The employee appeals from the compensation judge’s determination that the employee’s depression was not causally related to his work injury.  We affirm.

On January 24 and April 23, 2002, Thomas Penn [the employee] sustained rotator cuff tears to both of his shoulders in the course of his work as a journeyman plumber for NewMech Companies, [the employer].  The employee was fifty-four years old at the time of his injuries and was earning a weekly wage of $1,152.40.  The injuries each eventually required two surgical repairs - - on May 29 and September 17, 2002, for the left shoulder and on February 27 and June 12, 2003, for the right shoulder.  The employer and its workers’ compensation insurer admitted liability for the injuries and paid benefits, including compensation for a 6% whole-body impairment to each shoulder.

The employee underwent a three-day, eighteen-hour Functional Capacity Evaluation [FCE] at Saunders Therapy Centers on November 24, November 26, and December 3, 2003.  On the basis of the FCE, the employee was limited to intermittent reaching and essentially restricted from doing any lifting above his shoulders.  He was also limited to a maximum of ten pounds carrying and ten pounds lifting on a frequent basis.  On December 18, 2003, the employee and his Qualified Rehabilitation Consultant [QRC], Brett Krumhardt, met with the employee’s surgeon, Dr. Richard Strand, and discussed the employee’s FCE in detail.  Dr. Strand noted that the employee was still complaining of “bitter” pain in his shoulders, and he modified the FCE recommendations to restrict the employee to fairly limited duties.

About six weeks later, on February 3, 2004, the employee returned to Dr. Strand for a recheck of his rotator cuff repairs.  The employee reported that he was still having some pain in his shoulders and that it had gotten worse, particularly in his right shoulder, after his FCE.  The doctor noted that the right rotator cuff seemed to be working well on examination, but, noting tenderness along the anterior acromian, he recommended a shot of cortisone.  Dr. Strand supposed that the employee would need a job “that is significantly less [physically demanding] th[a]n his normal plumbing job,” based on the FCE.  The employee saw Dr. Strand in follow-up on March 16, 2004, on which date Dr. Strand noted that the employee still had subjective complaints in both of his shoulders but that he had excellent range of motion and his rotator cuffs were working.  The doctor advised the employee that there was no need for any further surgery and that he should return to work within the limits of his FCE.

About two months later, in about May of 2004, the employer offered the employee a position as estimator for plumbing operations.  The job evidently offered the same salary and benefit package as had the position held by the employee on the dates of his injuries.  A written job description was provided to Dr. Strand by the QRC, along with the QRC’s assessment of the job’s physical requirements.  The job was described as sedentary desk work, with some minimal computer work, no lifting over ten pounds, and no overhead reaching.  Dr. Strand approved the job on May 11, 2004, and the employee began the position two days later.

After the employee had worked at the estimator position for about two months, QRC Krumhardt prepared a closure report.  In his report, dated July 19, 2004, covering his activities from May 19 through July 19, 2004, the QRC reported that the employee hand remained in the estimator position since May “without significant incident.”  He noted that the position did involve some on-the-job training in computer work and estimating and that the “[employee’s] transferable skills as a Plumber are considered excellent background and experience for this position.”  On July 15, 2004, the QRC met with Tony Condon, the employer’s Safety Director, and, upon observing the employee’s work area and office, suggested only some minor ergonomic changes for the employee’s comfort.  At that time, Mr. Condon reported that the employee was performing his new duties very well.

On July 22, 2004, the employee returned to see Dr. Strand and reported that he was working at a computer, was finding it very stressful, and was not sure he could continue doing it.  With regard to his shoulders, the employee reported that they were feeling pretty good, although “the stress really bothers him.”  He reported some achiness in both shoulders and a little clicking in the right shoulder.  On examination, Dr. Strand noted that the employee’s rotator cuffs were working very well and that the employee had good range of motion.  No further treatment was deemed necessary, and Dr. Strand recommended that the employee’s permanent restrictions remain the same.

By August of 2004, the employee apparently began to feel overwhelmed by the demands of his new job and his work performance evidently fell off.  On September 14, 2004, concerned about his poor performance, the employer gave the employee an oral and written performance review that was negative in almost every current aspect.  The following day, the employee left the estimator job and sought treatment for depression.  The employee was seen by Dr. Kent Bratten at Park Nicollet Clinic on September 23, 2004, and was diagnosed with major depression.  Dr. Bratten prescribed antidepressants, recommended an eight week leave of absence from work, and referred the employee to psychiatrist Dr. Lon Augdahl.

