LYNN ADELMAN District Judge

DECISION AND ORDER

Plaintiff Kim Lang hurt her back at work on August 5, 2008; she briefly returned to work in the summer of 2009 but soon re-injured her back and did not return to work thereafter. She applied for social security benefits in November 2010, alleging disability since August 2008 based primarily on her back impairment. The Administrative Law Judge ("ALJ") assigned to the case found plaintiff disabled from August 5, 2008 to December 28, 2009, concluding that her back condition had improved by the end of that period, enabling her to work. Plaintiff challenged that determination in an action for judicial review, and on the parties' stipulation the court remanded the case for further proceedings. On remand, the ALJ re-affirmed the previous closed period determination and additionally awarded benefits from November 27, 2012 to February 25, 2014, based on a post-application knee injury. However, he found that neither plaintiff's back problem nor her knee injury rendered her disabled thereafter.

Plaintiff seeks judicial review of this partially favorable determination. I first set forth the applicable legal standards, then review the extensive medical and procedural record, and, finally, address the parties' arguments regarding the ALJ's decision.

I. LEGAL STANDARDS

A. Disability Determination

Social security regulations set forth a sequential, five-step test for determining whether a claimant is disabled. E.g., Varga v. Colvin, 794 F.3d 809, 812 (7^th Cir. 2015) (citing 20 C.F.R. § 404.1520). Under this test, the ALJ asks:

(1) Is the claimant currently working, i.e., doing substantial gainful activity? If so, she is not disabled.

(2) If not, does the claimant have a severe medically determinable physical or mental impairment? If not, the claimant is not disabled.

(3) If so, does the claimant's impairment meet or equal one of the presumptively disabling impairments set forth in the agency's Listings? If so, the claimant is disabled.

(4) If not, does the claimant retain the residual functional capacity ("RFC") to perform her past relevant work? If so, she is not disabled.

(5) If not, can the claimant, based on her RFC, age, education, and work experience make an adjustment to other work? If so, she is not disabled. If not, she is disabled. See 20 C.F.R. § 404.1520(a)(4).

In some cases, the ALJ may find the claimant disabled for a finite period of time, i.e., a "closed period." See Shepherd v. Apfel, 184 F.3d 1196, 1198 (10^th Cir. 1999). "Before limiting benefits to a closed period, an ALJ must conclude either that a claimant experienced 'medical improvement' as evidenced by changes in the symptoms, signs, or test results associated with her impairments, or else that an exception to this rule applies." Tumminaro v. Astrue, 671 F.3d 629, 633 (7^th Cir. 2011) (citing 20 C.F.R. § 404.1594). In this situation, the ALJ applies an eight-step test, see Tumminaro, 671 F.3d at 629, asking:

(1) Is the claimant engaging in substantial gainful activity? If so, disability ends.

(2) If not, does the claimant have an impairment which meets or equals the severity of an impairment set forth in the Listings? If so, disability continues.

(3) If not, has there been medical improvement? If there has been medical improvement as shown by a decrease in medical severity, proceed to step 4. If there has been no decrease in medical severity, proceed to step 5.

(4) If there has been medical improvement, is it related to the claimant's ability to do work, i.e., has there been an increase in RFC based on the impairment(s) present at the time of the most recent favorable determination? If medical improvement is not related to the ability to do work, proceed to step 5. If medical improvement is related to the ability to do work, proceed to step 6.

(5) If the ALJ found at step 3 that there has been no medical improvement or at step 4 that the medical improvement is not related to the claimant's ability to work, does an exception apply? If not, disability continues.

(6) If the medical improvement is related to the ability work, are the claimant's current impairments severe? If not, disability ends.

(7) If the current impairments are severe, will the claimant's current RFC permit the performance of her past work? If so, disability ends.

(8) If the claimant cannot perform past work, is she able, given her current RFC, age, education, and experience, to perform other work? If so, disability ends. If not, disability continues. See 20 C.F.R. § 404.1594(f).

B. Standard of Review

The court will uphold an ALJ's decision if he applied the correct legal standards and supported his decision with "substantial evidence." Bates v. Colvin, 736 F.3d 1093, 1097 (7^th Cir. 2013). Substantial evidence means such relevant evidence as a reasonable mind might accept as adequate to support a conclusion. E.g., Brown v. Colvin, 845 F.3d 247, 251 (7^th Cir. 2016). While the reviewing court will not, under this deferential standard, re-weigh the evidence or substitute its judgment for that of the ALJ, Shideler v. Astrue, 688 F.3d 306, 310 (7^th Cir. 2012), the ALJ must provide a logical bridge between the evidence and his conclusions. Varga, 794 F.3d at 813.

In reaching his decision, the ALJ must consider all medical opinions in the record. Roddy v. Astrue, 705 F.3d 631, 636 (7^th Cir. 2013). Opinions from the claimant's treating physician are entitled to "special significance," SSR 96-8p, 1996 SSR LEXIS 5, at *20, and will 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). If a treating source opinion does not meet the test for controlling weight, the ALJ must determine what value the opinion does merit, considering 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. Scott v. Astrue, 647 F.3d 734, 740 (7^th Cir. 2011). The reviewing court must ensure that the ALJ provided "good reasons" for discounting the opinion of a treating physician. See id. at 739.

The ALJ is also required to consider the claimant's statements regarding her symptoms and their effect on her ability to work. In evaluating a claimant's statements, the ALJ must first determine whether the claimant suffers from a medically determinable impairment that could reasonably be expected to produce her symptoms. SSR 16-3p, 2016 SSR LEXIS 4, at *5; SSR 96-7p, 1996 SSR LEXIS 4, at *5. If the claimant has such an impairment, the ALJ must then evaluate the intensity and persistence of the symptoms to determine the extent to which they limit the claimant's ability to work. SSR 16-3p, 2016 SSR LEXIS 4, at *9; SSR 96-7p, 1996 SSR LEXIS 4, at *5-6. At this second step, "the absence of objective medical corroboration for a complainant's subjective accounts of pain does not permit an ALJ to disregard those accounts." Ghiselli v. Colvin, 837 F.3d 771, 777 (7^th Cir. 2016). Rather, once the claimant has demonstrated the existence of an impairment that can reasonably be expected to produce the symptoms, the ALJ must evaluate the intensity, persistence, and limiting effects of those symptoms based on the entire record, considering the claimant's daily activities; the location, duration, frequency, and intensity of the symptoms; factors that precipitate and aggravate the symptoms; the type, dosage, effectiveness, and side effects of any medication the claimant takes; other treatment or measures the claimant receives or uses to relieve the symptoms; and any other factors concerning the claimant's functional limitations due to the symptoms. SSR 16-3p, 2016 SSR LEXIS 4, at *18-19; SSR 96-7p, 1996 SSR LEXIS 4, at *8. The ALJ must then provide specific reasons for his credibility determination, supported by substantial evidence in the record. See, e.g., Israel v. Colvin, 840 F.3d 432, 441 (7^th Cir. 2016).

On review, an ALJ's credibility determination is entitled to deference and will be overturned only if it is patently wrong. Bates, 736 F.3d at 1098. However, if the credibility finding rests on objective factors or fundamental implausibilities, rather than on the claimant's demeanor or other subjective factors, the court has greater leeway to evaluate the ALJ's determination. Id.

II. FACTS AND BACKGROUND

A. Medical Evidence

Plaintiff underwent back surgery with fusion at the L5-S1 level in April 1997 after a motor vehicle accident, resulting in resolution of her symptoms. (Tr. at 265, 324-26.) On August 5, 2008, plaintiff (re)injured her back lifting a large patient while working as a certified nursing assistant ("CNA"). Assessed with a lumbar strain, her primary physician, Dr. Walter Fuhr, prescribed Percocet and Flexeril. (Tr. at 352-53.) Plaintiff attempted physical therapy but was unable to tolerate it due to pain and so was sent to a pain management specialist. (Tr. at 361.)

On September 11, 2008, plaintiff saw Dr. Harry Tagalakis at Advanced Pain Management, with a chief complaint of left side low back pain, radiating into her left leg. (Tr. at 336.) Dr. Tagalakis assessed lumbosacral radiculopathy and degeneration of the lumbosacral disc, providing an epidural steroid injection. A lumbar MRI revealed a disc protrusion at L5-S1, as well as neural foraminal stenosis above this level. (Tr. at 338-41.) Dr. Tagalakis provided an additional injection on September 28, 2008, also prescribing Lyrica. (Tr. at 342-43.)

On November 7, 2008, plaintiff saw Dr. Max Lee for a neurological consultation, the injections having failed to resolve her pain. (Tr. at 364.) Based on his exam, Dr. Lee felt her pain generated not from the lumbar spine but the sacroiliac joint. He provided an SI joint injection, with a dramatic amount of pain relief. (Tr. at 365-67.)

