[T1] The Medical Commission sustained the Wyoming Workers' Safety and Compensation Division's (Division) permanent partial impairment (PPT) rating of Michael D. Hurt's lumbar spine condition. The Medical Commission ruled Mr. Hurt failed to prove he was entitled to a PPI rating greater than the 9% whole body rating assigned by the Division, and the district court upheld the Medi*377cal Commission's decision. We conclude the Medical Commission's decision is supported by substantial evidence and, therefore, affirm.
ISSUE
[12] The issue for our review is:
1. Was the Medical Commission's decision that Mr. Hurt did not prove he was entitled to a higher impairment rating under the AMA Guides to the Evaluation of Physical Impairment supported by substantial evidence and otherwise in accordance with the law?
FACTS
[13] Mr. Hurt worked as a concrete truck driver for Tilton Ready Mix, Inc. On September 7, 2006, he was helping unload concrete from a truck when he slipped on a large hose and fell onto his back and buttocks. Steven Beer, M.D. performed surgery on Mr. Hurt's back on December 19, 2006, including an L4-5 and L5-S1 interbody fusion, an L8 to S1 posterior instrumentation using various types of hardware, and an LS8-S1 intertransverse fusion using autologous bone.
[14] Although Mr. Hurt generally had good results from the first surgery, he received permanent partial impairment (PPT) benefits as a result of the work-related injury. On September 15, 2008, Michael Kaplan, M.D. rated Mr. Hurt, using the AMA Guides to the Evaluation of Permanent Impairment (6th ed.2008), as having 7% whole body impairment. Mr. Hurt accepted the Division's award of PPI benefits based upon that rating.
[15] Over time, Mr. Hurt began to experience additional problems with his back, including pain, weakness and numbness. On August 31, 2010, Dr. Beer performed a see-ond surgery on Mr. Hurt's back, removing the hardware installed in the first surgery and extending the lumbar spine fusion to L2 and L8. Mr. Hurt's back was, therefore, fused from L2 to S1. The second surgery improved his condition, but he could not return to work.
[T6] On August 2, 2011, Dr. Kaplan performed a second PPI rating. He concluded Mr. Hurt's impairment placed him in Class 1 for Motion Segment Lesions and assigned an impairment rating of 9% of the whole body, a 2% increase over the previous rating and award. On September 26, 2011, Mr. Hurt was evaluated in Dr. Beer's office by physician assistant Andy Beguin. The physician assistant placed Mr. Hurt in Class 4 for Motion Segment Lesions and assigned him a 25% whole body PPI rating. Given the disparity between the two ratings, the Division asked a third physician, Anne MacGuire, M.D., to review them.1 She did not examine Mr. Hurt; however, she did review his medi-eal records and the rating reports from Dr. Kaplan and Dr. Beer's office under the AMA Guides. Although Dr. MacGuire agreed with Dr. Kaplan that Mr. Hurt belonged in Class 1, she stated that he was only entitled to an 8% PPI rating.
[17] After considering the three rating reports, the Division issued a final determination assigning Mr. Hurt a 9% impairment rating. He objected and the Division referred the matter to the Medical Commission for a contested case hearing. Mr. Hurt was the only witness to testify at the hearing, but the evidence included his medical records and the deposition testimony of Dr. Beer. The Medical Commission ruled that Mr. Hurt had not met his burden of establishing he was entitled to a PPI rating higher than the 9% assigned by the Division.
[18] Mr. Hurt petitioned the district court for review of the Medical Commission decision. The district court affirmed, and Mr. Hurt appealed to this Court.
