Dunlavey v. Economy Fire & Casualty Co.

McGIVERIN, Chief Justice. This case presents the question of whether our workers’ compensation statutes, Iowa Code Chapter 85 (1993), permit an employee, Francis C. Dunlavey, to recover from his employer, Economy Fire and Casualty Company, for a nontraumatic mental injury caused by mental stimuli in the work environment. Because we conclude that the definition of the term “personal injuries” found in Iowa Code section 85.3(1) includes pure mental injuries, we agree with the district court and the industrial commissioner that an employee may recover for a nontraumatic mental injury. *847 We further conclude that the employee must satisfy two requirements. First, the employee must establish factual or medical causation; the employee must prove that the employee has a mental injury which was caused in fact by mental stimuli in the work environment. Second, the employee must meet the legal causation standard; the employee must prove that the mental injury was caused by workplace stress of greater magnitude than the day-to-day mental stresses experienced by other workers employed in the same or similar jobs, regardless of their employer. After viewing the record as a whole, we further agree with the district court that substantial evidence supports the commissioner’s decision that the employee in this case, Dunlavey, has satisfied the first of these two requirements. All of the doctors giving an opinion in this case agreed that Dunlavey’s work environment at Economy Fire and Casualty Company was the substantial or major cause of his depression. Accordingly, we affirm the district court’s judgment and the industrial commissioner’s decision regarding medical causation. However, we reverse the district court’s judgment and the industrial commissioner’s decision regarding the legal causation issue because they both applied an incorrect legal standard. Accordingly, we remand for application by the commissioner of the correct legal causation standard as described in this opinion. I. Background facts and proceedings. A. Employment and medical history. Francis C. Dunlavey (Dunlavey) is a sixty-two year old married man who resides in Mason City, Iowa and has a bachelors degree in business administration. He has been employed in the insurance industry for approximately thirty years and had been employed by Iowa Kemper Insurance Company (Iowa Kemper) from 1977 until March of 1986, when Iowa Kemper merged with Economy Fire and Casualty Company (Economy). Prior to the merger of Iowa Kemper and Economy, Dunlavey enjoyed his work as a claims adjuster for Iowa Kemper and denies having had any mental injuries or ever having seen a mental health professional. He claims he had always received good employment evaluations and periodically received increased responsibility and pay. After the announcement of the merger in 1985, however, Dunlavey insists his work environment grew increasingly more stressful, initially due to uncertainty about the continuation of his employment and later due to restructuring resulting from the merger of the two insurance companies. Dunlavey and one of his coworkers, Howard Anderson (Anderson), testified that the merger caused many changes in the work environment of former Iowa Kemper claims adjusters, including a revision of claims handling procedures, different managerial personnel, and an- increase in workload. According to Dunlavey, the most troublesome of these changes were the increase in workload and his new supervisors’ criticism of his performance. Dunlavey and Anderson testified that their supervisors gave them unrealistic demands and constantly downgraded them through unfavorable evaluations. Among other new responsibilities, Dunlavey was required to exercise more supervision over adjusters and to work on Economy files while converting pending Iowa Kemper files into Economy files. Dunlavey put in overtime in an effort to comply with these new requirements, generally working daily from six-thirty in the morning to six-thirty in the evening and several hours on the weekends. Despite this time investment, Dunlavey’s level of performance was evaluated only as marginally acceptable. Anderson testified that he also received unfavorable performance appraisals and was downgraded. He believes that such treatment was part of a process of intimidation and that Economy did not want him. Both Dunlavey and Anderson claim that former Iowa Kemper employees developed a fear that the new Economy management intended to get rid of them in one way or another. According to Anderson and Dunlavey, many claims examiners left because they felt they would be systematically eliminated. Dunla-vey testified that after the merger occurred he looked for other employment but was *848 unable to find any and, as a result, decided to remain with Economy until retirement. Dunlavey and Anderson claim that as part of the intimidation, the former Iowa Kemper employees had to perform more work than the personnel brought in by Economy and, by and large, worked more hours than the original Economy personnel. Both Dunlavey and Anderson testified that they could not feasibly perform all of the work which the Economy managers placed upon them without working many extra hours. They also testified that the stress the Economy managers placed upon them was greater than the stress they placed on other Economy employees. On the other hand, Denise Bollesen, an Economy claims examiner, Mary Joan Ward, Economy’s assistant claims manager, and Patrick Walters, Economy’s claims manager, all testified that Dunlavey’s workload was not unusual or out of the ordinary from that routinely carried by other claims examiners. Economy terminated Anderson in February 1987, just after he had completed converting his assigned Iowa Kemper files to the Economy system. By June 1987, only one claims examiner remained employed by Economy out of the five who had been employed by Iowa Kemper at the time of the merger. During that same year, Dunlavey’s wife, Lila, claims she noticed Dunlavey appeared depressed. Lila stated that her husband came home from work really tired and spoke of problems he had had at work. She stated that he would come home drained of any emotion and would not want to eat or do anything. Shortly after Lila noticed these changes in her husband, Dunlavey’s family, physician, Dr. James K. Coddington, M.D., diagnosed Dunlavey as having depression. Dr. Cod-dington wrote that “job stress is a definite causative factor.” At that time Dunlavey was taken off work and, except for approximately one week in the fall of 1987, has not since returned. Since Dr. Coddington’s diagnosis of depression, Dunlavey has been hospitalized, evaluated and treated