JUSTICE McMORROW delivered the opinion of the court:

In this appeal we are asked to resolve whether application of the "loss of chance" doctrine in medical malpractice cases lessens the plaintiff's burden of proving proximate cause. This question has caused conflicting opinions among the Illinois appellate court panels. The loss of chance concept refers to the harm resulting to a patient when negligent medical treatment is alleged to have damaged or decreased the patient's chance of survival or recovery, or to have subjected the patient to an increased risk of harm.

Defendant, Memorial Hospital, requests this court to reverse the judgment entered upon the jury's verdict in favor of plaintiffs, Patricia and Frank Holton. Defendant argues that plaintiffs failed to establish that any acts or omissions of defendant's staff proximately caused plaintiffs' injuries. Alternatively, defendant contends that if the judgment is not reversed outright it must be reversed and a new trial ordered because defendant was prejudiced by conduct of plaintiffs' attorney and by the failure of the trial court to maintain impartiality. In addition, defendant challenges certain jury instructions, and also argues that it is entitled to a setoff in the amount of Mrs. Holton's past medical expenses which were reimbursed by insurance.

We allowed defendant's petition for leave to appeal. 155 Ill.2d R. 315. For the reasons that follow, we reverse the judgment of the appellate court and remand for new trial.

Background

In 1991, Patricia Holton and her husband Frank filed suit in the circuit court of St. Clair County against Memorial Hospital, Radiological Associates, Limited, and William G. Doubek, M.D., seeking damages for Patricia Holton's personal injuries and for Frank Holton's loss of consortium, which allegedly resulted from the defendants' negligence. Subsequently, plaintiffs settled their claims against William Doubek, Radiological Associates, and three other defendants named in a companion case. The trial court found that the $2,950,000 settlement was made in good faith and subsequently dismissed Doubek and Radiological Associates from the case at bar, leaving Memorial Hospital as the sole remaining defendant.

Testimony at trial revealed that Patricia Holton began to suffer severe back pain in late November or early December 1990. Her primary care physician, Dr. Doubek, ordered an X ray and a bone scan. These procedures indicated that the patient suffered from a degenerative process or compression fracture to a vertebra in her thoracic spine. Dr. Doubek scheduled Mrs. Holton to undergo a magnetic resonance imaging (MRI) on January 4, 1991, the earliest available date for nonemergencies. He also told her to go to the emergency room if her pain worsened.

At approximately 9:30 p.m. on December 26, 1990, Mrs. Holton went to the emergency room of Memorial Hospital in Belleville, Illinois, complaining of numbness below the waist and a tingling sensation in her left leg. Dr. Mark Jergens, an emergency room physician, examined Mrs. Holton. He found evidence of a low fever and an elevated white blood count, which indicated the presence of an infection. At this time, according to the trial testimony of Dr. Jergens, Mrs. Holton had not lost any motor skills. The doctor ordered a blood culture and a CAT scan. On the medical records Dr. Jergens noted that a thoracic epidural abscess was a possible explanation for Mrs. Holton's symptoms. An epidural abscess is a collection of pus in the epidural area adjacent to the membrane which covers the spinal cord. Because emergency room physicians at the hospital did not have admitting privileges, Dr. Jergens called Mrs. Holton's primary care physicians to order her admission to Memorial Hospital.

Dr. Doubek examined Mrs. Holton early on the morning of December 27, 1990, and discovered that she had tingling, numbness, and weakness in her lower extremities. He was aware that her symptoms were consistent with either a bone infection called osteomyelitis or a tumor in her spine. Dr. Doubek ordered a neurological consultation.

Dr. Murphy, a neurosurgeon, examined Mrs. Holton in the late afternoon of December 27, 1990. She complained of numbness in her abdomen and legs but could still move her extremities. The CAT scan confirmed the existence of a compression fracture. The radiologist who interpreted the CAT scan was of the opinion that Mrs. Holton's pain was caused by a cancerous tumor.

Mrs. Holton testified that during the day of December 27, 1990, she noticed increasing difficulty in moving her left leg. She informed her attending nurses of this condition. However, the nurses' notes state only that the patient did not experience any significant change in her condition during the day. Between 6 and 7 p.m. on December 27, 1990, Mrs. Holton walked to the bathroom unassisted. After a few moments she attempted to rise and return to her bed. However, she could not move her legs or stand up and noticed a particular problem with numbness in her left leg. She rang the bell for help. Two nurses aides helped her into a wheelchair and from there, into her bed. Although she remarked when the nurses aides helped her from the toilet that her legs "didn't want to work," neither aide reported this incident to a supervisor, a nurse, or a doctor.

Registered nurse Barbara Ford cared for Mrs. Holton during the shift between 11 p.m. on December 27, 1990, and 7 a.m. on December 28, 1990. At some time between 1 a.m. and 5 a.m. on December 28, Nurse Ford determined that Mrs. Holton was having difficulties moving her left leg, but she did not believe that this was a significant change of condition.

After the next change in nursing shifts, between 8 and 9 a.m. on December 28, Mrs. Holton complained of numbness from the waist down and an inability to move her legs except for very slight movement in her right foot. She also lost bowel and bladder control. The nurse then on duty, Susan Schindler, informed Dr. Doubek and the neurosurgeon on call of these developments. Dr. Doubek came immediately and confirmed that Mrs. Holton had suffered a complete loss of motor control below the waist. Dr. Sprich, the neurosurgeon on call for Dr. Murphy, ordered tests to determine where the pressure on the spinal cord was located.

Dr. Doubek testified at trial that at the time of his diagnosis he was operating under the assumption that Mrs. Holton had suffered a sudden onset of paralysis because he had not been informed otherwise by the hospital staff or the charts and records. Accordingly, he initially thought it likely that her condition was caused by a tumor-caused infarct of the blood supply to the spinal cord, which would be consistent with sudden and complete loss of motor function. Defendant's nursing staff did not tell Dr. Doubek that Mrs. Holton's numbness and other symptoms of sensory deficits had been progressing over several hours to a state of motor impairment or partial paralysis. According to the neurological evidence adduced at trial, this escalation from tingling and numbness to paresis, or partial loss of motor function, indicates spinal cord compression caused by osteomyelitis (infection of the bone). Osteomyelitis can cause either an inflammation or an abscess that puts pressure on the spine, resulting in the progression of paralysis. Because Dr. Doubek was unaware that Mrs. Holton's apparently sudden paralysis had been preceded by a more gradual onset of declining motor function, Dr. Doubek determined that the most likely cause of plaintiff's condition was a cancerous tumor, requiring radiation.

Similarly, Dr. Sprich, the neurosurgeon on call for Dr. Murphy, testified that he would have come to the hospital immediately if he had been informed that Mrs. Holton was beginning to have trouble walking and moving her legs. By the time she lost bowel and bladder function, according to Dr. Sprich, her condition was irreversible. He further testified that patients suffering from neurological conditions whose symptoms are consistent with an epidural abscess can usually be beneficially treated because there is sufficient time to confirm the diagnosis and perform surgery to ease the pressure on the spinal cord. Dr. Sprich testified that if cord compression caused by an epidural abscess is diagnosed within 24 hours of the occurrence of paresis, patients often have "an excellent neurological recovery" and, if caught at an early enough point in the paresis, such patients will be able to "move their legs."

