OPINION
JANICE M. HOLDER, J„
delivered the opinion of the court,
in which E. RILEY ANDERSON, ADOLPHO A. BIRCH, JR., and WILLIAM M. BARKER, JJ., joined.
We granted this appeal, in part, to determine whether fault was properly assessed against the patient in this medical malpractice action. We overrule Gray v. Ford Motor Co., 914 S.W.2d 464 (Tenn.1996), and hold that fault may not be assessed against a patient in a medical malpractice action in which a patient’s negligent conduct provides only the occasion for the medical attention, care, or treatment which is the basis for the action. We also hold that the additional issues raised by the defendant are without merit. We therefore affirm the trial court’s post-trial ruling that the defendant is 100% at fault and is responsible for the full amount of damages found by the jury.
Factual and Procedural Background
In the early morning hours of May 30, 1998, Larry T. Qualls was seriously injured in a single-vehicle accident in which he sustained a mild-to-moderate concussion and multiple facial fractures. Upon arriving at the trauma center at Vanderbilt University Medical Center (‘Vanderbilt”) nearly four hours after the accident, Qualls had a blood alcohol level of .13%, which extrapolated to approximately .20% at the time of the accident. Qualls was connected to a ventilator to assist his breathing and remained on a ventilator while at Vanderbilt. Dr. Timothy Van Natta, Qualls’s treating trauma surgeon, admitted Qualls to the neuro-intensive care unit for observation.
Over the next two days, Qualls remained in stable condition but suffered from “severe agitation.” Dr. Van Natta attributed this agitation to alcohol withdrawal. Qualls was given unusually large doses of Valium, Fentanyl, and other drugs to keep him sedated.
On June 2, 1998, Qualls’s fourth day at Vanderbilt, he continued to show signs of alcohol withdrawal and received large doses of medication to control his agitation. CT scans were ordered that day in preparation for reconstructive surgery to treat his facial fractures. At approximately 10:20 p.m., Linda Starks, a registered nurse, Karita Turner, a respiratory therapist, and Peggy Fowler, a nurse’s aide, prepared to transport Qualls from the neu-ro-intensive care unit to the CT suite.
Qualls was connected to a portable cardiac monitor and to a portable ventilator. The ventilator was attached to one of three portable oxygen tanks that accompanied Qualls to the CT suite. One of the tanks was full, and the remaining tanks were half-full. Ms. Turner did not record the portable ventilator’s settings and alarm parameters prior to leaving the neuro-inten-sive care unit. Nurse Starks did not verify the alarms or the parameters on Qualls’s portable cardiac monitor and did not make any entry in his medical record of either the alarm parameters or Qualls’s vital signs.
Upon arriving at the CT suite at approximately 10:30 p.m., Nurse Starks administered a paralytic drug to Qualls in prepa[126]*126ration for the CT scans. As a result, Qualls was unable to move or breathe on his own and became completely dependent on the portable ventilator. Nurse Starks never informed the other health care providers that she had given Qualls the drug.
Qualls was then moved from his hospital bed to the CT table. After Qualls was secured on the table, Nurse Starks, Ms. Turner, and the CT technician, Wayman Bean, moved to the adjoining control room, and Ms. Fowler returned to the neuro-intensive care unit to obtain fresh linens for Qualls’s bed.
From the control room, Nurse Starks and Ms. Turner could see the CT table through a glass window and could observe Qualls’s face on a television monitor. Neither of them looked at the cardiac monitor or the ventilator while Qualls was in the CT scanner. Nurse Starks testified that she was watching Qualls’s face and was listening for alarms. Both she and Ms. Turner testified that no alarms ever sounded, either during or after Qualls’s CT scans.
Mr. Bean took three scans of Qualls’s face. The first scan was taken at 10:47 p.m., and the last scan was taken at 10:56 p.m. After the scans were completed, Mr. Bean retracted the table from the scanner, and Nurse Starks and Ms. Turner prepared to move Qualls back to his bed. All three testified that Qualls’s complexion was “pink” and that he was breathing when he came out of the scanner.
