 3, which she argued removed them from the unskilled category. "While this argument may have some merit, the undersigned is probably to blame for making the hypothetical question too narrow in it scope, as he believes [plaintiff] has the capacity to apply common sense, understand and carry out instructions furnished in written, oral or diagramic form and deal with problems involving several concrete variables in or from the standardized situations that she may encounter." (Tr. at 726.) In making this finding, the ALJ relied on the reports from Dr. Pushkash and the Wisconsin consultants, as well as plaintiff's past history. The ALJ accordingly found that plaintiff had the mental capacity for at least unskilled or low level semiskilled tasks, so long as she was allowed to be off task 5% of the time and provided that the work did not involve extended periods of concentration or attention to detailed types of tasks. He concluded that the jobs cited by the VE fell into that category. (Tr. at 726.)

Finally, the ALJ assumed that the VE had the standing and credentials to address vocational matters, including the DOT and related materials. (Tr. at 726.) The ALJ noted that plaintiff's counsel was able to cross-examine the VE on his qualifications and the source of his numbers, and the VE was able to explain how he arrived at those numbers. The ALJ cited Liskowitz v. Astrue, 559 F.3d 736 (7^th Cir. 2009), for the proposition that vocational testimony is not an exact science, and that when faced with competing opinions the ALJ must decide which opinion to credit. The ALJ chose "to credit the testifying impartial expert over Mr. Traver." (Tr. at 727.) In response to plaintiff's challenge to the accuracy of the VE's numbers and whether they were sufficient to constitute a significant number, the ALJ noted that given the fluctuating nature of the workforce extrapolations from available reports will be less than precise. "By the very nature of such fluid situations and the almost instant obsolescence of such reports once they come out, there is considerable room for debate as to who is correct and who is not." (Tr. at 727.) The ALJ cited Sullivan v. Lee, 988 F.2d 789 (7^th Cir. 1993), in which the court found fewer than 1400 jobs significant. "Here the VE was testifying as a vocational expert and not a statistician or census taker. And his information came from reliable sources, indeed, some of which the undersigned is required to take administrative notice." (Tr. at 727.)

The ALJ then determined that plaintiff suffered from the severe impairments of mild to moderate degenerative disc disease, right shoulder adhesive capsulitis, fibromyalgia, and a dysthymic disorder. The mental impairment resulted in mild restrictions of daily activities, mild difficulty in social functioning, and moderate deficiencies of concentration and pace as it related to more detailed types of tasks. He found that plaintiff had the mental capacity for unskilled to low level semi-skilled work involving simple, routine, repetitive tasks that would allow her to be off task no more than 5% of the day. Regarding her physical capacity, she was limited to light work with no more than occasional lifting above shoulder height with the right arm. Her arthritis would preclude constant fingering and handling but would allow occasional to frequent fingering and handling. Given these limitations, plaintiff could not perform her past work as a bookkeeper or store manager. (Tr. at 727.) However, relying on the VE's testimony, the ALJ found that she could do other jobs such as parking lot attendant, lobby attendant, and interviewer. Pursuant to SSR 00-4p, the ALJ found the VE's testimony consistent with the DOT. He therefore found plaintiff not disabled. (Tr. at 728.)

On November 12, 2013, the Appeals Council denied review. (Tr. at 692.) Plaintiff then commenced this action.