Dr. Doubek and Dr. Sprich also explained why they agreed that the apparently sudden onset of Mrs. Holton's paralysis supported the diagnosis of cancerous tumor rather than osteomyelitis. Osteomyelitis is a rare occurrence in the absence of a prior surgical intervention or trauma to the spinal cord, and Mrs. Holton did not have such a history. Also, statistical probabilities favored the existence of cancerous tumor under the perceived suddenness of the paralysis, especially where, as here, there was no involvement of the disc area of the spine. Consequently, the doctors who diagnosed and treated Mrs. Holton concluded that the cause of her condition was a cancerous tumor and proceeded to treat her based upon that mistaken assumption.

On January 17, 1991, three weeks after her loss of motor function, Mrs. Holton's family transferred her to another hospital because she was not improving. At that hospital, Mrs. Holton was diagnosed with and treated for osteomyelitis. Her current condition is paraplegia with bladder and bowel involvement due to a spinal injury caused by upper thoracic vertebrae osteomyelitis.

Following the trial, the jury returned a verdict against Memorial Hospital and in favor of Mrs. Holton in the amount of $8,706,500 and a verdict in favor of Mr. Holton on his loss of consortium claim in the amount of $110,000. Defendant filed a post trial motion seeking alternative relief including judgment n.o.v., a new trial on all issues, a new trial on damages only, and a remittitur of Mrs. Holton's future medical expenses. Defendant further sought to have its liability reduced by the amount of the plaintiffs' settlement with other defendants. The circuit court granted a remittitur in the amount of $1,500,000 but denied defendant's motion in all other respects.

The appellate court affirmed the judgment of the circuit court, as modified to reflect a reduction of the verdict by the amount of plaintiffs' settlement with other defendants, $2,950,000, pursuant to the Contribution Act. 274 Ill. App.3d 868. The amount awarded to plaintiffs, as reduced on appeal, totals $4,366,500.

Analysis

In this court, defendant challenges plaintiffs' proof of proximate causation and challenges the propriety of certain jury instructions. Defendant further argues that it was denied a fair trial because of plaintiffs' counsel's prejudicial remarks and a statement of the trial judge that damaged the credibility of defendant's attorneys and one of its witnesses. Finally, defendant contends that the trial court erred in refusing to set off from the judgment the amount of Mrs. Holton's medical expenses reimbursed by insurance. We address these contentions in turn.

I

Defendant first argues that the appellate court's opinion in the case at bar improperly altered, diminished, or diluted plaintiffs' burden of proving that defendant's negligence proximately caused plaintiffs' injuries. According to defendant, the appellate decisions "are in patent conflict on the sufficiency of evidence necessary to prove proximate causation in a medical malpractice case." See generally Comment, Lost Chance of Survival in Illinois: The Need for Guidance from the Illinois Supreme Court, 23 Loy. U. Chi. L.J. 155, 156 (1991). One line of decisions holds that proximate cause may be established by evidence, to a reasonable degree of medical certainty, that the hospital, doctor, or other health care provider "increased the risk of harm" to plaintiff or "lessened the effectiveness" of plaintiff's treatment by the defendant's negligent conduct. See, e.g., Hajian v. Holy Family Hospital, 273 Ill. App.3d 932, 939 (1st Dist. 1995); Galvin v. Olysav, 212 Ill. App.3d 399, 403 (5th Dist. 1991); Chambers v. Rush-Presbyterian-St. Luke's Medical Center, 155 Ill. App.3d 458, 463-65 (1st Dist. 1987); Northern Trust Co. v. Louis A. Weiss Memorial Hospital, 143 Ill. App.3d 479, 487-88 (1st Dist. 1986). The approach taken in these and similar cases has been termed the "loss of chance" or "lost chance" doctrine. See generally T. Lavin G. Ziebell, Lost Chance of Survival: Is it a Lost Cause in Illinois? 84 Ill. B.J. 458 (1996).

Another line of Illinois appellate cases has rejected the loss of chance doctrine as giving rise to a relaxed standard of proximate cause, one that is too conjectural to satisfy the traditional test of proximate causation. See, e.g., Netto v. Goldenberg, 266 Ill. App.3d 174, 180-81 (2d Dist. 1994); Hare v. Foster G. McGaw Hospital, 192 Ill. App.3d 1031, 1038 (1st Dist. 1989); Russell v. Subbiah, 149 Ill. App.3d 268 (3d Dist. 1986); Curry v. Summer, 136 Ill. App.3d 468, 476 (4th Dist. 1985).

Defendant contends that the appellate court's application of the "lost chance" doctrine in the case at bar allowed plaintiffs to establish their cause of action without being held to the traditional standard of proving causation approved by this court in Borowski v. Von Solbrig, 60 Ill.2d 418, 424 (1975). Borowski requires proof of causation by the preponderance of evidence, otherwise referred to as the "more probably true than not true" standard, that the negligence complained of caused plaintiffs injury. Before evaluating the conflicting appellate decisions cited above, we revisit the Borowski standard of proximate causation in medical malpractice actions, and determine whether, under that standard, plaintiffs' evidence was sufficient to withstand defendant's motion for judgment notwithstanding the verdict.

A

In Borowski, this court did not directly address the loss of chance doctrine. Instead, the issue relevant to causation was whether an injured plaintiff who allegedly received negligent medical treatment could establish proximate cause without presenting evidence that a better result would have obtained if proper treatment had been administered. The plaintiff in Borowski was struck by an automobile while crossing a street and received severe injuries to his legs. The medical malpractice claim arose from the defendant hospital's alleged negligence in its treatment of the plaintiff's injuries, which resulted in the amputation of the plaintiff's left leg. The plaintiff's expert witnesses testified that in their opinion, plaintiff's leg could have been successfully repaired had proper tests been performed and procedures followed, and that delay in the plaintiff's surgery was a proximate cause of the amputation. Defendants countered that there was insufficient evidence to support the verdict in favor of plaintiff on the causation issue. Further, defendants contended that they were "entitled to judgment because the evidence d[id] not establish that a better result would have been obtained if proper treatment were administered." Borowski, 60 Ill.2d at 424.

This court rejected the defendants' argument, stating that it was "unnecessary to extend the burden-of-proof requirements of a medical malpractice case beyond those of an ordinary negligence case by adding the further requirement that the plaintiff prove a better result would have been achieved absent the alleged negligence of the doctor." Borowski, 60 Ill.2d at 424. The court reiterated that under accepted Illinois Pattern Jury Instructions, plaintiff's burden of proving that defendants' negligence was the proximate cause of his injury was sustainable by proof that the proposition in issue — defendants' breach of duty caused plaintiff's injury — was more probably true than not true. Addressing the defendants' contention that the jury should not be permitted to speculate upon the relative amount of injury attributable to the fracture caused by the original accident and the amount attributable to the malpractice, the Borowski court stated, "[Speculation] can be guarded against by the use of appropriate instructions to the jury. It is not necessary to become involved in all of the collateral ramifications that the 'better result' test could inject into a case." Borowski, 60 Ill.2d at 424.

