OPINION
PHIL JOHNSON, Chief Justice.
Appellant Emily Archer, M.D., appeals from a judgment against her in a medical negligence case. Concluding that the evidence of proximate cause is legally insufficient, we reverse and render.
BACKGROUND
Dr. Emily Archer, a gynecologist, began treating Anita Warren in 1988. The initial treatment was for pelvic pain. Dr. Archer performed hysterectomy surgery which was successful and relieved Anita’s pain.
Anita continued consulting Dr. Archer, as well as other doctors, through the years. During her annual examination in 1990, Anita related symptoms of mild incontinence, which Dr. Archer diagnosed as stress urinary incontinence (“SUI”). Anita testified at trial that her symptoms in
1990 consisted of a feeling of unusual dampness during sexual intercourse. The 1990 pelvic exam by Dr. Archer revealed a cystocele. The cystocele was formed by herniation of part of Anita’s bladder through the wall of her vagina and the resultant protrusion of that part of her bladder into the vagina.
Anita returned to Dr. Archer in 1993. Her complaints included incontinence, which she described at trial as being about the same as in 1990. Examination by Dr. Archer revealed that Anita’s cystocele had progressed to a more advanced state, and that she had developed a rectocele. The rectocele was formed by herniation of part of the wall of Anita’s rectum through the wall of her vagina and the resulting protrusion of part of her rectal wall into the vagina.
On January 17, 1995, Anita returned to Dr. Archer for an annual examination. She had continued complaints of incontinence which, according to her trial testimony, by then had begun occurring with certain physically stressful activities such as coughing, sneezing, and lifting heavy objects, as well as with sexual intercourse. Dr. Archer found that Anita’s cystocele had progressed to involve the urethra (the canal for discharging urine from her bladder) and had compromised the urethra. Dr. Archer classified the cystocele as a second degree cystourethrocele.
Anita still had the rectocele, which Archer then classified as a first degree rectocele.
Dr. Archer recommended surgery which she believed would correct the cystoureth-rocele, the rectocele and the SUI. She did not discuss Kegel’s exercises with Anita. Kegel’s exercises are exercises designed to strengthen muscles in a woman’s pelvic floor and muscles supporting the urethra.
Anita testified that she understood the surgery was needed to repair her bladder because the bladder had dropped following her hysterectomy. Anita agreed to the recommendation for surgery and on January 27, 1995, Dr. Archer performed surgery. The surgery stopped Anita’s incontinence and corrected the anatomical defects.
Anita developed pain in her right leg postoperatively. Dr. Archer performed a second surgery to release two sutures which were suspected of impinging on Anita’s obturator nerve and causing the pain. Anita’s pain persisted after the second surgery, despite referrals to and treatments by specialists in physical medicine, pain management, and neurosurgery. At trial, Anita claimed continuing pain and impairment from her right leg pain, which was diagnosed as pain from nerve damage as a result of the surgery.
Anita and her husband filed suit alleging that Dr. Archer was negligent in various ways which proximately caused Anita’s continuing pain and impairment. The case eventually was tried on the theory that (1) Dr. Archer negligently failed to offer the non-surgical option of Kegel’s exercises
to Anita before doing surgery; (2) Anita
would have chosen and performed the nonsurgical option had it been offered; (3) the Kegel’s exercises probably would have corrected her incontinence without surgery; and (4) Anita’s nerve damage would have been avoided if the surgery had not been done. The jury found, in response to a broad form liability question, that Dr. Archer’s negligence was a proximate cause of Anita’s injury in question. Judgment was entered in favor of the Warrens for the amount of damages found by the jury, together with pre- and post-judgment interest.
Via ten issues, Dr. Archer challenges the (1) legal and factual sufficiency of evidence to support the findings of negligence and proximate cause; (2) factual sufficiency of evidence to support the damages findings for lost wages, lost earning capacity and future medical care; (3) failure of the trial court to give limiting instructions to the jury concerning evidence admitted for a limited purpose; (4) trial court’s written notations on an exhibit as a comment on the weight of the evidence; (5) broad form submission of the negligence issue; and (6) trial court’s refusal to hold a hearing on her motion for new trial which alleged jury misconduct. We determine that her third issue, which urges legal insufficiency of the evidence to support a finding that the alleged negligence proximately caused Anita’s injuries, is dispositive. We will only address that issue.
See
Tex.R.App. P. 47.1.
MEDICAL NEGLIGENCE
Plaintiffs in medical negligence eases are required to prove by a preponderance of the evidence that the allegedly negligent act or omission was a proximate cause of the harm alleged.