II. DISCUSSION

A. Standard of Review

If the ALJ's decision is supported by substantial evidence, the court on judicial review must uphold that decision even if the court might have decided the case differently in the first instance. Beardsley v. Colvin, No. 13-3609, 2014 WL 3361073, at *2 (7^th Cir. July 10, 2014). Substantial evidence means "'such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.'" Id. (quoting Richardson v. Perales, 402 U.S. 389, 401 (1971)). To determine whether substantial evidence exists, the court reviews the record as a whole but does not attempt to substitute its judgment for the ALJ's by re-weighing the evidence, resolving material conflicts, or reconsidering facts or the credibility of witnesses. Id. Where conflicting evidence would allow reasonable minds to differ as to whether the claimant is entitled to benefits, the court must defer to the ALJ's resolution of that conflict. Id. A reversal and remand may be required, however, if the ALJ committed an error of law, based the decision on serious factual mistakes or omissions, or failed to build an accurate and logical bridge between the evidence and the result. Id. Finally, because judicial review is confined to the reasons supplied by the ALJ, the Commissioner's lawyers may not later fill in gaps in the ALJ's analysis. See Steele v. Barnhart, 290 F.3d 936, 941 (7^th Cir. 2002).

B. Analysis

Plaintiff argues that the ALJ (1) credited unreliable vocational testimony and (2) overlooked or misinterpreted medical evidence supporting her claim. I address each contention in turn.

1. Vocational Evidence

a. Applicable Legal Standard

Where, as here, the claimant cannot because of her severe impairments perform her past relevant work, the Commissioner bears the burden of establishing that the claimant can perform other work that exists in significant numbers in the national economy. Overman, 546 F.3d at 464. "A VE's testimony can satisfy this burden only if that testimony is reliable." Id. "A finding based on unreliable VE testimony is equivalent to a finding that is not supported by substantial evidence and must be vacated." Britton v. Astrue, 521 F.3d 799, 803 (7^th Cir. 2008).

b. Analysis

Everyone agrees that the VE's testimony in this case conflicted with the DOT. The ALJ asked the VE to identify light jobs requiring occasional handling and fingering, and the VE came up with parking lot attendant, lobby attendant, and interviewer. As plaintiff pointed out after the hearing, however, according to the DOT/SCO all three of these jobs require frequent handling and fingering. The ALJ conceded the problem but addressed it by concluding that plaintiff could perform frequent handling and fingering after all. Remarkably, the ALJ later stated that the VE's "testimony is consistent with the information contained in the Dictionary of Occupational Titles." (Tr. at 728.)

SSR 00-4p states: "When there is an apparent unresolved conflict between VE . . . evidence and the DOT, the adjudicator must elicit a reasonable explanation for the conflict before relying on the VE . . . evidence to support a determination or decision about whether the claimant is disabled." 2000 WL 1898704, at *2. Nothing in the Ruling suggests that the ALJ may eliminate a conflict simply by adopting a different RFC.

The Commissioner responds that it was appropriate for the ALJ to formulate the ultimate RFC after the hearing, that his determination is supported by substantial evidence, and that there is no conflict between the ALJ's RFC finding and the jobs identified by the VE. The Commissioner's attempt to recast this as an RFC issue misses the point. SSR 00-4p seeks to maintain consistency between VE testimony and the DOT not only to ensure that the ALJ has before him probative evidence of jobs that the claimant can actually do but also as a means of ensuring the reliability of the VE's testimony. See Carey v. Apfel, 230 F.3d 131, 147 (5^th Cir. 2000) ("Certainly, a vocational expert's erroneous characterization of the exertional level or skills required to perform a particular job calls into question both the probative value and reliability of the expert's testimony."). In this case, the ALJ failed to consider whether the VE's mistakes impacted his credibility.

The ALJ also impeded plaintiff's efforts to challenge the basis for the VE's conclusions. Prior to the hearing, plaintiff demanded the VE's data, but the ALJ denied that request, finding it sufficient for counsel to cross-examine the expert about his sources at the hearing. The Seventh Circuit has ruled that the data underlying the VE's opinion "must be 'available on demand' if the claimant challenges the foundation of the vocational expert's opinions." McKinnie, 368 F.3d at 911 (quoting Donahue, 279 F.3d at 446). The Commissioner responds that this claim is moot because plaintiff's counsel was able to anticipate the sources upon which the VE would rely. However, it appears that counsel had with him at the hearing just one of the primary sources the VE used (the Bureau of Labor Statistics report), not the other (the U.S. Publishing report). It was only after the hearing that counsel was able to submit a sample of the second source. The Seventh Circuit has explained that the "data underlying a VE's testimony must be available on demand to facilitate cross-examination." Britton, 521 F.3d at 804. Pre-hearing guesswork and post-hearing written submissions are no substitute for counsel's ability to ask the VE precisely where in his data he found evidence of a specific number of jobs.