We reaffirm the Borowski holding. The traditional statement of proximate cause requires plaintiff to prove that defendant's negligence "more probably than not" caused plaintiff's injury. The "better result test" is not a part of plaintiff's burden of proof. Issues involving proximate cause are fact specific and therefore uniquely for the jury's determination. When a plaintiff comes to a hospital already injured, as in the case of Borowski, or has an existing undiagnosed medical condition, as in the case at bar, and while in the care of the hospital is negligently treated, the question of whether the defendant's negligent treatment is a proximate cause of plaintiff's ultimate injury is ordinarily one of fact for the jury.

In the case at bar, defendant asserts that it was entitled to judgment as a matter of law for the failure of plaintiffs to present expert testimony that an earlier call to Mrs. Holton's physicians would have prevented her paralysis. In light of Borowski's rejection of the "better result" test, Mrs. Holton was not required to prove that an earlier call to her physicians would have resulted in a more favorable outcome. Moreover, Mrs. Holton did not base her case solely on defendant's delay in the reporting of her condition. Instead, she contended that the failure of defendant's nursing staff to accurately report the progression of her decline into paresis was a proximate cause of her paralysis. The record contains evidentiary support for plaintiffs' theory. Both Dr. Sprich and Dr. Doubek explained that they based their erroneous diagnosis and treatment decisions upon inaccurate and incomplete information regarding Mrs. Holton's condition in the hours preceding her total loss of motor control. Dr. Sprich testified that when a patient's paresis (partial paralysis) is detected and treated early enough there is a good probability of avoiding or minimizing paralysis. Dr. Doubek testified that, to a reasonable degree of medical certainty, the preferred treatment for relieving pressure on the spinal cord caused by an abscess or edema is decompression or drainage. Had the doctors been given the opportunity to properly diagnose Mrs. Holton's condition based on accurate and complete information, they would have had the opportunity to treat her condition by ordering the appropriate treatment. Because of the hospital's negligent failure to accurately and timely report Mrs. Holton's symptomotology, the appropriate treatment was not even considered. In light of these facts, we conclude that defendant failed to establish that it was entitled to judgment n.o.v. based on plaintiffs' failure to prove that an earlier call to the treating physicians would have resulted in Mrs. Holton's recovery.

Defendant presents the additional argument that judgment n.o.v. is justified in the case at bar because Mrs. Holton's personal physicians rendered ineffective treatment both before and after being notified of her loss of motor skills; therefore, she did not establish that her physicians would have acted differently had they been earlier notified. In support, defendant cites to Gill v. Foster, 157 Ill.2d 304 (1993), where summary judgment was entered in favor of the defendant hospital despite the failure of a nurse to notify a physician that a patient being discharged from the hospital complained of chest pains. Gill held that summary judgment was appropriate in that case because there was no indication that the doctor, who was aware of his patient's complaint and had decided it was not significant, would have done anything differently had the nurse repeated the patient's complaint to the doctor.

Gill is inapposite to the case at bar because in that case the nurse's report of the complaint would not have caused any further action on the part of the doctor. In contrast, there is testimony in the instant case that the doctors would have undertaken a different course of treatment had they been accurately and promptly apprised of their patient's progressive paresis.

Judgment notwithstanding the verdict should not be entered unless the evidence, when viewed in the light most favorable to the opponent, so overwhelmingly favors the movant that no contrary verdict based on that evidence could ever stand. See, e.g., Pasquale v. Speed Products Engineering, 166 Ill.2d 337, 351 (1995). We do not believe that the evidence in the case at bar so overwhelming favors defendant that no verdict contrary to defendant could ever stand. Mrs. Holton's treating physicians testified that they based their erroneous diagnosis and treatment decisions upon their belief that their patient's paralysis was of sudden onset rather than having been preceded by approximately 14 to 18 hours of paresis. It is uncontested that neither of the nurses aides who assisted Mrs. Holton into a wheelchair when her legs stopped working during the evening of December 28, 1990, informed the duty nurse or a doctor of the patient's motor impairment. It also appears uncontested that the nurses who were on duty during the hours after the bathroom incident did not report Mrs. Holton's difficulties in moving her legs as a significant change in her condition. It was not until she lost bladder and bowel function along with all ability to move her legs that the nurse on duty reported Mrs. Holton's condition to her doctors. While the parties' expert opinion testimony differs with respect to the consequences flowing from the nursing staffs failure to accurately and promptly apprise the doctors of Mrs. Holton's deteriorating motor ability, the record evidence permits the inference that defendant's negligent acts and omissions prevented her physicians from correctly diagnosing and treating her condition. We conclude that the evidence supports the jury's verdict. See, e.g., Wodziak v. Kash, 278 Ill. App.3d 901 (1996) (holding that evidence was sufficient to establish that defendant's delay in diagnosing the decedent's illness lessened the effectiveness of the treatment and that plaintiff was not required to show in absolute terms that a different outcome would have occurred had the defendant made an earlier diagnosis of the decedent's condition).

We note that the Borowski court's formulation of proximate cause in the context of medical malpractice litigation is the same standard of proximate cause that is used in other types of negligence actions. Although it may appear more difficult to assess exactly what harm negligent medical treatment may have caused when the patient had a preexisting illness or injury, juries routinely are asked to determine whether, and to what extent, a defendant's negligent treatment proximately caused the injury upon which the patient's lawsuit is based. An Illinois Pattern Jury Instruction (IPI) on proximate cause that was given in the case at bar explains that the defendant's negligence need only be a cause of the harm, or "any cause which, in the natural or probable sequence, produced the injury" of the plaintiff, not "the only cause, nor the last or nearest cause. It is sufficient if it concurs with some other cause acting at the same time, which in combination with it, causes the injury." Illinois Pattern Jury Instructions, Civil, No. 15.01 (3d ed. 1995) (hereinafter IPI Civil 3d). Under the Borowski standard, plaintiff met her evidentiary burden of proving the elements of her case. Accordingly, we hold that, based on the evidence of record, defendant was not entitled to judgment as a matter of law, and the trial court did not err in denying defendant's motion for judgment notwithstanding the verdict.

B

Our conclusion that defendant was not entitled to judgment as a matter of law does not directly resolve defendant's contention that the appellate court's reliance on the loss of chance doctrine lessened plaintiffs' burden of proving proximate cause. Because our review of the conflicting appellate cases reveals a significant and ongoing dispute over the application of the loss of chance doctrine in medical malpractice actions, we further consider the issue.

