Monroe v. Zimmer U.S. Inc.

766 F. Supp. 2d 1012 (2011)

Sara MONROE, Plaintiff,
v.
ZIMMER U.S. INC., Zimmer, Inc., and Does 1 through 25, Defendants.

No. CIV. S-08-2944 FCD/EFB.

United States District Court, E.D. California.

February 14, 2011.

Stewart C. Altemus, Altemus & Wagner, Redding, CA, Paul J. Schachter, PHV, Penny Hendy, PHV, Ronald E. Johnson, PHV, Jr., Schachter, Hendy and Johnson, Fort Wright, KY, for Plaintiff.

Brian J. Mooney, Gordon & Rees LLP, San Francisco, CA, Catherine L. Manske, Gordon & Rees LLP, Sacramento, CA, Michael J. Kanute, PHV, Baker & Daniels, LLP, Chicago, IL, for Defendants.

MEMORANDUM AND ORDER

FRANK C. DAMRELL, JR., District Judge.

This matter is before the court on defendants Zimmer US, Inc. and Zimmer Inc.'s ("defendants") motion for summary judgment pursuant to Federal Rule of Civil Procedure 56 and defendants' motion to exclude the testimony of Martin T. Wells ("Wells") pursuant to Federal Rule of Evidence 702.^[1] For the reasons set forth below, defendants' motion for summary judgment is GRANTED in part and DENIED in part and defendants' motion to exclude Wells' testimony is DENIED.

BACKGROUND^[2]

Plaintiff Sara Monroe ("plaintiff") alleges that use of a Zimmer Ambulatory Pump ("ZAP")^[3] caused her to lose cartilage in her shoulder, resulting in unbearable pain and suffering. (Complaint, filed Oct. 16, 2008.) Plaintiff raises claims under California law for: (1) general negligence; (2) negligent products liability; and (3) strict products liability. As to the claim of general negligence, plaintiff alleges that defendants negligently designed, manufactured, and distributed a defective ZAP and negligently failed to warn doctors and patients that use of the ZAP in shoulder joints was unsafe.

1. Overview^[4]

A. Anatomy of the Shoulder and the Glenohumeral Joint

The ball-and-socket joint of the shoulder is called the glenohumeral joint. The area surrounding the glenohumeral joint contains articular cartilage and synovial tissue. Articular cartilage is a thin layer of tissue that covers the ends of bones in the joint and provides a smooth, gliding surface that enables the bones to move. Articular cartilage receives its nutrients exclusively from synovial fluid within the joint space.

B. Glenohumeral Chondrolysis

Articular cartilage cells are known as chondrocytes is the destruction of these cells. Glenohumeral refers to rapid and permanent destruction of in the shoulder joint. See D.J. Solomon et al., Chondrolysis After Arthroscopy: A Systematic Review and Causal Pathways, 25 ARTHROSCOPY: J. ARTHROSCOPIC & RELATED SURG. 1329, 1330 (Nov. 2009). Cartilage cell death may occur from the inability of chondrocytes to maintain or produce cartilage matrix. Id. If the cartilage matrix is not renewed, it wears away with normal use of the joint until all of the protective tissue is gone. This results in the bones of the joint rubbing against one another, causing debilitating pain and stiffness.

C. Pain Pump

A pain pump is a medical device designed to deliver targeted doses of pain management medication to a specific part of the body. A pain pump consists of a fluid reservoir that contains one or more pain medications, a mechanism that pumps the pain management medication to the patient, and a catheter that delivers the pain management medication to a specific part of the body.

2. Plaintiff's Medical History

On May 18, 2007, plaintiff's orthopedic surgeon, Dr. Richard Cross ("Cross"), performed surgery on plaintiff's left shoulder. (Deposition of Dr. Richard Cross ["Cross Dep."], filed Sep. 23, 2010 [Docket # 80], Ex. C at 23:5-7.)^[5] The surgery involved decompression of the acromion and a resection of the distal clavicle. (Id. at 23:16-21; DUF ¶ 5.) Prior to the surgery, on May 10, 2007, Cross injected plaintiff's left shoulder with the anesthetic Marcaine.^[6] (Cross Dep. at 12:3-7.) During the surgery, Cross found that plaintiff had a "degenerative or arthritic and/or arthritic acromioclavicular joint" and "some impingement in her shoulder or crowding of her acromion." (Id. at 23:10-13.) Cross noted that these issues were "not unusual" for a patient of plaintiff's age with plaintiff's medical background. (Id. at 23:25-24:4.)

Also on May 18, 2007, Cross used a radio frequency probe on plaintiff's glenoid, within the glenohumeral joint. (Id. at 71:15-25.) He used this probe, at a low frequency, to smooth a mild amount of damaged cartilage. (Id.) Cross noted that this damage was typical wear and tear damage for someone of plaintiff's age. (Id.)

Following the surgery on plaintiff's left shoulder, a ZAP was inserted into the subacromial space of plaintiff's left shoulder for a minimum of two days.^[7] (DUF ¶¶ 7-8, 10; Cross Dep. at 30:4-7.) A fluid pathway from the ZAP catheter to plaintiff's glenohumeral joint should not have been present at this time; however, it is possible that such communication occurred. (DUF ¶ 9; Cross Dep. at 54:2-25.)

After plaintiff's surgery, she was involved in an altercation where she was punched in her left shoulder. (DUF ¶¶ 11-12.) This altercation delayed plaintiff's treatment and led to Cross injecting corticosteroids into plaintiff's glenohumeral joint. (DUF ¶¶ 14-15.)

On August 31, 2007, Cross performed a second surgery on plaintiff's left shoulder. (DUF ¶ 16.) During this surgery, Cross noted a lesion, labral tearing, and an onset of degenerative changes in the articular cartilage of plaintiff's glenoid. (DUF ¶ 17.) Cross also noted a Hill-Sachs lesion of the humerus during this surgery. (DUF ¶ 18; Cross Dep. at 52:11-25-53:1-5.) As he did during the first surgery, Cross used a radio frequency probe on the cartilage of plaintiff's glenoid. (DUF ¶ 20.)

During plaintiff's surgery on August 31, 2007, a ZAP was inserted directly into plaintiff's glenohumeral joint.^[8] (DUF ¶ 21.) The ZAP remained in plaintiff's shoulder for six days. (Cross Dep. at 62:7-21.) After this surgery, plaintiff received two additional injections of corticosteroids into her glenohumeral joint. (DUF ¶ 22.)

After receiving the injections, plaintiff complained to Cross of increased pain and decreased range of motion in her shoulder. (Cross Dep. at 68:21-24.) Cross ordered an x-ray, which revealed global loss of cartilage in plaintiff's shoulder. (Id. at 69:2-6.) Plaintiff was subsequently diagnosed with chondrolysis in her left shoulder. (DUF ¶ 23.)

3. Causation

A. Specific Causation

Testimony on the issue of the specific cause of plaintiff's injury comes from plaintiff's treating physician, Cross. Plaintiff disclosed Cross as a non-retained expert. (DUF ¶ 25.) Cross admits that he is not an expert on the development of chondrolysis, nor is he an expert on whether chondrolysis is caused by pain pumps. (DUF ¶ 26.) As discussed below, Cross states that he implemented the "differential diagnosis" method to determine the cause of plaintiff's injury.

Initially, Cross believed that plaintiff's chondrolysis resulted from her altercation. (DUF ¶ 27; Cross Dep. at 74:11-21.) It was not until January 24, 2008 that Cross determined that plaintiff's chondrolysis was "possibly" due to the ZAP. (Cross Dep. at 74:24-25-75:1-5.) Cross testifies that this change occurred because he was not previously aware of an association between chondrolysis and pain pump use. (Id. at 75:17-19.) Cross testifies that he is not entirely sure, but he believes that a "blurb" on the internet drew his attention to potential problems associated with the use of pain pumps. (Id. at 75:24-25-76:1-4.) Once he became aware of this potential association, Cross reports that it "clicked" for him, that the pain pump may be the cause of plaintiff's chondrolysis. (Id. at 76:4-7.) During his deposition, defendants' counsel asked Cross if he ever formed an opinion that the pain pump was a substantial factor in causing plaintiff's chondrolysis. (Id. at 77:21-23.) Cross responded "[t]hats a good question. And I would say I think I have. I think I have, but it's not science. It's an opinion ... it's my gut feeling with [plaintiff]." (Id. at 77:24-25, 78:2-3.)

Cross later elaborated on his thought process, discussing how he considered other factors that could have caused plaintiff's injury but ultimately ruled them out, eventually deciding that the pain pump may have been a significant cause of plaintiff's injury. (Id. at 78:4-25-79:1-13.) Cross also testified that, "if I could do it all over again, I never would've put a pain catheter in her. Whether we would've had this problem or not, who's to say? ... Certainly you can argue that point." (Id. at 80:13-18.) Eventually, in response to questioning from plaintiff's counsel, Cross testified that it was his opinion, based on his review of medical literature, many years of orthopedic experience, and clinical experience with plaintiff, that the pain pump was "the most probable cause" of plaintiff's chondrolysis.^[9] (Id. at 102:XX-XX-XXX:1-11.)