extensively. Psychiatrists who have treated him include Ed Loon J. Chua, M.D., Ron M. Larsen, M.D., Bruce Pfohl, M.D., and Mary Hennessy, M.D. They uniformly agree that Dunlavey is afflicted with major depression. Doctors Cod-dington, Chua and Larsen have all expressed the opinion that Dunlavey’s employment stress was a causative or aggravating factor in the development of his depression. Dr. Pfohl stated that Dunlavey’s work stress is probably a substantial factor in causing Dun-lavey’s depression, but he declined to compare its significance to other known stresses in Dunlavey’s life. Dr. Hennessy, however, declined to express an opinion regarding the cause of Dunlavey’s mental injury. With regard to Dunlavey’s ability to resume employment, the doctors generally agreed that Dunlavey was not capable of returning to work at the time of the hearing. The doctors also generally agree that depression is usually a temporary condition and that most individuals who suffer from it recover and are able to return to work. However, they generally declined to express a specific prognosis for Dunlavey, since his depression has existed for an unusually long period of time. The doctors vacillated in their opinions, ranging from the opinion that Dunlavey was not capable of returning to any type of regular employment to the opinion that he could possibly try a low stress job, or that, with rehabilitation services, he could possibly resume employment. B. The present action. Dunlavey filed a petition for arbitration with the Iowa industrial commissioner against his employer Economy and its insurance carrier, Lumber-mens Mutual Casualty Company, alleging that he had sustained a psychological injury as a result of stress to which he had been exposed through his employment at Economy. See Iowa Code § 85.26(3). He sought compensation for permanent total disability and payment of medical benefits. Respondents denied petitioner’s allegations, insisting that without any physical injury Dunlavey had no basis for recovery under Iowa’s workers’ compensation statutes. After a hearing in this contested case, see id. section 86.14, the deputy commissioner found and concluded, and the industrial com *849 missioner agreed, that Dunlavey developed a depression as a result of stresses to which he was subjected through his employment at Economy and that those stresses exceeded the magnitude of the day-to-day stresses which must be endured by all employees. The deputy commissioner made these decisions without considering the deposition testimony and exhibits of respondents’ expert witness, Dr. Robert Smith. The deputy refused to admit Dr. Smith’s deposition and exhibits as a sanction for respondents’ failure to comply with Iowa rule of civil procedure 125(e) concerning timely identification of an expert and the proposed testimony. The industrial commissioner approved this sanction. The deputy also found, and again the industrial commissioner agreed, that Dunlavey has been totally disabled from June 17, 1987 to the present, except for the short period when he attempted to return to work, from performing work substantially similar to the work he performed at the time of his injury. Based on these findings, the deputy then ordered Economy to pay Dunlavey temporary total disability weekly compensation benefits and the unpaid balance of Dunla-vey’s medical expenses. The industrial commissioner affirmed, stating that the weekly compensation benefits should be paid pursuant to Iowa Code section 85.33(1). Respondents filed a petition for judicial review in district court, see Iowa Code § 86.26, and the court affirmed the industrial commissioner’s decision. Respondents then appealed, see id. § 17A.20, contending that the district court erred in (1) allowing petitioner to recover compensation based on a mental injury without any accompanying physical injury, (2) finding sufficient evidence that petitioner’s mental injury was caused both factually and legally by his employment at Economy, and (3) failing to find the commissioner erred in refusing to admit into evidence the deposition and exhibits of respondents’ medical expert. II. Scope of review and burden of proof. Our decision is largely controlled by the limited scope of review applicable to decisions made by an administrative agency. Kostelac v. Feldman’s, Inc., 497 N.W.2d 853, 856 (Iowa 1993); see Iowa Code § 17A.19(8). Our review is for correction of errors at law, not de novo, and the industrial commissioner’s findings have the effect of a jury verdict. Kostelac, 497 N.W.2d at 856. The burden rested upon petitioner Dunlavey to show by a preponderance of the evidence that the injury arose out of and in the course of his employment. See Anderson v. Oscar Mayer & Co., 217 N.W.2d 531, 535 (Iowa 1974). “[A] possibility is insufficient; a probability is necessary.” Id. The question is whether the industrial commissioner’s decision is supported by substantial evidence in the record made before the agency when that record is viewed as a whole. Schreckengast v. Hammermills, Inc., 369 N.W.2d 809, 811 (Iowa 1985) (citing Iowa Code § 17A.19(8)(f)). Evidence is substantial if a reasonable mind would accept it as adequate to reach a conclusion. John Deere Dubuque Works of Deere & Co. v. Weyant, 442 N.W.2d 101, 105 (Iowa 1989). The possibility of drawing inconsistent conclusions from the same evidence does not mean an agency’s decision lacks substantial support. Id. In the case of a conflict in the evidence, we are not free to interfere with the commissioner’s findings. Schreckengast, 369 N.W.2d at 811. We are not bound, however, by an agency’s erroneous conclusions of law. King v. City of Mount Pleasant, 474 N.W.2d 564, 565 (Iowa 1991); see also Iowa Code § 17A.19(8)(e). Following these principles, the main question we must address is whether a worker’s mental injury, allegedly caused by the work environment but not suffered in connection with a physical injury, is compensable under our workers’ compensation statutes. III. loica Code section 85.3(1) and interpretative case law. As workers’ compensation is based on Iowa Code chapter 85, our analysis begins with the applicable code section, section 85.3(1), and our prior case law interpreting that section. Iowa Code section 85.3(1) provides: *850 1. Every employer, not specifically excepted by the provisions of this chapter, shall provide, secure, and pay compensation according to the provisions of this chapter for any and all personal injuries sustained by an employee arising out of and in the course of the employment, and in such cases, the employer shall be relieved from other liability for recovery of damages or other compensation for such personal injury. (Emphasis added.) For the