Nancy Carmen Curnel and Ronald Curnel v. Houston Methodist Hospital-Willowbrook and Michael Esantsi

Opinion issued August 16, 2018 In The Court of Appeals For The First District of Texas ———————————— NO. 01-17-00088-CV ——————————— NANCY CARMEN CURNEL AND RONALD CURNEL, Appellants V. HOUSTON METHODIST HOSPITAL-WILLOWBROOK AND MICHAEL ESANTSI, Appellees On Appeal from the 55th District Court Harris County, Texas Trial Court Case No. 2016-36453 OPINION ON REHEARING1 1 Appellants, Nancy and Ronald Curnel, have filed a motion for rehearing of our October 31, 2017 memorandum opinion and judgment. We grant the motion, withdraw our memorandum opinion and judgment of October 31, 2017, and issue this opinion and a new judgment in their stead. This is an interlocutory appeal from the trial court’s order dismissing health care liability claims for failure to serve adequate expert reports. See TEX. CIV. PRAC. & REM. CODE §§ 51.014(a)(9), 74.351(a), (b). According to the expert reports, Nancy Curnel presented to the emergency room of Houston Methodist Hospital-Willowbrook with elevated liver enzymes caused by a recently prescribed antibiotic. Curnel was examined by a hospitalist, Dr. Michael Esantsi, who misdiagnosed her with viral hepatitis; ordered that she continue taking her current medications, including the antibiotic that was causing her elevated liver enzymes; and admitted her to the hospital. Once admitted, Curnel continued to receive the hepatotoxic antibiotic, and she began to receive another well-known hepatotoxic medication, acetaminophen, again by an order of Esantsi. On the third day of her hospitalization, Curnel was examined by a gastroenterologist, who noted that she might be suffering from drug-induced liver injury. He ordered a biopsy of Curnel’s liver to test for other potential causes. Later that same day, another hospitalist discontinued the antibiotic. Curnel’s liver enzymes began to improve. Curnel’s physicians did not cancel or postpone the biopsy. On the morning of the fifth day of Curnel’s hospitalization, a blood clotting test ordered by Esantsi to clear Curnel for the biopsy returned as normal, and a radiologist performed the biopsy as scheduled. During the biopsy, the radiologist nicked Curnel’s artery, causing her severe injuries. 2 Curnel and her husband, Ronald, asserted health care liability claims against Esantsi and Methodist, among others. They served a series of expert reports from a gastroenterologist, Dr. Todd Sheer, and a registered nurse, Julie Fomenko. Esantsi and Methodist both filed motions to dismiss. The trial court found that the expert reports were deficient as to both Esantsi and Methodist, denied the Curnels’ request for an extension to cure the deficiencies, and dismissed the Curnels’ claims with prejudice. The Curnels filed a motion for reconsideration, supported by amended expert reports, which the trial court denied as well. In three issues, the Curnels contend that the trial court abused its discretion by (1) granting Esantsi’s and Methodist’s motions to dismiss, (2) denying their request for an extension to cure, and (3) denying their motion for reconsideration. We hold that the expert reports were deficient but may be curable. Therefore, we reverse the trial court’s order dismissing the Curnels’ claims against Esantsi and Methodist and remand the case for further proceedings. Factual Background The expert reports of Sheer and Fomenko provide the background facts in this case. There are five reports from Sheer (one original and four supplemental) and three reports from Fomenko (one original and two supplemental). We accept the expert reports’ factual statements for the limited purpose of this appeal and do not address the merits of the Curnels’ claims. See Bowie Mem’l Hosp. v. Wright, 3 79 S.W.3d 48, 52 (Tex. 2002) (per curiam) (review of Chapter 74 report is limited to four corners of report). Curnel is prescribed an antibiotic that can cause elevated liver enzymes On October 4, 2015, Nancy Curnel presented to a local walk-in clinic “with a complaint of two days of subjective fever, rash, chronic cough, generalized pain, nasal congestion, dysuria, and diarrhea.” At the time, Curnel was on “multiple medications,” including “acyclovir, alprazolam, duloxetine, hydrochlorothiazide, lisdexamfetamine, levothyroxine, zolpidem, and nasonex.” Curnel was examined by a nurse practitioner, who diagnosed Curnel with a urinary tract infection and prescribed her the antibiotic nitrofurantoin, “100mg twice a day for 7 days.” Nitrofurantoin (also known as Macrodantin or Macrobid) is known for potential hepatotoxic effects and can cause drug-induced liver injury (DILI). Curnel presents to the Methodist ER with elevated liver enzymes Four days later, on Thursday, October 8, Curnel presented to the Methodist emergency room. According to Dr. Sheer’s expert reports, Curnel had been sent to the ER “by her primary medical provider to find out why recent blood work identified abnormal liver tests (hepatitis).” Curnel’s “medical history was negative for drug or alcohol abuse and chronic liver disease.” “She was taking several medications for longstanding medical conditions, with no history that these medications had caused hepatotoxicity in the past.” The only “new drug” she was 4 taking was the nitrofurantoin. After she began taking nitrofurantoin, “liver function studies performed by her primary care provider demonstrated elevated values.” Curnel told the nursing staff that she had been sent to the ER by her primary care physician because recent blood work indicated that she had elevated liver enzymes. An ER nurse took Curnel’s medical history, noting that “Curnel had recently