See Kramer v. Lewisville Mem’l Hosp.,
858 S.W.2d 397, 400 (Tex.1993). To establish proximate cause, the plaintiff must prove (1) foreseeability, and (2) cause-in-fact.
See Leitch v. Hornsby,
935 S.W.2d 114, 118-19 (Tex.1996). The cause-in-fact element of proximate cause requires proof that the alleged negligence was a substantial factor in bringing about the harm, and without which the harm would not have occurred.
See Park Place Hosp. v. Estate of Milo,
909 S.W.2d 508, 511 (Tex.1995). With regard to cause-in-fact, the plaintiff must establish a causal connection between the defendant’s negligence and the injuries based upon a reasonable medical probability.
Id.
at 511. Opinion evidence relied on as proof must be based on more than possibilities, speculation and surmise.
See Schaefer v. Tex. Employers’ Ins. Ass’n,
612 S.W.2d 199, 202-05 (Tex.1980). In evaluating opinion evidence we look to the basis of the expert’s opinion, and not the bare opinion alone. A claim cannot stand or fall on the mere
ipse dixit
of a credentialed witness.
See Burrow v. Arce,
997 S.W.2d 229, 235 (Tex.1999). Whether expert testimony on causal connection rests upon reasonable medical probability must be determined by the substance and context of the testimony rather than semantics or use of a particular term or phrase.
See Burroughs Wellcome Co. v. Crye,
907 S.W.2d 497, 500 (Tex.1995). If an expert’s opinion is based on assumed facts that vary materially from the actual, undisputed facts, the opinion is without probative value and cannot support a verdict or judgment.
Id.
at 499-500.
LEGAL SUFFICIENCY OF THE EVIDENCE
In reviewing legal sufficiency or “no evidence” complaints, we may consider only
the evidence and inferences that tend to support the finding and must disregard all contrary evidence and inferences.
See Continental Coffee Prods. Co. v. Cazarez,
937 S.W.2d 444, 450 (Tex.1996). A no evidence complaint is to be sustained when the record shows one of the following: (a) a complete absence of evidence to prove a vital fact; (b) the reviewing court is barred by rules of law or evidence from giving weight to the only evidence offered to prove a vital fact; (c) the evidence offered to prove a vital fact is no more than a mere scintilla; or (d) the evidence establishes conclusively the opposite of the vital fact.
See Maritime Overseas Corp. v. Ellis,
971 S.W.2d 402, 409 (Tex.1998). If more than a scintilla of evidence exists, the evidence is legally sufficient.
See Lee Lewis Constr., Inc. v. Harrison,
70 S.W.3d 778, 782 (Tex.2001). More than a scintilla of evidence exists if the evidence furnishes some reasonable basis for differing conclusions by reasonable minds about a vital fact’s existence.
Id.
at 782-83.
ISSUE THREE: PROXIMATE CAUSE
By issue three, Dr. Archer challenges the legal sufficiency of the evidence of proximate cause in two areas. Both areas challenge the evidentiary support for a cause-in-fact finding.
First, she asserts there is no evidence that Anita would have done Kegel’s exercises if they had been offered to her. Second, she alleges there is no evidence that even if Anita had done the exercises, Anita, to a reasonable medical probability, would have improved to the extent that surgery would not have been necessary. We will address the second part of the issue. In doing so, we will assume, without deciding, that legally sufficient evidence exists to support a finding that Anita would have tried and performed Kegel’s exercises if they had been discussed by Dr. Archer.
Would Anita’s Use of Kegel’s Have Avoided Surgery?
Dr. Archer approaches this aspect of her legal sufficiency evidentiary challenge in two ways. We will address them in the order presented.
First, Dr. Archer references her own testimony that she did not offer Ke-gel’s to Anita in January, 1995, because the exercises would not have corrected the anatomical defects of the cystourethrocele and rectocele. She maintains that those defects and the SUI were corrected by the surgery which she planned, explained to Anita, Anita consented to and Dr. Archer performed; and there is no evidence that the exercises would have corrected those anatomical defects.
The Warrens respond that Anita agreed to surgery in order to correct her SUI and not because of the physical defects. They argue that Dr. Archer’s position is a misstatement of their theory of liability.
