EGAN, J.

Dale C. Johnson, Springfield, argued the cause and filed the briefs for petitioner.

Daniel Edward Walker argued the cause for respondents. On the brief was Kim Shubin.

Before Tookey, Presiding Judge, and Egan, Judge, and Aoyagi, Judge.

EGAN, J. Claimant, the personal representative of the deceased worker's estate, seeks review of an order of the Workers’ Compensation Board upholding SAIF's denial of his claim for left L5-S1 radiculopathy. He contends that the board's order is not supported by substantial evidence or substantial reason, because the board failed, without reason, to defer to the opinion of the worker's treating physician and did not adequately explain its conclusion that the treating physician's opinion had been based on an incomplete medical history and inaccurate information. In reviewing the board's order for substantial evidence and substantial reason, ORS 656.298(7) ; 183.482(8)(c), we agree with claimant that the board erred and reverse and remand.

This case presents a true battle of the experts as to whether Robert Culley, the deceased worker, had L5-S1 radiculopathy, and whether a work incident was a material cause of the alleged radiculopathy. Beginning in 2013, Culley suffered from preexisting bilateral sciatica and low back pain, with pain radiating down both legs to the knees. Culley saw Dr. Essex, his primary care physician, in July 2013 for symptoms of pain in both legs. Essex diagnosed bilateral sciatica. An MRI in 2013 showed "multilevel disc disease producing relatively minimal canal, but up to moderate neuroforaminal narrowing as described above." There were disc bulges at L3-4 and L4-5 which minimally indented the ventral aspect of the thecal sac. There was nothing remarkable with respect to L5-S1.

Beginning in April 2014, Essex referred Culley to Dr. Zilkoski for treatment of constant pain in his left foot, which began after treatment for a left knee Baker's cyst. At that time, Culley also was experiencing low back pain, and he sought and obtained several months of chiropractic treatment for his foot and for low back pain, including mild to moderate pain in the sacrum region. A chiropractic exam revealed positive left leg straightening for low back pain as well as decreased sensation in the dermatomes of L5 and S2. The chiropractor's primary diagnosis was "lumber spine sprain/strain." Culley's foot symptoms gradually resolved. In July 2015, Culley, who worked for employer Oregon Department of Justice as a detective, was injured at work when, as he walked in employer's parking lot, he was hit by a bicycle. The bicycle hit the back of Culley's left side, and he fell to the pavement on his right leg and scraped his knees and elbows. Culley reported pain in his left ankle but initially he did not report any foot or back pain. SAIF, employer's workers’ compensation insurer, accepted a claim for left knee strain, right knee abrasion, left knee abrasion, left elbow abrasion, and left foot comminuted fracture of the 3rd proximal phalanx.

In early September 2015, Culley saw Dr. Yao for his knees and reported that for about two weeks he had been experiencing sharp pain and numbness in both feet, primarily on the left. Yao noted that Culley had previously seen Zilkoski "for this," a reference to Zilkoski's treatment of claimant for foot pain in 2014. Yao referred Culley to Dr. McCormick, an orthopedist, for his foot pain, who referred Culley for a nerve conduction study, which showed that Culley's left lateral plantar motor nerves "showed no response." McCormick stated in the chart note of December 19, 2015, that diagnostic imaging confirmed a diagnosis of neuralgia and neuritis, likely due to Culley's work injury. In February 2016, Culley began to report pain and numbness in his left great toe.

In June 2016, Culley saw his primary care doctor, Essex, for left foot pain. Essex noted that he had previously treated Culley for very similar symptoms in 2013 not related to work.

In October 2016, on Yao's referral, Culley began seeing Dr. Herring, a neurologist, for his continued left foot pain and numbness. Culley reported to Herring that, about a year before, he had experienced shooting electrical pains into the left big toe, but that the symptoms had resolved on their own and were not present at the time of the work injury.

Herring examined Culley and noted tenderness in the left lateral lumbosacral region with positive straight leg raising on the left. Herring identified symptoms of radiculopathy, including decreased pinprick at the bottom of Culley's left foot and the left lateral foot, with decreased pinprick throughout, including the lower leg. Herring also noted weakness in Culley's left foot and leg. Herring concluded that, although Culley had not experienced low back pain at the time of the work injury, Culley's symptoms were the result of a radiculopathy originating at L5-S1, with the work injury as the most likely cause. Herring recommended further imaging.

