77 F.3d 485 | 7th Cir. | 1996
NOTICE: Seventh Circuit Rule 53(b)(2) states unpublished orders shall not be cited or used as precedent except to support a claim of res judicata, collateral estoppel or law of the case in any federal court within the circuit.
WISCONSIN HEALTH CARE LIABILITY INSURANCE PLAN and Wisconsin
Patients Compensation Fund, Plaintiffs-Appellants,
v.
OHMEDA, a division of The Boc Group, Inc., and ABC Insurance
Company, Defendants-Appellees.
No. 94-3542.
Seventh Circuit.
Argued March 30, 1995.
Decided Feb. 15, 1996.
Before FLAUM, Circuit Judge, RIPPLE, Circuit Judge, and ROVNER, Circuit Judge.
ORDER
A jury found Ohmeda negligent for the failure to warn a Wisconsin hospital about the dangers in using an older model of its anesthesia machine and to apprise the hospital of an available device that made the machine simpler and thus more safe to use. The district court granted Ohmeda's motion for judgment as a matter of law pursuant to Fed.R.Civ.P. 50(b), and that ruling is challenged on appeal. We affirm.
I.
This tragic case stems from events that took place on August 11, 1988 at St. Luke's Hospital in Racine, Wisconsin. At approximately 12:30 a.m., nurse-anesthetist Nancy Myers was summoned to assist in the delivery of patient Wendy Olsen's twins. Having been on duty since 7:00 a.m. the previous day with almost no rest, Myers was tired and feeling stress. The first of Olsen's twins was delivered without incident, but complications in the delivery of the second child prompted the attending physician to order Myers to "put the patient to sleep" immediately.
The anesthesia machine present in the delivery room was an Ohmeda Model 2000 that St. Luke's had purchased in 1966. The machine was set up to deliver the anesthetic agent Forane via two carrier gases, oxygen and nitrous oxide. The anesthetic agent is introduced into these gases by means of a "Vernitrol" vaporizer, which bubbles the oxygen through liquid Forane. The Forane-laden oxygen is then delivered to the patient along with nitrous oxide through an endotracheal tube. The flows of oxygen and nitrous oxide, and the percentage of Forane introduced into the carrier gases, are controlled by a series of knobs and corresponding flowmeters. Two knobs control the volume of oxygen being delivered, for example, one controlling the "course" flow and the other controlling the "fine" flow; as these knobs are adjusted, the two corresponding flowmeters (one for each control knob) indicate the rate at which oxygen is being delivered. The same is true for nitrous oxide. A third set of two knobs and two flowmeters controls the Vernitrol vaporizer and establishes the amount of Forane that is mixed with the carrier gases. Determining the correct mixture is a somewhat complex calculation dependent upon the particular anesthetic agent being used, the volume of carrier gases being delivered, the room temperature, and the percentage of anesthetic agent desired. A circular slide rule supplied by the manufacturer enables the operator to arrive at the proper Vernitrol setting.
Although Myers routinely checked the anesthesia machine to make sure that it was prepared for use and had previously used it to administer oxygen and nitrous oxide, she had never before used this particular machine to administer general anesthesia. After making the necessary preparations for the delivery of anesthesia to Olsen, Myers properly set the flows of oxygen and nitrous oxide at three liters each, producing a carrier gas mixture that was 50 percent oxygen and 50 percent nitrous oxide. However, as she then turned to the Vernitrol vaporizer, she did not use the slide rule to calculate the appropriate setting. Instead she employed a rule of thumb which dictated that the lowest volume setting of a "1 followed by 0's" would yield a mixture containing the desired one percent of Forane. Myers reached for the first of the two knobs controlling the vaporizer (which controlled the course flow) and turned the dial until the flowmeter read 1000 cc's. In fact, this was ten times greater than the correct setting of 100 cc's for a 1 percent Forane mixture. Myers should have used the other of the two knobs, controlling the fine flow, to select a setting of 100 cc's. She testified at trial that she was unaware the machine had two flowmeters for the Vernitrol and believed the second flowmeter was hooked up to another vaporizer.
