OPINION ON REHEARING
Evelyn V. Keyes, JusticeAppellee, Tracy Windrum, moved for rehearing and en banc reconsideration of our April 19, 2016 opinion and judgment in this case. We withdraw our April 19, 2016 opinion and judgment and issue this opinion and judgment in their stead. Our disposition remains unchanged.
In this wrongful death case, Tracy Windrum, individually, as representative of the estate of Lancer Windrum, her hus*500band, and on behalf of her minor children, B.W, J.W., and H.W., sued Dr. Victor Kareh for medical malpractice. After a jury trial, the jury found Dr. Kareh 80% negligent and awarded a total of $4,239,464 to Windrum in damages. After applying settlement credits and statutory damages caps, the trial court entered judgment in favor of Windrum, awarding her $1,875,887.62 in damages. In seven issues, Dr. Kareh contends that (1) Windrum failed to present legally and factually sufficient evidence that he was negligent: (2) Windrum failed to present legally and factually sufficient evidence that his negligence caused Lance Windrum’s death: (3) the trial court erroneously admitted expert testimony and accompanying photographs that were not timely produced: (4) the trial court erroneously denied his motion for mistrial made after the court informed the jury that the parties had been to mediation and tried to settle: (5) the trial court erroneously excluded on the basis of the Texas Deadman’s Rule testimony from one of the physicians involved concerning statements made to her by the decedent: (6) the foregoing errors constituted cumulative error: and (7) the trial court erroneously applied the statutory damages caps applicable to the recovery of non-economic damages in wrongful death cases.
We reverse and render.
Background
A. Factual Background
On February 3, 2010, forty-six-year-old Lancer (“Lance”) Windrum was out shopping with his three children when he started slurring his speech, became confused and disoriented, and hit his head while trying to climb back into his car. An ambulance took Lance to the North Cypress Medical Center (“NCMC”), where he worked as the Director of Radiology. Lance reported to his treating physicians that he had had three similar “episodes” over the past several months, which involved “very mild” slurring of his speech that resolved over the course of several hours. During the third episode, which occurred on Christmas Eve 2009 and was “pretty similar” to the February episode, Lance had felt confused, his balance had been impaired, and he had had tremors in his left hand and leg. Lance told his physicians that, on each of these occasions, he “was back to his baseline” within a matter of hours. Lance also reported that he had contracted encephalitis, a brain infection, when he was six years old.
Dr. Carrie Blades, the attending emergency room physician, ordered that Lance undergo a CT scan of his head. The lateral and third ventricles of the brain produce cerebrospinal fluid, which flows through an aqueduct into the fourth ventricle of the brain and then into the spinal column before it is later absorbed into the body through the venous system. The CT scan report noted that the ventricles in Lance’s brain were “dilated out of proportion,” indicating hydrocephalus. Dr. Blades ordered that Lance undergo an MRI. Dr. Christina Payan, the neuroradiologist who read the MRI scan, reported the following findings: “The lateral and third ventricles are markedly dilated out of proportion with the fourth ventricle and sulci. The cerebral aqueduct is narrowed. These findings are indicative of aqueductal stenosis [i.e., the narrowing of the aqueduct that carries cerebrospinal fluid through the brain]. There is some white matter atrophy. No significant transependymal [cere-brospinal fluid] flow is evident. ... No masses are present.”1
*501Lance then consulted Dr. Harpaul Gill, a neurologist at NCMC.2 Dr. Gill agreed that, at the time he presented to NCMC, Lance was experiencing symptoms of a neurological condition. During the consultation, Dr. Gill came to the conclusion that Lance’s symptoms might be caused by an increase in intracranial pressure due to a build-up of cerebrospinal fluid in the ventricles of Lance’s brain, and he told Lance that a shunt was a possible treatment to drain the excess fluid from the brain. Dr. Gill referred Lance to Dr. Kareh, a neurosurgeon, to determine whether Lance had increased intracranial pressure which would require surgery to alleviate.3
Dr. Kareh first saw Lance around 6:00 a.m. on February 4, 2010. Dr. Kareh testified that he did not review Lance’s medical history prior to meeting with him. Lance did not have any of the symptoms that he had displayed when he presented to NCMC the previous evening. All of Lance’s cranial nerves exhibited normal functioning. Dr. Kareh testified that double vision and papilledema, or swelling around the optic nerve, are both common symptoms that occur when a patient has increased intracranial pressure. Lance did not have double vision or papilledema at the time Dr. Kareh examined him. Dr. Kareh informed Lance that if he had increased intracranial pressure, he might need to have a shunt placed to drain the built-up cerebrospinal fluid. Lance consented to the placement of a ventricular drain and a device to monitor his intracra-nial pressure to determine whether it was increased.
Dr. Kareh monitored Lance’s intracranial pressure over a twenty-four hour period. Lance did not have increased intracra-nial pressure at the time that Dr. Kareh placed the monitoring device inside his brain. During the monitoring period, Lance’s intracranial pressure spiked on several occasions to a higher level than what is considered “normal.” However, Lance’s intracranial pressure quickly returned to a normal level on each occasion, and he did not experience any periods of sustained increased intracranial pressure. After the monitoring period ended, Dr. Kareh concluded that Lance’s intracranial pressure levels were normal, his neurological examination was normal, and he was not suffering from any symptoms such as confusion, imbalance, weakness, or numbness. Dr. Kareh determined that, although Lance had hydrocephalus, he did not have increased intracranial pressure. He therefore did not place a shunt.
Dr. Gill saw Lance for a follow-up appointment on February 17, 2010. Lance reported that he had had “one to two headaches every week,” but he had not experienced nausea, vomiting, focal weakness, numbness, visual disturbances, or sensitivity to light or sound. Dr. Gill performed a neurological examination, and the results were “normal.” Dr. Gill and Lance discussed medication for Lance’s headaches, but Lance decided against this course of action because he was “feeling *502better.” Dr. Gill directed. Lance to visit the emergency room if he experienced any more neurological symptoms, and he recommended that Lance undergo another MRI scan in three months and that Lance keep track of the headaches he experienced. Dr. Gill gave Lance a “headache calendar” to keep track of the days on which he experienced headaches.
