McGRAW, Justice.
Two workers’ compensation claimants who suffered back injuries disagreed with the initial disability ratings granted them by the Workers’ Compensation Division and appealed. In each case, the Office of Judges found that examinations conducted under one examination model were unreliable, and granted the claimants higher disability ratings. The employer in each case appealed, and the Workers’ Compensation Appeal Board found the questioned method of examination to be valid, and reinstated the original awards granted by the Workers’ Compensation Division. The claimants appeal, arguing that the Office of Judges was correct in disregarding the questioned examination model because it is in conflict with our law. Because we concur with the logic of the Office of Judges, we reverse.
I.
BACKGROUND
Today we examine two cases, which we consolidated because they concern the same questions of law. Both cases concern the use of a particular diagnostic method, or “model,” for the evaluation of injuries to the spine. One method, the Range of Motion Model, or ROM, measures, as the name implies, the patient’s range of motion across several specific movements. The doctor then compares these ranges against presumed normal ranges for uninjured individuals. Doctors using the other method, the Diagnostic-Related Estimate Model, or DRE, examine a patient’s injury or symptoms and then assign the patient to one of eight categories, or “pigeon holes,” based on those particular symptoms.
Both models are found in the American Medical Association’s Guides to the Evaluation of Permanent Impairment, Fourth Edition (1993). Rules promulgated by the Workers’ Compensation Commissioner command doctors to conduct their examinations and prepare their reports in accordance with this publication when examining patients for permanent partial disability.1 The two models often produce very different impairment ratings, with the DRE usually producing a lower percentage of impairment, and the ROM producing a higher percentage of impairment; each model has its supports and detractors. However, the fourth edition of the [91]*91Guides demands that the DRE Model be used in most circumstances unless a doctor can show some special reason why that model should not be used.
Because of concerns about the DRE Model, discussed at greater length infra, the Workers’ Compensation Office of Judges deemed DRE-based examinations “unreliable” in late 1997. However, the Workers’ Compensation Appeal Board continued to accept DRE-based examinations. The two appellants in this ease both received lower ratings under the DRE Model and higher ratings under the ROM Model. Each appealed a decision of the Workers’ Compensation Division to the Office of Judges, where each prevailed. Then, each of them had that favorable decision reversed by the Workers’ Compensation Appeal Board, and subsequently appealed that adverse decision to this Court. For the reasons set forth below, we find that examinations performed under the DRE are invalid and unreliable, and reverse the decisions of the Appeal Board.
A.
Repass v. WCD and U.S. Steel Mining Co., Inc.
William R. Repass suffered an injury to his lower back on June 29, 1992 while working for his employer, U.S. Steel Mining Company. Mr. Repass filed a claim, and some time later the West Virginia Workers’ Compensation Division (the “Division”) referred Mr. Repass to a Dr. Landis for an evaluation to determine his degree of permanent partial disability, if any. During an examination based upon the DRE Model conducted on October 30, 1996, Dr. Landis found evidence of a healed fracture of one of Mr. Repass’ lumbar vertebrae, and a full range of motion of the lumbar spine. Dr. Landis concluded that Mr. Repass had reached maximum medical improvement, and that he suffered a 5 percent impairment as result of the June 29, 1992 injury. Upon receipt of Dr. Landis’ evaluation, the Division awarded Mi-. Repass an award for a 5 percent permanent partial disability.
Mr. Repass protested the Division’s award and submitted an evaluation report conducted by a Dr. Carlson on August 1, 1997. Dr. Carlson examined Mr. Repass using the ROM Model and concluded that Mr. Repass suffered a 9 percent impairment, specifically, 5 percent for the fracture and 4 percent for limitation of flexion in his back. Thereafter, Mr. Repass’ employer, U.S. Steel Mining Company, had yet another doctor examine Mr. Repass. Dr. Paul Bachwitt examined Mr. Repass on October 14, 1997, and his report referenced both the DRE and ROM methods, but apparently the doctor claimed that the ROM findings were not credible and that the ROM method should not be used. Under the DRE method, Dr. Bachwitt, in agreement with the report of Dr. Landis, found that Mr. Repass should be granted a 5 percent award.
The Workers’ Compensation Office of Judges considered Mr. Repass’ appeal and on March 11, 1998, issued a decision reversing the Division’s prior award of 5 percent, and granting Mr. Repass a 9 percent award based on Dr. Carlson’s findings. In reaching this conclusion, the administrative law judge referenced an earlier opinion of the Office of Judges, known as the Cottrell decision of August 1997,2 in which the chief administrative law judge had determined that examinations conducted using the DRE were unreliable. U.S. Steel Mining Company appealed, and on September 30, 1998, the Workers’ Compensation Appeal Board reversed and reinstated the initial 5 percent award for permanent partial disability granted by the Division. Mr. Repass now challenges the Appeal Board’s final decision.
B.
Bower v. WCD and Maple Meadow Mining Company
Mr. Randall Z. Bower suffered two separate injuries to his back, one on October 3, 1994, and one about two years later on July [92]*9216, 1996. Also, while Mr. Bower was in the midst of his examinations for these injuries, the Office of Judges released the Cottrell decision and accompanying policy memorandum, which necessitated even more examinations. Accordingly, Mr. Bower’s examination history is quite complicated, and the briefs actually contain a chart that sets forth the doctors who examined him and them corresponding findings.