B. General Causation

1. Dr. Dragoo

One of plaintiff's two retained experts on the issue of general causation is Dr. Jason Dragoo ("Dragoo"), a professor of Orthopedic Surgery at Stanford University. (Decl. of Dr. Dragoo ["Dragoo Report"], filed Feb. 19, 2010 [Docket # 29], Attachment 7 at ¶ 1.) Dragoo reviewed retrospective studies, case reports, and an in vivo animal study. Dragoo also conducted an in vitro human study, using articular cartilage from human knees and a "culture system" designed to replicate the normal physiology of the shoulder. (Dragoo Report ¶¶ 14-17.) Based on the results of his study and his review of prior studies, Dragoo determined, with a reasonable degree of medical certainty, that intra-articular pain pump usage can cause cartilage degeneration. (Id. at ¶ 31.)

2. Dr. Wells

Wells, the Chair of the Department of Statistical Sciences at Cornell University, is plaintiff's other expert on the issue of general causation. (Decl. of Martin T. Wells, Ph.D. ["Wells Report I"], filed Feb. 19, 2010 [Docket # 29], Attachment 3 at ¶ 2.) Wells reviewed epidemiological and scientific literature relating to intra-articular pain pump use and its connection to glenohumeral chondrolysis. (Wells Report I ¶ 1.) As discussed infra, Wells conducted statistical analysis on the data reported in these studies to determine whether they reported a statistically significant causal relationship between the use of intra-articular pain pumps and the development of chondrolysis. (Id. at ¶¶ 1, 14.)

In his analysis, Wells employed principles set forth in Bradford Hill's criteria of causation ("Hill analysis"), which are meant to assess the presence of a meaningful, statistically valid association between phenomena. (Id. at ¶ 14.) Wells' statistical analysis led him to conclude that there is a substantial difference in the risk of developing chondrolysis for patients that use a pain pump versus patients who do not. Wells also analyzed the data to determine whether other potential risk factors could be responsible for the development of chondrolysis, ultimately ruling them out.

STANDARD

1. Summary Judgment

The Federal Rules of Civil Procedure provide for summary judgment where "the movant shows that there is no genuine dispute as to any material fact and the movant is entitled to judgment as a matter of law." Fed.R.Civ.P. 56(a). The evidence must be viewed in the light most favorable to the nonmoving party. See Lopez v. Smith, 203 F.3d 1122, 1131 (9th Cir.2000) (en banc).

The moving party bears the initial burden of demonstrating the absence of a genuine issue of fact. See Celotex Corp. v. Catrett, 477 U.S. 317, 325, 106 S. Ct. 2548, 91 L. Ed. 2d 265 (1986). If the moving party fails to meet this burden, "the nonmoving party has no obligation to produce anything, even if the nonmoving party would have the ultimate burden of persuasion at trial." Nissan Fire & Marine Ins. Co. v. Fritz Cos., 210 F.3d 1099, 1102-03 (9th Cir.2000). However, if the nonmoving party has the burden of proof at trial, the moving party only needs to show "that there is an absence of evidence to support the nonmoving party's case." Celotex Corp., 477 U.S. at 325, 106 S. Ct. 2548.

Once the moving party has met its burden of proof, the nonmoving party must produce evidence on which a reasonable trier of fact could find in its favor viewing the record as a whole in light of the evidentiary burden the law places on that party. See Triton Energy Corp. v. Square D Co., 68 F.3d 1216, 1221 (9th Cir.1995). The nonmoving party cannot simply rest on its allegations without any significant probative evidence tending to support the complaint. See Nissan Fire & Marine, 210 F.3d at 1107. Instead, the nonmoving party must cite to "particular parts of materials in the record," or show that moving party's cited materials "do not establish the absence or presence of a genuine dispute, or that an adverse party cannot produce admissible evidence to support the fact." Fed.R.Civ.P. 56(c)(1).

2. Expert Testimony

The admissibility of expert testimony is governed by Federal Rule of Evidence 702 ("Rule 702"). When specialized knowledge will assist the trier of fact to understand the evidence in a case, Rule 702 permits testimony by experts with the knowledge, skill, experience, training, or education necessary to testify thereto, as long as their testimony meets several requirements. Fed.R.Evid. 702. A qualified expert's testimony is admissible if: (1) the testimony is based upon sufficient facts or data; (2) the testimony is the product of reliable principles and methods, and (3) the witness has applied the principles and methods reliably to the facts of the case. Id.

The role of the trial judge is to act as a gatekeeper, ensuring that only relevant and reliable testimony is admitted. See Daubert v. Merrell Dow Pharms., Inc., 509 U.S. 579, 589, 113 S. Ct. 2786, 125 L. Ed. 2d 469 (1993) ("Daubert I"); see also Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 152, 119 S. Ct. 1167, 143 L. Ed. 2d 238 (1999). Accordingly, the court must conduct a "preliminary assessment of whether the reasoning or methodology underlying the testimony is scientifically valid and of whether that reasoning or methodology properly can be applied to the facts in issue." Daubert I, 509 U.S. at 592-93, 113 S. Ct. 2786. "Pertinent evidence based on scientifically valid principles will satisfy those demands." Id. at 597, 113 S. Ct. 2786.

In Daubert I, the Supreme Court listed several key considerations to aid the trial court in its gatekeeping role. The trial court should consider: (1) whether the scientific knowledge can or has been tested; (2) whether the given theory or technique has been published or subjected to peer review; (3) the potential or known error rate; and (4) whether the theory or technique has gained general acceptance in the pertinent field. Daubert I, 509 U.S. at 592-94, 113 S. Ct. 2786.

The court's Rule 702 analysis is flexible. Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 150, 119 S. Ct. 1167, 143 L. Ed. 2d 238 (1999). The court may consider any or all of the Daubert I factors, depending on the "particular circumstances of the particular case at issue." (Id.)

The trial court may dismiss testimony that is either irrelevant or unreliable. See Stilwell v. Smith & Nephew, Inc., 482 F.3d 1187 (9th Cir.2007). In evaluating the expert's testimony, the court must focus on the principles or methodology involved, not the conclusions that they generate. Daubert I, 509 U.S. at 595, 113 S. Ct. 2786. Thus, "the test under Daubert is not the correctness of the expert's conclusions but the soundness of his methodology." Daubert v. Merrell Dow Pharms., Inc., 43 F.3d 1311, 1318 (9th Cir.1995) ("Daubert II"). "Ultimately, the court must ensure that an expert employs in the courtroom the same level of intellectual rigor that characterizes the practice of an expert in the relevant field," such that the expert's "work product amounts to `good science.'" McClellan v. I-Flow Corp., 710 F. Supp. 2d 1092, 1100 (D.Or.2010) (quoting Kumho, 526 U.S. at 152, 119 S. Ct. 1167 and Daubert II, 43 F.3d at 1315).

ANALYSIS

Defendants move for summary judgment, arguing that plaintiff's claims fail because her experts are unable to prove that the ZAP is capable of causing chondrolysis generally, or that the ZAP specifically caused plaintiff's injury. Further, defendants assert that plaintiff's claims should be dismissed because she has not demonstrated that defendants breached a specific duty of care to her.

Preliminarily, defendants attack the testimony of Wells, one of plaintiff's experts on general causation. As the resolution of defendants' motion for summary judgement turns, in large part, on whether plaintiff has sufficient, admissible evidence to establish that intra-articular use of a pain pump can cause chondrolysis, the court addresses defendants' motion to exclude Wells' testimony first, as without Wells' testimony, plaintiff cannot prevail on the motion. Second, because the court denies defendants' motion to exclude Wells' testimony, it next considers defendants' motion for summary judgment.

1. Motion to Exclude Wells' Testimony

Plaintiff offers Wells' testimony as proof that intra-articular use of pain pumps can cause glenohumeral chondrolysis. Defendants move to exclude Wells' testimony, stating that (1) he is not qualified to opine on the cause of glenohumeral chondrolysis, and (2) his testimony is unreliable because he lacks adequate scientific bases for his opinions.

A. Dr. Wells' Report

At plaintiff's request, Wells reviewed statistical aspects of scientific literature relating to intra-articular pain pump use and subsequent glenohumeral chondrolysis. (Wells Report I ¶ 1.) Wells analyzed the data reported in these studies to determine whether a causal relationship existed between the use of intra-articular pain pumps and the development of chondrolysis. (Id. at ¶¶ 1, 14.) In his analysis, Wells employed principles of the Hill analysis, which are designed to assess the presence of a meaningful, statistically valid association between phenomena. (Id. at ¶ 14.) These criteria include: (1) temporality; (2) strength of association; (3) consistency; (4) consideration of alternate explanations; (5) specificity; (6) dose response; (7) plausibility; (8) coherence; and (9) experiment. (Id.)

Wells reviewed a paper entitled Postarthroscopic Glenohumeral Chondrolysis ("Hansen study"). B. Hansen & C. Beck, 35 AM. J. SPORTS MED. 1628 (July 2007). The Hansen study examined two groups of individuals who had undergone shoulder surgery, one group that used an intra-articular pain pump ("pump group") and another that did not ("no pump group"). (Id.) Hansen's study reviewed the results of 177 shoulder arthroscopies performed on 152 patients. (Id.)

Nineteen of the 177 shoulders received intra-articular pain pump catheters filled with bupivacaine and epinephrine. (Id.) Of these 19 shoulders, 12 developed chondrolysis. (Id.) On the other hand, none of the 158 shoulders that did not receive a pain pump developed chondrolysis. (Id.)