started nitrofurantoin and was then referred to the hospital for evaluation of elevated liver enzymes.” However, there is no documentation that the nurse evaluated Curnel’s “current medications, including nitrofurantoin,” for “potential hepatotoxicity.” After the nurse took Curnel’s medical history, Curnel was examined by an emergency medicine physician, Dr. Scott Wiesenborn. Wiesenborn noted that Curnel had recently begun taking nitrofurantoin and had been referred to the ER for elevated liver enzymes. Wiesenborn ordered that Curnel’s liver enzymes be tested, and the results confirmed that several of her liver enzymes were abnormally high, including her alanine transaminase (ALT), aspartate transaminase (AST), and alkaline phosphatase (ALP). Wiesenborn diagnosed Curnel with a fever and “elevated liver function tests/probable acute hepatitis,” although he did not specify whether it was “drug, viral, autoimmune, etc.” 5 Curnel is hospitalized Wiesenborn called the on-duty hospitalist, Dr. Michael Esantsi, to determine whether to admit Curnel for hospitalization. According to Sheer’s expert reports, Esantsi’s “history and physical” was “not significantly different from the one performed in the ER” by Wiesenborn. Esantsi documented that Curnel had started nitrofurantoin, but he did “not list this in his Medications section.” He diagnosed Curnel with “abnormal liver function tests,” specifying that the probable cause was “viral-induced hepatitis.” According to Sheer, “[d]rug-induced liver injury [was] not considered in the history and physical,” and Esantsi’s “plan include[d] continuing current regular home medications.” Esantsi ordered that Curnel take “nitrofurantoin 100 mg capsule oral two times daily” and that she undergo a gastrointestinal consultation. One of the nurses documented Esantsi’s order concerning Curnel’s medications, “which included nitrofurantoin SR (Macrobid) 100 mg capsule oral two times daily.” There is no documentation that the nurses ever developed “a plan for an evaluation for the hepatotoxic potential of the medications, and there is no documentation elsewhere in the medical record that such evaluation was completed.” That afternoon, Curnel was transferred from the ER to the medical-surgical unit, where she was examined by a gastroenterologist, Dr. Steven Ugbarugba. 6 Ugbarugba’s consultation notes made “no mention” of Curnel “starting nitrofurantoin recently” and “omit[ted] a list of her medications.” Ugbarugba performed a number of tests, which ruled out a number of potential causes of Curnel’s elevated liver enzymes.2 Ugbarugba did not note nitrofurantoin as a potential cause of Curnel’s elevated liver enzymes. Again quoting from Sheer’s reports, there is “no documentation that drug-induced liver injury [was] considered” by Ugbarugba at that time. Later that evening, Curnel received a dose of nitrofurantoin, which was “dispensed from the pharmacy” and “administer[ed]” by one of the nurses. On the morning of Friday, October 9, Curnel’s liver enzymes were tested again. The test results showed a “slight increase” in both her ALT and AST. Curnel underwent additional testing, which indicated that she did not have viral hepatitis but was suffering from DILI instead. A complete blood count (CBC) revealed that Curnel had eosinophilia, a condition that is “indicative” or “suggestive” of DILI. According to Sheer, Esantsi’s notes indicate that, when he saw Curnel again that day, he “repeat[ed] the running diagnosis of probable viral hepatitis, despite negative test results,” with “no documentation that drug-induced liver injury [was] 2 Specifically, Ugbarugba ruled out autoimmune hepatitis, viral hepatitis, lupus hepatitis, celiac disease, Epstein-Barr virus, Parvovirus PCRs, primary biliary cholangitis, and ulcers. 7 considered.” At the nurses’ request, Esantsi ordered that Curnel begin to take acetaminophen, “another well-known hepatotoxic medication.” A nurse administered the medication to Curnel that afternoon. That evening, Curnel received another dose of nitrofurantoin, “dispensed from the pharmacy” and “administer[ed]” by another nurse. On the morning of Saturday, October 10, Curnel received yet another dose of nitrofurantoin. Her liver enzymes were tested again, and the results showed increases in her ALT and ALP, and a slight decrease in her AST, which nevertheless remained abnormally high. The results also showed a “slight rise” in her bilirubin. Later that day, she was examined again by Dr. Ugbarugba, who ordered that Curnel undergo a liver biopsy. Ugbarugba wrote a progress note, which “reaffirm[ed]” the general diagnosis of “acute hepatitis” but added that the etiology remained “unclear” and that there was a “possibility” that the hepatitis was “medication-induced.” Sheer’s expert reports state that this is the “first documentation” of a medical provider considering DILI as a potential cause of Curnel’s elevated liver enzymes. Ugbarugba “recommend[ed] holding Acyclovir, Vyvance, and Tylenol specifically,” and he made “a general suggestion to ‘hold hepatotoxic medications.’” 8 Later that afternoon, Curnel was examined by another physician, Dr. Yamini Naygandhi, who was “covering for Esantsi.” Naygandhi noted that Curnel might be suffering from “medication related hepatitis” (i.e., DILI) instead of “viral” hepatitis and ordered a review of Curnel’s medications “to find out what [was] causing [her] elevated LFT [liver function tests].” She also ordered that Curnel discontinue nitrofurantoin.3 By Sunday, nitrofurantoin was “no longer on [Curnel]’s patient medication list and was not administered.” Her bilirubin and ALP increased, but her ALT and AST decreased. Curnel’s artery is nicked during her liver biopsy On the morning of Monday, October 12, the day of Curnel’s scheduled liver biopsy, Curnel’s enzymes showed “further improvement.” Her bilirubin began to decrease, her AST continued to decrease, and her ALT and ALP underwent “non- significant changes.” At 9:00 a.m., Curnel was examined for a third time by Ugbarugba. His progress note contained the “exact” same “assessment from the day prior” except that it noted, “Liver bx today.” “A pre-procedure prothrombin time/INR ordered by 3 None of the physicians specifically ordered that Curnel discontinue acetaminophen. 