No fewer than eight doctors examined Mr. Bower between January 1996 and December 1998, using either the DRE Model, the ROM Model, or both, and giving him impairment ratings from 4 percent to 16 percent. By April 22, 1997, the Division hád made awards for a total of 5 percent permanent partial disability. After the issuance of Cottrell and a flurry of additional examinations, the Office of Judges reversed the Division and granted Mr. Bower a disability rating of. 16 percent. The Office of Judges apparently disregarded any examination based upon the DRE and applied the liberality rule, basing the award upon the highest impairment rating found under the ROM Model. The employer appealed, and on June 29, 2000, the Appeal Board reversed and reinstated the earlier award by the Division of a 5 percent permanent partial disability. Mr. Bower now asks this Court to reverse the Appeal Board’s June 29, 2000 decision.
II.
STANDARD OF REVIEW
If considering a question of fact, “[i]n most cases we show substantial deference to the factual findings of the Workers’ Compensation Appeal Board.” Plummer v. Workers’ Compensation Division, 209 W.Va. 710, 712, 551 S.E.2d 46, 48 (2001). However, when considering a question of law, as we are in the instant matter, our standard differs: “[w]hile the findings of fact of the appeal board are conclusive unless they are manifestly against the weight of the evidence, the legal conclusions of the appeal board, based upon such findings, are subject to review by the courts.” Barnett v. State Workmen’s Compensation Com’r., 153 W.Va. 796, 812, 172 S.E.2d 698, 707 (1970) (quoting Emmel v. State Compensation Director, 150 W.Va. 277, 145 S.E.2d 29 (1965)). We must also bear in mind that: ‘When the Workers’ Compensation Appeal Board reviews a ruling from the Workers’ Compensation Office of Judges it must do so under the standard of review set out in W. Va.Code § 23-5-12(b) (1995), and failure to do so will be reversible error.” Syl. pt. 6, Conley v. Workers’ Compensation Div., 199 W.Va. 196, 483 S.E.2d 542 (1997). That Code section provides that the Appeal Board may only reverse the Office of Judges when the judge’s findings are in violation of statute, in excess of the judge’s statutory authority, made upon unlawful procedure, or affected by some error of law. The Appeal Board must also reverse if the administrative law judge’s decision is clearly wrong, or arbitrary, capricious or characterized by abuse of discretion. W. Va.Code § 23-5-12(b) (1995).
III.
DISCUSSION
A. Liberality
Though often stated by the Court, it bears repeating that we look to one polar star for guidance in any workers’ compensation case. “ ‘The Workmen’s Compensation Law is remedial in its nature, and must be given a liberal construction to accomplish the purpose intended.’ Syl. pt. 3, McVey v. Chesapeake & Potomac Telephone Co., 103 W.Va. 519, 138 S.E. 97 (1927) (citation omitted).” Syl. pt. 1, Plummer v. Workers’ Compensation Division, 209 W.Va. 710, 551 S.E.2d 46 (2001). The West Virginia Workers’ Compensation system exists to the benefit of both employers and employees, freeing employers from lawsuits for simple negligence while ensuring employees compensation for them work-related injuries.
The Act is designed to compensate injured workers as speedily and expeditiously as possible in order that injured workers and those who depend upon them for support shall not be left destitute during a period of disability. The benefits of this system accrue both to the employer, who is relieved from common-law tort liability for negligently inflicted injuries, and to the [93]*93employee, who is assured prompt payment of benefits.
Meadows v. Lewis, 172 W.Va. 457, 469, 307 S.E.2d 625, 638 (1983). In restating this principle, we are not unmindful of the perception that our system makes our state less attractive to some businesses. But it is for the Legislature, not the Commissioner, or the courts, to make the difficult and sometimes unpopular decisions necessary to fund the system or limit its expenses. The Commissioner must follow the commands of the Legislature as expressed by statute. In the absence of express statutory authority, it is not the Commissioner’s duty to take major steps to limit compensation to deserving workers, or design systems that make recovery less likely. Indeed:
“[The Workers’ Compensation Act] re-quir[es] the state compensation commissioner in administering the workmen’s compensation fund, to ascertain the substantial rights of the claimants in such manner as will carry out justly and liberally the spirit of the act, unrestricted by technical and formal rules of procedure -” Syllabus, in part, Culurides v. Ott, 78 W.Va. 696, 90 S.E. 270 (1916) (citation omitted);
Martin v. Workers Compensation Div., 210 W.Va. 270, 275, 557 S.E.2d 324, 329 (2001) (quoting syl. pt. 2, Plummer v. Workers’ Compensation Division, 209 W.Va. 710, 551 S.E.2d 46 (2001)). This does not mean that any person claiming an injury should instantly be awarded the maximum possible compensation. We recognize that workers’ compensation is not the equivalent of a welfare system.3 The command of the law that the workers’ compensation process be “unrestricted by technical and formal rules of procedure” means that, if the statutes would provide a worker with compensation, the Commissioner cannot create barriers or hurdles that would prevent that recovery. If the system is not producing the desired results, then the Legislature can change the statute, and in so doing craft a solution suitable to a majority of the electorate. We recognize that the ultimate responsibility for the fiscal health of the West Virginia Workers’ Compensation system rests with the Legislature. Balancing the conflicting goals of minimizing premiums while providing full and fair compensation to injured workers is the exclusive province of our publicly-elected legislators, and is not to be invaded by the Commissioner, or the Courts.
And so bearing in mind that “although the rules and regulations governing the workers’ compensation system in this state are necessarily detailed and complex, we must be careful to prevent those deserving of compensation from being thwarted by technicalities or procedural niceties,” Martin v. Workers Compensation Div., 210 W.Va. 270, 275, 557 S.E.2d 324, 329 (2001), we turn to an examination of the instant dispute.