On November 26, 2008, plaintiff saw Dr. James Hollowell for evaluation of her back pain, noting that her back remained quite good following the 1997 surgery until her recent work related injury. She reported her pain varied from 7 to 10 out of 10, mostly in the left low back/buttock region, occasionally down the left leg. She had difficulty walking. She had received injections, which were not helpful. She had also been given Percocet, but it caused her to be nauseated. She had been unable to work since the injury. (Tr. at 265.) Dr. Hollowell believed it possible plaintiff sustained a disruption to the prior fusion, ordering dynamic films, a bone scan, and CT scan. She was given a release to be completely off work and had been provided Vicodin to provide pain relief with fewer nausea symptoms. (Tr. at 266.)

On December 31, 2008, plaintiff returned to Dr. Hollowell, for follow up after the additional imaging, which collectively demonstrated fracture of the upper portion of the pedicle fixation screw on the left at the S1 level, possibly at the L5 level as well. (Tr. at 264; see also Tr. at 227, 228, 230.) Dr. Hollowell suspected disruption of the fixation device, although this did not appear to be accompanied by disruption of the L5-S1 fusion. He found that she may be a candidate for an SI joint injection and should remain off work. (Tr. at 264.) On January 22, 2009, Dr. Gordon Mortenson performed a left SI joint injection. (Tr. at 398.)

On February 18, 2009, plaintiff returned to Dr. Hollowell, noting no improvement from the injection. Dr. Hollowell believed her pain likely unrelated to her hardware failure. The fusion appeared intact, and it was possible the hardware failure occurred quite some time ago and was entirely unrelated. He thought it reasonable she have some facet injections. (Tr. at 257.) On February 26, Dr. Mortenson performed a facet injection at L3-4. (Tr. at 399.)

On March 20, 2009, plaintiff advised Dr. Hollowell she continued to have similar symptoms; the facet injection did not offer any significant relief. Dr. Hollowell indicated that plaintiff should be permitted to return to part-time, light duty work, for four hours/day, three days/week, lifting no greater than ten pounds, avoiding prolonged sitting, and changing positions every 30 minutes. (Tr. at 256.)

On March 24, 2009, Dr. Mortenson performed another L4-5 facet injection (Tr. at 400), and on April 2 a hardware block to the left fusion hardware. (Tr. at 401.) On April 22, Dr. Hollowell suggested removal of the instrumentation, which sometimes offered good relief of symptoms. (Tr. at 254.) On May 18, Dr. Hollowell removed the fractured hardware. (Tr. at 219-20.)

On June 3, 2009, plaintiff followed up with Dr. Hollowell's assistant, reporting some degree of post-operative pain, with continued discomfort in the left low back and superior buttock region. She appeared to be progressing, and they initiated physical therapy. She would remain off work. They switched from Percocet to Norco. Plaintiff was encouraged to substantially increase her activity overall, particularly walking. (Tr. at 252, 369.)

On July 8, 2009, plaintiff returned to Dr. Hollowell, reporting she was better overall regarding low back pain, which she rated 4/10, improved by about 50%, mostly in the left buttock. Sometimes the pain went up into the medial scapular region and neck. She also had increased headaches. Dr. Hollowell indicated that plaintiff had made good progress with regard to her lumbar spine but unfortunately had other problems that had overwhelmed her low back pain. He prescribed Flexeril and indicated that she should continue with therapy and work in a part-time, light duty capacity (ten pounds, four hours/day, three days/week). (Tr. at 263, 370.)

On August 5, 2009, plaintiff saw Dr. Hollowell's assistant, reporting overall improvement but fairly persistent left superior buttock pain. The symptoms extended up her back, into the thoracic and cervical region, associated with a great deal of muscle spasm. They increased her work schedule to four hours/day, four days/week for one week, with a subsequent increase to four hours/day, five days/week. She would remain with a ten pound lifting restriction and continue with physical therapy. (Tr. at 262, 371-72.)

On September 2, 2009, plaintiff advised Dr. Hollowell's assistant that she had re-injured her back at work on August 31, 2009. On exam, she rose from the chair slowly, with clear discomfort, and exhibited an antalgic gait. They ordered updated imaging, took her off work, put therapy on hold, and gave her an updated prescription for Vicodin. (Tr. at 261, 373.)

On September 18, 2009, plaintiff reported fairly significant pain but felt she was overall improved. A bone scan noted some minimal stress changes, progressed since the bone scan done pre-operatively. A lumbar MRI again showed significant facet hypertrophy above her fusion at L4-5. (Tr. at 259; see also Tr. at 233, 236.) On exam, she was clearly uncomfortable rising from a chair and exhibited a left antalgic gait. Frustrated by her current symptoms, she was to revisit her pain concerns with Dr. Mortenson and remain off work. (Tr. at 259-60.)

On October 1, 2009, plaintiff advised Dr. Mortenson that she had been doing reasonably well until she re-injured her back at work. Dr. Mortenson provided Lyrica and Lidoderm patches. (Tr. at 402.) On November 2, plaintiff told Dr. Mortenson the Lidoderm did nothing for her and the Lyrica was too strong to take during the day so she only took it at night. Dr. Mortenson asked her to participate in some physical therapy and provided Tramadol. (Tr. at 403.)

On November 4, 2009, plaintiff saw Dr. Hollowell's assistant. She remained off work at the time, taking Lyrica and Vicodin. On exam, she had difficulty sitting squarely due to pain in the left low back and buttock, and continued to exhibit a left antalgic gait. (Tr. at 258.) She was continued on Vicodin and provided a Flector patch. (Tr. at 258.) On December 7, Dr. Mortenson listed an impression of neuritis and scheduled a cluneal nerve block. (Tr. at 404.)

During her December 16, 2009, follow up with Dr. Hollowell, plaintiff reported her condition was about the same. On exam, she rose slowly and walked deliberately, with a left antalgic component. She was able to partially squat and rise, and walk on her heels and toes, though this caused significant pain. The left lumbar region was exquisitely tender to palpation. There was no palpable abnormality or overlying changes of the skin. Dr. Hollowell recommended she reintegrate to work on a very gradual basis, four hours/day, three days/week, lifting no greater than ten pounds. He noted: "She seems to be quite opposed to this; overall, I think it is in her best interest." (Tr. at 250.) Irrespective of her response to the injection scheduled by Dr. Mortenson, Dr. Hollowell thought it appropriate for her to do some minimal amount of light duty work. (Tr. at 250, 374.) On December 17, Dr. Mortenson performed the nerve block. (Tr. at 405.)

On December 22, 2009, plaintiff underwent an independent medical evaluation with Dr. Richard Karr arranged by her employer's workers' compensation insurance carrier. (Tr. at 376.) Dr. Karr indicated on that on exam plaintiff exhibited pain behaviors, including grimacing, antalgic posturing, tearfulness, and audible pain complaints in response to benign examination methods, with several positive "Waddell signs." (Tr. at 385.) Dr. Karr opined that plaintiff had reached maximum medical improvement following her work related injuries, with 5% permanent partial disability to the body. (Tr. at 387.) He found that the could return to full-time work with the following restrictions: maximum lifting 35 pounds, maximum repetitive lifting/carrying 20 pounds, and no repetitive or prolonged bending at the waist beyond 45 degrees. (Tr. at 389.)

On February 18, 2010, plaintiff saw Dr. Mortenson, reporting no significant improvement from the cluneal nerve block. Physical therapy had helped with the headaches, upper back pain, and thoracic pain, but she continued to have pain in the buttocks area, with severe tenderness to palpation in the area of the piriformis, left SI joint, and cluneal nerve. Dr. Mortenson provided hydrocodone and asked her to think about the possibility of a spinal cord stimulator. (Tr. at 406.)

On April 15, 2010, plaintiff reported that physical therapy helped a great deal, but she still experienced severe pain that interfered with her daily activities. She was not able to do laundry or cook for any length of time. Dr. Mortenson indicated that she appeared to have hit maximum healing and was "likely not going to be able to work." (Tr. at 407.) On June 15, plaintiff rated her pain as 7 to 8 out of 10, with some sharp stabbing pains that radiated from her back all the way to her neck. Dr. Mortenson assessed recurrent and severe low back pain, asking her to set up an appointment for chiropractic care. (Tr. at 408.) On August 6, plaintiff again reported continued significant pain. She had tried multiple medications, including Vicodin, Tramadol, Neurontin, and Lyrica, which were either ineffective or caused nausea. Dr. Mortenson assessed chronic left low back pain secondary to a work injury, uncertain etiology, providing ibuprofen, and filled out her restrictions. (Tr. at 409.)

On October 18, 2010, plaintiff was seen for a physical therapy evaluation related to left knee pain, stating that she had fallen after her back gave out, causing her to twist her knee during the fall. She reported difficulty walking one to two blocks and standing five to ten minutes while cooking meals. She had decreased range of motion, decreased strength, and decreased functional mobility. (Tr. at 280.) The therapist planned to see her two times a week for five weeks. (Tr. at 281.)