Lance saw Dr. Kareh for a follow-up appointment on February 22, 2010. Lance reported that he had had one headache episode since he had been discharged from the hospital, which Dr. Kareh testified was expected due to the surgical procedure he had undergone, and one episode of slurred speech. Dr. Kareh recommended that Lance undergo a nuclear cisternogram to track the circulation of cerebrospinal fluid throughout his body, and he also recommended that Lance consult an endocrinologist to rule out a hormonal -cause, to his neurological symptoms. Dr. Kareh did not see Lance again after the February- 22 appointment. Lance did not have a nuclear cisternogram performed. Lance did see an endocrinologist on March 24, 2010, and testing conducted by this doctor revealed no problems with Lance’s endocrine system that might have caused his symptoms.
On his headache calendar, Lance self-reported taking two Lortabs for headache-related pain on two occasions during April 2010. He also underwent a second MRI scan in April 2010 with the findings reported to Dr. Gill, Dr. Payan again read the MR-1 scan and testified, that “[t]he ventricles looked as big, if not worse in size, and the angle- of the aqueduct had notably changed” since the February MRI. Dr. Payan called Dr. Gill and reported her findings to him. Dr. Gill did-not discuss the results of this MRI- with Lance, but Lance did undergo an EEG on April 29, 2010, at Dr. Gill’s direction. The results-of this test were normal. There is no evidence that either Dr. Gill or Dr. Payan informed Dr. Kareh of Lance’s symptoms after the February follow-up appointment or of the results of the April MRI scan.
Lance passed away in his sleep on May 2, 2010. Lance had reportedly complained to Windrum the previous day that he felt tired, sluggish, and irritable, and he had slurred speech. Lance did not self-report experiencing any headaches for the ten days prior to his death, which included his second MRI, showing a notably changed aqueduct and worsened ventricles, and a normal EEG.
Dr. Morna Gonsoulin, a medical examiner for the Harris County Institute of Forensic Sciences, performed an autopsy on Lance. Dr. Gonsoulin noted that Lance’s heart was enlarged and that the chambers of the heart were dilated. Dr. Gonsoulin made the following findings relevant to Lance’s brain:
The leptomeninges are clear. There is no epidural, subdural, or subarachnoid hemorrhage. The cerebral hemispheres are generally symmetrical with a relatively unremarkable gyral pattern. The vessels at the base of the brain are normally configured without atherosclerosis. The cranial nerves appear unremarkable. Sections through the cerebrum reveal markedly expanded lateral ventricles with rostral and caudal extensions to the frontal and occipital poles, respectively. The left hippocam-pus has slightly more prominent gray matter than the right hippocampus. There is decreased periventricular white matter surrounding the dentate nuclei of the cerebellum with expanded nuclear outlines abutting the ventricular border and no intervening white matter. A 0.5 centimeter cystic membrane is adjacent to the left dentate nucleus near the ventricle with interruption of the nuclear outline and slightly more white matter *503compared to that of the right. The per-iaqueductal gray matter is blurred with prominent stenosis of the aqueduct at the level of the cerebral pedicles. The diameter of the aqueduct ranges from pinpoint to non-visible, obscured by ill-defined light tan gelatinous gray material. Slightly increased gray matter is noted in the crossing fibers of the pons. No discrete areas of hemorrhage, infection or neoplasm are apparent.
(Emphasis added.) In the “Microscopic Examination” section of the autopsy report, Dr. Gonsoulin stated, “Sections from ros-tral pons through medulla show marked stenosis of aqueduct with gliosis[, i.e., scarring] of adjacent structures.” Dr. Gonsou-lin listed “[complications of hydrocephalus due to aqueductal stenosis” as Lance’s cause of death.
B. Procedural Background
Windrum, in her individual capacity, in her capacity as the representative of Lance’s estate, and on behalf of her three minor children, brought a negligence cause of action against Dr. Kareh and Dr. Gill pursuant to Texas’s wrongful death statute. Windrum alleged that the applicable standard of care when Lance was seen by Dr. Kareh at NCMC on February 4 required Dr. Kareh to install a shunt, or a permanent drain, in Lance’s brain to prevent a fatal build-up of cerebrospinal fluid and intracranial pressure. Dr. Gill settled before trial.
Windrum retained Dr. Robert Parrish, a neurosurgeon, to testify concerning the standard of care and causation, and she retained Dr. Ljubisa Dragovic, a forensic and neuropathologist, to testify concerning causation. Dr. Kareh filed a Daubert motion challenging both experts’ opinions on causation, arguing that neither doctor has “a sufficient scientific .and/or factual basis to render such opinions and such opinions are based on pure speculation and mere conjecture and do not pass the Analytical Gap test.” Dr. Kareh also argued that the methodology underlying Dr. Parrish’s and Dr. Dragovic’s opinions “is based on speculation and is unreliable.” The trial court overruled this motion.
Dr. Parrish testified that his opinion was that “Dr. Kareh should have put a shunt in when he saw Mr. Windrum in the hospital” on February 4 and that Lance “died of obstructive hydrocephalus.”4 When asked how Lance died, Dr. Parrish testified,
His aqueduct obstructed. There’s pressure in the ventricles. It put pressure on the red nuclei and the periaqueductal region right around where all that important stuff is. And those fibers made him stop breathing and his heart stop beating. ... But all those vital structures stopped because of pressure on the top of the brain stem where he is most susceptible with the aqueductal stenosis.
He stated that Lance “had these classic symptoms of increased intracranial pressure with staggering, slurred speech, and altered mental status that were periodic.” He discounted the significance of the absence of papilledema in Lance’s eyes— likewise a classic symptom of increased intracranial pressure—and he testified that papilledema can be intermittent and did not have to be present for Lance to have increased intracranial pressure. Relying on the February MRI results plus the “classic symptoms” of hydrocephalus, Dr. Parrish opined that this “equals a shunt ... every time.” He stated that although Lance’s being off-balance and confused *504and having slurred speech are “generic symptoms,” “in the fact of that M.R.I. scan showing severe aqueductal stenosis, they are the light bulb that needs to go off and say this requires a shunt.”