Shortly after Qualls was moved to his bed, however, someone noticed that he had turned gray and was not breathing. Nurse Starks and Ms. Turner looked at the cardiac monitor and discovered that Qualls’s heart had stopped. Nurse Starks and Ms. Turner disconnected Qualls from the ventilator and immediately began administering CPR. Ms. Fowler called a “code” for additional personnel at approximately 11:05 p.m., and within minutes, the “code team” arrived and took over the resuscitation effort.
Although Qualls was successfully resuscitated, he sustained severe and permanent brain damage. Prior to his discharge from Vanderbilt, Qualls’s facial fractures were successfully treated. Since his discharge, Qualls has been a patient at NHC Healthcare, a skilled nursing facility in Dickson, Tennessee, and has remained in a persistent vegetative state.
The pending medical malpractice action was filed by Qualls’s sister, Sally Qualls Mercer, who was appointed as conservator for her brother. The plaintiffs theory at trial was that Ms. Turner attached the ventilator to an oxygen tank that was only half-full and that the tank ran out of oxygen during the CT scans, that Nurse Starks and Ms. Turner failed to appropriately monitor Qualls during the procedure, and that they either failed to activate or turned off the alarms on the cardiac monitor and the portable ventilator.
At trial, Vanderbilt admitted that Nurse Starks had violated the applicable standard of care in her treatment of Qualls on the night in question. However, Vanderbilt asserted that her violation of the standard of care was not the cause of Qualls’s permanent brain injury. Instead, Vanderbilt’s theory was that Qualls suffered a catastrophic event, such as a seizure or a malignant heart arrhythmia, which was caused by his alcohol withdrawal. Vanderbilt contends that it was this event that caused Qualls’s brain injury.
At the conclusion of the proof, the plaintiff moved for a directed verdict on the issue of comparative fault. The plaintiff argued that the holding in Gray v. Ford Motor Co., 914 S.W.2d 464 (Tenn.1996), did not apply because Qualls’s permanent brain injury was separate and distinct [127]*127from the injuries that resulted from the accident he caused by driving under the influence and therefore asserted that the jury should not be allowed to apportion any fault to him. The trial court denied the plaintiffs motion and submitted the case to the jury.
The jury returned a verdict in favor of the plaintiff, setting damages in the amount of $7,366,000. The jury apportioned 70% of the fault to Vanderbilt and 30% of the fault to Qualls. Based on the jury’s verdict, the trial court entered an order of judgment awarding the plaintiff $5,156,200.
The plaintiff filed a motion pursuant to Rule 50.02 of the Tennessee Rules of Civil Procedure, seeking a judgment entered in accordance with the plaintiffs motion for directed verdict and requesting the trial court to enter a judgment for $7,366,000, the full amount of damages found by the jury. Vanderbilt filed a motion for a new trial or, alternatively, for remittitur. In its motion, Vanderbilt asserted that: 1) the trial court erred in excluding evidence of Qualls’s prior alcohol-related conduct; 2) the trial court erred in excluding the testimony of Dr. John Salyer, a treating physician, and Jim Hutchison, a biomedical engineer; 3) the trial court erred in excluding expert testimony on the subject of the cost of an annuity to cover Qualls’s future medical care; and 4) the trial court improperly commented on the evidence by stating in the presence of the jury that Ms. Fowler was a “hostile witness” who was “changing her testimony.” Vanderbilt also raised other issues that are not pertinent to this appeal.
The trial court denied Vanderbilt’s motions but granted the plaintiffs Rule 50.02 motion. The court ruled that the injury sustained by Qualls on June 2, 1998, was, as a matter of law, a separate and distinct injury for which Qualls was not at fault. The court therefore ruled that Vanderbilt was 100% at fault and entered a judgment against Vanderbilt for the full amount of the damages found by the jury.
The Court of Appeals held that the trial court erred in granting the plaintiffs Rule 50.02 motion. The intermediate appellate court also reversed the trial court as to the first three issues listed above but held that the trial court’s comment as to Ms. Fowler, though improper, did not warrant reversal. The Court of Appeals therefore reversed the judgment of the trial court and remanded the case for a new trial.
We granted permission to appeal.