B. Why the Guides are at issue
This case turns upon the propriety of the Commissioner’s adoption of what the parties refer to as “the Guides,” as the standard by which permanent impairment shall be determined. The full name of this publication is, The American Medical Association’s Guides [94]*94to the Evaluation of Permanent Impairment, Fourth Edition, and we shall refer to it simply as, “Guides, Fourth ” in the text of this opinion, so that it is not confused with earlier or later editions of the same publication.4 The Guides, Fourth are at issue in this case by virtue of the Legislature’s delegation of authority to the Commissioner, and a rule issued by the Commissioner adopting the Guides, Fourth for certain purposes.
Before addressing the application of this publication to examinations of spinal injuries, we first explain how it is that this publication is at issue in the cases below. As we have discussed in previous cases, the Legislature made significant changes to the workers’ compensation system in 1995.
On February 10, 1995, the West Virginia Legislature passed and enacted Committee Substitute for Senate Bill 250, which comprehensively revised numerous aspects of West Virginia workers’ compensation law. The purported goal of these sweeping reforms envisioned ameliorating the workers’ compensation fund’s fiscal crisis and restoring its financial integrity. See State ex rel. Blankenship v. Richardson, 196 W.Va. 726, 729-31, 474 S.E.2d 906, 909-11 (1996).
State ex rel. ACF Industries, Inc. v. Vieweg, 204 W.Va. 525, 529, 514 S.E.2d 176, 180 (1999) (footnote omitted). Of specific importance to this case is that the Legislature altered the way in which one discovers the extent of an injured worker’s permanent partial disability; under current law, one determines the level of permanent disability by first finding the worker’s degree of whole body medical impairment. The Legislature also granted the Commissioner authority to establish standards for the determination of that impairment:
[T]he degree of permanent disability other than permanent total disability shall be determined exclusively by the degree of whole body medical impairment that a claimant has suffered.... The workers’ compensation division shall adopt standards for the evaluation of claimants and the determination of a claimant’s degree of whole body medical impairment. Once the degree of medical impairment has been determined, that degree of impairment shall be the degree of permanent partial disability that shall be awarded to the claimant.
W. Va.Code § 23-4-6(i) (1999).5 Of course the ultimate question being asked is, “how much, if any do we (the State) owe the injured worker?” Before 1995, our system took into account factors beyond medical impairment when making a determination about permanent partial disability. As Justice Stareher explained in a concurring opinion,
Prior to the Legislature’s amendments in 1995, the Workers’ Compensation Act distinguished between impairment, which is a [95]*95medical question, and disability, which is a legal question. To determine impairment, a doctor would examine the claimant and render a scientific opinion regarding how much a claimant’s physical functions were impaired by a work-related injury. The Workers’ Compensation Commissioner would then determine disability by looking at the doctor’s opinion on impairment, and mix that opinion with the evidence of the claimant’s earning capacity, the effect of the impairment on the claimant’s efficiency at work, and the effect of the impairment on the claimant’s pursuit of normal everyday living. From a mix of these factors, the Commissioner would compute the claimant’s percentage of permanent partial disability. The Commissioner’s permanent partial disability award would, in theory, only partially take into account the doctor’s determination of impairment.
In 1995 the Legislature amended W. Va.Code, 23-4-6(i) to state that “the degree of permanent disability other than permanent total disability shall be determined exclusively by the degree of whole body medical impairment that a claimant has suffered.” This amendment altered the Workers’ Compensation Act in two ways significant to this case. First, after 1995 doctors are to make impairment evaluations using a standardized, “whole body” impairment rating system — in other words, they are to use the American Medical Association’s Guides to the Evaluation of Permanent Impairment, Fourth Edition (1993). Second, the Commissioner is to make permanent partial disability awards solely on the basis of the doctor’s impairment evaluation. Put another way, the percentage of medical impairment now equals the percentage of permanent partial disability, and the Commissioner cannot take into consideration any other factors.
Wagner v. Workers’ Compensation Div., 205 W.Va. 186, 191-92, 517 S.E.2d 283, 288-89 (1998) (per curiam) (Starcher J., concurring) (footnote omitted).
Moreover, the Commissioner has equated an impairment rating conducted under this standard to be the equivalent of a disability rating for permanent impairment. The authors of The Guides themselves urge against any system that equates medical impairment with disability. As we noted in a case challenging the constitutionality of the 1995 changes:
See 85 CSR 16-3.4 (defining “permanent impairment” and “impairment,” according to the Guides, and indicating that “[a] claimant’s degree of permanent whole body medical impairment is to be determined in keeping with the determination of whole person permanent impairment as set forth in the Guides.” Id., in relevant part.)
The exclusive use of the Guides for evaluation of impairment and, therefore, permanent partial disability, is not recommended by its author:
“The critical problem is that no formula is known by which knowledge about a medical condition can be combined 'with knowledge about other factors to calculate the percentage by which the employee’s industrial use of the body is impaired. Accordingly, each commissioner or hearing official must come to a conclusion on the basis of assessment of the available medical and nonmedical information. The Guides may help resolve such a situation, but it cannot provide complete and definitive answers. Each administrative or legal system that uses permanent impairment as a basis for disability ratings should define its own means for translating knowledge about an impairment into an estimate of the degree to which the impairment limits the individual’s capacity to meet personal, social, occupational, and other demands or to meet statutory requirements.
It must be emphasized and clearly understood that impairment percentages derived according to Guides criteria should not be used to make direct financial awards or direct estimates of dis-. abilities.”