As a result of the mistaken setting, Olsen was administered an overdose of Forane and suffered permanent brain damage. Neither of the twins was physically harmed.
Olsen filed suit against Myers and the hospital in Wisconsin state court. That suit concluded in April 1990 with a settlement funded by the Wisconsin Health Care Liability Plan ("WHCLIP"), which insured both Myers and the hospital, and the Wisconsin Patients Compensation Fund ("WPCF"), a legislatively created fund paying claims in excess of WHCLIP's coverage. WHCLIP and WPCF then filed this diversity suit against Ohmeda, alleging that the manufacturer was both strictly liable for the manufacture of a defective and unreasonably dangerous product and negligent in its failure to warn the hospital of a defect that it discovered after the sale of the machine and for which a corrective safety device was available. A jury found in favor of Ohmeda on the product liability claim at the liability phase of the trial but against Ohmeda on the post-sale negligent failure to warn claim. The jury also found Myers and St. Luke's negligent, but Ohmeda was deemed the most culpable, the jury finding it 55 percent responsible for the Olsen's injuries, St. Luke's 35 percent responsible, and Myers only 10 percent.
After the liability phase of the trial, Ohmeda renewed its motion for judgment as a matter of law pursuant to Fed.R.Civ.P. 50(b) and in the alternative asked for a new trial pursuant to Fed.R.Civ.P. 59. The district court granted the motion for judgment as a matter of law. Extrapolating four prerequisites to liability from the leading Wisconsin case on a manufacturer's post-sale duty to warn customers of defects, Kozlowski v. John E. Smith's Sons Co., 275 N.W.2d 915 (Wis.1979), the court found the proof insufficient to meet three of them. The court found no evidence that Ohmeda had become aware of any defect in the Model 2000, no evidence that a safety device had been developed in response to such a defect, and no evidence that St. Luke's as a user of the Model 2000 required notice from Ohmeda of the availability of a device that, in the plaintiffs' view, would have prevented the anesthetic overdose administered to Ms. Olsen. Finally, the court concluded in the alternative that even if the evidence were sufficient to sustain the plaintiffs' post-sale failure to warn claim, Ohmeda was entitled to a new trial. Olsen v. Ohmeda, A Div. of the Boc Group, Inc., 863 F.Supp. 870 (E.D.Wis.1994).
From this order, WHCLIP and WPCF appeal.
II.
We review de novo the district court's determination that Ohmeda was entitled to judgment as a matter of law. Williams v. O'Leary, 55 F.3d 320, 323 (7th Cir.), cert. denied sub nom. Williams v. Brewer, 116 S.Ct. 527 (1995); Jackson v. Bunge Corp., 40 F.3d 239, 246 (7th Cir.1994). Although Wisconsin law of course defines the substance of the plaintiffs' claims in this diversity case, it is federal law that supplies the standard for purposes of Rule 50(a). Leon v. Caterpillar Indus., Inc., 69 F.3d 1326, 1333 (7th Cir.1995); LaFollette v. Savage, 63 F.3d 540, 543 (7th Cir.1995); Mayer v. Gary Partners & Co., 29 F.3d 330, 335 (7th Cir.1994). We will thus uphold the grant of judgment as a matter of law so long as reasonable persons could not have found that the plaintiffs satisfied their burden to prove each essential element of their claim. E.g., Leon, 69 F.3d at 1333. Of course, at this juncture, we must consider the evidence in the light most favorable to the plaintiffs. See LaFollette, 63 F.3d at 544.
