ORDER AND JUDGMENT
Paul J. Kelly, Jr., Circuit Judge
Edward D. Kellams appeals a magis
trate judge’s
order affirming the Commissioner’s denial of disability and supplemental security income benefits. Mr. Kellams claims an administrative law judge (ALJ) discredited his testimony without substantial evidence and improperly evaluated his medical providers’ opinions. Exercising jurisdiction under 28 U.S.C. § 1291 and 42 U.S.C. § 405(g), we reverse and remand for further proceedings.
I
On September 22, 2009, Mr. Kellams had a seizure while driving. He rolled his vehicle several times and sustained two burst fractures of his vertebrae. The next day, he underwent spinal fusion surgery to stabilize the fractured vertebrae, but imaging in November 2009 showed “multiple chronic compression deformities in the thoracic spine” and “displacement of the anterior vertebral body fragment” in the lumbar spine. Aplt. App. at 386-87. Nearly a year later, in October 2010, Mr. Kellams filed for benefits, claiming he was disabled on the date of his accident by epilepsy and his back condition.
An ALJ denied benefits, but the Appeals Council remanded for additional proceedings. On remand, another ALJ held a second hearing at which Mr. Kellams described his conditions and their limiting effects. After considering his testimony and other evidence, the ALJ found that Mr. Kellams was severely impaired by two burst fractures of his T12 and L5 vertebrae, degenerative changes of the lumbar spine status post L4-S1 fusion, and a seizure disorder. Despite these impairments, however, the ALJ concluded at step five of the five-step evaluation process,
see Wall v. Astrue,
561 F.3d 1048, 1052 (10th Cir. 2009) (explaining the five-step process), that Mr. Kellams was not disabled because he retained the residual functional capacity (RFC) to perform a limited range of light work. Specifically, the ALJ determined Mr. Kellams had the RFC for light work but must avoid hazards and repetitive lifting; cannot drive;- could occasionally engage in postural activities; and could not be exposed to heights, ladders, or scaffolds. In reaching this conclusion, the ALJ discounted Mr. Kellams’ credibility and gave little weight to his medical providers’ opinions. The magistrate judge affirmed, and Mr. Kellams appealed.
II
“We review the Commissioner’s decision to determine whether the correct legal standards were applied and whether the Commissioner’s factual findings are supported by substantial evidence in the record.”
Langley v. Barnhart,
373 F.3d 1116, 1118 (10th Cir. 2004). “Substantial evidence is such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.”
Id.
(internal quotation marks omitted). “It is more than a scintilla, but less than a preponderance.”
Newbold v. Colvin,
718 F.3d 1257, 1262 (10th Cir. 2013) (internal quotation marks omitted). “In addition to a lack of substantial evidence, the [Commissioner’s] failure to apply the correct legal standards, or to show us that she has done so, are also grounds for reversal.”
Winfrey v. Chater,
92 F.3d 1017, 1019 (10th Cir. 1996). Under our standard of review, “we neither reweigh the evidence nor substitute our
judgment for that of the agency.”
Newbold,
718 F.3d at 1262.
Mr. Kellams raises two issues on appeal. He first contends the ALJ’s partially adverse credibility finding is unsupported by substantial evidence. He says the evidence cited by the ALJ actually bolsters his credibility and there is other evidence of pain and dysfunction that was ignored by the ALJ. We have held that an ALJ cannot mischaracterize or downplay evidence to support his findings.
See Talbot v. Heckler,
814 F.2d 1456, 1463-64 (10th Cir. 1987). And although an ALJ need not “discuss every piece of evidence,” he “must discuss the uncontroverted evidence he chooses not to rely upon, as well as significantly probative evidence he rejects.”
Clifton v. Chater,
79 F.3d 1007, 1009-10 (10th Cir. 1996). As more fully discussed below, the ALJ mischaracterized or downplayed the severity of some evidence to support his adverse credibility finding, without considering other probative evidence that tends to support Mr. Kellams’ allegations of pain and limited functioning. This was error.
Mr. Kellams’ second point of contention is that the ALJ incorrectly evaluated his medical providers’ opinions. He argues that the ALJ discounted these opinions without citing substantial evidence and without properly considering the relevant factors set forth at 20 C.F.R. §§ 404.1527(c), 416.927(c). It is well established that an ALJ must consider the opinion of every medical source and provide specific, legitimate reasons for rejecting it.