9 Dr. Esantsi return[ed] as normal (this was the first time checked since presentation).”4 At 10:30 a.m., a radiologist, Dr. Mark Brodie, performed the biopsy. He obtained two “cores,” which showed that “the liver function abnormalities were due to medication effects.” During the biopsy, Brodie nicked Curnel’s artery, causing severe injuries. Curnel required multiple blood transfusions, medications to maintain circulation, mechanical ventilation, prolonged resuscitation, and extended ICU care. According to Sheer, Curnel “will have a slow and painful recovery.” Procedural History The Curnels asserted health care liability claims against Esantsi and Methodist, among others. The Curnels served Esantsi and Methodist with a series of expert reports from Sheer and Fomenko. Sheer’s reports addressed both Esantsi and Methodist, while Fomenko’s addressed only Methodist. Esantsi and Methodist filed objections and motions to dismiss for failure to serve adequate expert reports. The Curnels responded that the reports were adequate. They requested that the trial court deny the motions to dismiss or, alternatively, grant them a 30-day extension to cure any deficiencies in the reports. 4 “A prothrombin time test measures how quickly your blood clots.” Prothrombin time test, MAYO CLINIC (May 10, 2018), https://www.mayoclinic.org/tests- procedures/prothrombin-time/about/pac-20384661. 10 The trial court heard the motions to dismiss and found that the combined expert reports of Sheer and Fomenko were deficient as to both Esantsi and Methodist. The trial court sustained Esantsi’s and Methodist’s objections, denied the Curnels’ request for a 30-day extension to cure, and dismissed the Curnels’ claims with prejudice. The Curnels filed a motion for reconsideration, supported by amended expert reports. The trial court denied the motion, and the Curnels appealed. Motions to Dismiss In their first issue, the Curnels contend that the trial court abused its discretion in granting Esantsi’s and Methodist’s motions to dismiss for failure to serve adequate expert reports. A. Applicable law and standard of review Under the Medical Liability Act, a plaintiff asserting health care liability claims must timely serve each defendant physician and health care provider with one or more expert reports and a curriculum vitae of each expert whose opinion is offered to substantiate the merits of the claims. TEX. CIV. PRAC. & REM. CODE § 74.351(a), (i); see Mangin v. Wendt, 480 S.W.3d 701, 705 (Tex. App.— Houston [1st Dist.] 2015, no pet.). The expert report must provide a “fair summary” of the expert’s opinions regarding the (1) applicable standards of care, (2) manner in which the care rendered by the physician or health care provider 11 failed to meet the standards, and (3) causal relationship between that failure and the injury, harm, or damages claimed. TEX. CIV. PRAC. & REM. CODE § 74.351(r)(6). “No particular words or formality are required, but bare conclusions will not suffice.” Scoresby v. Santillan, 346 S.W.3d 546, 556 (Tex. 2011). Instead, the report must explain the basis of the expert’s statements and link the expert’s conclusions to the facts of the case. Jelinek v. Casas, 328 S.W.3d 526, 539 (Tex. 2010). For standard of care and breach, the expert report must explain what the physician or health care provider should have done under the circumstances and what the physician or health care provider did instead. Am. Transitional Care Ctrs. of Tex., Inc. v. Palacios, 46 S.W.3d 873, 880 (Tex. 2001). For causation, the expert report must explain how and why the physician’s or health care provider’s breach proximately caused the plaintiff’s injury. Columbia Valley Healthcare Sys., L.P. v. Zamarripa, 526 S.W.3d 453, 459–60 (Tex. 2017). Proximate cause has two components: (1) cause-in-fact and (2) foreseeability. Id. at 460. A physician’s or health care provider’s breach was a cause-in-fact of the plaintiff’s injury if the breach was a substantial factor in bringing about the harm, and absent the breach (i.e., but for the breach) the harm would not have occurred. Id. Even if the harm would not have occurred absent the defendant’s breach, “the connection between the defendant and the plaintiff’s 12 injuries simply may be too attenuated” for the breach to qualify as a substantial factor. Allways Auto Grp., Ltd. v. Walters, 530 S.W.3d 147, 149 (Tex. 2017) (per curiam) (quoting Union Pump Co. v. Allbritton, 898 S.W.2d 773, 776 (Tex. 1995)). A breach is not a substantial factor if it “does no more than furnish the condition that makes the plaintiff’s injury possible.” Id. A physician’s or health care provider’s breach is a foreseeable cause of the plaintiff’s injury if a physician or health care provider of ordinary intelligence would have anticipated the danger caused by the negligent act or omission. See Price v. Divita, 224 S.W.3d 331, 336 (Tex. App.