Even if a post-hearing written submission could suffice, in this case the ALJ failed to meaningfully address plaintiff's challenges to the data the VE said he used. Plaintiff presented a report from Traver including a detailed critique of the method U.S. Publishing uses in its Occupational Employment Quarterly ("OEQ"), but the ALJ dismissed this as a mere disagreement among experts, choosing to "credit the impartial expert over Mr. Traver." (Tr. at 727.) The ALJ noted that "vocational testimony is not an exact science" and that given the "ever changing economic dynamics in this country," "there is considerable room for debate as to who is correct and who is not." (Tr. at 727.) ALJs are required to provide some explanation as to why they credit one opinion over another. See O'Connor-Spinner v. Astrue, 627 F.3d 614, 621 (7^th Cir. 2010) ("An ALJ must explain why he does not credit evidence that would support strongly a claim of disability, or why he concludes that such evidence is outweighed by other evidence."). The ALJ concluded that "the VE was testifying as a vocational expert and not a statistician or census taker. And his information came from reliable sources, indeed, some of which the undersigned is required to take administrative notice." (Tr. at 727.) While the ALJ was required to take administrative notice of the government source the VE cited, U.S. Publishing is not, as the VE mistakenly believed, a governmental agency. In any event, the ALJ also failed to address Traver's doubts as to how specific DOT codes could be pulled from the government sources. (See Tr. at 1062: "We are not aware of any data source or methodology for reliably determining the number of jobs by DOT code." (quoting BLS official).)

Perhaps these apparent flaws in the data could be overlooked if the VE had demonstrated a reliable method for extracting specific numbers of jobs from that data. But his testimony regarding his methodology was vague. The VE explained that he looked at the available date and tried "to cluster it around what we know based on DOT's classifications of jobs and how they fit in to some of the clusters that makes some logical sense, at least as far as providing some kind of predictor around jobs." (Tr. at 1248.) Pressed on how he came up with such specific numbers, the VE said "you do the best job as a vocational expert using your background and experience to at least provide some predictability around what jobs are available." (Tr. at 1249.) Pointing out that neither of the sources the VE cited used his "clustering method," counsel asked what source he could consult to see the clusters the VE was using. The VE responded, "Well, some of what I do is proprietary." (Tr. at 1250.) Pressed to describe that proprietary methodology, the VE essentially restated his previous testimony: "Well, again, I take the best data that's available and break it down into jobs that have similar physical requirements and occupational demands. So when I talked about things like receptionists and information clerks, I'm talking about jobs that cluster around that same type of field." (Tr. at 1251.) The ALJ concluded: "Like it or not, the vocational expert was able to explain how we arrived at those numbers." (Tr. at 727.)

"An expert must offer good reason to think that his approach produces an accurate estimate using professional methods, and this estimate must be testable. Someone else using the same data and methods must be able to replicate the result." Zenith Electronics Corp. v. WH-TV Broadcasting Corp., 395 F.3d 416, 419 (7^th Cir. 2005). It is not sufficient for an expert to simply invoke his own expertise or experience. Id. A proposed expert who cannot or will not explain his conclusions should not be allowed to give expert testimony. Id.; see, e.g., Veryzer v. Sec'y, HHS, No. 06-522V, 2010 WL 2507791, at *23-25 (Fed. Cl. June 15, 2010) (granting motion to exclude an expert witness who refused to explain how his causation theory worked, describing that information as "proprietary").

Given the VE's admitted contradiction of the DOT, the ALJ's refusal to require production of the VE's data, the serious doubts plaintiff's expert raised about the data the VE ostensibly relied upon, and the VE's inability to cogently explain his method of drawing conclusions from that data, I cannot find the ALJ's step five determination supported by substantial evidence. The matter must be remanded for a new step five determination based on reliable vocational testimony, obtained in compliance with Seventh Circuit case-law.