On November 16, 2010, Dr. Mortenson prepared a letter stating that plaintiff had reached a healing plateau in April 2010, with permanent partial disability of 3% on top of her disability from the prior fusion. (Tr. 393.) In an attached physical capacity form dated August 6, 2010 (Tr. at 396), Dr. Mortenson opined that plaintiff could continuously sit for 45 minutes and stand for 60 minutes. In an eight hour working day, she could sit for less than two hours and stand/walk less than two hours. She could walk one block without rest or severe pain. She needed a job permitting shifting positions at will and sometimes needed to take unscheduled breaks. (Tr. at 394.) She could frequently lift less than ten pounds, occasionally lift ten pounds, and never bend or stoop. Her impairment would likely produce good days and bad days, with more than four absences per month. (Tr. at 395.)

On June 14, 2011, plaintiff saw Dr. Jonathan Berry for consultation regarding her knee pain. Plaintiff explained that she was bringing chicken in from the grill when her back gave out, causing her to fall and injure her left knee. She had injections and therapy, which did not help. Examination of the left knee revealed that she was exquisitely tender along the lateral joint line, but x-rays were normal. (Tr. at 414.) Dr. Berry suspected a torn lateral meniscus and so obtained an MRI. (Tr. at 415.)

On June 17, 2011, plaintiff returned to Dr. Mortenson with increasing pain, now radiating up into her head. She continued to use hydrocodone sparingly. Cyclobenzaprine did not help much. On exam, she had decreased range of motion of the neck and tenderness to palpation over the musculature in her trapezius and low back. Dr. Mortenson tried a TENS unit and switched her from cyclobenzaprine to Skelaxin. (Tr. at 315, 410.)

A June 23, 2011, left knee MRI revealed moderate to severe chondromalacia patella. (Tr. at 417.) Dr. Berry recommended an arthroscopy (Tr. at 418), which he performed on July 13 (Tr. at 420-21). On July 19, plaintiff came in for follow-up, using crutches and reporting significant pain. Dr. Berry's assistant provided a referral for physical therapy and prescribed Celebrex. (Tr. at 422.) Plaintiff started therapy on July 25 (Tr. at 423-24), and on September 27 returned to Dr. Berry reporting no more pain and very appreciative. (Tr. at 426.)

On October 27, 2011, plaintiff saw Dr. Mortenson with increasing pain radiating from her left SI region up her left back to her neck and head. She had failed on cyclobenzaprine and Skelaxin, and wanted to switch back from Percocet to hydrocodone. Dr. Mortenson assessed chronic low back pain with associated thoracic pain and headaches, referring her for physical therapy. (Tr. at 317, 411.)

On December 13, 2011, plaintiff returned to Dr. Berry, indicating that she had recently re-injured her knee when, after returning from Florida, she twisted as she fell over her daughter's clothes. Dr. Berry obtained an MRI (Tr. at 427), which revealed moderate joint effusion and small osteochondral defects unchanged from the previous examination (Tr. at 416). On January 3, 2012, Dr. Berry assured plaintiff that there did not appear to be any new findings on the MRI. She was to increase her activity as tolerated, ice and elevate as necessary. (Tr. at 428.)

On June 25, 2012, plaintiff saw Dr. Mortenson, continuing to report persistent left-sided low back pain, axial or myofascial in nature, which extended superiorly up the dorsal back into the neck and left sided posterior head. (Tr. at 319.) Dr. Mortenson continued her current medication regimen and ordered a cervical MRI (Tr. at 320), which revealed mild multi-level spurring, as well as bulging at C4-5. On August 22, plaintiff reported being stable on her current medication regimen with improvement in pain, increased function, and no significant side effects. On exam, range of cervical motion was mildly reduced. (Tr. at 322.) Dr. Mortenson assessed chronic low back pain following a work-related injury and persistent neck pain likely secondary to exacerbation of her multi-level degenerative spondylosis. He continued her medication regimen and referred her for chiropractic treatment. (Tr. at 323.)

In September 2012, plaintiff started treatment with Kenneth Koch, D.C. In a March 18, 2013, letter, Dr. Koch indicated that he treated plaintiff for chronic neck and upper back pain. She also suffered from low back pain, but because of the lumbar fusion surgery coupled with the inability of this area to tolerate even mild manual therapy he limited his efforts to the cervical and upper thoracic region. (Tr. at 429.) During his initial evaluation on September 5, 2012, plaintiff exhibited reduced range of motion of the low back. (Tr. at 430.)

On November 27, 2012, plaintiff saw Lori Hubacek, PA-C, for evaluation of her left knee, indicating she injured it on November 24, 2012. She stated she had been doing very well until that time when she was out dancing with friends and felt a significant pain in her left knee. Exam of the left knee revealed effusion and tenderness along the lateral aspect of the knee. X-rays obtained at the emergency room on November 25, 2012, showed no fracture or dislocation. PA Hubacek encouraged plaintiff to use ibuprofen, ice, and crutches as needed; they would consider an MRI if the symptoms did not improve. (Tr. at 842.)

A December 12, 2012, left knee MRI revealed large joint effusion, a high-grade partial-thickness tear, and a new cartilaginous defect at the articular surface of the lateral femoral condyle (Tr. at 830), and Dr. Berry subsequently performed arthroscopic surgery (Tr. at 690). On January 8, 2013, plaintiff returned for follow up, complaining of moderate pain but not using an assistive device. She elected to pursue physical therapy and was given a referral. (Tr. at 694.) On February 12, plaintiff advised that she continued to have significant pain and swelling and was not improving in physical therapy. (Tr. at 698.) Dr. Berry provided gabapentin. (Tr. at 699.)

On March 8, 2013, plaintiff saw Sally Vahovick, PA-C, Dr. Mortenson's assistant, reporting pain in the low back on the left side radiating up to the left side of the back into the neck, with no significant change since the last evaluation. (Tr. at 673.) PA Vahovick assessed chronic left-sided low back pain, continued medications at the current dose, and discussed implantable options for pain management, focusing on a neuro-stimulator. (Tr. at 674.)

On April 30, 2013, plaintiff returned to Dr. Berry for follow up of her left knee, continuing to have significant pain and constant swelling, unimproved with therapy. (Tr. at 703.) Dr. Berry recommended a second look arthroscopy (Tr. at 704), which he performed on May 8 (Tr. at 725). On May 14, plaintiff reported significantly more pain than with the previous surgery, ambulating with crutches. (Tr. at 725.) She was to start physical therapy, increase weight bearing as tolerated, remain off work, and return for follow up in three to four weeks. (Tr. at 726.) On June 11, plaintiff advised Dr. Berry that she continued to have significant swelling. (Tr. at 730.) Dr. Berry provided an injection of cortisone to decrease the inflammation. (Tr. at 731.) On July 9, plaintiff returned with continued left knee pain and swelling. She and Dr. Berry were both frustrated because she had these recurrent symptoms despite two knee scopes and a number of cortisone injections. (Tr. at 735.) Dr. Berry assessed continued internal derangement. He wanted to review her MRIs with other physicians to come up with a plan of action; she would continue taking anti-inflammatories and "gutting it out" until then. (Tr. at 736.) On August 20, Dr. Berry recommended a patellofemoral arthroplasty. (Tr. at 742.)

On September 10, 2013, plaintiff returned to PA Vahovick, reporting pain primarily in the left low back, radiating up to the left shoulder. There had been no significant change in her pain since the last visit. She was then taking Norco twice daily, which adequately controlled her pain and allowed her to maintain her current level of functioning. (Tr. at 671.) They continued medications at the current dose. (Tr. at 672.)

On September 16, 2013, Dr. Berry performed a left total knee arthroplasty. (Tr. at 768). Plaintiff returned for follow up on October 1, doing OK despite the fact that she did not tolerate pain medication. She was taking Tylenol and ibuprofen and ambulated with a walker. (Tr. at 768.) On October 29, plaintiff reported that her kneecap felt great, but she still had significant pain both medially and laterally in the knee. (Tr. at 773.) She had been working hard in therapy and, although her range of motion was stellar, she still struggled with pain. (Tr. at 773-74.) It was difficult for her to wear pants because of hyper-sensitivity of the skin. Because the pain seemed out of proportion, Dr. Berry recommended a trial of gabapentin. (Tr. at 774.) On November 26, plaintiff reported continued left knee pain (Tr. at 778), and Dr. Berry tried a lidocaine patch (Tr. at 779). On January 14, 2014, plaintiff reported a recent fall, which caused a setback. (Tr. at 787.) Exam of the right knee revealed effusion and tenderness medially and laterally. Exam of the left knee revealed a well-healed incision with excellent range of motion. Her biggest complaint was laterally at the "IT band." She was to continue physical therapy and try a stretch for the IT band. (Tr. at 789.)

On February 24, 2014, plaintiff returned to Dr. Berry, with some left lateral knee pain, otherwise doing well. She was very inconsistent with therapy secondary to a child in the hospital. She stated her right knee was bothering her a little bit but wanted to wait before an evaluation. On review of symptoms, she reported no back or joint pain. (Tr. at 794.) Her gait was normal. She continued to do well status post left total knee replacement, certainly better than pre-op. They would see her back as needed. (Tr. at 795.)