—Houston [1st Dist.] 2006, pet. denied). The plaintiff may serve reports of separate experts regarding different physicians or health care providers or regarding different issues arising from the conduct of a single physician or health care provider. TEX. CIV. PRAC. & REM. CODE § 74.351(i). However, only a qualified physician may give opinion testimony about the causal relationship between the claimed injury, harm, or damages and the alleged departure from the applicable standard of care. See id. § 74.351(r)(5)(C). The expert report is not required to prove the plaintiff’s case but only to provide notice of the conduct forming the basis of the plaintiff’s claim. Gracy Woods I Nursing Home v. Mahan, 520 S.W.3d 171, 189 (Tex. App.—Austin 2017, no pet.). The report “need not anticipate or rebut all possible defensive theories that may ultimately be presented” in the case. Owens v. Handyside, 478 S.W.3d 172, 13 187 (Tex. App.—Houston [1st Dist.] 2015, pet. denied). Nor must the report “rule out every possible cause of the injury, harm, or damages claimed.” Baylor Med. Ctr. at Waxahachie, Baylor Health Care Sys. v. Wallace, 278 S.W.3d 552, 562 (Tex. App.—Dallas 2009, no pet.). In reviewing the adequacy of an expert report, a trial court may not consider an expert’s credibility, the data relied upon by the expert, or the documents that the expert failed to consider at this pre-discovery stage of the litigation. See Mettauer v. Noble, 326 S.W.3d 685, 691 (Tex. App.—Houston [1st Dist.] 2010, no pet.); Gonzalez v. Padilla, 485 S.W.3d 236, 245 (Tex. App.—El Paso 2016, no pet.). Instead, the trial court must limit its review to the “four corners” of the expert report and, when the question of adequacy hinges on the expert’s qualifications, the “four corners” of the expert’s curriculum vitae. Mangin, 480 S.W.3d at 706. The statute’s purpose is not to determine the merits of the claim but to rule out frivolous lawsuits at the onset of litigation, before the parties have conducted full discovery. Ross v. St. Luke’s Episcopal Hosp., 462 S.W.3d 496, 502 (Tex. 2015); Mangin, 480 S.W.3d at 706. As we have explained: The requirement to serve an expert report arises at the outset of litigation and before the opportunity for the plaintiff to engage in significant discovery, including taking oral depositions of the defendants. As such, the statute itself contemplates that the amount and quality of evidence available at the time of drafting the expert reports will be less than that available at trial on the merits or even the summary-judgment stage. 14 Mangin, 480 S.W.3d at 713 (citations omitted). Thus, the requirements of the statute have been variously described as a “lenient standard,”5 “low threshold,”6 and “relatively low bar.”7 If the plaintiff “fails to timely serve an expert report, then on the affected health care provider’s motion the trial court must dismiss the pertinent health care liability claim with prejudice and award attorney’s fees.” Baty v. Futrell, 543 S.W.3d 689, 692 (Tex. 2018) (citing TEX. CIV. PRAC. & REM. CODE § 74.351(b)). “However, if the motion challenges the adequacy of an otherwise timely report, the trial court may grant the motion ‘only if it appears to the court, after hearing, that the report does not represent an objective good faith effort to comply with the [Act’s] definition of an expert report.’” Baty, 543 S.W.3d at 692–93 (quoting TEX. CIV. PRAC. & REM. CODE § 74.351(l)). A report qualifies as an objective good faith effort to avoid dismissal if it discusses each element with sufficient specificity that it (1) informs the defendant of the specific conduct the plaintiff questions and (2) provides a basis for the trial 5 Scoresby v. Santillan, 346 S.W.3d 546, 549 (Tex. 2011). 6 Loaisiga v. Cerda, 379 S.W.3d 248, 264 (Tex. 2012) (Hecht, J., concurring in part and dissenting in part) (“An expert report, as we have interpreted it, is a low threshold a person claiming against a health care provider must cross merely to show that his claim is not frivolous.”). 7 Baty v. Futrell, 543 S.W.3d 689, 698 (Tex. 2018) (Johnson, J., dissenting) (describing medical expert report requirements as interpreted by majority). 15 court to conclude that the plaintiff’s claims have merit. Miller v. JSC Lake Highlands Operations, LP, 536 S.W.3d 510, 513 (Tex. 2017) (per curiam). In determining whether an expert report constitutes an objective good faith effort to address each element, “a trial court may not draw inferences; instead, it must exclusively rely upon the information contained within the four corners of the report.” Cornejo v. Hilgers, 446 S.W.3d 113, 123 (Tex. App.—Houston [1st Dist.] 2014, pet. denied); see Baty, 543 S.W.3d at 693. We review a trial court’s ruling on a motion to dismiss a health care liability claim for an abuse of discretion. Van Ness v. ETMC First Physicians, 461 S.W.3d 140, 142 (Tex. 2015) (per curiam). Under this standard, we “defer to the trial court’s factual determinations if they are supported by evidence, but review its legal determinations de novo.” Id. “A trial court abuses its discretion if it rules without reference to guiding rules or principles.” Id. B. Adequacy of expert reports on Esantsi The Curnels supported their claim against Esantsi with expert reports from Sheer. The trial court found that Sheer’s reports were deficient on all three elements. 