"Lost chance" or "loss of chance" in medical malpractice actions refers to the injury sustained by a plaintiff whose medical providers are alleged to have negligently deprived the plaintiff of a chance to survive or recover from a health problem, or where the malpractice has lessened the effectiveness of treatment or increased the risk of an unfavorable outcome to the plaintiff. Under the traditional formulation of proximate cause, as reflected in Borowski, the plaintiff must prove that defendant's alleged medical malpractice more probably than not caused the claimed injury. Where there is evidence that a plaintiffs estimated chance of surviving or recovering from an existing illness or injury, absent the malpractice, is 50% or less, some courts have concluded that proximate cause under the traditional definition is lacking. In such cases, courts have entered judgments in favor of defendants as a matter of law. See, e.g., Russell v. Subbiah, 149 Ill. App.3d 268 (1986) (entering summary judgment for defendant where evidence showed a 50/50 chance that a doctor's negligence contributed to plaintiff's prolonged recovery); see also Curry v. Summer, 136 Ill. App.3d 468, 476-80 (1985) (noting in dicta that plaintiff must show better than even chance of survival absent alleged malpractice to sustain burden of proof on proximate cause).

Other courts have recognized that victims of medical malpractice should be able to seek damages arising from their doctors' or hospitals' negligent treatment, notwithstanding that the patients' chance of recovering from existing illnesses or injuries may be less than 50%. See, e.g., Chambers v. Rush-Presbyterian-St. Luke's Medical Center, 155 Ill. App.3d 458, 463-65 (1987) (evidence indicating that decedent had only a 33% chance of surviving an undetected cancer even with prompt diagnosis and treatment did not require conclusion that defendant's negligent inducement of brain-damaging coma did not proximately cause patient's death).

Although the legal literature and case law reflect varying perceptions and applications of the loss of chance doctrine, in general two opposing views of the doctrine have emerged — the "relaxed causation" approach and the "separate injury" approach. See 84 Ill. B.J. at 459. In Illinois, most of the controversy stems from a divergence in viewpoint as to whether the loss of chance doctrine relaxes the traditional proximate cause standard in medical malpractice actions (see, e.g., Netto v. Goldenberg, 266 Ill. App.3d 174, 180-81 (1994); Hare v. Foster G. McGaw Hospital, 192 Ill. App.3d 1031, 1038 (1989)) or whether the traditional principles of proximate cause are satisfied by and can be harmonized with the loss of chance concept (see, e.g, Hajian v. Holy Family Hospital, 273 Ill. App.3d 932, 940 (1995); Chambers v. Rush-Presbyterian-St. Luke's Medical Center, 155 Ill. App.3d 458, 463-65 (1987)).

Northern Trust Co. v. Louis A. Weiss Memorial Hospital, 143 Ill. App.3d 479 (1986), was the first Illinois case to specifically identify and approve the loss of chance concept, in the context of deciding whether proximate causation could be established by evidence that the defendant's conduct "lessened the effectiveness" of the treatment or "increased the risk of harm" to a patient. In Northern Trust Co., the defendant hospital was found liable for injuries arising from the medical care given to a newborn infant. Although the individual nurses and doctors involved were found not liable in negligence, the jury's verdict against the hospital was upheld on the ground that the hospital was negligent for its failure to provide a specially trained nurse to supervise the nursery in accordance with board of health regulations.

With respect to the issue of causation, the hospital in Northern Trust Co. claimed that its failure to provide a specially trained nurse was not a proximate cause of the baby's brain damage. However, the court cited evidence which indicated that a trained nurse would have known from the symptoms that the baby's condition was deteriorating and, had notification of such condition been made without the 6 1/2-hour delay that occurred in the case, the baby's brain damage and retardation could have been prevented or reduced. The expert evidence indicated that the delay increased the likelihood of permanent damage. The court reviewed cases from other jurisdictions and examined section 323 of the Restatement (Second) of Torts (1965), which provides that "[o]ne who undertakes * * * to render services to another which he should recognize as necessary for the protection of the other's person or things, is subject to liability to the other for physical harm resulting from his failure to exercise reasonable care to perform his undertaking, if (a) his failure to exercise such care increases the risk of such harm * * *." The Northern Trust Co. court concluded that "the better rule is that '[e]vidence which shows to a reasonable certainty that negligent delay in diagnosis or treatment * * * lessened the effectiveness of treatment is sufficient to establish proximate cause.' " (Emphasis added.) Northern Trust Co., 143 Ill. App.3d at 487.

We believe that the "reasonable certainty" language referenced above conforms to traditional principles of proximate cause. Similarly, in Chambers v. Rush-Presbyterian-St. Luke's Medical Center, 155 Ill. App.3d 458 (1987), the court cited the reasoning in Northern Trust Co. and upheld a medical malpractice verdict for a plaintiff against the hospital's claim that the plaintiff failed to establish that the cause of death was defendant's negligence rather than the decedent's preexisting terminal cancer. In Chambers, the plaintiff asserted that the defendant hospital negligently treated the plaintiff's decedent, causing him to lapse into a coma caused by high blood sugar. The patient suffered brain damage as a result of the coma and received no treatment for a cancer that subsequently was discovered during his autopsy. A contested issue raised in defendant's appeal centered upon the cause of death and whether plaintiff had adequately proved that defendant's negligence proximately caused the patient's death, as opposed to the cancer itself. The Chambers court cited the Borowski decision for the proposition that a plaintiff is not required to prove that a better result would have been obtained absent the malpractice. The court also acknowledged that the circumstances were more problematic when evidence existed that both medical malpractice and an underlying disease or injury caused the patient's death. Nevertheless, the Chambers court rejected defendants' contention that the plaintiff did not sustain her burden of proving proximate cause because of her failure to demonstrate that decedent had more than a 50% chance of surviving the cancer had he not been put into a coma through the negligence of the hospital. Chambers, 155 Ill. App.3d at 463.

We conclude that Northern Trust Co. and Chambers reflect the correct application of proximate causation principles when a defendant's negligent medical care is alleged to have denied the patient a chance of survival or recovery. We note, however, that other appellate decisions have expressly departed from the rationale of Northern Trust Co. and Chambers on the ground that the loss of chance analysis in those decisions altered or eliminated plaintiffs' burden of proving proximate cause. See Hare v. Foster G. McGaw Hospital, 192 Ill. App.3d 1031 (1st Dist. 1989); Netto v. Goldenberg, 266 Ill. App.3d 174, 180-81 (2d Dist. 1994). Another recent appellate decision has, in turn, rejected the Hare and Netto analyses and held that loss of chance theory may be harmonized with the traditional standard of proof set forth in Borowski. See Hajian v. Holy Family Hospital, 273 Ill. App.3d 932 ( 1st. Dist. 1995), citing Pumala v. Sipos, 163 Ill. App.3d 1093 (2d Dist. 1987). We briefly discuss each of these cases.

Hare was a survival and wrongful death action in which the decedent, who suffered from an untreatable condition, had been turned away from defendant hospital. The appellate court affirmed a directed verdict entered in favor of defendant on the ground that plaintiff failed to establish the requisite causal connection between the defendant's allegedly negligent failure to hospitalize plaintiff's decedent and the patient's death from his untreatable fatal illness. Because the evidence revealed that no medical treatment was available for the decedent's fatal illness, the Hare court determined that plaintiff had not established that the defendant's failure to hospitalize the decedent "more probably than not" was a cause of his death.