purposes of the issue presented by this case, whether Dunlavey may recover from Economy for his depression allegedly caused by nontraumatie mental stimuli at work, section 85.3(1) requires us to answer two preliminary questions: (1) whether Dun-lavey’s depression is a “personal injury” within the scope of section 85.3(1), and (2) whether Dunlavey’s depression “arose out of and in the course of [his] employment” at Economy. We have had occasion in several recent decisions to recognize these issues concerning the compensability of such “mental/mental” cases, characterized as such by Professor Larson in his treatise on workers’ compensation, see IB Arthur Larson, The Law of Workmen’s Compensation § 42.20, at 7-813 (1993). See Kostelac v. Feldman’s, Inc., 497 N.W.2d 853, 857 (Iowa 1993); Newman v. John Deere Ottumwa Works, 372 N.W.2d 199, 202-03 (Iowa 1985); Schreckengast v. Hammermills, Inc., 369 N.W.2d 809, 810-11 (Iowa 1985). In fact, in our most recent decision, Kostelac, we explicitly recognized that resolution of the issue of the compensa-bility of “mental/mental” injuries under our workers’ compensation statutes involves answering the two aforementioned questions regarding the scope of the statutes’ coverage. We stated: [W]e have not yet had occasion to rule whether mental injury, standing alone, will give rise to a claim for workers’ compensation. ... Nor have we squarely faced the issue of what degree of work related stress would justify an award of benefits in the event nontraumatie mental injury were deemed compensable. 497 N.W.2d at 857. In each of these cases, however, we resolved the appeal without deciding these issues. See id. at 857; Newman, 372 N.W.2d at 202-08; Schreckengast, 369 N.W.2d at 810-11. Now we are squarely presented with the compensability of a “mental/mental” injury, and we therefore analyze and answer each of these questions in turn. IV. Interpretation of “personal injuries” and compensability for “mental/mental” injuries. The first issue we must resolve is whether the phrase “personal injuries” in Iowa Code section 85.3(1) encompasses a mental injury standing alone. Respondents advance several arguments in support of their contention that a mental injury only qualifies as a “personal injury” if the mental injury is suffered in connection with a physical injury or a physical stimulus. After analyzing our past interpretations of the phrase “personal injuries,” the district court rejected respondents’ arguments and concluded that “personal injuries” included a pure nontrau-matic mental injury. We agree with both the district court’s analysis and its conclusion. The Iowa Code does not provide a complete definition of “personal injuries.” The Iowa Code’s definition of “personal injury” at section 85.61(4) merely provides: 4. The words “injury ” or “personal injury” shall be construed as follows: a. They shall include death resulting from personal injury. b. They shall not include a disease unless it shall result from the injury and they shall not include an occupational disease as defined in section 85A.8. In one of our earlier cases, we expanded on this incomplete definition of “personal injury” by stating: A personal injury, contemplated by the Workmen’s Compensation Law, obviously means an injury to the body, the impairment of health, or a disease, not excluded by the act, which comes about, not through the natural building up and tearing down of the human body, but because of a traumatic or other hurt or damage to the health or body of an employee.... The *851 injury to the human body here contemplated must be something, whether an accident or not, that acts extraneously to the natural processes of nature, and thereby impairs the health, overcomes, injures, interrupts, or destroys some function of the body, or otherwise damages or injures a part or all of the body. This is the personal injury contemplated by ... the ... Code.... Almquist v. Shenandoah Nurseries, Inc., 218 Iowa 724, 732, 254 N.W. 35, 39 (1934) (citations omitted). We believe a pure mental injury falls within this definition of “personal injury.” As the district court stated: The expansive concept of personal injury outlined in Almquist would, applying the language of the case, include a mental impairment produced by a mental stimulus. A major depressive disorder is clearly an impairment of health, and if caused by unusual mental stress is not the product of a natural process, but rather something which acts extraneously to damage mental health. Moreover, our more recent case law has explicitly “interpreted ‘personal injuries’ for purposes of workers’ compensation coverage to include mental conditions.” Mortimer v. Fruehauf Corp., 502 N.W.2d 12, 16 (Iowa 1993); see also Deaver v. Armstrong Rubber Co., 170 N.W.2d 455, 466 (Iowa 1969) (following the Indiana court’s interpretation of the word “injury” in the context of its workers’ compensation act and adopting its statement that the term “injury” “includ[es] mental ailments or nervous conditions”). Although our interpretation of “personal injury” in both Mortimer and Deaver occurred in the context of a mental injury accompanied by a physical injury, we view this distinction as irrelevant. In both cases we concluded that the term “personal injuries” as used in Iowa Code section 85.3(1) encompasses a mental injury as well as a physical injury. See Mortimer, 502 N.W.2d at 16-17; Deaver, 170 N.W.2d at 466; see also Honeywell v. Allen Drilling Co., 506 N.W.2d 434, 437 (Iowa 1993) (reiterating Mortimer’s statement that no logical reason exists to distinguish a mental injury from a physical injury because both result in a disability). Compare Bailey v. American Gen. Ins. Co., 154 Tex. 430, 279 S.W.2d 315, 318-19 (1955) (determining that a statute defining “injury” as “damage or harm to the physical structure of the body” included compensation for a pure mental injury). Our conclusion finds support in Professor Larson’s treatise where he urges that “there is no really valid distinction between physical and ‘nervous’ injury.” See Larson, § 42.23(a), at 7-906. As the district court stated: In the final analysis it is not possible to limit the concept of injury to a physical mechanism or stimulus and be consistent with the line of cases from Almquist to Mortimer or the principle of liberal interpretation and construction they reflect. It seems clear from the cases that the requirement of “injury” is intended to distinguish “hurt or damage to the health or the body” of an employee not the product of natural processes, which is compensable, from “the natural building up and tearing down of the human body” which is not. Almquist, 218 Iowa at'732, 254 N.W. at 39. There is no indication the legislature intended to