PLURALITY OPINION Appellants, Ronda Neely, Michelle Patrick, and Royce Goehring, the adult children of decedent Janet Goehring, brought this wrongful death action against appellee Union Carbide Corporation, alleging that asbestos fibers designed, manufactured, or marketed by Union Carbide were a substantial factor in causing Janet’s fatal mesothelioma. The trial court granted Union Carbide’s motion for summary judgment, which asserted that appellants could produce no evidence that a Union Carbide product was a substantial factor in causing Janet’s mesothelioma. In a single issue, appellants now challenge that ruling on appeal. We affirm. Governing Law No-evidence Summary Judgment To defeat a no-evidence motion for summary judgment, the responding party must present evidence raising a genuine issue of material fact supporting each element contested in the motion. Timpte Indus., Inc. v. Gish, 286 S.W.3d 306, 310 (Tex. 2009). When reviewing a trial court’s grant of such a motion, we consider the evidence presented in the light most favorable to the party against whom summary judgment was rendered, crediting evidence favorable to that party if reasonable jurors could and disregarding contrary evidence unless reasonable jurors could not. Id. We indulge every reasonable inference and resolve any doubts in the nonmovant’s favor. Cantey Hanger, LLP v. Byrd, 467 S.W.3d 477, 481 (Tex. 2015). We review a no-evidence summary judgment de novo. See Joe v. Two Thirty Nine Joint Venture, 145 S.W.3d 150, 156–57 (Tex. 2004). A no-evidence summary judgment is improperly granted if the respondent presents more than a scintilla of probative evidence to raise a genuine issue of material fact on each challenged element. King Ranch, Inc. v. Chapman, 118 S.W.3d 742, 751 (Tex. 2003). More than a scintilla of evidence exists when the evidence “rises to a level that would enable reasonable and fair-minded people to differ in their conclusions.” Merrell Dow Pharms., Inc. v. Havner, 953 S.W.2d 706, 711 (Tex. 1997). Toxic Tort Causation Causation in toxic tort cases is often discussed in terms of general and specific causation. See, e.g., id. at 714. “General causation is whether a substance is capable of causing a particular injury or condition in the general population, while specific causation is whether a substance caused a particular individual’s injury.” Id. In its motion for summary judgment, Union Carbide challenged that appellants could not provide evidence establishing specific causation—whether asbestos fibers designed, manufactured, or marketed by Union Carbide were a substantial factor in causing Janet’s mesothelioma. Union Carbide does not raise a general causation challenge. As the parties acknowledge, the proper framework for analysis of specific causation here is set forth in two Texas Supreme Court opinions: Borg–Warner Corp. v. Flores, 232 S.W.3d 765 (Tex. 2007) (an asbestosis case), and Bostic v. Georgia-Pacific Corp., 439 S.W.3d 332 (Tex. 2014) (a mesothelioma case). In Flores, the supreme court considered the proper standards for reviewing the sufficiency of the evidence on specific causation in asbestosis cases. The court held that to establish causation in fact against a particular defendant, a plaintiff must prove that the defendant’s product was a substantial factor in causing the plaintiff’s asbestosis, and mere proof that the plaintiff was exposed to some asbestos fibers traceable to the defendant is insufficient. Flores, 232 S.W.3d at 766, 770. In defining the word “substantial” in this context, the court explained that it “denote[s] the fact that the defendant’s conduct ha[d] such an effect in producing the harm as to lead reasonable men to regard it as a cause, using that word in the popular sense, in which there always lurks the idea of responsibility.” Id. at 770 (quoting Restatement (Second) of Torts § 431 cmt. a (1965)). The Flores court also explained that while evidence of the frequency, regularity, and proximity of exposure is necessary to establish causation, it is not sufficient absent quantitative evidence about the dose or level of exposure. Id. at 772.[1] While a plaintiff need not establish causation with “mathematical precision,” a plaintiff must produce “[d]efendant-specific evidence relating to the approximate dose to which the plaintiff was exposed, coupled with evidence that the dose was a substantial factor in causing the asbestos-related disease.” Id. at 773. Regarding the use of epidemiological studies as proof, the court reiterated prior guidance that such studies can be considered in toxic tort cases so long as they are properly designed and executed and the plaintiff can show that his or her exposure or dose level was comparable to or greater than those in the studies. Id. 771-72 (citing Havner, 953 S.W.2d at 720-21). The court further noted that to be useful, such studies must show the exposure level in question at least doubled the risk of contracting the disease. Id. at 772 (citing Havner, 953 S.W.2d at 715, 717-18). The court ultimately held the evidence was insufficient in Flores because the plaintiff presented no evidence of the approximate quantum of the particular defendant’s fibers to which he had been exposed or any evidence that such exposure sufficiently contributed to the aggregate dose of asbestos the plaintiff inhaled, such that it could be considered a substantial factor in causing his asbestosis. Id. at 772. The court further noted the lack of epidemiological studies in