CRAIG L. RUSSELL v. CALL/D, LLC

THOMPSON, Associate Judge: In June 2012, appellant Craig Russell brought a lawsuit asserting “negligence— premises liability” and “strict liability/negligent failure to warn” claims against ap-pellee Call/D, LLC (“Call/D”), the owner of the apartment building where appellant was residing when he began to suffer the symptoms of Legionnaires’ disease, with which he was diagnosed in May 2011. Russell alleged in his Complaint and subsequent deposition that sewage back-ups in the apartment building and sewage-contaminated water that was allowed to stand in a vacant basement apartment, caused him to contract the disease. The Superior Court (the Honorable Natalia Combs Greene) granted Call/D’s motion in limine to exclude the testimony of Russell’s designated causation expert and then granted Call/D’s motion for summary judgment. Russell argues on appeal that Judge Combs Greene abused her discretion by ruling without considering his supplemental Super. Ct. Civ. R. 26(b)(4) statement and also that she erred in granting Call/ D’s motions in limine and for summary judgment. We disagree and therefore affirm. I. Factual and Procedural Background The following background facts, which we consider in the light most favorable to Russell, 1 are undisputed. In 2004, Russell leased an apartment on the second flooi; of 5218 Fitch Street, S.E. (Apartment 6), and lived there until May 2011. Sometime during the fall of 2010, the tenants living in the basement-level Apartment 1 vacated the premises, leaving behind their cats. The windows to Apartment 1 were left open, allowing the cats (and perhaps other animals) to enter and exit the vacant apartment. As a result, feces and dead animals were on the floor of the abandoned apartment, and the apartment had a horrible stench. The deposition testimony also establishes that in November 2010, a pipe burst in Apartment 1, causing flooding in the apartment. Additionally, sometime during or before May 2011, sewage and sewage-contaminated water backed up and overflowed into Apartment 1, covering its floor with a half-inch to inch of water that, at one point, also overflowed into the adjacent hallway. A video recording made on *863 May 28,. 2011 — after Russell had fallen ill with Legionnaires’ disease — shows that the apartment was still filthy at that time, with standing pools of water on the floor. Around May 8, 2011, Russell noticed that the water coming out of the faucets in his apartment smelled bad and was discolored for several moments before running clean. Sewage or sewage-like water would also bubble up out of his drains after use. During early May 2011, the entire apartment building also began to smell, prompting residents to make calls to management starting around May 9, 2011. On May 10, 2011, Russell began to feel fatigued. Over the next few days, he felt increasingly worse. On May 21, 2011, when he was barely able to stand, he went to the hospital and was diagnosed with Legionnaires’ disease, a serious form of pneumonia. Before the illness, Russell was a very healthy 89-year-old who routinely engaged in vigorous exercise. The parties’ designated experts agree that Legionella bacteria, the cause of Legionnaires’ disease, infect humans through the inhalation of contaminated airborne water droplets; that is, the water source containing the bacteria must be aerosol-ized and inhaled. Legionnaires’ disease has an incubation period of 2-14 days, meaning that “symptoms typically manifest within 2 to 14 days after the exposure.” During the two weeks before he became symptomatic, Russell went to many different places. He worked five days a week for ADT Security Systems as a service technician, a job that required him to go to an average of seven different locations every day. From May 6-8, 2011, Russell and his girlfriend were on a weekend trip to Bryce Resort, a ski resort in Mount Jackson, Virginia, where, according to the girlfriend’s deposition testimony, he took at least one shower. No samples were taken of the stagnant-water in Apartment 1 before the apartment was professionally cleaned. Testing performed in September 2011 on water from taps and drain traps in Apartment 6 (to which Russell never returned after he was stricken in May 2011) uncovered no Legionella bacteria.. As far as the record reveals, no one else living in the Fitch Street apartment building and no one else who visited Bryce Resort during the same time period when Russell was there contracted Legionnaires’ disease. As noted above, Russell filed his lawsuit in June 2012. On February 5, 2013, he served Call/D with a Rule 26(b)(4) statement in which he designated Dr. Steven Zimmet, a pulmonologist who treated Russell when he had Legionnaires’ disease, as an expert who would testify as to “how Legionnaires’ disease is contracted in general and specifically how Plaintiff contracted it due to his exposure to fumes and smells at his apartment building.” 2 No expert report or academic articles were attached to the Rule 26(b)(4) statement. Call/D’s attorneys deposed Dr. Zimmet on April 19, 2018. Although Russell’s counsel provided Dr. Zimmet. with two academic articles regarding Legionella bacteria immediately before the deposition, Dr. Zim-met testified that he did not rely on these articles in coming to the opinion about which he planned to testify at trial. 