imply a limitation to harm resulting from a physical stimulus. Based on the foregoing analysis, we conclude that the term “personal injuries,” as used in Iowa Code section 85.3(1), includes a mental injury standing alone. Having so determined, it naturally follows that an employee’s pure nontraumatic mental injury “arising out of and in the course of the employment” is compensable under chapter 85 of the Iowa Code. Accord Stokes v. First Nat’l Bank, 298 S.C. 13, 377 S.E.2d 922, 927 (Ct.App.1988), aff'd, 306 S.C. 46, 410 S.E.2d 248 (1991). As the district court summarized: [Ojnce all of the arguments are considered, and giving due deference to the commissioner’s views, the court believes the conclusion most consistent with the broad, inclusive language of the statute, the case law interpreting it, and its salutary purpose, is that mental injury resulting from a mental stimulus may in appropriate cases *852 be a compensable work related injury under chapter 85, Iowa Code. Several other jurisdictions facing the issue of the compensability of “mental/mental” injuries agree. At least fifteen state courts have permitted recovery for a mental injury caused solely by a mental stimulus under the state’s workers’ compensation laws. 1 See, e.g., Brock v. Industrial Comm’n, 15 Ariz.App. 95, 486 P.2d 207 (1971); Owens v. National Health Lab., Inc., 8 Ark.App. 92, 648 S.W.2d 829 (1983); Baker v. Workmen’s Compensation Appeals Bd., 18 Cal.App.3d 852, 96 Cal.Rptr. 279 (1971); Battista v. Chrysler Corp., 517 A.2d 295 (Del.Super.Ct.1986); Royal State Nat’l Ins. Co. v. Labor & Indus. Relations Appeal Bd., 53 Haw. 32, 487 P.2d 278 (1971); Pathfinder Co. v. Industrial Comm’n, 62 Ill.2d 556, 343 N.E.2d 913 (1976); Hansen v. Von Duprin, Inc., 496 N.E.2d 1348 (Ind.Ct.App.1986), rev’d on other grounds, 507 N.E.2d 573 (1987); Sargent v. Board of Educ., 49 Md.App. 577, 433 A.2d 1209 (Ct.Spec.App.1981); Br own & Root Constr. Co. v. Duckworth, 475 So.2d 813 (Miss.1985); Simon v. R.H.H. Steel Laundry, Inc., 25 N.J.Super. 50, 95 A.2d 446 (Hudson County Ct.), aff'd, 26 N.J.Super. 598, 98 A.2d 604 (Super.Ct.App.Div.), cer t. denied, 13 N.J. 392, 99 A.2d 859 (1953); Stokes v. First Nat’l Bank, 298 S.C. 13, 377 S.E.2d 922 (Ct.App.1988), aff'd, 306 S.C. 46, 410 S.E.2d 248 (1991); Jose v. Equifax, Inc., 556 S.W.2d 82 (Tenn.1977); Bailey v. American Gen. Ins. Co., 154 Tex. 430, 279 S.W.2d 315 (1955); Burlington Mills Corp. v. Hagood, 177 Va. 204, 13 S.E.2d 291 (1941); Consolidated Freightways v. Drake, 678 P.2d 874 (Wyo.1984). Also, through legislation, several other states have provided workers’ compensation for mental injuries arising solely from a mental stimulus. See, e.g., Alaska Stat. § 23.30.265(17) (1990); Colo.Rev.Stat.Ann. § 8-41-301(2)(a)-(d) (West Supp.1994); La. Rev.Stat.Ann. § 23:1021(7)(b) (West Supp. 1994); Me.Rev.Stat.Ann. tit. 39-A, § 201(3) (West Supp.1994); Mass.Gen.L. ch. 152, § 1(7A) (Supp.1994); N.M.Stat.Ann. § 52-1-24(A)-(C) (Michie 1978 & Supp.1994), (But see Douglass v. State, Regulation & Licensing Dep’t, 112 N.M. 183, 812 P.2d 1331 (Ct.App.), ce rt. denied, 112 N.M. 77, 811 P.2d 575 (1991) (holding stress-caused neurochemical depression was not compensable as a “primary mental impairment”).); N.Y. Workers’ Compensation Law § 2(7) (McKinney 1992); N.D.Cent.Code § 65-01-02(9)(a)(3) & (b)(9) (Supp.1993); Or.Rev.Stat. § 656.802(l)(b) & (3) (1993); R.I.Gen.Laws § 28-34-2(36) (1986); Wis.Stat. § 102.01(2)(c) (1988). On the other hand, other state courts have denied workers’ compensation for mental claims by interpreting their workers’ compensation statutes as not authorizing recovery for mental injuries based on job related stresses. See, e.g., Superior Mill Work v. Gabel, 89 So.2d 794 (Fla.1956); Hanson Buick Inc. v. Chatham, 163 Ga.App. 127, 292 S.E.2d 428 (1982); Followill v. Emerson Elec. Co., 234 Kan. 791, 674 P.2d 1050 (1984); Lockwood v. Independent Sch. Dist. No. 877, 312 N.W.2d 924 (Minn.1981); Sorensen v. City of Omaha, Pub. Safety/Fire Div., 230 Neb. 286, 430 N.W.2d 696 (1988); Wolf v. Northmont City Schs., 38 Ohio App.3d 118, 528 N.E.2d 589 (1987); Vernon v. Seven-Eleven Stores, 547 P.2d 1300 (Okla.1976); Lather v. Huron College, 413 N.W.2d 369 (S.D.1987). Similarly, some state legislatures have denied workers’ compensation for mental injuries unless the mental injury is suffered in connection with a physical injury. See, e.g., Ala.Code § 25-5-1(9) (1992); Conn.Gen.Stat. Ann. § 31-275(16)(B)(ii) & (iii) (West Supp. 1994); Idaho Code § 72-451 (1994); Ky.Rev. Stat.Ann. § 342.0011(1) (Michie/Bobbs-Mer-rill Supp.1994); Mont.Code Ann. § 39-71-105(5) (1993); N.H.Rev.Stat.Ann. § 281-A:2(XI) (Supp.1993); Wash.Rev.Code Ann. *853 § 51.08.142 (West 1990); W.Va.Code § 23-4-lf (1994). We have considered these other states’ resolution of the issue. To the extent other states have made a decision contrary to ours, they do not affect our determination here. In each case, the state’s workers’ compensation statutes, either implicitly, through a more detailed definition of “personal injury,” or explicitly, by a specific provision on the matter, provide a basis for the state to deny recovery for “mental/mental” claims. Lacking such legislative guidance, we join the majority of states with workers’ compensation statutes similar to ours in allowing workers to recover for mental injuries caused by unusual stress in the work environment. V. Medical and legal causation. Having concluded that an employee may recover under Iowa Code section 85.3(1) for a pure nontraumatic mental injury, we now turn to the second requirement for recovery contained in section 85.3(1), whether the injury “ar[ose] out of and in the course of the employment.” We have previously divided this causation requirement into two separate determinations: (1) factual or medical causation and (2) legal causation. See Newman, 372 N.W.2d at 202-03; Schreckengast, 369 N.W.2d at 810-11 & n. 3. Causation in fact involves whether a particular event in fact caused certain consequences to occur. Schreckengast, 369 N.W.2d at 810, n. 3. In the context of this workers’ compensation case, factual causation means medical causation, that is whether the employee’s injury is causally connected to the employee’s employment. See id. at 810. Distinct from medical causation, we must also consider the legal causation standard. Legal causation presents a question of whether the policy of the law will extend responsibility to those consequences which have in fact been produced. Id. at 810, n. 3 (citing State v. Marti, 290 N.W.2d 570, 584-85 (Iowa 1980)). Factual or medical causation presents an issue of fact while the standard of legal causation presents an issue of law. See id. A. Medical causation. Respondents’ contend