On March 21, 2014, plaintiff saw PA Vahovick, reporting no significant change in her back pain since the last visit. She was taking Norco twice daily, which adequately controlled her pain and enabled her to maintain her current level of functioning. (Tr. at 669.) On exam, she appeared seated comfortably in the chair, rose to standing with little difficulty, and ambulated with a normal gait over short distances. She did not display any aberrant pain behavior. They continued current medications. (Tr. at 670.)

On September 30, 2014, plaintiff reported left-sided low back pain radiating up into the left shoulder and neck. Overall, there had been a mild increase in her pain since the last evaluation. She had been undergoing chiropractic treatments with no overall improvement in her pain and wanted to try something different. (Tr. at 667.) On exam, she appeared seated comfortably in the chair, rose to standing with little difficulty, and ambulated with a normal gait over short distances. She was very tender over the left low back and SI joint, with diffuse myofascial tenderness throughout the left mid- and upper back. PA Vahovick continued current medications, tried Lidoderm and Flector patches to see if this helped, and started physical therapy. (Tr. at 668.)

On October 14, 2014, plaintiff was seen for left foot pain after she hit her toe on a door. (Tr. at 799-800.) Exam showed swelling at the second toe, and x-rays revealed a minimally displaced fracture of the proximal phalanx of the second toe. (Tr. at 800, 833.) She was fitted with a post-op shoe and instructed on "buddy taping" the great and second toes. (Tr. at 801.)

At her April 3, 2015, visit with PA Vahovick, plaintiff reported no significant change since the last evaluation. Her pain was improved with her current medications of Norco and Lidoderm patches. She continued to receive chiropractic treatments with some pain relief. She had tried physical therapy, which aggravated the pain. (Tr. at 665.) On exam, she was tender to palpation over the left lumbar paraspinal muscles, left latissimus dorsi, and left trapezius, but did not display any aberrant pain behavior. They continued her medications at the current dose. (Tr. at 666.)

On May 12, 2015, plaintiff returned to Dr. Berry, reporting a fall in March of 2014, with continued pain since that time; however, she did not come in to see Dr. Berry until then, 15 months after the fall. (Tr. at 806.) Exam of the knee revealed full range of motion, with some tenderness over the lateral aspect. X-rays looked perfect. (Tr. at 807; see also Tr. at 834.) Dr. Berry recommended physical therapy, as well as glucosamine for the patellofemoral symptoms. (Tr. at 807.) On June 23, Dr. Berry Dr. Berry assessed tight IT band with lateral knee pain status post total knee arthroplasty. (Tr. at 813.) He recommended she continue therapy; she declined a cortisone injection. They would see her back in a couple months or as needed. (Tr. at 814.)

On August 31, 2015, plaintiff advised PA Vahovick of a moderate increase in chronic low back pain since the past evaluation. She used Norco twice daily and cyclobenzaprine three times daily, as well as Lidoderm patches over the painful area. Her pain was improved with these medications but not adequately controlled. (Tr. at 663.) On exam, she arose to a standing position with modest difficulty and ambulated with a grossly normal gait. She was tender to palpation over the left SI joint. She did not display any aberrant pain behavior. They continued her medications and scheduled an injection (Tr. at 664), which Dr. Mortenson administered on September 11, 2015 (Tr. at 662).

On October 23, 2015, chiropractor Koch provided a letter report, indicating he had treated plaintiff since September 2012, focusing on her upper back pain, with plaintiff reporting a slow, gradual reduction of pain and improved cervical range of motion. Dr. Koch indicated a review of his notes showed a pattern of several months of therapy followed by several months of remission, again followed by an exacerbation of cervical and upper thoracic spinal pain. Based on this history, he expected plaintiff to experience periodic exacerbations leaving her incapacitated for up to one to two weeks several times a year. (Tr. at 675.) He suggested limiting her lifting to less than ten pounds maximum, with standing/walking limited to one to two hours per day, minimal repetitive bending, and sitting limited to four hours or less per day. (Tr. at 676.)

On October 27, 2015, plaintiff was seen for a physical therapy evaluation on referral from PA Vahovick, reporting pain radiating into the left scapular region, upper trap, and into the left side of the neck, also causing tension headaches. (Tr. at 781.) She reported functional limitations in reaching overhead, looking over her shoulders, and driving. She got assistance from her husband and children with carrying laundry, groceries, and heavier housework. (Tr. at 782.) Her neck disability index score of 29 placed her in the "severe disability" range. She was instructed on a home exercise program. She was to continue with chiropractic treatments and use of a TENS unit. (Tr. at 783.)

On November 6, 2015, Dr. Mortenson prepared a letter report indicating:

We have treated Ms. Kim Lang since 2009. She has the diagnosis of Post Laminectomy Syndrome. We have tried Physical Therapy, SI Joint Injections, medications and Lumbar Median Branch Blocks. We have not improved her in any meaningful way and her prognosis is poor. She is not capable of doing competitive work at this time.

B. Procedural History

1. Application and Initial Decisions

Plaintiff applied for benefits in November 2010, alleging a disability onset date of August 5, 2008. (Tr. at 144.) She indicated that she could not work due to back problems, headaches, and trouble turning her head. (Tr. at 172.) The agency denied the application initially in January 2011 (Tr. at 76, 87) based on the assessment of consultant Philip Cohen, M.D., that plaintiff could perform sedentary work (Tr. at 83), and on reconsideration in July 2011 (Tr. at 86, 92) based on the assessment of Ronald Shaw, M.D., who agreed with Dr. Cohen (Tr. at 312). Plaintiff requested a hearing (Tr. at 96), and on October 18, 2012, she appeared with counsel before an ALJ (Tr. at 37-75). The ALJ also summoned a vocational expert ("VE").

2. First Hearing

a. Plaintiff

Plaintiff testified that she was married and lived with her husband and four children, ages 16, 13, and 11 year old twins. (Tr. at 561.) Forty-three years old, she stood 5'3" tall and weighed 195 pounds. (Tr. at 561-62.) She testified that she graduated from high school in 1987 and from 1994 to 1999 worked for an accounting company doing data entry, payroll, and filing. From 1999 to 2005, she worked for a trucking company doing data entry, filing, and answering phones. In 2005, she became a certified nursing assistant ("CNA"), working in that capacity from November 2005 until August 5, 2008, when she hurt her back. She briefly returned to work, part-time, as a CNA in the summer of 2009, before re-injuring her back. She had not worked in any other capacity since August 2008. (Tr. at 562-64.)

Plaintiff testified that in 1997 she underwent lumbar fusion surgery. (Tr. at 564.) She re-injured her back in August 2008, and in 2009 she underwent surgery to remove the hardware from the prior surgery. She testified that she primarily treated with Dr. Mortenson for her back. She reported pain in her lower back going into her buttocks, off to the left side. (Tr. at 565.) She experienced pain every day, most of the day. (Tr. at 565-66.) She indicated that the heaviest thing she could lift was a gallon of milk. (Tr. at 566.) She also reported difficulty bending, such as when she tried to wash dishes. (Tr. at 566-67.) She said that her back also bothered her with prolonged sitting. She was unable to estimate how long she could sit (Tr. at 567) but later testified that she could sit at the computer for half an hour to an hour (Tr. at 568). She also reported back problems with prolonged standing; she had to take breaks when washing dishes. (Tr. at 568.) She initially indicated that she could not estimate how long she could stand, stating "I don't have a time limit, just do as much as I can." (Tr. at 568.) On follow up questioning she did indicate that she could not do a half hour worth of dishes without stopping. She testified that she could walk about a block. (Tr. at 569.) In order to relieve pain, plaintiff would take pain pills (two Percocet per day) and muscle relaxers (cyclobenzaprine three times per day); lay down; and use a heating pad and a TENS unit. (Tr. at 569-70.)

Plaintiff testified that on a typical day, she got up at about 6:00, woke up her husband for work, and helped her children get ready for school. (Tr. at 572.) After they left, she would drink coffee, watch the news, and then go lay down for about an hour after taking her pills. She would get up, take a shower, and try to do some chores around the house, such as folding laundry, doing the breakfast dishes, and wiping cupboards. (Tr. at 573.) She did not carry laundry up and down the stairs. After lunch, she would sit down with a heating pad or an ice pack until her kids came home from school. (Tr. at 574.) She would cook dinner with her kids. The kids would clean the table and do the dishes. After dinner, they would do homework, watch TV, and read. (Tr. at 575.) Her kids did the mopping and her husband helped with grocery shopping. (Tr. at 576.)

Plaintiff testified that she could not do her past desk jobs because of all the sitting and filing. (Tr. at 577.) Plaintiff testified that she used to coach her children's sports teams but now just watched their games. She also used to bowl and skate with her kids but no longer did. (Tr. at 582.) She reported that she had settled her workers' compensation case for about $100,000. (Tr. at 577.)

b. VE

The VE classified plaintiff's past work as a CNA as medium generally, very heavy as plaintiff did it; as a general clerk as sedentary generally, light as plaintiff did it; and as a payroll clerk as sedentary generally, light as plaintiff did it. (Tr. at 587.) The VE testified that the clerk positions would allow flexibility to shift positions throughout the workday. (Tr. at 588.)