3 Dr. Zimmet assumed that Russell had regularly walked through sewage water when en *864 tering the apartment building, and he opined that Russell thereby came into contact with a sufficient number of aerosolized Legionella bacteria so as to become infected. After the April 30, 2013, close of discovery, Call/D filed a motion in limine to exclude Dr. Zimmet’s testimony, arguing that his training and experience as a clinician in treating patients diagnosed with Legionnaires’ disease did not render him qualified to pinpoint the source of Russell’s exposure to Legionella bacteria. On October 15, 2013, Judge Combs Greene issued an Omnibus Order in which she granted Call/D’s motion in limine to exclude Dr. Zimmet’s testimony, reasoning in part that his testimony regarding the source of Russell’s illness was “entirely speculative.” That ruling was the premise for Judge Combs Greene’s additional ruling that there was “no genuine issue of material fact ... as to [the] proximate cause” of Russell’s illness and her granting of Call/ D’s motion for summary judgment. II. Disregard of the Supplemental Rule 26(b)(4) Statement and the Decision Not to Hear Testimony from Dr. Zimmet On September 18, 2013, Judge Combs Greene issued an order setting September 26, 2013, as the date for a hearing on Call/D’s motion in limine. She asked the parties to be prepared to address “[t]he basis for Dr. Zimm[e]t’s opinion that sewage is a possible source for Legionella bacteria,” the extent of Dr. Zimmet’s “knowledge concerning the sources of [the] infections” of- his previous patients who had been diagnosed with Legionnaires’ disease, and “[a]ny and all scholarly literature Dr. Zimm[e]t used to guide or assist him in reaching his conclusion that sewage can be a source of the Legionella bacteria.” Judge Combs Greene also stated that, while she found Dr. Zimmet’s attendance “essential,” his attendance would be excused if the scheduled hearing was inconvenient and if Russell could provide sufficient answers and documentation to address the court’s concerns. Alternatively, Judge Combs Greene would “consider Dr. Zimm[e]t’s appearance by telephone” if the parties agreed. Two days before the scheduled hearing, apparently while Judge Combs Greene was on leave, Russell’s counsel advised the judge’s staff that Dr. Zimmet would be unable to attend a hearing on September 26 and requested a continuance until September 27. 4 Staff instructed the parties to appear at the scheduled time. During the September 26 hearing, Russell’s counsel gave the court the two academic articles counsel had shown Dr. Zimmet prior to his deposition. On September 30, 2013, four days after the hearing, Russell filed a supplemental Rule 26(b)(4) statement (“the supplemental statement”). The supplemental statement slightly amended the earlier description of Dr. Zimmet’s expected testimony, and attached to it were a number of journal articles and other publications. In her October 15, 2013, order granting Call/ D’s motion in limine, Judge Combs Greene stated that she “did not view or consider” the supplemental statement in ruling on the motion. Russell now contends that Judge Combs Greene abused her discretion in ruling on the motion in limine without reading the supplemental statement and without hearing testimony from Dr. Zimmet. This court reviews a trial judge’s decision whether to accept a supplemental Rule 26(b)(4) statement for abuse of discretion. See Ferrell v. Rosenbaum, 691 A.2d 641, 646 (D.C.1997). We “take into *865 account the reason for the supplement, the importance of the proffered testimony and any prejudice to the opposing party.” Id. at 647. In this case, we can find no abuse of discretion in Judge Combs Greene’s decision not to consider Russell’s supplemental statement. Russell was made aware by (1) the questioning at Dr. Zimmet’s deposition on April 19, 2013, (2) Call/D’s May 15, 2013, motion in limine, and (3) the affidavits of Call/D’s experts attached to Call/D’s May 15, 2013, motion for summary judgment (to which Russell filed an opposition on May 24, 2013), that Call/D was challenging Dr. Zimmet’s qualifications to testify, but Russell waited until September 30, 2013, to file the supplemental statement. Russell provided Judge Combs Greene (and he has provided this court) with no explanation for the delay; quite the contrary, his counsel asserted that Dr. Zimmet’s deposition was “very clear[,]” never mentioned a need to supplement the expert witness statement, and did not request leave to do so. Nor do we discern that the information provided by the supplemental statement was important for appellant’s case. The supplemental statement slightly modified the description of Dr. Zimmet’s anticipated trial testimony: the earlier statement said that he would testify about how Russell contracted Legionnaires’ disease “due to his exposure to fumes and smells at his apartment building[,]” while the supplemental statement said that Dr. Zimmet would testify about how Russell contracted Legionnaires’ disease “due to his exposure to stagnant water and their [sic] associated fumes and smells at his apartment building” (emphasis added). The supplemental statement’s references to “stagnant water” and “longstanding stagnant water” were new, but Judge Combs Greene could see, from Dr. Zimmet’s deposition, that his opinion was that “sewage water was the most likely source of [Russell’s] exposure[,]” meaning that Russell cannot establish prejudice from the court’s not having read the revised text. 5 The publications attached to the supplemental statement and the citations to some of them in the text of the statement did add something new, because some of them pertain to investigations of the presence of Legionella bacteria in sewage treatment plants. However, nothing in the supplemental statement indicated that Dr. Zim-met had approved the statement or had ever read the articles and publications attached to it. 6 Dr. Zimmet testified at his *866 deposition that he arrived at his opinion about the source of Russell’s exposure without relying on journal articles, 7 and, as Judge Combs Greene recognized, Call/D could rightly claim prejudice from a late supplemental statement that “would potentially subject them to different allegations ... at a late stage of the litigation.” 