that when the record is viewed as a whole no credible substantial evidence exists to support a finding of causal connection between Dunlavey’s employment and his claimed depressive condition. Specifically, respondents argue that certain medical and psychological experts testifying in this matter did not have full, objective knowledge of other potential stressors in Dunlavey’s life, such as Dunlavey’s wife’s recent medical problems, his adopted children’s substance abuse and learning disability, and Dunlavey’s recent awareness of being adopted. Respondents further challenge the experts’ opinions as based solely on what Dunlavey, with his mental disturbances, told them. On the basis of the experts’ opinion testimony, however, the deputy commissioner found as a fact that Dunlavey “developed depression as a result of stresses to which he was subjected in his employment with Economy,” and the industrial commissioner agreed. The district court concluded the evidence supported that finding. We agree. Basically, respondents’ complaints go to the weight and credibility of the experts’ testimony. Whether an injury has a direct causal connection with the employment or arose independently thereof is essentially within the domain of expert testimony. Deaver, 170 N.W.2d at 464 (citing Bradshaw v. Iowa Methodist Hosp., 251 Iowa 375, 383, 101 N.W.2d 167, -171-72 (I960)). The weight to be given such an opinion is for the finder of fact, in this case the commissioner, and that may be affected by the completeness of the premise given the expert and other surrounding circumstances. See id. (citing Bodish v. Fischer, Inc., 257 Iowa 516, 521-22, 133 N.W.2d 867, 870 (1965)). When an expert’s opinion is based upon an incomplete history, the opinion is not necessarily binding upon the commissioner. Id. The commissioner as trier of fact has the duty to determine the credibility of the witnesses and to weigh the evidence, together with the other disclosed facts and circumstances, and then to accept or reject the opinion. Id. (citing Gwynne v. Vance, 258 Iowa 875, 879, 140 N.W.2d 917, 919 (1966)). *854 Our review of such a determination by the commissioner is limited to whether the commissioner’s finding is supported by substantial evidence in the record made before the agency when that record is viewed as a whole. Iowa Code § 17A.19(8)(f); Schreckengast, 369 N.W.2d at 811. We believe substantial evidence exists in this case to support the commissioner’s findings and conclusion. Respondents do not dispute, and the medical testimony was uniform, that Dunlavey suffered from a severe depression. Moreover, the experts’ testimony, given by Dunla-vey’s family physician, Dr. Coddington, and four psychiatrists who examined or treated him at various times, is substantial that Dun-lavey’s work was the substantial or major cause of his depression. The district court summarized the experts’ testimony as follows: —Dr. Coddington testified there was a “significant causal relationship” and that work factors were a “substantial factor” in Dunlavey’s depression in the sense they were more than half of the total cause. —Dr. Chua, a Mayo Clinic psychiatrist who treated Dunlavey in connection with his first two hospitalizations in June and August 1987, testified the most significant cause of Dunlavey’s depression was job-related stress. —Dr. Larsen, a Mason City psychiatrist who was involved with Dunlavey in connection with a November 1987 hospitalization, felt his depression was caused by a combination of stress and biochemistry, the latter precipitating the job stress. Dr. Larsen could not state the cause with certainty, but agreed with the suggestion that job-related stress was “a significant or predominant or major contributor” to Dunlavey’s depression. —Dr. Hennessy, another Mason City psychiatrist who examined Dunlavey in November 1988 and who has since managed his medications, did not feel she was close enough to the onset of depression to state an opinion to its cause but indicated that she would defer to Dr. Chua’s opinion. —Finally, Dr. Bruce Pfohl, a University of Iowa psychiatrist with clinical, teaching and research responsibilities, felt job-related stress was a possible causal factor. He noted that, in analyzing the impact of stressors, the things “temporally close to onset” [i.e. the merger and change in working conditions] are the most critical. Cf. Patnode v. Rome Dev. Ctr., 150 A.D.2d 868, 540 N.Y.S.2d 899, 901 (App.Div.1989) (stating long gap between 1983 alleged interrogation by supervisor and 1985 alleged anxiety attacks supported board’s conclusion that claimant’s disorder was not caused by work experience). Dr. Pfohl was reluctant to express an opinion because of incomplete information about Dunlavey’s work. However, later in his testimony he referred to job stress as a “major role” or “significant factor” in Dunlavey’s depression. We believe such testimony supports the district court’s conclusion that “[t]hough there were other stressors in Dunlavey’s life outside his employment, the medical testimony in general was that these were either less significant or a symptom rather than a cause of his depression.” We also conclude that such testimony reflects substantial evidence in the record as a whole that Dunlavey’s work at Economy caused his depression and therefore do not disturb the commissioner’s outcome determinative finding on this issue. See Schreckengast, 369 N.W.2d at 811. B. Legal causation. Because we have found that Dunlavey met his burden of proving medical causation by substantial evidence, we next are confronted with the task of establishing a standard for legal causation. On previous occasions we have recognized the potential issue of what degree of work related stress would justify an award of benefits if a worker’s pure nontraumatic mental injury were deemed compensable but, until today, had not been squarely faced with the issue. See Kostelac, 497 N.W.2d at 857; Schreckengast, 369 N.W.2d at 811. Petitioner Dunlavey urges us to adopt the so-called “Wisconsin Standard” of legal causation which the industrial commissioner applied in this and previous agency cases. See School Dist. # 1 v. Department of Indus., Labor & Human Relations, 62 Wis.2d 370, 215 N.W.2d 373, 377 (1974). The Wisconsin standard provides that a “mental injury non-traumatieally caused must have resulted *855 from a situation of greater dimensions than the day-to-day emotional strain and tension which all employees must experience.” Id. (emphasis added). On the other hand, respondents contend that the “Wisconsin Standard” cannot be applied effectively or consistently. Upon review of other states’ opinions and statutes considering the matter, we adopt an “unusual stress” standard but decline to follow the Wisconsin supreme court’s characterization of it. Rather, we adopt the “unusual stress” standard as formulated by the Wyoming supreme court and thus conclude that a “mental/mental” injury is com-pensable under Iowa Code section 85.3(1) if, after proving medical causation, an employee establishes that the mental injury “was caused by workplace stress of greater magnitude than the day-to-day mental stresses experienced by other ivorkers employed in the same or similar jobs,” regardless of their employer. See Graves v. Utah Power & Light Co., 713 P.2d 187, 193 (Wyo.1986) (emphasis added). 