Instead of simply holding that the directed verdict was proper because of insufficient evidence of proximate cause, the Hare court went on to discuss the development of the loss of chance doctrine and the conflicting interpretations of what standard of proof is required to establish proximate cause. Apparently believing that the Borowski standard of proximate cause would be undermined by recognition of the loss of chance concept, the Hare court expressly declined to follow the lead of such cases as Northern Trust Co. and Chambers, stating, "[A]ny alteration in the burden of proof regarding proximate cause as was done in [ Hamil v. Bashline, 481 Pa. 256, 392 A.2d 1280 (1978)] would have to come from the supreme court." Hare, 192 Ill. App.3d at 1038.

The Second District of the Appellate Court, in Netto, adopted the Hare court's view that the loss of chance doctrine lessened plaintiffs' burden of proving proximate cause and further expressed its belief that "the Northern Trust Co. and Chambers courts removed the proximate cause element from medical negligence actions." (Emphasis added.) Netto, 266 Ill. App.3d at 181. The Netto court's disapproval of the loss of chance doctrine occurred in the context of reviewing the plaintiff's non-IPI instruction purporting to state a loss of chance theory. However, the Netto court did not hold that the plaintiff's evidence of proximate cause was inadequate and, in fact, reversed the judgment entered upon the jury verdict in favor of defendant and remanded for a new trial because of error.

In Pumala v. Sipos, 163 Ill. App.3d 1093 (1987), another panel of the Second District observed that a plaintiff need not prove that a better result would have occurred absent the alleged malpractice, consistent with Borowski, but also emphasized that a plaintiff, to establish proximate cause, must show with a reasonable degree of medical certainty that the negligent delay in diagnosis or treatment lessened the effectiveness of the medical services rendered to the plaintiff. Pumala, 163 Ill. App.3d at 1098, citing Northern Trust Co, 143 Ill. App.3d 479. The Pumala court affirmed the trial court's order directing a verdict in favor of the defendant in that case, however, because the evidence did not establish the requisite nexus between the alleged malpractice and plaintiff's injury.

Finally, in Hajian v. Holy Family Hospital, 273 Ill. App.3d 932 (1995), a panel of the First District of the Appellate Court engaged in a thorough analysis of the loss of chance doctrine and followed the reasoning of Pumala. The court in Hajian stated, "We do not agree with [ Netto's] interpretation of Northern Trust and its progeny, but rather, would resolve the conflict by harmonizing the language of Northern Trust and Borowski as was attempted in Pumala [ v. Sipos, 163 Ill. App.3d 1093 (1987)]." Hajian, 273 Ill. App.3d at 940.

As the above examination of the case law reveals, individual decisions may be properly decided upon their facts yet still conflict with other decisions in their analysis of the loss of chance doctrine. The reasoning of the appellate panels in Pumala and Hajian is consistent with Northern Trust Co. and Chambers and, in our view, reflects the correct understanding of the loss of chance concept. Conversely, decisions such as Hare and Netto misleadingly suggest that loss of chance analysis ipso facto results in a lowered standard of plaintiff's proof of causation. Although the specific dispositions of the appeals in Hare and Netto may have been justified, the loss of chance analysis contained in the decisions has perpetuated confusion as to the proper standard of proof in medical malpractice cases. Accordingly, we overrule those cases to the extent that their analysis conflicts with our holding in the instant case.

There is nothing novel about requiring health care professionals to compensate patients who are negligently injured while in their care. To the extent a plaintiff's chance of recovery or survival is lessened by the malpractice, he or she should be able to present evidence to a jury that the defendant's malpractice, to a reasonable degree of medical certainty, proximately caused the increased risk of harm or lost chance of recovery. We therefore reject the reasoning of cases which hold, as a matter of law, that plaintiffs may not recover for medical malpractice injuries if they are unable to prove that they would have enjoyed a greater than 50% chance of survival or recovery absent the alleged malpractice of the defendant. E.g., Curry, 136 Ill. App.3d 468; Russell, 149 Ill. App.3d 268. See Gatlin v. Ruder, 137 Ill.2d 284, 292-93 (1990) (specifically disapproving the Russell court's entry of summary judgment upon evidence that the patient's chance of survival absent the malpractice was "50/50"). To hold otherwise would free health care providers from legal responsibility for even the grossest acts of negligence, as long as the patient upon whom the malpractice was performed already suffered an illness or injury that could be quantified by experts as affording that patient less than a 50% chance of recovering his or her health.

Disallowing tort recovery in medical malpractice actions on the theory that a patient was already too ill to survive or recover may operate as a disincentive on the part of health care providers to administer quality medical care to critically ill or injured patients. Moreover, it has been noted that "[i]t is impossible to divine who would fall into one category [survivor] or the other [nonsurvivor]. Not allowing such a case to be decided by a jury means that statistical proof of a less than 50% chance would be dispositive, even though no expert in the world could prospectively state who would survive and who would die. That is why doctors treat all patients, not just those with better than even odds." 84 Ill. B.J. at 462.

We hold that the loss of chance concept, when properly analyzed, does not relax or lower plaintiffs' burden of proving causation. Rather, the concept comports with the Borowski standard.

II. Trial Error

Although we have held that defendant was not entitled to judgment notwithstanding the verdict, we conclude that defendant is entitled to a new trial because of certain errors which had a strong probability of prejudicing the jury. Specifically, we determine that the trial judge's admonition to the jury that one of the defense witnesses gave inaccurate testimony on a collateral issue and was encouraged to do so by defendant's attorneys, coupled with plaintiffs' counsel's repeated charges of defendant's fraudulent misconduct, constituted reversible error under the circumstances of this case.

To explain why these errors require the granting of a new trial it is necessary to consider the factual context in which the errors occurred. During trial, plaintiffs' counsel attempted to demonstrate that defendant and its attorneys had coached key witnesses for defendant to change their deposition testimony to better fit the defense theory. By vigorous cross-examination and through repeated remarks during closing argument, plaintiffs' counsel vehemently argued the possibility that the hospital had engaged in an attempt to cover up the nursing staff's negligence by falsifying testimony. Additionally, because of an incident that occurred just after Dr. Jergens' testimony in defendant's case, plaintiffs' counsel apparently became convinced that defendant's attorneys had encouraged Dr. Jergens to give false testimony with regard to whether the doctor was a defendant in a companion lawsuit filed by plaintiffs. Although the issue of Dr. Jergens' status as a defendant in a different case was collateral to the trial issues, plaintiffs' counsel subsequently asserted to the trial judge in conference that Dr. Jergens had committed perjury, and that defendant's counsel had aided and abetted such perjury.

The following facts taken from the record are relevant to this issue. Dr. Jergens, the emergency room physician on duty at the time of Mrs. Holton's admission, was asked at the end of his cross-examination the following question by plaintiffs' counsel, Bruce Cook:

"Q. You understand, do you not, sir, that you are a defendant in a companion case to this one? Are you aware of that?

