Defendant appeals a judgment for plaintiff in this personal injury action, raising four assignments of error. Plaintiff contends that all of defendant’s assignments of error are without merit. We reverse.
Defendant’s first three assignments of error implicate the trial court’s handling of evidence related to plaintiff s claim for lost earnings. We need not reach those assignments of error, however, because we find defendant’s fourth assignment of error dispositive. That assignment of error requires us to review the necessity of expert witness testimony on the issue of causation when a plaintiffs medical situation is complicated. In this case, the trial court partially denied defendant’s motion for directed verdict on that issue; therefore, we view the evidence, including attendant inferences, in the light most favorable to the nonmoving party, here plaintiff. Hudjohn v. S&G Machinery Co., 200 Or App 340, 342, 114 P3d 1141 (2005) (citing Mauri v. Smith, 324 Or 476, 479, 929 P2d 307 (1996); Joshi v. Providence Health System, 198 Or App 535, 536, 108 P3d 1195 (2005)).
In early 1999, plaintiff purchased and attempted to operate a bookstore, which turned out not to be as profitable as he desired because, in his opinion, it was in an undesirable location. That spring, plaintiff reported to his physician that he was experiencing popping in his knees, and, by late 1999, plaintiffs knees started to give out and he had trouble walking. After a course of medication and therapy, plaintiffs knees felt better and he was able to walk again. During that period, however, plaintiff was repeatedly hospitalized and ultimately diagnosed with AIDS and a number of AIDS-related ailments, including general aches and pains, back problems, nerve injury, loss of muscle mass, weight loss, foot drop, joint disease, anemia, and pneumonia. In February 2000, plaintiff began a regimen of pain medications to combat the generalized pain associated with those ailments, and by early June he had started taking regular doses of morphine. Those medications treated most of plaintiffs ailments successfully. It was then that plaintiff decided to move his bookstore to a new location. The physical labor involved with that *528move, combined with the effects of plaintiffs illness and medications, left him weak, ill, and in pain.
On June 26, 2000, plaintiff rode a bus operated by defendant. When plaintiff reached his desired location, he informed the driver that he needed to retrieve his bicycle from a rack located on the front of the bus. As he attempted to retrieve his bicycle from the rack, the bus accelerated forward and made contact with the left side of plaintiffs body, spinning him back on his right leg. Plaintiff immediately felt pain and experienced shock as a result of the incident.
Later that day, plaintiff contacted the office of his regular medical provider, James McDonald. McDonald could not immediately see him, so plaintiff reported to the emergency room of an area hospital. While there, a nurse examined plaintiff but found no sign of injury. According to plaintiff, the nurse felt that he “was looking to score off of Tri-Met.” Plaintiff later contacted McDonald again, and McDonald advised plaintiff to take the pain medication he had been prescribed as part of his AIDS-related therapy.
Approximately three weeks after the incident, plaintiff reported to McDonald and complained of a number of ailments, including weakness, pain in his joints, arms, and knees, difficulty with heavy lifting and typing, and emotional stress. McDonald diagnosed most of those ailments, including a “sprained strain” in plaintiffs right knee. A subsequent MRI revealed a meniscal tear in that knee. Plaintiff consulted an orthopedist who opined that plaintiffs knee condition was at least one year old. The orthopedist operated on plaintiffs knee in late August. Afterwards, plaintiff still felt that he had “bad knees” and sought a surgeon who would perform a second operation on his right knee. After a few surgeons refused to do so, plaintiff obtained a second MRI, which again revealed a meniscal tear in the right knee. After a second surgery in April 2001, plaintiff’s complaints of pain in that knee subsided.
At trial, plaintiff claimed that, as a result of the incident, he incurred noneconomic damages as well as economic damages in the form of medical expenses and lost earnings related to his inability to operate his bookstore. Plaintiff did not offer expert testimony as to the cause of any his injuries. *529At the close of plaintiffs case, defendant moved for directed verdict with regard to liability and all damages. The trial court granted defendant’s motion as to plaintiffs medical expenses1 and defendant’s liability for the meniscal tear in plaintiffs right knee, but denied the motion as to defendant’s liability for the “sprained strain” and other injuries, as well as plaintiffs related noneconomic damages and economic damages of lost earnings. Thereafter a jury found defendant liable and awarded plaintiff $20,000 in noneconomic damages and $25,000 in economic damages. This appeal ensued.