8 French v. Levitt, 997 A.2d 701, 704 (D.C. 2010). As for the court’s not hearing testimony from Dr. Zimmet before ruling on the motion in limine, Russell highlights Judge Combs Greene’s statement (in the order setting the. hearing) that the description of Dr. Zimmet’s testimony in Russell’s Rule 26(b)(4) statement was “very vague.” But Judge Combs Greene told the parties at the outset of the hearing that she had probably been “overly cautious” in scheduling the hearing and thought she could rule on the motion after rereading Dr. Zimmet’s deposition, without hearing from him. 9 We discern no reason to question *867 Judge Combs Greene’s assessment, especially since we, too, think that Dr. Zim-met’s deposition furnished a sufficiently clear basis for the court to rule on the motion in limine. 10 Thus, we cannot agree with Russell’s claim that Judge Combs Greene abused her discretion by ruling “when she lacked the very information ... she needed to make an informed decision[.]” III. The Ruling on Call/D’s Motion In Limine Russell contends that Judge Combs Greene also abused her discretion in excluding Dr. Zimmet’s testimony, arguing that Dr. Zimmet’s “training and experience provided a sufficient foundation to permit his expert opinion testimony on the issue of the source of [Russell’s] exposure to Legionella bacteria.” We review a trial court’s decision about whether to admit expert testimony for abuse of discretion. District of Columbia v. Anderson, 597 A.2d 1295, 1299 (D.C.1991). That means that our review is deferential, as “deference ... is the hallmark of abuse of discretion review.” General Elec. Co. v. Joiner, 522 U.S. 136, 143, 118 S.Ct. 512, 139 L.Ed.2d 508 (1997); see also Girardot v. United States, 92 A.3d 1107, 1113 (D.C. 2014) (“The concept of ‘exercise of discretion’ is a review-restraining one.” (quoting Johnson v. United States, 398 A.2d 354, 362 (D.C.1979))). Accordingly, the trial court’s decision will be “sustained unless it is manifestly erroneous.” Coates v. United States, 558 A.2d 1148, 1152 (D.C.1989). To be permitted to testify as an expert, a witness “must have sufficient skill, knowledge, or experience in that field or calling as to make it appear that his opinion or inference will probably aid the trier in his search for truth[.]” Dyas v. United States, 376 A.2d 827, 832 (D.C. 1977) (citation omitted). “Implicit in that requirement is that the expert [must] have a reliable basis for [his] theory steeped in fact or adequate data, as opposed to offering a mere guess or conjecture.” Perkins v. Hansen, 79 A.3d 342, 345 (D.C.2013) (citation and internal quotation marks omitted). “The purpose of expert opinion testimony is to avoid jury findings based on mere speculation or conjecture[,]” and thus the court must weigh the “sufficiency of the foundation for those opinions ... with this purpose in mind.” Washington v. Washington Hosp. Ctr., 579 A.2d 177, 181 (D.C.1990). “Without more than credentials and a subjective opinion, an expert’s testimony that ‘it is so’ is not admissible.” Wilson Sporting Goods Co. v. Hickox, 59 A.3d 1267, 1273 (D.C.2013) (quoting Viterbo v. Dow Chem. Co., 826 F.2d 420, 424 (5th Cir.1987)). “Expert testimony may be excluded when the expert is unable to show a reliable basis for their theory.” Haidak v. Corso, 841 A.2d 316, 327 (D.C.2004). In this case, Judge Combs Greene articulated many reasons for her ruling that Dr. Zimmet was “not qualified to testify as to the source of [Russell’s] disease,” among them that Dr. Zimmet had never been trained in or “involved with the identification of the source of exposure” of Legion *868 naires’ disease, was “not aware of any specific type of bacteria present in sewage water[,]” “did not indicate any literature to support his belief that Legionella bacteria could exist in sewage water[,]” had “never had responsibility for determining the source of a person’s Legionella bacterial exposure as part of his clinical practiee[,]” did not know “what strains of Legionella bacteria are most commonly associated with Legionnaire’s [sic] Disease in humans[,]” was not familiar with the temperatures at which or the conditions under which Legionella bacteria amplify, and had “no data to support his opinion that Le-gionella bacteria [were] at sufficient levels [in Apartment 1] to cause [Russell’s] Legionnaire’s [sic] disease.” Dr. Zimmet’s deposition answers support these findings as well as Judge Combs Greene’s finding that his opinion testimony on causation was “without factual basis.” It is true that, as Russell emphasizes, an expert witness may rely upon his experience alone when providing an opinion, see Perkins, 79 A.3d at 345 (“[A] physician’s experience may provide a reliable basis for his or her expert opinion.”), and that an expert witness need not cite a peer-reviewed journal article as the basis for every opinion he provides and need not use the “data that will provide the highest degree of certainty[.]” Id. at 346. In addition, “[w]here an expert otherwise reliably utilizes scientific methods to reach a conclusion, lack of textual support may ‘go to the weight, not the admissibility’ of the expert’s testimony.” Knight v. Kirby Inland Marine Inc., 482 F.3d 347, 354 (5th Cir.2007). But, as Judge Combs Greene recognized, Dr. Zimmet relied on neither experience in investigating the source of a Legionella infection, nor knowledge about what had been determined to be the source of his other Legionnaires’ disease patients’ exposures, nor peer-reviewed journal articles, nor data from testing at the apartment building, and he did not utilize a scientific method to conclude that Legionella bacteria were present in the Fitch Street apartment building. 