Additionally, Wells' review of the Hansen study included 30 shoulders that underwent arthroscopic stabilization procedures. (Supplemental Decl. of Martin T. Wells, Ph.D. ["Wells Report II"], filed July 27, 2010 [Docket # 68], Attachment 2 at ¶ 3.) Of these 30 shoulders, 11 of them were not exposed to a pain pump. (Id.) None of these 11 shoulders developed chondrolysis. (Id.) Whereas, of the 19 shoulders that were exposed to a pain pump, 12 developed chondrolysis. (Id.) Accordingly, the authors of the Hansen study concluded that the "[u]se of intra-articular pain pump catheters eluting bupivacaine with epinephrine appear highly associated with postarthroscopic glenohumeral chondrolysis." (See Hansen Study.)

Wells conducted various analyses of the Hansen study data, concluding that, "the risk difference between the pump and no pump groups with respect to chondrolysis is extremely large and is well beyond what would occur by chance and far too large to be credibly explained by any other factor that has been shown linked to chondrolysis;" "the association of postsurgical use of intra-articular pump with chondrolysis is well beyond what would occur by chance;" and "diagnostic bias is not a viable explanation for the enormous association of intra-articular pain pump use with chondrolysis." (Wells Report I ¶¶ 8-11; Wells Report II ¶ 4.)

Wells explained his methodology for reaching each of these conclusions, describing numerous statistical calculations derived from accepted methodologies in this industry, including: p-values using Fisher's exact test; the odds ratio of chondrolysis from the pump group to the no-pump group; the median unbiased odds ratio estimate; the continuity corrected odds ratio; the Fisher's mid-p exact 95% confidence limit; and the risk difference between the two groups. (See Wells Report I, II).

Wells also reviewed the data reported in an unpublished study of arthroscopic shoulder surgeries conducted by Dr. Frederick Matsen ("Matsen"). (Wells Report I ¶ 12.) Matsen reviewed 396 surgeries performed by Dr. Brooke Benz ("Benz") from 1996 through 2008 ("Matsen study"). 281 of Benz's surgeries did not involve the use of an intra-articular pain pump. (Id. at ¶ 13.) No instances of chondrolysis were reported after those surgeries. (Id.) Whereas, of the 115 of Benz's surgeries that involved the use of an intra-articular pain pump, forty-eight, or 41.74%, of the shoulders developed chondrolysis. (Id.)

Wells performed a similar statistical analysis for the Matsen study as he did for the Hansen study. Once again, Wells calculated the estimated risk difference between the two groups, the Fisher exact p-values, the median-unbiased estimate for the odds ratio, and the continuity corrected odds ratio. (Id.) Based on these calculations, Wells determined that the data presents "an extremely high association" between the use of an intra-articular pump and the development of chondrolysis "that is far too large to be credibly explained by any other factor that has been shown linked to chondrolysis." (Id.) Further, based on Wells' calculations, he found that the Matsen study results are consistent with the results of the Hansen study. (Id.)

After Wells' initial declaration was filed, a draft of Matsen's study became available. (Wells Report II ¶ 5.) This draft was authored by Dr. Wiater, Dr. Matsen, and others ("Wiater study").^[10] (Id.) Wells obtained the datasets and program files used in this study. (Id. ¶ 6.) Wells then conducted numerous independent data analyses, finding results that were consistent with the results of the Wiater study. (Id.)

By conducting regression analyses, and examining results by chondrolysis onset date and follow-up status definitions, Wells reports that the Wiater study's results provide strong evidence that selection bias and confounding by anchors, radiofrequency, sutures, or other factors were not viable alternative explanations for the elevated correlation between intra-articular pain pump use and chondrolysis. (Id. ¶ 7.) Wells also reports that in his view, the Wiater study is consistent with the results of the Hansen study. (Id.) Further, Wells states that his Wiater study analysis allows an inference that intra-articular pain pump use was a primary causative factor of the chondrolysis cases reported in the Hansen and Wiater studies. (Id.)

Wells also analyzed a study conducted by Rapley et al. entitled Glenohumeral Chondrolysis After Shoulder Arthroscopy Associated with Continuous Bupivacaine Infusion ("Rapley study"). Rapley et al., ARTHROSCOPY: 25 THE JOURNAL OF ARTHROSCOPIC AND RELATED SURGERY 12 (2009). Rapley's study compared a group of individuals who used a pain pump with a higher infusion flow rate (4.16 mL/h) with a group of individuals that used a pain pump with a lower infusion flow rate (2.08 mL/h). (Wells Report II ¶ 8.) The higher infusion group experienced a 19% chondrolysis rate (3/16), while the lower infusion group's chondrolysis rate was 0% (0/13). (Id.)

Wells then took the users from the pump group in Wiater's study that reported a flow rate and divided them into one group with a flow rate higher or equal to 4.16 mL/h, and one group with a flow rate lower than 4.16 mL/h. (Id.) 75% of the higher flow group developed chondrolysis (9/12), compared to 0% of the lower flow group (0/2). (Id.)

Wells then calculated the mid p-values for the Wiater and Rapley studies and determined that the association of higher flow rate intra-articular pump use with chondrolysis is higher than what could occur by chance. (Id.) Additionally, Wells compared his analysis of the Wiater and Rapley studies to the Hansen study data and found that the results regarding higher flow intra-articular pump use were consistent across all three studies. (Id.)

B. Dr. Wells' Qualifications

Wells is a biostatistician. He received his Ph.D. in Mathematics from the University of California. (Wells Report I, ¶ 11.) He is currently the Chair of the Department of Statistical Sciences at Cornell University, and formerly the Chair of the Department of Biological Statistics and Computational Biology at Cornell University. (Id. ¶ 2.) Wells has received numerous professional honors and has published well over one hundred scholarly articles. (Id. ¶¶ 13-20.)

Defendants assert that Wells' general causation testimony is not admissible under Rule 702 because he is not qualified to testify to whether pain pumps can cause glenohumeral chondrolysis. Based on the qualifications described above, Wells undoubtedly has the skill, experience, training, and education necessary to testify regarding mathematics and statistical analyses.^[11] Defendants argue that despite Wells' mathematics and statistics expertise, he is not qualified to opine as to general causation because he is not a qualified expert in medicine or epidemiology.

In re Welding Fume Products, No. 1:03-17000, 2005 WL 1868046 (N.D.Ohio. Aug. 8, 2005), presented issues similar to the current case. As here, the plaintiffs in In re Welding Fume Products designated Wells "to opine about the strengths and weaknesses of the statistical analyses contained in various epidemiology studies." Id. at *15. In both cases, Wells analyzed studies to determine whether a statistical association existed between a product and a harm.^[12]Id.

Defendants in In re Welding Fume Products made similar arguments to defendants in this case in moving to exclude Wells' testimony. They argued that Wells is "`simply a statistician,' and not a medical doctor or epidemiologist." Id. Defendants in In re Welding Fume Products further argued that Wells did not have experience with the product or harm involved. Id. Accordingly, they asked the court to "preclude Dr. Wells from testifying about any studies in the epidemiological and scientific literature discussing the existence vel non of an association between [the product and the harm]." Id.

The court in In re Welding Fume Products denied the defendants' request, determining that "ultimately the opinions offered by Dr. Wells are relevant to the central epidemiology issues in this case, and Dr. Wells is qualified to offer them." Id. at *16. The court further noted that "[t]hat Dr. Wells is not an epidemiologist is not at all decisive, as his understanding of statistical principles and methodology relevant to the design and interpretation of epidemiological studies is thorough and clear." Id. at *15.

The court finds the same in this case. As a prolific publisher in the field of biostatistics, Wells is capable of understanding the data reported in studies conducted by medical professionals. Wells analyzed the data reported in prior studies to determine the statistical strength of the correlations that the authors report. He did not determine what data would be collected in any of these studies, nor did he collect the data himself.

Nonetheless, Wells' findings are all reported in the language of a statistician. At no time does Wells attempt to offer a medical or epidemiological opinion. Rather, he is an expert in applying statistical principles relevant to the design and interpretation of epidemiological studies. Wells is aptly qualified to provide expert testimony. Accordingly, defendants' motion to exclude Wells' testimony based on lack of qualifications is DENIED.

C. Relevance of Dr. Wells' Statistical Analysis

Defendants also argue that Wells' testimony is inadmissible because he is not qualified to assess all of the factors of the the Hill analysis. Wells admits that he is not qualified to comment upon five of Hill's nine criteria of causation. This issue falls under the relevance prong of the Daubert standard. Specifically, whether the four criteria of the Hill analysis that Wells purports to be qualified to comment upon are sufficient to provide relevant testimony as to general causation.

Pursuant to Rule 702, the court may not admit expert testimony, unless it determines that it will "assist the trier of fact to understand the evidence or determine a fact in issue." Daubert I, 509 U.S. at 591, 113 S. Ct. 2786. "Expert testimony which does not relate to any issue in the case is not relevant and, ergo, non-helpful." (Id.) Accordingly, in its role as gatekeeper, the court must ensure that proffered testimony "logically advances a material aspect of the proposing party's case." Daubert II, 43 F.3d at 1315.

Wells testifies that his analysis confirms Hill's temporal criterion because the Hansen and Matsen studies involved pain pump use prior to the occurrence of chondrolysis, and because the incidence of chondrolysis completely ceased once the doctors stopped using pain pumps. (Id. at 15.) Wells also testifies that the "startlingly high" association and risk differences between the pump and no-pump groups confirms Hill's strength of association criterion. (Id. at 16.) Wells then testifies that, because this association is consistent when replicated in multiple studies in multiple settings, Hill's criterion of consistency is confirmed. (Id. at 17.)