The ALJ then asked a hypothetical question, assuming a person of plaintiff's age, education, and work experience, limited to light and/or sedentary work, with no repetitive or prolonged bending. The VE testified that this person could do plaintiff's past clerk jobs. (Tr. at 589.) The ALJ then asked about a person restricted to sedentary work, sitting no more than 45 minutes at a time, no standing more than 60 minutes at a time, no walking more than one block at a time, and no bending or stooping. The VE testified that such a person could perform the general clerk and payroll clerk jobs as generally performed. Adding additional restrictions of standing and walking less than two hours, sitting less than two hours, absent from work more than four times per month, and needing unscheduled breaks, would preclude work. (Tr. at 590.) The VE explained that the sitting, standing, and walking in this hypothetical did not equal a full workday, and four absences from work per month would not be tolerated. (Tr. at 590-91.)

3. ALJ's First Decision

On January 25, 2013, the ALJ issued a partially favorable decision (Tr. at 16), finding plaintiff disabled from August 5, 2008, through December 28, 2009. The ALJ found that medical improvement occurred as of December 29, 2009, and that plaintiff had been able to work from that date through the date of the decision. (Tr. at 20.)

In reaching this conclusion, the ALJ indicated that he credited the opinions of Drs. Hollowell and Karr, and rejected the opinion of Dr. Mortenson. The ALJ found Dr. Hollowell's December 2009 recommendation that plaintiff reintegrate to work supported by the record, including evidence that the surgery was successful, diagnostic tests showing mild to moderate degenerative changes, and plaintiff's activities, including caring for four children, cooking, performing household chores, shopping, and attending her children's school events. (Tr. at 26.) The ALJ gave Dr. Mortenson's report endorsing significant limitations little weight. (Tr. at 26-27.) First, Dr. Mortenson's suggestion that plaintiff's 1997 fusion left her permanently disabled was contrary to the evidence that she fully recovered from that surgery and returned to the work force, including in the physically demanding job of CNA. Second, Dr. Mortenson's finding that plaintiff reached a healing plateau in April 2010 was contrary the findings of the treating surgeon (Dr. Hollowell) and the independent medical examiner (Dr. Karr), who, the ALJ said, both found that she had reached maximum medical improvement and was ready to return to work in December 2009. Third, the ALJ concluded that Dr. Mortenson's extreme restrictions were not supported by plaintiff's daily activities. The ALJ found Dr. Karr's opinion that plaintiff could return to full-time work in December 2009 with limitations in lifting and bending supported by the evidence. (Tr. at 27.)

The ALJ determined that, from August 2008 to December 2009, plaintiff could not sustain full-time work. However, the ALJ stated that Drs. Hollowell and Karr both found that by the end of December 2009 plaintiff had reached maximum medical improvement and was ready to return to work. (Tr. at 29.)

The ALJ found plaintiff's statements concerning the intensity, persistence, and limiting effects of her symptoms after December 2009 "not entirely credible." (Tr. at 30.) Both Dr. Hollowell and Dr. Karr opined that she was ready to return to work by the end of December 2009, and Dr. Karr concluded that behavioral factors influenced plaintiff's pain, with her subjective pain appearing out of proportion to the objective physical findings. The ALJ found plaintiff's testimony at the hearing not very credible, as she was vague and evasive about her limitations. With follow up questions, she admitted various activities. (Tr. at 30.) The ALJ also concluded that plaintiff was not interested in returning to work based on her opposition to Dr. Hollowell's recommendation that she do so. Finally, he found that she had the "secondary gain" factor of enhancing her workers' compensation claim. (Tr. at 31.)

The ALJ concluded that, beginning December 29, 2009, plaintiff was capable of performing her past work as a payroll clerk and general office clerk. He accordingly found that her disability ended at that time. (Tr. at 32.)

4. First Round of Review

Plaintiff requested review (Tr. at 14-15), but on March 18, 2014, the Appeals Council denied her request (Tr. at 1). Plaintiff commenced an action for judicial review, and on December 31, 2014, the court approved the parties' stipulation that the matter be remanded for further proceedings. On remand, the ALJ was to hold a de novo hearing and further evaluate the medical opinion evidence, plaintiff's subjective complaints, and whether medical improvement occurred. (Tr. at 544.)

On August 14, 2015, the Appeals Council issued its remand order, affirming the ALJ's finding that plaintiff was disabled from August 5, 2008 to December 28, 2009, vacating the ALJ's decision with respect to the issue of disability after December 28, 2009, and remanding for resolution of several issues. First, the Council noted that the previous decision did not contain an adequate evaluation of Dr. Hollowell's opinion. Contrary to the ALJ's finding, Dr. Hollowell did not find plaintiff ready for full-time work in December 2009; rather, he recommended that she return to work on a very gradual basis, continuing the same restrictions he had given in July 2009, during a time the ALJ found plaintiff disabled. (Tr. at 551.) Second, the Council found that the prior decision did not contain an adequate evaluation of Dr. Mortenson's August 2010 opinion. The ALJ discounted that opinion as contrary to the finding of Drs. Karr and Hollowell that plaintiff had reached maximum medical improvement and was ready to return to work in December 2009. However, only Dr. Karr's report appeared to support medical improvement, not Dr. Hollowell's. The Council further noted that Dr. Mortenson treated plaintiff before and after Dr. Karr's report, and neither he nor Dr. Hollowell reported that plaintiff exaggerated her symptoms. Third, the Council found that the prior decision did not contain an adequate discussion of plaintiff's credibility. The Council noted that the ALJ relied on plaintiff's ability to perform household chores; however, plaintiff performed some chores sitting. The ALJ also found a secondary gain factor of enhancing her workers' compensation claim, but Dr. Karr was the only source to suggest pain behaviors influenced plaintiff's reported symptoms. The Council directed the ALJ to further evaluate plaintiff's subjective complaints, further evaluate whether medical improvement occurred on December 29, 2009, and give further consideration to plaintiff's maximum RFC, evaluating the medical opinions in so doing. (Tr. at 552.) The ALJ was to offer plaintiff the opportunity for a hearing, take any action needed to complete the record, and issue a new decision for the period after December 28, 2009. (Tr. at 553.)

5. Hearing on Remand

On November 9, 2015, plaintiff appeared with counsel for her hearing before the ALJ on remand. The ALJ again summoned a vocational expert ("VE"). (Tr. at 465.)

a. Plaintiff

Plaintiff reiterated her background information. Married with four children, then ages 19, 16, and twins age 14, she previously worked as a CNA, which she described as heavy work, and as an office clerk and payroll clerk for an accounting firm, which she described as sedentary work. (Tr. at 469-72.)

Plaintiff testified that Dr. Hollowell performed her back surgery, then transferred care to Dr. Mortenson, a pain management doctor. (Tr. at 472-73.) She saw Dr. Mortenson every six months or so, depending on her pain level. Dr. Mortenson tried pain medications, injections, therapy, and a TENS unit. (Tr. at 473.) They had discussed use of a spinal cord stimulator, but this procedure scared her. She also received chiropractic treatment with Dr. Koch. No further surgery was planned. (Tr. at 474.)

Plaintiff testified that her back was getting worse in terms of her pain level and ability to move and do things. (Tr. at 474.) Her pain level averaged between 6 and 7 on a 0-10 scale, sometimes higher depending on her activities. (Tr. at 475.) She also related her left knee injury and subsequent surgeries. (Tr. at 475-76.) Since the replacement surgery, her knee was doing a lot better but still hurt and swelled at times, such as when she did a lot of activities. (Tr. at 47879.) She used a knee brace, ice to reduce the swelling every other day, and elevated it in her recliner every day due to stiffness and pain. (Tr. at 479-80.) Plaintiff also reported treatment for her neck since 2012, with an MRI showing some degeneration. (Tr. at 480-81.) Plaintiff testified that when her lower back started hurting, the pain flared up her left side, went into her shoulder blade, and then she got bad headaches. (Tr. at 481.) Plaintiff reported using hydrocodone, a lidoderm pain patch, and the muscle relaxer cyclobenzaprine. The medications caused side effects of fatigue and nausea. (Tr. at 483-84.)