1. Sheer’s opinions on elements of standard of care and breach We begin by considering whether Sheer’s reports provide adequate opinions on the first two elements: standard of care and breach. In his reports, Sheer 16 explains that the standard of care required Esantsi to take certain actions and to refrain from taking others throughout the course of Curnel’s treatment at Methodist. Sheer writes that, as Curnel’s “primary physician” at the hospital, Esantsi was “responsible” for the “decision to admit and discharge” her. At the “most basic” level, a hospitalist like Esantsi must “understand the effect of medications on a patient, especially when evaluating liver function.” When a patient presents with elevated liver enzymes, “medications should be at the top of the list of potential causes, which is especially true when the patient has recently initiated a course of a frequently implicated agent such as nitrofurantoin.” And, Sheer continues, when a patient’s elevated liver enzymes are “drug-induced,” a liver biopsy is “rarely helpful.” According to Sheer, when Curnel presented to the hospital after having recently initiated an antibiotic well-known for its hepatotoxic potential, the standard of care required Esantsi to “evaluate the hepatotoxic potential” of Curnel’s medications; “recognize drug-induced liver injury as the probable cause of [her] elevated liver enzymes”; “discontinue the offending drug,” nitrofurantoin; refrain from ordering other hepatotoxic medications; and “discharge her to be followed as an outpatient with her primary care physician to confirm normalization of the liver tests on serial lab testing.” Sheer explains that Esantsi breached this 17 standard by failing to take any of these steps. Esantsi failed to evaluate the hepatotoxic potential of her medications, recognize DILI as the probable cause of her elevated liver enzymes, discontinue the antibiotic, refrain from ordering other hepatotoxic medication, and discharge her for outpatient monitoring. Instead, Esantsi misdiagnosed Curnel with probable viral hepatitis, ordered that she continue to take nitrofurantoin and begin to take hepatotoxic acetaminophen, and admitted her for hospitalization and “unnecessary testing.” After Curnel was admitted for hospitalization, Sheer explains, Esantsi continued to be responsible for deciding when to discharge her, and the standard of care required him to recognize that a liver biopsy was not “warranted” or “justified” under the circumstances.8 Sheer further explains that Esantsi breached this standard by “maintaining [Curnel’s] admission to the hospital for the biopsy” and ordering a pre-procedure prothrombin time test “in preparation of the liver biopsy.” Thus, in his reports, Sheer explains in detail what he believes the standard of care required Esantsi to have done under the circumstances and what Esantsi did 8 Sheer explains: “A physician must have evidence that the information to be obtained from a liver biopsy warrants the risk before the procedure is performed. That did not exist in this case. A liver biopsy is most often performed for evaluation of abnormal liver tests once medication review, thorough history and physical examination, and complete blood testing has been performed and are not indicative of a cause. Liver biopsy is not indicated for a stable patient without evidence of acute liver failure (which Ms. Curnel did not have) when the workup (including medication review) is incomplete.” 18 instead. See Palacios, 46 S.W.3d at 880. We hold that Sheer’s reports provide adequate opinions on the standard of care applicable to Esantsi and the manner in which the care rendered by Esantsi failed to meet that standard. See TEX. CIV. PRAC. & REM. CODE § 74.351(r)(6). 2. Sheer’s opinion on element of causation We next consider whether Sheer’s reports provide an adequate opinion on the third element: causation. To show causation, Sheer explains a chain of events that begins with Esantsi’s initial breaches—Esantsi’s failure to evaluate Curnel’s medications for hepatotoxic potential when she first presented, to diagnose her with DILI, and to discharge her for outpatient monitoring. He instead misdiagnosed her with viral hepatitis, ordered that she continue to take the contraindicated antibiotic and another hepatotoxic medication, and improperly admitted her for hospitalization and unnecessary testing. The chain of the events then continues to Esantsi’s subsequent breaches—Esantsi’s act and omissions in continuing to misdiagnose Curnel with viral hepatitis, maintaining her admission to the hospital, and ordering the prothrombin time test in preparation for the biopsy. And the chain ends with the unwarranted biopsy taking place as scheduled. The chain of events leading up to Curnel’s injuries includes many other acts by many other actors. These actors notably include Ugbarugba, who ordered the liver biopsy even though he considered DILI a potential cause of Curnel’s elevated 19 liver enzymes and Curnel showed improvement after nitrofurantoin was discontinued, and Brodie, who actually performed the biopsy and nicked Curnel’s artery. An event that starts a chain of events can be too attenuated from an injury to cause it. See Providence Health Ctr. v. Dowell, 262 S.W.3d 324, 330 (Tex. 2008) (holding that medical providers’ “negligence was too attenuated from the [harm] to have been a substantial factor in bringing it about”); Shenoy v. Jean, No. 01-10- 01116-CV, 2011 WL 6938538, at *9 (Tex. App.