A. I'm not aware of that."

At the time of this question, Dr. Jergens had not been formally served with summons and a copy of the complaint in a companion lawsuit that plaintiffs had filed approximately a year before the trial in the instant case. In fact, according to a discussion in the transcript it appears that at the beginning of trial, Cook had told defendant's lead trial counsel, Sandberg, that he believed the other lawsuit had been dismissed. Moreover, as is clear from the attorneys' arguments to the judge outside the presence of the jury, the two attorneys disagreed as to whether a person is properly considered a party defendant if he or she has been named in a complaint but has never been served with process. Sandberg maintained that Dr. Jergens was not a party in the companion lawsuit until he was served with process. More importantly, Sandberg argued that the doctor's potential status as a defendant in a separate lawsuit was a collateral issue of no relevance to the pending trial. In contrast, plaintiffs' counsel took the position that if Dr. Jergens was aware of the existence of the other lawsuit, he did not answer truthfully when questioned during cross-examination as to his awareness of being a defendant in that suit. Over objection, and in the jury's presence, the court took judicial notice of Dr. Jergens' party status and allowed plaintiffs' counsel to inform the jury, "Mark Jergens, who just testified, is a party defendant in a pending lawsuit where my clients have filed suit against him."

If that had been the end of the matter, no real prejudice to defendant would have resulted because Dr. Jergens' status as a party defendant in another lawsuit would have been relevant only to show a potential bias in his testimony, which is an appropriate use of cross-examination. However, instead of treating the dispute over Dr. Jergens' party status as a difference of legal opinion, plaintiffs' attorney chose to treat the doctor's answer as perjury and to accuse opposing counsel of suborning or inducing such perjury. In support of these serious charges, Cook informed the court that after Dr. Jergens left the courtroom following his testimony, the doctor fled when an employee of Cook's law firm attempted to serve him with the complaint and summons in the separate suit. Although service was effected, plaintiffs' counsel demanded to have the witness brought back for further questioning, claiming that the incident proved that Dr. Jergens knowingly deceived the court when he denied awareness of being a defendant in the other lawsuit. According to Cook, defendant's attorneys must have warned Dr. Jergens to evade service, which indicated that defendant's attorneys "again" were engaged in "inducing" a witness to "tell a lie under oath."

Sandberg acknowledged that Dr. Jergens was aware of the existence of the separate lawsuit but denied that he had instructed Dr. Jergens how to answer the question because the issue had "nothing to do with this lawsuit." Sandberg also told the court that when he saw the papers for the lawsuit lying on counsel table he informed Dr. Jergens of the likelihood that plaintiffs' counsel would attempt to effect service. However, Sandberg reiterated his belief that Dr. Jergens, who had not been served with the separate suit at the time he testified, honestly stated his understanding of his party status and did not give a false or inaccurate response to Cook's question.

When Dr. Jergens returned to court the same afternoon in response to the trial judge's order, he was questioned at some length outside the presence of the jury. Dr. Jergens testified that he knew of the existence of the companion lawsuit. He had formerly worked at Memorial Hospital and had been deposed in the pending lawsuit. It was his understanding, based on information from "many sources," including defendant's attorneys, that he was not a party defendant in the companion suit until formal service of summons was effected. Dr. Jergens, who is not an Illinois resident, explained that he ran from plaintiffs' process server as he was leaving court because he did not want to be served with the lawsuit. He had asked Kristen Hines, the hospital's in-house counsel, what to do to avoid service and she told him to leave quickly. Dr. Jergens denied that either Hines or Sandberg had told him how to answer if Cook inquired upon cross-examination whether he was a defendant in the other lawsuit.

During his questioning of Dr. Jergens outside the presence of the jury, Cook argued with the witness over the honesty of his belief that he was not a party in the separate suit, and outright accused Dr. Jergens of lying to the jury at trial. Then, when Sandberg remarked that the questioning was getting a "bit far afield," Cook replied, "You're about ready to go in the slammer here in a second * * *." He referred to Hines as Sandberg's "henchman" who advised the witness to flee. Continuously during the hearing, Cook made remarks questioning the ethics of opposing counsel and impugning their integrity and honesty.

Despite his hostile characterization of the situation, Cook offered the trial court no authority to support his contention that Dr. Jergens' testimony regarding his status as a party was legally incorrect. Nor did he explain in what manner Dr. Jergens' answer was factually untruthful. Nonetheless, he moved the court to hold Dr. Jergens and the two attorneys for the hospital in contempt. He stated that defendant's lawyers, Hines and Sandberg, should be reported to the Attorney Registration and Disciplinary Commission. Finally, Cook orally moved the court to strike the hospital's pleadings, which would have the effect of defaulting defendant. He told the court, "As a practical matter, Judge, you probably should issue a directed verdict in my behalf anyway on the issue of negligence. We'll save a lot of time. The doctor won't have to spend another day. Ms. Hines and Mr. Sandberg rather than having to be jailed will * * * get their names in the books in the Appellate Court when they appeal it."

Although the trial judge did not impose the severe sanction of striking the hospital's answer, the court concluded that Sandberg had been untruthful when he denied discussing the companion suit with Dr. Jergens, and further concluded that the veracity of Dr. Jergens on this collateral matter of his party status was "important, very, very important." Consequently, the trial judge read the following statement to the jury at the next session of court, just before closing statements:

"Yesterday afternoon out of your presence, we conducted a hearing concerning Dr. Mark Jergens' testimony. I determined at the hearing that the doctor's testimony concerning his knowledge of a lawsuit pending against him was not true. I further determined that Ms. Hines and Mr. Sandberg knew the statement was false and had done certain things that encouraged the doctor to believe his answer was accurate. You may consider this fact in determining the credibility of Dr. Jergens' testimony."

In our view, the controversy over Dr. Jergens' party status in a separate lawsuit was a collateral issue of limited if any relevance to the instant case. This court has noted that if a question concerns a collateral matter the cross-examining attorney must accept the answer given and is not allowed to attempt impeachment on the issue. See Esser v. McIntyre, 169 Ill.2d 292, 305 (1996); see also Tzystuck v. Chicago Transit Authority, 124 Ill.2d 226, 242 (1988) (collateral evidence should be excluded to prevent the jury from becoming distracted from the main issues). Moreover, because the trial court took judicial notice of Dr. Jergens' party status, plaintiffs' counsel was able to inform the jury of the point he wished to make through Dr. Jergens' testimony, i.e., that the doctor was a party defendant in the collateral lawsuit. Therefore, we conclude that the trial court should not have entertained Cook's request to hold a mini-trial into the side issue of whether Dr. Jergens' party status answer was complete and accurate.

Of greater concern are Cook's groundless accusations of perjury and subornation of perjury in connection with this matter. The record does not bear out such charges. Dr. Jergens told the court it was his honest belief that he was not a defendant in the other suit at the time of his trial testimony because he had not been served with process. Opinions on points of law, even if incorrect, are not perjurious. Similarly, a person's honest understanding of his legal status cannot be viewed as constituting an intentional misrepresentation of fact. Therefore, Cook's charge of perjury should have been rejected.