evidence showing at least a doubling of the risk. Id. While recognizing that mesothelioma can be caused by a much lower level of exposure to asbestos than can asbestosis, the supreme court in Bostic extended and amplified the standards established in Flores for use in the mesothelioma context. Bostic, 439 S.W.3d. at 338. In doing so, the court noted that both diseases are dose-related, meaning that the risk of contracting each disease rises along with the level of exposure. Id. at 338-39. The court therefore again rejected the “any exposure theory” in favor of the substantial factor test. Id. Like the court in Flores, the Bostic court also emphasized the role of epidemiological studies and other scientific evidence in toxic tort cases. “Where direct evidence of specific causation is unavailable, specific causation may be established through an alternative two-step process whereby the plaintiff establishes general causation through reliable studies, and then demonstrates that his circumstances are similar to the subjects of the studies.” Id. at 351. In other words, proof of a doubling of the risk must “be shown through reliable expert testimony that is based on epidemiological studies or similarly reliable scientific testimony.” Id. at 350. As to what makes an epidemiological study reliable, the court noted that the confidence level must be at least 95 percent, the confidence interval must not include the number 1, and the study should be peer reviewed or at least published. Id. at 347 (citing Havner, 953 S.W.2d at 713-727).[2] The court explained that the doubling of the risk requirement goes toward the preponderance of the evidence burden of proof and “strikes a balance between the needs of our legal system and the limits of science.” Id. at 349 (citing Havner, 953 S.W.2d at 718). The Bostic court further emphasized that in cases involving exposure from multiple sources, proof of a doubling of the risk from a particular defendant’s product alone may not suffice to establish substantial factor causation. Id. at 350. The plaintiff in such cases also may need to present evidence regarding his or her aggregate exposure to asbestos. Id. at 351, 353. In a hypothetical, the court suggested that a defendant’s product may not be considered a substantial factor in causing a plaintiff’s disease even when the plaintiff’s risk was more than doubled by that exposure if other exposures increased the risk by a factor of 10,000. Id. at 352. The test is whether reasonable persons would regard the defendant’s product as a cause of the disease. Id. at 353. The court held that the evidence of causation in Bostic was legally insufficient, specifically noting that the plaintiff’s expert on specific causation urged that any exposure to asbestos was a contributing cause of plaintiff’s mesothelioma. Id. at 354-55. The plaintiff made no effort to quantify his aggregate exposure even though evidence established he had other exposures than just to the defendant’s product and his experts acknowledged that mesothelioma is “dose- related.” Id. at 355, 359. The court also found the epidemiological studies on which the plaintiff relied wanting, as they did not establish a statistically significant link between mesothelioma and the type of exposure the plaintiff claimed or support the plaintiff’s position that any exposure would be a cause in fact. Id. at 356-58. The Summary Judgment Evidence In their response to Union Carbide’s motion, appellants offered several pieces of evidence, including Janet’s medical records; the depositions of all three of Janet’s children (Ronda, Michelle, and Royce); Royce’s affidavit; and an expert affidavit from Dr. Richard Cohen. The plaintiffs posited that Janet contracted mesothelioma because of exposure to asbestos fibers designed, manufactured, or marketed by Union Carbide when she washed Royce’s clothing during periods in which he worked with products containing those fibers. Royce’s Affidavit and Deposition Royce explained that he had worked with and around asbestos drilling mud additives for periods of time while living with his mother from 1975 to 1979 and in 1986. From 1975 to 1977, Royce had a summer job unloading 50-pound bags of drilling mud additives from freight cars onto a flatbed truck. He identified the bags as including bags of six different products: Flosal, Visquik, Visbestos, IMCO Superbest, IMCO Shurlift, and Super Visbestos.[3] Royce said that the freight cars also contained bags of additives that were not asbestos but the majority were asbestos additives. In fact, he said most of the bags in the freight cars were of those six products. However, he did not recall what percentage each product was of the whole number of bags because he did not pay a lot of attention to the bags. Indeed, he acknowledged he did not recall the specific names of the products until his lawyer showed him photographs. Royce described the work as very dusty, with several busted bags in each car allowing the asbestos powder to cover the other bags, fill the air inside the car, coat his hair, shirt, and jeans, and even get inside his pockets. Royce said that they generally got called in two or three times a week, would unload for a day, day and a half, or so, and then would get a day or two off in between. He said that the work was steady throughout the summers. In 1978, and again for a short period in 1986, Royce worked for Flournoy Drilling Company as a floor hand on oil drilling rigs. Each day, one of his job duties