Dr. Archer testified that she would not recommend surgery for a second degree cystourethrocele and a first degree recto-cele if the patient was not having problems and was living with the defects, provided that the patient’s SUI symptoms could be corrected without surgery. She had diagnosed Anita with a second degree cystour-ethrocele and a first degree rectocele. The Warrens do not contend that they alleged or offered proof that Kegel’s would have corrected the cystourethrocele or the rectocele. They contend that their theory, as alleged and proved, was that Kegel’s probably would have cured or improved Anita’s SUI to the extent that she then would have been within the group of patients on which Dr. Archer testified she would not have performed surgery.
We agree with the Warrens. We do not perceive their allegations against Dr. Archer to have been that Kegel’s would have corrected the cystourethrocele and the rectocele, as Dr. Archer has framed the allegations.
This part of Dr. Archer’s issue three is overruled.
Second, Dr. Archer urges that Dr. Rosenfeld’s testimony as to causation consisted of general statistics which addressed percentages of patients whose SUI was “improved or cured” by the use of Kegel’s, but which did not differentiate between (1) percentages of patients whose SUI was cured or improved to the extent and duration that surgery was not necessary, and (2) patients whose SUI was improved, but not to the extent or for the duration that surgery was not necessary. Dr. Archer also maintains that the Warrens offered no evidence that Anita, as an individual with her particular conditions, to a reasonable medical probability, would have been one of the group of patients whose SUI was cured or improved to the extent or for the duration that surgery was not necessary.
In responding to this part of the issue, the Warrens refer us to (1) Dr. Rosenfeld’s testimony that “There is absolutely no reason that this particular patient could not be helped to the tune of maybe 60 or 80 percent improvement or of pure continence by the use of some well-explained nonsurgical treatment such as Kegel exercises”; (2) Dr. Rosenfeld’s reference to Department of Health Guidelines which “strongly recommended” pelvic muscle exercises (such as Kegel’s) for SUI and reported statistical studies showing “54 to 60 percent reduction in incontinent episodes”; (3) Dr. Archer’s testimony that Kegel’s exercises strengthen the pelvic floor muscles, help keep the bladder in place and lessen symptoms in patients with mild SUI; (4) Dr. Archer’s recognition of authorities citing high percentages of improvement or cure of SUI following an intensive program of physical therapy over a three-month period of time; and (5) Dr. Rosen-feld’s testimony that studies concerning the effects of Kegel’s exercises on women with SUI were studies of women who had the same or a similar condition as Anita had when she presented to Dr. Archer in 1995.
We will examine these areas for evidence that, to a reasonable medical probability, the prescribing of Kegel’s by Dr. Archer and use of Kegel’s by Anita would have improved her incontinence to the extent that surgery would have been avoided.
First, as to Dr. Rosenfeld’s testimony quoted above referencing “this particular patient,” the testimony did not reference Anita or her individualized situation. The line of questioning was objected to initially
by Dr. Archer’s counsel when Anita was referred to in a question. Following the objection and the trial court’s response, Anita’s counsel rephrased his question and referenced patients in general, and not Anita in particular.
Moreover, the question and answer dealt with cure or improvement of SUI symptoms, but did not differentiate between the two possible outcomes. Nor did the question and answer differentiate between patients who improved to some extent or for a temporary time with Kegel’s, but for whom surgery was still necessary to reach an acceptable reduction in SUI, and those who were cured or improved with Kegel’s to the extent and for the time duration that they did not require surgery to reach such a level. The cited portions of the record do not, therefore, indicate to a reasonable medical probability that even if Kegel’s had been offered to, accepted by and “well-done” by Anita, and Kegel’s had improved her incontinence, that she would have been cured or reached a level and duration of improvement such that surgery would have been avoided, as opposed to merely having some unspecified level or time period of improvement but still having fallen within the group of patients for whom surgery would have been necessary to reach their acceptable level of incontinence.
The second area of evidence referenced by the Warrens is Dr. Rosenfeld’s response when he was asked his opinion of whether, in January, 1995, Kegel’s exercises were a viable option that should have been discussed and tried with Anita. He opined that he thought the exercises should be tried at any time,
... but there are studies in this particular book that indicate that regardless of whether the patient had severe or moderate SUI, those people got good results with well-done Kegel’s exercises. They improved to the point that they did not need surgery ... and the answer to your question, that if — that the patient is never at a point — never at a point where you should not try a conservative method of holding this urethra up and trying to avoid surgery. Because if you have a 60 or 70 percent chance of avoiding surgery, why not?