Analysis
I. Trial Court’s Ruling on Plaintiffs Rule 50.02 Motion
A. Comparative Fault
The first issue we must address in this appeal is the propriety of the trial court’s ruling that Vanderbilt was 100% at fault and therefore responsible for the full amount of the damages found by the jury. The plaintiff argues that the jury should not have been charged on the issue of comparative fault. Specifically, the plaintiff contends that this Court’s holding in Gray v. Ford Motor Co., 914 S.W.2d 464 (Tenn.1996), does not control this case. While we acknowledge that the present case could be distinguished from Gray, we are convinced that the need for clarity in this area of the law requires a reexamination of the underpinnings of Gray and its continued viability.
In Gray, the plaintiffs decedent was involved in a single-vehicle accident that was caused by her negligent operation of an automobile while under the influence of alcohol. 914 S.W.2d at 465. As a result of the accident, the decedent sustained a ruptured spleen. Id. at 465-66. The dece[128]*128dent’s injury was negligently misdiagnosed by her treating physician and ultimately led to her death. Id. The plaintiff brought a wrongful death action, claiming that the decedent’s death was proximately caused by the physician’s negligence in treating the decedent’s injury. Id. at 465. In its opinion, this Court reaffirmed its holding in Volz v. Ledes, 895 S.W.2d 677 (Tenn.1995), and held that the principles of comparative fault apply in medical malpractice actions, thus permitting the apportionment of fault between a patient who acts negligently in causing his or her initial injury and a physician who acts negligently in the treatment of the patient for that injury. Id. at 467. The Court concluded, without analysis, that the decedent suffered an indivisible injury and limited the scope of Gray’s holding to cases in which “[t]here [is] one indivisible injury proximately caused by the separate, independent acts of the [patient] and the physician.” Id. The Court explicitly stated that “[t]his case does not present, and the Court declines to address in this opinion, the rights and liabilities of the parties where there are multiple, separate injuries.” Id. The Court gave no analytical framework for the adoption of this distinction or for its implied premise that its holding was a necessary consequence of Volz.
In the present case, we find that this indivisible/separate injury distinction defies meaningful application. The plaintiff makes a strong argument that Qualls’s brain injury was separate and distinct from the injuries he sustained in the automobile accident. The plaintiff acknowledges that the mild-to-moderate concussion and facial fractures that Qualls sustained in the automobile accident were a result of his own negligence. However, the plaintiff argues that the brain injury that left Qualls in a persistent vegetative state occurred nearly four days after Qualls was admitted to Vanderbilt and was not a result of the accident or any negligence on Qualls’s part. An equally strong argument, however, could be made that Qualls suffered one, indivisible injury. Qualls’s negligence caused his facial fractures, and the incident that resulted in his brain injury occurred during the treatment of those fractures. Similarly, in Gray, the Court’s conclusion that the decedent had sustained an indivisible injury evidently was based on the fact that the decedent’s own negligence had caused the very injury that the physician negligently misdiagnosed. However, a valid argument could be made that the decedent suffered two separate injuries: her ruptured spleen, which was caused by her negligent operation of a motor vehicle, and her death, which was caused by her physician’s negligent misdiagnosis of her ruptured spleen.