Doyal v. Barnhart,
331 F.3d 758, 764 (10th Cir. 2003). In weighing a medical source opinion, the ALJ must evaluate “all of the [relevant] factors set out in [20 C.F.R. §§ 404.1527(c), 416.927(c)],” mindful that an examining source opinion “is presumptively entitled to more weight than a doctor’s opinion derived from a review of the medical record.”
Chapo v. Astrue,
682 F.3d 1285, 1291 (10th Cir. 2012). Here, the ALJ failed to provide specific, legitimate reasons for discounting the medical source opinions and presumptively dismissed the examining source opinions in favor of a reviewing physician’s opinion. This, too, was error. We elaborate on both of these issues raised by Mr. Kellams below.
A
Credibility
Mr. Kellams first challenges the ALJ’s credibility finding. The ALJ found that Mr. Kellams’ “medically determinable impairments could reasonably be expected to cause the alleged symptoms; however, [his] statements concerning the intensity, persistence and limiting effects of these symptoms are not entirely credible for the reasons explained in this decision.” Aplt. App. at 20. Mr. Kellams acknowledges that the ALJ cited some evidence in support of this finding, but he says that evidence does not actually support discrediting his complaints of pain and the limiting effects of his conditions. We agree.
“Credibility determinations are peculiarly the province of the finder of fact, and we will not upset such determinations when supported by substantial evidence. However, findings as to credibility should be closely and affirmatively linked to substantial evidence and not just a conclusion in the guise of findings.”
Newbold,
718 F.3d at 1267 (internal quotation marks omitted).
When a claimant alleges disabling pain,
[w]e must consider (1) whether Claimant established a pain-producing impairment by objective medical evidence; (2) if so, whether there is a “loose nexus” between the proven impairment and the Claimant’s subjective allegations of pain; and (3) if so, whether considering all the evidence, both objective and subjective, Claimant’s pain is in fact disabling.
Kepler v. Chater,
68 F.3d 387, 390 (10th Cir. 1995) (internal quotation marks omitted). In evaluating a claimant’s credibility, the ALJ should consider:
the levels of medication and their effectiveness, the extensiveness of the attempts (medical or nonmedical) to obtain relief, the frequency of medical contacts, the nature of daily activities, subjective measures of credibility that are peculiarly within the judgment of the ALJ, the motivation of and relationship between the claimant and other witnesses, and the consistency or compatibility of non-medical testimony with objective medical evidence.
Id.
at 391 (internal quotation marks omitted).
Mr. Kellams testified that before his accident, he had not had a seizure in nearly twelve years. But on the date of his accident, he was supposed to be taking a slow-metabolizing anti-seizure medication, which he could not afford, so instead he was taking a cheaper medication that metabolized faster. He stated that since his accident, he was having three to four seizures a year and most recently had a grand mal seizure in February 2013. On that occasion, he did not go to the emergency room, but he woke up in bed and learned his sons had found him on the floor. He said he was very tired and fatigued after having the seizure and slept the rest of the day and night. Before that episode, he said he experienced a seizure in 2010, when he fell “straight back ... onto the concrete sidewalk.” Aplt. App. at 87-88.
Mr. Kellams further testified that his seizure disorder was only “somewhat controlled.”
Id.
at 85. He stated that he had been using medical marijuana to help control his seizures until he “got caught with too many plants” and was put “on probation.”
Id.
at 95. He indicated that he smoked once per day, right before bed, which helped control his seizures the next day. Moreover, Mr. Kellams testified that after he stopped using medical marijuana in November 2012, he had the aforementioned seizure in February 2013 and had not been seizure-free for more than three to four months without using medical marijuana and his other medication.
Regarding his back injury, Mr. Kellams testified that his back pain limited his .ability to work more than his seizure disorder. He acknowledged that x-rays taken when he had a seizure and fell on the sidewalk did not indicate displacement of the surgical hardware, but he testified that it seemed his pain was getting progressively worse. He said he had been on pain medication since 2011. In February of that year, he attempted to pick up a laundry basket, which caused acute back pain that required treatment and medication at the emergency room. Thereafter, in July 2011, he asked his doctors to increase his pain medication. He testified that he had been taking 20 mg of Percocet, which was increased to 30 mg. He also stated that his doctor had discussed chronic pain syndrome with him but did not recommend that he try any other treatment beside pain medication.