Dr. Rosenfeld then referenced a study in a book he was referring to which indicated two treatment groups with a 54 to 60 percent reduction in incontinent episodes. He did not testify that the referenced study suggested a percentage of patients who were either cured of SUI without surgery, or were improved to the extent that surgery was not necessary. His statement that “They improved to the point that they did not need surgery” could be interpreted to imply a study wherein 100 percent of patients improved to the point that they did not need surgery. If Dr. Rosenfeld intended such an implication, the implication would be in the nature of an
ipse dixit
statement because no other evidence supported the statement. To the contrary, the statement stands in direct conflict with the remainder of his testimony when considered as a whole, and studies he cited and relied on as authoritative.
See Arce,
997 S.W.2d at 235. For example, Dr. Rosenfeld’s testimony elsewhere in the record is that Ke-gel’s exercises are performed for the purpose of (1) strengthening muscles in a woman’s pelvic floor to prevent her bladder from dropping, or dropping further, out of its correct anatomical position, and (2) strengthening muscles supporting the urethra so its ability to fully close and remain closed is enhanced. He referenced studies which reported that the strengthening of pelvic muscles via Kegel’s exercises, along with educational aspects of performing the exercises, resulted in certain
percentages of females with SUI having decreased incontinence or becoming completely continent. The studies reported, however, and Rosenfeld opined, that some females with SUI did not improve with use of the exercises. He specifically testified at one point that certain patients had no improvement from the exercises, even if properly done, because their muscles could not be tightened beyond a certain point.
His statement, “Because
if
you have a 60 or 70 percent chance of
avoiding surgery”
(emphasis added), assumed that Anita had a 60 or 70 percent chance of avoiding surgery via “weE-done” Kegel’s exercises. The assumption does not comport with (1) the studies he elsewhere referenced, (2) the balance of the opinions he offered which addressed only cure or improvement by the use of Kegel’s without a differentiation between patients who improved and avoided surgery and patients who improved but did not avoid surgery, or (3) his testimony that the most important factor in a patient’s improvement or cure through the prescription and performance of Kegel’s was the confidence expressed in the exercises by the treating physician. As to the latter, Dr. Rosenfeld was firm in his opinion that if the prescribing physician did not express confidence in the exercises as a mode of treatment, and communicate such confidence to the patient, then the patient’s use of Kegel’s was not going to have a successful result.
Dr. Rosenfeld was present in the courtroom during Dr. Archer’s testimony. Dr. Archer testified that her experience with Kegel’s was not positive; she did not have confidence that, given Anita’s condition, the exercises would improve Anita to the point that surgery could be avoided; she beheved Kegel’s would not work for Anita; and her lack of confidence that the exercises would work for Anita was the reason she did not discuss them in January, 1995.
Dr. Rosenfeld agreed that the reported percentages of females whose SUI was improved or stopped completely by use of Kegel’s varied by study, and that in the medical literature on the subject, “everybody cites a different percentage based on their own theory.” He acknowledged that some studies (1) reported improvement rates of less than 50 percent, and (2) reported that only a smaU number of patients studied were able to avoid surgery by using Kegel’s. He testified that if Ke-gel’s exercises were effective “it may be” that they would be a temporary solution, or they may be a permanent solution, but each patient has to be dealt with individually.
On cross-examination Dr. Rosenfeld was asked specifically if he could give a percentage as to the number of patients improved by Kegel’s to the degree that they could avoid surgery. He did not give such a percentage. Rather, he reiterated an earlier statement: “I quoted figures that 60 to 80 percent of the patients are either markedly improved or are continent, meaning that they are holding their urine. Now, that — you could imply from that, that, say, 20 percent of them to that point may not be improved but a little bit or not at aU.”
See Bradley v. Rogers,
879 S.W.2d 947, 956 (Tex.App.-Houston [14th Dist.] 1994, writ denied).
Dr. Rosenfeld testified consistently that each patient had to be dealt with individually. When he was asked directly what characteristics or description fit the kind of patient that Kegel’s would not work on, he responded, “Well, I’m not sure. The reason that Kegel’s exercises do not work may be on a basis of a variety of things_” Dr. Rosenfeld testified that surgery was a viable option if a patient’s SUI was serious enough that in the patient’s mind the SUI was socially unacceptable. He did not opine as to what level of
improvement was necessary to reach such a state of soeially-acceptable SUI, he did not reference Anita’s particular circumstances or level or duration of improvement which probably would have occurred, nor did he articulate standards by which to predict, by either individualized factors or a general percentage, which patients who used and had improvement from Kegel’s, to a reasonable medical probability, would remain with socially-unacceptable SUI requiring surgery for cure.