STANDARD OF REVIEW
[19] When an appeal is taken from a district court's review of an administrative agency's decision, we examine the case as if *378it came directly from the agency, giving no deference to the district court's decision. Guerrero v. State ex rel. Wyo. Dep't of Workforce Servs., Workers' Comp. Div., 2015 WY 88, ¶11, 352 P.3d 262, 265 (Wyo.2015). See also Dale v. S & S Builders, LLC, 2008 WY 84, ¶8, 188 P.3d 554, 557 (Wyo.2008). Wyo. Stat. Ann. § 16-8-114(c) (LexisNexis 2015) governs judicial review of administrative decisions:
(c) To the extent necessary to make a decision and when presented, the reviewing court shall decide all relevant questions of law, interpret constitutional and statutory provisions, and determine the meaning or applicability of the terms of an agency action. In making the following determinations, the court shall review the whole record or those parts of it cited by a party and due account shall be taken of the rule of prejudicial error. The reviewing court shall:
(i) Compel agency action unlawfully withheld or unreasonably delayed; and
(ii) Hold unlawful and set aside agency action, findings and conclusions found to be:
(A) Arbitrary, capricious, an abuse of discretion or otherwise not in accordance with law;
(B) Contrary to constitutional right, power, privilege or immunity;
(C) In excess of statutory jurisdiction, authority or limitations or lacking statutory right;
(D) Without observance of procedure required by law; or
(E) Unsupported by substantial evidence in a case reviewed on the record of an agency hearing provided by statute.
[110] In accordance with § 16-8-1l14(c), we review the agency's findings of fact by applying the substantial evidence standard. Dale, 122, 188 P.3d at 561. Substantial evidence means " 'such relevant evidence as a reasonable mind might accept as adequate to support a conclusion.'" Guerrero, ¶12, 352 P.3d at 266, quoting Bush v. State ex rel. Wyo. Workers' Comp. Div., 200WY 120, ¶5, 120 P.3d 176, 179 (Wyo.2005) (citation omitted). "Findings of fact are supported by substantial evidence if, from the evidence preserved in the record, we can discern a rational premise for those findings." Kenyon v. State ex rel. Wyo. Workers' Safety & Comp. Div., 2011 WY 14, ¶11, 247 P.3d 845, 849 (Wyo.2011) (citation omitted).
If the hearing examiner determines that the burdened party failed to meet his burden of proof, we will decide whether there is substantial evidence to support the agency's decision to reject the evidence offered by the burdened party by considering whether that conclusion was contrary to the overwhelming weight of the evidence in the record as a whole. If, in the course of its decision making process, the agency disregards certain evidence and explains its reasons for doing so based upon determinations of credibility or other factors contained in the record, its decision will be sustainable under the substantial evidence test. Importantly, our review of any particular decision turns not on whether we agree with the outcome, but on whether the agency could reasonably conclude as it did, based on all the evidence before it.
Dale, ¶22, 188 P.3d at 561 (citations omitted). "We review an agency's conclusions of law de novo, and will affirm only if the agency's conclusions are in accordance with the law." Middlemass v. State ex rel. Wyo. Workers' Safety & Comp. Div., 2011 WY 118, ¶ 13, 259 P.3d 1161, 1164 (Wyo.2011) (citation omitted).
DISCUSSION
[111] The workers' compensation law applicable to PPI benefits is set out in Wyo. Stat. Ann. § 27-14-405 (LexisNexis 2015), which states in pertinent part:
(£) An injured employee suffering an ascertainable loss may apply for a permanent partial impairment award as provided in this section.
(g) An injured employee's impairment shall be rated by a licensed physician using the most recent edition of the American Medical Association's guide to the evaluation of permanent impairment. The award shall be paid as provided by W.S. 27-14-*379408 for the number of months determined by multiplying the percentage of impairment by sixty (60) months.
[[Image here]]
(m) If the percentage of physical impairment is disputed, the division shall obtain a second opinion and if the ratings conflict, shall determine the physical impairment award upon consideration of the initial and second opinion. Any objection to a final determination pursuant to this subsection shall be referred to the medical commission for hearing by a medical hearing panel acting as hearing examiner pursuant to W.S. 27-14-616.
[T12] When a worker's compensation claimant contests the Division's PPI rating, he has the burden of proving he is entitled to a higher rating. Green v. State ex rel. Dep't of Workforce Servs., Workers' Safety & Comp. Div., 2013 WY 81, ¶ 28, 304 P.3d 941, 950 (Wyo.2013). See also Himes v. Petro Eng. & Constr., 2003 WY 5, ¶ 16, 61 P.3d 393, 398-99 (Wyo.2003). The Medical Commission ruled that Mr. Hurt did not prove he was entitled to a higher PPI rating than the 9% assigned by the Division.