2. Medical Evidence

a. Applicable Legal Standards

"An ALJ has the obligation to consider all relevant medical evidence and cannot simply cherry-pick facts that support a finding of non-disability while ignoring evidence that points to a disability finding." Denton v. Astrue, 596 F.3d 419, 425 (7^th Cir. 2010). However, this does not mean that the ALJ is required to discuss every piece of evidence in the record; he is prohibited only from ignoring an entire line of evidence that supports a finding of disability. Jones v. Astrue, 623 F.3d 1155, 1162 (7^th Cir. 2010).

Opinions from a social security claimant's treating physician are entitled to "special significance." SSR 96-8p, 1996 WL 374184, at *7. Such opinions must be given "controlling weight" if well-supported by medically acceptable clinical and laboratory diagnostic techniques and not inconsistent with the other substantial evidence in the record. 20 C.F.R. § 404.1527(c)(2). Treating source opinions that do not meet the test for controlling weight are still entitled to deference and must be weighed according to a checklist of factors, see SSR 96-2p, 1996 WL 374188, at *4, including the length, nature, and extent of the treatment relationship; frequency of examination; the physician's specialty; the types of tests performed; and the consistency and supportability of the physician's opinion. 20 C.F.R. § 404.1527(c)(2); Scott v. Astrue, 647 F.3d 734, 739 (7^th Cir. 2011). The ALJ must always provide "good reasons" for discounting the opinion of a treating physician. Scott, 647 F.3d at 739. The ALJ is also required to consider any opinions in the record from state agency medical and psychological consultants, as they "are highly qualified physicians and psychologists who are experts in the evaluation of the medical issues in disability claims under the Act." SSR 96-6p, 1996 WL 374180, at *2. In weighing these opinions, the ALJ must consider the supportability of the opinion in the evidence, the consistency of the opinion with the record as a whole, any explanation for the opinion provided by the consultant, and any specialization of the consultant. Id. Generally, more weight is given to the opinion of a source who examined the claimant than to the opinion of a source who did not. 20 C.F.R. § 404.1527(c)(2).

b. Analysis

Plaintiff argues that the ALJ overlooked medical evidence regarding her degenerative disc disease, failed to address Dr. Johnston's January 2007 report, improperly determined that she could use her hands frequently, and erred in his evaluation of Dr. Shields's report.

i. Degenerative Disc Disease

Plaintiff first challenges the ALJ's characterization of the severity of her degenerative disc disease, citing various pieces of diagnostic and neurological evidence. As noted, however, the "ALJ is not required to discuss every snippet of information from the medical records that might be inconsistent with the rest of the objective medical evidence." Pepper v. Colvin, 712 F.3d 351, 363 (7^th Cir. 2013). Plaintiff makes no allegation that the ALJ ignored an entire line of contrary evidence, and I am not authorized to re-weigh the evidence myself.

ii. Dr. Johnston's Report

Plaintiff next contends that the ALJ erred by failing to discuss Dr. Johnston's January 2007 report, in which the doctor opined that plaintiff could not sustain full-time sedentary or light work, even with a sit/stand option. (Tr. at 481-82.) Dr. Johnston further opined that, due to pain and other symptoms, plaintiff had a moderately severe limitation in her ability to maintain attention and concentration for extended periods, perform activities within a schedule, and complete a workday without interruptions based on her symptoms. (Tr. at 483.) Plaintiff admits that ALJ Malloy incorporated by reference ALJ Kingrey's previous evaluation of the medical evidence (Tr. at 715), which included a rejection of Johnston's report (Tr. at 746), but argues that reliance on the prior decision must fail because that decision was vacated by the Appeals Council. The Commissioner responds that the district court found no fault with ALJ Kingrey's rejection of the Johnston report, as ALJ Malloy also noted (Tr. at 715); since the district court's remand order did not require re-weighing of Dr. Johnston's report, ALJ Malloy did not err by failing to mention it. In reply, plaintiff concedes that the district court found ALJ Kingrey's analysis of the Johnston report sufficient but contends that after the district court issued its decision the Appeals Council vacated ALJ Kingrey's entire decision. Plaintiff points to a provision in the HALLEX stating: "When the Appeals Council vacates a final decision of the Commissioner, the ALJ must consider all pertinent issues de novo."