Plaintiff testified that she had help with household chores such as laundry, cooking, and cleaning; her kids and husband usually did the laundry. She could handle personal care herself. She avoided bending and lifting over 10 pounds. She testified that she could sit for ½ hour, stand about hour, and walk less than a block. (Tr. at 485-86.) She took breaks when trying to do things around the house. (Tr. at 487.) She testified that she could not work full-time but could work part-time, a couple hours per day. (Tr. at 488.) She did not think she could return to her past office jobs because of the sitting, filing, and bending required. (Tr. at 491.)

b. VE

The VE classified plaintiff's past work as a payroll clerk as sedentary, data entry clerk as sedentary, and CNA as medium generally, heavy as plaintiff did it. (Tr. at 494.) The ALJ then asked a hypothetical question assuming a person of plaintiff's age, education, and work experience limited to sedentary work that would allow a change of positions such that the person would not have to sit or stand longer than ½ hour at a time. The VE testified there would be jobs for such a person. He explained that for sedentary jobs it is "fairly common that employers would provide an accommodation . . . where the desk will move. . . . So, I think the way the jobs are done in the economy would provide for being able to sit and stand." (Tr. at 495.) The VE further indicated that these types of sedentary jobs did not require crawling, kneeling, squatting, or stooping, and that in clerical type jobs there was room for changing positions. (Tr. at 496.) The VE testified that once the person exhausted leave, absences of more than one per month would not be tolerated. (Tr. at 497.) Nor would employers tolerate a person taking extra breaks or being off task more than 10% of the time. (Tr. at 498.)

6. ALJ's Decision on Remand

On February 5, 2016, the ALJ issued his second decision. (Tr. at 440.) He stated that, ordinarily, when the Appeals Council remands a case it vacates the entire prior decision; here, however, the Council affirmed the previous closed period of August 5, 2008 to December 28, 2009, remanding only that portion of the previous decision pertaining to plaintiff's status thereafter. The ALJ accordingly incorporated that portion of his previous decision addressing plaintiff's status and the medical evidence from 2008 to the end of 2009. (Tr. at 449.)

Summarized, the previous medical evidence showed that plaintiff injured her back in a motor vehicle accident, undergoing lumbar fusion surgery in 1997. On August 5, 2008, plaintiff re-injured her back at work. After conservative treatment failed, plaintiff underwent surgery to remove the instrumentation. By July 2009, Dr. Hollowell placed plaintiff on part-time, light duty, with the plan of gradually increasing her hours. (Tr. at 449.) However, in August 2009, plaintiff aggravated her back at work. (Tr. at 449-50.) An MRI from September 10, 2009, showed no new injury, and in December 2009 Dr. Hollowell recommended part-time work to reintegrate plaintiff back into the workforce, although plaintiff was "quite opposed" to that suggestion. (Tr. at 450.) Plaintiff never did return to work. (Tr. at 450.)

On December 28, 2009, Dr. Karr performed an independent medical examination, expressing skepticism about the extent of plaintiff's pain complaints. Dr. Karr offered a permanent partial disability rating of only 5%. (Tr. at 450.)

The ALJ acknowledged the Council's intimation that he had given too much weight to Dr. Karr's report and too little to the opinions of Drs. Hollowell and Mortenson. The ALJ further acknowledged that he may have misconstrued Dr. Hollowell's opinion to some extent, as Dr. Hollowell during the December 16, 2009 visit still limited plaintiff to part-time, sedentary work. (Tr. at 450.) However, the ALJ concluded that Dr. Hollowell's post-operative exam results were not particularly impressive; the doctor seemed to be of the position that plaintiff should have ultimately been able to resume work (Tr. at 450); and, while he may not have expressed the skepticism of Dr. Karr, Dr. Hollowell did seem to be "of the belief that some functional component was present." (Tr. at 451, citing Tr. at 250, note from 12/16/09.)

The ALJ found that Dr. Mortenson's "objectivity seemed particularly clouded." (Tr. at 451.) The ALJ noted the minimal findings from an October 1, 2009 exam and December 2009 diagnosis of simply, "neuritis." (Tr. at 451.) In a subsequent letter dated November 16, 2010, Dr. Mortenson found plaintiff capable of sitting or standing less than two hours in an eight-hour workday, walking no more than one block, and lifting up to 10 pounds occasionally, yet he gave a permanent partial disability estimate of just 3%, which the ALJ found inconsistent with those extreme limitations. (Tr. at 451.)

In addition to the medical evidence, the ALJ surmised that a desire to help plaintiff get Medicare and disability payments may have influenced Dr. Mortenson and Hollowell's opinions. The ALJ further noted that federal courts had taken ALJs to task for failing to build a "logical bridge" between the evidence and their findings, yet if treating source opinions are to be given weight there should be a reciprocal duty on their part to offer support for extreme limitations. (Tr. at 451.)

The ALJ continued that, after she obtained a large workers' compensation settlement, plaintiff's medical visits became less frequent. "While one cannot necessarily take from that that the degree of [plaintiff's] pain complaints had diminished, it certainly does raise questions as to the presence of incapacitating levels of discomfort or the degree of limitation which [plaintiff] claims." (Tr. at 451.) The ALJ noted that plaintiff saw Dr. Mortenson and others at the pain clinic six times in 2010-11, primarily to get pain medications. The ALJ also found plaintiff's pain complaints during those visits inconsistent with lumbar radiculopathy. Rather, those complaints were more reflective of what Dr. Mortenson characterized as "myofascial in nature." (Tr. at 451.)

The ALJ noted that much of the focus in 2011 involved left knee complaints. Plaintiff saw Dr. Berry in June 2011, with an MRI showing some thinning of the cartilage, and in July 2011 Dr. Berry performed an arthroscopy, after which she did extremely well, later taking a trip to Florida. She did aggravate her knee after tripping in November 2011, but in January 2012 she was reassured that there was nothing serious and could increase her activities as tolerated. The records contained little reference to the knee until November 27, 2012, when plaintiff re-injured her knee while out dancing with friends, an activity which, the ALJ found, suggested that up to that point the knee was fairly stable. (Tr. at 452.)

The ALJ noted that most of the treatment records for the latter part of 2012 came from Koch Chiropractic and focused on plaintiff's back and neck. The ALJ indicated that Dr. Koch's notes, other than his initial exam on September 5, 2012, consisted largely of plaintiff's subjective complaints. (Tr. at 452.) Dr. Koch provided a letter report in which he related a pattern of several months of therapy followed by several months of remission, again followed by exacerbation of cervical and thoracic pain, which pattern he expected to continue. (Tr. at 452.) The ALJ gave this letter "little weight" (Tr. at 453), after noting that under social security regulations chiropractors are not "acceptable medical sources" (Tr. at 452).

The ALJ noted that, besides visits to the chiropractor, plaintiff was seen several times at the pain clinic between 2013 and 2015 by a physician's assistant for her back and neck complaints, primarily for the dispensing of pain medications. Dr. Mortenson also administered a left sacroiliac joint injection on September 11, 2015. In most of those visits, the physician's assistant cited no aberrant pain behavior; exam was notable largely for tenderness. (Tr. at 453.)

The ALJ found the more significant issue raised in the later medical records was plaintiff's left knee following the November 2012 injury. This injury resulted in a series of surgeries, culminating in a total left knee replacement performed on September 16, 2013. Despite a setback following a fall in December 2013, plaintiff improved following the surgery. In a February 25, 2014 note, Dr. Berry indicated she was doing well, despite inconsistent therapy attendance. Nothing further was noted until October 14, 2014, when plaintiff broke her toe after hitting it on a door. (Tr. at 453.) In May 2015, plaintiff returned to Dr. Berry, complaining of knee pain, but x-rays looked "perfect," physical exam noted some tenderness, and plaintiff had full range of motion. (Tr. at 453.) The doctor recommended physical therapy. (Tr. a 453-53.) During a June 2015 follow-up, Dr. Berry noted that the pain complaints were likely secondary to a "tight IT band." (Tr. at 454.)

Just before the hearing on remand, Dr. Mortenson provided a letter dated November 6, 2015, summarizing plaintiff's medical history and his treatment of her since 2009, and indicating that in his opinion she was not capable of working. At the hearing, plaintiff claimed her back pain was getting worse. She acknowledged that her left knee had gotten better since the knee replacement surgery in 2013, but she still experienced some pain and swelling with activity and used a knee brace. She estimated that she could sit for no more than ½ hour before experiencing stiffness, stand for 30-60 minutes, and walk less than a block. She stated that she more recently began experiencing neck pain and headaches. (Tr. at 454.)

The ALJ reviewed his previous decision that there had been sufficient medical improvement by December 28, 2009, such that plaintiff could perform a range of work. Despite the criticism that he gave too much weight to Dr. Karr's opinion to the exclusion of Drs. Hollowell and Mortenson, the ALJ found Dr. Karr's examination far more detailed and supported by the evidence, marked by a more realistic and critical eye than Dr. Mortenson. Subsequent medical records and references to a number of activities also showed plaintiff to have been far more functional. (Tr. at 454.) The ALJ found that Dr. Hollowell tried to be sympathetic toward plaintiff while at the same time considering the surgery to remove the prior fusion hardware to have been successful. Dr. Hollowell's "subsequent comment that [plaintiff's] complaints seemed out of proportion to the examination results was as close as he came to questioning [her] actual degree of . . . pain." (Tr. at 455.)