Mr. Kellams also described his limited daily activities. He said he lives with his adult sons, who help around the house. He testified that he could no longer do laundry, work on the car, or perform yard work. Although he has no dishwasher, he said he could wash some dishes by hand, which took about ten minutes. Even sitting, however, he said he could not wash dishes for more than fifteen minutes because of his back pain. Additionally, he denied having the ability to reach for things high or low when grocery shopping, and he said he could pick up a gallon of milk but not two gallons. Mr. Kellams
testified that he could push a shopping cart without difficulty, but when asked if he could push a cart full of groceries, he said he did not have much money to spend at the grocery store, so he would have to say no.
As for his ability to work, Mr. Kellams testified that after his accident, he was unable to do his past work as an HYAC technician. He said he gave jobs to one of his sons and sometimes he would ride along to supervise. This arrangement in which he accompanied his son approximately ten hours per week lasted less than a year.
The ALJ summarized this testimony and provided several explanations why it was not entirely credible. Mr. Kellams challenges the following explanations.
1, Objective Diagnostic Evidence
The ALJ first ruled that Mr, Kellams’ allegations were not supported by the objective diagnostic evidence. The ALJ reasoned that CT and MRI scans from September 2009 and an x-ray from December 2010 confirmed burst fractures at the T12 and L5 vertebrae, but there was no evidence of hardware failure. Although the ALJ recognized some abnormalities of the lumbar spine, he determined that these abnormalities did not equate to specific functional limitations. Moreover, he concluded that the CT and MRI scans showed “stable findings with no evidence of severe stenosis, nerve root impingement, or herniated discs.” Aplt. App. at 21.
The ALJ’s rationale appears to acknowledge both a pain-producing impairment and a loose nexus to Mr. Kellams’ allegations of pain. Nevertheless, the evidence cited by the ALJ does not support discounting his allegations. Although the ALJ correctly noted there was no evidence of hardware failure, the September 2009 CT scan revealed not only the two burst fractures but also a 40% loss of the vertebral body height at T12 with a 20% canal-diameter compromise. Further, it showed old fractures of T5 through T9, with a 70% loss of vertebral body height at T8. It also showed “a complex distracted [] fracture of the L5 vertebral body with loss of approximately 26% of the vertebral body height.”
Id.
at 373. The ALJ did not discuss these findings but instead downplayed them by broadly characterizing the record as revealing “some abnormalities of the lumbar spine.”
Id.
at 21.
Additionally, the ALJ cited the September 2009 MRI and the December 2010 x-ray, which again confirmed the two burst fractures, but he did not discuss an MRI from November 2009 indicating that Mr. Kellams had a history and clinical impression of pain.
See Clifton,
79 F.3d at 1009-10 (holding that an ALJ “must discuss the uncontroverted evidence he chooses not to rely upon, as well as significantly probative evidence he rejects”). This MRI reflected a 50% to 75% loss of vertebral body height at T5 and “[djiffuse osteopenia with multiple chronic compression deformities in the thoracic spine.” Aplt. App. at 386. Another contemporaneous MRI of the lumbar spine showed instability by virtue of “a severe compression fracture” at L5 with “[t]he anterior vertebral body fragment ... ante-riorly displaced by roughly 1.7 cm. This is further displaced than on the previous CT images from September of 2009.”
Id.
at 387. This evidence, which again the ALJ
did not discuss, refutes his conclusion that Mr. Reliaras had stable findings.
Also, the ALJ observed there was “no evidence of severe stenosis, nerve root impingement, or herniated discs.”
Id.
at 21. But “[t]he absence of evidence is not evidence,”
Thompson v. Sullivan,
987 F.2d 1482, 1491 (10th Cir. 1993), and there were other probative findings that Mr. Rellanas had significant loss of vertebral body-height at multiple levels and a 20% canal diameter compromise. At the very least, these uncontroverted findings undermine the ALJ’s determination that there was no evidence of stenosis or nerve root impingement, yet the ALJ failed to discuss them.