1. In the states which have permitted “mental/mental” claims, the state courts have developed primarily three different tests for legal causation. 2 At least one state court, the Michigan court, ruled that compensation should be awarded even if there is only a strictly subjective causal nexus between the trauma and the mental injury — the so-called “subjective causal nexus” standard. Deziel v. Difco Lab., Inc., 403 Mich. 1, 268 N.W.2d 1, 11 (1978). This subjective test focuses on the employee’s own perception of reality in that if the claimant honestly perceives that there is some work related stimuli causing a mental disability the employee is entitled to recover. Deziel, 268 N.W.2d at 11. Subsequently, however, the Michigan legislature enacted a statute repudiating the Deziel court’s subjective standard. See Mich.Comp. Laws Ann. § 418.301(2) (West 1985) (“Mental disabilities shall be compensable when arising out of actual events of employment, not unfounded perceptions thereof.”); Gardner v. Van Buren Pub. Sch., 445 Mich. 23, 517 N.W.2d 1 (1994) (recognizing the Deziel opinion had been superseded by statute). No other known states have adopted this subjective standard. Showing a different view, several states recognizing compensability of nervous injuries caused by mental stimuli have adopted a “broad causation” test. See, e.g., Albertson’s, Inc. v. Workers’ Comp. Appeals Bd., 131 Cal.App.3d 308, 182 Cal.Rptr. 304 (1982); Royal State Nat’l Ins. Co. v. Labor & Indus. Relations Appeal Bd., 53 Hawaii 32, 487 P.2d 278 (1971); Hansen v. Von Duprin, Inc., 507 N.E.2d 573, 575-76 (Ind.1987). The “broad causation” test allows a worker to be compensated for normal work related stress as long as the worker’s mental injury is causally connected to the employment. See, e.g., Von Duprin, 507 N.E.2d at 576. “This approach is premised on the view that the basic purpose of a workers’ compensation system mandates that a worker disabled as a result of work related stress receive treatment identical to a worker disabled by a work related physical injury.” Candelaria v. General Elec. Co., 105 N.M. 167, 730 P.2d 470, 477 (App.), cert. quashed, 105 N.M. 111, 729 P.2d 1365 (1986). A third type of causation standard is the “Wisconsin Standard.” The “Wisconsin Standard” requires the employee to show that the work stress which led to the injury was unusual or greater than that endured by all other workers. See Swiss Colony, Inc. v. Department of Indus., Labor & Human Relations, 72 Wis.2d 46, 240 N.W.2d 128, 130 (1976). The Wisconsin court developed this standard as a response to the difficulties in the evaluation of psychological injuries, such as the ease with which such claims may be feigned and the difficulty with which fraudulent claims can be detected. See School Dist. # 1, 215 N.W.2d at 376-77. Several states have agreed with the Wisconsin court’s logic *856 and have chosen this type of test over the “subjective causal nexus” test and the “broad causation” test. See Larson, § 42.25(f), at 7-966. See, e.g., McClain v. Texaco, Inc., 29 Ark.App. 218, 780 S.W.2d 34 (1989); City of Springfield v. Industrial Comm’n, 214 Ill.App.3d 301, 158 Ill.Dec. 23, 573 N.E.2d 836 (1991); Stokes v. First Nat’l Bank, 298 S.C. 13, 377 S.E.2d 922 (Ct.App.1988), aff'd, 306 S.C. 46, 410 S.E.2d 248 (1991). 2. Like these states, we agree with the Wisconsin court that an “unusual stress” standard is the preferable standard of legal causation for “mental/mental” claims. We have already impliedly rejected the “subjective causal nexus” standard in Newman v. John Deere Ottumwa Works, 372 N.W.2d 199, 203 (Iowa 1985), where we refused to “permit recovery when employment merely provided- a stage for the nervous injury.” Moreover, neither the “subjective causal nexus” test nor the “broad causation” test correspond to our statutory goal of making sure the mental injuries are causally connected to the worker’s employment. We believe an “unusual stress” standard will better achieve that statutory goal. As the district court stated: [Sjueh a standard assures the injury is appropriately tethered to the employee’s employment. If an employee cannot endure the ordinary stress of employment, it cannot really be said his or her mental condition arises out of “the employment” as opposed to general inability to cope with the stress of everyday life. See Iowa Code § 85.3(1)_ Further, in many' cases there will be other contributing stressors not related to the workplace and/or preexisting mental health problems. In such cases an unusual stress requirement helps insure the requisite causal connection to employment. Additionally, as recognized by the Wisconsin court, the standard will help to prevent the evils respondents contend will occur from our determination that “mental/mental” claims are compensable under Iowa Code chapter 85. See Sara J. Sersland, “Mental Disability Caused by Mental Stress: Standards of Proof in Workers’ Compensation Cases,” 33 Drake L.Rev. 751, 807 (1983-1984). Among other arguments, respondents assert that nondetectable fraudulent claims, conversion of workers’ compensation into general mental health insurance, chilling effects on Iowa businesses, and enormous direct and indirect costs to Iowa businesses will result from the compensation of “mental/mental” claims. At the heart of these arguments is the “slippery slope” argument that once “mental/mental” claims are deemed compensable, employees will increasingly make fraudulent claims which the courts will not be able to detect and which will ultimately force employers out of business. Although we recognize respondents’ arguments, we agree with both the Wisconsin court and the district court that an “unusual stress” standard, which places the burden on the employee to prove that the work stress suffered is unusual, is an effective means of evaluating employees’ claims. See School Dist. # 1, 215 N.W.2d at 377. At the same time, an “unusual stress” standard furthers the intent of the legislature to compensate those workers who suffer bona fide “personal injuries” caused by their employment. A more specific response to respondents’ public policy arguments must come from a contrary expression of intent by the legislature, which is the most appropriate forum to address the stated concerns. Also, the industrial commissioner’s past consistent use of an “unusual stress” standard persuades us to adopt such a standard as the legal causation standard for “mental/mental” claims. “Although final interpretation and construction of the statute is for this court, we give deference to an interpretation by the responsible administrative agency.” Mortimer, 502 N.W.2d at 14. Finally, as discussed earlier, several other jurisdictions allowing compensation for “mental/mental” claims support our decision as to the appropriateness of an “unusual stress” standard. See Larson, § 42.25(f), at 7-966. 