Following a jury verdict in favor of appellee Troy Mattis, appellants Carlon Electrical Products, Lamson and Sessions, and Oatey Co. challenge the district court's evidentiary rulings and its refusal to grant judgment as a matter of law in appellants' favor. We affirm.

In the summer of 1995, Mattis was 25 years old and was working as an apprentice electrician. At the beginning of the summer, he had a well-documented history of good health.1 On July 13, Mattis worked on a project near Wagner, South Dakota and used Carlon All Weather Quick Set Cement ("Carlon cement"). Carlon cement is manufactured by Oatey Co., and it is labeled and sold by Carlon Electrical Products, a business unit of Lamson and Sessions. Carlon cement includes six ingredients, four of which are respiratory irritants: tetrahydrofuran, acetone, cyclohexanone, and methyl ethyl ketone.

On July 13, 1995, Mattis spent most of the day outside laying ten foot sections of three-quarter inch polyvinyl chloride (pvc) pipe in a trench that was six inches wide and eighteen inches deep. The temperature was 109 degrees. After laying the sections of pipe, he worked on his hands and knees using Carlon cement to connect the sections. Before performing this work, Mattis read the safety warnings on the label of the Carlon cement.2 When Mattis was not using the glue, he put the lid back on the can according to his supervisor's instructions. The can sat in the sun when not in use. To reopen the can of cement, Mattis held it close to his chest and opened the lid. As he did this, the can made sizzling sounds, like a can of soda-pop. He testified that he opened the can at least ten to twelve times during the course of the day.

On the morning of July 14, 1995, Mattis woke up with a severe headache, nausea, vomiting, and soreness in his chest. Nevertheless, he returned to work and again used the Carlon cement. On Saturday morning, July 15, Mattis awoke with another severe headache, continued nausea, and tightness in his chest. His symptoms were worsening, and it was difficult for him to breathe. On Sunday, he went to a doctor and was treated for dehydration and released. He called in sick on Monday and went to see another doctor on Tuesday. His physician, Dr. Weber, admitted him to the hospital, where he stayed for six days. X-rays of his chest were taken on July 18, 1995, and again on July 19, 1995. The board certified radiologist, Dr. Frank Messner, who read the x-ray concluded:

The patient demonstrates a rather extensive bilateral infiltrate, which has a somewhat nodular component to it.... The distribution would suggest some type of unusual organism or cause.... IMPRESSION. Fairly extensive bilateral infiltrates suspicious for unusual pneumonia or some type of response to a noxious agent.

Appellee's App. at 4. Dr. Weber diagnosed Mattis with bilateral pneumonia, reactive airways disease, and exposure to PVC glue fumes, among other things. Dr. Weber referred Mattis to Dr. Hansen, a board certified pulmonologist. She first met with Mattis on August 18, 1995. At that time, she performed a differential diagnosis using a specific medical methodology for reactive airways dysfunction syndrome (RADS). After performing the differential diagnosis, Dr. Hansen diagnosed Mattis with RADS and concluded that it was caused by his inhalation of the Carlon cement fumes.

Since his release from the hospital, Mattis has experienced persistent intermittent symptoms and some decline in his lung function. He must avoid exposure to irritants such as smoke, dust, chemical odors, and extremes in temperatures. Because of his health problems and work restrictions, he gave up working as an electrician.

Mattis and his wife brought a cause of action in federal court against Oatey, Carlon, and Lamson and Sessions claiming negligence, strict liability based on failure to warn, and negligence per se. The jury returned a verdict in favor of Mattis in the amount of $600,000 and in favor of his wife for $300,000 on her loss of consortium claim. Following the verdict, the defendants filed a renewed motion for judgment as a matter of law and a motion for a new trial. The district court denied these motions and Carlon, Oatey, and Lamson and Sessions appeal.

Appellants raise four issues on appeal. They contend that the district court erred in refusing to grant judgment as a matter of law in their favor because: (1) Mattis failed to establish causation; and (2) he failed to establish liability on the failure to warn claim. Appellants also maintain that the district court erred in excluding the results of field studies Oatey conducted after Mattis's injury, and that the court erred in admitting reports of health problems caused by Carlon cement.

We first address appellants' argument that the district court should have granted judgment as a matter of law in their favor because the expert testimony offered by Mattis failed to prove that his exposure to Carlon cement caused his illness. We review de novo the district court's denial of a motion for judgment as a matter of law and view the facts in the light most favorable to the nonmoving party. Cardenas v. AT & T Corp., 245 F.3d 994, 998 (8th Cir.2001) (citation omitted). Judgment as a matter of law is only appropriate when no reasonable jury could have found for the nonmoving party. Id. (citation omitted).