Kozlowski establishes the contours of the post-sale failure to warn claim. In that case, a worker in a meat packing plant was killed in the process of cleaning a sausage stuffing machine. The worker had removed the cover of the machine for cleaning, which left only an undersized safety ring to restrain the free-floating piston inside. Air pressure was used in order to raise the piston so that it might be wiped down. The pressure applied in this case was excessive, and the piston shattered the safety ring. Flying fragments of the broken ring breached cooling pipes above the machine, flooding the area with ammonia and asphyxiating the worker. The evidence revealed that the manufacturer of the machine had become aware at some point after initial sales of the machine of the dangers attending excessive pressurization of an uncovered machine, and had developed a by-pass valve to relieve excess pressure some twenty-five years earlier. Citing the possibility that the safety ring might shatter as a defective condition, the worker's widow sought to hold the manufacturer liable for failing to warn the packing company of that condition and for failing to apprise the company of the safety valve developed to alleviate the risk. At the close of the plaintiff's case, the trial court directed a verdict in favor of the manufacturer. The Wisconsin Supreme Court reversed, finding the evidence sufficient to present a jury question as to whether the manufacturer had given the worker's employer adequate notice of the purported defective condition and the safety valve. 275 N.W.2d at 923. The court took care to emphasize that it was not recognizing "an absolute continuing duty, year after year, for all manufacturers to warn of a new safety device which eliminates potential hazards." Id. In this case, the defendant did not produce consumer goods that might wind up in countless households; on the contrary, "the sausage stuffer machine industry is far more limited in scope." Id. at 924. "[A] jury in determining a manufacturer's duty in this restricted area must look to the nature of the industry, warnings given, the intended life of the machine, safety improvements, the number of units sold and reasonable marketing practices, combined with the consumer expectations inherent therein." Id.; see also Gracyalny v. Westinghouse Elec. Corp., 723 F.2d 1311, 1318-20 (7th Cir.1983).
The district court derived from Kozlowski four prerequisites to liability for the post-sale failure to warn: (1) a relatively limited industry and market, in which it is feasible for the manufacturer to know who has purchased its product and to communicate safety improvements to them; (2) proof that the manufacturer learned, or should have learned, about a defect in the product after it was sold; (3) development of a safety device by the manufacturer in response to the defect; and (4) the manufacturer's failure to notify its customers of the defect and of availability of the safety device. 863 F.Supp. at 874-75. Although the court agreed that the market for anesthesia machines was sufficiently limited to support a post-sale duty to warn customers of defects learned of later (id. at 875), it concluded that the plaintiffs had failed to satisfy any of the three remaining prerequisites to liability.
First, the court found the evidence insufficient to establish a defect in the Ohmeda 2000. There was no proof that any other incident like the overdose of Olsen had ever occurred before. The plaintiffs' theory, instead, seemed to be that as new models with enhanced safety features were put into use, operators might become reliant on such features and at the same time increasingly unfamiliar with the precautions one needed to take with older models like the Model 2000--for example, the need to calculate the flow of gas through the Vernitrol vaporizer, and the need not to confuse the parallel flowmeters. This would be exacerbated as hospitals shunted the older models into secondary locations like obstetric suites, where they would be used less often to administer general anesthesia. Yet, the district court reasoned, safety advances are a common phenomenon, and the fact that newer models have enhanced safety features lacking in older models does not as a matter of course render the latter "defective." The fact that older models will be shifted to secondary locations as newer models are purchased is likewise a common phenomenon, and the lack of familiarity with the older models a natural result. Moreover, the need for caution for in the use of older machines with parallel features like the Model 2000 Vernitrol is a matter of common sense. Thus, the shift from older to newer models that are more foolproof does not signify a "defect" requiring manufacturers to warn hospitals about the dangers of complacency and the need to take care with the older models as they fall into disuse. 863 F.Supp. at 875-77.
Second, there was no proof that any safety features of Ohmeda's newer anesthesia machines had been developed in response to a known defect. The plaintiffs had cited the calibrated vaporizer as such a feature. It permits the user to simply to turn a dial to the desired percentage of anesthetizing agent without having to engage in any of the calculations or settings required by the Vernitrol vaporizer. The danger of using the wrong control knob or of selecting the wrong flow on the Vernitrol is thus eliminated. But, the district court pointed out, there was no evidence that the calibrated vaporizer had been developed in response to a known defect. There was no proof that Ohmeda had ever considered the calibrated vaporizer to be a safety device; in fact, it had distributed the calibrated vaporizer as part of a giveaway program to promote the use of Forane. Moreover, calibrated vaporizers were frequently installed along side of the dual flowmeter Vernitrol as an alternative to, but not a replacement for, the Vernitrol, permitting the user to select the control of his or her preference. 863 F.Supp. at 877-78.
