This appeal arises out of a $1,333,279.31 jury verdict for personal injuries in favor of Delmas R. Hose against the Chicago and North Western Transportation Company ("CNW") under the Federal Employers' Liability Act, 45 U.S.C. Secs. 51-60 ("FELA"). Hose worked as a welder at CNW's reclamation center in Council Bluffs, Iowa, from 1976 to 1991. The reclamation process exposed Hose to substantial amounts of fumes and dust containing manganese, a toxin. Hose's evidence shows that CNW did not provide adequate ventilation, warnings, monitoring, or safety training in relation to manganese hazards in the work place.

In April 1991, Hose reported that he was suffering memory loss and had fallen in the welding shop when his right leg gave out. CNW directed him to take sick leave and not to return to work until he obtained medical approval.

In April 1992, Hose filed suit in federal district court against CNW under FELA. Hose also sued Stoody Deloro Stellite, Inc., and Arcair Company, who designed, manufactured, and sold the welding equipment used by CNW, on state law counts. CNW filed cross-claims against Stoody and Arcair. Prior to trial, Hose settled with Stoody and Arcair, and CNW's cross-claims against them were dismissed. A three-week trial was held in 1994. The jury found CNW negligent under FELA and returned a verdict in Hose's favor. The jury apportioned Hose's damages finding CNW was ninety percent at fault, and Hose was ten percent contributorily at fault. At CNW's request, the jury was instructed to evaluate the contributing fault of Stoody and Arcair and found no fault on their part. The district court, the Honorable Harold D. Vietor, rejected CNW's motions for judgment as a matter of law and a new trial. This appeal followed.

On appeal, CNW challenges the admissibility of certain expert medical testimony, the sufficiency of the evidence of medical causation, the exclusion of evidence of Hose's claims against Stoody and Arcair and their subsequent settlement, and certain jury instructions. Hose cross-appeals the jury's finding of ten percent contributory negligence. We affirm.

In February 1992, after a series of medical examinations and varying diagnoses by different specialists, Hose was examined by Dr. Carol Angle, who was then director of clinical toxicology and professor of pediatrics at the University of Nebraska Medical Center. Dr. Angle made the diagnosis that Hose was suffering from manganese encephalopathy. In March 1992, Dr. Jan Golnick, a neurologist, concurred in Dr. Angle's diagnosis and removed Hose from seizure medication prescribed by another physician. Various other examinations and treatments followed.

Manganese encephalopathy is a dementia, or difficulty in thinking, that usually results from chronic exposure to manganese, and is usually associated with reduced ability to control one's movements. (Tran. 491) (testimony of Dr. Angle). CNW challenges the expert opinion evidence of Hose's physicians under Daubert v. Merrell Dow Pharmaceuticals, Inc., --- U.S. ----, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993) (interpreting Fed.R.Evid. 702).1 Under Daubert, the trial court plays a "gatekeeping role" to "ensure that any and all scientific testimony or evidence admitted is not only relevant, but reliable." Id. at ----, ----, 113 S. Ct. at 2798, 2795. Relevance means there must be "a valid scientific connection to the pertinent inquiry" in the case. Id. at ----, 113 S.Ct. at 2796. Reliability means the evidence must be based upon scientific knowledge, i.e., "ground[ed] in the methods and procedures of science," and must represent "more than [a] subjective belief or [an] unsupported speculation." Id. at ----, 113 S.Ct. at 2795.2

We review a trial court's decision on the admissibility of expert medical evidence for a clear abuse of discretion. Gier v. Educational Serv. Unit No. 16, 66 F.3d 940, 942 (8th Cir.1995). If the party opposing the evidence does not properly object, however, we review for plain error. Owen v. Patton, 925 F.2d 1111, 1115 (8th Cir.1991). Finally, we will not reverse a jury verdict if an erroneous admission of expert evidence is harmless. McKnight v. Johnson Controls, Inc., 36 F.3d 1396, 1405 (8th Cir.1994).