Significantly, no other jurisdiction appears to utilize this indivisible/separate injury approach in determining whether principles of comparative fault or contributory negligence apply to medical malpractice actions. To the contrary, most jurisdictions have held that a patient’s negligence that provides only the occasion for medical treatment may not be compared to that of a negligent physician. See, e.g., Harvey v. Mid-Coast Hosp., 36 F.Supp.2d 32 (D.Me.1999) (holding that patient’s intentional or negligent ingestion of a drug may not be compared with the defendant physician’s subsequent, negligent treatment); Shinholster v. Annapolis Hosp., 255 Mich.App. 339, 660 N.W.2d 361 (2003) (holding that patient’s failure to regularly take her blood pressure medication in the year before her death could not be compared with the defendant physician’s negligent treatment and diagnosis of her condition); Harding v. Deiss, 300 Mont. 312, 3 P.3d 1286 (2000) (holding [129]*129that patient’s negligence in riding a horse when she had asthma and was allergic to horses could not be compared to the defendant physician’s failure to immediately intubate her upon her arrival at the hospital); Jensen v. Archbishop Bergan Mercy Hosp., 236 Neb. 1, 459 N.W.2d 178 (1990) (holding that patient’s failure to lose weight could not be compared with defendant physician’s negligence); Eiss v. Lillis, 233 Va. 545, 357 S.E.2d 539 (1987) (holding that patient’s negligent ingestion of aspirin and heart medication could not be compared with the defendant physician’s negligence). Several of these jurisdictions have concluded that a patient’s negligence in causing a motor vehicle accident may not be compared with the defendant physician’s subsequent, negligent treatment of the injuries that the patient sustained in the accident. See, e.g., Martin v. Reed, 200 Ga.App. 775, 409 S.E.2d 874 (1991); Fritts v. McKinne, 934 P.2d 371 (Okla.Ct.App.1996); Sendejar v. Alice Physicians & Surgeons Hosp. Inc., 555 S.W.2d 879 (Tex.Civ.App.1977); Rowe v. Sisters of the Pallottine Missionary Soc’y, 211 W.Va. 16, 560 S.E.2d 491 (2001). These jurisdictions conclude that a health care provider may not reduce or avoid liability for negligent treatment by asserting that the patient’s injuries were originally caused by the patient’s own negligence. The Restatement of Torts reiterates this view. According to the Restatement, “in a case involving negligent rendition of a service, including medical services, a factfinder does not consider any plaintiffs conduct that created the condition the service was employed to remedy.” Restatement (Third) of Torts: Apportionment of Liability § 7 cmt. m (2000). The reporter’s note to this comment explains that it would be unfair to allow a defendant doctor to complain about the patient’s negligence because this negligence caused the very condition the doctor undertook to treat. Restatement (Third) of Torts: Apportionment of Liability § 7 reporter’s note to cmt. m (2000).
A majority of jurisdictions have allowed a patient’s fault to be considered in medical malpractice cases only under very limited circumstances. See Fritts, 934 P.2d at 374. For example, some jurisdictions have allowed juries to apportion fault to a patient who delays in seeking or returning for medical treatment, see LeBlanc v. N. Colfax County Hosp., 100 N.M. 494, 672 P.2d 667, 669-70 (1983), who fails to follow a physician’s -advice or instructions, see Musachia v. Rosman, 190 So.2d 47, 50 (Fla.Dist.Ct.App.1966), who furnishes false, incomplete, or misleading information to his or her physician, see Rochester v. Katalan, 320 A.2d 704, 708 (Del.1974), or who attempts to treat his or her own injury before seeking medical attention, see Sales v. Bacigalupi, 47 Cal.App.2d 82, 117 P.2d 399, 402 (1941). In Volz v. Ledes, this Court upheld the jury’s allocation of fault to the patient who delayed in returning for a follow-up examination with his physician. 895 S.W.2d 677, 678, 680 (Tenn.1995). As such, Volz falls into the category of cases in which a patient delays seeking or returning for medical treatment. Consequently, our decision in Volz did not dictate the result in Gray, and Volz does not control the present case.
Although we recognize that Gray is supported by the principle of stare decisis, we conclude that the majority of jurisdictions have adopted the better-reasoned view. As one court has aptly stated,
It would be anomalous to posit, on the one hand, that a health care provider is required to meet a uniform standard of care in its delivery of medical services to all patients, but permit, on the other hand, the conclusion that, where a breach of that duty is established, no [130]*130liability may exist if the patient’s own preinjury conduct caused the illness or injury which necessitated the medical care.
Harvey, 36 F.Supp.2d at 38. We also agree that “ ‘patients who may have negligently injured themselves are nevertheless entitled to subsequent non-negligent medical treatment and to an undiminished recovery if such subsequent non-negligent treatment is not afforded.’ ” Fritts, 934 P.2d at 374 (quoting Martin, 409 S.E.2d at 877). We therefore hold that a patient’s negligent conduct that occurs prior to a health care provider’s negligent treatment and provides only the occasion for the health care provider’s subsequent negligence may not be compared to the negligence of the health care provider. To the extent that Gray holds otherwise, it is hereby overruled.
In the present case, Qualls’s negligence merely provided the occasion for the medical care, attention, and treatment that gave rise to this medical malpractice action. We therefore hold that the principles of comparative fault do not apply so as to allow fault to be assessed to Qualls. We recognize that Qualls’s medical treatment was complicated by his alcohol withdrawal and that evidence concerning his alcohol consumption was clearly relevant to his treatment and to Vanderbilt’s theory of causation. We hold, however, that Qualls’s antecedent negligence should not have been considered by the jury in assessing fault.