—Houston [1st Dist.] Dec. 29, 2011, pet. denied) (mem. op.) (“A causal link can be too attenuated to satisfy the causation requirement for an expert report.”). It is not enough that one event occurred before the other; that is only evidence of but-for causation. Shenoy, 2011 WL 6938538, at *9; see Jelinek, 328 S.W.3d at 533 (“Care must be taken to avoid the post hoc ergo propter hoc fallacy, that is, finding an earlier event caused a later event merely because it occurred first.”). Rather, the event must have been a substantial factor in bringing about the harm. Zamarripa, 526 S.W.3d at 460. And an event is not a substantial factor if it is “too attenuated” from the harm or “does no more than furnish the condition that makes the plaintiff’s injury possible.’” Walters, 530 S.W.3d at 149 (quoting Allbritton, 898 S.W.2d at 776). Esantsi’s initial breaches—the breaches he committed when Curnel first presented to Methodist—are too attenuated from Curnel’s injuries to be considered 20 a substantial factor in bringing those injuries about. It is true that, if Esantsi had followed the identified standard of care when determining whether to admit Curnel—had he recognized that Curnel was likely suffering from DILI caused by the recently-prescribed nitrofurantoin, discontinued the nitrofurantoin, and discharged her for outpatient monitoring—then Curnel’s liver would have never been biopsied, and her injuries would have never occurred. Nevertheless, Esantsi’s initial breaches, which occurred days before the biopsy and before she was even admitted to the hospital, did “no more than furnish the condition” that made Curnel’s injury “possible.” Id. That is, as a result of Esantsi’s initial breaches, Curnel was admitted for hospitalization, which made it possible for Ugbarugba to order the biopsy and for Brodie to perform the biopsy, among other causal links in the chain. Given the numerous different acts by other physicians and nurses during the multiple days between Curnel’s admission and her biopsy, Esantsi’s initial breaches do not constitute a cause-in-fact of Curnel’s injuries. See Zamarripa, 526 S.W.3d at 461 (holding that expert reports were deficient because they failed to explain how hospital proximately caused death of pregnant patient); Dowell, 262 S.W.3d at 330 (holding that “the defendants’ negligence was too attenuated from the [harm] to have been a substantial factor in bringing it about”); Shenoy, 2011 WL 6938538, at *9 (holding that expert report was deficient on causation because physician’s negligence in clearing patient for surgery was “too attenuated” from 21 patient’s death from post-surgery complications); cf. Allways Auto Grp., Ltd. v. Walters, 530 S.W.3d 147, 147–49 (Tex. 2017) (per curiam) (holding that car dealership did not proximately cause collision between intoxicated motorist and other driver by providing loaner vehicle to motorist 18 days before collision occurred, despite fact that motorist was also intoxicated when dealership provided him vehicle). In his reports, Sheer identifies two subsequent breaches by Esantsi: (1) “maintaining [Curnel’s] admission to the hospital for the biopsy” and (2) ordering a pre-procedure prothrombin time test “in preparation of the liver biopsy.” To conclude that either of these subsequent breaches was a substantial factor in bringing about Curnel’s injuries, we would have to infer that Esantsi had a duty to prevent the biopsy from taking place or that he participated in the decision to biopsy Curnel’s liver in a manner that was “collaborative,” much like the “the screening, diagnosis, and treatment” of the two defendant-physicians in Bustamante v. Ponte was “collaborative.” 529 S.W.3d 447, 457 (Tex. 2017). This is particularly true given that Ugbarugba possessed the information that DILI was a potential cause of Curnel’s elevated liver enzymes and nevertheless ordered a biopsy. The reports do not state how Esantsi “had either the right or the means to persuade” Ugbarugba and Brodie to cancel the biopsy. Zamarripa, 526 S.W.3d at 461. Nor do they otherwise explain whether Esantsi “had any say in the matter.” 22 Id. We hold that Sheer’s reports on Esantsi do not provide an adequate opinion on cause-in-fact. We further hold that Sheer’s reports do not provide an adequate opinion on foreseeability, either. In his reports, Sheer explains that, because the liver is “very vascular” and “it is not possible to visualize all of the blood vessels during the biopsy,” “in the process of removing pieces of liver tissues, there is [a] risk of cutting one or more blood vessels.” Sheer does not state, however, whether this risk was known to Esantsi—who is not a gastroenterologist and does not perform liver biopsies—or should have been recognized before the biopsy. Nor can we draw this inference from Sheer’s reports. Cornejo, 446 S.W.3d at 123. Sheer does not address whether the risk was generally known or recognized by hospitalists like Esantsi before the surgery. Nor does Sheer provide information demonstrating that the risk is part of the informed consent disclosures or that a hospitalist of ordinary intelligence would have anticipated the danger of a patient’s blood vessel being cut during this type of procedure. See Price, 224 S.W.3d at 336. We hold that Sheer’s reports on Esantsi fail to adequately address both components of proximate cause. C. Adequacy of expert reports on Methodist The Curnels supported their claim against Methodist with expert reports from Fomenko and Sheer. Fomenko’s expert reports address standard of care and 23 breach,9 while Sheer’s address causation. The trial court found that the combined expert reports were deficient on all three elements. 1. Fomenko’s opinions on elements of standard of care and breach We begin our analysis by considering whether Fomenko’s reports provide adequate opinions on the first two elements: standard of care and breach. See TEX. CIV. PRAC. & REM. CODE § 74.351(r)(6). With respect to the Curnels’ claim against Methodist for its allegedly inadequate policies, Fomenko opines that the standard of care required the hospital to have in place and enforce policies and procedures requiring all physicians, nurses, and pharmacists providing care to a patient to evaluate the patient’s medications for hepatotoxicity and other negative effects and contraindications through in-house computer formulary programs and pharmaceutical publications. She further opines that the policies and procedures should require such an evaluation to occur when the patient is admitted to the hospital and, once admitted, when the patient is prescribed a given medication. According to Fomenko, Methodist breached this standard because such policies and procedures were either not in place or not enforced, as evidenced by the fact that Curnel’s medications 9 Because Fomenko is a registered nurse and not a physician, she is not qualified to offer an opinion on causation. See TEX. CIV. PRAC. & REM. CODE § 74.351(r)(5)(C). 