Claimant appeals from an order of the Worker’s Compensation Board reversing the referee’s determination that claimant’s knee injury is compensable. We reverse.
The facts surrounding claimant’s injury are essentially undisputed. He was employed as a mechanic. On June 18, 1982, he suffered a knee injury while walking across the shop floor after answering his employer’s telephone. He testified that he felt a sudden “snap or a terrific pain” in his left knee and then could not put any weight on it. He stumbled but did not fall to the floor. Claimant was hospitalized that day and, on June 23, underwent corrective surgery. He was determined to have a longitudinally torn left medial meniscus with the medial portion retracted into the intercondylar notch; an arthoscopy with partial resection of the medial meniscus was performed. SAIF, the employer’s insurer, denied claimant’s worker’s compensation claim on the basis that there was insufficient evidence to show that the injury was a result of his work activities.
The evidence at the hearing included reports of two physicians. Claimant’s treating physician, Dr. Weinman, reported:
“A meniscus may be torn just by an abnormal step or a twist on a knee. The patient reported that this happened to him while he was walking, while he was on the job. In looking at the meniscus with an arthroscope, it is impossible to tell if there was a pre-existing small tear that was merely enlarged by his activity on the job. If his job includes a lot of squatting down or twisting on his knee, it could contribute to his torn meniscus.”
SAIF’s medical administrator, Dr. Norton, reported:
“[C]laimant’s medical records were reviewed for an opinion regarding the probable cause of his torn meniscus. The review discloses that the claimant developed acute knee pain while walking. There evidently was no prior history of symptoms or injury of any consequence.
“For an acute tear of the medial meniscus to occur in this fashion would be unusual, although not impossible. For a previously torn meniscus to extend enough to displace into the intercondylar notch while walking would be far more *162likely. A degenerative fissuring or tearing could occur without a prior incident of accidental injury.
“There appears to be little question that the sudden displacement of the meniscus with locking of the knee occurred at the time of the work incident based on the reported information. The meniscus was still locked into the intercondylar notch at the time of the arthroscopic examination, (an untenable situation) so it is evident that the need for the urgent surgery was occasioned by the incident occurring at work. Under those circumstances it appears that whether or not the condition represented a pre-existing partial tear, or an acute new injury, the result was the same ie. [sic] the claimant’s first reported episode of knee locking persisted and required urgent surgery. It was probably happenstance that it occurred at work.
“You asked whether job activities such as kneeling, twisting etc. could cause a meniscus tear. This type of activity could result in an acute meniscus tear, although ordinarily it would be a memorable experience. A degenerative tear could develop gradually, (asymptomatically) and kneeling and twisting activities could contribute over an extended period of time if carried out in a particularly frequent, repetitious, and abusive fashion. The commonest cause of meniscus tears, however, are probably single episodes of trauma. Squatting and rising could produce an acute meniscus tear, but the first pain would be experienced during the commission of the act, not at some remote time. In this claimant’s case, if the activity of squatting or working on the knees had caused the meniscus to ‘tear’ or a previously torn meniscus to ‘displace and lock,’ the pain and limitation would be experienced immediately upon rising and extending the knee, not after walking a distance. The displacement and locking occurred at the time the acute pain was experienced (while walking).”
The referee found the injury compensable. He analyzed the issue in terms of whether the injury was sufficiently work connected to justify a holding that it arose out of claimant’s employment. He found that the injury occurred while claimant was walking across the shop floor after answering the employer’s telephone. Relying on Hubble v. SAIF, 56 Or App 154, 641 P2d 593, rev den 293 Or 103 (1982), in which we held that a torn medial meniscus sustained by a worker while walking down a straight corridor was compensable, the referee concluded that “when walking is a part of a worker’s *163job, as it was here, the risk of injury from the walking is a risk of the job.”
The Board reversed. It concluded that the case should be analyzed “in terms of whether claimant’s fall was caused by idiopathic factors or was truly work related,” according to the standard approved in Phil A. Livesley Co. v. Russ, 296 Or 25, 672 P2d 337 (1983). Under that standard, an idiopathic fall is not compensable, nor is a fall compensable if it is equally possible that its cause was idiopathic or work related; a truly unexplained fall that occurs on the employer’s premises while the employe is performing required duties is compensable if the employe can eliminate idiopathic causes. 296 Or at 30. The Board concluded that “[t]he evidence in this case indicates * * * an equal possibility that the cause of claimant’s fall was a pre-existing (i.e., idiopathic) problem or that it was connected to claimant’s employment” and that, therefore, claimant had failed to prove that his fall was connected to a risk of his employment. It reversed the order of the referee and ordered SAIF’s denial reinstated.