Third, even if the calibrated vaporizer had been developed as a safety device, the court found that St. Luke's had been aware of the availability of this device prior to the incident at issue in this case. In particular, the evidence indicated that St. Luke's was already using calibrated vaporizers on some of its anesthesia machines at the time Ms. Olsen was injured. 863 F.Supp. at 878.
The plaintiffs do not quarrel with the elements of a Kozlowski claim as the district court articulated them. Instead, they argue that the evidence was sufficient to support each of the three elements that the district court found unmet. We need not undertake a review of each of these elements, however. Because we agree with the district court that the plaintiffs failed to adduce sufficient proof of a defect in the Ohmeda 2000, our inquiry need proceed no further.
So far as the record reveals, no other patient anesthetized using Model 2000 had been injured in the manner Olsen was; nor do the plaintiffs point to any "near misses" that might have placed Ohmeda on notice of a defect in its product. See Kozlowski, 275 N.W.2d at 921. Instead, as the district court observed:
It is crystal clear from the evidence that during the long post sale period that the Model 2000 was used, it was used without any incidents of the type that are the subject of this lawsuit. The evidence further disclosed that, to Ohmeda's knowledge, the Model 2000, put to its proper use, functioned in all types of hospital environments over a long period of years without drawing any complaints and without causing any damage or injury. To Ohmeda's knowledge, the machine was never put to an unintended use by any user, and, while Ohmeda made many observations of its product over a long period of time, none of those observations were of defects or defective conditions of the type found in Kozlowski. Indeed, the evidence disclosed that Ohmeda was aware of no defect or defective condition in its product at all.
863 F.Supp. at 875 (footnote omitted).
What the plaintiffs rely on to establish a defect in the machine is evidence that Ohmeda was aware of the possibility that operators of the Model 2000 might be confused by the dual knobs and flowmeters that controlled the Vernitrol vaporizer. For example, they point to a 1975 anesthesiology equipment textbook which, in its discussion of the "hazards" associated with the Vernitrol, notes that "[w]hen there are flowmeters in parallel, care must be taken that the fine and coarse flows are not confused." Plaintiffs' Ex. 155, Suppl.App. 142. Plaintiffs also cite the American National Standards Institute's standard Z.79.8, published in 1979. Plaintiffs' Ex. 81. This standard, which outlines the minimum performance and safety requirements for anesthesia systems and components, indicates that an anesthesia machine should be equipped with only one flowmeter per gas. The standard does not in any way bar or deem unsafe machines with dual flowmeters (it instead permits their continued production for a period of five years), but provides that machines supplied with a dual flowmeter setup should display appropriate warnings. Id. p 11.1.1. Like the textbook, this standard seems implicitly to acknowledge the prospect of operators becoming confused between the fine and course flowmeters on machines like the 2000. See id. App. C, pp C24, C28.2.
We find this evidence insufficient to establish a defect in the machine triggering a duty on the part of the manufacturer to issue warnings. The mere possibility that an operator might make a mistake does not alone establish that the product is defective and unreasonably dangerous. Human nature being what it is, mistakes can be made in the use of virtually any machine; as the complexity of the machine increases, so too does the number of mistakes that might be made. Many of these mistakes, and the dangers they pose, will be foreseeable to the manufacturer, but many of them will be equally apparent to the users of the machine. Kozlowski's focus is on defects engendering dangerous conditions that are not apparent to the user. As the Restatement explains, for a product to be deemed "unreasonably dangerous," it "must be dangerous to an extent beyond that which would be contemplated by the ordinary consumer who purchases it, with the ordinary knowledge common to the community as to its characteristics." RESTATEMENT (SECOND) OF TORTS § 402A cmt. i (1965), adopted in Vincer v. Esther Williams All-Aluminum Swimming Pool Co, 230 N.W.2d 794, 798 (Wis.1975); see also Kozlowski, 275 N.W.2d at 920. Thus, the manufacturer only becomes obligated to issue warnings when it learns of a danger that purchasers of the product would not themselves appreciate. In Kozlowski, for example, it was the risk of the piston exploding through the safety ring when the cover was removed from the machine, a circumstance as to which the plaintiff claimed appropriate warnings had never been given.