Dr. Rosenfeld’s response referenced by the Warrens was to the question of whether Kegel’s should have been discussed and tried with Anita. It does not purport to be an opinion that, to a reasonable medical probability, Anita had a 60 to 70 percent chance (or probability) of avoiding surgery. It assumes a state of evidence at variance with undisputed facts and Dr. Rosenfeld’s own direct opinions. Such testimony has no probative weight as evidence that, to a reasonable medical probability, surgery would have been avoided had Kegel’s been discussed with and performed by Anita.
See Burroughs Wellcome Co.,
907 S.W.2d at 499-500.
Dr. Rosenfeld’s testimony cited by the Warrens amounts to no more than a scintilla of probative evidence that had Anita used Kegel’s, she would, to a reasonable medical probability, have avoided surgery. Such testimony is legally insufficient evidence of proximate cause.
See Ellis,
971 S.W.2d at 409.
The third and fourth areas referenced by the Warrens relate to Dr. Archer’s testimony on cross-examination in which she acknowledged that Kegel’s are designed to strengthen the pelvic musculature, improve SUI symptoms, and that some studies show a high rate of improvement or cure of SUI via the patient’s performance of Kegel’s. Dr. Archer, however, did not testify that, to a reasonable medical probability, Anita would have avoided the need for surgery if Kegel’s had been prescribed for her and if she had performed them properly. To the contrary, Dr. Archer consistently maintained that she doubted the benefit of Kegel’s in SUI cases beyond the mild stage and Anita’s particular situation was one in which Dr. Archer’s opinion was that Kegel’s would not have prevented the need for surgery. Dr. Archer’s referenced testimony is not evidence that if she had offered Kegel’s to Anita, and Anita had performed them, that Anita would have avoided surgery to' a reasonable medical probability.
See Glenn v. Prestegord,
456 S.W.2d 901, 902 (Tex.1970).
Finally, the Warrens reference Dr. Ro-senfeld’s testimony that, except for some percentage of women who had congenital defects, all women who had SUI had urethral deviations which caused their SUI and that studies describing the effects of Kegel’s exercises on women with SUI were studies of women who had a compromised urethral condition which was either the same as or similar to Anita’s condition when she presented to Dr. Archer in 1995. As we have discussed above, such testimony is not evidence that Anita, to a reasonable medical probability, was one of the SUI patients who would be improved by “well-done” Kegel’s to the extent that surgery would not be required. If all women who had SUI had a compromised urethra such as Anita, then according to the evidence, performance of the exercises by the similarly-situated women (1) would not improve a percentage of the women, (2) would improve some percentage, but the improvement would not be enough to avoid surgery, (3) would improve some percentage enough to avoid surgery, and (4) would cure some percentage. But, predicting to a reasonable medical probability which particular patients would fall into which
category could logically only be based on the individual circumstances and conditions of each particular patient. There is an analytical gap between the referenced testimony and the conclusion that Anita would have avoided surgery to a reasonable medical probability.
See Gammill v. Jack Williams Chevrolet, Inc.,
972 S.W.2d 713, 728 (Tex.1998);
Bradley,
879 S.W.2d at 956-57. The referenced testimony is not legally sufficient evidence to support a finding that to a reasonable medical probability Anita would have avoided surgery by the use of Kegel’s.
In sum, Dr. Rosenfeld and the other physicians agreed that each patient must be considered individually. Dr. Rosenfeld was unable to set out any criteria which could be used to identify patients who would not be improved or cured by Ke-gel’s, such as patients with muscles which simply could not be strengthened by the exercises. The evidence showed that statistically, well-done Kegel’s would cure some patients of SUI so that surgery would not be necessary; would not improve some patients at all and surgery would still be necessary; and would improve SUI symptoms in some patients to some degree and for some time duration in which case surgery might or might not be necessary, depending on the degree of improvement, and whether the improvement was temporary or permanent. Such general statistical studies are not legally sufficient evidence that, to a reasonable medical probability, Anita’s SUI would have been cured or improved by Kegel’s to the extent that surgery would not have been necessary.
See Gammill,
972 S.W.2d at 728;
Havner,
953 S.W.2d at 720, 724.
Because the evidence is legally insufficient to support a finding that, to a reasonable medical probability, Anita would have avoided surgery by the use of Kegel’s, issue 3 is sustained.
See Ellis,
971 S.W.2d at 409. Our disposition of issue 3 is dis-positive of the appeal and we do not consider any other issues.
See
Tex.R.App. P. 47.1.
The judgment of the trial court is reversed. Judgment is rendered that the Warrens take nothing.