11 Instead, as he acknowledged during his deposition, he “essentially ma[de] the assumption ... that the water in the apartment [building] contained amplified levels of Legionella bacteria.” He acknowledged that the same logic by which he eliminated the environment at the Bryce Resort as a source of Russell’s infection — that “[n]obody else got sick” there — was equally applicable to the Fitch Street apartment building (since no one else living in the apartment building got sick), and he offered only the “beliefs]” that “taking a couple of showers” at Bryce would not be enough to make the previously healthy Russell sick and that Russell “really had no other source of Legionella disease.” 12 Further, Dr. Zimmet invoked “common sense” and “intuiti[on]” as his basis for saying that the Bryce Resort was not a probable source of Russell’s infection and that Russell’s “walking] through ... sew *869 er water” at the apartment building was the likely culprit. 13 And, when asked at his deposition about his “training, knowledge and experience with respect to being able to connect a diagnosis of Legionnaire’s [sic] disease to a suspected souree[,]” Dr. Zimmet answered that the basis of his opinion in this ease was that “[y]ou know it when you see it.” That response — I know it when I see it — was the “quintessential ipse dixit justification.” TASER Int’l, Inc. v. Karbon Arms, LLC, No. 11-426, 2013 WL 6773663, at *1 (D.Del. Dec. 19, 2013). Judge Combs Greene was not required to admit opinion evidence that was “connected to existing data only by the ipse dixit of the expert.” Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 157, 119 S.Ct. 1167, 143 L.Ed.2d 238 (1999) (quoting Joiner, 522 U.S. at 146, 118 S.Ct. 512); cf. Brown v. Bray & Gillespie III Mgmt. LLC, No. 6:06-cv-661-Orl-22GJK, 2008 WL 2397601, at *6 (M.D.Fla. June 10, 2008) (concluding that an internist’s opinion that plaintiffs “contracted] an acute Legionella infection as a result of [their] hotel stay ... at the Sea Garden Inn in Florida” “lack[ed] a substantial basis,” and explaining that “[t]he opinions of an expert will not be accepted if based on merely the ipse dixit of the expert”). 14 Russell further contends that Judge Combs Greene “impermissibly demanded that [he] adduce direct (rather than circumstantial) evidence of the presence of Legionella bacteria in the building before permitting Dr. Zimmet to testify.” While we agree that the source of a plaintiffs Legionnaires’ infection can be proven by (and perhaps can be proven only by) circumstantial evidence, 15 we reject Russell’s claim of error, because we do not agree that Judge Combs Greene excluded Dr. Zimmet’s testimony on the ground that *870 he had no direct evidence of the presence of Legionella in the building. Judge Combs Greene found it “significante ]” that Dr. Zimmet “did not indicate any literature to support his belief that Legionella bacteria could exist in sewage water.” She reasoned that Dr. Zimmet’s opinion that sewage water was the most likely source of Russell’s exposure was “not supported by published scholarly data” and that Dr. Zimmet was “unfamiliar with several areas of scientific scholarship or clinical work that might provide a substantive basis for the conclusion offered ” (emphasis added). She also cited portions of Dr. Zimmet’s deposition in which he testified that he did not regularly read “published medical or scientific or hygiene literature with respect to Legionnaire[s’] disease”; that he had formed his opinions in this case before seeing, just prior to his deposition, an article Russell’s counsel had given him entitled “Health Effects of Sewage Aerosols: Additional Serological Surveys and Search for Legionella Pneumophila in Sewage”; that the article would not play a role in the opinions he would render in the case; and that he had not reviewed any peer-reviewed and published literature discussing the prevalence of Legionella in sewage-contaminated water. In addition, Judge Combs Greene explicitly recognized that case law has held that “it is not an abuse of discretion for courts to allow expert testimony when the expert did not conduct independent testing” and that “[a]s long as the expert relied on published data generated by another expert in the pertinent field, the opinion is' admissible.” We read all these statements in the Omnibus Order as indicating that even though Dr. Zimmet had no direct evidence that Legionella bacteria were in water at the apartment building, Judge Combs Greene would likely have permitted him to testify if he had had a basis in the scientific literature to say that the sewage-contaminated water in the basement of the apartment building was a more likely source of Russell’s exposure than a shower at Bryce Resort or any of the scores of locations Russell had visited through his work and other activities during the incubation period. We next address Russell’s contention that Judge Combs Greene applied a legally erroneous standard in ruling on the motion in limine. He argues that the judge held Dr. Zimmet “to a burden of proof greater than the preponderance of evidence (more likely than not) standard[.]” 