Dr. Parrish testified that Lance had “pre-existing” large ventricles. He considered it significant that Lance had contracted encephalitis when he was six years old. He testified that he believed the encephalitis “had something to do with scarring in the aqueduct which led to [Lance’s] increased intracranial pressure and enlarged ventricles.” Dr. Parrish opined that the encephalitis caused an inflammation in Lance’s brain, which led to scarring, or gliosis, which then led to the narrowing of the aqueduct. Dr. Parrish testified that a narrowed, or partially obstructed, aqueduct “means it’s more difficult for fluid to flow through” and thus requires a higher amount of intracranial pressure to force fluid through the aqueduct.
Dr. Parrish also testified that “[t]he contour of the ventricles and even the contour of the aqueduct is proof that there is at some time increased intracranial pressure, increased intraventricular pressure.” Dr. Parrish described Lance’s third ventricle, as seen in the February 2010 MRI, as “huge,” and he stated that “the top part of the aqueduct is enlarged compared to the bottom part, which is extremely small.” He testified that this was evidence of “increased intracranial pressure at some time.” Dr. Parrish testified that the “obvious indications of pressure” on the February 2010 MRI scan included the “[b]ig third ventricle,” “enlargement of the proximal part of the aqueduct of Sylvius and constriction of the bottom part [of the aqueduct],” and a slightly enlarged fourth ventricle. He stated, “Those ventricles got big somehow, and they were blown up by the increased pressure.”
Dr. Parrish reviewed the April MRI and testified that, although Lance’s ventricles looked the same size in the April MRI, he concluded that “the aqueduct here is more dilated proximally on the inside” than the aqueduct in the February MRI. Dr. Parrish suggested that “the pressure has increased, or it may be intermittently increasing,” and he testified that the April MRI reflected that Lance was “getting worse.” Dr. Parrish agreed that the April MRI indicated that “the angle of the aqueduct was different and it indicated pressure.” He also testified that Lance demonstrated “typical compensated hydrocephalus,” in which the ventricles expand to compensate for the obstructed flow of cerebrospinal fluid through the aqueduct, but, at some point, because the brain is constrained by the skull, the ventricles reach the limit of the amount they can expand, the increasing intracranial pressure has “to go somewhere” and so it is “exerted down through the brain stem,” which affects the heart and respiratory rates. Dr. Parrish stated that the April MRI demonstrated compensation and that “you can compensate up to a point, and at some point the time bomb goes off.”
On cross-examination, Dr. Parrish agreed that the autopsy showed a “normal looking brain” and revealed no microscopic evidence of increased intracranial pressure, such as herniation, swelling, or bleeding within the brain. He also agreed that he could not determine how long Lance had had enlarged ventricles and that the MRI could not pinpoint when the changes in Lance’s brain structure had occurred. Dr. Parrish also agreed that although Lance had several symptoms associated with increased intracranial pressure when he presented to NCMC, such as slurred speech, confusion, a headache, and balance problems, he did not have other “classic” symptoms such as nausea and vomiting, increased blood pressure, increased pulse *505pressure, papilledema, and a low heart rate.
Dr. Parrish agreed that Lance’s symptoms could have been caused by “some other process” rather than increased intra-cranial pressure and that Lance’s symptoms all disappeared while he was in the hospital. Dr. Parrish suggested that Lance “opened up his pathway somehow,” such as by having “enough [intracranial] pressure that he opened up the aqueduct” and “relieved his own pressure,” which could account for the rapid dissipation of Lance’s symptoms. Dr. Parrish further agreed that no other doctor called Dr. Kareh to inform him of the April 2010 MRI results and that Dr. Kareh, therefore, would not have had any knowledge of Lance’s worsening hydrocephalus and aqueductal stenosis as shown on the April MRI. Dr. Parrish also agreed that at the time Lance left NCMC in February 2010, his aqueduct was not completely closed. He further agreed that placing a, shunt in a patient can result in the patient’s death. Dr. Parrish agreed that Lance had an MRI performed nine days before he died and he “could have survived his problem ... if he’d had a shunt done the day before he died.”
Dr. Dragovic testified that, in his opinion based on a reasonable degree of medical probability, Lance “died of complications of obstructive hydrocephalus.” Factors relevant to Dr. Dragovic’s opinion included the fact that Lance had had “some problems and neurological deficits that were occurring on and off over a period of time,” the “established clinical diagnosis [of] enlarged ventricles,” and Lance’s history of having suffered from encephalitis.
Dr. Dragovic stated that after reviewing the microscopic slides prepared during the autopsy, he “now know[s] beyond any reasonable doubt in [his] mind that there was acute blockage, acute obstruction of the aqueduct at the lower level [leading to the fourth ventricle]” when Lance died, and he opined that a build-up of glial tissue, or scar tissue in the brain, caused the blockage. Dr. Dragovic also testified that Lance’s enlarged ventricles “reflect[ed] sudden increase of [intracranial] pressure as a result of increased blockage.” He stated that it was “clear that this condition had been present for a long time.” Dr. Dragovic thus concluded that, in his opinion, this case involved an acute blockage of the aqueduct and that the “sudden rise of intracranial pressure because of the blockage creating the pressure on the brain stem and pressure on the structures above the brain stem to lose control of respiratory function and allow the quick accumulation of fluid in the lungs.”5
Dr. Gill, Lance’s treating neurologist, who settled before trial, testified by video deposition. He testified that although Lance was suffering from obstructive hydrocephalus, he did not wish that he had insisted that Dr. Kareh place a shunt in Lance’s brain. Dr. Gill agreed that “the applicable standard of care is that the treatment for obstructive hydrocephalus is either a shunt or a third ventriculostomy.”6 He testified, however, that he believed discharging Lance without placement of a *506shunt was proper because the monitoring of Lance’s intracranial pressure revealed no sustained increased in pressure and because his headache had improved and he was feeling better. Dr. Gill agreed that intracranial pressure fluctuates and that increased pressure could be intermittent.
Windrum also called Dr. Randolph Evans, a neurologist who had been retained by Dr. Gill, to testify. Dr. Evans testified that Lance was “perhaps symptomatic” when he presented to NCMC in February 2010 and that he was “not entirely sure that these symptoms [that he had upon presentment] were due to aqueductal sten-osis,” although he later testified, based on a reasonable degree of medical probability, that Lance’s symptoms were caused by aqueductal stenosis. He stated that the symptoms with which Lance presented to NCMC “can be consistent with a number of different neurological problems, including increased intracranial pressure.”