B. Tennessee Rule of Civil Procedure 50.02
After receiving the jury’s verdict, which apportioned 30% of the fault to Qualls, the trial court granted the plaintiffs Rule 50.02 motion and ruled that Vanderbilt was 100% at fault. In accordance with this ruling, a judgment was entered against Vanderbilt for the full amount of the damages found by the jury. We now must evaluate the trial court’s ruling on the motion for judgment in accordance with the plaintiffs motion for directed verdict. Under Rule 50.02, if a motion for directed verdict is not granted, the case is deemed to have been submitted to the jury subject to a later determination of the legal questions raised by the motion. See Tenn. R. Civ. P. 50.02.2 Therefore, when the jury’s verdict rests upon an error of law, a party who has moved for a directed verdict may request the trial court to set aside the verdict and enter a judgment in accordance with the party’s motion for directed verdict. See id.
In ruling on such a motion, the standard applied by both the trial court and the appellate court is the same as that applied to a motion for directed verdict made during trial. See Holmes v. Wilson, 551 S.W.2d 682, 685 (Tenn.1977); Mairose v. Fed. Express Corp., 86 S.W.3d 502, 511 (Tenn.Ct.App.2001). Therefore, the trial court and appellate court are required to take the strongest legitimate view of the evidence in favor of the opponent of the motion, allow all reasonable inferences in his or her favor, discard all countervailing evidence, and deny the motion when there [131]*131is any doubt as to the conclusions to be drawn from the evidence. See Williams v. Brown, 860 S.W.2d 854, 857 (Tenn.1993). A verdict should not be directed during, or after, trial except where a reasonable mind could draw but one conclusion. See id.
Having held that the principles of comparative fault do not apply to the present case, we conclude that the trial court acted properly in ruling that Vanderbilt was 100% at fault and therefore responsible for the full amount of the damages found by the jury. By assessing Vanderbilt 70% of the fault, the jury clearly found that Vanderbilt’s negligence caused Qualls’s brain injury. Because Qualls’s antecedent negligence cannot be considered, the trial court did not err in removing the jury’s finding of fault against him, leaving Vanderbilt as the sole, remaining cause of Qualls’s injuries. As such, the trial court’s actions were not a “reallocation” of fault in contravention of our holding in Turner v. Jordan, 957 S.W.2d 815, 824 (Tenn.1997). Furthermore, the court was not exercising its role as thirteenth juror. A trial court exercises its authority as thirteenth juror when it finds that the verdict is contrary to the weight of the evidence, as opposed to being based upon an error of law. See Turner, 957 S.W.2d at 828; James E. Strates Shows, Inc. v. Jakobik, 554 S.W.2d 613, 615 (Tenn.1977). Here, the trial court was ruling on a valid Rule 50.02 motion and adjusted a verdict that was based on an error of law in accordance with that rule.
II. Prior Alcohol-Related Conduct
In addition to the issue of comparative fault, Vanderbilt advances many of the issues that it raised in its Motion for New Trial. The first of these issues involves the trial court’s exclusion of evidence regarding Qualls’s alcohol-related conduct prior to his May 30,1998 accident. Specifically, the trial court excluded evidence of the following two incidents. First, on April 13, 1996, a Lewis County Sheriffs Deputy saw Qualls erratically operating a motor vehicle. The deputy stopped Qualls’s automobile and administered a breathalyzer test, which showed that Qualls’s blood alcohol level exceeded .10%. Qualls pleaded guilty to the offense of driving under the influence and was sentenced to forty-eight hours in jail and eleven months, twenty-nine days of probation. Second, on November 8, 1997, a Perry County Sheriffs Deputy arrested Qualls for the aggravated assault of his wife. The deputy observed that Qualls was intoxicated at the time of arrest and would have testified that Qualls resisted arrest and attempted to flee from arresting officers. Qualls pleaded guilty to the offense of aggravated assault and was sentenced to thirty days in jail and was ordered to attend Alcoholics Anonymous meetings. The trial court ruled that this evidence was not relevant. We agree.