CNW first challenges Dr. Naresh Gupta's deposition testimony based on a positron emission tomography ("PET") scan of Hose's brain. The PET scan was ordered by Dr. Golnick to aid his ability to make a diagnosis. Dr. Gupta testified that the PET scan ruled out alternative diagnoses of Hose's injuries, such as alcoholism, a stroke, and Alzheimer's disease, but was "consistent" with manganese encephalopathy.

CNW sought to exclude use of the PET scan in a motion in limine and timely objected to Dr. Gupta's testimony.3 CNW argued the PET scan was not scientifically reliable for use in making a diagnosis of manganese encephalopathy and was not relevant for any other purpose. The district court, however, admitted Dr. Gupta's testimony because it was relevant in terms of excluding other diagnoses of Hose's injuries and was limited to showing consistency with, not diagnostic proof of, manganese encephalopathy. Dr. Gupta further testified to the limited scientific experience and literature about using PET scans in manganese cases.

We find the district court did not abuse its discretion by allowing Dr. Gupta's testimony based on the PET scan. Dr. Gupta's testimony clearly showed the limited use of the PET scan, but that use was nonetheless relevant. In determining the cause of a person's injuries, it is relevant that other possible sources of his injuries, argued for by defense counsel, have been ruled out by his treating physicians. Indeed, ruling out alternative explanations for injuries is a valid medical method. See McCullock v. H.B. Fuller, Inc., 61 F.3d 1038, 1043-44 (2d Cir.1995). There is also no question that the PET scan is scientifically reliable for measuring brain function. The fact that Hose's treating physician ordered the PET scan prior to the initiation of litigation is another important indication that this technique is scientifically valid. Cf. Daubert v. Merrell Dow Pharmaceuticals, Inc., 43 F.3d 1311, 1317 (9th Cir.) (expert testimony based on "legitimate, preexisting research unrelated to the litigation provides the most persuasive basis" for ensuring scientific validity of expert testimony), cert. denied, --- U.S. ----, 116 S. Ct. 189, 133 L. Ed. 2d 126 (1995).

CNW next challenges Dr. Richard Andrews's deposition testimony regarding the polysomnogram conducted under his supervision. A polysomnogram assesses whether a person has any sleep disorders. Like the PET scan, the polysomnogram was ordered by Hose's attending physician, Dr. Golnick. Dr. Andrews testified that Hose has a sleep disorder consistent with an exposure to a toxic substance.

CNW did not object to the admission of Dr. Andrews's deposition testimony until plaintiff's counsel prepared to read the deposition at trial. At that point, CNW objected on the grounds that the polysomnogram was scientifically unreliable in the diagnosis of manganese encephalopathy and was otherwise irrelevant. To support its argument, CNW presented the district court a statement in a deposition by Dr. Angle that "there is no clearcut distinction at this point of the type of sleep disorders from toxic and nontoxic encephalopathy." (Deposition of Dr. Angle, April 28, 1994, at 26-27). Hose's counsel argued the sleep study was relevant because Dr. Korn would testify, as he ultimately did, that sleep disorders are common in people suffering from encephalopathy. On this basis, the district court allowed the plaintiffs to introduce Dr. Andrews's deposition.

We find the district court did not abuse its discretion in rejecting CNW's last-minute objection to the scientific validity of the polsomnogram. The statement by Dr. Angle on which CNW relied did not undermine the validity of the sleep study, but rather only explained that there are both toxic and non-toxic explanations for sleep disorders such as Hose's. Moreover, Dr. Andrews's testimony clearly showed the limited value of the sleep study: the polysomnogram was not diagnostic of manganese encephalopathy; there was no "cause-and-effect" relationship between manganese and sleep disorders; and sleep disorders were commonly associated with alcohol, stroke, or aging. In these circumstances, CNW was free to argue to the jury that this testimony should carry little weight, but it was not reversible error to admit it.