16 The evidence of this error, he asserts, is the following statement on page 9 of Combs Greene’s Omnibus Order: “Dr. Zimmet concludes that he ‘believes’ that the sewage water was the most likely source of exposure as compared to the other possible sources because it looked like a ‘cesspool’ ” (bold font in the original). We note that Judge Combs Greene made a similar statement in footnote 6 of the Omnibus Order: “Dr. Zimmet does not, in his deposition, states [sic] that he believes the sewage was the actual source of the Le-gionella bacteria. Rather, ... he ‘believes’ the apartment complex was the most likely source” (bold font in the original). Russell is correct that his burden was to prove the “most likely” source of his exposure. See Snyder v. George Washington Univ., 890 A.2d 237, 248 (D.C.2006) (“We require only that a causation expert state an opinion, based on a reasonable degree of medical certainty, that the defendant’s conduct was a likely cause of the plaintiffs injuries.”). Because it is not clear to us what Judge Combs Greene intended to *871 convey through her emphasis on the words “believes,” “actual,” and “most likely,” 17 we are skeptical of Russell’s claim that she misapprehended the legal standard or incorrectly applied it. We are, however, satisfied that even if Judge Combs Greene erroneously held Russell to a heightened standard of proof, the basis for her ruling that Dr. Zimmet was not qualified to testify as an expert on causation was that Dr. Zimmet did not provide “any real basis (based on facts or science) for his opinions.” That was a sufficient basis for precluding him from testifying as a causation expert regardless of the degree of certainty that was required. For that and all the foregoing reasons, we will not disturb Judge Combs Greene’s ruling on the motion in limine. Russell argues in the alternative that Dr. Zimmet was entitled to testify about the source of Russell’s exposure to Legionella as a fact witness rather than as an expert witness. It is true that “[i]nso-far as a physician obtains and develops his information and opinions in the course of his treatment of a patient, he becomes an ‘actor or viewer’ who should be treated as an ordinary witness rather than as an expert covered under Rule 26(b)(4).” Adkins v. Morton, 494 A.2d 652, 657 (D.C. 1985) (explaining that “the crucial inquiry is whether the facts and opinions possessed by the expert were obtained for the specific purpose of preparing for the litigation in question”); Gubbins v. Hurson, 885 A.2d 269, 277 (D.C.2005) (quoting Adkins and reasoning that “Dr. Kelly’s opinion, which he expressed to Gubbins while she still was under his care, that her nerves were injured by the medication she received during her surgery” “clearly fell within this ‘exempt-from-Rule 26(b)(4)’ category”). 18 However, “a witness may be an ‘expert’ as to some matters and an ‘actor’ as to others.” Id. (internal quotation marks omitted). The question here is one of timing: When did Dr. Zimmet reach the opinions about which Russell sought to have him opine? If he reached them only “while preparing for his trial appearance[,]” id. at 278, he needed to be qualified as an expert to be permitted to testify. What the record reveals is the following: Dr. Zimmet testified that he had formulated his opinions as of February 5, 2018 (the date of Russell’s previous Rule 26(b)(4) statement). There was evidence that, as recently as September 2012, Dr. Zimmet or (much more frequently) his staff had provided care to Russell in connection with a blood clot in his leg that stemmed from his being bedridden for weeks while recovering from Legionnaires’ disease. However, Russell testified that the last time he went to any doctor for anything relating to the Legionnaires’ disease itself was in June 2011. In addition, Dr. Zimmet testified that Russell’s diagnosis of Legionnaires’ disease was made by other doctors prior to Dr. Zimmet’s involvement in Russell’s treatment, and one of those doctors (“the infectious disease doctor”) wrote in *872 the medical records (apparently incorrectly) that Russell reported that he had water leakage in his apartment from a ventilation system and attributed the Legionella exposure to that. 19 Dr. Zimmet described the “grilling” about “[w]hat’s your exposure” that “we” doctors conduct when encountering a patient who has Legionnaires’ disease, but also explained that Russell was asked about exposures “at the time he came in the hospital in his May admission,” clarified that Russell did not answer such questions to Zimmet specifically, and acknowledged he had never “had the responsibility for determining the source of an individual’s Legionella bacterial exposure as part of [his] clinical practice.” Further, Dr. Zimmet had never visited the Fitch Street apartment building. He relied on what he read in Russell’s deposition transcript and other depositions and on what he saw in the video of Apartment 1 for his knowledge of the conditions that existed in the building, and, importantly, he did not see the video or any photos of Apartment 1 until the day of his deposition, which was April 19, 2013. 