entailed tearing open bags of asbestos drilling mud additives and pouring the contents of the bags into the rig’s mud system. He recalled using a significant amount of the same six products mentioned above while on the rigs in 1978, explaining that certain procedures might require “anywhere from 15 to 200 bags” of the six products. He particularly recalled using a lot of Flosal in 1978. When he returned to Flournoy for a short period in 1986, Royce recalled using only Super Visbestos. He explained that sometimes he would add the products into a hopper and other times he had to carry the bags up a flight of stairs to pour the products directly into the mud tank. The additives were usually stored in a ten-foot by twenty-foot “mud house,” and Royce’s clothes would become covered in the asbestos powder. During his shift, Royce also would be responsible for cleaning up used bags by flattening and stacking them, a process that generated even more dust. He said that everyone on the rigs complained about the dust. Beginning in late 1978, Royce went to work as a warehouseman for Mutual Supply & Rental. Asbestos mud additives were stored in the warehouse, and Royce’s jobs included moving and stacking the products, sweeping the floors daily, and delivering supplies to drilling rigs. He recalled the same six products as being ones stored at the warehouse. Again, he stated this was a dusty job as bags often would break open and he would get asbestos dust on his clothes daily. Royce also stated that throughout each of these periods, Janet would wash his work clothes two or three times a week in a small laundry room. She first would shake the asbestos-covered clothes out, which “would create a huge dust cloud in the laundry room.” She also would turn the clothes upside down and empty the powder from the pockets. Janet often would comment to Royce about how dusty the clothes were. Once the clothes were in the washer, Janet would sweep the powder from the floor, which would cause some of the powder to become airborne again. She might be required to perform this routine more than once on the days she washed Royce’s work clothes. She also swept the room on other days once the dust had had a chance to settle to the floor. Ronda’s and Michelle’s Depositions Ronda testified that Janet would complain about how dusty Royce’s clothes were and that it was ruining her washing machine. Ronda also described the small laundry room and stated she saw Janet washing Royce’s work clothes on multiple occasions, shaking out the clothes and emptying dust from the pockets. Ronda also testified that her father had worked as a parts salesman and possibly went to oil field locations on occasion, but his clothes were rarely dirty. In her deposition, Michelle also recalled Janet washing Royce’s work clothes. Michelle said that the clothes were “very, very dirty” and left dust all over the place, and her mom would tell her to leave the room and stay out because it was so dusty in there. Janet washed Royce’s work clothes separately because she did not want to mix them with the rest of the family’s clothes. Michelle also said that after her parents divorced, Janet lived for a time with a boyfriend who was a welder. As to Janet’s work history, Michelle noted that at different times, Janet had worked for a daycare, a car dealership, a law firm, a drilling company, and the Houston Grand Opera. Although Michelle was not able to provide much information about what Janet did for these employers, Michelle stated that for several Janet worked in the employer’s offices or did administrative work. Janet’s medical records reflected that she once told her doctors that her parents had worked for a steel mill, but there is no evidence in the record about what they did at the mill. Expert’s Affidavit Cohen stated that he is board certified in General Preventative Medicine and Occupational Medicine and has a graduate degree in public health, specializing in epidemiology. He has engaged in private practice in occupational medicine and industrial toxicology for over thirty years and has been a clinical professor in occupational and environmental medicine since 1998 at the University of California San Francisco School of Medicine. Cohen reported that he has reviewed medical literature on asbestos and industrial health and safety relating to asbestos “throughout the decades.” In an exhibit to his affidavit, Cohen listed the literature on which he relied in forming his opinions, stating that the studies and articles were all either peer-reviewed published studies or “generally accepted and reliable statements and standards issued by governmental entities and other medical/public health organizations.” Cohen attached three epidemiological studies to his affidavit as exhibits. Further, for purposes of this case, Cohen reviewed Janet’s death certificate and medical records, the appellants’ depositions and Royce’s affidavit, Royce’s social security records, and an expert report by Dr. James Bruce that does not appear in the record. Cohen notes that Janet was diagnosed with mesothelioma in 2005 and her death certificate from 2006 lists mesothelioma as the underlying cause of death. Cohen then recounts Royce’s and Janet’s exposure to asbestos drilling mud additives and draws conclusions about Janet’s risk of contracting mesothelioma through that exposure. Cohen highlights details from Royce’s deposition and affidavit about the work he did with the asbestos additives and concludes that Royce “encountered heavy, significant and repeated exposure to airborne asbestos and clothing contamination.” Cohen supported these conclusions with material arranged in three charts, listing scientific studies and other reference materials that he identified as relating to asbestos exposure from dumping, sweeping and cleanup, and loading and handling. The charts show exposure levels for those tasks in terms of fibers of asbestos per cubic centimeter of air breathed.