CNW challenges Dr. Angle's opinion that her diagnosis at the time of trial was manganese encephalopathy caused by inhalation of manganese fumes at CNW's Reclamation Center, since, CNW argues, her opinion depended on the truthfulness of Hose and his wife in telling Dr. Angle about his physical symptoms. Relying on United States v. Whitted, 11 F.3d 782 (8th Cir.1993), CNW argues that allowing an unqualified opinion based on a patient's statements to a doctor is improper since such an opinion vouches for the truthfulness of the patient's story.4

Assuming CNW made a timely objection to Dr. Angle's testimony,5 we find no abuse of discretion in overruling CNW's objection. "As a general rule, the factual basis of an expert opinion goes to the credibility of the testimony, not the admissibility, and it is up to the opposing party to examine the factual basis for the opinion in cross-examination." Loudermill v. Dow Chem. Co., 863 F.2d 566, 570 (8th Cir.1988). See also Fed.R.Evid. 703.6 Only if an expert's opinion is "so fundamentally unsupported that it can offer no assistance to the jury" must such testimony be excluded. Loudermill, 863 F.2d at 570. Dr. Angle's opinion testimony clearly had sufficient factual basis. In addition to patient history, Dr. Angle relied on laboratory studies showing elevated levels of manganese in Hose's body and on his work clothes, her own clinical examinations of Hose showing physical impairment, a series of magnetic resonance images ("MRIs") suggesting the presence of manganese in Hose's brain, and reports from other doctors showing memory loss, sensory loss, slow cognition, and other ailments.

It is also significant that CNW had a thorough opportunity to cross-examine Dr. Angle about the physical symptoms in Hose that she actually observed. See Daubert, --- U.S. at ----, 113 S. Ct. at 2798 ("Vigorous cross-examination [is one] traditional and appropriate means of attacking shaky but admissible evidence."). However, Dr. Angle did not give CNW the answers it sought. On cross-examination, Dr. Angle testified that she personally observed that Hose had an "intention tremor" in his hands, a "mask-like face," and a weak hand grip, all of which Dr. Angle considered physical symptoms of manganese encephalopathy. (Tran. 572, 573, 610). Under cross-examination, Dr. Angle also agreed she did not fully trust the patient's history because Hose's wife already was "convinced" Hose had manganese poisoning after reading about manganese at a library.7 Finally, the district court instructed the jury to evaluate whether the plaintiff's "subjective symptoms" reported to the experts were accurate, see Jury Instruction 6, and the jury had a full opportunity to evaluate the credibility of Hose and his wife as they described his symptoms.

Much of what we have said about the admissibility of Dr. Angle's and Dr. Gupta's testimony also relates to CNW's challenge to the sufficiency of the medical evidence of causation. In addition, Dr. Kevin Nelson, a radiologist, testified to his opinion, based on the MRIs, that Hose had manganese in his brain, and he testified the MRIs excluded stroke, calcium, copper, and other explanations for Hose's injuries. Dr. Golnick testified that his initial examination of Hose suggested an encephalopathy, possibly of toxic origin, and that his ultimate diagnosis of Hose's condition was manganese encephalopathy. Dr. Thomas Korn, a neuropsychologist, testified Hose suffered from an encephalopathy, possibly toxic. We conclude there is more than sufficient evidence of medical causation to sustain the jury's verdict.8

CNW argues, however, it is not enough that certain experts expressed opinions favorable to Hose's position based on scientifically valid methods. CNW contends the plaintiff's experts have failed to rebut its experts' scientific understanding about the manifestations of manganese encephalopathy, and thus CNW claims it is entitled to judgment as a matter of law. See Daubert, --- U.S. at ----, 113 S. Ct. at 2798 (judgment as matter of law is proper where expert evidence, even if admissible, is insufficient). CNW presented expert evidence that manganese encephalopathy involves disabling physical symptoms, but no progressive deterioration, with clinically verifiable signs of Parkinson's disease affecting both sides of the body. In contrast, CNW argues Hose has no disabling physical symptoms, no physical symptoms on his left side, no signs of Parkinson's disease, but has appeared to deteriorate since 1991, and thus cannot be suffering from manganese encephalopathy.