2. Exam Findings
The ALJ next ruled that Mr. Rellanas’ allegations were inconsistent with his exam findings. Although the ALJ gave little weight to the opinions of all examining sources (which we discuss belojv), he acknowledged that Mr. Rellanas’ emanas indicated he had tenderness to palpation of the lumbar spine, discomfort with hip range of motion, and hypertonicity of the musculature of the lumbar spine. Indeed, these findings bolster, rather than undermine, Mr. Rellanas’ complaints of pain and limited functionality. In any event, the ALJ reasoned that despite these findings, Mr. Rellanas had full strength in all muscle groups, normal sensation and range of motion, normal gait and ambulation, and negative straight-leg raise testing.
The absence of strength and sensory deficits does not necessarily undermine Mr. Rellanas’ subjective allegations of pain, however, and there were other findings that corroborate his testimony. For example, the ALJ discounted Mr. Rellanas’ testimony because he had normal range of motion, but Dr. Charlene Borja, who made that finding, restricted Mr. Rellanas from repetitive bending on account of his pain symptoms. Moreover, another provider determined that Mr. Rellanas had “multiple range of motion deficits of his Lumbar spine, Thoracic spine and Shoulders,” Aplt. App. at 428, but the ALJ completely overlooked these findings. Similarly, the ALJ determined Mr. Rellanas had a normal gait, but Dr. Borja observed he had an antalgic gait to the left.
Id.
at 411. The ALJ cited Dr. Borja’s report to confirm hypertonicity, but he failed to mention her finding of an antalgic gait. He also neglected to discuss evidence that Mr. Rellanas suffered from muscle spasms, which were repeatedly noted by his doctor.
See id.
at 470 (“[pjersistent muscle spasms”);
id.
at 472 (“thoracolumbar dysfunction with spasm”). An ALJ may not “pick, and choose among medical reports, using portions of evidence favorable to his position while ignoring other evidence.”
Hardman v. Barnhart,
362 F.3d 676, 681 (10th Cir. 2004).
3. Treatment History
The ALJ also discredited Mr. Rellanas by downplaying his treatment history, recalling that he underwent minimally invasive spine stabilization surgery. Minimally invasive or not, Mr. Rellanas had his spine fused over three vertebral levels at two different locations (Tll-Ll and L4-S1) using pedicle screws and rods.
See
Aplt. App. at 357-58. This was not substantial evidence for discounting his complaints of pain. Further, the ALJ noted that Mr. Rellanas regularly saw his primary care provider, was prescribed Percocet, anti-seizure medication, and medical marijuana, and took hot baths to relieve his pain. Such efforts to find relief generally bolster, rather than detract from, the credibility of subjective complaints of pain.
See Luna v. Bowen,
834 F.2d 161, 165-66 (10th Cir. 1987) (recognizing that “persistent attempts to find relief’ may help assess the
credibility of subjective claims of pain). Indeed, the ALJ conceded this level of treatment was consistent with some degree of symptomology. Yet the ALJ failed to consider Mr. Kellams’ February 2011 visit to the emergency room, where he reported back pain after attempting to pick up a laundry basket. This was another attempt to obtain relief for back pain experienced as a result of doing an ordinary daily activity, but the ALJ failed to consider it.
Instead, the ALJ discredited Mr. Kel-lams for not exploring other treatment options such as physical therapy, acupuncture, chiropractic manipulation, or massage. On this score, however, the ALJ failed to consider “(1) whether the treatment at issue would restore claimant’s ability to work; (2) whether the treatment was prescribed; (3) whether the treatment was refused; and if so, (4) whether the refusal was without justifiable excuse.”
Thompson,
987 F.2d at 1490. Except for a physical therapist’s functional capacity evaluation, there is no evidence that any of the ALJ’s alternative treatments were prescribed to Mr. Kellams. In fact, Mr. Kel-lams testified that his doctor discussed with him chronic pain syndrome but did not discuss any other treatment options, except pain medication.