Id. at 1/4 — 1/5 (bold provided [in text of Guides ]).
In light of the AMA’s admonition against the exclusive use of its Guides for evaluation of permanent disability, we question the Commissioner’s wisdom in adopting them.
[96]*96State ex rel. Blankenship v. Richardson, 196 W.Va. 726, 735-36, n. 13, 474 S.E.2d 906, 915-16, n. 13 (1996) (quoting Guides, Fourth, emphasis in original). While we still have our concerns about a system that makes a one-to-one correlation between impairment and disability, and while we continue to question the wisdom of employing the Guides, Fourth in a fashion contrary to the intent of the authors, neither issue is squarely before the Court today. What is before the Court, is whether the adoption of the Guides, Fourth as a mandatory set of instructions for performing an examination of a claimant with a spinal injury is consistent with our workers’ compensation law.
Pursuant to the W: Va.Code § 23 — 4—6(i) (1999), the Division6 issued a rule that established standards for determining an injured worker’s degree of permanent impairment:
§ 85-16-4. Adoption of Standards.
4.1. Except as provided for in section 6 of this rule, on and after the effective date of this rule all evaluations, examinations, reports, and opinions with regard to the degree of permanent whole body medical impairment which a claimant has suffered shall he conducted and composed in accordance with the “Guides to the Evaluation of Permanent Impairment,” (4th ed.1993), as published by the American Medical Association. If in any particular claim, the examiner is of the opinion that the Guides or the section 6 substitutes cannot be appropriately applied or that an impairment guide established by a recognized medical speciality group may be more appropriately applied, then the examiner’s report must document and explain the basis for that opinion. Deviations from the requirements of the Guides or the section 6 substitutes shall not be the basis for excluding evidence from consideration. Rather, in any such instance such deviations shall be considered in determining the weight that will be given to that evidence. An example of an acceptable recognized medical speciality group’s own guides is the “Orthopaedic Surgeons Manual in Evaluating Permanent Physical Impairment.”
85 C.S.R. § 85-16-4(1996)(emphasis added)7.
We find it extremely important to note that the rule commands a doctor to conduct the examination and compose the report “in accordance with the guides” and not simply to use the models or suggestions contained in the publication. Indeed, the basis of the appellees’ argument is that, minus the exceptions specified in the rule, the whole of the Guides, Fourth, including the DRE section, must be followed by any doctor performing an examination on a claimant. As a result of this phrasing, the Division has essentially incorporated by reference into the rule, the entirety of the Guides, Fourth. While the publication runs some three hundred pages and encompasses all aspects of impairment of many bodily systems, we are concerned in this case with injuries of the spine. In the section dealing with spinal injuries, the Guides, Fourth describes two methods of determining impairment. One is called the [97]*97ROM, or Range of Motion Model. The other is called the DRE, or Diagnosis-Related Estimate Model, or simply, the Injury Model, and was newly introduced in the fourth edition of the Guides.8 We shall make reference to either ROM or DRE for the remainder of this opinion.
C. Conñict with Statute
The Guides, Fourth is far more than a chart or reference table consulted by doctors; it is a complete set of extremely detailed instructions as to how a doctor should evaluate a claimant’s injury. Of particular importance in the instant case is that the Guides, Fourth mandates the use of the DRE model for most back injuries.
The evaluator assessing the spine should use the Injury Model [the DRE], if the patient’s condition is one of those listed in Table 70 (p. 108). That model, for instance, would be applicable to a patient with a herniated lumbar disk and evidence of nerve root irritation. If none of the eight categories of the Injury Model is applicable, then the evaluator should use the Range of Motion Model.
Guides, Fourth, § 3.3, p. 94. Implicit in this instruction is that very few claimants would ever be evaluated under the ROM Model, since most would fall into one of the eight categories. The Guides, Fourth, goes on to list other specific instructions as to not only how a doctor should perform an examination, but what sort of data a doctor should consider, and in some eases, what weight the doctor should give that data. Again, we stress that, by virtue of incorporation by rule, the Guides, Fourth section on spinal injuries is far more than a guide, model, or illustration; it is a mechanical set of rules that compels a doctor to consider “x” but exclude “y.”
It is with this controlling aspect of the Guides, Fourth that appellants most strongly disagree. Appellants point out that several commands of the Gtddes, Fourth appear to be in conflict with specific dictates of the Legislature, and that the spine injury section of the Guides, Fourth is also internally inconsistent. In essence, appellants echo the logic expressed in a “policy memorandum” issued by the then-chief administrative law judge, which rejected the DRE Model. Chief Administrative Law Judge Robert Smith issued the policy memorandum shortly after rendering a decision in the case of Thelma Cottrell v. Workers’ Compensation Division and Community Health Associates, Claim No. 92-56811 (August 22, 1997). In that case, the chief administrative law judge determined that the DRE Model ran afoul of several Code provisions, and as a result, reports conducted in accordance with the DRE are “unreliable” and thus should not be considered in a determination of disability. Ap-pellees argue that the decision of the chief administrative law judge was in excess of his statutory authority, and cannot bind either the Division or the Appeal Board.
Before reaching the question of the administrative law judge’s authority, we examine appellants’ argument that the DRE conflicts with our workers’ compensation statutes. First, appellants point out that the text of the Guides, Fourth states that the DRE Model considers a claimant’s impairment at the time of the injury, not the time of evaluation, that the text of the DRE Model excludes consideration of occupational wear and tear or so called “developmental findings,” and finally, that the Guides, Fourth rates one specific injury, but cannot be used where there are sequential injuries to the same body part or when an injury grows worse over time. It is clear that several statutes and opinions of this Court are in apparent conflict with the Guides, Fourth.