[4] As for Janet’s exposure, Cohen recounts the testimony that she washed Royce’s work clothes two or three times a week during the relevant time periods in a small laundry room, shook out the clothes first, and swept up the resulting dust. Cohen provided another chart that he represents shows “documented asbestos exposures associated with laundering asbestos contaminated work clothing.” Cohen also compared Janet’s exposure to that found for laundering exposures in a laboratory simulation. In the simulation, the eight-hour exposure rate was determined to be an average of 0.13 fibers per centimeter of air breathed. Cohen asserted, however, that due to Janet working in a much smaller space than used in the simulation and there being no evidence of exhaust ventilation in the laundry room as opposed to the regular air exchanges recorded for the simulation, Janet’s exposure from laundering would be significantly higher at 1.49 fibers per centimeter of air breathed (an over elevenfold increase in the exposure calculated for the laboratory simulation). Cohen also noted Janet had additional exposure from sweeping up the asbestos dust, which, based on studies of similar exposures, he calculated as an additional 1 fiber per centimeter of air breathed, for a total average exposure of approximately 2.5 fibers per cubic centimeter of air breathed. To determine Janet’s cumulative exposure level from laundering Royce’s work clothes, Cohen calculated that she laundered Royce’s asbestos-covered work clothes for eight hours a day, 2.5 days a week, over a total of 2.1 years, with an average exposure level of 2.5 fibers per centimeter. This calculus resulted in an aggregate or cumulative exposure of 2.63 fiber years per cubic centimeter or “2.63 fiber-cc/years.” Cohen further explained that “mesothelioma is a low dose disease, with no known threshold of asbestos exposure below which there is no risk. Even an exposure below 0.2 fiber-cc/years can be a substantial factor in the development of the disease.” Cohen supported these assertions with a brief discussion of studies in which low exposures were found to have caused mesothelioma. He also noted that “[s]ince 1965, it has been repeatedly and consistently demonstrated in the medical and scientific literature that family members whose sole asbestos exposure resulted from asbestos dust brought into the home on another family member’s clothing have developed mesothelioma.” Cohen, however, did not cite any specific studies as supporting this statement. Cohen concluded that Janet’s exposures, as described by her children, “were substantial contributing factors for increasing [her] risk for asbestos-related disease” and, in fact, “more than doubled her risk of developing mesothelioma.” He supported these conclusions by reference to epidemiological studies that he says, “show a greater than doubling of the risk of malignant mesothelioma with cumulative asbestos exposures well below the 2.63 f-cc/years experienced by Janet.” The studies he references were listed in a fifth chart attached to his affidavit and cited in an attached bibliography of supporting materials. Cohen lists and discusses studies showing more than a doubling of the risk of mesothelioma from lifetime exposure at or even below 0.1 f-cc/years.[5] Analysis In its motion, Union Carbide asserted that appellants could produce no evidence that a Union Carbide product was a substantial factor in causing Janet’s mesothelioma and, more specifically, that appellants “lack evidence of the dose of asbestos attributable to Union Carbide to which Mrs. Goehring was allegedly exposed and have no scientifically reliable expert testimony or epidemiological evidence demonstrating that such dose more than doubled her risk of developing mesothelioma.” We begin by discussing the evidence on Janet’s exposure to Union Carbide products and then we will turn to the scientific evidence appellants presented. A. Evidence of Exposure The Union Carbide Products As an initial matter, the parties dispute which of the six products Royce identified by name as ones he worked with were actually shown to be Union Carbide products. Evidence, of course, showed that Janet was exposed to the same products as Royce when she washed his work clothes. Royce mentioned six asbestos drilling mud additives by name: Flosal, Visquik, Visbestos, IMCO Superbest, IMCO Shurlift, and Super Visbestos. Evidence showed that Flosal was produced by other defendants that are no longer in the lawsuit. Union Carbide does not contest that appellants timely presented evidence that Visbestos, IMCO Superbest, IMCO Shurlift, and Super Visbestos were Union Carbide products. Union Carbide asserts, however, that the evidence submitted to show that Union Carbide produced Visquik (i.e., Union Carbide’s answers to interrogatories) was untimely filed with the trial court and therefore should not be considered on appeal. Under the rules governing summary judgments, except with leave of court, a nonmovant must file responsive evidence