In November 2016, Herring's associate Dr. Balm, a neurophysiologist, performed an electrodiagnostic study of Culley's left foot. Balm reported findings of "electrophysiologically mild, old, or chronic inactive left S1 radiculopathy." He concluded that the study provided no electrophysiologic evidence for the presence of any ongoing active radiculopathy, nor for the presence of lumbosacral plexopathy, sciatica or other mononeuropathy affecting the left lower extremity. Culley also had an MRI of the lumbar spine in November 2016. That imaging showed mild multilevel spondylosis and mild L2-3 spinal stenosis with moderate bilateral lateral recess narrowing and no foraminal narrowing. All levels had facet degenerative changes and some degree of central disc bulging with no compression of the nerve roots. During and after that MRI, claimant began to experience pain in his low back on the left and pain radiating into his leg.

Herring reported that, although the 2016 MRI did not show any definitive etiology for Culley's symptoms, he was still of the opinion that Culley's symptoms and findings were suggestive of radiculopathy /nerve root irritation. Because Culley's symptoms persisted and had begun to include back pain, Herring recommended more imaging and the opinion of a spine surgeon. An x-ray confirmed mild degenerative changes of the lumbar spine.

Herring continued to believe that the onset of Culley's increased radiculopathy symptoms was related to his work injury.

Dr. Rosenbaum examined Culley on SAIF's behalf in October 2017 and disagreed. Culley reported to Rosenbaum that he was experiencing left foot numbness and low back pain, which had developed after his work injury. Rosenbaum's exam revealed "no true spasm" and moderate pain to palpation at L5-S1 and L4-5, moderate left trochanteric pain, bilateral sacroiliac pain and moderate left sciatic notch pain. Rosenbaum believed that the disc bulge findings on the November 2016 MRI and x-ray were consistent with the degenerative process with no acute findings. He said that Culley's various MRIs did not reveal a pathology that would indicate nerve root compression, displacement, or impingement at any level. He stated further that neither his chart notes nor those of any other examiner indicated lumbar radiculopathy in a specific dermatomal pattern as evidenced by motor, sensation, or reflex loss. He did not see clinical signs of radiculopathy and was of the opinion that Culley's lumbar condition was preexisting.

Herring rejected Rosenbaum's opinion as having been based on an incomplete examination, or an inaccurate recording of that examination, and maintained that Culley's symptoms were radiculopathy "coincident with a work injury where he was struck by a heavy bicyclist with a mechanism of injury that would certainly be consistent with a subsequent lumbar spine injury."

In October 2017, SAIF denied Culley's request to accept left L5-S1 lumbar radiculopathy as a new/omitted medical condition, and Culley filed a request for hearing.

Culley underwent an MRI in November 2017, which had findings similar to the November 2016 MRI. Findings from an MRI in November 2018 were also unchanged.

Dr. Button, an orthopedic surgeon, examined Culley at SAIF's request. He explained that radiculopathy means irritation of a cervical or lumbar nerve that produces pain, weakness, and/or numbness radiating down an extremity. Button found no complaints of low back pain or evidence of any pain, weakness, or numbness radiating down the extremity (except for the foot numbness) until Culley saw Herring 14 months after the injury. Button stated that, because there are many causes of numbness in an extremity and the cause is often unknown, to support a diagnosis of radiculopathy, one needed to have "either electrical support from a nerve test, a history or exam consistent with pain, weakness, or numbness radiating down a leg, or it should generally be in conjunction with nerve compression seen in the lumbar spine." Button opined that, because MRI testing had not revealed nerve root compression, and Culley had none of the symptoms typically present with radiculopathy until 14 months after the injury, it was "medically highly unlikely" that Culley developed radiculopathy from the work incident. Button diagnosed lumbar spondylosis, preexisting, without nerve root compression to explain any motor or sensory changes in the lower extremities. Button opined that Culley's symptoms were related to preexisting degenerative disease.