We conclude that the Board erred in applying an “unexplained fall” analysis to this case. That analysis is applicable only when the claimant falls for unknown reasons and injury results from the fall. In such cases, the question is whether there is a sufficient connection between the factors precipitating the unexplained fall and the claimant’s employment. Because evidence of that causal nexus cannot be directly established, the claimant is required affirmatively to exclude idiopathic factors as the cause of the fall in order to permit an inference that the fall was traceable to some risk of the employment. Phil A. Livesley Co. v. Russ, supra, 296 Or at 32.
In applying that analysis, the Board relied primarily on our decision in Mackey v. SAIF, 60 Or App 536, 654 P2d 1144 (1983). However, we find that case to be inapposite. There the claimant, a bus driver, injured her low back when her knee buckled and she fell. There was no question but that the fall caused the low-back injury; the issue was whether the cause of her fall was work connected. Because the evidence showed no more than that the cause of the fall, her buckling knee, was as likely idiopathic as it was work connected, and because there was no evidence that her knee buckled as a risk *164of her employment, we held that she had failed to satisfy her burden of proving work connection. 60 Or App at 538-39.
The issue in this case, as in Mackey, is whether the evidence shows that the cause of the collapse of the claimant’s leg was work connected. Here, however, the cause of the collapse is known; it was the displacement of claimant’s medial meniscus while he was walking across the shop floor after answering his employer’s telephone. That is not contested. The evidence from both doctors establishes that walking can cause an acute medial meniscus tear and displacement. Claimant here is not burdened with establishing the connection to his employment of an unexplained fall that resulted in injury. Rather, the identified event causing the injury was the sudden, unexpected displacement of the medial meniscus, which SAIF agrees should be evaluated under the law governing accidental injury.1 Because the medical evidence showed that walking was the immediate cause of claimant’s acute medial meniscus displacement, the fact that he could not by medical evidence eliminate the possibility that he could have had a pre-existing Assuring or susceptibility to tearing does not defeat his claim. See Harris v. Albertson’s Inc., 65 Or App 254, 257, 670 P2d 1059 (1983). Under these circumstances, claimant is not required to disprove all idiopathic causes to permit an inference of work connection; his only burden of proof is to show by a preponderance of the evidence that his injury “arose out of and in the course of employment.” See ORS 656.005(8)(a).
In determining whether an injury is one arising out of and in the course of employment, we apply a unitary work-connection approach: “[I]s the relationship between the injury and the employment sufficient that the injury should be compensable?” Rogers v. SAIF, 289 Or 633, 642, 616 P2d 485 (1980). Under that approach, “course of employment” and “arising out of employment” are components of a single work connection analysis. The accident here occurred during working hours, on the work premises and while claimant was performing a task required by the employer for its benefit— walking back to his work site after answering the employer’s *165telephone. SAIF acknowledges that the “course of employment” component is satisfied.
The focus of our inquiry is on “arising out of employment.” To satisfy that part of the test, the employe must show a causal link between the occurrence of the injury and a risk connected with the employment. Phil A. Livesley Co. v. Russ, supra, 296 Or at 29. At the time of the injury claimant was walking across the shop floor. The issue, therefore, is narrowed to whether that walking provides the necessary causal link. We conclude that it does.
We note, as did the referee, that this case is factually similar to Hubble v. SAIF, supra, where the claimant suffered a torn medial meniscus while walking down a corridor. SAIF argues that Hubble is distinquishable on the basis that there the claimant’s job required a significant amount of walking and that occasional walking cannot be considered a risk of employment. We disagree. Our holding in Hubble did not turn on the fact that the claimant’s job required considerable walking, and we do not find the quantum of requisite walking to be the determining factor here. Where a specific work activity, whether isolated or repetitive, is a part of a claimant’s job, the risk of injury from that activity is a risk of that job. Here, the evidence showed that walking was a part of claimant’s job. The risk of injury from that walking was a risk of the job. Claimant has met his burden of proving that his claim is compensable.
Reversed and remanded with instructions to reinstate the referee’s order.
Claimant argues alternatively that his claim is compensable as an occupational disease. Because we determine that he suffered an accidental injury, we do not discuss that contention.