Dr. Evans also agreed that the two major alternatives for treating aqueductal stenosis are shunt surgery and a third ventriculostomy, but he stated, “[T]he [medical] literature suggests that surgical treatment should be offered to patients where the symptoms are felt to be due to aqueductal stenosis.” He testified that placing a shunt “has a high risk of complications,” although he also agreed that shunt surgery is successful in a high percentage of cases and that the mortality rate for this treatment is “close to zero.” He testified that “for many patients, [shunt surgery] will be a good treatment, but there are risks and benefits of these surgical treatments, like any others,” and the neurosurgeon must determine whether “the risk of treatment outweigh[s] the risk of not having treatment.” Dr. Evans agreed that unless the patient has specific impairments such as advanced age or a heart condition, surgical intervention is appropriate. Dr. Evans also noted that the medical records reflected that Dr. Kareh offered to place a shunt in Lance’s brain, but Lance had “declined.”
Dr. Warren Neely, a neurosurgeon, testified on behalf of Dr. Kareh. Dr. Neely testified that, in his opinion, although Lance had aqueductal stenosis, it was not obstructive and Lance did not die from aqueductal stenosis.7 Dr. Neely opined that none of the radiological scans demonstrated evidence of increased intracranial pressure, that the ventricular monitoring demonstrated intracranial pressure within a normal range, and that the autopsy revealed “normal findings of the brain” and did not show any indication of elevated intracranial pressure at the time of death. Dr. Neely testified that the major symptoms consistent with obstructive hydrocephalus are extreme drowsiness, severe headaches, nausea, vomiting, eye movement problems, swelling of the optic nerve, and papilledema. He stated that the symptoms that Lance presented with were all “nonspecific symptoms” that could be indicative of several conditions and do not necessarily indicate increased intracranial pressure.
*507Dr. Neely testified that the standard of care did not require Dr. Kareh to install a shunt in Lance’s brain. He stated:
[T]his is an initial assessment. You’re seeing someone that has very nonspecific symptoms. You have a CAT scan and an MRI scan that do not show increased intracranial pressure. Yes, there are certainly abnormalities in his ventricular system. We see that all the time. This is a very common finding in patients that we see.
Again, in this situation, I would not install a shunt based on the history or the findings on the MRI scan or CAT scan.
Dr. Neely further testified that the medical records reflected that Dr. Kareh explained to Lance that he might have increased intracranial pressure, that the pressure needed to be monitored, that, if it was elevated, they would consider placing a shunt, and that they discussed the risks of the procedures involved. Based on his review of the ventricular monitoring procedure, Dr. Neely agreed with Dr. Kareh that Lance was not suffering from increased intracranial pressure at the time he saw Dr. Kareh, although there were several instances in which Lance’s intra-cranial pressure spiked to above-normal levels.
Dr. Neely testified that, based on the intracranial pressure readings, he “absolutely” would not have recommended the installation of a shunt and that the standard of care did not require a shunt based on those readings. He stated that he would not install a shunt in a patient who had normal levels of intracranial pressure because draining cerebrospinal fluid from a patient with normal pressure levels could cause chronic headaches, dizziness, fainting spells, and complications in which the surface of the brain moves away from the skull and the resulting space fills up with either fluid or blood, which could lead to a tear in a vein and a subdural hematoma. He also testified that installation of a shunt’itself can have complications, such as risks from anesthesia, the possibility of infection, failure of the shunt, and rupture of a blood vessel in the brain or chest or abdominal cavities. Dr. Neely testified that, based on the possibility of complications from installing a shunt and the fact that Lance did not have increased intra-cranial pressure, it was “very appropriate” for Lance to be discharged from NCMC without placement of a shunt.
Dr. Kent Heck testified as Dr. Kareh’s neuropathology expert. Dr. Heck agreed that Lance’s aqueduct was narrowed and that this finding was consistent with Lance’s history of hydrocephalus with aqueductal stenosis. He testified that if a patient died from hydrocephalus and aque-ductal stenosis, he would expect to find during the autopsy evidence of brain swelling and herniation, which he did not see in the pathology slides from Lance’s autopsy. Dr. Heck testified that he saw no evidence of increased intracranial pressure at the time of Lance’s death and that he saw no evidence of Lance’s dying from complications from hydrocephalus due to aqueduc-tal stenosis.
Dr. Heck testified that other pathology slides revealed that Lance had an enlarged heart and dilation of the chambers of the heart, indicative of congestive heart failure. He testified that if he had had the responsibility of filling out the death certificate in this case he would have listed “undetermined” as the cause of death. He stated that, in this case, “the two primary suspects” for Lance’s cause of death were the heart and the brain but that “neither [had] enough conclusive evidence to determine which [was] the true cause of death.” He agreed with Dr. Kareh’s counsel that “there is absolutely no evidence of any kind of a complication from hydrocephalus *508due to aqueductal stenosis as a cause of death in Mr. Windrum.”
The jury found both Dr. Kareh and Dr. Gill to be negligent, and it assigned eighty percent responsibility to Dr. Kareh and twenty percent responsibility to Dr. Gill. The jury awarded to Tracy Windrum, in her individual capacity, $211,280 for past pecuniary loss, $1,177,176.96 for future pecuniary loss, $30,000 for past loss of companionship and society, $200,000 for past mental anguish, and $250,000 for future mental anguish. The jury awarded B.W. $39,615 for past pecuniary loss, $220,720.68 for future pecuniary loss, $30,000 for past loss of companionship and society, $50,000 for future loss of companionship and society, $200,000 for past mental anguish, and $500,000 for future mental anguish. The jury awarded J.W. $39,615 for past pecuniary loss, $220,720.68 for future pecuniary loss, $30,000 for past loss of companionship and society, $50,000 for future loss of companionship and society, $100,000 for past mental anguish, and $275,000 for future mental anguish. The jury awarded H.W. $39,615 for past pecuniary loss, $220,720.68 for future pecuniary loss, $30,000 for past loss of companionship and society, $50,000 for future loss of companionship and society, $75,000 for past mental anguish, and $200,000 for future mental anguish.