The employee first saw Dr. Augdahl on September 27, 2004, when his chief complaint was identified as the “emergence of depressive symptoms secondary to employment difficulties.”  In his report, Dr. Augdahl noted that the employee

On October 6, 2004, the employee filed a claim petition, seeking payment of temporary total disability benefits continuing from September 15, 2004, as a result of his bilateral rotator cuff tears and a consequential major depressive disorder.  The employer and insurer denied liability for the alleged consequential major depressive disorder and for the claimed temporary total disability benefits.

On January 28, 2005, the employee was examined at the request of the employer and insurer by psychiatrist Dr. Thomas Gratzer.  Dr. Gratzer reviewed the employee’s medical records, obtained a Minnesota Multiphasic Personality Inventory-2 (MMPI-2), and conducted a psychiatric evaluation.  In a report dated February 17, 2005, Dr. Gratzer diagnosed adjustment disorder with mixed anxiety and depressed mood, resolving.  He also opined that the employee had preexisting psychiatric conditions showing underlying characterological issues, which he diagnosed as mixed personality traits.  Dr. Gratzer opined that the employee did not develop any psychiatric symptoms consequent to his January 24 and April 23, 2002, injuries.  He concluded that the employee’s psychological problems stemmed “from psychosocial stressors principally from his inability to perform his computer job effectively.  This is separate from physical stresses of the January 24, 2002 and April 23, 2002 injuries.”  Dr. Gratzer did not believe that the employee’s adjustment disorder was related to the work injuries or that the employee was temporarily totally disabled as claimed.  He also concluded that the employee had reached maximum medical improvement with respect to his adjustment disorder.  Dr. Gratzer stated:

Dr. Augdahl testified by deposition on October 24, 2005.  As of that date, he continued to be of the opinion that the employee was totally disabled from all employment as a result of his psychological condition.  Dr. Augdahl testified that he had noted some improvement in the employee’s condition over the course of the employee’s treatment with him but that it was not enough to allow a return to gainful employment.  In fact, in the two visits immediately preceding his deposition, Dr. Augdahl had noted an increase in the employee’s anxiety that was likely related to the upcoming court date and ongoing concerns about legal issues and work status.  While he continued to diagnose an adjustment disorder with mixed emotional features, anxiety, and depression, Dr. Augdahl indicated that he believed the employee’s psychological condition to be evolving more into a major depressive disorder.

With regard to issues of causation, Dr. Augdahl testified as follows:

The employee’s claim for benefits came on for hearing before a compensation judge on October 26, 2005.  At hearing, the employee testified that at the time he was offered the job as an estimator his shoulders hurt constantly and he was having difficulty sleeping due to his pain.  He testified that he was apprehensive about his ability to perform the job from the very beginning, in that he had never worked in an office before and had no computer skills, that, while he initially received positive feedback from the employer about his performance on the job, he was really not sure what was expected of him and did not feel that he was provided with adequate training.  He testified further that the stress of the transition into this new job affected his already impaired sleep pattern until he was only sleeping two or three hours per night and that, by late July or early August, his work performance suffered because he was overwhelmed by the demands of the job.  He testified that he began to feel that he was failing in all areas of his job and that he found the eventual bad performance review devastating and humiliating.

The employee’s wife, Barbara, also testified at hearing, in part that she was not aware of significant stress in the employee’s life prior to his shoulder injuries.  She testified that the employee was apprehensive about the estimator job and was very anxious about working in an office situation, which he had never done before.  She testified that the employee had initially seemed to enjoy having a job to go to and to be learning new skills but that by the second or third week of July everything started to “cave in” and his demeanor began to grow progressively worse, as the job seemed to overwhelm him.  Testimony was also offered by the employee’s supervisor, Estimating Manager David Secker, who testified that the employee’s performance at the time of his performance review was indeed not acceptable.  Mr. Secker testified that he oversees seven or eight estimators, who come to him from college training, from the field experience, and from other companies.  Mr. Secker testified that he puts a premium on people with field experience because they can read and understand reports and because their skills are easily transferable into the estimator position.  He testified that he had confidence in the employee’s expertise and actually used bids generated by the employee.  Mr. Secker testified that he provided regular contact, support, teaching, and reassurance to the employee.

By findings and order issued February 1, 2006, the compensation judge determined that the work injuries of January 24 and April 23, 2002, were not a substantial contributing factor to the development of the employee’s major depressive disorder or anxiety disorder with depressive mood.  She concluded that the employee’s psychological condition was not a result of his pain complaints or loss of ability to function but of anxiety related to commencing the new position as an estimator.  She therefore denied the employee’s claim for temporary total disability benefits and for medical expenses related to the employee’s psychological condition.  The employee appeals.