The scientific evidence, including articles introduced by CNW on cross-examination, does not show that manganese encephalopathy must involve clinically-verifiable signs of Parkinson's disease, only that those signs are "usual," "the hallmark," "most characteristic," "similar," or "typical" in cases of manganese encephalopathy.9 (Tran. 567-68, 593, 1244, 2300). Both Dr. Angle and Dr. Snyder testified manganese encephalopathy involves symptoms that are broader than found in Parkinson's disease. Dr. Angle testified that, unlike Parkinson's disease, manganese patients suffer personality changes and emotional instability prior to suffering physical symptoms. Nonetheless, Dr. Angle also testified Hose had "episodic weakness" and "postural instability" consistent with Parkinson's disease. (Tran. 573).

The record also does not show that manganese encephalopathy must produce disabling physical symptoms substantially worse than Hose's. Although on cross-examination Dr. Angle agreed that people who are disabled due to manganese encephalopathy have disabling physical problems, earlier she explained that manganese encephalopathy is "usually associated with decrease in control of movements." (Tran. 491). The evidence of Hose's physical symptoms was clearly sufficient to show physical impairment in control over his movements. Pat Carrigan, Hose's supervisor at work, testified she directed Hose to take sick leave in part because his right leg collapsed at work. Dr. Jason Ohr, a neurologist who thought Hose had suffered a stroke but was admittedly uncertain, found clumsiness in Hose's right hand and diminished sensation on his right side. Both Dr. Riedler and Dr. Angle observed Hose having difficulty walking and saw tremors. Hose himself testified he has difficulty grasping things, falls easily, and "jerk[s] and tremble[s]." (Tran. 2060).

The evidence also does not support CNW's contention that the physical symptoms involved in manganese encephalopathy must affect both sides of the body in the same way. Dr. Nelson testified "symptoms can be variable" between the sides. (Tran. 388-89). Dr. Angle testified one-sided symptoms are "not typical, but it doesn't exclude manganese poisoning." (Tran. 614). Although the evidence shows most of Hose's symptoms on his right side, Dr. Angle testified Hose has a tremor in both hands, and Dr. Golnick found sensation deficiencies in Hose's left leg. Thus, the evidence shows Hose's physical symptoms are not "confined" to his right side as CNW so strenuously contends.

Finally, the evidence does not support the contention that manganese encephalopathy never manifests itself in progressive deterioration. Dr. Snyder testified there are patients suffering manganese-related health effects who deteriorate over time, and that scientific evidence to the contrary is limited to "one particular patient population." (Tran. 2300). Although Dr. Angle testified she did not order treatment for Hose initially, because she "expected he was stable," she also testified that "progressive stages ... are seen in most but certainly not all cases of manganese." (Tran. 590, 519).

Although it is common that medical experts often disagree on diagnosis and causation, questions of conflicting evidence must be left for the jury's determination. We find this true on the record in this case. We must uphold a jury verdict where the evidence supports a reasonable inference supporting the prevailing party's position. See United States v. Articles of Drug for Veterinary Use, 50 F.3d 497, 502 (8th Cir.1995); McKnight, 36 F.3d at 1400; Johnson Int'l Co. v. Jackson Nat'l Life Ins. Co., 19 F.3d 431, 435 (8th Cir.1994); Cashman v. Allied Prods. Corp., 761 F.2d 1250, 1253 (8th Cir.1985).

CNW next challenges the district court's exclusion of evidence of Hose's initial complaint making state law claims against Stoody and Arcair and their ultimate settlement. CNW sought to use Hose's complaint to show that Stoody and Arcair were at fault, but the district court excluded the evidence as unduly prejudicial. Instead, CNW was allowed to introduce evidence blaming Stoody and Arcair for Hose's injuries. In addition, the jury was instructed to evaluate whether Stoody and Arcair were responsible for Hose's injuries, and, at CNW's request, the verdict form provided a line for Stoody and Arcair's relative fault. Under these circumstances, we find no clear abuse of discretion by the district court. Cf. Garman v. Griffin, 666 F.2d 1156, 1158 (8th Cir.1981) (using pleading to show dismissed third party's fault was reversible error because pleadings often represent tentative, inconsistent, and alternative theories); id. at 1162-63 (Arnold, J., dissenting) (whether undue prejudice outweighs probative value should be left to trial court).

