Opinion
No. 02A03-9907-CV-265
June 27, 2000.
APPEAL FROM THE ALLEN SUPERIOR COURT, The Honorable Paul D. Mathias, Judge, Cause No. 02D01-9410-CP-1405.
Marianne L. Vorhees, Charles R. Clark, Beasley Gilkison LLP, Muncie, Indiana, Attorneys for Appellants.
Robert T. Keen, Jr., Diana C. Bauer, Miller, Carson, Boxberger Murphy LLP, Fort Wayne, Indiana, Attorneys for Appelle Meadow Acres Manufactured Housing, Inc.,
Darla S. Brown, Kelley, Belcher Brown, Bloomington, Indiana Joseph G. Eaton, Michael D. Moon, Jr., Barnes Thornburg Indianapolis, Indiana, Attorneys for Appellees Redman Homes, Inc. and Redman Industries, Inc.,
OPINION
Diana Wallace, Doug Wallace, Lora Wallace, Andru Wallace, Brandin Wallace, and Evin Wallace (collectively, the Wallaces) appeal the trial court's order granting summary judgment in favor of Appellees, Meadow Acres Manufactured Housing, Inc. (Meadow Acres), Redman Homes, Inc., and Redman Industries, Inc. (Redman). However, because the Wallaces concede that they are unable to prove causation with regard to their cause of action without certain expert testimony on formaldehyde levels which was excluded by the trial court, essentially, they are challenging the trial court's conclusion that their expert's testimony is unreliable. Because the trial court did not abuse its discretion in excluding the expert's testimony, we affirm.
Facts and Procedural History
We heard oral argument in this case on March 27, 2000, in Indianapolis.
Redman manufactures mobile homes. Meadow Acres sells Redman's mobile homes at retail. On March 23, 1989, Doug and Diana Wallace purchased a Redman mobile home from Meadow Acres. After the Wallaces moved into the mobile home, they began to develop health problems.
On December 1, 1992, Diana Wallace contacted the Indiana State Department of Health, which collected an air sample from the Wallaces' home to analyze the formaldehyde concentrations. Subsequent testing revealed a formaldehyde concentration level of 0.05 parts per million (ppm) in the kitchen. This level was below the occupational health permissible exposure limit for formaldehyde of 1.0 ppm, an "action level" of 0.5 ppm, and the suggested indoor air guideline for formaldehyde of 0.1 ppm.
In February of 1994, the Wallaces moved out of their mobile home. From February 1994 to October 8, 1994, a period of approximately eight months, the Wallace home was completely closed. There was no airflow into or out of the home.
The Wallaces filed this lawsuit against Meadow Acres and Redman on October 7, 1994. On October 8, 1994, Thaddeus J. Godish, Ph.D., a professor at Ball State University in the Natural Resources and Environmental Management Department who had studied formaldehyde for twenty years, tested the air in the home by obtaining ambient air samples. Record at 404. Before performing the test, he instructed Diana Wallace to keep the doors and windows in the manufactured home closed. He also instructed her to heat the home to 78 degrees Fahrenheit in order to simulate the high end of the temperature range that would usually exist in a manufactured home in Indiana in October. On the day of testing, the relative humidity in the Wallace home was 70%. At the time of the test, the home was heated to 71 1/2 degrees Fahrenheit. After the air samples were taken, Dr. Godish recorded the following levels of formaldehyde in different areas of the home:
Kitchen 0.105 ppm
Master Bedroom 0.114 ppm
East Bedroom 0.111 ppm
Inside Kitchen Cabinet 0.230 ppm
In order to determine the formaldehyde levels that would have been present at 78 degrees Fahrenheit on October 8, 1994, Dr. Godish applied a formula called the "Berge equation." Using this formula, he determined that the formaldehyde levels would have been "as high as .17 parts per million" at 78 degrees Fahrenheit. Record at 157. Dr. Godish then calculated what he believed the formaldehyde levels would have been when the Wallaces purchased the home in March 1989 by a process of extrapolation. He based his calculations on the probability that the formaldehyde levels had decreased twice during that time period (decay rate). Dr. Godish's calculation revealed a formaldehyde concentration in the Wallace home exceeding .4 ppm in March of 1989.
On April 30, 1999, Redman filed a Motion to Exclude Testimony of Plaintiff's Expert, Dr. Godish. In their motion, the defendants contended that Dr. Godish was not qualified to render ultimate expert opinions on medical causation. They further argued that his methods and opinions were: (1) unreliable, (2) not based on sound science, (3) not generally accepted in the relevant scientific community, and (4) not grounded in scientific knowledge, as required by Ind. Evidence Rule 702 and Daubert v. Merrell Dow Pharm., Inc., 509 U.S. 579 (1993). Record at 243. The Wallaces filed a memorandum in opposition on May 13, 1999. During a hearing on the motion, Dr. Godish testified regarding the different levels of formaldehyde present in the home on October 8, 1994, and the levels he believed were present in March 1989 based on the Berge equation and the process of extrapolation.
Following the hearing, the trial court ruled that although Dr. Godish was permitted to testify to the formaldehyde levels in the home on October 8, 1994, he was precluded from testifying regarding either the formaldehyde levels determined using the Berge equation or the estimated levels found using the extrapolation method. As the Wallaces conceded they were unable to prove causation without Dr. Godish's testimony, the trial court granted summary judgment in favor of the defendants.
Discussion and Decision
I. Standard of Review
The trial court excluded the results Dr. Godish arrived at using the Berge equation and the extrapolation method because the scientific principles the calculations rested upon were unreliable and therefore, inadmissible under Evid. R. 702(b). The Wallaces assert the trial court erred in excluding this evidence.