[113] Mr. Hurt challenges the Medical Commission's decision on various grounds. First, he claims the Medical Commission incorrectly applied the AMA Guides. This Court provided a general explanation of the use of the AMA Guides for impairment ratings in State ex rel. Wyo. Workers' Safety & Comp. Div. v. Singer, 2011 WY 57, ¶ 18, 248 P.3d 1155, 1160-61 (Wyo.2011):
According to Wyo. Stat. Ann. § 27-14-405(g), an impairment "shall be rated by a licensed physician using the most recent edition of the American Medical Association's guide to the evaluation of permanent impairment." The Guides explain that an "impairment rating enables the physician to render a quantitative estimate of losses to the individual as a result of their health condition, disorder, or disease. Impairment ratings are defined by anatomic, structural, functional, and diagnostic criteria." AMA Guides, at 5. The Guides use the concept of "whole person impairment," which takes into account "the severity of the organ or body system impairment and the resulting functional limitations of the whole person." Id. at 21 (emphasis in original).
[T14] Everyone involved in this case agreed that Mr. Hurt suffered from alteration of motion segment integrity [AOMSI], making the "Motion Segment Lesions" grid in the AMA Guides applicable to him. The dispute is whether Mr. Hurt's level of impairment was properly classified as Class 1 or Class 4 under the Motion Segment Lesions grid. The grid includes classes ranging from 0 to 4, with the severity of the impairment increasing in accordance with the class numbers. Table 17-4 of the AMA Guides describes Class 1 lumbar spine impairments involving AOMSI as: °
Intervertebral disk herniation(s) or documented AOMSI, at a single level or multiple levels with medically documented findings; with or without surgery and for disk herniation(s) with documented resolved radiculopathy or nonverifiable radi-cular complaints at clinically appropriate level(s), present at the time of examination. [Footnote]: Or AOMSI in the absence of radiculopathy, or with documented resolved radiculopathy or nonverifiable radicular complaints at the clinically appropriate levels present at the time of examination.
Class 4 Iumbar spine impairments involving AOMSI include:
Intervertebral disk herniations and/or AOMSI, at multiple levels, with medically documented findings; with or without surgery and with documented signs of residual bilateral or multiple-level radiculopathy at the elinically appropriate levels present at the time of examination (see Table 17-7 to grade radiculopathy ).
[115] After the contested case hearing, the Medical Commission concluded:
15. Hurt was rated for his permanent impairment by Dr. Kaplan and in Dr. Beer's office utilizing the most recent edition of the AMA Guides [6th ed.]. On August 2, 2011, Dr. Kaplan rated Hurt's permanent partial impairment at 9% whole body. Following Hurt's objection to the IME rating by Dr. Kaplan, on September *38026, 2011, Hurt's permanent impairment was rated by Physician Assistant, Andy Be[gluin at 25% whole body. This rating was apparently reviewed after the fact by Dr. Beer who concurred with the rating....
16. Dr. MacGuire did not examine Mr. Hurt. However, Dr. MacGuire reviewed the medical records and commented on the two impairment rating opinions of Dr. Kaplan and Dr. Beer's office, applying her knowledge of the Guides, 6th edition. The value of her opinion is in helping the Hearing Panel evaluate the relative merit of the impairment ratings of Dr. Kaplan and Dr. Beer's office. Pohl v. The Bailey Company, 980 P.2d 816, 821 (1999).
17. Dr. Beer's office incorrectly placed Hurt into class 4 on Table 17-4 on page 570 of the Guides, with alteration of motion segment integrity at multiple levels with documented signs of bilateral radicu-lopathy at the clinically appropriate levels. However, in order to properly document a radiculopathy placing Hurt in grade 4, Hurt must have only slight contraction of motor strength then [sic] no movement or no contraction: an inability to move. Hurt does not have documented evidence of residual bilateral or multilevel radiculopathy at the clinically appropriate levels; Table 17-9 specifies EMG [electromyographic testing] evidence consistent with multiple nerve root radiculopathy. Guides, Chap. 17, Table 17-9, p. 581. EMG and NCV studies have never been completed for Hurt.