First, we note that the section we quoted, supra, states that “the degree of permanent disability other than permanent total disability shall be determined exclusively by the degree of whole body medical impairment that a claimant has suffered.” W. Ya.Code § 23 — 4—6(i) (1999) (emphasis added). We be[98]*98lieve “has suffered” implies that any rating of permanent partial impairment must be made when the claimant has healed as much as possible. Furthermore, the Code suggests that the permanent partial impairment rating of an injured worker must be made after he or she has reached the maximum degree of improvement.
(c) When the authorized treating physician concludes that the claimant has either reached his or her maximum degree of improvement or is ready for disability evaluation, or when the claimant has returned to work, such authorized treating physician may recommend a permanent partial disability award for residual impairment relating to and resulting from the compensa-ble injury.
W. Va.Code § 23-4-7a(c)(1995); accord, Dalton v. Spieler, 184 W.Va. 471, 473, 401 S.E.2d 216, 218 (1990). Clearly, this Code section does not contemplate an impairment rating made based upon the claimant’s condition at the time of injury, or even at some time after injury that is prior to reaching the maximum degree of improvement. However, on its face, the DRE Model suggests that once an estimate of impairment is made, subsequent treatment or impi’ovement of the injury should not change the initial estimate.
With the Injury Model [the DRE], surgery to treat an impairment does not modify the original impairment estimate, which remains the same in spite of any changes in signs or symptoms that may follow the surgery and irrespective of whether the patient has a favorable or unfavorable response to treatment.
Guides, Fourth at § 3.3d, p. 100. Although this statement seems at odds with other language in the spine section of the Guides, Fourth,9 it nonetheless directs a physician to ignore the outcome of a claimant’s treatment, and to rely upon the first estimate of impairment. By virtue of the Commissioner’s decision to incorporate the Guides, Fourth, we are left with a nonsensical command that conflicts with the statute. Although during oral argument before this Court, certain parties intimated that this command is, in practice, ignored, we note that ignoring it doesn’t make it disappear'. Thus, we believe that this aspect of the DRE Model conflicts with W. Va.Code §§ 23-4-7a (1995) and 23-4-6(i) (1999).
Appellants also argue that the DRE Model, as set forth in the Guides, Fourth, is geared more toward rating single incident traumas to the spine, and purports to exclude any injury that appears over time.
The Injury Model [the DRE] attempts to document physiologic and structural impairments relating to insults other than common developmental findings, such as (1) spondylolysis, found normally in 7% of adults; (2) spondylolisthesis, found in 3%; 10(3) herniated disk without radiculo-pathy, found in more than 30% of individuals by age 40 years; and (4) aging changes, common in 40% of adults after age 35 years.
Guides, Fourth, § 3.3d, p. 100 (footnotes omitted) (footnote added). This sort of analysis would be inappropriate where a workers’ injury has occurred over time. We have made clear that: “An employee who is injured gradually by reason of the duties of employment and eventually becomes disabled is, under our workmen’s compensation law, no less the recipient of a personal injury than one who suffers a single disabling trauma.” Syl. pt. 3, Lilly v. Workmen’s Compensation Commissioner, 159 W.Va. 613, 225 S.E.2d 214 (1976); accord, Syl. pt. 2, Sansom v. Workers’ Compensation Commissioner, 176 W.Va. 545, 346 S.E.2d 63 (1986). We have also stated that:
A compensable injury which does not initially or of itself produce a permanent total disability may become progressively worse over time or combine with prior impairments under the second injury statute, W. [99]*99Va.Code, 23-3-1, so as to result in a permanent total disability.”
Syl. pt. 1, Lambert v. Workers’ Compensation Division, 211 W.Va. 436, 566 S.E.2d 573 (2002) (quoting Miracle v. Workers’ Compensation Commissioner, 181 W.Va. 443, 383 S.E.2d 75 (1989)). Obviously, our law contemplates that an injury is not necessarily static, and can become worse over time, a concept to which the DRE is at odds, and to which it must yield.
Another argument advanced by appellants is that W. Va.Code § 23-4-16 (1995) allows a claimant to “re-open” his or her claim if a compensable injury worsens. The Code provides that “in any claim in which an award of permanent disability was made, any such request must be made within five years of the date of the initial award. During that time period, only two such requests may be filed.” As we noted above, the authors of the DRE state that the original impairment estimate remains the same, even if a claimant’s condition worsens or improves. Thus, a claimant who attempts to reopen a claim would appear to be blocked by this language. Moreover, even if we accept the appellees’ argument that this language can be ignored or explained, a claimant evaluated under the DRE Model who attempts to reopen still faces the enormous challenge of getting out of the “box” or category of his or her initial rating.
This same problem also befalls a claimant who does not attempt to reopen for a change in an injury, but is simply injured a second time in a separate incident. The Code provides for additional compensation for claimants who become further impaired as a result of a second injury. See, W. Va.Code § 23-4-6(d) (1999). Because the DRE puts a claimant in one category or another, his or her rating will be 5 percent, 10 percent, 15 percent, etc. Once rated at 5 percent, for example, a claimant must suffer a severe enough second injury to make the quantum leap between the 5 percent category and the 10 percent eategoi-y. Thus, a claimant whose second injury under the ROM Model would have moved him or her from 5 percent to 7 percent, would remain stuck at 5 percent under the DRE Model. Thus, there is effectively no compensation for a second injury under this scenario, which we believe conflicts with our law.