20 In short, the record provides no foundation for a conclusion that Dr. Zimmet “formulated the opinions in question while he was treating” Russell, Gubbins, 885 A.2d at 278, rather than in preparation for his deposition and for trial, such that it was error not to permit him to present the opinions as a fact witness. IV. Summary Judgment Russell’s final contention is that he presented sufficient circumstantial evidence for a reasonable jury to fairly infer that Legionella bacteria were present in his apartment building and that he was exposed to the bacteria there. He argues that whether sewer water is a hospitable breeding ground for Legionella is a disputed issue that could not be resolved on summary judgment. We can agree with the latter contention, but the problem for Russell, who bore the burden of proof, was that, upon the court’s ruling on the motion in limine, he was left without an expert to testify that sewage-contaminated water likely contains a strain or strains of Le-gionella associated with Legionnaires’ disease and that Russell’s exposure to sewage-contaminated water was the likely source of his illness. Russell needed to prove more than the “bare possibility” that the contaminated water in the apartment building was the exposure source. Ferrell, 691 A.2d at 650. Meanwhile, Call/D’s environmental health expert John David Krause, whose affidavit describes his involvement in “at least 10 investigations involving Legionnaires’ disease” and cites his consideration of the “relevant peer-reviewed scientific literature,” was poised to testify that “Le-gionella are not normally found in sewage”; that Legionella generally “corn- *873 pete[] poorly with other bacteria in the environment[,]” which “typically out compete and may even prey upon Legionella and that “aerosols of water droplets, such as those created by showering, are capable of carrying Legionella bacteria[.]” 21 Call/ D’s expert Tim Keane, a water treatment consulting engineer who runs a company called Legionella Risk Management, Inc., and whose expert report describes his “lead role in [Legionella ] outbreak investigations” for state and county health departments, was prepared to testify that 24% of Legionnaires’ outbreaks are associated with travel and that travel-related outbreaks are difficult to detect because of the dispersal of persons from the source of the outbreak; and that samples of stagnant water taken from Apartment 6, which Russell had vacated, in September 2011 were negative for Legionella, 22 In addition, Call/D’s expert Dr. Hung Cheung, a specialist in environmental medicine who had testified as an expert in epidemiology and toxicology, was poised to testify, inter alia, that Russell’s visit to Bryce Resort was consistent with the incubation period for Legionnaires’ disease. 23 Reviewing the matter de novo, we conclude that, presented with the foregoing and other testimony by Call/D’s experts and no expert causation testimony to the contrary, 24 a jury would have had no basis other than speculation for finding that sewage-contaminated water or sewage vapors at the apartment building were the source of Russell’s Legionella exposure. 25 Accordingly, Judge Combs Greene did not err in granting summary judgment in favor of Call/D. For the foregoing reasons, the judgment of the trial court is Affirmed. Footnotes [1] "This court reviews a summary judgment decision de novo, construing the facts independently in the light most favorable to the nonmoving party.” Ivey v. District of Columbia, 949 A.2d 607, 611 (D.C.2008). [2] Dr. Zimmet estimated that, in his clinical practice, he had each year evaluated “one or more” patients who had Legionnaires’ disease. [3] In preparation for his deposition, Dr. Zim- • met consulted only an online service called UpToDate, which he agreed was a "quick way for a physician to make sure ... he hasn’t missed anything current!.]” [4] Russell thereafter filed a motion to the same effect. [5] Russell's Reply Brief draws a distinction between "stagnant water” and "sewer water,” apparently to suggest that the revised language in the supplemental statement did add something different that Judge Combs Greene needed to consider. Remarkably, the brief asserts that from the very beginning, "this case has been about [Russell's] ‘exposure to standing water in his apartment buildIng[,]’ " and that "[wjhether sewer water is a hospitable breeding ground for Legionella is simply not at issue in this litigation and amounts to a straw man argument.” That assertion cannot be squared with Russell’s Complaint, which avers that his Legionnaires' disease was proximately caused by his exposure to raw sewage, or with his opposition to Call/D’s motion for summary judgment, which repeats that claim. Moreover, Dr. Zimmet opined at his deposition that the apartment building was the most likely source of Russell's exposure to Legionella bacteria because of the "contaminated sewer water” there. Therefore, if the references to "stagnant water” in the supplemental statement were meant to be substantively different from the prior claims about sewage-contaminated water, that change, in itself, was reason for Judge Combs Greene to decline to consider the post-discovery-period supplemental state.ment. [6] Call/D may be correct that the court's disregard of the publications did not prejudice Russell even if Dr. Zimmet had read them. We say that because Judge Combs Greene could reasonably have found that the