Hurt does not fall under class 4 either on table 17-4 or with the required use of table 17-7 to document radiculopathy. Placement in class 4 requires not only the "documented signs of bilateral or multiple-level radiculopathy at the clinically appropriate levels present at the time of evaluation" but also requires the "inability to ambulate - without - assistive - devices." Guides, Chap. 17, Table 17-4, p. 572. Hurt is able to walk, sit, stand, and drive. His motor strength is essentially intact and he falls appropriately under the grade modifier 1.
18. Hurt failed to meet his burden of proof to show by a preponderance of the evidence that on August 2{ ], 2011 he had a permanent impairment greater than the 9% rating upon which he was awarded PPI benefits; he needed to show that he was entitled to more than a 9% whole person permanent impairment rating using the AMA Guides to the Evaluation of Permanent Impairment (6th ed.2008). The Hearing Panel finds that Dr. Kaplan correctly applied the AMA Guides, 6th edition to assign the 9% whole person impairment rating, while Andy Be[gluin, a Physician Assistant, incorrectly placed Hurt in class 4 of the Table 17-4. His rating of 25% permanent impairment was in error, and not credible based upon the medical evidence. The Division properly awarded Hurt PPI benefits based on the 9% permanent impairment rating by Dr. Kaplan.
(footnote and some citations omitted).
[T16] Mr. Hurt claims, under the AMA Guides, the Medical Commission incorrectly used criteria from grade 4 in Table 17-7, which is entitled "Physical Examination Adjustment: Spine," in determining that he belonged in Class 1. He asserts the commission's statement that he did not have "only slight contraction of motor strength then [sic] no movement or contraction: an inability to move," improperly mixed grade criteria with class criteria. To support his argument, Mr. Hurt refers to some instructions on how to use the AMA Guides which apparently dictate that a class be assigned first and then a grade may be assigned within a class. He claims, therefore, that the Medical Commission erred when it referred to Table 17-7 in determining his class designation.
[117] By insisting that Table 17-7 can be used only to assign a grade within a class, not to determine the class, Mr. Hurt ignores the statement in Table 17-4 which specifically directs the use of Table 17-7 when radieu-lopathy is present (see Para 14, supra). More importantly, the medical evaluators did not apply the procedure advocated by Mr. Hurt in arriving at their ratings. In fact, Dr. MacGuire's report specifically referred to the grade 4 radiculopathy factors when she concluded Mr. Hurt's impairment fell within *381Class 1. Mr. Hurt did not offer any medical evidence to show that Dr. MacGuire's interpretation of the AMA Guides was incorrect.
[118] The legislature was clear when it directed that impairment ratings be assigned by licensed physicians using the most recent version of the AMA Guides. That instruction confirms medical training and analysis are necessary to the proper interpretation and application of the AMA Guides It would be improper for us to disregard the expert opinions and conduct our own ratings analysis based upon our interpretation of the AMA Guides and medical records.
[119] Furthermore, while Mr. Hurt conducts an interesting exercise in applying the AMA Guides, we do not find it to be very helpful in determining whether the Medical Commission's decision is supported by substantial evidence. In the end, the difference between Class 1 and Class 4 is determined by whether the person suffers from radiculo-pathy or nonverifiable radicular complaints. The AMA Guides define those terms as follows:
Radiculopathy. For the purposes of the Guides, radiculopathy is defined as significant alteration in the function of a single or multiple nerve roots and is usually caused by mechanical or chemical irritation of one or several nerves. The diagnosis requires clinical findings including specific dermatomal distribution of pain, numbness, and/or parasthesias. Subjective reports of sensory changes are more difficult to assess; therefore, these complaints should be consistent and supported by other findings of radiculopathy. There may be associated motor weakness and loss of reflex. A root tension sign is usually positive. The identification of a condition that may be associated with radieulo-pathy (such as a herniated disk) on an imaging study is not sufficient to make a diagnosis of radiculopathy; clinical findings must correlate with radiographic findings in order to be considered.