In its final judgment, the trial court applied the statutory cap on damages in wrongful death cases and awarded a total of $1,875,887.62 to Tracy Windrum. The trial court apportioned the award as follows: $1,123,301.89 for Tracy Windrum in her individual capacity, $277,840.33 for the benefit of B.W., $241,869.10 for the benefit of J.W., and $232,876.30 for the benefit of H.W. The trial court denied Dr. Kareh’s motion for judgment notwithstanding the verdict and motion for new trial, and this appeal followed.
Sufficiency of the Evidence of Medical Negligence
In his first issue, Dr. Kareh contends that Windrum failed to present legally and factually sufficient evidence that his actions or omissions caused Lance’s death. In his second issue, Dr. Kareh contends that Windrum failed to present legally and factually sufficient evidence that he breached the standard of care, and thereby committed negligence, by failing to install a shunt in Lance’s brain.
A. Standard of Review
When conducting a legal sufficiency review, we credit favorable evidence if a reasonable fact-finder could do so and disregard contrary evidence unless a reasonable fact-finder could not. See City of Keller v. Wilson, 168 S.W.3d 802, 827 (Tex. 2005): Brown v. Brown, 236 S.W.3d 343, 348 (Tex. App.-Houston [1st Dist.] 2007, no pet.). We consider the evidence in the light most favorable to the finding under review and we indulge every reasonable inference that would support the finding. City of Keller, 168 S.W.3d at 822. We sustain a no-evidence contention only if: (1) the record reveals a complete absence of evidence of a vital fact: (2) the court is barred by rules of law or evidence from giving weight to the only evidence offered to prove a vital fact: (3) the evidence offered to prove a vital fact is no more than a mere scintilla: or (4) the evidence conclusively establishes the opposite of the vital fact. Id. at 810: Merrell Dow Pharms., Inc. v. Havner, 953 S.W.2d 706, 711 (Tex. 1997).
In a factual sufficiency review, we consider and weigh all of the evidence. See Cain v. Bain, 709 S.W.2d 175, 176 (Tex. 1986) (per curiam): Arias v. Brookstone, L.P., 265 S.W.3d 459, 468 (Tex. App.-Houston [1st Dist.] 2007, pet. denied). When the appellant challenges a *509jury finding on an issue on which it did not have the burden of proof at trial, we set aside the verdict only if the evidence supporting the jury finding is so weak as to make the verdict clearly wrong and manifestly unjust. See Cain, 709 S.W.2d at 176: Reliant Energy Servs., Inc. v. Cotton Valley Compression, L.L.C., 336 S.W.3d 764, 782 (Tex. App.-Houston [1st Dist.] 2011, no pet.). The jury is the sole judge of the witnesses’ credibility and it may choose to believe one witness over another. See Golden Eagle Archery, Inc. v. Jackson, 116 S.W.3d 767, 761 (Tex. 2003). We may not substitute our judgment for that of the jury. Id. “Because it is the jury’s province to resolve conflicting evidence, we must assume that jurors resolved all conflicts in accordance with their verdict.” Figueroa v. Davis, 318 S.W.3d 53, 60 (Tex. App.-Houston [1st Dist.] 2010, no pet.).
B. Evidence of Negligence
“To meet the legal sufficiency standard in medical malpractice cases ‘plaintiffs are required to adduce evidence of a “reasonable medical probability” or “reasonable probability” that their injuries were caused by the negligence of one or more defendants, meaning simply that it is “more likely than not” that the ultimate harm or condition resulted from such negligence.’” Jelinek v. Casas, 328 S.W.3d 526, 532-33 (Tex. 2010) (quoting Kramer v. Lewisville Mem’l Hosp., 858 S.W.2d 397, 399-400 (Tex. 1993)). The elements of a health care liability claim sounding in negligence are (1) a legal duty, (2) a breach of duty, and (3) damages proximately caused by the breach. Creech v. Columbia Med. Ctr. of Las Colinas Subsidiary, L.P., 411 S.W.3d 1, 5-6 (Tex. App.-Dallas 2013, no pet.). The standard of care for a health care provider is what an ordinarily prudent health care provider would do under the same or similar circumstances. Creech, 411 S.W.3d at 6. In a medical malpractice case, the plaintiff ordinarily must produce expert testimony to establish the applicable standard of care and causation if those matters are not within the experience of a layperson. Id. Thus, to establish negligence in this case, Windrum had to demonstrate, by a preponderance of the evidence, (1) that Dr. Kareh had a duty to place a shunt in Lance’s brain when he saw him on February 4, 2010, (2) that Dr. Kareh’s failure to place the shunt in Lance’s brain at that time fell below the standard of care of a reasonably prudent neurosurgeon, and (3) that, but for Dr. Kareh’s failure to place the shunt in Lance’s brain at that time, Lance would not have suffered sudden death on May 2,2010.
Texas Rule of Evidence 702 provides that “[i]f scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training, or education may testify thereto in the form of an opinion or otherwise.” Tex. R. Evid. 702, 61 Tex. B.J. 374, 392 (Tex. & Tex. Crim. App. 1998, amended 2015).8 “ ‘It is the basis of the witness’s opinion, and not the witness’s qualifications or his bare opinions alone, that can settle an issue as a matter of law: a claim will not stand or fall on the mere ipse dixit of a credentialed witness.’ ” Coastal Transp. Co. v. Crown Cent. Petroleum Corp., 136 S.W.3d 227, 232 (Tex. 2004) (quoting Burrow v. Arce, 997 S.W.2d 229, 235 (Tex. 1999)): Gammill v. Jack Williams Chevrolet, Inc., 972 S.W.2d 713, 726 (Tex. 1998) *510(“[T]here must be some basis for the. opinion offered to show its reliability. Experience alone may provide a sufficient basis for an expert’s testimony in some cases, but it cannot do so in every case,”).
Opinion testimony that is con-clusory or speculative is not relevant evidence because it does not tend to make the existence of a material fact “more probable or less probable.” Coastal Transp. Co., 136 S.W.3d at 232 (quoting TEX. R. EVID. 401): see also Havner, 953 S.W.2d at 712 (“When the expert ‘br[ings] to court little more than his credentials and a subjective opinion,’ this is not evidence that would support a judgment.”) (quoting Viterbo v. Dow Chem. Co., 826 F.2d 420, 421 (5th Cir. 1987)): Cooper Tire & Rubber Co. v. Mendez, 204 S.W.3d 797, 801 (Tex. 2006) (“If the expert brings'only his credentials and a subjective opinion, his testimony'is fundamentally unsupported and therefore of no assistance to the. jury.”). “It is incumbent on an expert to connect the data relied on and his or her opinion and to show how that data is valid support for the opinion reached.” Whirlpool Corp. v. Camacho, 298 S.W.3d 631, 642 (Tex. 2009).