We also question the basis upon which Cook insisted that Sandberg lied to the court when he initially denied having discussed the separate suit with Dr. Jergens. The transcript indicates that Sandberg subsequently corrected himself by acknowledging that after seeing the papers for the separate lawsuit lying on counsel table, he informed the doctor of his belief that plaintiffs planned to serve him in court. Sandberg admitted discussing the legalities of service of process, but denied that he had specifically told Dr. Jergens how to answer any questions at trial concerning Dr. Jergens' party status. Indeed, Sandberg stated that he did not believe the issue was relevant. Dr. Jergens, in his testimony, corroborated Sandberg's representations that the subject of how to answer questions regarding Dr. Jergens' party status was not discussed. Dr. Jergens also acknowledged having talked to Sandberg and others in the past regarding the legal significance of service of process.

We conclude that the record contains no sound basis for finding that Sandberg intentionally misled the court. Even if Sandberg misspoke in his initial denial of having discussed the separate lawsuit with Dr. Jergens, it is clear from the context that he was attempting to respond to Cook's specific accusation that he had coached the witness on how to answer the party status question. We conclude that Cook's motion for sanctions based on perjury or subornation of perjury was not supported by the record and was meritless.

We hold that the trial court abused its discretion in bringing to the jury's attention its personal belief that Dr. Jergens had testified falsely and had been led to do so by defendant's attorneys. This court has stated that a trial judge should refrain from conveying to the jury his or her opinions on ultimate matters of fact or the credibility of witnesses and the weight to be given their testimony. See People v. Santucci, 24 Ill.2d 93, 98 (1962). In the case at bar, the prejudicial impact of the trial court's statement upon the jury can hardly be overstated. The jury was not apprised of the legal basis upon which Dr. Jergens' opinion of his party status rested, nor was the jury told of the circumstances under which the charge of attorney misconduct was made. Instead, the trial court told the jury, as proven fact, that Dr. Jergens had, in effect, lied at the behest of defendant's attorneys and that the jury could consider that "fact" in determining the witness' credibility. This undoubtedly had a devastating impact on the jury's perception of defendant, its lawyers, and its witnesses. By its unwarranted remarks to the jury, the trial court placed its neutrality in issue. Therefore, we hold that defendant was denied a fair trial and is entitled to a new trial. See, e.g., Forest Preserve District v. Wike, 3 Ill.2d 49, 57 (1954) (judge should not make comments reflecting on the integrity of counsel in the presence of the jury); see generally A. Hartman, The "Whys" and "Whynots" of Judicial Comments on Evidence in Jury Trials, 23 Loy. U. Chi. L.J. 1, 19-21 (1991).

We note further that the prejudicial impact of the trial judge's remarks was heightened by the immediately following closing argument of Cook, who riddled his summation with references to coached and deceitful hospital witnesses and manipulative attorneys. For example, Cook claimed that the rules of trial had been "shamelessly ignored" by the defense, including the rule that attorneys should not "counsel or assist a witness to testify falsely." Cook commented that the nurses and nurses aides were "nice people" who had been encouraged to "modify their testimony." According to Cook, his was the profession of Abraham Lincoln and Daniel Webster and the Declaration of Independence, while the defense attorneys' profession was that of "John Dean, John Erlichmann, people who were so interested in winning that they violated the rules." Cook told the jury that the probable reason for the "lawyer misconduct in this case" was that Mrs. Holton sustained terrible damages. He further informed the jury that it had been the victim of "distortion of the truth" by "the named partner in a large St. Louis law firm."

We believe that these remarks were of a type likely to arouse passion and prejudice in the jury, particularly in light of the trial judge's comments undermining Dr. Jergens' credibility and the lawyers' integrity. Therefore, the prejudicial remarks contributed to the trial court's error and combined with that error to deny defendant a fair trial under the circumstances.

It is true that where there is record evidence in support of a claim that opposing counsel or parties falsified evidence or encouraged witnesses to change their trial testimony, counsel may fairly comment upon such evidence. If a witness' trial testimony significantly differs from his or her deposition testimony, opposing counsel may exploit such changes by traditional means of impeachment. However, modifications or additions to a witness' trial testimony which were not expressly stated in that witness' pretrial deposition do not, in and of themselves, suggest deliberate distortion of evidence by the witness or fraudulent coaching by lawyers. In the case at bar, plaintiffs' counsel believed that he had obtained damaging admissions from certain witnesses that defendant or its agents had encouraged them to change their testimony with respect to Mrs. Holton's condition in the hours preceding her paralysis. Because we are remanding this cause for new trial we make no express finding regarding whether there may have existed a sufficient evidentiary basis to support some of Cook's remarks indicating that defense counsel coached the nursing staff witnesses to lie and change their testimony. Although the appellate court in the instant case observed that there was evidence sufficient to raise such an inference, the court further stated that the evidence did not "conclusively" establish that hospital employees were manipulated or told to testify falsely. We hold only that the trial in the instant case was fatally tainted by the improper comments of the trial judge to the jury, coupled with closing remarks of an unduly prejudicial nature. The evidence of the hospital's liability was not so overwhelming that we can characterize the trial error as harmless under the circumstances. Nor can we conclude that the jury's verdict was unlikely to have been influenced by the trial court's comments upon the integrity of defense counsel and the credibility of one of its witnesses, especially in light of Cook's repeated charges of dishonesty and attorney misconduct.

III. Jury Instructions

Defendant also challenges certain jury instructions. First, defendant contends that plaintiffs' instruction No. 13, a non-IPI instruction, was misleading. This instruction stated that defendant "owed a duty, independent of any relationship between physician and patient, to review and supervise the medical care administered to the plaintiff." According to defendant, this instruction could be understood as making the hospital liable not only for the acts and omissions of the nursing staff and emergency room personnel, but also for any negligence on the part of Dr. Doubek and the other private physicians who treated Mrs. Holton. According to defendant, because there was no evidence that the hospital knew or should have known that plaintiffs' private treating physicians had misdiagnosed Mrs. Holton's condition or rendered ineffective treatment, defendant was not vicariously liable for the acts or omissions of plaintiffs' physicians. Furthermore, defendant argues that the potential jury confusion over the scope of the hospital's vicarious liability was demonstrated by the jury's written inquiry to the court, asking whether "a staff member, i.e., Dr. Doubek, [is] an officer or employee of the hospital."

Plaintiffs concede that instruction No. 13 could have been drafted more narrowly to prevent the jury from speculating whether any physicians other than Dr. Jergens, the emergency room physician on call, were included in the group of defendant's employees whose acts or omissions were legally attributable to defendant. However, plaintiffs assert that the instruction correctly states the applicable law, that a hospital has an independent duty to its patients to review and supervise treatment, and therefore any error in the instruction was harmless, particularly in view of the "overwhelming" evidence against defendant.

In light of our decision to reverse and remand this cause for a new trial we need only note that the instruction should not be resubmitted in its presently drafted form. We believe that the jury's question to the court reflects potential confusion over whether private physicians are considered "staff members" for purposes of holding the hospital legally responsible for acts and omissions of the private doctors. Accordingly, on remand any instruction attempting to state the hospital's duty to supervise the medical care given its patients should be clearly and accurately drafted to avoid jury speculation or confusion.