D. The Office of Judges
As we noted, it is because of these specific conflicts that the Workers’ Compensation Office of Judges developed a policy that rejected examinations performed under the DRE Model as being contrary to law and thus unreliable. Appellees argue that the chief administrative law judge had no authority to issue a so-called policy memorandum directing other administrative law judges to exclude examinations conducted pursuant to the DRE. We have noted that the Office of Judges has the authority to create its own rules of practice and procedure: “The Legislature has empowered the OOJ to craft rules of practice and procedure for the review of disputed claims.” Plummer v. Workers’ Compensation Division, 209 W.Va. 710, 715, 551 S.E.2d 46, 51. The authorizing statute states:
Subject to the approval of the compensation programs performance council pursuant to subdivisions (b) and (c), section seven, article three, chapter twenty-one-a of this code, the office of judges shall from time to time promulgate rules of practice and procedure for the hearing and determination of all objections to findings or orders of the workers’ compensation division pursuant to section one of this article. The office of judges shall not have the power to initiate or to promulgate legislative rules as that phrase is defined in article three, chapter twenty-nine-a of said code.
W. Va.Code § 23-5-8(e) (2001). We also note that the Code provides the chief administrative law judge with broad authority to exercise such powers as may be necessary for the proper administrative review of disputed claims.
The chief administrative law judge shall continue to have the power to hear and determine all disputed claims in accordance with the provisions of this article, establish a procedure for the hearing of disputed claims, take oaths, examine witnesses, issue subpoenas, establish the amount of witness fees, keep such records and make such reports as are necessary [100]*100for disputed claims and exercise such additional powers, including the delegation of such powers to administrative law judges or hearing examiners as may be necessary for the proper conduct of a system of administrative review of disputed claims. The chief administrative law judge shall make such reports as may be requested of him or her by the compensation programs performance council.
W. Va.Code § 23-5-8(f) (2001).
Appellees still argue that the decision in Cottrell, and subsequent policy memorandum, could not act as binding precedent on other administrative law judges, the Division, or the Appeal Board. Appellants argue that the Cottrell policy memorandum was never intended to be applied in that fashion, but rather provided notice to all litigants that medical evaluations relying upon the DRE Model were unreliable to support a finding regarding a claimant’s degree of impairment before the Office of Judges, only. Thus, they argue, this memorandum did not promulgate a legislative rule as asserted by defense counsel nor did it establish a precedent before the Division or Appeal Board. That is to say, argue appellants, the policy memorandum about Cottrell was necessary precisely because the decision in Cottrell did not have precedential value.
We concur that W. Va.Code § 23-5-8(f) (2001) gives the chief administrative law judge broad authority to manage cases before the Office of Judges.11 We agree with the appellees that neither a policy memorandum, nor a decision by the Office of Judges can bind the Division or the Appeal Board in all future claims, but we find no authority that would prohibit the chief administrative law judge from enacting a uniform policy for his or her office to follow.
E. Incorporation by Reference
Some might argue that the use of the Guides, Fourth amounts to an impermissible delegation of the Legislature’s power to a group of non-elected doctors and experts. However, we note that the delegation of authority to an agency for rule-making purposes is permitted. It is clear that the Legislature may delegate certain powers or responsibilities to an agency: “ ‘The delegation by the legislature of broad discretionary powers to an administrative body, accompanied by fitting standards for their exercise, is not of itself unconstitutional.’ Point 8 Syllabus, Chapman v. Huntington, West Virginia, Housing Authority, 121 W.Va. 319 [, 3 S.E.2d 502 (1939) ].” Syl. pt. 5, State ex rel. W. Va. Hous. Dev. Fund v. Copenhaver, 153 W.Va. 636, 171 S.E.2d 545 (1969). And we note that incorporation of a standard not created by the Legislature is permitted, provided that the standard is static. Because of changes made in the Guides, some might argue that the proper course of action would be to apply the language of the most recent version. However, changes to an already adopted standard cannot be made “automatically.”
In a case from North Dakota, the relevant statute required examinations be conducted in accordance with “the most recent edition” of the Guides. On the date of the statute’s enactment, the most recent edition was the third revised edition of the Guides, but when the worker was evaluated in 1994, the Guides, Fourth was the most recent edition. The worker received a 20 percent evaluation from his doctor under the ROM Model, but the Bureau had him evaluated by another doctor under the DRE Model, which resulted in an impairment rating of 10 percent. The worker appealed and the Supreme Court of North Dakota held that, while incorporation of an existing standard is permissible, the attempted adoption of a future standard would be unconstitutional.
[101]*101Numerous other courts hold that a statute that attempts to incorporate future changes of another statute, code, regulation, standard, or guideline is an unconstitutional delegation of legislative power. See, e.g., International Ass’n of Plumbing and Mechanical Officials v. California Bldg. Standards Comm’n, 55 Cal.App.4th 245, 64 Cal.Rptr.2d 129, 134 (1997); People v. Pollution Control Bd., 83 Ill.App.3d 802, 38 Ill.Dec. 928, 932-933 404 N.E.2d 352, 356-357 (1980); Gumbhir v. Kansas State Bd. of Pharmacy, 228 Kan. 579, 618 P.2d 837, 842-843 (1980); Board of Trustees v. City of Baltimore, 317 Md. 72, 562 A.2d 720, 731 (1989); Michigan Mfrs. Ass’n v. Director of Workers’ Disability Compensation Bureau, 134 Mich.App. 723, 352 N.W.2d 712, 715 (1984); Meyer v. Lord, 37 Or.App. 59, 586 P.2d 367, 371 (1978); City of Chamberlain v. R.E. Lien, Inc., 521 N.W.2d 130, 132-133 (S.D.1994); Independent Community Bankers Ass’n v. State, 346 N.W.2d 737, 744 (S.D.1984); Woodson v. State, 95 Wash.2d 257, 623 P.2d 683, 685 (1980).... We agree with those courts taking this view.