Herring disagreed with Button. Herring noted that while it was true there was no MRI evidence of nerve root compression, compression was not the only source of radiculopathy. He explained that, in the absence of obvious compression, sometimes a tear in the annulus fibrosis will release disc fluid with an inflammatory component that will provoke an auto-immune reaction resulting in radiculopathy of the nerve root, and that such tears do not necessarily show up on MRIs. In Culley's case, he believed that the bicycle incident was of sufficient force to "probably cause a small rent in his annulus which leaked irritants and caused chronic inflammatory changes around the S1 nerve root." Herring opined that the work injury was the major contributing cause of Culley's radiculopathy condition. He based his opinion on the mechanism of injury (a "forceful event"), Culley's consistent symptoms and credible examinations, the transient relief of symptoms that Culley experienced after receiving anti-inflammatory steroid injections, and Balm's abnormal EMG findings that showed objective evidence of S1 radiculopathy. Although Culley had had prior left toe issues, Herring noted that the problem had resolved and that Culley was asymptomatic at the time of the work injury. In light of Culley's symptoms of left foot weakness, numbness and pain, a positive straight leg test on the left, and the EMG showing S1 radiculopathy, Herring concluded that Culley had a pattern of nerve root symptoms and findings consistent with nerve root irritation, and that, to a reasonable probability, Culley's complaints were consistent with a lumbar radiculopathy related to the work injury. An administrative law judge (ALJ) upheld SAIF's denial of the claim. The ALJ directed his analysis to the proof of causation, reasoning that claimant had failed to establish that work was a material contributing cause of the claimed radiculopathy. The ALJ explained that Herring had not addressed claimant's prior low back and sciatic symptoms. The board, in adopting the ALJ's order with supplementation, reasoned that the deference commonly given to a treating physician's diagnosis was not applicable in this case, in view of the fact that Herring did not begin treating Culley until 14 months after the injury. The board discounted Herring's opinion for several additional reasons.

For example, the board reasoned that the record did not show that Herring was aware that Culley had suffered and been treated for similar sciatic symptoms in 2013, including left foot pain; thus, the board concluded that Herring's opinion was based on an incomplete medical history. The board further reasoned that Herring had mistakenly relied on a chart note referencing treatment of a complaint about foot numbness two weeks after the injury when, in fact, the treatment had occurred before the injury; thus, the board concluded that Herring's opinion was inaccurate. Finally, the board discounted Herring's opinion as unpersuasive, reasoning that Herring had not sufficiently responded to Button's contrary opinion and had not adequately explained "his conclusion by describing how plaintiff's signs and symptoms on examination close to the time of the injury" "fit within the dermatomal pattern for the left L5-S1 radiculopathy condition." Having discounted Herring's opinion, the board concluded that Culley had not met his burden of proof to establish that the symptoms of radiculopathy are work-related.

On judicial review, claimant contends that the board erred in rejecting Herring's opinion. As Culley's treating physician, claimant contends, Herring's opinion was entitled to deference in his evaluation of Culley's current symptoms, whether or not he began treating Culley immediately after the injury. Cf. Dillon v. Whirlpool Corp. , 172 Or. App. 484, 489, 19 P.3d 951 (2001) ("The Board properly may or may not give greater weight to the opinion of the treating physician, depending on the record in each case.") Claimant further challenges the board's conclusion that the record did not show that Herring had a complete medical record, pointing out that Herring had for his review Culley's full medical history, including records from his primary care physician, as well as all of the subsequent medical records. Finally, claimant challenges the board's conclusion that Herring's causation opinion was based on inaccurate information, asserting that the board's conclusion is a misreading of the medical record. Claimant asserts that, as a specialist in the field of neurology who had treated Culley for his symptoms for two years, Herring was in the best position to evaluate the cause of Culley's symptoms. See Weiland v. SAIF , 64 Or. App. 810, 814, 669 P.2d 1163 (1983) ("When the medical evidence is divided, we have tended to give greater weight to the conclusions of a claimant's treating physician, absent persuasive reasons not to do so.").

We have reviewed the record and agree with claimant that the board misread it with regard to Herring's reference to Yao's report of symptoms two weeks after the injury. The record requires the finding that Herring's opinion was based on accurate information and that Culley did complain of foot pain to Yao two weeks after the injury.

We also agree with claimant that the board erred in determining that Herring had an incomplete record because he did not refer in his reports to Culley's history of and treatment for sciatic in 2013 and may not have been aware of Culley's prior history of sciatica. The record requires a finding that Herring had for his review all of Culley's medical records and was aware of his history.

Finally, we agree with claimant's contention that the board erred in discounting Herring's opinion because he did not adequately explain his disagreement with Button's view that Culley had not experienced symptoms of radiculopathy. Herring's reports do explain his reasoning that, despite the absence of back symptoms immediately following the injury, Culley's history and diagnostics, which included a positive leg straightening test, decreased pinprick in the left foot and leg, indicative of sensory loss, and an EMG showing S1 radiculopathy, were indicative of L5-S1 radiculopathy. SAIF argues that whatever the board's rationale in rejecting Herring's opinion relating to the existence of L5-S1 radiculopathy, the record supports the board's determination as to a lack of causation. SAIF is correct that the medical record includes evidence that, before the 2015 injury, in 2014, Culley was suffering from low back symptoms on the left as well as symptoms that Herring identified as radiculopathy, including decreased sensation along the L5 dermatome. The record also includes medical evidence from Rosenbaum and Button that Culley's symptom complex after the work injury did not constitute radiculopathy. For those contradictory reasons, SAIF argues, this court should conclude that the board's order upholding SAIF's denial of the claim is supported by substantial evidence.