Generally, the admissibility of evidence is within the sound discretion of the trial court. Otis v. Cambridge Mut. Fire Ins. Co., 850 S.W.2d 439, 442 (Tenn.1992). The trial court’s decision to admit or exclude evidence will be overturned on appeal only where there is an abuse of discretion. Id. A trial court abuses its discretion “only when it ‘applie[s] an incorrect legal standard, or reache[s] a decision which is against logic or reasoning that cause[s] an injustice to the party complaining.’ ” Eldridge v. Eldridge, 42 S.W.3d 82, 85 (Tenn.2001) (quoting State v. Shirley, 6 S.W.3d 243, 247 (Tenn.1999)).
In personal injury actions, damages primarily compensate the wronged party for his or her injuries and are intended to make the wronged party whole. Overstreet v. Shoney’s, Inc., 4 S.W.3d 694, 703 (Tenn.Ct.App.1999). Loss [132]*132of earning capacity is an element of damages in personal injury actions, see id. (citing Wolfe v. Vaughn, 177 Tenn. 678, 152 S.W.2d 631, 635 (1941)), as is the injured party’s future medical expenses, see Newman v. Aluminum Co. of Am., 643 S.W.2d 109, 111 (Tenn.Ct.App.1982). As such, evidence relating to either of these elements would be relevant in a personal injury action.3
Qualls’s prior alcohol-related convictions, however, clearly are not relevant to his loss of earning capacity or to his future medical expenses. Qualls’s occupation did not involve driving, and there was no contention that Qualls drank alcohol during working hours or that his use of alcohol affected his ability to perform competently at work. In addition, the proffered evidence of Qualls’s prior convictions would not be relevant to determine the amount of future medical expenses he would require over his expected lifetime. In contrast, Qualls’s alcohol consumption leading to his admission to Vanderbilt clearly was relevant. The jury heard a considerable amount of evidence related to Qualls’s alcoholism. This evidence was properly admitted to show the medical treatment Qualls received that was related to his alcohol withdrawal and to support Vanderbilt’s theory that Qualls suffered a catastrophic event caused by his alcohol withdrawal.
Vanderbilt asserts that Qualls’s prior convictions involving his use of alcohol are relevant because they show the pecuniary value of his life. Vanderbilt contends that we should rely upon wrongful death cases that address the pecuniary value of the life of a decedent because the issues of a decedent’s life expectancy and loss of earning capacity are similar to that of a plaintiff who is in a persistent vegetative state. We disagree. Pecuniary value is judicially defined as a part of the incidental damages that may be recovered by the decedent’s survivors to compensate them for the decedent’s death. Jordan v. Baptist Three Rivers Hosp., 984 S.W.2d 593, 600 (Tenn.1999). Pecuniary value takes into account all aspects of a decedent’s life, including his or her services and companionship, so as to place a value on the decedent’s life. Id. These damages are not available in a personal injury action, regardless of the medical condition of the injured party. Consequently, we conclude that the trial court did not commit reversible error by excluding the proffered evidence.
III. Witness Exclusion
Vanderbilt also contends that the trial court committed reversible error by excluding the testimony of Dr. John Sal-yer, one of Qualls’s treating physicians at NHC Healthcare, and Mr. James Hutchi-son, a biomedical engineer who is employed by Vanderbilt. The trial court excluded this evidence in response to Vanderbilt’s failure to supplement its answers to plaintiffs interrogatories pursuant to Rule 26.05(1) of the Tennessee Rules of Civil Procedure. Under this rule, “[a] party is under a duty seasonably to supplement the party’s response with respect to any question directly addressed to ... the identity and location of persons having knowledge of discoverable matters.... ” Tenn. R. Civ. P. 26.05(1). The court found that the plaintiff would need an additional three to six weeks to retain additional experts and [133]*133prepare for these “surprise witnesses.” Therefore, the court concluded that witness exclusion was an appropriate sanction for Vanderbilt’s discovery abuse.