CNW also sought to introduce evidence of the amount of the settlement with Stoody and Arcair to show Hose had the financial ability to obtain treatment for his depression but had unreasonably failed to do so. CNW also contends it offered this evidence, as well, to rebut testimony that Hose had not obtained certain treatment because of his inability to pay.10 Virginia Hose and Dr. Golnick both testified Hose discontinued certain rehabilitative treatment during 1992 in part due to his inability to pay. On cross-examination, however, Hose's wife testified that Hose was currently receiving therapy. We find the district court's refusal to allow CNW to introduce evidence about the settlement amount with Stoody and Arcair was not an abuse of discretion.11

CNW challenges Jury Instructions 12 and 16 which, it claims, told the jury that CNW should be liable to Hose for injuries even if CNW had reasonably warned Hose about welding hazards and additional warnings would not have helped him avoid any hazards. Instruction 12 said in part that Hose proved negligence if "defendant was negligent ... [i]n failing to warn plaintiff of any hazardous effects of welding, grinding and/or air-arcing with or upon manganese rods and/or manganese-containing alloys and/or materials." Instruction 16 said:

Where there is a duty to warn, the employer is negligent when it fails to so warn, and it cannot escape responsibility for its failure, even though the danger is one which in the ordinary course of business cannot be avoided ... nor is it material how or by whom the dangerous conditions were created or that the employee could not have avoided the danger had he known of it.

A district court has " 'wide discretion' " in formulating instructions for the jury. Davis v. Merrill Lynch, Pierce, Fenner & Smith, Inc., 906 F.2d 1206, 1212 (8th Cir.1990) (quoting United States v. Ridinger, 805 F.2d 818, 821 (8th Cir.1986)). Our review is limited to "whether the instructions, when taken as a whole and viewed in light of the evidence and applicable law, fairly and adequately present the issues in the case to the jury." Id. "[A] single erroneous instruction [does not] necessarily require reversal if the error was cured by a subsequent instruction or by consideration of the entire charge." Id. We find the jury instructions in this case were within the district court's discretion.

As a whole, the instructions told the jury CNW had a duty to use "ordinary" or "reasonable" care. See, e.g., Jury Instructions 10, 13, 15. The fact that the court elaborated on the meaning of "negligence" in the factual context of this case in Jury Instruction 12 does not undermine the validity of the instructions as a whole. Moreover, we reject CNW's assertion that its duty was to warn Hose only where he could have altered his behavior by heeding the warnings. There are many reasons employees want to know about work place dangers aside from taking additional safety precautions. An employee may want more information in order to watch for symptoms of illness, advocate for work place safety improvements, or even seek other employment if the hazards are too great. We find no prejudicial error in the instructions.THE CROSS-APPEAL

On his cross-appeal, Hose challenges the sufficiency of the evidence to support the district court's instruction on his contributory negligence and the jury's ultimate finding of ten percent contributory negligence.

Under FELA, the burden of proving contributory negligence is on the defendant, who is entitled to a jury instruction on contributory negligence if there is any evidence to support that theory. Wilson v. Burlington N., Inc., 670 F.2d 780, 782 (8th Cir.), cert. denied, 457 U.S. 1120, 102 S. Ct. 2934, 73 L. Ed. 2d 1333 (1982). But "when there is no evidence from which a jury could reasonably find a lack of due care by the plaintiff, it is reversible error to submit the issue of contributory negligence to the jury." Id. at 784. It is well-settled that assumption of risk is not a proper defense under FELA, and thus evidence that relates solely to assumption of risk cannot constitute evidence of contributory negligence to support a jury verdict. See Birchem v. Burlington N. R.R. Co., 812 F.2d 1047, 1049 (8th Cir.1987). Evidence of contributory negligence cannot be excluded, however, "merely because the evidence ... may also be pertinent to assumption of risk." Beanland v. Chicago, Rock Island & Pac. R.R. Co., 480 F.2d 109, 116 n. 5 (8th Cir.1973).