3 *857 3. However, although we agree with the petitioner that an objective, “unusual stress” standard is the appropriate standard, we decline to follow the petitioner’s suggestion to adopt the Wisconsin supreme court’s formulation of the standard. Rather, we adopt the Wyoming supreme court’s formulation of the “unusual stress” standard. See Graves v. Utah Power & Light Co., 713 P.2d 187, 193 (Wyo.1986). Specifically, we hold that in order for an employee to establish legal causation for a nontraumatic mental injury caused only by mental stimuli, the employee must show that the the mental injury “was caused by workplace stress of greater magnitude than the day-to-day mental stresses experienced by other workers employed in the same or similar jobs,” regardless of their employer. See id. Three reasons exist for our decision to follow the Wyoming supreme court’s formulation of the “unusual stress” causation standard. First, the Wyoming standard strikes a balance between the overall purposes embodied in our workers’ compensation statutes. By comparing the stresses endured by similarly situated employees, the Wyoming standard provides the employees with compensation for legitimate work related injuries while at the same time limits the employers’ liability to injuries caused by its industry. As the Wyoming supreme court insightfully noted: It might be considered modern and forward looking for workers’ compensation to be expanded to provide full coverage health insurance. But that was not its intended purpose, and it would be unfair to the employer who pays premiums on the assumption that they cover only injuries caused by [its] industry. Id. at 190. Second, the Wyoming standard avoids any confusion over the meaning and the application of the phrase “all employees” embodied in the Wisconsin standard. The Wyoming court has described the possibility for confusion as being threefold: The objective “all employees” standard could be based upon three different groups. First, “all employees” could consist of a worker’s “fellow employees” employed in the same or similar jobs by the same employer. Second, it could consist of workers in the same or similar jobs, including those who work for .other employers. Finally, it could consist of the “working world at large.” Id. at 192. Indeed, states which have adopted a generic “unusual stress in comparison to all employees” standard have developed varying formulations for its application. Some states, including Wisconsin, have characterized the level of stress which a worker must prove as stress which is greater than the stress that all employees in the “working world at large”, must experience. See Probst v. Labor & Indus. Review Comm’n, 153 Wis.2d 185, 450 N.W.2d 478, 480 (1989) (emphasis added). See also Caron v. Maine Sch. Admin. Dist. No. 27, 594 A.2d 560, 562-63 (Me.1991) (statute with the language “extraordinary and unusual [work stress] in comparison to pressures and tensions experienced by the average employee” did not require the commission to limit its comparison to those performing a similar job with the same duties). Other states initially following the generic Wisconsin standard, including Wyoming, later characterized the work stress which a worker must prove as stress which is greater than the work stress experienced by similarly situated employees. See Graves, 713 P.2d at 193 (emphasis added); see also McClain v. Texaco, Inc., 29 Ark.App. 218, 220-21, 224, 780 S.W.2d 34, 35, 37 (1989) (interpreting its “all workers” causation standard and stating that “the ultimate test is whether the stress constitutes an abnormal working condition for [the pertinent] type of employment”). In choosing its “similarly situated” formulation of the “unusual stress” standard, the Graves court reasoned: Unlike the “fellow employees” test, this test could be used in cases where the worker has no fellow employee holding the same job in his company. Moreover, under the same or similar job standard, an employer who puts excessive stress on several employees could not avoid the payment of benefits by simply making that excessive stress equal for all employees. *858 The stress on [its] workers would be compared to the stress suffered by those holding similar jobs in other companies. The objective test based on workers with the same or similar jobs is also superior to a test based on the working world at large. It is impossible to determine, except in the broadest fashion, the stress to which the working world at large is exposed. In every worker’s compensation ease heard under this test, the parties could call witnesses whose job related stress is either significantly greater or significantly smaller than the stress suffered by the worker seeking compensation. The standard would be too amorphous to be practical. 713 P.2d at 193. We find this reasoning persuasive. Finally, those states which have had experience with “mental/mental” claims are increasingly enacting statutes which require proof that the employee’s stress is greater than that of similarly situated employees. See, e.g., Alaska Stat. § 23.30.265(17)(A) (1990) (work stress must be “extraordinary and unusual in comparison to pressures and tensions experienced by individuals in a comparable work environment”); Colo.Rev.Stat. Ann. § 8-41-301(2)(a)-(d) (West Supp.1994) (the psychologically traumatic event must be “generally outside of a worker’s usual experience and would evoke significant symptoms of distress in a worker in similar circumstances”); N.M.Stat.Aim. § 52-l-24(A) (Mi-chie 1978 & Supp.1994) (same). For all these reasons, we adopt an objective standard of legal causation and place the burden on the employee to establish that the mental injury was caused by workplace stress of greater magnitude than the day-today mental stresses experienced by other workers employed in the same or similar jobs, regardless of their employer. Although evidence of workers with similar jobs employed by a different