As to the issue of deference to the opinion of the treating physician, SAIF points out correctly that it is not a rule of law; rather, it is a method of factual analysis that the board is free to apply in its judgment. In Dillon , we explained that, in view of the fact that the court no longer reviews the board's orders de novo , but for substantial evidence, the question of deference to the treating physician is for the board, as a factual analytical construct. We must affirm the board's deference determination if it is supported by substantial evidence. Id. at 488, 19 P.3d 951.

In Dillon , 172 Or. App. at 489, 19 P.3d 951, we went on to explain the nature of substantial evidence review, quoting from this court's watershed opinion in Armstrong v. Asten-Hill Co. , 90 Or. App. 200, 752 P.2d 312 (1988). As relevant here, the takeaway from Armstrong is that, to be supported by substantial evidence, the board's order must indicate what findings the board makes and how those findings led the board to its ultimate conclusion—that is, it must be supported by substantial reason. Armstrong , 90 Or. App. at 206, 752 P.2d 312 ("The requirement of findings leads to a requirement that the agency state its reasoning."); see Guild v. SAIF , 291 Or. App. 793, 800, 422 P.3d 376 (2018) ("The board can reject an expert's medical opinion as unpersuasive, but it must explain its reasons for doing so."); see also Minor v. SAIF , 290 Or. App. 537, 545, 415 P.3d 1107 (2018) ("In reviewing for substantial evidence, we must also determine whether the board's analysis comports with substantial reason. To satisfy that requirement, the board must ‘provide a rational explanation of how its factual findings lead to the legal conclusions on which the order is based.’ ") (Quoting Arms v. SAIF , 268 Or. App. 761, 767, 343 P.3d 659 (2015) (citing Drew v. PSRB , 322 Or. 491, 500, 909 P.2d 1211 (1996) )).

As we pointed out in Guild , this court does not "reweigh the evidence or ‘substitute our judgment for that of the board as to any issue of fact supported by substantial evidence.’ " 291 Or. App. at 796, 422 P.3d 376 (quoting Elsea v. Liberty Mutual Ins ., 277 Or. App. 475, 483, 371 P.3d 1279 (2016) ); ORS 183.482(7). Nevertheless, the factfinder must meticulously review the entire record to correctly decide a case. If the board makes a finding and conclusion based on one doctor's opinion, then the finding and conclusion must be based on an analysis of the entirety of the information provided by that doctor. Guild , 291 Or. App. at 798-800, 422 P.3d 376. If it is not, then the order lacks substantial evidence and substantial reason. See Garcia v. Boise Cascade Corp. , 309 Or. 292, 296, 787 P.2d 884 (1990) ("An assertion of a finding of fact as part of an explanation for disregarding evidence is subject to attack if that fact relied upon is not, itself, supported by substantial evidence."). That requirement makes a difference in this case: The board's findings that are not based on substantial evidence led it to misapply the factual analytical model concerning deference to Herring, the treating physician.

In Garcia , the court said:

"In cases where evidence is rejected by the [board], and such action purports to be based on facts, it is appropriate for the reviewing court to examine whether the [board's] decision to disregard or discount the evidence in the record is supported by substantial evidence. Put another way: An assertion of a finding of fact as part of an explanation for disregarding evidence is subject to attack if that fact relied upon is not, itself, supported by substantial evidence."

309 Or. at 296, 787 P.2d 884. Under our standard of review, it is not appropriate for us to correct the board's findings, but it is incumbent upon us to point out errors in the board's analysis that could have affected the outcome of the case. Guild , 291 Or. App. at 796, 422 P.3d 376. As we have determined, the evidence in the record does not support the board's several rationales for discounting Herring's opinion. Thus, we conclude that the board's findings, including its rejection of the treating physician's opinion, are not supported by substantial evidence or substantial reason. In light of that conclusion, we reverse and remand the board's order for reconsideration under the correct standard relating to consideration of the treating physician's opinion.

Reversed and remanded.