not later than seven days before the hearing on the motion. Tex. R. Civ. P. 166a(c). Consequently, unless the record indicates the trial court granted leave or actually considered the evidence, we generally presume the trial court did not consider untimely filed summary judgment evidence. See, e.g., Benchmark Bank v. Crowder, 919 S.W.2d 657, 663 (Tex. 1996). Such an indication may arise from a recital in the summary judgment order, a separate written order, or an oral statement during the summary judgment hearing. B.C. v. Steak N Shake Operations, Inc., 598 S.W.3d 256, 259-60 (Tex. 2020). Here, appellants point to language in the summary judgment order in which the trial court stated that “having heard the Motion, any Response, and arguments of counsel, if any, [the court] is of the opinion that the Motion is meritorious and should in all things be GRANTED.” Appellants submit that in stating it “heard . . . any Response,” the trial court indicated it considered the late-filed evidence, which was attached to a supplemental response to the motion for summary judgment. We agree. In B.C., the supreme court cited with approval Stavron v. SureTec Insurance Co., No. 02-19-00125-CV, 2019 WL 6768125, at *6 (Tex. App.—Fort Worth Dec. 12, 2019, no pet.) (mem. op.), in which the court of appeals found similar language in a summary judgment order to be an indication that the trial court granted leave for the late-filed evidence. 598 S.W.3d at 261 n.27. As the Stavron court noted, “‘Courts of appeals have found that late-filed summary judgment evidence was considered by a trial court when . . . an order granting summary judgment stated that it had considered the response’ to which that evidence was attached.” 2019 WL 6768125, at *6 (quoting Foussadier v. Triple B Servs., LLP, No. 01-18-00106- CV, 2019 WL 2127604, at *4 (Tex. App.—Houston [1st Dist.] May 16, 2019, pet. filed) (mem. op.)). Accordingly, the evidence that Union Carbide produced Visquik is part of the summary judgment evidence and reviewable on appeal. Union Carbide additionally suggests that there is no evidence Royce used Visquik made with Union Carbide asbestos as opposed to another company’s asbestos. In its answers to interrogatories, Union Carbide acknowledged that it “packaged . . . very small amounts [of Calidria asbestos] for Dresser-Magcobar under the trade name Visquik.” Union Carbide neither explains what it meant by “very small amounts” nor points to any evidence that Visquik was made using asbestos from any other manufacturer or, if so, what percentage was made by Union Carbide. As stated above, in the summary judgment context, we indulge every reasonable inference and resolve any doubts in the nonmovant’s favor. See Cantey Hanger, LLP, 467 S.W.3d at 481. In the absence of evidence to the contrary, it is reasonable to infer from Union Carbide’s answers to interrogatories that it was the only producer to package Visquik and that it packaged the Visquik with which Royce worked. Parsing the Evidence of Exposure to Union Carbide Products Appellants presented significant evidence of Janet’s exposure to asbestos drilling mud additives including ones produced by Union Carbide. As discussed in detail above, Royce identified five Union Carbide asbestos drilling mud additives as ones he frequently worked with (1) during a summer job in the years 1975 to 1977 unloading 50-pound bags from freight cars, (2) while working on drilling rigs in 1978 and 1986, and (3) when working as a warehouseman beginning in late 1978. Royce explained in detail how his work clothes would become coated with asbestos dust from these products each day, and Royce and his sisters testified that during each of these stints, Janet regularly washed Royce’s work clothes, first shaking the dust out of the clothes in a small laundry room and then sweeping it up. This evidence established Janet’s exposure to specific Union Carbide products on a regular basis over an extended period of time in close proximity to where she was working. In other words, this evidence suffices to pass the frequency, regularity, and proximity test. See Flores, 232 S.W.3d at 769-70 (citing Lohrmann, 782 F.2d at 1162–63). As explained above, however, Texas jurisprudence requires more than that; it requires quantitative evidence regarding the level of exposure, i.e., “[d]efendant- specific evidence relating to the approximate dose to which the plaintiff was exposed, coupled with evidence that the dose was a substantial factor in causing the asbestos-related disease.” Id. at 772. And in cases involving exposure from multiple sources, a plaintiff also must produce evidence of aggregate or cumulative exposure and the ratio or percentage of that aggregate exposure attributable to the specific defendant’s products. See Bostic, 439 S.W.3d at 350; Flores, 232 S.W.3d at 772-73. Union Carbide argues that appellants’ evidence was lacking in that it did not specifically quantify Janet’s exposure to asbestos from Union Carbide products—the dose amount—and failed to account for the ratio of that exposure to Janet’s aggregate or cumulative exposure to asbestos. Union Carbide thereby challenges both the specific factual statements Royce made regarding Janet’s exposure level and the calculations of their expert, Cohen, based on those statements. According to Union Carbide, the problem with Royce’s testimony lies in his spotty recounting of the details of the asbestos products with which he worked. As Union Carbide emphasizes, Royce