In this case, the purported defect in the Model 2000 lies in the fact that dual knobs and flowmeters control the delivery of anesthesia and the danger is the possibility that the user will confuse the two or, as Myers testified she did, fail to realize that there even are two.
We are uncertain, in the first instance, whether the risk of confusion establishes the dual control Vernitrol as defective. The fact that the Model 2000 Vernitrol has two knobs and two flowmeters is plain from even a cursory look at a photograph of the machine; and the need to take care not to confuse the fine and course flows would seem to be a matter of common sense even to the layperson. Surely the manufacturers of medical devices as complex as anesthesia machines are entitled to expect that only personnel properly trained and familiarized with the machine will use it, and nothing in the record indicates that such persons would not realize that the Model 2000 Vernitrol is controlled by two, rather than one, set of knobs and flowmeters. Indeed, Myers herself admitted at trial that her mistake would not have occurred if she had been familiar with the machine. Tr. 265. The dual control design of the Model 2000 is obvious, as is the danger in the event that the user fails to distinguish between the fine and course flows.
What plaintiffs are left with is the argument that although the dual-control Vernitrol may not have presented the type of dangerous condition requiring a warning in its heyday, when users were familiar with its peculiarities, the duty to warn arose as the machines were reassigned to secondary locations and used infrequently. In other words, as users become more accustomed to newer models with enhanced safety features, they may fail to appreciate the care that must be taken with older machines that lack such features. Thus, plaintiffs reason, medical personnel who use newer anesthesia machines that employ calibrated vaporizers rather than the dual flowmeter design of the Model 2000 do not appreciate the risks posed by the older machine. Ohmeda, which continued to service the older machines at St. Luke's, was aware that these machines had been removed to secondary locations like obstetric suites, where, as Myers' testimony revealed, they were used to administer oxygen and occasionally a whiff of nitrous oxide, but less often for purposes of general anesthesia. And, as a draft of an Ohmeda flyer on risk management revealed, the company was aware of the fact that as these machines were placed in locations where they would be used less often, the risk of mistakes or confusion in using these machines was greater. Plaintiffs' Ex. 228; see also Plaintiffs' Ex. 217.
We are not persuaded by this argument. Plaintiffs' theory takes a significant step beyond Kozlowski to the extent that the dangerous condition of which they complain arises not solely from the product itself, but from the safety advances of later models. Kozlowski and its progeny do not, of course, ignore safety devices; indeed, the point of Kozlowski is that a manufacturer must warn an otherwise ignorant consumer of a safety device that it has developed in response to a known defect. But the plaintiffs direct us to no case that encompasses the more expansive duty to warn consumers of the comparative dangers of older models as newer models with enhanced safety features come into common use. That theory stretches far beyond the rationale of Kozlowski, and we find no support for the expansion in this case. Hospitals are intimately familiar with complicated machinery requiring a high degree of attention on the part of the user. That newer machines with foolproof features may cause the vigilance that older models required of their users to atrophy is a matter of common sense, certainly to trained medical personnel. Moreover, the care required in use of the Model 2000 should have been obvious to any trained individual, as the dual-knob, dual-flowmeter design of the Vernitrol was in no sense "hidden."
III.
For the reasons we have discussed, we agree with the district court that the evidence before the jury did not reasonably support a finding that Ohmeda learned of any defect in its Model 2000 anesthesia machine that would trigger a duty to issue warnings in compliance with Kozlowski v. John E. Smith's Sons Co. On that basis, we affirm the grant of judgment as a matter of law in favor of Ohmeda.
AFFIRMED.