24 were not evaluated for hepatotoxicity when she was admitted to the hospital or before the nurses administered nitrofurantoin to her for three consecutive days. With respect to the Curnels’ vicarious liability claim based on negligence by the Methodist nursing staff, Fomenko opines that the standard of care required the nurses to evaluate Curnel’s medications; recognize that nitrofurantoin was hepatotoxic and thus contraindicated given Curnel’s elevated liver enzymes; refrain from administering nitrofurantoin to Curnel; notify the ordering practitioner, Esantsi, of the reason for their decision; and seek clarification of his order. According to Fomenko, the nurses breached the standard of care by failing to take any of these steps. The nurses failed to evaluate the hepatotoxic potential of Curnel’s medications when Curnel presented to the ER or when she was admitted to the hospital, and they failed to document the need to perform such an evaluation in Curnel’s plan of care. They noted Curnel had been taking nitrofurantoin, but they failed to recognize that nitrofurantoin is hepatotoxic and thus failed to clarify the contraindicated nitrofurantoin order with Esantsi or any other practitioner. And instead of holding the medication, they administered it to Curnel for three consecutive days. In her reports, Fomenko explains in detail what she believes the standard of care required Methodist and its nursing staff to have done under the circumstances 25 and what they did instead. Palacios, 46 S.W.3d at 880; see Gardner v. U.S. Imaging, Inc., 274 S.W.3d 669, 671–72 (Tex. 2008) (per curiam) (when health care liability claim involves vicarious liability theory, expert report that adequately implicates actions of party’s agents or employees is sufficient to implicate party itself). We hold that Fomenko’s expert reports provide adequate opinions on the standard of care applicable to Methodist and the manner in which the care Methodist rendered breached that standard. See TEX. CIV. PRAC. & REM. CODE § 74.351(r)(6). 2. Sheer’s opinion on element of causation We next consider whether Sheer’s reports provide an adequate opinion on the third element: causation. In his reports, Sheer opines that, by failing to promptly evaluate Curnel’s mediations and instead administering the contraindicated antibiotic for three consecutive days, Methodist and its nursing staff deprived Curnel’s physicians of “necessary data and information for making the correct diagnosis and considering discharge from the hospital.” According to Sheer, had the Methodist nurses complied with the standard of care identified by Fomenko, the physicians would have had the benefit of data showing a declining trend in Curnel’s liver enzymes, which, in turn, would have led them to diagnose her with DILI and discharge her for further monitoring on an 26 outpatient basis, thereby avoiding the liver biopsy and the injuries that Curnel received from it. [A]dministration of the nitrofurantoin to [Curnel] during her hospitalization (due to failure to evaluate the hepatotoxic potential of the medication) perpetuated the elevated liver enzymes. In reasonable medical probability, if the nurses had not administered nitrofurantoin, [Curnel]’s liver enzymes would have immediately improved. Thus, the nurses’ administration of the drug deprived the physicians of data relevant to a diagnosis of probable drug induced liver injury and discharge for outpatient monitoring. Sheer’s reports about the nurses, like his reports on Esantsi, attempt to show proximate cause by explaining a chain of events that begins with the Methodist nurses’ failure to evaluate Curnel’s medications and to refrain from administering the contraindicated nitrofurantoin and ends with Curnel’s artery being nicked and the resulting injuries. And like the alleged negligence of Esantsi, the alleged negligence of Methodist and its nursing staff is too attenuated from Curnel’s injuries to be considered a substantial factor in bringing those injuries about. Curnel’s physicians identified DILI as a potential cause of her elevated liver enzymes two days before the biopsy. Ugbarugba and Naygandhi both considered DILI a potential cause of Curnel’s elevated liver enzymes, both ordered that Curnel stop taking all hepatotoxic medications, and Naygandhi specifically ordered that Curnel stop taking nitrofurantoin. Once Curnel stopped taking nitrofurantoin, her enzymes began to improve. And, as Sheer states in his reports, by October 12, the liver tests showed even “further improvement.” None of Curnel’s physicians 27 ordered that the biopsy not take place. The physicians had the information to diagnose Curnel with DILI and discharge her from the hospital before the biopsy. Instead, armed with the very information that Curnel claims they needed, they elected to go forward with the biopsy on its scheduled date. Sheer’s reports do not explain how and why additional information from the nurses would have led the physicians to cancel the biopsy if the information the physicians already had did not or how the nurses “had either the right or the means to persuade” the physicians to cancel the biopsy. Zamarripa, 526 S.W.3d at 461. Nor do Sheer’s reports state that Methodist’s nurses were part of the decision to perform the biopsy or its timing. We hold that Sheer’s reports do not adequately address cause- in-fact for the claims against Methodist. We further hold that Sheer’s reports do not adequately address foreseeability, either. They do not explain how and why Methodist’s nurses should have anticipated Curnel’s artery being nicked because of either Methodist’s failure to implement and enforce policies and procedures requiring the evaluation of hepatotoxic medication or the nurses’ failure to evaluate Curnel’s medications for hepatotoxicity and to refuse to administer the drug. Thus, we hold that Sheer’s reports on Methodist fail to adequately address both components of proximate cause. 