On appeal, defendant argues that the “trial court erred in submitting to the jury plaintiffs claim for nonecon-omic damages for personal injuries without expert witness testimony on the issue of causation.” In that regard, defendant alludes to the trial court’s partial denial of defendant’s motion for directed verdict based on plaintiffs lack of expert testimony regarding causation. Before proceeding, however, we must make two points clear. First, although defendant apparently argues only that the lack of expert testimony regarding causation invalidates the award of noneconomic damages, we note that defendant’s challenge inherently extends to the findings of liability and economic damages as well.2 There is no basis in law to conclude that a plaintiffs lack of proof as to causation in the context of the facts of this negligence action would invalidate only some part of plaintiffs damages award, and not the entire judgment. Second, although defendant’s assignment of error would lead one to believe that the trial court had denied defendant’s motion in its entirety, that is not the case. The trial court granted defendant’s motion as to plaintiffs meniscal tear and denied the motion only as to plaintiffs “sprained strain” and other injuries. Thus, when the trial court removed plaintiffs meniscal tear from the jury’s consideration, and the jury subsequently found defendant liable and awarded damages, the *530liability and damages findings must have been based on plaintiffs “sprained strain” and other injuries. If proof of causation as to those injuries required expert testimony, the entire verdict must fall.
In reviewing the denial of a motion for directed verdict based on lack of causation, we are required to reverse only if we determine that no evidence of causation exists in the record. Peery v. Hanley, 135 Or App 162, 165, 897 P2d 1189 (1995); see also Mauri, 324 Or at 479 (on review of a denial of a motion for directed verdict, we review for errors of law). However, in complex medical situations, that evidence must take the form of expert testimony, and a lack of such testimony will constitute a lack of evidence regarding causation. As we recently stated:
“When the element of causation involves a complex medical question, as a matter of law, no rational juror can find that a plaintiff has established causation unless the plaintiff has presented expert testimony that there is a reasonable medical probability that the alleged negligence caused the plaintiffs injuries. See Uris v. Compensation Department, 247 Or 420, 424, 427 P2d 753 (1967) ([W]here injuries complained of are of such character as to require skilled and professional persons to determine the cause and extent thereof, the question is one of science and must necessarily be determined by testimony of skilled, professional persons.’ (Internal quotation marks omitted.)); see also Chouinard[ v. Health Ventures], 179 Or App [507, 512, 39 P3d 951 (2002)]. The rule prevents jurors from speculating about causation in cases where that determination requires expertise beyond the knowledge and experience of an ordinary lay person. Howerton v. Pfaff, 246 Or 341, 347-48, 425 P2d 533 (1967).”
Baughman v. Pina, 200 Or App 15, 18, 113 P3d 459 (2005); see also Hudjohn, 200 Or App at 351.
Defendant asserts that, given plaintiffs complicated medical history, plaintiff was required to present expert witness testimony on the issue of causation. Plaintiff disputes that assertion and contends that, in any event, he presented evidence sufficient to establish causation. We agree with defendant.
*531The Supreme Court in Uris set out several factors to consider when determining whether medical testimony is required to prove causation. 247 Or at 426. Those factors include the complexity of the situation, “the immediate appearance of symptoms,” prompt consultation with a medical authority, whether the plaintiff was in good health or suffered from a disability of a kind similar to the present injury, and whether the defendant offered expert testimony showing “that the alleged precipitating event could not have been the cause of the injury.” Id.
The spectrum of cases cited by the parties illustrates how factual this determination is in practice, especially under the standard of review for a denial of a motion for directed verdict, in which the court is required to review the entire record. See Peery, 135 Or App at 165. For example, defendant asserts that this case most closely resembles Cleland v. Wilcox, 273 Or 883, 884, 543 P2d 1032 (1975), in which the plaintiff was involved in a minor automobile accident. Initially the plaintiff had a small bruise on one hand, but experienced no back pain. Id. About four weeks later, the plaintiff began experiencing back pain and reported that pain to his physician. Id. Months later the physician determined the source of the pain to be a protruding intervertebral disc, a condition that required surgery to correct. Id. at 885. At trial, the plaintiffs physician testified that only half of all cases involving a back injury led to such a protruded disc and that the plaintiff could have had some previously unknown preexisting condition that led to the disc problem. Id. at 885-86. Based on the equivocal nature of the physician’s testimony, the Supreme Court affirmed the trial court’s granting of the defendant’s motion for nonsuit. Id. at 889.
On the other side of the spectrum, plaintiff contends that this case is more like Wheeler v. LaViolette, 129 Or App 57, 877 P2d 665 (1994). There the plaintiff ran down a dock in aid of a small child. Id. at 59. As she ran, the plaintiff fell into a hole where a plank was missing. Id. She immediately felt pain. Id. Four hours later, she went to the hospital where evidence of her knee injury was noted in the emergency room report. Id. Based on that record, we held that the plaintiffs claim was sufficient to survive a motion for directed verdict. Id. at 61.
*532Here, the trial court classified plaintiffs injuries in two groups: those associated with the meniscal tear and all others. On the one hand, the court barred the jury from considering the meniscal tear because plaintiff had not offered expert testimony showing a causal link between the incident and the tear. On the other hand, the court ruled that the jury could find that plaintiff suffered “some pain and injury” when he was struck by defendant’s bus, and that “it would be a jury question as to whether those injuries were the cause of an earnings loss.”