Finally, Wells testifies as to Hill's criterion of considering alternative explanations. Defendants' expert, Judith K. Jones, M.D., Ph.D. ("Jones"), has presented numerous factors that have been cited in medical literature as explanations for the onset of chondrolysis. Specifically, Jones declares that the use of: (1) certain diagnostic dyes; (2) irrigating solutions; (3) thermal/radiofrequency devices; (4) local anesthetics; (5) pain pumps; (6) sutures; (7) anchors; (8) pins; and (9) arthroscopic surgery, have all been reported as factors that may have played a role in the development of chondrolysis. (Report of Opinions of Judith K. Jones, M.D., Ph.D. ("Jones Report") ¶ 8, filed February 19, 2010 [Docket # 28], Attachment 2.)

Wells states that the factors cited by Jones are addressed in the discussion section of the Hansen study and ruled out by Matsen, both in his study and in Matsen's testimony regarding his study in another case. Further, Wells states that if one of these potential cofounders was a primary contributory cause of chondrolysis, then it would have a higher association than intra-articular pump use. This did not occur. (Wells Report I ¶ 19.)

This testimony as to the four Hill analysis factors logically advances plaintiff's case (see Daubert I):(1) plaintiff must establish general causation, and the studies that Wells cites provide evidentiary support for such a finding; (2) this testimony will assist the trier of fact in understanding the evidence in order to determine a fact at issue; (3) Wells' analysis enables the fact finder to better understand the importance of what other researchers have found; Wells' testimony reveals that the "the association of postsurgical use of intra-articular pump with chondrolysis is well beyond what would occur by chance," ruling out other suggested explanations and identifying a consistency amongst multiple studies; and (4) this testimony will assist the jury in understanding whether the medical literature has found that pain pumps can cause chondrolysis.

Certainly, the fact that Wells has not offered testimony as to the other five Hill analysis criteria give defendants fair grounds to attack Wells' opinion. However, defendants' argument as to faults in Wells' Hill analysis go to the weight, not the admissibility of Wells' testimony. See Kennedy v. Collagen Corp., 161 F.3d 1226 (9th Cir.1998). Accordingly, defendants' motion to exclude Wells' testimony based on a lack of relevance is DENIED.

D. Reliability of Wells' Statistical Analysis

Defendants also argue that Wells' testimony is inadmissible because he lacks an adequate scientific basis for his opinions. In its role as gatekeeper, the court must determine whether the reasoning and methodology underlying Wells' testimony is scientifically valid. Daubert I, 509 U.S. at 592-593, 113 S. Ct. 2786. In doing so, the court must ensure that Wells has employed scientifically valid principles. Id. at 597, 113 S. Ct. 2786. The Supreme Court listed several key considerations to aid the trial court in this task. The trial court should consider: (1) whether the scientific knowledge can or has been tested; (2) whether the given theory or technique has been published or subjected to peer review; (3) the potential or known error rate; and (4) whether the theory or technique has gained general acceptance in the pertinent field. Daubert I, 509 U.S at 592-94, 113 S. Ct. 2786. In employing this analysis, the court must focus on the principles or methodology involved in the expert's testimony, not the conclusions that they generate. Daubert I, 509 U.S. at 595, 113 S. Ct. 2786.

Defendants argue that Wells' analysis of the Hansen study is inadmissible due to various faults with the Hansen study itself. Defendants point out that the Hansen study was a case study, rather than a cohort study; that Wells cannot determine whether the Hansen study applied predefined screening criteria; the Hansen study does not allow for an analysis of confounding factors because it does not include non-chondrolysis patient data; and that the Hansen study did not have a follow-up monitoring period.

Defendants' reliance upon Kilpatrick v. Breg Inc., 613 F.3d 1329 (11th Cir.2010) in support of this argument, does not persuade the court to exclude Wells' testimony regarding the Hansen study. In this pain pump negligence and products liability case, the district court determined that the plaintiff's causation expert's methodology was not sufficiently reliable to pass Daubert muster, and the Eleventh Circuit affirmed. Id. at 1341.

The expert, a medical doctor, relied in part on the Hansen study to form his conclusions regarding causation. The district court in Kilpatrick faulted the Hansen study for lacking statistical analysis, failing to (1) explain whether its results where statistically meaningful, (2) account for other causes of chondrolysis, (3) account for the shoulders that were exposed to pain pumps but did not develop chondrolysis, and (4) firmly conclude that pain pumps caused chondrolysis. Id. at 1337-38.

As discussed supra, Wells accounted for the Kilpatrick court's first three criticisms of the Hansen study. Wells analyzed the data to determine whether a statistically meaningful association between pain pump use and the development of chondrolysis was present. He also analyzed the data to determine whether any of the co-founders that Jones mentioned could have caused the shoulders to develop chondrolysis, ultimately ruling them out.

As other courts have, this court disagrees with the Kilpatrick court's determination that the Hansen study contained "an unexplained 40% error rate" because 7 out of 19 shoulders that were exposed to pain pump use did not develop chondrolysis. Kilpatrick v. Breg, Inc., No. 08-10052, 2009 WL 2058384 at *6 (S.D.Fla. Jun. 25, 2009). Just as the court found in Schott v. I-Flow Corp., 696 F. Supp. 2d 898, 905 (S.D.Ohio.2010), this court is unaware of any authority for the proposition that, where 12 out of 19 shoulders developed chondrolysis, the minority of shoulders that did not, constitutes an "error rate." Further, as the court noted in McClellan, 710 F.Supp.2d at 1106, "[i]f 100% of patients receiving continuous infusion must develop chondrolysis before an expert may reliably opine on causation, expert testimony would hardly be necessary."

Finally, unlike the Kilpatrick court, this court will not exclude testimony regarding the Hansen study for failure to conclusively determine that pain pumps cause chondrolysis. As other courts have found, this court likewise holds that excluding testimony on these grounds goes beyond the court's gatekeeping role. See McClellan, 710 F.Supp.2d at 1106 ("I find that the district court in Kilpatrick went far beyond its gatekeeping function and `failed to distinguish between the threshold question of admissibility of expert testimony and the persuasive weight to be accorded such testimony by a jury.'") The court must "ensure that the proffered expert testimony is sufficiently reliable to be admitted at trial for consideration by the trier of fact," not determine whether "the proffered expert testimony sufficiently proves that continuous infusion can cause chondrolysis." Id. Accordingly, for all the reasons discussed supra, the decision in Kilpatrick does not provide a basis for excluding testimony regarding the Hansen study in this case.

The court in McClellan v. I-Flow Corp., 710 F. Supp. 2d 1092 (D.Or.2010), provides a helpful analysis of Wells' methodology in conducting statistical analysis of the Hansen study. The analysis in McClellan is particularly pertinent in this case because, unlike in Kilpatrick, McClellan involved Wells' analysis of the Hansen study, rather than just the Hansen study itself.

The court in McClellan discussed the Hansen study's faults, finding it "questionable whether the Hansen[] study constitutes helpful relevant evidence to support an opinion of general causation between continuous infusion and chondrolysis." Id. at 1108. However, the court explicitly stated that it would not exclude the study for a lack of epidemiological evidence, particularly because there is no epidemiological evidence that rules out general causation.^[13]Id.

Indeed, "the fact that a study is associational—rather than an epidemiological study intended to show causation— does not bar it from being used to inform an expert's opinion." Id. (quoting U.S. v. W.R. Grace, 504 F.3d 745, 765 (9th Cir. 2007)). Further, "`the absence of definitive scientific studies' should not `deprive a jury of having before it scientific opinions,' though `less definitive and more qualified.'" McClellan, 710 F.Supp 2d at 1109 (quoting In re Ephedra, 393 F.Supp.2d at 189-190 (S.D.N.Y.2005)). Moreover, "[n]on-epidemiological sources are frequently utilized by experts in rendering scientific opinions and, under Daubert, should be considered by the court in assessing the reliability of those opinions." McClellan, 710 F.Supp.2d at 1109 (quoting In re Phenylpropanolamine (PPA) Prods. Liab. Litig., 289 F. Supp. 2d 1230, 1242 (W.D.Wash.2003)). Accordingly, the court will not exclude Wells' testimony regarding the Hansen study.

Next, defendants argue that Wells' testimony should be excluded because he analyzed the work of others rather than participating in the underlying studies. In a situation such as this, the fact that the expert has not created his own published research is not dispositive. Rather, "[i]n the absence of independent research or peer review, experts must explain the process by which they reached their conclusions and identify some type of objective source demonstrating their adherence to the scientific method." In re Phenylpropanolamine (PPA) Products Liability Litigation, 289 F. Supp. 2d 1230, 1238 (W.D.Wash.2003) (citing Daubert II).

Wells has met this requirement. As set forth above, Wells utilized widely accepted statistical methods to support all of his reported conclusions. Further, each of the individual statistical tests that Wells utilized were part of the Hill analysis, a larger, widely accepted statistical method.

Defendants also argue that Wells' testimony should be excluded because it was conducted solely for the purpose of litigation, at the request of plaintiff's attorneys. It is true, as defendants point out, that "independent research, rather than research conducted for the purposes of litigation, carries with it the indicia of reliability." Carnegie Mellon University v. Hoffmann-LaRoche, Inc., 55 F. Supp. 2d 1024 (N.D.Cal.1999) (citing Daubert II). However, testimony that is based directly on legitimate, preexisting research unrelated to the litigation provides the most persuasive basis for concluding that the opinions expressed were "derived by the scientific method." Daubert II, 43 F.3d at 1317. Such is the case here. While Wells conducted his research for the purposes of litigation, his research was based directly on legitimate, albeit in part flawed, research unrelated to the litigation.