The ALJ found Dr. Karr's restrictions consistent with plaintiff's daily activities. At the initial hearing, the ALJ found plaintiff vague and evasive when testifying about her limitations, but on follow up questioning she admitted activities like childcare, meal preparation, folding laundry, paying bills, computer work, cleaning, washing dishes, attending sporting events, and helping children with homework. Regarding the finding of medical improvement by the end of 2009, the ALJ found that Dr. Karr's report remained persuasive. Dr. Hollowell last saw plaintiff in December 2009, with the most notable aspect of that entry being plaintiff's opposition to returning to work. (Tr. at 455.)

The ALJ found plaintiff's subsequent complaints of pain radiating up the spine "questionable to say the least," and that whatever degenerative disc disease may be present in the cervical spine did not "appear to be all that severe." (Tr. at 455.) He continued:

As for Dr. Mortenson, for reasons already enunciated, including the wide discrepancy between the permanent partial disability estimates he offered versus the ridiculous accompanying residual functional capacity figures given, not to mention the discrepancy between those estimates and some of [plaintiff's] actual activities, causes the undersigned to question his objectivity. That and the fact that at this point [plaintiff's] visits to him appear[] to be nothing more than for the periodic dispensing of oxycodone suggest he may be nothing more than enabler.

Regarding her knee complaints, plaintiff first saw Dr. Berry in June 2011 and after undergoing arthroscopy in July 2011 and physical therapy thereafter, by September 2011 she was "doing good." (Tr. at 456.) After her injury in November 2012 while out dancing with friends, the knee became particularly problematic requiring a series of procedures, ultimately resulting in a total knee replacement in September 2013. By February 2014, her knee appeared to be doing well (and she did not report back pain during this time). She sought no further treatment until October 2014 when she broke her toe. On return to Dr. Berry in May 2015, she displayed full range of motion and x-rays of the knee looked perfect. She continued to deny back pain in June 2015. The ALJ found that these denials of back pain supported the finding that, while the spinal impairment remained severe, it was not all that incapacitating. (Tr. at 456.)

In what he called a generous concession, the ALJ awarded a closed period from November 27, 2012, to February 24, 2014, based on plaintiff's left knee impairment. During that time, she was precluded from much in the way of weight-bearing on a consistent basis. Since February 24, 2014, however, the ALJ again found sufficient medical improvement to the point where plaintiff regained the capacity for some degree of light work and a full range of sedentary work, including past jobs as a general clerk and payroll clerk. (Tr. at 457.)

The ALJ concluded that, beginning December 29, 2009, plaintiff did not have an impairment that met or equaled a Listing. Medical improvement occurred as of that date, related to her ability to work, such that between December 29, 2009 and November 27, 2012, plaintiff had the RFC to perform light work with no squatting or kneeling and no more than occasional bending, allowing performance of past relevant work. (Tr. at 457.) From November 27, 2012, through February 25, 2014, plaintiff had the severe impairments of status post lumbar fusion and removal of the hardware, along with some degenerative disc disease; and moderate to severe chondromalacia leading to total left knee replacement, which precluded her from engaging in even sedentary work. She developed no new impairments after this time, and beginning February 26, 2014, she did not suffer from an impairment that met or equaled a Listing. Finally, the ALJ concluded that plaintiff experienced medical improvement as of that date, such that she had the RFC for a full range of sedentary work and some degree to light work so long as there was no prolonged standing nor any squatting, crawling, or kneeling. As of that date, she had the ability to perform past work. Disability accordingly ended, the ALJ concluded, on February 26, 2014. (Tr. at 458.)

This action followed.

III. DISCUSSION

I briefly address a preliminary issue before turning to the parties' primary contentions. In supplemental memoranda, the parties dispute whether the medical improvement standard applies in closed period cases such as this. It does. See Tumminaro, 671 F.3d at 633; Jones v. Shalala, 10 F.3d 522, 524 (7^th Cir. 1993); see also Attmore v. Colvin, 827 F.3d 872, 876 (9^th Cir. 2016) (collecting cases). In any event, under both the five-step test and the eight-step test, the ultimate issues are basically the same - whether plaintiff can, given any medical improvement and her resulting RFC, perform her past or other work. As the Commissioner notes, plaintiff appears to concede that her knee impairment improved by February 2014; accordingly, her back impairment would provide the primary basis for disability between December 2009 and November 2012 and continuing past February 2014.

A. Plaintiff's Arguments

Plaintiff argues that the ALJ's decision, and in particular his assessment of the medical source opinions and the credibility of her statements regarding her symptoms, is based on speculation and conjecture, rather than evidence. I first address the medical opinions, then credibility.

1. Medical Opinions

The ALJ's assessment of Dr. Mortenson's opinions is particularly harsh. The ALJ called Dr. Mortenson's RFC figures "ridiculous" and suggested that Dr. Mortenson was nothing more than "enabler," dispensing oxycodone. The record contains no support for an accusation that Dr. Mortenson behaved unprofessionally; nor does it contain evidence that plaintiff abused prescription medications or otherwise engaged in drug-seeking behavior. Further, the record shows that Dr. Mortenson treated plaintiff for an extended time, from January 2009 through November 2015, offering a variety of treatments, including different injections and therapy referrals, in addition to prescribing medications. He also considered implanting a neuro-stimulator. Despite these various treatments, Dr. Mortenson concluded in 2015 that plaintiff's pain persisted at a level precluding full-time work.

In Carradine v. Barnhart, 360 F.3d 751, 755 (7^th Cir. 2004), the court reversed under similar circumstances, where the claimant's doctors provided an extensive course of pain management treatment in the absence of the kind of objective findings the ALJ thought should be present. The court explained:

Since severe pain is consistent with "the absence of significant findings upon diagnostic testing and physical examination," which would not reveal a psychological origin of pain, the doctors had no choice but to take Carradine's complaints of pain "at face value" and treat her. What is significant is the improbability that Carradine would have undergone the pain treatment procedures that she did, which included not only heavy doses of strong drugs such as Vicodin, Toradol, Demerol, and even morphine, but also the surgical implantation in her spine of a catheter and a spinal-cord stimulator, merely in order to strengthen the credibility of her complaints of pain and so increase her chances of obtaining disability benefits; likewise the improbability that she is a good enough actress to fool a host of doctors and emergency-room personnel into thinking she suffers extreme pain; and the (perhaps lesser) improbability that this host of medical workers would prescribe drugs and other treatment for her if they thought she were faking her symptoms. Such an inference would amount to an accusation that the medical workers who treated Carradine were behaving unprofessionally.

The Commissioner attempts to provide an alternate basis for rejecting Dr. Mortenson's November 2015 letter - that it offered only an opinion on the ultimate issue without a discussion of specific functional limitations or supporting medical findings. The ALJ did not reject the letter on this basis, and judicial review is limited to the reasons he provided. See, e.g., Pierce v. Colvin, 739 F.3d 1046, 1050 (7^th Cir. 2013) (citing SEC v. Chenery Corp., 318 U.S. 80, 87-88 (1943)). "Further, although a medical opinion on an ultimate issue such as whether the claimant is disabled is not entitled to controlling weight, the ALJ must consider the opinion and should recontact the doctor for clarification if necessary." Barnett v. Barnhart, 381 F.3d 664, 669 (7^th Cir. 2004). The ALJ cannot simply reject a doctor's report on this basis. See Garcia v. Colvin, 741 F.3d 758, 760 (7^th Cir. 2013).

The ALJ's assessment of Dr. Hollowell's opinions is also problematic. In its remand order, the Appeals Council highlighted the ALJ's erroneous statement in the first decision that Dr. Hollowell found maximum medical improvement as of December 2009. Despite acknowledging that he may have misconstrued Dr. Hollowell's opinion to some extent in the previous decision, the ALJ did not grapple with the fact that Dr. Hollowell imposed these same restrictions in July 2009, a time when the ALJ found plaintiff disabled. While the ALJ indicated that Dr. Hollowell "seemed to be of the position that [plaintiff] should have been able to ultimately resume work" (Tr. at 450), the record contains no subsequent report from Dr. Hollowell returning plaintiff to full-time work.

The ALJ also stated that Drs. Hollowell and Mortenson may have been sympathetic to plaintiff, influenced by a desire to help her get Medicare and disability benefits. While the Seventh Circuit has acknowledged the possibility that treating doctors may "bend over backwards to assist a patient in obtaining benefits," Hofslien v. Barnhart, 439 F.3d 375, 377 (7^th Cir. 2006), the ALJ cited no record evidence that Drs. Hollowell and Mortenson were so motivated. Indeed, Dr. Hollowell encouraged plaintiff to try to return to work. The ALJ stressed plaintiff's reluctance to return to work as Dr. Hollowell recommended in December 2009, but given the fact that she quickly re-injured her back when she made such an attempt in the summer of 2009 her reticence was understandable.