To make clear the basis of our holding in this case, we turn first to whether plaintiff needed to present expert testimony regarding his meniscal tear.3 Although we find the evidence in that regard less speculative than the evidence in Cleland and cases of its ilk, we nonetheless believe expert testimony was required here. First, although plaintiff admitted himself to the emergency room on the same date as the incident, unlike in Wheeler the examining nurse found no sign of injury. Second, this was not an “uncomplicated situation.” Uris, 247 Or at 426; see also Hudjohn, 200 Or App at 353 (noting that exceptions to the rule of expert testimony “appear to have been limited to cases of simple injuries, generally without a substantial possibility of alternative causation”). In addition to the incident at issue, plaintiffs meniscal tear could have been caused by a variety of other factors. Plaintiff had complained of similar knee problems in the year preceding the incident, and, indeed, plaintiffs own orthopedist opined after the incident that plaintiffs knee problems were at least one year old. Plaintiff had also been on a pain medication regimen dating back to February 2000 and had begun taking morphine shortly before the incident. Those pain medications could have masked damage preceding the incident. Finally, the AIDS-related illnesses from which plaintiff suffered, as well as the physical labor involved with moving and operating his bookstore, constitute other potential sources of plaintiffs injuries. Given the totality of the factual record in this matter, we cannot conclude that this is the type of case in which the ordinary layperson could determine the cause of plaintiffs meniscal tear.
*533Having concluded that plaintiff was required to present expert testimony regarding his meniscal tear, we next turn to plaintiffs “sprained strain.” As an initial matter, we believe that we would likely conclude that plaintiff needed to present expert testimony regarding the “sprained strain” even if he had not suffered the meniscal tear. The same factors that complicate the issue of causation as to the meniscal tear would seem to complicate that issue as to the “sprained strain” as well. Even if the jury could find that the incident caused plaintiff some pain and injury, how was it to distinguish that injury from the ailments caused by plaintiffs past knee problems, AIDS-related illnesses, medications, and heavy workload? We need not answer that question, however, because we reject the distinction made by the trial court between the meniscal tear and the “sprained strain.” We find the distinction, although understandable at first, ultimately unconvincing. The trial court in this case attempted to distill two discrete knee injuries from one complex situation. The problem with that method is that it is impossible for a layperson to draw the line between two similar or related injuries, one requiring expert testimony and one not, except through uninformed speculation. Cf. Uris, 247 Or at 424, 426 (expert testimony might not be necessary when plaintiff is “free from any disability of the kind involved” or when one injury “was not connected with the injury of which plaintiff complains”). If plaintiffs medical situation was so complicated that the jury needed expert testimony to understand the cause of plaintiffs meniscal tear, how was the jury to know what caused plaintiffs “sprained strain” as a separate matter? And, if the jury did not know how to separate the cause of the “sprained strain” from the cause of the meniscal tear, how was it to measure damages arising solely from the “sprained strain” and segregate those damages from the damages arising from the meniscal tear?4 See Uris, 247 Or at 424 (expert testimony required “where injuries complained of are of such character as to require skilled and professional persons to determine the cause and extent thereof’ (citations omitted; emphasis added)). In short, just as plaintiff needed to offer *534expert testimony regarding causation of his meniscal tear, he needed to offer expert testimony regarding causation of all similar and related injuries, including his “sprained strain.”5
We therefore conclude, as a matter of law, that plaintiffs medical situation was so complicated as to require expert testimony on the issue of causation, not only as to plaintiffs meniscal tear, but also his “sprained strain” and other injuries. The trial court therefore erred in denying defendant’s motion for directed verdict as it related to plaintiffs “sprained strain” and other injuries.
Reversed and remanded.
The trial court granted defendant’s motion as to plaintiff’s medical expenses because plaintiff presented no evidence at trial to prove the amount of those expenses.
Before the trial court, defendant directed its causation challenge at the findings of liability and damages in general. Further, although the phrasing of defendant’s fourth assignment of error is inartful, the substance of defendant’s argument clearly encompasses a general challenge to the findings of both liability and damages.
Plaintiff has abandoned his cross-appeal challenging, among other things, the trial court’s ruling regarding his meniscal tear. We nonetheless discuss the issue because it is important to our analysis regarding plaintiffs “sprained strain.”
As plaintiff testified, his right knee pain subsided after his second meniscal tear surgery. That testimony underlines the complexity and interrelatedness of his knee injuries. If the same surgery that corrected plaintiffs meniscal tear also corrected his “sprained strain,” then the jury had no way of knowing, except through speculation, that the two injuries were not related.
That analysis holds true for plaintiffs other injuries as well, each of which was similar or related to ailments from which plaintiff suffered before the incident, and the causation of none of which was proved at trial by expert testimony. Although a jury could find that the incident caused plaintiff some general pain and injury, it is difficult to conceive how the jury, except through speculation, could conclude that that injury was new, different, or greater than what plaintiff suffered before or would have suffered had the incident not occurred.