Hose contends that to the extent the record shows any fault on his part, the evidence relates to assumption of risk (i.e., knowingly performing ordinary job duties in dangerous conditions) rather than contributory negligence. In Birchem, we held the railroad could not claim contributory negligence even though the employee knowingly operated defective equipment in violation of the railroad's safety rules, so long as the employee exercised due care in operating the equipment. 812 F.2d at 1049. We agreed with the district court that the railroad's contributory negligence argument was "an impermissible effort to transfer to [the employee] its non-delegable duty to provide safe equipment and a safe working environment." Id. Given the railroad's duty to provide safe equipment, we found that the employee did not " 'add new dangers to the conditions that the employer negligently created or permitted to exist' " by operating the defective machine. Id. (emphasis omitted) (quoting Taylor v. Burlington N. R.R. Co., 787 F.2d 1309, 1316 (9th Cir.1986)). Without adding new dangers to the already dangerous work conditions attributable to the employer, the employee was not contributorily negligent. Id.

Although we deem it a close question, we find CNW offered sufficient evidence that Hose did not consistently exercise due care for his own safety to support a finding of ten percent contributory negligence. CNW elicited testimony from Joyce Henderson, a clerical worker at the reclamation center, that Hose said he "didn't always feel it was necessary to use the exhaust equipment" provided by CNW. (Tran. 2484). Pat Carrigan, Hose's supervisor, testified she had to remind Hose "[a]t times" to use ventilation. (Tran. 780). Carrigan also testified that she talked with the welders in general to "make sure that they used the exhaust equipment." (Tran. 719). CNW also read part of Hose's deposition in which Hose admitted he may not have read warning labels. The jury may have reasonably inferred from this testimony that Hose occasionally failed to take advantage of the ventilation equipment and warnings (however inadequate) provided by CNW, which goes beyond accepting a dangerous condition as an ordinary job duty and constitutes the lack of due care by the employee.

Hose points out, however, that Carrigan's testimony shows CNW was aware its employees did not consistently use ventilation equipment and thus permitted the dangerous condition to exist. Cf. Wilson, 670 F.2d at 782-83 & n. 2 (no contributory negligence by employees where railroad was "aware" that employees had not been using the mechanical jack for over a year to lift 290-pound "pans" that attach to railroad hopper). Indeed, Carrigan's testimony shows that no employee was ever disciplined in any manner for failing to use ventilation equipment. Nonetheless, Carrigan's occasional verbal reminders belie the notion that CNW acquiesced in Hose's failure to use the safety equipment and read the warnings provided.

Hose also argues, however, that any evidence he failed to use the ventilation equipment provided by CNW can be explained by the fact that this equipment often did not work or did not effectively remove the fumes resulting from certain welding procedures. Hose testified he did not use suckers during certain welding procedures because "[i]t would burn up the tubing and it would also burn up the internal parts of the fan." (Tran. 2038). Rather than simply not use the equipment, as Hose did, Hose had a duty in such circumstances to tell his supervisors the equipment was ineffective. He did not do so.

Finally, Hose contends the railroad's negligent failure to warn Hose about manganese hazards should vitiate any contributory negligence on his part. Since he had not been warned about the extent of the risks associated with welding fumes, Hose argues, his failure to use ventilation equipment could not constitute contributory negligence. We do not think Hose should be able to avoid his own responsibility on this basis. CNW's failure to warn related to the scope of the warning provided, not the absence of any warning at all. Hose gave clear testimony that he knew he should use ventilation equipment and take other precautions to avoid breathing welding fumes. Thus, we conclude CNW's failure to warn specifically about manganese does not vitiate Hose's failure to abide by the admittedly inadequate warnings given.

For the reasons given, the judgment is affirmed.