Nonverifiable - Radicular - Complaints. Nonverifiable radicular complaints are defined as chronic persistent limb pain or numbness, which is consistently and repetitively recognized in the medical records in the distribution of a single nerve root that the examiner can name and with the following characteristics: preserved sharp vs. dull sensation and preserved muscle strength in the muscles it innervates, is not significantly compressed on imaging and is not affected on electrodiagnostic studies (if performed). Although there are subjective complaints of a specific radicular nature, there are inadequate or no objective findings to support the diagnosis of radiculopathy.
[120] A diagnosis of true radiculopathy is required for a Class 4 rating. In order to meet the requirements for that diagnosis, a claimant must have significant alteration of nerve function. A Class 4 impairment requires documented signs of residual bilateral or multiple level radiculopathy at the elinically appropriate levels. By contrast, a patient with a Class 1 impairment may have radieu-lar complaints that are "inadequate" to be classified as true radiculopathy. The AMA Guides' use of terms like "significant" and "inadequate" recognize that medical judgment is required in applying the AMA Guides and determining the level of radicular problems a person is experiencing. Mr. Hurt would have the Court conduct its own analysis of the meaning and significance of the clinical findings without regard to the medical evaluators' and the Medical Commission's application of medical judgment. That is not the Court's role in performing a substantial evidence review of an agency's decision.
The second aspect of Mr. Hurt's challenge to the Medical Commission's decision is based upon the evidence actually presented at the hearing and the medical evaluator's analysis of that evidence. Mr. Hurt claims the Medical Commission failed to properly recognize the existence and/or severity of his radiculopathy when it accepted Dr. Kaplan's opinion that he belonged in Class 1 rather than Class 4 as surmised by Dr. Beer's physician assistant. Mr. Hurt asserts the report from Dr. Beer's office establishes bilateral radiculopathy because the physician assistant noted that reflexes were absent in both of Mr. Hurt's knees and ankles.
*382[122] The AMA Guides state that when radiculopathy is present there may be a loss of reflexes. However, the definition of radi-culopathy does not state that a finding of loss of reflexes automatically establishes the condition. The AMA Guides specifically note that not all radicular symptoms or conditions are severe enough to qualify as true radiculo-pathy. As the definition of Class 1 states, a patient may have radicular complaints and still fall within that range of impairment. Medical examiners use their judgment in determining the degree of radicular problems.
[T23] Furthermore, Dr. Beer's physician assistant's finding of loss of reflexes is contradicted by other medical records. Another physician in Dr. Beer's office, Radu Segal, M.D., examined Mr. Hurt in the months following the second surgery and preceding the impairment rating and consistently found reflexes bilaterally at the knees and ankles. When Dr. Beer was asked about the apparent contradiction between Dr. Segal's and the physician assistant's findings, he acknowledged they were significantly different. Dr. Beer did not explain the discrepancy or relate the loss of reflexes to any deterioration in Mr. Hurt's condition. Dr. Kaplan, who examined Mr. Hurt just a few weeks before the physician assistant, did not note any loss of reflexes.
[124] Mr. Hurt also asserts the physical findings in Dr. Kaplan's report establish that he suffers from radiculopathy. He directs us to Dr. Kaplan's finding that "[wlith sensory examination, there was some apparent decrease to pinwheel in the right L4 through S1 dermatomes2 This also included his right lateral thigh." It is true that a radieu-lopathy diagnosis under the AMA Guides requires "clinical findings including specific dermatomal distribution of pain, numbness, and/or parasthesias." However, as we said above there are other indicators of true radi-culopathy and the diagnosis depends on the severity of the symptoms. Dr. Kaplan noted the dermatomal findings, but he did not conclude that those findings mandated a diagnosis of radiculopathy for a Class 4 designation. Dr. MacQuire analyzed Mr. Hurt's medical records and Dr. Kaplan's and the physician assistant's ratings and ultimately agreed with Dr. Kaplan that under the AMA Guides, Mr. Hurt was properly placed in Class 1 despite the dermatomal findings.