See
Aplt. App. at 90. Moreover, Mr. Kellams points out that he was uninsured for most of the relevant timeframe and could not afford these therapies. Indeed, there is evidence that Mr. Kellams could not afford the proper anti-seizure medication,
see id.
at 84, 86, 350-51, 412, 426, but the ALJ did not consider that factor,
see Thompson,
987 F.2d at 1489-90 (holding that ALJ should consider whether claimant’s inability to afford treatment justified her failure to pursue it). Without considering these factors, the ALJ simply speculated on the efficacy and availability of these alternative, unpres-cribed treatments, which is not substantial evidence for discrediting Mr. Kellams’ testimony.
Neither was there substantial evidence to discredit Mr. Kellams based on the ALJ’s finding that he had not been prescribed any new, stronger medications. While technically Mr. Kellams was not prescribed a new, stronger medication (although he apparently switched from Per-cocet to a slightly different medication, Oxycodone), his
dosage
was increased repeatedly. Indeed, the ALJ acknowledged that Mr. Kellams’ dosage of Percocet was increased from 20 mg to 30 mg.
See
Aplt. App. at 21. And the record shows that his dosage began at 10 mg,
see id.
at 425, was increased to 15 mg,
see id.
at 472, was increased again to 20 mg,
see id.
at 477, and eventually graduated to 30 mg,
see id.
at 539. These progressively stronger doses of medication do not support discounting Mr. Kellams’ testimony.
A
Daily Activities
Finally, the ALJ discredited Mr. Kel-lams’ testimony based on his limited daily activities. The ALJ ruled that his activities could not be objectively verified and it was difficult to attribute his limitations to his medical conditions in light of the “relatively weak medical evidence.”
Id.
at 22. But the ALJ failed to identify the “relatively weak” evidence he was referring to, and
this explanation does not support an adverse credibility finding. Although we normally defer “to the ALJ as the trier of fact on credibility, deference is not an absolute rule.”
Thompson,
987 F.2d at 1490 (citations omitted). Accordingly, we remand this case to the agency for a proper credibility assessment.
B. Medical Source Opinions
Mr. Kellams also challenges the ALJ’s evaluation of the medical source opinions. The ALJ gave little weight to the opinions of three examining sources, but he gave great weight to the opinion of a state agency physician, Dr. Henry Fieger, who only reviewed Mr. Kellams’ medical record.
When evaluating the opinion of any medical source, an ALJ must consider:
(1) the length of the treatment relationship and the frequency of examination; (2) the nature and extent of the treatment relationship, including the treatment provided and the kind of examination or testing performed; (8) the degree to which the physician’s opinion is supported by relevant evidence; (4) consistency between the opinion and the record as a whole; (5) whether or not the physician is a specialist in the area upon which an opinion is rendered; and (6) other factors brought to the ALJ’s attention which tend to support or contradict the opinion.
Goatcher v. U.S. Dep’t of Health & Human Servs.,
52 F.3d 288, 290 (10th Cir. 1995);
see
20 C.F.R. §§ 404.1527(c), 416.927(c). If an ALJ rejects an opinion, he “must provide specific, legitimate reasons for rejecting it.”
Chapo v. Astrue,
682 F.3d 1285, 1291 (10th Cir. 2012) (internal quotation marks omitted).
1. Dr. Charlene Borja
The ALJ considered a consultative examination report prepared by Dr. Borja. The ALJ gave the report little weight, in part, because Dr. Borja examined Mr. Kel-lams only once. This rationale may justify refusing to give Dr. Borja the status of a treating physician and according her opinion controlling weight, but as an examining source her opinion still was entitled to particular consideration. Indeed, “an examining medical-source opinion is,
as such, ...
presumptively entitled to more weight than a. doctor’s opinion derived from a review of the medical record.”
Id.; see
20 C.F.R. §§ 404.1527(c)(1), 416.927(c)(1). The ALJ erred in discounting Dr. Borja’s opinion based on her single exam, yet according great weight to Dr. Fieger’s opinion, which was based on a review of the then-incomplete medical record.
The ALJ’s other reasons for rejecting Dr. Borja’s opinion were equally invalid. He determined the limitations she assessed for Mr. Kellams’ postural activities were “not particularly consistent with those findings that demonstrated no loss of range of motion of the spine and few other findings of significance.” Aplt. App. at 22. But the ALJ failed to specify “other findings of significance.”