In reviewing cases on appeal, the Workers’ Compensation Court of Appeals must determine whether “the findings of fact and order [are] clearly erroneous and unsupported by substantial evidence in view of the entire record as submitted.”  Minn. Stat. § 176.421, subd. 1 (1992).  Substantial evidence supports the findings if, in the context of the entire record, “they are supported by evidence that a reasonable mind might accept as adequate.”  Hengemuhle v. Long Prairie Jaycees, 358 N.W.2d 54, 59, 37 W.C.D. 235, 239 (Minn. 1984).  Where evidence conflicts or more than one inference may reasonably be drawn from the evidence, the findings are to be affirmed.  Id. at 60, 37 W.C.D. at 240.  Similarly, “[f]actfindings are clearly erroneous only if the reviewing court on the entire evidence is left with a definite and firm conviction that a mistake has been committed.”  Northern States Power Co. v. Lyon Food Prods., Inc., 304 Minn. 196, 201, 229 N.W.2d 521, 524 (1975).  Findings of fact should not be disturbed, even though the reviewing court might disagree with them, “unless they are clearly erroneous in the sense that they are manifestly contrary to the weight of the evidence or not reasonably supported by the evidence as a whole.”  Id.

At Finding 1, the compensation judge concluded in part that the employee’s post-injury employment with the employer as an estimator was economically, physically, and vocationally suitable.  At Finding 3, the judge found that the employee’s work-related shoulder injuries in January and April 2002 were not substantial contributing factors in the development of the employee’s major depressive disorder or anxiety disorder with depressive mood, noting that “evidence fails to prove that the employee developed a psychological condition as a result of the pain complaints and loss of ability to function.”  In Finding 4, the judge suggested that the employee was subject to a preexisting psychological condition that was aggravated by anxiety associated with the estimator  position and that, rather than the work-related disability, “[t]he job circumstance in general caused the anxiety and aggravated the underlying psychological condition.”  The employee contends first, on a factual basis, that substantial evidence does not support the judge’s finding that loss of function arising from the employee’s work injuries was not a substantial contributing factor in his disabling psychological condition.  The employee contends also, on a legal basis, that the compensation judge erred as a matter of law in concluding at Finding 4 that the causal link between the employee’s shoulder injuries and his psychological condition was “so indirect as to break the chain of causation.”

With regard to his contention on the facts as to the effect of the employee’s functional loss, the employee argues that both Dr. Augdahl and Dr. Gratzer agreed that loss of function from the employee’s work injuries was a substantial contributing factor in the employee’s psychological condition.  Such agreement, the employee argues, constitutes “unopposed” medical opinion that cannot be ignored.  See Ruether v. State of Minnesota, 455 N.W.2d 475, 478, 42 W.C.D. 1118, 1122 (Minn. 1990).  The employee argues that, because both Dr. Augdahl and Dr. Gratzer attribute the employee’s psychological condition to “a forced career transition,” and because that transition would not have been necessary but for his injury-related loss of ability to function, the employee’s psychological condition is a direct and natural consequence of the work injuries.  He argues further that “the record contains no evidence of any kind that would competently establish that the Employee’s forced career transition, from plumber to estimator, was not a substantial contributing factor to his psychological condition” and that “[t]he Employee’s loss of his plumbing career is not, as the Compensation Judge below described it, a concern ‘peripheral to the workers’ compensation system’.”  We are not persuaded.

Where a work-related physical injury causes, aggravates, accelerates or precipitates a mental injury, that mental injury is compensable.  See Hartman v. Cold Spring Granite Co., 243 Minn. 264, 67 N.W.2d 656, 18 W.C.D. 206 (1954).  It is not necessary that the physical injury be the sole cause of the mental injury; it is sufficient if the work-related physical injury is a substantial contributing factor in the mental injury.  Miels v. Northwestern Bell Tel. Co., 355 N.W.2d 710, 715, 37 W.C.D. 164, 170 Minn. 1984).  It does not necessarily follow, however, that the physical injury caused the mental injury just because the mental injury followed the physical injury.  In the present case, it is undisputed that the employee, in September 2004, suffered from an adjustment disorder with depressed mood.  Dr. Augdahl subsequently concluded that the employee’s diagnosis was evolving into a major depressive disorder, while Dr. Gratzer viewed the employee’s adjustment disorder and depressive symptoms as resolving.  Regardless of the diagnosis or its resolution, however, the issue is whether the employee’s physical injuries substantially contributed to his psychological condition.  While they may have agreed that the employee’s career change and related stress was a substantial cause of his psychological condition, Dr. Augdahl and Dr. Gratzer did not agree that the work injuries were a substantial contributing factor in that psychological condition.  Dr. Gratzer quite clearly did not relate the employee’s psychological condition to the work injuries, and the compensation judge was entitled to rely on that opinion.  See Nord v. City of Cook, 360 N.W.2d 337, 342-43, 37 W.C.D. 364, 372-73 (Minn. 1985) (a trier of fact’s choice between experts whose testimony conflicts is usually upheld unless the facts assumed by the expert in rendering his opinion are not supported by the evidence).  While Dr. Augdahl did relate the employee’s psychological injury to the employee’s pain and related loss of function, the judge was not required to accept that opinion.  In addition to Dr. Gratzer’s opinion, the judge noted further that the time elapsed between the work injuries and surgeries and the onset of psychological complaints, together with the records and history provided at the time of the employee’s psychological evaluation and treatment, failed to prove that the employee developed a psychological condition as a result of his pain complaints and loss of ability to function.  We cannot conclude that the judge’s assessment of the evidence was unreasonable.