[125] In addition, although Mr. Hurt is relying upon Dr. Beer's physician assistant's evaluation to support the Class 4 rating, the physician assistant's report did not include any dermatomal findings. When questioned about the report during his deposition, Dr. Beer testified:
Q. And then in terms of the dermato-mal distribution of any radiculopathy, what was the dermatomal distribution in this case?
A. If I-I don't see it documented clearly here. I would say it's an L2-L3 distribution.
Q. It's not documented in the report in terms of findings at the time of the impairment evaluation, correct?
A. Andy's indication [sic] does not specify exactly where the loss was at, no.
[126] To determine whether the Medical Commission's decision on an impairment rating is supported by substantial evidence, we recognize:
"It is the obligation of the trier of fact to sort through and weigh the differences in evidence and testimony, including that obtained from medical experts." In re Worker's Comp. Claim of David v. State ex rel. Wyo. Workers' Safety and Comp. Div., 2007 WY 22, ¶ 15, 151 P.3d 280, 290 (Wyo.2007). Further, we have noted that "The Commission is in the best position to judge and weigh medical evidence and may disregard an expert opinion if it finds the opinion unreasonable or not adequately supported by the facts upon which the opinion is based." Spletzer v. Wyo. ex rel. Wyo. Workers' Safety & Comp. Div., 2005 WY 90, ¶21, 116 P.3d 1103, 1112 (Wyo.2005).
Willey v. State ex rel. Wyo. Workers' Safety & Comp. Div., 2012 WY 144, ¶18, 288 P.3d 418, 426 (Wy.2012).
*383[127] In choosing to reject Dr. Beer's physician assistant's rating and accept Dr. Kaplan's rating as supported by Dr. MacGuire's analysis, the Medical Commission explained that, under § 27-14-405(g), a "licensed physician," rather than a physician's assistant was supposed to perform the rating. The Medical Commission noted the irregularity but apparently did not completely disregard the physician assistant's rating. Instead, the commission decided that Dr. Kaplan's and Dr. MacGuire's opinions were more credible. The Medical Commission's decision to accept Dr. Kaplan's rating is not against the overwhelming weight of the evidence. Dr. Kaplan's rating was consistent with Mr. Hurt's other medical records and supported by Dr. MacGuire. The physician assistant's rating did not include an analysis of the dermatomal distribution and his finding of loss of reflexes was contradicted by the other medical records. Thus, the Medical Commission's determination that Mr. Hurt's radicular problems fell within the definition of nonverifiable radicular complaints in Class 1 instead of true radiculopathy for Class 4 was not against the overwhelming weight of the evidence.
[128] One more matter bears mention in our review of this case. The Medical Commission referenced language from Table 17-4 regarding a patient's inability to ambulate without assistive devices. The Division concedes that criterion did not apply to Mr. Hurt's AOMSI condition but asserts any error in that regard was harmless. We agree. The medical experts did not refer to the factor or use it in determining Mr. Hurt's impairment rating. The Medical Commission's conclusion as to that factor was, therefore, extrancous to the analysis of Mr. Hurt's degree of impairment and its error in referring to it was harmless. See W.R.AP. 9.04 ("any error, defect, irregularity or variance which does not affect substantial rights shall be disregarded by the reviewing court").
[129] Affirmed.
. In Pohl v. Bailey Co., 980 P.2d 816, 821 (Wyo.1999), overruled on other grounds, Torres v. State ex rel. Wyo. Workers' Safety & Comp. Div., 2004 WY 92, 95 P.3d 794 (Wyo.2004), we approved the Division's practice of engaging a third medical professional to assist in evaluating two conflicting impairment ratings.
. Dermatome is defined as "an area of skin that is supplied with the nerve fibers of a single, posterior, spinal root." - Webster's Third New Int'l Dictionary 608 (2002).