McCabe v. North Dakota Workers Compensation Bureau, 567 N.W.2d 201, 204-05 (N.D.1997).
The distinction is that, when an existing standard is incorporated by reference, there is the presumption that a legislature is familiar with that standard in its entirety and approves of it. However, by attempting to incorporate a standard, plus any modifications it might undergo, a legislature is delegating its authority to the non-elected authors of the standard, who could then change the standard in some way not contemplated by the legislature. See, Michigan Mfrs. Ass’n v. Director of Workers’ Disability Compensation Bureau, 134 Mich.App. 723, 352 N.W.2d 712, 715 (1984) (statute was not an impermissible delegation of authority to an association to create a standard because statute made a valid adoption of a standard already in existence.)
However, while permissible, the incorporation of a publication like the Guides, Fourth is far more troublesome than the incorporation of pure data, or a printed chart, or a schedule of impairment values.12 Statutes and rules are written by lawyers with the understanding that any word written may eventually be subject to construction or attack by other lawyers. Thus, statutes and rules generally have very specific definitions and the authors take pains to make them internally consistent and as unambiguous as possible. An instruction manual written by doctors for doctors is not usually composed with these concerns in mind.
Appellees argue that the advantage of the DRE is that it is somehow more objective than the ROM. Indeed, the Guides state that the purpose of the DRE is to counter objection that the ROM is too subjective:
One of the purposes of the Guides is to lead to similar results when different clinicians evaluate illnesses and impairments. For evaluating spine impairments, past Guides editions have used a system based on assessing the degree of spine motion and assigning impairment percentages according to limitations of motion. Impairment percents related to the range of motion were to be combined with percents based on diagnoses or therapeutic approaches and neurologic impairments.
One concern with the range of motion system has been that in applying it, other clinical data and diagnostic information tend to be ignored. Also some physicians are concerned about the accuracy and reproducibility of mobility measurements, while others believe the system fails to account for the effects of aging.
Guides, Fourth § 3.3. p. 94. However, we find it important to restate that the spinal [102]*102section of the Guides, Fourth is not simply a “guide” as the name would imply — it is a set of instructions to be followed. While a uniform set of instructions, on its face, would appear to offer some degree of objectivity in examinations, one must be convinced that .the process that created that set of instructions was itself objective, reasonable, or impartial. We remain unconvinced that the DRE is somehow perfectly objective. As one scholar' has explained, it is wrong to assume that the Guides are somehow perfect or flawless.
[T]he Guides is not the objective, medical evaluative system that it purports to be and that has been so appealing to legislators and other decision makers. Instead, like any impairment rating scheme, it rests in large part on important and difficult normative judgments. Yet the Guides obscures this from the x*eader; it is laden with hidden or poorly explained value judgments that frequently are gender-biased. The Guides’ flawed promises of objectivity are especially troubling because they appeal to the craving of legislators and other decision makers for certainty and clarity in the difficult arena of impairment and disability assessment. By uncritically embracing the Guides, these decision makers delegate significant normative decisions to the book’s medical authors.
Ellen Smith Pryor, Flawed Promises: A Critical Evaluation of the American Medical Association’s Guides to the Evaluation of Permanent Impairment, 103 Harv. L.Rev. 964, 965-66 (1990) (footnote omitted).13
Of course, the Legislature can adopt any system it wants, within the ambit of our constitution, to evaluate injuries, impairments, or disabilities. But we must recognize that merely labeling an evaluation system as objective and making uniform application of it does not somehow make that system a perfect one.
This vast literature on measurement, at least implicitly, makes a critical point: it is impossible to define and measure impairment (both organ and whole-person level) except in relation to some norm. Hence, any organ-level impairment evaluation requires (1) a decision about what organ-level qualities (e.g., range of motion) are relevant to the measurement, and (2) a decision about what degree of that quality will serve as the norm. Similarly, a whole-person impairment evaluation requires (1) a decision about what activities or qualities are relevant to the measurement, and (2) a decision about what levels of ability will serve as the norm.
The need for these decisions dooms the Guides’ claims that its impairment evaluation scheme is objective and purely medical. These claims seem most appealing with respect to some organ-level impairments, such as those to the orthopedic extremities, where one might argue that a basic consensus exists on the qualities that should serve as the norm (such as range of motion). But even this argument does not justify the Guides’ claims of objectivity and of its medical ñatee.
Id. at 968-69 (footnote omitted). In short, the Commissioner, in concert with the Performance Council may, within the bounds of them authority, adopt the standard of their choosing, but we must remember that the introduction of the Guides, Fourth into our workers’ compensation system was not accompanied by a burning bush, or even direct action of the Legislature. Thus, when we find it to be in conflict with our existing statutory law, we must adhere to the law.
There is no question that when the rules of an agency come into conflict with a statute that the statute must control:
Any rules or regulations drafted by an agency must faithfully reflect the intention of the Legislature, as expressed in the controlling legislation. Where a statute contains clear and unambiguous language, an agency’s rules or regulations must give that language the same clear and unambiguous force and effect that the language commands in the statute.