"To prove causation in a toxic tort case, a plaintiff must show both that the alleged toxin is capable of causing injuries like that suffered by the plaintiff in human beings subjected to the same level of exposure as the plaintiff, and that the toxin was the cause of the plaintiff's injury." Bonner v. ISP Techs., 259 F.3d 924, 928 (8th Cir.2001) (citing Wright v. Willamette Indus., Inc., 91 F.3d 1105, 1106 (8th Cir. 1996)). To meet his burden of proving causation, Mattis presented the testimony of two expert witnesses, Dr. Hansen, his treating pulmonologist, and Roger Wabeke, an industrial hygienist. The district court found that this testimony, in addition to the testimony by appellants' expert, Dr. Kapp, was sufficient for a reasonable jury to find that Mattis's exposure to the organic solvents in Carlon cement was capable of causing RADS and that exposure to those solvents did, in fact, cause his illness. We agree. Dr. Kapp admitted that the organic solvents in Carlon cement were capable of causing RADS at high exposure levels. Wabeke's testimony established that Mattis was exposed to dangerous levels of those organic solvents, and Dr. Hansen's testimony provided evidence that Mattis's exposure to the organic solvents in the cement caused him to develop RADS.

Appellants argue, however, that Dr. Hansen's and Wabeke's testimony was insufficient as a matter of law to establish causation. They fault Wabeke's testimony because he could not determine Mattis's exact exposure level. To prove exposure levels, plaintiffs need not produce a "`mathematically precise table equating levels of exposure with levels of harm.'" Bednar v. Bassett Furniture Mfg. Co., 147 F.3d 737, 740 (8th Cir.1998) (quoting Wright, 91 F.3d at 1107). Rather, a plaintiff need only make a threshold showing that he or she was exposed to toxic levels known to cause the type of injuries he or she suffered. Id. Wabeke testified that experts have known for a long time that the organic solvents in Carlon cement are respiratory irritants capable of injuring respiratory mucous membranes in the nostrils, throat, trachea, and lungs. In addition, Wabeke used a vapor concentration test to determine whether Mattis was exposed to a dangerous level of fumes. To perform the test,3 he heated a can of Carlon cement to 112 degrees and cut PVC pipe into five inch sections. He then transferred the pipe and the cement to an air sampling chamber and spread the cement on the pipe. The test showed that the solvent vapors accumulated rapidly at extreme concentrations, far in excess of safe exposure levels. At trial, Wabeke noted that Mattis's exposure levels would have been higher than the levels in his test because the hissing and sizzling sounds the can made indicated that it was hotter than 112 degrees. This evidence is admissible and created a question of fact for the jury about whether Mattis was exposed to an unsafe level of fumes, capable of causing respiratory problems.

Next, appellants argue that Dr. Hansen's testimony was legally insufficient. We disagree. A medical opinion based upon a proper differential diagnosis is sufficiently reliable to satisfy Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579, 593-94, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993). Turner v. Iowa Fire Equip., Co., 229 F.3d 1202, 1208 (8th Cir.2000). Dr. Hansen performed a valid differential diagnosis to determine that Mattis's exposure to the cement caused his RADS. Cf. id. (noting that a proper differential diagnosis is one that identifies the cause of a medical condition by eliminating the likely causes until the most probable cause is isolated). While treating Mattis, Dr. Hansen ruled out other possible causes of Mattis's illness, such as smoking, asthma, or ammonia, and concluded that he developed RADS as a result of his exposure to Carlon cement fumes. She also relied on published studies linking RADS to organic solvents like those in Carlon cement. This testimony was properly admitted at trial and by relying on it, a reasonable jury could find that exposure to Carlon cement was the cause of Mattis's injury.

Appellants also fault Dr. Hansen's testimony because she could not identify Mattis's exposure level. As stated, exposures levels were adequately proven by Wabeke's testimony. The district court properly admitted the testimony of Wabeke and Dr. Hansen and properly held that there was sufficient evidence from which a reasonable jury could find that fumes from the Carlon cement caused Mattis's injury.