The trial court, as the “gatekeeper” of expert testimony,-has the threshold responsibility of “ensuring that an expert’s testimony both rests on a reliable foundation and is relevant to the task at hand.” Gammill, 972 S.W.2d at 728 (quoting Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 597, 113 S.Ct. 2786, 2799, 125 L.Ed.2d 469 (1993)). Expert testimony is conclusory if there is no factual basis for it or if the basis offered does not, on-its face, support the opinion. CCC Grp., Inc. v. S. Cent. Cement, Ltd., 450 S.W.3d 191, 202 (Tex. App.-Houston [1st Dist.] 2014, no pet.) (citing City of San Antonio v. Pollock, 284 S.W.3d 809, 817 (Tex. 2009)). Where experts rely on experience or .training to reach their opinions, rather than on a particular methodology, a reviewing court considers whether there is too great an analytical gap between the data and the opinion proffered for the opinion to be reliable. Moreno v. Ingram, 454 S.W.3d 186, 193 (Tex. App.-Dallas 2014, no pet.) (citing Gammill, 972 S.W.2d at 726). In conducting a no-evidence review involving expert testimony, we “cannot consider only an expert’s bare opinion, but must also consider contrary evidence showing it has no scientific basis.” Mendez, 204 S.W.3d at 804 (quoting City of Keller, 168 S.W.3d at 813). “[I]f an expert’s opinion is based on certain assumptions about the facts, we cannot disregard evidence showing those assumptions were unfounded.” Id. (quoting City of Keller, 168 S.W.3d at 813). “It is not enough for an expert simply to opine that the defendant’s negligence caused the plaintiffs injury. The expert must also, to a reasonable degree of medical probability, explain how and why the negligence caused the injury.” Jelinek, 328 S.W.3d at 536.
We conclude that, here, Windrum failed to carry her burden of proving by a preponderance of the evidence the elements of medical negligence required to hold Dr. Kareh liable in this case.
1. Duty to Place a Shunt on February 4, 2010, and Breach of that Duty
To prove that Dr. Kareh’s care of Lance fell below the standard of care of an ordinarily prudent neurosurgeon seeing a patient with symptoms of hydrocephalus for the first time, Windrum had to establish by a preponderance of the evidence that Dr. Kareh had a duty to place a shunt in Lance’s brain immediately following that visit on February 4, 2010, or, at the latest, at the time Dr. Kareh last treated Lance on February 22,2010.
Windrum relied on the expert testimony of Dr. Parrish to establish the essential *511elements of the standard of care applicable to neurosurgeons, Dr. Kareh’s breach of the standard of care, and causation. Dr. Parrish testified that, in his opinion, Dr. Kareh “should have put a shunt in when he saw Mr. Windrum in the hospital” in February because Lance “had these classic symptoms of increased intracranial pressure with staggering, slurred speech, and altered mental status that were periodic.” He testified that the presence of a classic symptom like papilledema—which was absent in this case—is “very significant” but not a necessary for a finding of increased intracranial pressure and that its absence is “not so significant.” Dr. Parrish stated that the February MRI indicated “that there is at some time increased intracranial pressure, increased intraventricular pressure.” He opined that “[t]hose ventricles got big somehow, and they were blown up by the increased pressure.”
Dr. Parrish also testified that Lance’s symptoms, plus the February 2010 MRI, which revealed aqueductal stenosis, “equals a shunt” “[e]very time.” He testified that the standard of care required Dr. Kareh to offer a shunt to Lance. He acknowledged that the balance problems, slurred speech, and confusion were “generic symptoms,” but, combined with the February 2010 MRI, those symptoms “are the light bulb that needs to go off and say this requires a shunt.” Dr. Parrish also acknowledged that there are “real risks” to performing surgery to install a shunt, but he state that the risks were “very rare” and “fairly low.”
Dr. Parrish concluded that, to comply with the standard of care, a reasonable, prudent neurosurgeon would have:
[M]ade the right diagnosis, obstructive hydrocephalus. Symptomatic obstructive hydrocephalus. Number two, he would have recommended a shunt or some definitive procedure to treat the hydrocephalus. And, three, he would have properly informed the patient and the patient’s family what would happen if he got a shunt, the reasonable things that would happen if he got a shunt. But even more importantly or as important I guess I would say, the benefit of getting the shunt and the risk of not getting a shunt.
Dr. Parrish presented no medical literature to support his opinion that the standard of care required the placement of a shunt "every time” when Dr, Kareh saw Lance in early February 2010. And his testimony- that Dr. Kareh should have “recommended a shunt or some definitive procedure to treat the hydrocephalus” and that Lance and his family should have been informed of the risks and benefits of a shunt-is-some evidence that a patient presenting with- Lance’s symptoms does not “equal[ ] a shunt” “every time.”
Other testimony by Dr. Parrish also undermined his claim that it was professional negligence, or malpractice, for Dr. Kareh not to install a shunt in Lance’s brain on February 4, 2010. On cross-examination, Dr. Parrish agreed that there was no “microscopic evidence” of increased intracra-nial pressure at the time of Lance’s autopsy in May 2010. He also agreed that Lance had increased intracranial pressure “at some point” in his life and that it was possible that his ventricles had enlarged and then remained the same size ever since he had had encephalitis as a child. Dr. Parrish acknowledged that, while Lance had some “classic symptoms” of increased intracranial pressure when he was seen by Dr. Kareh, such as slurred speech, confusion, and balance problems, he did not have other classic symptoms, such as widened pulse pressure, low heart rate, papilledema, nausea, or vomiting. He further agreed that the symptoms with which Lance presented to NCMC were consis*512tent with other conditions and that Lance “got better really fast” while in the hospital. Dr. Parrish opined that Lance’s symptoms could have been relieved because he “had enough [intraventricular] pressure that he opened up the aqueduct, and he started draining [cerebrospinal fluid] again” without a shunt.