Accordingly, defendants' motion to exclude Wells' testimony based on a lack of reliability is DENIED.

For all of the above reasons, defendants' motion to exclude Wells' testimony is DENIED.

2. Motion for Summary Judgment

A. Causation

Plaintiff alleges claims under California law for: (1) general negligence; (2) negligent products liability; and (3) strict products liability. Each of these claims require plaintiff to establish causation. Defendants move for summary judgment, primarily on the basis that plaintiff lacks sufficient evidence to establish general or specific causation,^[14] and thus, none of her claims can survive.^[15]

"To prevail in a negligence action, a plaintiff must show that the defendant owed a legal duty, the defendant breached that duty and the breach proximately caused injury to the plaintiff." Garcia v. W & W Community Development, Inc., 186 Cal. App. 4th 1038, 1044, 112 Cal. Rptr. 3d 394 (2010). For a strict products liability claim, plaintiff must show "that the injury to the plaintiff was caused by the defective condition." Gonzalez v. Autoliv ASP, Inc., 154 Cal. App. 4th 780, 793, 64 Cal. Rptr. 3d 908 (2007).

Once a defendant properly raises the issue of causation on a motion for summary judgment, as defendants have here, the burden shifts to the plaintiff to demonstrate a genuine issue of material fact as to causation. Casey v. Ohio Medical Products, 877 F. Supp. 1380, 1382 (N.D.Cal.1995). Plaintiff must meet this burden as to general causation, that the product is capable of causing the harm, and specific causation, that the product caused the harm in this specific case. Id. Where demonstrating causation requires a discussion of complex medical and scientific issues, expert testimony is required. Jones v. Ortho Pharm. Corp., 163 Cal. App. 3d 396, 403, 209 Cal. Rptr. 456 (1985).

1. General Causation

Plaintiff's experts on the issue of general causation are Wells and Dragoo. For the reasons set forth in detail supra, Wells' testimony regarding general causation is admissible. Wells statistically analyzed the data reported in scientific studies and determined that the risk difference with respect to chondrolysis between individuals who received a pain pump and those that did not is extremely large. Wells further determined that the association between postsurgical use of an intra-articular pump with the development of chondrolysis is well beyond what would occur by chance and far too large to be credibly explained by any other factor that has been linked to chondrolysis. Additionally, Wells testifies that his analysis rules out other possible explanations for the subjects' chondrolysis. Accordingly, Wells' testimony creates a triable issue of fact as to general causation.

Dragoo based his opinion upon his own research and his review of several studies conducted by others. Dragoo testifies that his opinion, to a reasonable degree of medical certainty, is that the use of intra-articular pain pumps with a local anesthetic can cause cartilage degeneration in a dose-dependent matter, regardless of whether the pump contains epinephrine. (Dragoo Report ¶ 31.) Dragoo's study, which was published in the American Journal of Sports Medicine in August of 2008, utilized human articular cartilage taken from individuals who were undergoing total knee replacement surgery. (Id. ¶¶ 15, 23.) As a result of this study, Dragoo found that continuous infusion of local anesthetics combined with epinephrine are chondrotoxic. (Id. ¶¶ 21, 24.) He also found that local anesthetics infused without epinephrine may be chondrotoxic when infused for less then forty-eight hours and are likely to be chondrotoxic when infused for longer periods. (Id.)

Defendants criticize Dragoo's testimony for failing to present epidemiological evidence. Epidemiology is "a field that concerns itself with finding the causal nexus between external factors and disease." Rider v. Sandoz Pharmaceuticals Corp., 295 F.3d 1194, 1198 (11th Cir.2002). As defendants point out, epidemiological studies are traditionally considered the best form of statistical evidence for proving causation. Kilpatrick v. Breg, Inc., 2009 WL 2058384 at *4. Defendants also criticize Dragoo's human study as being in vitro rather than in vivo.

However, conducting an epidemiological or in vivo human study would involve knowingly exposing the study's subjects to harm. Dragoo has testified that there is enough underlying research demonstrating a link between pain pumps and chondrolysis, that conducting epidemiological research at this point would be unethical. (Deposition of Dragoo, filed Sep. 23, 2010 [Docket # 81], Ex. G at 128:12-18). Accordingly, Dragoo states that scientists are forced to "piecemeal" the studies that are out there to do what they can "to build a story" of causation. (Id.)

Because epidemiological or in vivo research would be unethical at this point, other courts have refused to require plaintiffs to submit such studies in pain pump product liability litigation. For instance, the court in Schott, 696 F.Supp.2d at 905 found "plaintiffs' argument persuasive that they are unable to obtain epidemiological studies, as conducting any such studies would be unethical." Further stating that, "[i]t therefore strikes the Court as unreasonable for Defendant to clamour for such studies." (Id.) Additionally, the court in McClellan, found that "ethical considerations preclude randomized, controlled epidemiological studies of continuous infusion given the potential for irreversible harm." McClellan, 710 F.Supp.2d at 1109. Thus, the McClellan court found that "while epidemiological evidence is significant and can be helpful, it is not necessary to establish general causation." Id.

This court reaches the same conclusion, and thus will not require plaintiff to produce a study, which by its very nature contains a likelihood of causing harm. Accordingly, plaintiff's failure to submit an epidemiological or in vivo study is not fatal to its claim, particularly where defendants are unable to present epidemiological or in vivo evidence to "rule out" general causation. See Grace, 504 F.3d at 765 (the fact that a study is associational-rather than an epidemiological study intended to show causation-does not bar it from being used to inform an expert's opinion.); see also McClellan, 710 F.Supp.2d at 1108 ("Regardless, the lack of epidemiological evidence is not fatal to the admission of plaintiffs' experts' testimony, particularly in a case where no epidemiology "rules out" continuous infusion as a cause of chondrolysis.")

Defendants then criticize the underlying studies that Dragoo relies upon. Defendants note that the Hansen study used bupivicaine combined with epinephrine, rather than strictly bupivicaine. However, Dragoo's causation conclusion, based on his overall review of the research, was not dependent on the presence of epinephrine. He specifically states that, to a reasonable degree of medical certainty, it is his opinion that "the use of intra-articular pain pumps, with or without epinephrine, can cause cartilage degeneration." (Dragoo Report ¶ 31.)

Defendants next assert that the Hansen study cannot be used to make causation inferences because it is a case series. However, both Wells and Jones have testified that case series are helpful regarding general causation, though they do not establish causation in isolation. (Jones Report, at 7; Wells Dep. 72:11-16).

As discussed supra, the Hansen study has its shortcomings. As was the case in regards to Wells' testimony regarding the Hansen study, defendants can criticize Dragoo's reliance upon the Hansen study at trial. However, "it is not the court's role to decide whether the proffered expert testimony sufficiently proves that continuous infusion can cause chondrolysis; rather it is the court's duty to ensure that the proffered expert testimony is sufficiently reliable to be admitted at trial for consideration by the trier of fact." McClellan, 710 F.Supp.2d at 1106 (citing Primiano v. Cook, 598 F.3d 558, 565 (9th Cir.2010) ("Under Daubert, the district judge is a `gatekeeper, not a factfinder.'")). Because defendants' arguments against Dragoo's Hansen study analysis go towards the weight, not the admissibility of Dragoo's testimony, Dragoo's Hansen study-related testimony further creates a triable issue of material fact as to general causation.

Additionally, defendants criticize Dragoo's use of an animal study to testify regarding the development of chondrolysis in humans. The "Dragoo study" involved continuous administration of Marcaine over a forty-eight hour period to live rabbits. The authors of this study concluded that "[c]ontinuous intra-articular infusion of bupivacaine, with and without epinephrine, led to significant histopathologic and metabolic changes in articular cartilage." (Andres Gomoll et al.), Chondrolysis After Continuous Intra-Articular Bupivacaine Infusion: An Experimental Model Investigating Chondrotoxicity in the Rabbit Shoulder ("Gomoll study"), 22 ARTHROSCOPY 813-19 (August 2006).

Defendants again cite to Kilpatrick to support their argument that the Gomoll study does not provide a factual basis for Dragoo's general causation determination. The Kilpatrick court questioned the plaintiff's expert's reliance on the Gomoll study because the study failed to account for possible differences in dose-response relationship between humans and rabbits making the study unreliable for extrapolation to humans. Kilpatrick, 613 F.3d at 1338-39. The Kilpatrick court noted that "a dose-response relationship is `the single most important factor to consider in evaluating whether an alleged exposure caused a specific adverse effect.'" Id. at 1339. (quoting McClain v. Metabolife Intern., Inc., 401 F.3d 1233, 1242 (11th Cir.2005)).

Contrary to the Kilpatrick court, the courts in both Schott and McClellan admitted expert testimony regarding the results of the Gomoll study. As the McClellan court noted, "`analogy, inference and extrapolation can be sufficiently reliable' when the expert's opinion is the `kind that a reasonable scientist or physician would make in a decision of importance arising in the exercise of his profession outside the context of litigation.'" McClellan, 710 F.Supp.2d at 1110 (quoting In re Ephedra, 393 F. Supp. 2d 181, 189 (S.D.N.Y.2005)). Further, the Ninth Circuit has determined that animal studies can "provide useful data about human health," and thus, should be subjected to scientific analysis just like any other scientific evidence. Metabolife Int'l, Inc. v. Wornick, 264 F.3d 832, 842 (9th Cir.2001); see also Hopkins v. Dow Corning, Corp., 33 F.3d 1116, 1125 (9th Cir.1994) (finding testimony regarding "corroborating evidence found in studies conducted on animals" admissible); see also In re Silicone Gel Breast Implants Prods. Liab. Litig., 318 F. Supp. 2d 879, 910-911 (C.D.Cal.2004) (noting reliance by researchers and agencies on relevant animal studies).