Defendant next argues that the statement of law in plaintiffs' instruction No. 24 does not apply to the instant case and should not have been given. Instruction No. 24 stated, "If a health care provider is guilty of professional negligence which creates a condition of the plaintiff's body, then the health care provider is liable not only for plaintiffs damages resulting from that condition, but also liable for any damages sustained by the plaintiff arising from the efforts of subsequent health care providers to treat the condition caused by the initial health care provider." According to defendant, this instruction is applicable only in cases where one tortfeasor "creates a condition" in the plaintiff and subsequent medical providers improperly treat the condition, making it worse. See Gertz v. Campbell, 55 Ill.2d 84 (1973) (negligent driver sued for plaintiff's injuries sustained in collision and also for subsequent aggravation of injuries from medical malpractice). Defendant states that it did not create the condition of osteomyelitis in Mrs. Holton and therefore instruction No. 24 does not apply to any issue in the case.

Plaintiffs relied on the case of Daly v. Carmean, 210 Ill. App.3d 19 (1991), as authority for submitting this instruction. In Daly, the appellate court held that an identically worded instruction was appropriate in the case before it, which involved a podiatrist who had negligently performed foot surgery upon plaintiff and caused numerous complications needing subsequent correction. Daly held that although the instruction properly stated the doctor's liability for aggravating the plaintiff's initial condition, the trial court did not abuse its discretion in refusing the instruction. In the case at bar, plaintiffs do not explain in what manner Daly assists their contention that instruction No. 24 should have been given. Instead, plaintiffs simply argue that the instruction is a correct statement of the law "where subsequent medical treatment may have caused damage to plaintiff."

We believe that the circumstances of the case at bar are distinguishable from those of cases such as Daly and Gertz, where an initial tortfeasor was legally responsible for subsequent aggravation of the original injury caused by the initial tortfeasor. The case at bar would be analogous to Gertz and Daly if, for example, some individual had negligently injured Mrs. Holton's spine, causing trauma, which resulted in the osteomyelitis that was subsequently misdiagnosed and incorrectly treated. In such an example, the original tortfeasor presumably could be found liable for the damages caused not only by the initial injury to the spine but also for the subsequent failure on the part of the health care providers to detect and correct the condition. In the instant case, however, the facts do not appear to support the giving of an instruction involving the aggravation of an original tortfeasor's injury by subsequent medical malpractice. Therefore, we hold that instruction No. 24 should not be given upon retrial.

Defendant's remaining challenge to the jury instructions is that the proximate cause instructions were one-sided, favoring plaintiffs. According to defendant, the trial court erred in rejecting one of the optional provisions of an IPI instruction, which would have allowed the jury to enter verdict in favor of defendant if the jury found that plaintiffs' injuries were solely caused by the "conduct of some person other than defendant."

In instruction No. 15, the jury was given the long form proximate cause instruction (IPI Civil 3d No. 15.01), which defines proximate cause as "any cause which, in natural or probable sequence, produced the injury complained of. It need not be the only cause, nor the last or nearest cause. It is sufficient if it concurs with some other cause acting at the same time, which in combination with it, causes the injury." The jury was also given IPI Civil 3d No. 12.05, submitted by defendant, which states that if the jury finds the defendant's negligence was a proximate cause of plaintiff's injury it is not a defense that "something else may also have been a cause of the injury." The second paragraph of this instruction instructs that the verdict should be for defendant if the jury decides "that the sole proximate cause of injury to the plaintiff was something other than the conduct of the defendant." (Emphasis added.) IPI Civil 3d No. 12.05.

Defendant's specific challenge is to the omission from another instruction, instruction No. 14, of a sole proximate cause provision based on conduct of third parties. As given to the jury, instruction No. 14 stated that "more than one person may be to blame for causing an injury. If you decide that the defendant was negligent and that its negligence was a proximate cause of injury to the plaintiff, it is not a defense that some third person who is not a party to the suit may also have been to blame." See IPI Civil 3d No. 12.04. According to defendant, the second paragraph of IPI Civil 3d No. 12.04 should have been included, as follows: "However, if you decide that the sole proximate cause of injury to the plaintiff was the conduct of some person other than the defendant, then your verdict should be for the defendant." Defendant argues that the inclusion of this provision would have properly allowed the jury to find that the conduct of Dr. Doubek, Dr. Murphy, or Dr. Sprich in failing to properly diagnose and treat Mrs. Holton, was the sole proximate cause of her injury.

We reject defendant's contention that the trial court abused its discretion in declining to include the above-quoted provision in instruction No. 14. A defendant is not automatically entitled to a sole proximate cause instruction wherever there is evidence that there may have been more than one, or concurrent, causes of an injury or where more than one person may have been negligent. Instead, a sole proximate cause instruction is not appropriate unless there is evidence that the sole proximate cause (not "a" proximate cause) of a plaintiff's injury is conduct of another person or condition. See Ballweg v. City of Springfield, 114 Ill.2d 107, 121 (1986); cf. Leonardi v. Loyola University, 168 Ill.2d 83 (1995). The usage notes following IPI Civil 3d No. 12.04 caution that the sole proximate cause provision "should be used only where there is evidence tending to show that the sole proximate cause of the occurrence was the conduct of a third person."

In the case at bar, defendant did not present evidence or argue that it was only the negligence of persons other than the hospital employees which proximately caused plaintiffs' injury. Instead, defendant attempted to establish that no medical negligence had occurred at all. Defendant did not charge that plaintiffs' treating physicians were negligent in their acts or omissions. On the contrary, much of the defense relied on the rationale that the treating physicians' diagnosis and treatment decisions were proper in light of the circumstances in which the decisions were made. For example, there was evidence that a cancerous tumor appeared to be the most likely diagnosis based on the information upon which the treating doctors based their decisions. The jury, however, found in favor of plaintiffs, under whose theory of the case defendant's negligence proximately caused the treating physicians' misdiagnosis. Because neither plaintiffs nor defendant asserted at trial that the treating physicians themselves were negligent, we conclude that the trial court did not err in denying defendant's request for a sole proximate cause instruction based on the negligence of third parties.

IV

Finally, defendant claims that it was entitled to a setoff against the damages award for expenses paid by plaintiffs' insurance company. Plaintiffs counter that this court's resolution of the issue would be premature because the issue of the insurer's subrogation rights has not yet been adjudicated. We decline to address this issue in light of our reversal of the cause for new trial because there is no longer a judgment against which to set off the expenses paid by the insurance company. The parties may raise any issues regarding insurance reimbursement or subrogation rights at the appropriate time in the trial court.

For the foregoing reasons, we agree with the denial of defendant's motion for judgment notwithstanding the verdict and hold that plaintiffs sustained their burden of proving proximate cause pursuant to the Borowski standard. We reverse the judgment of the appellate court and the judgment of the circuit court entered upon jury verdicts in favor of plaintiffs and remand for a new trial because defendant was denied a fair trial.

Appellate court judgment reversed; circuit court judgment reversed; cause remanded.