Dr. Kareh saw Lance one other time after his initial presentment to NCMC—on February 22, 2010—and Lance reported at that appointment that he had had one headache episode and one episode of slurred speech. Dr. Kareh did not see Lance after that. Rather, Lance returned to Dr. Gill, and another MRI was ordered. The April 2010 MRI revealed changes both in the size of Lance’s ventricles and in the angle of the aqueduct relative to the February 2010 MRI. It is undisputed that no one informed Dr. Kareh of the headaches that Lance experienced in April 2010 or of the April. MRI scan. Windrum’s experts concurred that Lance’s sudden death was due to a complete obstruction of the aqueduct. All of the experts agreed, however, that, when Dr. Kareh saw Lance in February, the aqueduct, although narrowed, was open and cerebrospinal fluid was passing through the aqueduct. Although the April MRI revealed a worsening problem, no evidence showed that Dr. Kareh was advised of the results of that MRI.
Dr. Parrish did not point to any medical literature, such as peer-reviewed studies or authoritative treatises or texts, which stated that the immediate placement of a shunt is required when monitoring of in-tracranial pressure reveals no sustained increase in pressure and the patient’s symptoms have subsided. And, although Dr. Kareh presented evidence that shunt placement is not appropriate when intra-cranial pressure levels are within normal range and pressure monitoring does not reflect a sustained increase in pressure, Windrum presented no evidence other than Dr. Parrish’s testimony that shunt placement is necessary “every time.” Thus, there was no evidence other than Dr. Parrish’s unsupported opinion testimony, which did not consider the April 2010 MRI changes, to establish that the standard of care always requires placement of a shunt under-the circumstances presented to Dr. Kareh on February 4, 2010. See Coastal Transp. Co., 136 S.W.3d at 232 (providing that opinion testimony that is conclusory or speculative is not relevant evidence and cannot support judgment): Burrow, 997 S.W.2d at 235. Although Dr. Parrish has experience placing shunts in patients, he identified nothing in that experience to support the conclusion that every failure to place a shunt in those circumstances constitutes a breach of the standard of care by omission.
Although Dr. Parrish testified that shunt-placement surgery has its risks, as is true of all surgeries, and that he considered the risks in this case to be “fairly low,” neither Dr. Parrish nor any of Wind-rum’s other witnesses addressed the risks that Dr. Neely testified to concerning placement of a shunt in a patient who at the time of placement does not have increased intracranial pressure. See Ponte v. Bustamante, 490 S.W.3d 70, 80 (Tex. App.-Dallas 2015, pet. granted) (‘When the evidence shows that a particular treatment helps some patients and not others, the expert must explain the facts justifying a conclusion that a particular patient probably would have been helped by the treatment.”). Such evidence is particularly critical when the alleged negligence is the failure to perform an operation as opposed to negligence in actually performing it.
Windrum argues that all 'of the testifying physicians agreed that the standard of care required either a shunt or a “third ventriculostomy,” and she points to the *513testimony of Dr. Gill, the treating neurologist in this case, and Dr. Evans, a neurologist who had been retained by Dr. Gill. Dr. Gill agreed with Windrum’s counsel that “the applicable standard of care is that the treatment for obstructive hydrocephalus is either a shunt or a third ven-triculostomy.” He also testified, however, that he agreed with Dr. Kareh’s suggestion that a shunt was not necessary in this case: that, if he had not agreed, he would have “done something,” such as refer Lance to another neurosurgeon: and that he agreed with the decision to discharge Lance without surgical intervention because Lance did not demonstrate a sustained increase in intracranial pressure, his headaches had improved, and he “was feeling better.” Dr. Evans agreed that for most patients, unless they have a “specific physical impairment like age or a heart condition,” “surgical intervention is going to be the appropriate thing to do,” although he acknowledged there are risks associated with shunt surgery.
Neither Dr. Gill nor Dr. Evans testified concerning the specific risks of placing a shunt when the patient does not have increased intracranial pressure. Dr. Kareh also presented evidence that shunt placement was not appropriate in this case due to the monitoring results, which indicated that Lance was not suffering from increased intracranial pressure at the time Dr. Kareh was consulting on his case. See Mendez, 204 S.W.3d at 804 (stating that, in conducting no-evidence review involving expert testimony, courts “cannot consider only an expert’s bare opinion, but must also consider contrary evidence showing it has no scientific basis”). Dr. Parrish’s opinion fails to account for these risks or for Lance’s worsening symptoms and test results two months later, which the jury heard evidence about but which Dr. Kareh did not have in February when he treated Lance.
In sum, Windrum presented no evidence concerning the standard of care and Dr. Kareh’s breach of the standard of care beyond Dr. Parrish’s conelusory testimony. See Coastal Transp. Co., 136 S.W.3d at 232 (“It is the basis of the witness’s opinion, and not the witness’s qualifications or his bare opinions alone, that can settle an issue as a matter of law: a claim will not stand or fall on the mere ipse dixit of a credentialed witness.”): see also Mendez, 204 S.W.3d at 801 (“If the expert brings only his credentials and a subjective opinion, his testimony is fundamentally unsupported and therefore of no assistance to the jury.”). We therefore conclude that Windrum failed to present legally or factually sufficient evidence of an essential element of her cause of action. See Creech, 411 S.W.3d at 5-6 (stating that essential element of medical malpractice cause of action is breach of legal duty and that standard of care in medical malpractice suit is what ordinarily prudent health care provider would do under same or similar circumstances).
2. Proximate Cause of Lance’s Death
We further conclude that, even if Dr. Kareh’s actions did fall below the standard of care, Windrum failed to establish that Dr. Kareh’s actions proximately caused Lance’s death. Thus, Windrum failed to prove the essential causation element of negligence.