The HALLEX does not specify what "pertinent" means in this context. I therefore turn to the actual order from the Council in this case, which states:

The Appeals Council hereby vacates the final decision of the Commissioner of Social Security and remands this case to an Administrative Law Judge for further proceedings consistent with the order of the court.



In compliance with the above, the Administrative Law Judge will offer the claimant the opportunity for a hearing, take any further action needed to complete the administrative record and issue a new decision.

iii. Frequent Use of the Hands

Plaintiff's challenge to the ALJ's finding regarding use of the hands gains traction. As indicated above, the ALJ admitted that he made this finding in order to "moot" plaintiff's challenge to the VE's testimony:

Much was made at and following the hearing as to the vocational expert's response to the hypothetical questions posed in which he was asked to assume no more than occasional handling or fingering, the argument being made that the jobs he cited would involve frequent handling and fingering. To make that a moot point, the undersigned finds that claimant would be capable of frequent, but not constant finding or handling.

Perhaps this result-oriented approach could be overlooked if the ALJ had provided a convincing explanation for his conclusion, but he did not. The entirety of his analysis was as follows:

[The ALJ bases this finding on] the fact that there is no rheumatoid arthritis and whatever degenerative changes claimant may have, they would be mild. An EMG-nerve conduction velocity study done in November of 2006 did [show] "minimal" right median sensory neuropathy and "borderline" right ulnar sensory neuropathy, the comprehensive testing otherwise normal.

As plaintiff notes, the record also contains evidence from Dr. Greene, a rheumatologist, who in June 2005 diagnosed degenerative joint disease of the hands with inflammatory arthritis overlap (Tr. at 520); Dr. Melzer, the examining consultant, who diagnosed osteoarthritis in 2011 (Tr. at 902); and Dr. Ryzka, another rheumatologist, who in 2012 noted nodular changes in the joints of the hands consistent with degenerative disease (Tr. at 1035). At the hearing, plaintiff testified to limited use of her hands, including problems typing, writing, and grasping. (Tr. at 1209, 1225.) The Commissioner attempts to bolster the ALJ's conclusion by citing additional evidence of normal strength, grip, and range of motion, but my review is limited to the reasons provided by the ALJ. See Kastner, 697 F.3d at 648. In any event, this evidence of normal strength and range of motion does not necessarily account for limitations caused by pain. See id. at 649. The ALJ must revisit this issue on remand.

iv. Dr. Shields

Finally, plaintiff challenges ALJ Malloy's rejection of Dr. Shields's consultative report. As indicated, Dr. Shields conducted a neuropsychological evaluation, which included a battery of tests measuring plaintiff's intelligence, academic skills, and memory. He concluded that plaintiff was "likely to have moderate to severe impairment in her capacity to sustain concentration, persistence, and pace over the course of a full-time work week." (Tr. at 552.) He assessed major depressive disorder, recurrent, moderate to severe, without psychotic features; anxiety disorder; learning disorder (rule out reading disorder); psychological factors affecting pain syndrome; and histrionic and borderline personality traits, with a GAF of 50-55. (Tr. at 552.) He noted IQ scores in the low average and borderline range (Tr. at 554), reading scores in the extremely low range, with suspected dyslexia (Tr. at 551, 558), and spelling and arithmetic scores in the average and low average range (Tr. at 552, 558). In a medical source statement, Dr. Shields endorsed slight limitation in understanding and carrying out simple instructions, moderate limitation in understanding and carrying out detailed instructions, and no limitation in making judgments on simple work related decisions. (Tr. at 562.) He endorsed no limitation in plaintiff's ability to interact appropriately with the public, supervisors, and coworkers, and moderate limitation in her ability to respond to work pressures and respond appropriately to changes. He further noted limited energy levels, with moderate fatigue after four hours of testing. (Tr. at 563.)