28 D. Conclusion on adequacy of expert reports In sum, we hold that Sheer’s reports on Esantsi provide adequate opinions on the standard of care and breach but do not provide an adequate opinion on either component of proximate cause. We further hold that Fomenko’s reports on Methodist provide adequate opinions on the standard of care and breach but that Sheer’s reports on Methodist do not provide an adequate opinion on either component of proximate cause. Therefore, the trial court did not abuse its discretion in finding the reports inadequate as to both Esantsi and Methodist. Accordingly, we overrule the Curnels’ first issue. Motions for Extension to Cure In their second and third issues, the Curnels contend that the trial court abused its discretion by denying their motion for a 30-day extension to cure the deficient expert reports and by denying their motion for reconsideration. Under the Act, if the plaintiff timely serves an expert report, and the trial court concludes that the report is an objective good faith effort to comply with the statute but nevertheless deficient in some way, the trial court has the discretion to grant the plaintiff one 30-day extension to cure the deficiencies. TEX. CIV. PRAC. & REM. CODE § 74.351(c); Mangin, 480 S.W.3d at 705–06. The trial court should err on the side of granting the extension. Samlowski v. Wooten, 332 S.W.3d 404, 416 (Tex. 2011) (Guzman, J., concurring) (“In order to preserve the highest number of 29 meritorious claims, trial courts should err on the side of granting claimants’ extensions . . . .”); see Samlowski, 332 S.W.3d at 411 (plurality op.) (agreeing with concurrence that trial court should err on side of granting extension). And the trial court must grant the extension if the deficiencies are curable. Zamarripa, 526 S.W.3d at 461. The Texas Supreme Court established a “minimal” standard for determining whether a deficient report is curable: “a 30–day extension to cure deficiencies in an expert report may be granted if the report is served by the statutory deadline, if it contains the opinion of an individual with expertise that the claim has merit, and if the defendant’s conduct is implicated.” Scoresby, 346 S.W.3d at 557. We review a trial court’s ruling on motion for an extension to cure a deficient expert report for an abuse of discretion. Quintero v. Hous. Methodist Hosp., No. 01-14-00448-CV, 2015 WL 831955, at *2 (Tex. App.—Houston [1st Dist.] Feb. 26, 2015, pet. denied) (mem. op.); Henry v. Kelly, 375 S.W.3d 531, 535 (Tex. App.—Houston [14th Dist.] 2012, pet. denied). The Curnels served Methodist and Esantsi with expert reports before the statutory deadline, Methodist and Esantsi filed objections to the reports, and the Curnels served Methodist and Esantsi with additional reports in response to the objections. As discussed above, despite their deficiencies, these reports contain the opinions of qualified experts that the Curnels’ claims had merit and implicated the 30 conduct of Methodist and Esantsi; they qualify as objective good faith efforts to comply with the statute. See Scoresby, 364 S.W.3d at 557. The trial court dismissed their claims without affording them an opportunity to cure their deficient reports. Part of the purpose of the extension is to afford a plaintiff who made a good faith effort the chance to cure a defective report after the deficiencies have been identified by the trial court. Given the Supreme Court’s minimal standard and the Curnels’ objective good faith efforts, we cannot say that the Curnels’ reports were incurable. See Zamarripa, 526 S.W.3d at 461. Moreover, when the trial court dismissed the Curnels’ claims, it was unclear whether the statute required expert reports to address foreseeability, and at least two courts of appeals had held that it did not. See Rio Grande Reg’l Hosp. v. Ayala, No. 13-11-00686-CV, 2012 WL 3637368, at *19 (Tex. App.—Corpus Christi Aug. 24, 2012, pet. denied) (mem. op.), abrogated by Columbia Valley Healthcare Sys., L.P. v. Zamarripa, 526 S.W.3d 453 (Tex. 2017); Adeyemi v. Guerrero, 329 S.W.3d 241, 246 (Tex. App.—Dallas 2010, no pet.). The Texas Supreme Court resolved the issue while this appeal was pending when it issued Zamarripa and held that an expert report must address both cause-in-fact and foreseeability. 526 S.W.3d at 460. Given this development, the Curnels’ should be afforded the opportunity to amend their reports to address foreseeability and to cure the other deficiencies identified in this opinion. 31 Therefore, we hold that the trial court abused its discretion in denying their motion for an extension to cure and motion for reconsideration. Accordingly, we sustain the Curnels’ second and third issues. Conclusion We reverse the trial court’s order and remand for further proceedings. Harvey Brown Justice Panel consists of Justices Jennings, Bland, and Brown. 32