“Proximate cause” includes both cause in fact, meaning that “the act or omission was a substantial factor in bringing about the injuries, and without it, the harm would not have occurred,” and foreseeability. IHS Cedars Treatment Ctr. of DeSoto, Tex., Inc. v. Mason, 143 S.W.3d 794, 798-99 (Tex. 2004): Tejada v. Gernale, 363 S.W.3d 699, 709 (Tex. App.Houston [1st Dist.] 2011, no pet.) (noting *514that evidence showing only that defendant’s negligence furnished condition that made injuries possible is insufficient to show proximate cause and that proximate cause cannot be established by “mere conjecture, guess, or speculation”). Cause in fact is not established where a defendant’s actions do no more than furnish a condition which makes the injuries possible. Givens v. M&S Imaging Partners, L.P., 200 S.W.3d 735, 738 (Tex. App.-Texarkana 2006, no pet.). In such a case, the defendant’s conduct is too attenuated from the resulting injuries to be a substantial factor in bringing about the harm. Id.: see also Providence Health Ctr. v. Dowell, 262 S.W.3d 324, 328-29 (Tex. 2008) (holding that discharge of patient from emergency room, when patient had presented to emergency room with self-inflicted cut on wrist and then committed suicide thirty-three hours after discharge, “was simply too remote from his death in terms of time and circumstances” and, thus, plaintiffs presented insufficient evidence of proximate cause). “Foreseeability means the actor, as a person of ordinary intelligence, should have anticipated the dangers his negligent act created for others,” but it does not “require a person to anticipate the precise manner in which injury will occur once the person creates a dangerous situation through his negligence.” Taylor v. Carley, 158 S.W.3d 1, 9 (Tex. App.— Houston [14th Dist.] 2004, pet. denied).
Dr. Parrish agreed with defense counsel that the April MRI revealed that Lance’s symptoms were progressing and that Lance could have survived “if he’d had a shunt done the day before he died,” indicating that any failure by Dr. Kareh to place a shunt when he saw Lance in February 2010 was not an immediate cause of death. All of the doctors who testified in this case, including Dr. Kareh’s experts, agreed that placement of a shunt can be an appropriate treatment for a patient presenting with obstructive hydrocephalus caused by aqueductal stenosis when there is a build-up of cerebrospinal fluid in the brain. There was no such evidence of cere-brospinal fluid buildup in February 2010. Instead, Lance’s intracranial pressure was normal, with occasional spikes in the pressure above a normal range and no sustained increase in pressure. All of the neurological symptoms with which Lance had presented to NCMC were resolved by the time the period of intracranial pressure monitoring ended. When Lance saw Dr. Kareh for a follow-up appointment almost three weeks later, he had had only one additional headache episode and one additional episode of slurred speech. Lance did not see Dr. Kareh again, and there is no evidence Dr. Kareh was ever informed of the changes to Lance’s aqueduct visible on the April MRI or of the additional headache episodes that he experienced in April. We conclude that, as a matter of law, Dr. Kareh’s decision not to recommend placement of a shunt on February 4, 2010, was too remote from Lance’s death on May 2, 2010, to be a proximate cause of Lance’s death. See Dowell, 262 S.W.3d at 328-29: Givens, 200 S.W.3d at 742.
We hold that because essential elements of Windrum’s medical malpractice cause of action are not supported by legally sufficient evidence, the trial court erred in entering judgment in favor of Windrum on that claim.
We sustain Dr. Kareh’s first and second issues.9
Conclusion
We reverse the judgment of the tidal court and render judgment that the trial *515court enter a take-nothing judgment against Windrum on Windrum’s medical malpractice claim.
Chief Justice Radack and Justices Jennings, Keyes, Higley, Bland, Massengale, Brown, Huddle, and Lloyd participated in the vote to determine en banc consideration.A majority of the Court voted to deny en banc consideration. See Tex. R. App. P. 49.7
Justice Jennings, joined by Justices Hig-ley and Lloyd, dissenting from the denial of en banc consideration.
Justice Brown, dissenting from the denial of en banc consideration.
. "Transependymal flow” is the flow of cere-brospinal fluid outside of the ventricular system.
. Windrum originally sued Dr. Gill, as well as North Cypress Medical Center, North Cypress Medical Center Operating Company, GP, LLC, North Cypress Medical Center Operating Company, Ltd., and Coresource, Inc. Windrum settled with Dr. Gill and the North Cypress entities pre-trial, and she nonsuited her claims against Coresource,
. Placement of a shunt involves threading a tube from the brain down into the patient’s abdomen. When there is a blockage in the ventricular system, excess cerebrospinal fluid flows through the shunt down into the abdomen, where it is then absorbed into the body. This mechanism helps relieve the elevated intracranial pressure that can occur with the build-up of cerebrospinal fluid in the ventricles.
. Dr. Parrish testified that "obstructive hydrocephalus" does not necessarily mean a complete blockage of the aqueduct and that a "partial” obstruction, such as the narrowed aqueduct seen in cases of aqueductal stenosis, is considered "obstructive hydrocephalus.”
. Dr. Dragovic testified that the photographs taken by the medical examiner’s office of Lance as he was found in bed on May 2, 2010, support this conclusion, as they show “purging from his nostrils, purging from his mouth,” indicative of a build-up of fluid in his lungs. He testified that this evidence is inconsistent with death from cardiac arrhythmia. Dr. Dragovic stated that he was able to exclude a heart problem as a possible cause of Lance’s death.
. A ventriculostomy involves puncturing the bottom of the third ventricle to create another method by which cerebrospinal fluid can flow out of the third ventricle.
. Dr. Neely defined "obstructive hydrocephalus” as "a blockage somewhere in the flow of spinal fluid from where it’s being made to actually where it’s being reabsorbed,” and he testified that obstructive hydrocephalus and aqueductal stenosis are not necessarily the same thing, although “compensated” or "partial obstructive hydrocephalus” "could mean the same thing as compensated aqueductal stenosis.” Dr. Kareh similarly defined obstructive hydrocephalus as "a blockage of the normal pathway [of cerebrospinal fluid.]” He also acknowledged that obstructive hydrocephalus can be total or partial. Dr. Kareh defined aqueductal stenosis as "[a] dysfunction through the aqueduct” that affects the proper circulation of cerebrospinal fluid.
. Effective April 1, 2015, the Texas Supreme Court adopted amendments to the Texas Rules of Evidence. See 78 Tex. B.J. 42, 42 (Tex. 2015). The revisions to Rule 702 was stylistic and does not affect the substance of the rules. All further citations to the Rules of Evidence refer to the rules as they existed at the time of the parties' trial.
. Because we hold that no evidence supports essential elements of Windrum’s cause of action, we need not address Dr. Kareh’s remaining issues on appeal.