ALJ Kingrey largely accepted Dr. Shields's conclusions and incorporated them into her RFC. (Tr. at 747.) ALJ Malloy disagreed, finding the report "questionable." (Tr. at 726.) First, he found the academic testing inconsistent with plaintiff's ability to graduate high school in regular classes, work as a bookkeeper, and operate a business. Second, ALJ Malloy criticized Dr. Shields for discussing medical records, a subject on which he had no expertise. Third, ALJ Malloy stated that Dr. Shields "seemingly appeared to accept many of claimant's various statements and discussion of her history at face value." (Tr. at 716.) "[G]iven the circumstances surrounding the referral to him by Judge Kingery, it should have at least prompted [Dr. Shields] to be a little suspicious, especially given claimant's past academic . . . and employment . . . history." (Tr. at 726.) ALJ Malloy found plaintiff capable of unskilled and low level semi-skilled tasks, noting that the Wisconsin consultants' reports suggested such capacity. (Tr. at 726.)

While it was within ALJ Malloy's discretion to credit the Wisconsin consultants over Dr. Shields, the record does not support his conclusion that Dr. Shields relied on plaintiff's subjective complaints in reaching his conclusions. In addition to a clinical interview, Dr. Shields conducted eight different tests in evaluating plaintiff's mental abilities. (Tr. at 545.) He also reviewed a significant number of treatment records (Tr. at 545-46), finding plaintiff's self-report "fairly consistent with the information obtained from available collateral information." (Tr. at 552.) Dr. Shields also noted that plaintiff exhibited adequate effort during the testing, making the results a "fairly reliable indication of her current intellectual functioning and memory skills." (Tr. at 552.) Dr. Shields examined plaintiff on behalf of the agency; he was not her doctor, so there is no reason to believe he would have slanted his opinion in her favor. See Garcia v. Colvin, 741 F.3d 758, 761 (7^th Cir. 2013). Absent some indication in Dr. Shields's report that he improperly relied on plaintiff's subjective claims, it was error for the ALJ to so assume.

III. CONCLUSION

Ordinarily, the district court will on finding the agency's denial unsupported or unexplained remand the matter for further proceedings; a direct award of benefits is proper only if all factual issues have been resolved and the record yields but one supportable conclusion. Allord v. Astrue, 631 F.3d 411, 415 (7^th Cir. 2011); Briscoe ex rel. Taylor v. Barnhart, 425 F.3d 345, 355 (7^th Cir. 2005). This case must be remanded for a fresh step five determination based on reliable vocational evidence and for reconsideration of the evidence regarding plaintiff's manipulative and mental limitations. The ALJ must in constructing hypothetical questions and determining RFC account for any limitations in concentration, persistence, and pace. See O'Connor-Spinner, 627 F.3d at 619 (indicating that the ALJ must orient to the VE to all of the claimant's limitations, including those in concentration, persistence, and pace, and that limiting the claimant to simple, repetitive tasks will not necessarily suffice). Finally, I urge the Commissioner to assign the matter to a different ALJ on remand, as the manner in which ALJ Malloy resolved the issues discussed in this decision suggests a "commitment to the denial of this applicant's claim." Sarchet, 78 F.3d at 309.

THEREFORE, IT IS ORDERED that the ALJ's decision is REVERSED, and this case is REMANDED to the Commissioner for further proceedings consistent with this decision. The Clerk is directed to enter judgment accordingly.

Dated at Milwaukee, Wisconsin this 19^th day of August, 2014.

/s Lynn Adelman

LYNN ADELMAN

District Judge