Although the Tennessee Rules of Civil Procedure do not provide a sanction for abuse of the discovery process, trial judges have the authority to take such action as is necessary to prevent discovery abuse. Lyle v. Exxon Corp., 746 S.W.2d 694, 699 (Tenn.1988); Strickland v. Strickland, 618 S.W.2d 496, 501 (Tenn.Ct.App.1981). Trial courts have wide discretion to determine the appropriate sanction to be imposed. Strickland, 618 S.W.2d at 501. Such a discretionary decision will be set aside on appeal only when “the trial court has misconstrued or misapplied the controlling legal principles or has acted inconsistently with the substantial weight of the evidence.” White v. Vanderbilt Univ., 21 S.W.3d 215, 223 (Tenn.Ct.App.1999) (citing Overstreet v. Shoney’s, Inc., 4 S.W.3d 694, 709 (Tenn.Ct.App.1999)). Appellate courts should allow discretionary decisions to stand even though reasonable judicial minds can differ concerning their soundness. White, 21 S.W.3d at 223; Overstreet, 4 S.W.3d at 709.
Excluding a witness’s testimony may be an appropriate sanction for failure to supplement answers to interrogatories. See Ammons v. Bonilla, 886 S.W.2d 239, 243 (Tenn.Ct.App.1994); see also Lyle, 746 S.W.2d at 699. In determining whether such a sanction is appropriate, the trial court should consider several factors: 1) the party’s reasons for not providing the challenged evidence during discovery; 2) the importance of the evidence; 3) the time needed for the other side to prepare to meet the evidence; and 4) the propriety of granting a continuance. Lyle, 746 S.W.2d at 699. In the present case, the trial court considered each of these factors and determined that witness exclusion was an appropriate sanction for Vanderbilt’s failure to supplement its answers to the plaintiffs interrogatories. The court found that the plaintiff would need an additional three to six weeks to prepare to meet the challenged evidence and concluded that a continuance was therefore not a reasonable option. Moreover, in pronouncing its ruling, the court stated, “I could not ascertain any good reason for the delay of the disclosure.” We therefore conclude that the trial court acted within its discretion by excluding the testimony of these witnesses.
IV. Testimony of Annuity Specialist
Vanderbilt argues that the trial court erred in precluding testimony related to the cost of an annuity. Vanderbilt contends that this evidence would have assisted the jury in determining an appropriate amount of damages for future medical expenses. Generally, questions regarding the admissibility, qualifications, relevancy, and competency of expert testimony are left to the trial court’s discretion, and the trial court’s ruling in this regard may be overturned only if the discretion is arbitrarily exercised or abused. McDaniel v. CSX Transp. Inc., 955 S.W.2d 257, 263-64 (Tenn.1997).
In the present case, the trial court excluded the testimony of Bruce Wolfe, a structural annuity specialist, offered by Vanderbilt on the issue of damages. Vanderbilt’s offer of proof indicated that Mr. Wolfe would have testified about the cost of an annuity policy that could have been purchased to ensure a stream of cash payments to cover Qualls’s future medical expenses. The trial court ruled that this evidence was “too speculative.”
We conclude that the trial court did not abuse its discretion by excluding this testimony. Many changing variables affect the [134]*134quote that an annuitist delivers to the jury. For instance, time limits and market factors both impact annuity rates. Moreover, an insurance company is in no way bound to the quoted rate or to its initial underwriting decision. These factors not only make the testimony as to the cost of an annuity speculative, but they also raise questions about its potential to mislead the jury. See Gold, Vann & White, P.A. v. DeBerry, 639 So.2d 47, 55 (Fla.Dist.Ct.App.1994). Furthermore, such testimony invites the jury to depart from its legal duty to award present cash value. See Gusky v. Candler Gen. Hosp. Inc., 192 Ga.App. 521, 385 S.E.2d 698, 701 (1989); Herman v. Milwaukee Children’s Hosp., 121 Wis.2d 531, 361 N.W.2d 297, 306 (1984); see also Waller v. Skeleton, 31 Tenn.App. 103, 212 S.W.2d 690, 698 (1948) (suggesting that an award of future medical expenses must be reduced to present value). Therefore, the trial court did not err in excluding this testimony.