stated that he worked with asbestos drilling mud additives “including” the six he identified by name (five of which were Union Carbide products) but did not testify that he worked only with those six asbestos products. Royce also acknowledged that he remembered those specific products only once his lawyers showed him photographs of bags of those products. Union Carbide also points out that Royce did not offer even an estimate on how much of each product he handled. It is certainly understandable that Royce would lack total recall of the products and the amounts of each product he worked with decades earlier. See Bostic, 439 S.W.3d at 344 (noting inherent difficulties of proof in asbestos cases given the long latency period for asbestos-related diseases); Flores, 232 S.W.3d at 772-73 (same). However, Royce did provide many details. In regard to the work he did in the summers of 1975, 1976, and 1977, Royce specifically stated that the majority of the bags of drilling mud additives he worked with were asbestos and that the majority of the bags were the six he named, including the five shown by other evidence to be produced by Union Carbide: Visquik, Visbestos, IMCO Superbest, IMCO Shurlift, and Super Visbestos. As for his time spent working on oil rigs in 1978 for Flournoy Drilling Company, Royce stated that he used a “significant amount” of all six of the products he named. He made similar statements about the products he handled as a warehouseman beginning in late 1978. Although Royce did not expressly rule out the possibility that he worked with other asbestos products in those jobs, he also did not explicitly state that he did work with other asbestos products. Those six were the ones that he recalled. The only product Royce recalled using When he returned to the Flournoy oil rigs in 1986, was Super Visbestos, a Union Carbide product. For each of these work experiences, Royce recounted how dusty his clothes would get while working with the bags of asbestos drilling mud additives. And through all of these experiences, Janet washed Royce’s clothes two to three times a week, shaking the dust from the clothes in the small laundry room and sweeping up the dust. Although Royce said that Janet might have had to repeat her clothes- washing process more than once on the days that she washed his work clothes, there was no specific evidence regarding how much time she would spend doing so or how long the dust from the process would remain airborne. While dose must be quantified, exposure-level evidence is not required to be mathematically precise. See Bostic, 439 S.W.3d at 353; Flores, 232 S.W.3d at 773. Royce did not exhibit perfect recall of the products he used, but reasonable people could interpret his statements as indicating that the majority of asbestos which with he worked in those jobs, and thus the majority of the asbestos that he brought home and to which Janet was exposed, came from Union Carbide products. To require someone to remember precise percentages of how many bags of each product they used decades before would not be reasonable and is not the law. See generally Bostic, 439 S.W.3d at 344; Flores, 232 S.W.3d at 772-73. To the extent Royce spoke inconsistently or without certainty at times would be matters for a jury to assess and do not render his testimony no evidence under the summary judgment standards of review. See Cantey Hanger, LLP, 467 S.W.3d at 481; Timpte Indus., 286 S.W.3d at 310. Regarding Janet’s cumulative asbestos exposure, Union Carbide highlights the fact that one of the six products Royce identified as ones he worked with frequently, Flosal, was manufactured by other defendants. The summary judgment record contains no evidence indicating exactly what percentage of Janet’s total exposure from washing Royce’s clothes came from Union Carbide products as opposed to Flosal. There is also no indication that the percentage of the total attributable to Flosal was particularly high compared to the other products. While Royce recalled using a lot of Flosal while working for Flournoy in 1978, he also testified that he used a “significant amount” of all six products, and he noted that he only recalled using Super Visbestos, a Union Carbide product, on his second stint with Flournoy Drilling Company. It therefore does not appear that Flosal played an out-sized role in the asbestos exposure of either Royce or Janet. Union Carbide also asserts appellants failed to account for other asbestos exposures Janet may have suffered. See Bostic, 439 S.W.3d at 350 (“[W]hen evidence is introduced of exposure from . . . other sources, proof of more than a doubling of the risk may not suffice to establish substantial factor causation.”). The record, however, contains little evidence of such exposures in this case, much less at levels approaching the exposure Janet received from washing Royce’s work clothes. Union Carbide notes that Janet’s husband worked in parts sales for oilfield customers, but there is scant evidence he was regularly exposed to asbestos or brought it home on his clothing. Ronda, in fact, testified that her father’s clothes were rarely dirty. There was some evidence that Janet’s parents may have worked at a steel mill at one time and that after her divorce she lived with a boyfriend who worked as a welder, but there is no indication in the record about what her parents did at the mill or whether any of these three people were exposed to asbestos or, if they were, whether they brought it home and