Next, appellants complain that the district court erred in refusing to grant their motion for judgment as a matter of law on Mattis's failure to warn claim. They argue that Mattis failed to prove a violation of the Federal Hazardous Substances Act (FHSA) and, therefore, cannot prevail. The FHSA was enacted in 1960 to "`provide nationally uniform requirements for adequate cautionary labeling of packages of hazardous substances which are sold in interstate commerce and are intended or suitable for household use.'" Milanese v. Rust-Oleum Corp., 244 F.3d 104, 109 (2d Cir.2001) (quoting House Comm. on Interstate and Foreign Commerce, Federal Hazardous Substances Labeling Act, H.R.Rep. No. 1861, 86th Cong., 2d Sess. 2 (1960), reprinted in 1960 U.S.C.C.A.N. 2833, 2833). As enacted, the FHSA did not mention federal preemption, but the 1966 Amendments to the Act added a provision to preempt any state cause of action that seeks to impose a labeling requirement different from the requirements in the FHSA or the regulations promulgated thereunder. See 15 U.S.C. § 1261 note (b)(1)(A) ("no state .. may establish or continue in effect a cautionary labeling requirement applicable to such substance or packaging and designed to protect against the same risk of illness unless such cautionary labeling requirement is identical to the labeling requirement under [this Act]"); see also Moss v. Parks Corp., 985 F.2d 736, 739 (4th Cir. 1993). Other circuits interpreting this provision have held that a plaintiff may not bring a claim for failure to warn based on state requirements that are more elaborate than the FHSA; however, a common law tort action based on failure to warn may be brought for noncompliance with the federal labeling requirements. See e.g. Milanese, 244 F.3d at 109-110; Moss, 985 F.2d at 740. Similarly, our circuit has held that "when a statute only preempts state requirements that are different from or in addition to those imposed by federal law, plaintiffs may still recover under state tort law when defendants fail to comply with the federal requirements." National Bank of Commerce of El Dorado v. Kimberly-Clark Corp., 38 F.3d 988, 993 (8th Cir.1994). Therefore, in order to prevail on his failure to warn claim, Mattis had the burden of proving the Carlon cement label did not comply with the FHSA.

The FHSA requires that a warning label contain, among other things, the following information: "(E) an affirmative statement of the principal hazard or hazards, such as ... `Vapor Harmful' ...; (F) precautionary measures describing the action to be followed or avoided ...; (I) instructions for handling and storage ...; and (J) the statement (i) `Keep out of the reach of children.'" 15 U.S.C. § 1261(p)(1). The district court held that there was a question of fact for the jury about whether these requirements were met because the label did not state "[k]eep out of the reach of children" and because the evidence about the harmful effects of Carlon cement called into question whether the label's statements about principal hazards, precautionary measures, or instructions for handling were inadequate. We agree. Although the label stated "vapor harmful," this warning was followed by the statements, "may irritate eyes and skin" and "vapors may cause flash fires." The label does not make it clear that inhalation of the vapors is harmful. The label did not state handling instructions or specify any precautionary measures regarding inhalation of fumes from the cement other than to say, "if inhaled, get fresh air." Moreover, the jury was properly instructed that if the label contained all the information required by the FHSA, the appellants were not negligent. Neither party objected to this instruction.4 We find that the court properly refused to grant judgment as a matter of law on this issue.

Appellants also argue that even if the label was deficient under the FHSA, those deficiencies were not the proximate cause of Mattis's injury. A deficient warning cannot be considered the cause of a user's injury when the user was fully aware of the danger, see Jaurequi v. Carter Mfg. Co., 173 F.3d 1076, 1084-85 (8th Cir.1999), but that rule cannot be applied to the case at hand. Mattis was not an experienced electrician who should have been aware of the dangers associated with breathing fumes from hot cement; rather, he was an apprentice electrician. He testified that he read the label and abided by its instruction not to get the cement on his skin or in his eyes. He further testified that had the label warned him not to inhale the fumes, he would not have held the can in his breathing zone to open it. This evidence was sufficient for a reasonable jury to find that the deficient label was the proximate cause of Mattis's injuries.

Appellants also appeal the district court's denial of their motion for a new trial on the grounds that the district court erroneously excluded field monitoring studies performed by the appellants and erroneously admitted evidence concerning health complaints related to Carlon cement. When a motion for new trial is based on rulings regarding the admissibility of evidence, the district court will not be reversed absent a clear and prejudicial abuse of discretion. First Sec. Bank v. Union Pac. R.R. Co., 152 F.3d 877, 879 (8th Cir.1998) (citation omitted). The field monitoring studies in question were performed after Mattis's exposure. They measured the air quality that two California construction workers breathed while they were working with another Carlon cement product in trenches outside in 70 degree weather with wind. The district court excluded the studies because they were performed under significantly different working conditions and because they used different product types. The district court concluded that the studies were not substantially similar to Mattis's exposure, but they were "confusingly similar." We agree and find no abuse of discretion by the district court.

Appellants also argue that the district court abused its discretion in admitting reports from the Rocky Mountain Poison Control Center. The Center received telephone calls regarding questions or problems relating to consumers' exposure to Carlon cement. The district court admitted the Center's reports because they proved that the organic solvents in Carlon cement were respiratory irritants and because the reports were unlikely to confuse the jury. The district court also noted, however, that the reports could not be used to prove causation. This ruling was not an abuse of discretion.

For the reasons stated above, we affirm the district court's decision.