See Hamlin v. Barnhart,
365 F.3d 1208, 1217 (10th Cir. 2004) (observing that ALJ should “specifically highlight those portions of the record [that were] allegedly inconsistent”). And although Dr. Borja did find that Mr. Kel-lams had normal range of motion in his cervical and dorsolumbar spine, she imposed postural limitations to accommodate his pain symptoms, reasoning that “[routine postural activities
are feasible;
however, repetitive bending, squatting, crouching
and stooping are not recommended until [Mr. Kellams’] low back symptoms are better managed and established.” Aplt. App. at 413 (emphasis added). This was no basis for discounting Dr. Borja’s opinion.
The ALJ also ruled that Dr. Borja’s opinion was not “consistent with the exam findings of others or the claimant’s treatment history.”
Id.
at 22. But the ALJ failed to cite any inconsistent exam findings, and although a physical therapist assessed multiple range-of-motion deficits almost a year after Dr. Borja’s exam, Dr. Borja noted Mr. Kellams’ pain symptoms and restricted him from repetitive bending. The ALJ failed to explain how these opinions were inconsistent. Moreover, although Dr. Borja did assess a greater ability to lift—up to 50 pounds—she similarly tempered that finding by stating it was “feasible” but “[rjepetitive carrying and lifting is not recommended until [his] low back symptoms are better managed.”
Id.
Apart from these findings, Dr. Borja discussed Mr. Kellams’ treatment history, including his accident, surgical history, and seizure medication. She also reviewed post-operative imaging of his spine, noting areas of compression and the .status of his hardware. Further, she identified “bilateral hypertonicity along the lumbar spine” and diagnosed “[l]ow back pain without radiculopathy.”
Id.
at 412. The ALJ did not explain how any of these findings are inconsistent with Mr. Kellams’ treatment history or the findings of other medical providers.
2. Michael Moore
The ALJ also gave little weight to the opinion of Michael Moore, a physical therapist who administered a functional capacity evaluation (FCE) on Mr. Kellams. Mr. Moore indicated that Mr. Kellams “gave a reliable effort [during the exam], with 27 of 27 consistency measures within expected limits.”
Id.
at 428. He found that Mr. Kellams did not tolerate stooping, crouching, or kneeling, noting Mr. Kellams’ reports of mid and low back pain. Additionally, Mr. Moore detected “multiple range of motion deficits” of the lumbar and thoracic spine,
id.,
although he determined that Mr. Kellams could frequently lift up to ten pounds and occasionally lift up to twenty pounds. He further determined that Mr. Kellams could frequently walk, stand, and climb stairs with occasional sitting and could occasionally reach overhead.
As a physical therapist, Mr. Moore is not an “[acceptable medical source.” 20 C.F.R. §§ 404.1502(a),-416.902(a). Nevertheless, ‘“[o]pinions from these medical sources ... are important and should be evaluated on key issues such as impairment severity and functional effects.’ ”
Bowman v. Astrue,
511 F.3d 1270, 1274-75 (10th Cir. 2008) (quoting Social Security Ruling 06-03p, 2006 WL 2329939, at *3 (Aug. 9, 2006)).
“[A]n opinion from a medical source who is not an ‘acceptable medical source’ may outweigh the opinion of an ‘acceptable medical' source’ ... if he or she has seen the individual more often than [an ‘acceptable medical source’] and has provided better supporting evidence and a better explanation for his or her opinion.”
Id.
at 1275 (quoting SSR 06-03p, 2006 WL 2329939, at *5).
Mr. Moore’s FCE spanned 18 pages, evaluating and detailing Mr. Kellams’ strength, range of motion, and functional abilities in a host of categories. It also evaluated the reliability of his efforts across 27 consistency measures. Despite these detailed findings and reliability controls, the ALJ discounted the FCE, in part, because Mr. Moore examined Mr. Kellams only once. That, however, was not a valid reason, standing alone, for dis
counting the FCE.
See Chapo,
682 F.3d at 1291.
The ALJ also discounted the FCE for lack of objective support. On this score, the ALJ conceded that an FCE is “a somewhat objective way of determining a person’s objective limitations.” Aplt. App. at 22. Moreover, he acknowledged that Mr. Kellams was “noted to have given good effort on this FCE exam.”
Id.
Still, the ALJ concluded there was “little objective support for the endorsed limitations on standing, sitting, lifting overhead, or engaging in postural activities.”