exposed Janet. Union Carbide does not suggest that any of the other evidence about Janet’s life and work history contained any indication of other asbestos exposures. In the summary judgment context, we disregard evidence contrary to that of the nonmovant unless reasonable jurors could not. Timpte Indus., 286 S.W.3d at 310. Having reviewed the historical, factual evidence on Janet’s exposure, we now turn to consideration of Cohen’s expert affidavit and the epidemiological materials on which he relied. See Bostic, 439 S.W.3d at 350 (requiring that proof of a doubling of the risk must “be shown through reliable expert testimony that is based on epidemiological studies or similarly reliable scientific testimony”). The Expert and Scientific Evidence Although Cohen calculated a dose level for Janet’s exposure, there are numerous difficulties with his affidavit and the studies on which he relies that render his opinion unreliable. Unreliable expert testimony is, legally, no evidence. See Gunn v. McCoy, 554 S.W.3d 645, 661-63 (Tex. 2018). “An expert’s opinion may be considered unreliable if it is based on assumed facts that vary materially from the actual facts, or if it is based on tests or data that do not support the conclusions reached.” Id. at 662. Such “testimony may also be unreliable if ‘there is simply too great an analytical gap between the data [relied upon] and the opinion proffered.’” Id. at 663 (quoting Hous. Unltd. Metal Processing v. Mel Acres Ranch, 443 S.W.3d 820, 835 (Tex. 2014)). Moreover, if no basis for an opinion is offered, or the basis offered provides no support, the opinion is not probative evidence. Id. at 662. In his affidavit, Cohen calculated Janet’s lifetime asbestos exposure from washing Royce’s work clothes at 2.63 fiber years per cubic centimeter. He also concluded that this exposure “more than doubled her risk of developing malignant mesothelioma” and “constituted a significant contributing factor in the development of her malignant mesothelioma and subsequent death.” Cohen did not account for the fact that Union Carbide did not make all of the asbestos products with which Royce worked and, by extension, to which Janet was exposed; instead, Cohen spoke to Janet’s exposure as a unified whole. Cohen’s analysis depended heavily on his use of a laboratory simulation that attempted to determine the exposure level for laundering contaminated clothes. According to Cohen, results of this simulation, which were not attached to his affidavit, showed that launderers would experience an 8-hour time-weighted average exposure of 0.13 fibers per centimeter of air breathed. Cohen, however, took this number and multiplied it by over 11 times to calculate the level of asbestos to which Janet was exposed each day that she washed Royce’s clothes at an 8-hour time-weighted average exposure of 1.49 fibers per centimeter of air breathed. Cohen’s only explanation for this over-elevenfold increase was that Janet worked in a smaller space and there was no evidence about whether the laundry room in which she worked had ventilation, whereas the room in the simulation had regular exchanges of air. Cohen did not offer any specific calculations to support the elevenfold increase, much less show that he used any reliable method to reach this conclusion. Cohen also did not offer any analysis regarding the actual time for which Janet was exposed to asbestos fibers. He calculated an 8-hour time-weighted average exposure for Janet apparently by referencing the 8-hour time-weighted average determined in the laboratory simulation, but there is no evidence that the duration of Janet’s exposure on the days she washed Royce’s clothes was comparable to the duration of exposure considered in the simulation. Testimony suggested that Janet washed Royce’s clothes two or three times a week and might have had to repeat her laundry process more than once on those days. There is no evidence in the record on whether Janet’s laundering process took her ten minutes, an hour, or any specific period of time. Cohen then went further and added an additional 1 fiber per centimeter of air breathed for Janet’s exposure based on the fact that she also swept up the asbestos dust after handling the clothes. Cohen again did not offer any specific calculations or methodology in support of this additional exposure but simply referenced a chart attached to his affidavit listing various studies and lab simulations for sweeping and cleanup. Again, Cohen did not provide the actual studies or many pertinent details to support his conclusion. In ultimately determining Janet’s cumulative exposure level from laundering Royce’s asbestos-covered work clothes, Cohen calculated that she laundered Royce’s work clothes 2.5 days a week, over a total of 2.1 years, with an 8-hour time-weighted average daily exposure of 2.5 fibers per centimeter (apparently rounding up from the 2.49 fibers per centimeter he actually calculated). Cohen asserted that “mesothelioma is a low dose disease” and “[e]ven an exposure below 0.2 fiber-cc/years can be a substantial factor in the development of the disease.” He supported these assertions with a brief discussion of studies in which low exposures were found to have caused mesothelioma, but he did not provide copies of the studies themselves. He also did not provide risk ratios found in the named studies with the exception of one, and he did not provide the confidence levels or confidence intervals for any of the studies he names