As discussed supra, in vivo studies on the development of chondrolysis in humans are not available and cannot be ethically conducted. Accordingly, the only in vivo studies available involve animals. Under these circumstances, animal studies are particularly useful in aiding the jury in understanding chondrolysis development in humans. Such a study becomes even more valuable where, as is the case with Dragoo's Report, an animal study is supported by human research that accounts for dose-response relationship. (See Dragoo Report ¶ 31, "it is my opinion to a reasonable degree of medical certainty that the use of intra-articular pain pumps... can cause cartilage degeneration in a dose-duration dependent manner.") Thus, as the courts did in McClellan and Schott, this court finds that Dragoo's testimony regarding the Gomoll study is supported by science and will assist the trier of fact in determining general causation.

Accordingly, for all of the reasons stated above, plaintiff has established a material issue of fact regarding general causation.

2. Specific Causation

Plaintiff's lone expert on specific causation is Cross, plaintiff's orthopedic surgeon. Cross admits that he is not an expert on the development of chondrolysis. (Cross Dep. at 111:15-18.) However, as plaintiff's treating physician, Cross is familiar with plaintiff's medical history and treatment. Thus, he is qualified to render opinion testimony on causation, diagnosis, or other matters based on his treatment of plaintiff. First Nat'l. Mortg. Co. v. Federal Realty Inv. Trust, No. 03-02013, 2006 WL 2228941, *14 (N.D.Cal. Aug. 3, 2006).

Defendants argue that Cross does not have a factual basis for determining that use of the ZAP caused plaintiff's injury. Plaintiff contends that Cross' specific causation determination is based in fact because he methodically ruled out other potential causes of chondrolysis.

"[E]xpert's opinions that are without factual basis and are based on speculation or conjecture are inadmissible at trial and are inappropriate material for consideration on a motion for summary judgment." California ex rel. Brown v. Safeway, Inc., 615 F.3d 1171, 1181 (9th Cir. 2010). However, "[l]ack of certainty is not, for a qualified expert, the same thing as guesswork." Primiano v. Cook, 598 F.3d at 565. "Expert opinion testimony is relevant if the knowledge underlying it has a valid connection to the pertinent inquiry." Id. (quoting U.S. v. Sandoval-Mendoza, 472 F.3d 645, 655 (9th Cir.2006)).

Defendants point to changes in Cross' opinion regarding the specific causation of plaintiff's injury as evidence that Cross is unable to testify that the ZAP caused her chondrolysis. On December 27, 2007, Cross noted that it was without significant question that plaintiff's injuries were "directly and completely related to [the] injury she sustained" as a result of her May 2007 altercation. (Cross Dep. at 74:11-15.) Then on January 24, 2008, Cross wrote that plaintiff's chondrolysis may be due, in part, to the trauma she suffered, but also, in part, to the postoperative pain pump use. (Id. at 74:25-75:1-5.)

Plaintiff argues, citing Cross' testimony, that Cross' evolving opinion coincided with his developing knowledge of a potential connection between chondrolysis and pain pump usage. Eventually, after reviewing medical literature on the association between intra-articular pain pump use and chondrolysis, Cross developed the opinion that the pain pump was "the most probable cause" of plaintiff's chondrolysis. (Id. at 102:XX-XX-XXX:1-11.)

Cross testified that his determination regarding the cause of plaintiff's injury is "not science," but "an opinion." (Id. at 77:25.) Accordingly, defendants assert that Cross' testimony is inadmissible because he admitted that his opinion is "not science."

Plaintiff argues that, despite Cross' comment that his opinion is "not science," his opinion is based in fact and is the result of an appropriate method for reaching a specific causation opinion. Plaintiff asserts that Cross implemented the "differential diagnosis" method in developing his specific causation opinion. The differential diagnosis method is a "process of elimination approach," that "doctors routinely use to identify the most likely cause" of harm to a particular individual. In re Silicone Gel Breast Implants Products Liability Litigation, 318 F. Supp. 2d 879, 916 (C.D.Cal. 2004.) This method involves an expert compiling "a comprehensive list of potential causes and then engag[ing] in a process of elimination to reach the most likely cause." In re Hanford Nuclear Reservation Litigation, 534 F.3d 986, 1014 (9th Cir.2008). The differential diagnosis method has been accepted as scientifically valid methodology. Morin v. U.S., 244 Fed.Appx. 142, 143 (9th Cir.2007). However, the Ninth Circuit has determined that a district court is within its discretion in determining that an expert's specific causation opinion is not admissible where the expert does not show whether or how he applied this method. Id. at 144 (citing Kumho Tire Co. v. Carmichael, 526 U.S. 137, 141-42, 119 S. Ct. 1167, 143 L. Ed. 2d 238 (1999)).

Plaintiff submits evidence to support that Cross accounted for and ruled out other potential causes of chondrolysis. Specifically, plaintiff provides evidence that Cross considered, and ruled out, single injection of steroids and local anesthetics, use of a radiofrequency probe, infection, wear and tear, and trauma suffered during plaintiff's altercation.

For instance, Cross testified that he has "done hundreds and hundreds of steroid injections and Marcaine injections and ... never experienced any significant chondrolysis or seen any significant chondrolysis develop." (Cross Dep. at 94:18-22.) Cross testified any prior cartilage damage to plaintiff's shoulder was "very typical" and that he "didn't see anything unusual" in this regard. (Id. at 23:25-24:1-4.) Cross also testified that "there was just no clinical indication of an infection." (Id. at 92:18.)

Defendants point out that there are "multiple potential causes" for plaintiff's chondrolysis, and Cross acknowledged as much. (Id. at 95:1.) However, Cross' lack of certainty does not mean that his analysis is not factually based or methodically sound. See Primiano, 598 F.3d at 565. Furthermore, the court's role at this juncture is not to weigh conflicting evidence, it is to determine whether plaintiff has established an issue of material fact. See T.W. Elec. Service, Inc. v. Pacific Elec. Contractors Ass'n, 809 F.2d 626 (9th Cir.1987). Accordingly, plaintiff has submitted admissible evidence to demonstrate an issue of material fact whether the ZAP caused plaintiff's injury.

B. General Negligence, Negligent Products Liability and Failure to Warn

To sustain her claim of negligence or negligent products liability, plaintiff must establish that defendants owed her a cognizable legal duty. Gonzalez v. Autoliv ASP, Inc., 154 Cal. App. 4th 780, 793, 64 Cal. Rptr. 3d 908 (2007). As to her claim of negligent manufacture or design, plaintiff must show that defendants breached their duty of care. See id. As to plaintiff's claim of negligent failure to warn, she must show that defendants "did not warn of a particular risk for reasons which fell below the acceptable standard of care, i.e., what a reasonably prudent manufacturer would have warned about." Anderson v. Owens-Corning Fiberglas Corp., 53 Cal. 3d 987, 1002, 281 Cal. Rptr. 528, 810 P.2d 549 (1991). Accordingly, plaintiff must provide evidence that demonstrates defendants failed to give adequate warning of a risk associated with their product that defendants knew or should have known about at the time the product was distributed. See Phillippi v. Stryker Corp., No. 2:08-02445, 2010 WL 2650596 (E.D.Cal. July 1, 2010) (citing Brown v. Superior Court, 44 Cal. 3d 1049, 1057, 245 Cal. Rptr. 412, 751 P.2d 470 (1988) ("The manufacturer of a ... medical device, which is available only through the prescription of a licensed physician, must give adequate warnings of dangerous propensities in its product of which the manufacturer knows or should know in the application of existing scientific knowledge at the time of distribution.")).

Plaintiff asserts defendants breached their duty to plaintiff in two respects: (1) failing to investigate the medical literature which would have put defendants on notice of the risk of pain pumps causing chondrolysis; and (2) failing to conduct required testing of the ZAP which would have revealed that the ZAP carried a risk of harm to users.

1. Medical Literature

Plaintiff asserts that defendants were on notice to investigate a risk of chondrolysis based on existing medical literature on this issue. Plaintiff cites nine articles published between 1983 and 2006 in support of her assertion that existing medical literature indicated the potential risk of cartilage damage as a result of continuous anesthetic infusion. (See Pl.'s Opp'n to Defs.' Motion for Summary Judgment, filed Nov. 5, 2010 [Docket # 86], at 19-22.)

A similar argument was presented in Krumpelbeck v. Breg, Inc., No. 1:09-91, 759 F. Supp. 2d 958, 2010 WL 5475616 (S.D.Ohio. Dec. 27, 2010). In Krumpelbeck, the plaintiff claimed that she developed chondrolysis as a result of using the defendant's pain pump. Just as in this case, the plaintiff pointed to multiple published articles, which allegedly put defendant Breg, Inc. on notice that continuous exposure to local anesthetics could damage cartilage. (Id. at 968-69, at *7.) The Krumpelbeck court ultimately found that the plaintiff's cited articles were not sufficiently reliable to support her foreseeability argument. (Id. at 969-70, at *8.)

The Krumpelbeck court's decision was based in part on the plaintiff's failure to provide expert testimony to support her counsel's conclusion that relevant scientific literature existed at the time the pain pump was distributed. (Id. at 968-69, at *7.) Plaintiff's argument in this case fails for the same reason. Plaintiff's opposition to defendants' motion simply provides counsel's analysis of the literature. It does not include expert testimony regarding the scientific value of these nine reported studies. Furthermore, plaintiff's opposition does not include any expert testimony that draws any connection between the literature and a duty to warn.