employer is relevant, evidence of the stresses of other workers employed by the same employer with the same or similar jobs will usually be most persuasive and determinative on the issue. Accord Graves, 713 P.2d at 193, n. 5. Having determined as a matter of law that the legal causation standard involves comparison of the stress endured by “similarly situated employees,” rather than that endured by “all employees,” we must reverse the district court’s judgment and the industrial commissioner’s decision regarding the legal causation standard used by them. In both instances, they applied the generic Wisconsin “all employees” standard rather than the Wyoming “similarly situated employees” standard of legal causation we have adopted in this opinion. Accordingly, we reverse and remand the case to the industrial commissioner for reconsideration of petitioner Dun-lavey’s claim for benefits under the existing record in light of the “similarly situated employees” standard of legal causation we have now adopted. VI. Exclusion of respondents’ expert’s deposition and accompanying exhibits. Respondents’ final argument in this appeal is that the deputy commissioner’s decision to exclude the deposition testimony and exhibits of respondents’ only expert witness, Dr. Robert Smith, at the arbitration hearing for noncompliance with Iowa rule of civil procedure 125(c) was an abuse of discretion. See Iowa Code § 17A.19(8)(g); Milks v. Iowa Oto-Head & Neck Specialists, P.C., 519 N.W.2d 801, 805 (Iowa 1994). The commissioner and the district court affirmed the deputy’s decision to exclude Dr. Smith’s deposition and exhibits. We find no abuse of discretion by the commissioner. Discovery procedures applied in civil cases are available in contested cases before an agency. Iowa Code § 17A.13(1); Miller v. Lauridsen Foods, Inc., 525 N.W.2d 417, 420 (Iowa 1994). Rule 125(c) is a discovery procedure which requires supplementation of responses regarding identity of experts and other information not less than thirty days prior to hearing. By the pre-hearing assignment order in this case, the parties were admonished by the commissioner that the service of witness lists did not relieve the parties from meeting the demands of rule 125(c). Furthermore, industrial commissioner rule 343-4.35(86) made rule 125 applicable *859 to this case. See Iowa Admin.Code 343— 4.35(86). The hearing before the deputy was scheduled for July 31, 1989. On June 27, 1989, respondents supplemented their answers to interrogatories to identify Dr. Smith, but they failed to supply Dr. Smith’s opinions, conclusions or the factual basis of same. In fact, respondents’ answer stated that no evaluation had been received. A couple of weeks later, respondents attempted to schedule Dr. Smith’s deposition. Upon application by petitioner Dunlavey, the deputy found that to proceed with Dr. Smith’s deposition would be prejudicial to Dunlavey and granted a protective order protecting Dunlavey from necessity to appear at the deposition. The deputy also ordered that any such deposition of Dr. Smith would be excluded from evidence at the hearing. We agree with the commissioner and the district court that rule 125(c) was applicable to this case, see Iowa Code § 17A.13(1); Iowa Admin.Code 343-4.35(86), and that Economy failed to comply with both the rule and the hearing assignment order by failing to provide the required information thirty days prior to the hearing. Furthermore, as rule 125(c) explicitly allows the court to exclude expert testimony as a sanction for noneomplianee with the rule and industrial commissioner rule 343-4.36(86) permits sanctions for failure to comply with agency rules or orders, see Iowa Admin.Code 343-4.36(86), we agree with the commissioner and the district court that the sanction of exclusion of Dr. Smith’s deposition and exhibits was appropriate for respondents’ noncompliance. 4 No abuse of discretion occurred. Cf. Stephenson v. Furnas Elec. Co., 522 N.W.2d 828, 831-32 (Iowa 1994) (deputy excluding expert from testifying at hearing for noncompliance with rule 125(c)’s “as soon as practicable” requirement was not an abuse of discretion). VII. Conclusion. We have considered all the matters urged by respondents on this appeal- Based on this consideration, we conclude that mental injuries caused by work related stress are compensable under Iowa Code section 85.3(1) if, after demonstrating medical causation, the employee shows that the mental injury was caused by workplace stress of greater magnitude than the day-today mental stresses experienced by other workers employed in the same or similar jobs, regardless of their employer. Thus, we affirm the district court judgment in part and reverse it in part, and remand for further proceedings before the industrial commissioner in accordance with this opinion. JUDGMENT OF DISTRICT COURT AFFIRMED IN PART AND REVERSED IN PART; REMANDED. Footnotes [1] Some states allowing compensability of "mental/mental” claims make a distinction between nontraumatic or gradual, and traumatic or sudden mental stimuli, allowing recovery only for injuries caused by the latter. See, e.g., Ayer v. Industrial Comm'n, 23 Ariz.App. 163, 531 P.2d 208, 211 (1975); Jose v. Equifax, Inc., 556 S.W.2d 82, 84 (Tenn.1977). Because we agree with Professor Larson that "the distinction is unsound,” see Larson, section 42.23(b), at 7-923, especially in light of our definition of “personal injuries," and because we have previously allowed workers' compensation for gradual injuries, see McKeever Custom Cabinets v. Smith, 379 N.W.2d 368 (Iowa 1985), we decline to adopt such an artificial limitation on recovery. [2] Specific formulations of these general standards have been developed for determining the compensability of claims brought by workers who have had pre-existing mental conditions. See, e.g., Townsend v. Maine Bureau of Pub. Safety, 404 A.2d 1014, 1020 (Me.1979); McGarrah v. State Accident Ins. Fund Corp., 296 Or. 145, 675 P.2d 159, 172 (1983). As petitioner Dunlavey did not have a pre-existing mental condition, we do not discuss or evaluate these various specific standards. [3] We have previously recognized an "unusually strenuous” causation standard in the context of determining workers' compensation for heart claims. See Briar Cliff College v. Campolo, 360 N.W.2d 91, 94-95 (Iowa 1984). [4] Our conclusion is further supported by the fact that respondents assert no reason why the interrogatory could not have been timely supplemented. Also, the industrial commissioner’s later finding on appeal stated that, even if the opinion of Dr. Smith, previously excluded, was considered, there was overwhelming medical evidence to support a finding that Dunlavey’s work caused his depression.