We also reject the employee’s argument that, because a career change was necessitated by the work injuries, the mental stress subsequently caused by the job transition was a direct and natural result of the employee’s work injuries.  While there is arguably a kind of chain of causation here, conditions resulting from such indirect causal relationships are generally not compensable as workers’ compensation injuries.  See e.g., Koenig v. Northern Insulation Co., 358 N.W.2d 644, 37 W.C.D. 266 (Minn. 1984) (aggravation of ulcer condition was not compensable where it was a result of employee’s experience with his QRC, the requirements of his job search and reporting, and the necessity of litigation); Hendrickson v. George Madsen Constr. Co., 281 N.W.2d 672, 31 W.C.D. 608 (Minn. 1979) (heart attack sustained while pursuing workers’ compensation claim in court was not compensable); Schander v. Northern States Power Co., 320 N.W.2d 84, 34 W.C.D. 695 (Minn. 1982) (injury sustained in automobile accident while returning from retraining class was not compensable); Melartin v. Mavo Sys., Inc., No. WC04-328 (W.C.C.A. May 5, 2005) (aggravation of pre-existing alcoholism by time off after work injury was not compensable).  Here, the compensation judge found that the evidence did not adequately relate the employee’s psychological condition to his physical injuries.  We find this case factually similar to the cases cited above.  Substantial evidence supports the compensation judge’s decision that the employee’s work injuries were not a substantial contributing factor in the development of the employee’s psychological condition.  That decision must, therefore, be affirmed.  Hengemuhle v. Long Prairie Jaycees, 358 N.W.2d 54, 59, 37 W.C.D. 235, 239 (Minn. 1984).

With regard to his contention that the judge erred as a matter of law in concluding that stress relating to the employee’s forced career change was not a direct result of the employee’s work injury, the employee argues that the determinative principle is that “stress resulting from loss of function is not a superseding cause breaking the chain of causation between a physical work-related injury and a consequential psychological injury.”  Here, the employee contends, “stress related to the Employee’s forced career transition must be recognized as a part of an unbroken chain of causation between the work-related injuries to the Employee’s shoulders and his disabling psychological condition.”  In other words, “[o]nce the physical limitations caused by the employee’s shoulder injuries precluded further plumbing work, it became inevitable that the employee would have to find another line of work, and stress related to such a transition is reasonably foreseeable.”  We are not persuaded.

We find that the employee’s argument seeks to blur the line between stress resulting from loss of function and stress resulting from a career change due to that loss.  While there clearly may be cases in which stress related to the physical injury overlaps other stressors arguably consequential to an injury, a compensable consequential injury is dependent upon a medical connection between the primary injury and a later condition.  In Melartin v. Mavo Systems, Inc., this court reviewed Minnesota case law and discussed at length the causation rules that determine how far the range of compensable consequences is carried, once the primary injury is causally connected with the employment.  See Melartin v. Mavo Sys., Inc., No. WC04-328, slip op. (W.C.C.A. May 5, 2005).  There we stated,

Here, we have already affirmed the judge’s factual determination that there was no medical connection between the employee’s primary injuries and his psychological condition.  The employee’s legal contention now - - that his physical work injuries caused his need for a change in careers which in turn caused stress related to that change which in turn caused his psychological condition - - depends on our granting substantial import to several intervening non-medical causative factors, all separating the personal injury from the claimed psychological consequence.  It is generally true that the more remote the claimed consequence is from the personal injury the less likely it is that the claimed consequence is compensable.  Even were the causative chain in this case not so extended, however, we would conclude on the facts of this case that, because the personal injury was not the medical cause of the claimed psychological consequence - - that is, because the employee’s psychological condition was not a direct consequence of physical pain or physical inability to perform his post-injury job - - the employee’s claimed psychological consequence is not compensable.

Concluding that it was neither unsupported by substantial evidence nor clearly erroneous as a matter of law, we affirm the compensation judge’s denial of the employee’s claim for benefits for a psychological injury consequent to stress related to his inability to adjust to his post-injury employment.  Hengemuhle v. Long Prairie Jaycees, 358 N.W.2d 54, 59, 37 W.C.D. 235, 239 (Minn. 1984).