Plaintiff's general assertion that the medical literature put defendants on notice of a risk of harm involves complex medical issues. As such, plaintiff's position must be supported by expert testimony. Jones v. Ortho Pharm. Corp., 163 Cal. App. 3d 396, 403, 209 Cal. Rptr. 456 (1985). The court cannot accept counsel's interpretation of the medical literature, counsel's unsupported determination that defendants had a duty to investigate the medical literature, nor counsel's unsupported determination that the medical literature triggered defendants' alleged duty to "further investigate [the] risk or at least warn of [the] risk." (Pl.'s Opp'n to Defs.' Motion for Summary Judgment at 22.)

For all of these reasons, plaintiff's citation to the nine scientific articles does not raise a material issue of fact that defendants were on notice of a risk of pain pumps causing chondrolysis.

In addition to the nine articles, plaintiff also submits testimony from Peggy Pence, Ph.D. ("Pence"), an expert on compliance with United States Food and Drug Administration ("FDA") regulations, regarding specific information plaintiff asserts put defendants on actual notice of a potential harm. Pence testified regarding certain, adverse event reports that Stryker, another pain pump manufacturer, submitted to the FDA. Pence testified that Stryker submitted to the FDA seven adverse event reports in September of 2005 regarding patients who developed chondrolysis after intra-articular infusion of a local anesthetic via a pain pump. (Pence Dep. at 136:19-25.) Pence further testified that defendants "should have been monitoring those reports," which would have "put them on alert that that might be an issue." (Id. at 139:5-18). Pence states that defendants had an obligation to monitor these reports based on "standard of practice." (Id. at 140:5).

Plaintiff cites to Hamilton v. Breg, Inc.^[16], a recent Southern District of Ohio case decided January 20, 2011, in support of her argument. (See Defs.' Second Supplemental Memorandum in Support of Their Motion for Summary Judgment, filed Jan. 24, 2011 [Docket # 108], Ex. 1.) However, contrary to plaintiff's suggestion, the court in Hamilton did not recognize a duty to warn based on a manufacturer's knowledge of adverse event reports. Indeed, the court recognized that generally, the state of the medical evidence that pain pumps could cause "chondrolysis was at best fragmentary at the [relevant] time." (Id. at 7.) The court emphasized that to the extent the Hamilton plaintiffs based their duty to warn claim on a theory of actual notice of the potential harm based on the state of the medical literature or any adverse event reports, they likely could not prevail. (Id.) Instead, the court held that it was only through the plaintiffs' claim that Breg, Inc. should have known of a potential harm as a result of a duty to conduct testing of the product, that plaintiffs stated a cognizable claim against the defendant. (Id. at 7-8.) This court discusses that latter theory of liability, also raised by plaintiff in this case, below. However, with respect to the issue of actual notice, Hamilton does not support plaintiff's theory that defendants had a duty to warn based on knowledge of prior adverse event reports.

Additionally, it is noteworthy that while Pence testifies that monitoring adverse incident reports would have complied with the "standard of practice," significantly, she did not testify that such monitoring is an FDA requirement. (Deposition of Pence ("Pence Dep."), filed Sep. 23, 2010 [Docket # 81], Ex. H at 140:5-8.) ("[T]hat's a standard of practice. I don't believe there's an actual statement to that effect in the [FDA] regulations.") Accordingly, plaintiff has not provided testimony that demonstrates a duty on defendants' part to monitor adverse event reports.

Additionally, even if Pence had shown that defendants had a legal duty to discover Stryker's adverse event reports, she does not testify that these seven incidents, out of millions of pain pumps in use, created a duty for defendants to conduct further testing. Without expert testimony or legal authority providing a reason to do so, this court will not bridge the gap between defendants being "on alert that [chondrolysis] might be an issue" and defendants having a legal duty to conduct further testing based on prior adverse event reports.

Accordingly, plaintiff has failed to raise a material issue of fact that the state of the medical literature, including certain, prior adverse incident reports, created a duty for defendants to warn of a risk of chondrolysis associated with intra-articular pain pump usage. Thus, with respect to plaintiff's argument that defendants should have known of a risk of chondrolysis based on existing scientific literature at the time the ZAP was distributed, defendants' motion is GRANTED.

2. FDA Approval

Plaintiff next asserts that defendants had a duty to warn because they should have known of the risk associated with intra-articular pain pump use as a result of testing required by the FDA. In support of this argument, Pence testified that defendants marketed the ZAP for intra-articular use, though defendants had only obtained FDA clearance for intra-operative use of its pain pumps. (Pence Dep. at 84:17-20). According to Pence, any intra-articular use of a ZAP, or use of a ZAP in the joint space or subacromial space, constitutes use that was not approved by the FDA. (Id. at 94:20-24; 81:8-9). In fact, as the FDA has specifically stated, "[t]he approved drug labels for local anesthetics do not include an indication for continuous intra-articular postoperative infusions.... The FDA has not cleared any infusion devices with an indication for use in intra-articular infusion of local anesthetics." (Pl.'s Exhibit 3, FDA Report, INFORMATION FOR HEALTHCARE PROFESSIONALS —CHONDROLYSIS REPORTED WITH CONTINUOUSLY INFUSED LOCAL ANESTHETICS, at 1.) Thus, according to Pence, defendants were knowingly marketing the ZAP for off-label use, or use that had not been cleared by the FDA, by marketing the ZAP for intra-articular use. (Pence Dep. at 86:12-22).

Pence further testified that defendants would have had to undergo a strenuous pre-market approval process if they attempted to obtain the proper FDA approval for the ZAP. (Pence's Expert Report, attached to Pl.'s Desig./Discl. Exp. Witnesses, filed Feb. 19, 2010 [Docket # 29], Attachment 4 at 11.) This process would have required defendants to demonstrate the device's safety and effectiveness through the presentation of valid scientific evidence. (Id.) Accordingly, the FDA would have required defendants to conduct controlled clinical trials, animal research and/or in vitro studies. (Id.)

As in the present case, the plaintiff in Krumpelbeck argued that the defendant promoted its pain pump for a specific use that had not been cleared by the FDA. Krumpelbeck, 759 F.Supp.2d at 970-71, 2010 WL 5475616 at *9. The plaintiff there argued that, had the defendant sought proper FDA clearance, they would have had to undergo the FDA's pre-market approval process which would have required extensive testing. (Id. at 972-73, at *11.) Further, the plaintiff argued, had the defendant conducted additional pre-marketing testing, "it would have foreseen that intra-articular infusion of anesthetic can cause chondrolysis." (Id. at 971, at *10.)

The Krumpelbeck court found that the plaintiff failed to show that defendant owed her a duty. This finding was based on the court's determination that the plaintiff failed to present evidence to establish that the FDA had not approved the defendant's product for its marketed purpose. (Id. at 972-73, at *11.) Based on this finding, the Krumpelbeck court found that plaintiff had not shown that defendant was required to engage in the pre-market approval process. (Id.)

The present case is distinguishable from Krumpelbeck in regards to expert testimony on FDA compliance and the pre-market approval process. Unlike in Krumpelbeck, here, plaintiff presents expert testimony to establish that defendants were marketing the ZAP for a purpose that was not approved by the FDA. Also unlike Krumpelbeck, plaintiff presents evidence to establish that defendants were required to undergo extensive testing in order to obtain FDA approval for the product's marketed purpose. (See generally Pence Dep. and Report.)

The present case is similarly distinguishable from Rodriguez v. Stryker Corp., No. 2:08-0124, 2011 WL 31462 (M.D.Tenn. Jan. 5, 2011), which defendants heavily rely upon. In this pain pump products liability case, the plaintiff argued that the defendant should have conducted further testing on its product because of the "FDA's repeated rejection of the proposed indication of the pain pump for use in the synovial cavity." Id. at *7. The Rodriguez court rejected this argument after finding that the evidence demonstrated that the pain pump was being used in a manner that was consistent with its FDA approved purpose. Id. ("[T]he FDA did (twice) clear the pump for use, without limitation, at the "intraoperative" site-with nothing to suggest that, if the "intraoperative" site was at the synovial cavity, the pump was not indicated for use.")

Unlike in Rodriguez, here, via Pence's testimony, plaintiff has submitted evidence to demonstrate that the ZAP was being used for a purpose that was not approved by the FDA. This critical, distinguishing factor renders the Rodriguez court's decision unpersuasive to this court regarding a duty to conduct further testing.

Having provided evidence that defendants should have tested the ZAP for intra-articular use, plaintiff must next provide evidence that such testing would have revealed an association between this use and the development of chondrolysis. In a rather conclusory manner, plaintiff states that, if defendants conducted this required testing, "these tests would have revealed then, as they have revealed now, that use of pain pumps in and around the joint space is not safe due to a risk of chondrolysis." (Pl.'s Opp'n. to Defs.' Motion for Summ. Judgment at 18.)