publications provided no support for Dr. Zimmet’s *866 opinion. One of the publications reports that “the risk of legionellosis to sewage workers was considered low.” Another of the articles explains that L. pneumophila is "the organism responsible for the majority (80%) of legionel-losis outbreaks” and that the Legionella bacteria found in sewage samples were "predominantly non-L. pneumophila species[.]” The article further states that using one method, Legionella pneumophila was recovered "in only 10% of primary effluent samples and 5% of secondary effluent samples.” Under a different method, Legionella pneumophila was detected "with much greater frequency!,]" although whether that method "actually detected L. pneumophila [is unclear] since the PCR and plate culture methods did not support the DFA method findings.” The authors explain that the test results could have been a false reading explained by antibodies having "cross-reacted with other organisms in the sewage.” [7] Regarding the publications Russell’s counsel handed to Judge Combs Greene in open court during the September 26 hearing, counsel incorrectly told the judge that Dr. Zimmet said at his deposition that he "was going to rely upon” the publications. In fact, Dr. Zim-met testified at his deposition that the publications would not "play[] any role in the opinions [he was] going to render in this case[.]” He further agreed that there was "no peer-reviewed published literature that [he was] citing in support of [his] opinions" and that he was not relying on any specific literature "to indicate that sewer water is a vector for distributing Legionnaire’s [sic] disease or Legionella bacteria.” [8] Judge Combs Greene explained at the September 26 hearing that Russell had to "put [the literature he was relying on] out there so the other side knows, can question it, can contest that at the deposition and say [']did you know that this source that you relied on has been discredited by every major author, peer review,!’] so on and so forth ... and let the person say what they have to say about it.” An impetus for Russell’s submitting the publications attached to the supplemental statement was no doubt that in May 2013, Call/D had filed affidavits from its experts stating, inter alia, that "conditions favorable to Legionella are not normally found in sewage.” Thus, analogous to the situation in Ferrell, a case on which Russell relies, in which this court held that the plaintiff’s supplemental Rule 26(b)(4) statement should have been considered by the trial court, "the need for supplementation was caused by new information from” Call/D’s experts. Ferrell, 691 A.2d at 647. However, in Ferrell, the plaintiff’s counsel advised defense counsel of the proposed expanded testimony of plaintiff's expert five days after learning certain factual information during defendant's deposition and served plaintiff's formal supplemental Rule 26(b)(4) submission a few weeks later. Id. at 645, 645 n. 6. Here, by contrast, Russell’s supplemental statement was served and filed over four months after Call/D had filed its experts’ affidavits and its motions in limine and for summary judgment, and after the September 26 hearing on the motion in li-mine. [9] Judge Combs Greene also told counsel at the outset of the hearing that her reason for not calling counsel and saying "don’t come in” was that she thought that one or both of the parties had "wanted to make [her] aware *867 of something” and thus would have objected to a postponement. [10] Russell complains that Dr. Zimmet could have testified regarding his experience with previous patients and his knowledge concerning the sources of their infections. But Dr. Zimmet had an opportunity to do that at his deposition (including during the questioning by Russell’s counsel, when he simply agreed with counsel’s vague leading question that "what [he] did in this case” was "tr[y] to figure out all of the possibilities that ... might explain the illness ... that [he is] trying to treat the patient for”), and did not do so. [11] "Scientific method” refers to "[t]he process of generating hypotheses and testing them through experimentation, publication, and replication.” Black's Law Dictionary 1547 (10th ed.2014). [12] Dr. Zimmet did not need to rule out every possible alternative source of exposure. See Westberry v. Gislaved Gummi AB, 178 F.3d 257, 265 (4th Cir.1999). But, to offer an opinion to a reasonable degree of medical certainty, he needed — and, in Judge Combs Greene's reasonable estimation, he lacked— “an objectively well founded conviction that the likelihood of one cause [of Russell’s exposure was] greater than any other[.]" Robinson v. Group Health Ass’n, Inc., 691 A.2d 1147, 1150 (D.C.1997) -(quoting Clifford v. United States, 532 A.2d 628, 640 n. 10 (D.C. 1987)). [13] We note, too, that although Russell testified that he sometimes checked on an elderly woman who was living on the basement level of the apartment building, the deposition testimony contains no evidence that Russell actually "crossed through” or "walked through” any water or did so "twice a day,” as Dr. Zimmet assumed. Russell's testimony indicated that, upon entering the apartment building on the ground floor, he needed to walk up to his apartment rather than down to the basement level. He also testified that he at no time entered Apartment 1. Judge Combs Greene could reasonably "conclude that there [was] simply too great an analytical gap between the data and the opinion [Dr. Zimmet] proffered." Joiner, 