V. Trial Court’s Comment on the Evidence
Vanderbilt also takes issue with a comment the trial court made in ruling on an objection. During the direct examination of Ms. Fowler, Vanderbilt objected to certain questions as leading. The trial judge overruled the objection, stating that the witness is a “hostile witness” who is “changing her testimony.” At the conclusion of Ms. Fowler’s testimony and outside the presence of the jury, Vanderbilt objected to the court’s comment.
Judges are prohibited from commenting upon the credibility of witnesses or upon the evidence in a case. See Tenn. Const. art. VI, § 9 (stating that “[t]he judges shall not charge juries with respect to matters of fact, but may state the testimony and declare the law”). Therefore, trial judges must be “very careful not to give the jury any impression as to his [or her] feelings or to make any statement which might reflect upon the weight or credibility of evidence or which might sway the jury.” State v. Suttles, 767 S.W.2d 403, 407 (Tenn.1989); see also Kanbi v. Sousa, 26 S.W.3d 495, 498-99 (Tenn.Ct.App.2000). These restrictions apply to comments made when ruling on an objection. Loeffler v. Kjellgren, 884 S.W.2d 463, 474 (Tenn.Ct.App.1994).
Even though judges need to be circumspect in this area, not every comment on the evidence made by a judge is grounds for a new trial. Kanbi, 26 S.W.3d at 499. We must consider the trial court’s comment in the overall context of the case to determine whether the comment was prejudicial. State v. Caughron, 855 S.W.2d 526, 536-37 (Tenn.1993). Although we agree with Vanderbilt that this comment was improper, we do not believe that it rises to the level of prejudice that would require the granting of a new trial.
First, the court’s comment was directed at a small part of Ms. Fowler’s testimony. The questions posed by plaintiffs counsel concerned the location of the three other witnesses at the time that Ms. Fowler returned to the CT suite. During her deposition, Ms. Fowler stated that when she returned to the CT suite Nurse Starks, Ms. Turner, and Mr. Bean were “standing around.” However, at trial, she testified that they were “standing around the bed.” In the context of the entire trial, which lasted approximately two weeks, any prejudicial effect that resulted from the court’s comment would have been slight.
Second, at the end of the proof, the trial court charged the jury concerning its role and its function as the trier of fact. It is presumed that jurors understand and follow the court’s instructions. Memphis [135]*135St. Ry. Co. v. Cooper, 203 Tenn. 425, 313 S.W.2d 444, 449 (1958); Bass v. Barksdale, 671 S.W.2d 476, 489 (Tenn.Ct.App.1984). The court instructed the jury that it was “the sole and exclusive judge[ ] of the credibility and believability of the witnesses in this case.” The court also instructed the jury that “[tjhere may be discrepancies or differences within a witness’ testimony” but that “[tjhis does not necessarily mean that a witness should be disbelieved.” In light of these instructions, we conclude that the trial court’s isolated remark did not have any prejudicial effect on the outcome of the trial.
Finally, the trial court offered to give a curative instruction with regard to its comment, but Vanderbilt refused the court’s offer and failed to request a more appropriate instruction. Therefore, Vanderbilt cannot rely on the trial court’s improper comment on the evidence as grounds for overturning the jury’s verdict on appeal. See State v. Griffis, 964 S.W.2d 577, 599 (Tenn.Crim.App.1997) (“If a party fails to request a curative instruction, or, if dissatisfied with the instruction given and [sic] does not request a more complete instruction, the party effectively waives the issue for appellate purposes.”).
Conclusion
We overrule Gray v. Ford Motor Co., 914 S.W.2d 464 (Tenn.1996), and hold, as a matter of law, that the principles of comparative fault do not apply in medical malpractice actions in which a patient’s negligent conduct provides only the occasion for the medical attention, care, or treatment which is the basis for the action. Therefore, we conclude that the trial court properly granted the motion for judgment in accordance with the plaintiffs motion for directed verdict. We also hold that the issues raised by Vanderbilt are without merit and do not require us to grant a new trial. Accordingly, the judgment of the Court of Appeals is affirmed in part and reversed in part, and the jury’s verdict in the amount of $7,366,000 is reinstated without regard to the jury’s assessment of fault to Qualls. Costs of this appeal are taxed to the appellee, Vanderbilt, and its surety, for which execution may issue if necessary.
FRANK F. DROWOTA, III, C.J., filed a dissenting opinion.