The court in Hamilton was presented with a similar argument. In Hamilton, the court determined that the plaintiff had demonstrated an issue of material fact that the defendant should have been aware of a risk of chondrolysis because it should have conducted extensive pain pump testing in order to comply with FDA requirements. (Defs.' Second Supplemental Memorandum in Support of Their Motion for Summary Judgment, filed Jan. 24, 2011[Docket # 108], Ex. 1 at 12.) This decision was based, in part, on the testimony of the plaintiff's FDA expert that "[b]ecause [defendant] failed to conduct appropriate ... testing, it remained uninformed about the seriousness of the risk it was exposing patients to with its products." (Id. at *10.) The Hamilton decision was also based on Dragoo's testimony that the defendant could have easily conducted his study, which Dragoo says proves that continuous local anesthetic exposure causes chondrolysis. (Id. at *10-*11.)^[17]

In the present case, plaintiff does not provide this sort of testimony, which directly links the required testing to a discovery of risk of harm. However, as discussed at length supra, plaintiff's general causation experts have testified that numerous studies have found an association between chondrolysis and intra-articular pain pump use. For instance, Wells reviewed prior studies and determined that "the association of postsurgical use of intra-articular pump with chondrolysis is well beyond what would occur by chance." (Wells Report I ¶ 9.) Similarly, Dragoo conducted his own study and reviewed prior studies, concluding "that continuous infusion of local anesthetics combined with epinephrine are chondrotoxic." (Dragoo Report ¶ 21, 24.)

While Wells' and Dragoo's review of prior studies is presented in the context of demonstrating that intra-articular pain pump use can cause chondrolysis, this same testimony supports plaintiff's argument that a similar study conducted by defendants would also have found that intra-articular pain pump use can cause chondrolysis. Though plaintiff does not provide an expert that explicitly states that defendants would have discovered a risk of harm if they conducted the testing required for FDA approval, plaintiff's experts have created an issue of material fact whether defendants negligently breached a duty to plaintiff by failing to warn of a risk that defendants should have known about.^[18] Accordingly, for reasons akin to those in Hamilton, defendants' motion for summary judgment in regards to a duty to warn based on plaintiff's "testing" theory is DENIED.

C. Strict Products Liability

"For the cause of action for strict products liability there is no necessity to show duty or breach of duty but only that the product was defective and that the injury to the plaintiff was caused by the defective condition." Gonzalez v. Autoliv ASP, Inc., 154 Cal. App. 4th 780, 793, 64 Cal. Rptr. 3d 908 (2007). "The concept of strict liability imposes legal responsibility for injury upon the manufacturer of a product without proof of negligence." Johnson v. Honeywell Intern. Inc., 179 Cal. App. 4th 549, 558, 101 Cal. Rptr. 3d 726 (2009).

"This doctrine of strict liability extends to products which have design defects, manufacturing defects, or `warning defects.'" Id. "A product may be found defective in design if the plaintiff demonstrates that the product failed to perform as safely as an ordinary consumer would expect when used in an intended or reasonably foreseeable manner." Id. Also, "a product may be found defective in design, even if it satisfies ordinary consumer expectations, if through hindsight the jury determines that the product's design embodies `excessive preventable danger,' or, in other words, if the jury finds that the risk of danger inherent in the challenged design outweighs the benefits of such design." Id.

Had defendants been able to demonstrate the absence of a genuine issue of fact as to general or specific causation, plaintiff's strict liability design defect claim would not survive. However, because the court has found that a genuine issue of material fact exists as to both general and specific causation, plaintiff's strict liability claims remains viable.

Neither party's briefs address the test for a strict liability design defect claim. Because the moving party bears the initial burden of demonstrating the absence of a genuine issue of fact, and defendants fail to specifically address this issue in their motion for summary judgment, plaintiff had no corresponding obligation to address the issue. Accordingly plaintiff's strict liability design defect claim survives.

Nevertheless, the court notes that had defendants specifically addressed the consumer expectations test on its motion for summary judgment, plaintiff has submitted sufficient evidence to demonstrate a material issue of genuine fact as to whether, in hindsight, ZAP's design carries with it a risk of inherent danger that outweighs the benefits of its design. Id. Further, plaintiff has submitted evidence to raise a genuine issue of material fact whether the ZAP failed to perform as safely as an ordinary consumer would expect when used in an intended manner. Id. Thus, defendants' motion for summary judgment as to plaintiff's strict liability claim is DENIED.

CONCLUSION

For the foregoing reasons, the court DENIES defendants' motion to exclude Wells' testimony. Defendants' motion for summary judgement is DENIED in part and GRANTED in part. Defendants' motion for summary judgment is:

(1) DENIED as to plaintiff's claim that defendants' ZAP was the specific cause of her chondrolysis.

(2) DENIED as to plaintiff's claim that defendants' ZAP can generally cause chondrolysis.

(3) GRANTED with respect to plaintiff's argument that defendants should have known of a risk of chondrolysis based on existing scientific literature at the time the ZAP was distributed.

(4) DENIED as to plaintiff's argument that defendants should have known of a risk of chondrolysis because they were required to conduct extensive testing in order to comply with FDA requirements.

(5) DENIED as to both the "excessive preventable danger test" and the "consumer expectations test" for strict products liability.

Accordingly, plaintiff's general negligence, negligent products liability, and strict products liability claims remain pending.

IT IS SO ORDERED.

NOTES

[1] Because oral argument will not be of material assistance, the court orders these matters submitted on the briefs. E.D. Cal. L.R. 230(g).

[2] Unless otherwise noted, the facts cited herein are undisputed.

[3] A ZAP is a medical device called a "pain pump," marketed and distributed by defendants. (Defs.' Reply and Objections to Pl.'s Response to Separate Statement of Undisputed Material Facts in Support of Motion for Summary Judgment ["DUF"], filed Nov. 12, 2010 [Docket #95], ¶ 1.)

[4] While the parties generally discuss the topics covered in this Overview section, they do so largely without citation. Accordingly, the court gathered the general background information for this section from McClellan v. I-Flow Corp., 710 F. Supp. 2d 1092 (D.Or.2010). The plaintiffs in McClellan claimed they developed glenohumeral chondrolysis after pain pumps were used to administer local anesthetics during and after arthroscopic surgery. The court notes that this background information is provided strictly to assist the reader's general understanding. This information is not dispositive of any issue on this motion.

[5] DUF ¶ 4 states that the first surgery occurred on May 10, 2007. However, both parties cite Cross' testimony stating that the first surgery occurred on May 18, 2007.

[6] Marcaine is a trade name for the drug Bupivicaine.

[7] DUF ¶ 7 states that the ZAP was inserted into plaintiff's shoulder after the surgery, while DUF ¶ 8 states that the ZAP was inserted during the surgery.

[8] This was the first instance of insertion of a ZAP directly into the glenohumeral joint. The first ZAP was inserted into the subacromial space of plaintiff's left shoulder.

[9] At Cross' deposition, defendants objected to the following question from plaintiff's counsel as providing an incomplete hypothetical: "Based on your review of the literature that you have seen concerning this phenomenon of the development of chondrolysis following the use of intra-articular pain pumps and based on your many years of experience as an orthopedic surgeon and your clinical experience with this particular patient, is it your opinion that it is more likely than not that the use of the pain pump contributed to the chondrolysis that [plaintiff] suffered from?" This objection is overruled as Cross is qualified to provide an opinion, utilizing his own review of the literature, his own orthopedic experience, and his own clinical experience with plaintiff. Fed R. Evid. 702.

[10] B.P. Wiater et al., A Study of 375 Shoulder Arthroscopic Surgeries in the Practice of an Individual Surgeon.

[11] Defendants state that Wells is "at most, qualified to quantitatively determine statistical correlations between variables within a set of data." (Defs.' Motion to Exclude the Testimony of Wells, filed Sept. 23, 2010 [Docket # 73], at 4-5.) Wells has published well over 100 scholarly articles that required him to "quantitatively determine statistical correlations within a set of data." Accordingly, Wells' credentials in this arena are beyond reasonable challenge.

[12] In In re Welding Fume Products, Wells testified regarding an association between exposure to welding fumes and the development of neurological damage.

[13] The court in McClellan ended up excluding Wells' testimony after finding that it had little relevance in that case. This was based on the court's opinion that "the degree of association between continuous infusion and chondrolysis reflected in the Hansen [] study is so high that expert analysis is not necessary to infer a causal connection-it is `common sense.'" McClellan, 710 F.Supp.2d at 1136. This court does not share the opinion that general causation is so obvious that the fact finder would not benefit from the guidance provided by Wells' analysis.

[14] Defendants' motion for summary judgment also argues that plaintiff lacks sufficient evidence to demonstrate that defendants breached a specific duty of care to her.

[15] Defendants do not argue that plaintiff cannot support the other elements of her claims.

[16] Presently, there is not a published citation for this case. The court therefore cites to the case attached as an Exhibit to defendants' supplemental brief.

[17] Defendants argue that Hamilton is not useful because the court's analysis "lacks legal support." However, the Hamilton court was presented with unique evidence that rendered prior decisions in various pain pump cases inapplicable. Unlike previous cases, the evidence in Hamilton demonstrated that the defendant was promoting its product for a "new and unapproved use." (Defs.' Second Supplemental Memorandum in Support of Their Motion for Summary Judgment, filed Jan. 24, 2011, [Docket #108], Ex. 1 at *10.) Thus, the Hamilton court would have been misguided had it relied upon cases that presented contradictory evidence.

[18] The expert testimony in this case distinguishes it from Phillippi. The Phillippi court found that the "[p]laintiff offers no admissible evidence as to what testing would have been conducted ... and what the results of that unknown testing would have shown." Phillippi, 2010 WL 2650596 at *2. Plaintiff in this case offers admissible evidence regarding both of these factors. Pence's testimony discusses the type of testing that would have been conducted in order to obtain FDA approval, and Wells' and Dragoo's general causation testimony discusses what the results of this testing would have shown.