522 U.S. at 146, 118 S.Ct. 512. [14] We also defer to Judge Combs Greene’s reasoning that, in light of the facts that the sewage-contaminated water in the apartment building had not been tested for Legionella bacteria and that Dr. Zimmet had cited no basis in the scientific literature for his conclusion that the sewage water was the most likely source of exposure, to allow his speculative testimony to that effect would have been more prejudicial than probative. [15] Epidemiological evidence, which typically is used to prove the source of a Legionnaires’ outbreak, see Brown v. Bray, 2008 WL 2695230, at *2 (M.D.Fla. July 2, 2008) (”[E]p-idemiologic data indicated that the source of the outbreak was the hotel.”), is circumstantial evidence. See In re Hanford Nuclear Reservation Litig., 1998 WL 775340, at *5 (E.D.Wash. Aug. 21, 1998) (”[P]laintiffs here are forced to rely on experts who present circumstantial proof of causation, in particular, epidemiological proof.”); Land v. United States, 37 Fed.Cl. 231, 235 (Fed.C1.1997) (”[T]he two primary forms of circumstantial evidence in a toxic tort case are etiological evidence and epidemiological evidence.”) (citing Renaud v. Martin Marietta Corp., 749 F.Supp. 1545, 1553 (D.Colo. 1990)); see also Silivanch v. Celebrity Cruises, Inc., 171 F.Supp.2d 241, 255 (S.D.N.Y.2001) (describing circumstantial evidence that sufficed to prove causation: the Legionella bacteria recovered from the sand filters of a cruise ship spa system "were indistinguishable from organisms found in the respiratory system of one of the victims of the outbreak”). [16] A trial court erroneously exercises its discretion if it applies an incorrect legal standard. See Herbin v. United States, 683 A.2d 437, 443 (D.C. 1996). [17] At one point, Judge Combs Greene appeared to be distinguishing between Dr. Zim-met’s opinion that the apartment building "was the actual and only source of [Russell’s] Legionnaire's [sic] Disease” and his not opining that the sewage "was the actual source of the Legionella bacteria" (emphasis in original). At another point, however, she characterized Dr. Zimmet’s view as an opinion that the building "was the most likely source of exposure as compared to other possible sources ([Russell’s] job and Bryce Resort)” because of the contaminated sewer water. [18] See also Folks v. District of Columbia, 93 A.3d 681, 685 (D.C.2014) ("[T]he medical records suggest that the treating physicians would be able to testify at trial about their opinions on the issue of causation based on the information that they had obtained in their treatment of Mr. Folks.”). [19] Russell stated during his deposition that there were no ventilation ducts in his apartment and that he had never “spent any time in a residence where a ventilation system leaks,” and one of Call/D's experts also stated in his expert report that “[t]here is no vent system, no central air system and no air ducting system that could have dripped water into [Russell's] apartment.” [20] Russell’s girlfriend, who made the video of the standing water in Apartment 1, testified that she decided to do this on her own after hearing Russell’s doctors say "that his illness was a result of some sort of bacteria in water.” Her thinking was that "maybe this [i.e., the conditions in Apartment 1] contributed to [Russell's illness].” Explaining why he did not go back to Apartment 6 after he was discharged from the hospital, Russell testified non-specifically that "[d]octors” (he had "seven doctors”) told him not to breathe the air in his apartment. He did not attribute that advice to Dr. Zimmet. [21] To be sure, Russell filed a motion in limine (on which the court never ruled) to strike Krause’s testimony. However, Russell’s argument was that Krause should not be allowed to testify that Legionnaires' disease "cannot be contracted through the respiration of fumes ... or ... odors” or that "a specific determination of source and exposure pathway is required in order to establish causation with regard to Legionella exposure." We cite only other opinions by Krause. [22] Russell filed a motion in limine (on which the court never ruled) to strike Keane’s testimony, but the specific opinions he asked the court to strike were as to other points. [23] Again, Russell filed a motion in limine (on which the court never ruled) to strike Dr. Cheung’s testimony, but the specific opinions he asked the court to strike were as to other points. [24] In his motion in the trial court to strike Keane's testimony, Russell suggested that this is a case "where the jury can make the decision on medical causation without expert testimony!)]” but in his briefs on appeal he does not appear to take issue with Judge Combs Greene’s finding that "determining the source of contracting Legionnaire[s'] Disease is beyond the 'ken of the average layman.’ ” [25] Cf. Sudbeck v. Sunstone Hotel Props., Inc., No. 2:04-cv-1535, 2006 WL 2728624, *8 (D.Ariz. July 26, 2006) (reasoning that even where the evidence established that plaintiff contracted Legionnaires' disease sometime prior to July 8, 2002, and stayed at the defendant’s resort between June 25 and 27, 2002, i.e., within the incubation period of the Le-gionella pneumophila bacteria, and even where there was evidence that Legionella bacteria were found in water samples collected at the resort in May 2003, "it would be sheer speculation to say that it is more likely than not that the bacteria was present at the resort in June 2002 when [plaintiff] was a guest”).