Id.
He reasoned: “The claimant’s limitations in these areas ostensibly stem from his back condition, but the objective diagnostic evidence has revealed fairly stable findings in the spine.”
Id.
We have already discussed evidence that Mr. Kellams’ spine was not stable,
see id.
at 387, and as the ALJ conceded, the FCE was objective evidence of Mr. Kellams’ limitations. Those limitations were corroborated by Dr. Borja’s consultative exam, which similarly advised against repetitive bending, squatting, crouching and stooping. The ALJ failed to justify giving the FCE little weight.
3. Pete Smith
Pete Smith, a physician’s assistant who followed Mr. Kellams, wrote three short letters describing Mr. Kellams’ limitations. On December 13, 2011, he wrote that Mr. Kellams had been a patient at his practice for ten years, had a history of back pain and seizures, and was unable to stand, sit, or stoop for long periods of time. Mr. Smith advised that Mr. Kellams’ activities were limited to twenty minutes and he was not to lift anything over five pounds. Mr. Smith indicated that Mr. Kellams was limited to sedentary work and was disabled from his past work. He wrote a similar letter on January 17, 2012. And, on July 18, 2013, Mr. Smith wrote that he had closely reviewed Mr. Moore’s FCE and agreed with it.
The ALJ acknowledged that Mr. Smith was a treating medical source, but he still gave the three letters little weight. The ALJ reasoned that Mr. Smith’s opinions were inconsistent with his own exam findings and treatment recommendations, but the record reflects that Mr. Smith consistently confirmed and treated Mr. Kellams’ back pain.
See, e.g., id.
at 425 (noting in July 2011 “paralumbar and parathoracic muscle tenderness” and prescribing Perco-cet);
id.
at 474 (noting in December 2011 report of “Chronic back pain/seizure disorder” and refilling Percoeet);
id.
at 472 (noting in February 2012 “positive thoraco-lumbar dysfunction with spasm,” request for medical marijuana “due to chronic back pain,” and prescribing Percoeet);
id.
(noting in March 2012 complaint of back pain, report that “Percoeet does okay, but he says he has been having to take five a day,” and prescribing Oxycodone);
id.
at 520 (noting in May 2012 diagnosis of chronic pain syndrome and prescribing Ox-ycodone);
id.
at 522-23 (noting same in July 2012);
id.
at 524 (noting same in August 2012);
id.
at 527 (noting same in September 2012);
id.
at 530 (noting same in October 2012);
id.
at 531-33 (noting same in January 2013);
id.
at 539 (noting same in May 2013). This evidence refutes the ALJ’s conclusion that Mr. Smith’s opinions were inconsistent with his examinations and treatment.
The ALJ also ruled that Mr. Smith’s opinions were inconsistent with the findings of other providers and the objective diagnostic evidence. Yet the objective diagnostic evidence complements, rather than undermines, Mr. Smith’s opinions. And although he assessed greater limitations than other providers, he also had the longest treatment relationship of any provider, with the most frequent interactions. Dura
tion, frequency, and consistency are all relevant factors under SSR 06-03p. The ALJ purported to consider these factors, but he merely referenced them in general, without discussing the specific evidence in the record. Our review indicates that the ALJ failed to properly evaluate the evidence in discounting Mr. Smith’s opinion.
Jp. Dr. Henry Fieger
Lastly, the ALJ gave Dr. Fieger’s opinion great weight. Dr. Fieger was a state agency physician who reviewed Mr. Rel-iaras’ medical record and concluded that he could occasionally lift up to twenty pounds, frequently lift up to ten pounds, and frequently engage in postural limitations (stooping, kneeling, crouching, and crawling). The ALJ gave this opinion great weight because “[i]t was rendered after a thorough review of the record, and is consistent with the record as a whole.”
Id.
at 23. None of the other providers allowed for frequent postural limitations, however, and Dr. Fieger’s review of the record was not itself a proper basis for according his opinion greater weight than those of the examining sources,
Chapo,
682 F.3d at 1291. Accordingly, we remand to the agency so the Commissioner may properly evaluate the medical source opinions.
Ill
The judgment of the district court is reversed, and this case is remanded to the district court with instructions to remand to the Commissioner for further proceedings consistent with this Order and Judgment.