on this point. See Bostic, 439 S.W.3d. at 347 (listing risk ratio, confidence level, and confidence interval as hallmarks of reliability for epidemiological studies); Havner, 953 S.W.2d at 713-727 (same). Cohen cited no studies to support his assertion that “it has been repeatedly and consistently demonstrated in the medical and scientific literature that family members whose sole asbestos exposure resulted from asbestos dust brought into the home on another family member’s clothing have developed mesothelioma.” In support of his ultimate conclusion that Janet’s exposures to Royce’s work clothes “more than doubled her risk of developing mesothelioma,” Cohen cited five studies. One of these studies, he neither attached to his affidavit nor provided any information beyond the fiber years per cubic centimeter found by the study and the risk ratio calculated. He did not explain the subject matter or conditions considered in the study or provide its confidence level or confidence interval. See generally E.I. du Pont de Nemours & Co. v. Hood, No. 05-16-00609-CV, 2018 WL 2126935, at *18 (Tex. App.—Dallas May 8, 2018, no pet.) (“An expert’s assurance that a study establishes causation does not make it so.”) (mem. op.) (citing In re Allied Chem. Corp., 227 S.W.3d 652, 656 (Tex. 2007)). Cohen did provide copies of four studies on which he relied: N. Offermans, et al., “Occupational Asbestos Exposure and Risk of Pleural Mesothelioma, Lung Cancer, and Laryngeal Cancer in the Prospective Netherlands Cohort Study,” 56 J. of Occupational and Envtl. Med. 6 (2014); A. Lacourt, “Pleural Mesothelioma and Occupational Coexposure to Asbestos, Mineral Wool, and Silica,” Am. J. Respiratory and Critical Care Med. (2013); K. Rödelsperger, et al., “Asbestos and Man-Made Vitreous Fibers as Risk Factors for Diffuse Malignant Mesothelioma: Results From a German Hospital-Based Case-Control Study,” 39 Am. J. Indus. Med. 262 (2001); and Y. Iwatsubo, et al., “Pleural Mesothelioma: Dose-Response Relation at Low Levels of Asbestos Exposure in a French Population-based Base- Control Study,” 148 Am. J. Epidemiology 139 (1998). Apart from citing these studies, however, Cohen does not offer any analysis of how they support his conclusions about Janet’s asbestos exposure and mesothelioma. Moreover, although each of these studies confirms that mesothelioma may be caused by low-dose asbestos exposures, none of the studies appeared to look at conditions similar to those Janet experienced. See Flores, 232 S.W.3d at 771-72. For example, none of the studies separated out exposures by fiber type. Evidence in the summary judgment record indicates that the Union Carbide products at issue contained chrysotile fibers, but none of the attached studies specifically looked at that type of fiber. See Yates v. Ford Motor Co., 113 F. Supp. 3d 841, 857-61 (E.D.N.C. 2015) (rejecting expert’s testimony utilizing the Rödelsperger and Iwatsubo studies in part because they did not distinguish between asbestos fiber types); Butler v. Union Carbide Corp., 712 S.E.2d 537, 542-43 (Ga. App. 2011) (rejecting expert’s testimony relying on Iwatsubo study because it failed to provide scientifically reliable evidence regarding exposure to only chrysotile asbestos); Smith v. Kelly–Moore Paint Co., Inc., 307 S.W.3d 829, 837–39 (Tex. App.—Fort Worth 2010, no pet.) (rejecting expert’s reliance on the Rödelsperger and Iwatsubo studies in part because they were “inconclusive regarding the effect of exposure to only chrysotile fibers [and did] not support a minimum threshold dose for chrysotile only exposure that would increase one’s risk of developing mesothelioma”); see also Rödelsperger, supra, at 263 (“It has been demonstrated that . . . the type of asbestos [is] important for the quantification of the risk. [A] reliable dose-response relationship between the concentration of long amphibole fibers and the risk of mesothelioma has been consistently established . . . , while no relationship was observed for chrysotile fibers.”); Iwatsubo, supra, at 141 (“We could not examine mesothelioma risk according to fiber types because our study design . . . did not allow us to identity those subjects whose exposure was only to chrysotile fibers.”).[6] Cohen failed to account for the fact that not all of Janet’s asbestos exposure traced to Union Carbide products. The opinions expressed in Cohen’s affidavit also depended on unsupported factual assumptions, unrevealed and unexplained calculations and methodologies, studies for which he did not provide necessary information, and other studies that were not demonstrated to be sufficiently similar to Janet’s exposure to support causation. Accordingly, under the standards set forth by the Texas Supreme Court, Cohen’s affidavit is unreliable and no evidence of causation. See Gunn, 554 S.W.3d at 661-63; Bostic, 439 S.W.3d at 347-53; Flores, 232 S.W.3d at 770-73; Havner, 953 S.W.2d at 717-23. Conclusion In response to Union Carbide’s no-evidence motion for summary judgment, appellants failed to present scientifically reliable expert testimony or epidemiological evidence demonstrating that Janet’s exposure to Union Carbide asbestos products more than doubled her risk of developing mesothelioma. Accordingly, the trial court did not err in granting summary judgment for Union Carbide. We affirm the trial court’s judgment. /s/ Frances Bourliot Justice Panel consists of Justices Bourliot, Poissant, and Wilson. (Wilson, J., concurring) (Poissant, J., dissenting).