The ALJ discounted the opinion of plaintiff's chiropractor, Dr. Koch, on the ground that he was not an "acceptable medical source" under the regulations. While only "acceptable medical sources" (e.g., physicians or psychologists) can establish the existence of medically determinable impairments and provide medical opinions that may be entitled to controlling weight, opinions from "other medical sources," such as nurse practitioners, physician assistants, and chiropractors, "are important and should be evaluated on key issues such as impairment severity and functional effects, along with the other relevant evidence in the file." SSR 06-3p, 2006 SSR LEXIS 5, at *8. Opinions from other medical sources may not be given "little weight" just because they are not acceptable medical sources. See Voigt v. Colvin, 781 F.3d 871, 878 (7^th Cir. 2015); see also Thomas v. Colvin, 826 F.3d 953, 961 (7^th Cir. 2016).

Citing a Sixth Circuit case, Walters v. Comm'r of Soc. Sec., 127 F.3d 525, 530-31 (6^th Cir. 1997), the Commissioner argues that the ALJ was not required to adopt the opinions of chiropractor Koch or give them controlling weight. As indicated, however, Dr. Koch's status as an "other medical source" does not permit rejection of his opinion on that basis alone. The Commissioner suggests that the ALJ may have rejected Dr. Koch's opinion because his treatment notes consisted largely of plaintiff's subjective complaints with little in the way of detailed examination results. However, the ALJ stated: "Even if his findings had been in greater detail, the weight they would be accorded would be limited given the fact that under the Regulations chiropractors are not considered 'acceptable medical sources.'" (Tr. at 452.) After summarizing Dr. Koch's October 23, 2015 letter, the ALJ stated, without further elaboration, that he gave "that letter and the opinion contained therein little weight." (Tr. at 453.)

Because the ALJ did not provide "good reasons" for discounting the opinions of Drs. Hollowell and Mortenson, and his rejection of Dr. Koch's opinion conflicts with SSR 06-3p, the matter must be remanded.

2. Credibility

Plaintiff contends that the ALJ barely discusses credibility in the second decision, other than to reiterate his assessment from the first decision (i.e., that plaintiff's testimony at the first hearing was vague and evasive). The Commissioner responds that plaintiff's credibility argument, focused on the ALJ's first decision, is limited to the ALJ's finding of medical improvement in 2009; the Commissioner contends that plaintiff has thus waived any argument regarding the ALJ's credibility finding as of February 2014. Plaintiff has not waived the argument; she simply noted that, in the second decision, the ALJ adopted findings from the first.

The second decision does not contain a definite credibility finding. Rather, scattered throughout the decision are various comments that appear to bear on credibility. First, the ALJ noted that after plaintiff received her workers' compensation settlement, her treatment frequency decreased. (Tr. at 451.) The record does show that plaintiff's visits with Dr. Mortenson and his assistant decreased somewhat after 2012. However, the ALJ made no finding as to how often a patient experiencing the plaintiff's stated level of pain would be expected to see her pain management doctor. See Dominguese v. Massanari, 172 F. Supp. 2d 1087, 1096 (E.D. Wis. 2001) (reversing where, in the absence of medical evidence, ALJ made his own independent medical determination about the appropriateness of doctor visits). By that point, plaintiff had undergone surgery with Dr. Hollowell, received injections from Dr. Mortenson, participated in physical therapy, and declined the implantation of a spinal cord stimulator; aside from medication management, it is unclear what else Dr. Mortenson had to offer. Dr. Mortenson did suggest chiropractic treatment (Tr. at 323), and the record shows that plaintiff saw Dr. Koch dozens of times from 2012 to 2015. (Tr. at 429-39, 675-87.)

Second, the ALJ suggested that, because plaintiff was out dancing when she hurt her knee in November 2012, her condition at that point could not have been all that bad. (Tr. at 452, 456.) However, the ALJ did not question plaintiff about her dancing, e.g., how often she did it, what sort of dancing she was doing on this occasion, or how long she had been dancing when her knee gave out. If plaintiff's knee popped the moment she stepped on the dance floor it is hard to see how this defeats her claim.

Third, the ALJ noted that, at the initial hearing, plaintiff tended to be vague and evasive when testifying about her limitations, but on follow up questioning she admitted activities like childcare, meal preparation, folding laundry, paying bills, computer work, cleaning, washing dishes, attending sporting events, and helping children with homework. (Tr. at 455.) However, the ALJ did not link any of those activities with plaintiff's specific claims regarding her limitations or otherwise explain how they undermined her credibility. See Brown v. Colvin, No. 13-C-262, 2013 U.S. Dist. LEXIS 144740, at *22 (E.D. Wis. Oct. 7, 2013) ("[T]he ALJ must explain why particular activities undercut the claimant's credibility; it is not enough to simply list various chores, declare them 'significant,' and then find the claimant incredible."). Nor were the activities the ALJ listed particularly significant in terms of their demands. See, e.g., Zurawski v. Halter, 245 F.3d 881, 887 (7^th Cir. 2001) ("While the ALJ did list Zurawski's daily activities, those activities are fairly restricted (e.g., washing dishes, helping his children prepare for school, doing laundry, and preparing dinner) and not of a sort that necessarily undermines or contradicts a claim of disabling pain."). Moreover, plaintiff testified that she took breaks and got significant help from her kids and husband with household chores (Tr. at 485-87), which the ALJ did not consider. See Clifford v. Apfel, 227 F.3d 863, 872 (7^th Cir. 2000) (rejecting credibility finding based on limited daily activities done with help and rest). Finally, the ALJ did not appreciate the critical differences between such activities of daily living and the demands of a full-time job. See Bjornson v. Astrue, 671 F.3d 640, 647 (7^th Cir. 2012).

Fourth, the ALJ found plaintiff's claim of pain migrating up the spine "questionable to say the least," and that whatever cervical degeneration may be present "would not appear to be all that severe." (Tr. at 455.) However, he cited no medical opinion evidence in support of this statement. See Israel, 840 F.3d at 439 ("There is always a danger when lawyers and judges attempt to interpret medical reports and that peril is laid bare here."); Browning v. Colvin, 766 F.3d 702, 705 (7^th Cir. 2014) (noting that administrative law judges are not permitted to "play doctor").

Finally, the ALJ noted that plaintiff denied back pain when she saw Dr. Berry in 2014 and 2015. (Tr. at 456.) This was a correct observation, but it cannot, standing alone, support the ALJ's decision. The matter must be remanded for reconsideration of plaintiff's statements regarding her symptoms and their functional effects.

B. The Commissioner's Arguments

Focusing on the December 2009 medical improvement determination, the Commissioner argues that the ALJ reasonably adopted the opinion of Dr. Karr that plaintiff had reached maximum medical improvement and could return to full-time work over the contrary opinions of Drs. Hollowell and Mortenson. The Commissioner points to the ALJ's statement that the treating doctors' objective findings were lackluster, and that even Dr. Hollowell found plaintiff's complaints out of proportion to the examination results. (Tr. at 455; see also Tr. at 451, citing Tr. at 250.) However, I can find no statement by Dr. Hollowell to that effect. The most Dr. Hollowell said on December 16, 2009 was that plaintiff was "exquisitely tender to palpation." (Tr. at 250.) It appears that only Dr. Karr suggested symptom exaggeration, as the Appeals Council noted. Further, to the extent Dr. Karr relied on positive Waddell signs, that is not necessarily evidence of secondary gain or malingering. See Minor, 513 Fed. Appx. at 422 n.15.

The Commissioner notes that the ALJ found Dr. Karr's restrictions more consistent with plaintiff's activities, such as caring for her children, going to Florida, and dancing. (Tr. at 454-55.) As indicated above, however, it is hard to see how such activities are, without further explanation or exploration, inconsistent with disabling pain.

The Commissioner contends that many of these same reasons (Dr. Karr's suspicion of symptom exaggeration, plaintiff's daily activities, and plaintiff's opposition to returning to work as Dr. Hollowell suggested) support the ALJ's credibility finding. The court considers the ALJ's decision as a whole, see, e.g., Rice v. Barnhart, 384 F.3d 363, 370 n.5 (7^th Cir. 2004), but the scattered comments in this decision do not provide specific reasons supported by substantial evidence. See Moss v. Astrue, 555 F.3d 556, 561 (7^th Cir. 2009).

IV. CONCLUSION

Plaintiff requests that she be found disabled; in the alternative, she requests remand for further proceedings before a different ALJ. A judicial award is appropriate only if all factual issues have been resolved and the record clearly supports a finding of disability. Allord v. Astrue, 631 F.3d 411, 417 (7^th Cir. 2011). Plaintiff develops no argument that this standard is satisfied. The record contains medical support for the ALJ's conclusion, and as discussed above, the matter must be remanded so the ALJ can reconsider treating provider opinions and plaintiff's credibility.

THEREFORE, IT IS ORDERED that the ALJ's decision is reversed, and the matter is remanded for further proceedings consistent with this decision. The Clerk is directed to enter judgment accordingly. I recommend - but do not require - that the Commissioner assign the case to a different ALJ on remand. See Sarchet v. Chater, 78 F.3d 305, 309 (7^th Cir. 1996).

Dated at Milwaukee, Wisconsin this 29^th day of April, 2017.

/s Lynn Adelman

LYNN ADELMAN

District Judge