Syl. pt. 4, Maikotter v. University of West Virginia Bd, of Trustees/West Virginia Univ., 206 W.Va. 691, 527 S.E.2d 802 (1999). Or in other words: “Although an agency may have power to promulgate rules and regulations, the rules and regulations must be rea[103]*103sonable and conform to the laws enacted by the Legislature.” Anderson & Anderson Contractors, Inc. v. Latimer, 162 W.Va. 803, 807-08, 257 S.E.2d 878, 881 (1979) (citing Sheppe v. West Virginia Bd. of Dental Exmrs., 147 W.Va. 473, 128 S.E.2d 620 (1962)).
The power of the Legislature is paramount when a court is faced with a conflict between a statute and a rule:
It is fundamental law that Legislature may delegate to an administrative agency power to make rules and regulations to implement the statute under which the agency functions. In exercising that power, however, an administrative agency may not issue a regulation which is inconsistent with, or which alters or limits its statutory authority.
Syl. pt 3, Rowe v. W. Va. Dept. of Corrections, 170 W.Va. 230, 292 S.E.2d 650 (1982); accord, CNG Transmission Corp. v. Craig, 211 W.Va. 170, 564 S.E.2d 167 (2002); State ex rel. McKenzie v. Workers’ Compensation Commissioner, 212 W.Va. 288, 569 S.E. 809 (2002.)14 Though the courts have the power to harmonize a rule with an ambiguous statute, we must follow the will of the Legislature when expressed with clarity. “The judiciary is the final authority on issues of statutory construction, and we are obliged to reject administrative constructions that are contrary to the clear language of a statute.” Syl. pt. 5, CNG Transmission Corp. v. Craig, 211 W.Va. 170, 564 S.E.2d 167 (2002).
As we noted earlier in this opinion, in those instances where an agency rule addresses some issue that is already the subject of Legislative action, “[i]f the intention of the Legislature is clear, that is the end of the matter, and the agency’s position only can be upheld if it conforms to the Legislature’s intent.” Syl. pt. 3, in part, Appalachian Power Co. v. State Tax Department of West Virginia, 195 W.Va. 573, 466 S.E.2d 424 (1995).
We must apply the same standard to the rule in this case. Thus, we hold that a rule promulgated by the workers’ compensation division that mandates the use of a non-legislatively created guide for the examination of certain injuries is valid only to the extent that the mandated guide does not conflict with the specific dictates of the Legislature as expressed by statute. Those aspects of the mandated guide that are in conflict are invalid.
The DRE Model for the evaluation of spinal injuries conflicts with our law in several areas. The DRE disagrees with statutes that control: the proper time for making an impairment rating, the proper treatment of progressive injuries, the procedure for reopening a claim, and the consideration of a second injury. Any aspect of the Guides, Fourth that conflicts with these statutes must fail. Accordingly, we hold, because the Diagnosis-Related Estimate Model for the examination of spine injury claims, as set forth in The American Medical Association’s, Guides to the Evaluation of Permanent Impairment, Fourth Edition (1993), cannot be reconciled with several specific workers’ compensation statutes promulgated by the West Virginia Legislature, any medical examination conducted in accordance with that model is invalid and unreliable.
Some might argue that the Guides, Fourth were adopted in an effort to hold down costs in the workers’ compensation system. In arguments before this Court, counsel for appellants suggested that compensation for back injuries was greatly reduced by the application of the DRE Model. We note that at least one observer suggests that the problems of our workers’ compensation system are not the result of overpayment, but rather of under-collection.15 But we take the more [104]*104charitable view that the Commissioner adopted the Guides, Fourth in an effort to determine once and for all “objective” rules or procedures that produce a “fair” result. But only one thing is clear: no matter what sort of guides, rules, procedures, or standards the Legislature or Commissioner may adopt, none can yield an “exact” or “correct” estimate of impairment or disability that is beyond reproach. “The rating of disabilities, regardless of legislative precision or medical expertise, remains an inexact science.” Griffith v. State Workmen’s Compensation Comm’r, 157 W.Va. 837, 843, 205 S.E.2d 157, 161 (1974); Accord, Miracle v. Workers’ Compensation Comm’r, 181 W.Va. 443, 446, 383 S.E.2d 75, 78 (1989); Lambert v. Workers’ Compensation Division, 211 W.Va. 436, 566 S.E.2d 573 (2002).
We are also concerned that, in a quest for bright line rules and predictability, participants in the workers’ compensation system may latch onto “buzzwords” or phrases to decide a case. We fear that most any sentence in this, or any opinion, could be fastened upon by the Commissioner, Office of Judges, or Appeal Board and be imbued with near magical powers and employed almost as a spell or incantation that can “open the lock” for a claimant seeking compensation. We wish to do all that we can to discourage the use of such “magic words,” as we recently noted in a case concerning permanent total disability claims:
In practical application, this holding seems to have resulted in the Division searching medical, psychiatric, and vocational reports for a magical phrase stating “the claimant is permanently and totally disabled from future employment,”... If the magical words have not been included, the content to the report appears to often be disregarded .... [W]e specifically discourage compete reliance on buzzwords contained in medical evaluations. Instead, the Division should look at the record as a whole and give due consideration to the actual content of medical reports to determine whether they support a finding of PTD and may, thus, indicate the onset thereof.
Lambert v. Workers’ Compensation Division, 211 W.Va. 436, 446, 566 S.E.2d 573 (2002). We again encourage the Division to look at a claimant’s record as a whole, and to resist a simplistic review based upon “magic words.”
IV.
CONCLUSION
For the reasons stated, the judgment of the Workers’ Compensation Appeal Board is reversed, and these cases are remanded for further proceedings consistent with this opinion.
Reversed and remanded.