The proponent of expert testimony has the burden of proving the reliability of the principles upon which the expert's testimony rests. McGrew v. State, 682 N.E.2d 1289, 1290 (Ind. 1997). The burden of proving the reliability of Dr. Godish's testimony in this case rests upon the Wallaces. In determining whether expert testimony is reliable, the trial court assumes the function of "gatekeeper" in ensuring that an expert's testimony rests on a reliable foundation and is relevant to the issue at hand. Hottinger v. Trugreen Corp., 665 N.E.2d 593, 596 (Ind.Ct.App. 1996).
Deciding whether expert testimony is reliable and therefore admissible is a matter within the sound discretion of the trial court. Ford Motor Co. v. Ammerman, 705 N.E.2d 539, 550 (Ind.Ct.App. 1999), trans. denied. The trial court determines whether the expert testimony has a sufficiently reliable basis so that the expert's testimony will "assist the trier of fact to understand the evidence or determine a fact in issue." Evid. R. 702(a); Ford Motor Co., 705 N.E.2d at 550. We reverse a trial court's decision to exclude evidence only if that decision "is clearly against the logic and effect of the facts and circumstances before the Court, or the reasonable, probable and actual deductions to be drawn therefrom." Burkett v. State, 691 N.E.2d 1241, 1245 (Ind.Ct.App. 1998), trans. denied.
Here, the trial court entered findings of fact and conclusions of law pursuant to Rule 52(A) of the Indiana Rules of Trial Procedure. Therefore, in reviewing the exercise of the court's discretion, we must apply a two-tiered standard of review. We decide whether the "evidence supports the findings and the findings support the judgment." Chidester v. City of Hobart, 631 N.E.2d 908, 910 (Ind. 1994). We construe the findings liberally in support of the judgment. OVRS Acquisition Corp. v. Community Health Servs., Inc., 657 N.E.2d 117, 124 (Ind.Ct.App. 1995), reh'g denied, trans. denied. The findings are erroneous only when a review of the record indicates to the court that a mistake has been made. Id. When a judgment is not supported by the findings of fact and the conclusions, it is clearly erroneous. DeHaan v. DeHaan, 572 N.E.2d 1315, 1320 (Ind.Ct.App. 1991), reh'g denied, trans. denied. The court must consider only the evidence favorable to the judgment and must draw all reasonable inferences therefrom. Id.
II. Rule of Evidence 702
Indiana Rule of Evidence 702 provides:
(a) If scientific, technical, or other specialized knowledge will assist the trier of fact to understand the evidence or to determine a fact in issue, a witness qualified as an expert by knowledge, skill, experience, training or education may testify thereto in the form of an opinion or otherwise.
(b) Expert scientific testimony is admissible only if the court is satisfied that the scientific principles upon which the expert testimony rests are reliable.
Under Evid. R. 702, evidence which is not first established to be reliable under 702(b) will not "assist" the judge or jury pursuant to 702(a). Therefore, reliability is a threshold issue for a Evid. R. 702 analysis of scientific evidence.
Although Federal Rule of Evidence 702, which is the basis for the Supreme Court's decision in Daubert, and Evid. R. 702 are not identical:
[t]he concerns driving Daubert coincide with the express requirement of Indiana Rule of Evidence 702(b) that the trial court be satisfied of the reliability of the scientific principles involved. Thus, although not binding upon the determination of state evidentiary law issues, the federal evidence law of Daubert and its progeny is helpful to the bench and bar in applying Indiana Rule of Evidence 702(b).
Steward v. State, 652 N.E.2d 490, 498 (Ind. 1995).
In deciding whether scientific evidence is reliable, the trial court must determine whether it appears sufficiently valid or, in other words, trustworthy so that it assists the jury. Daubert, 509 U.S. at 590 n. 9. In order to make such a determination, the trial court must "make a preliminary assessment of whether the reasoning or methodology underlying the testimony is scientifically valid and whether that reasoning or methodology properly can be applied to the facts in issue." Hottinger, 665 N.E.2d at 596. Certain factors may be considered to determine the reliability of expert testimony: (1) whether the theory or technique at issue can be and has been tested; (2) whether the theory or technique has been subjected to peer review and publication; (3) whether there is a known or potential rate of error; (4) whether standards controlling the technique's operation exist and are maintained; and (5) whether the technique is generally accepted within the relevant scientific community. Daubert, 509 U.S. at 593-94. While these factors may be useful, "there is no specific `test' or set of `prongs' which must be considered in order to satisfy Indiana Evidence Rule 702(b)." McGrew, 682 N.E.2d at 1292. See also Kumho Tire Co., Ltd. v. Carmichael, 526 U.S. 137, 151 (1999) (asserting that the list of Daubert factors is "helpful, not definitive").
III. The Berge Equation
Our initial inquiry focuses on the trial court's exclusion of Dr. Godish's calculations reached using the Berge equation. With regard to the Berge equation, Dr. Godish was prepared to testify to what the formaldehyde level would have been if the home had been heated to 78 degrees Fahrenheit on October 8, 1994, instead of the actual temperature of the home: 71 1/2 degrees Fahrenheit.
The Wallaces contend that the trial court improperly required absolute scientific certainty rather than reliability. They argue that the trial court should have given greater weight to Dr. Godish's education, training, experience, and background. Of particular importance, according to the Wallaces, is the fact that Dr. Godish researched and published extensively in the area of formaldehyde and had previously tested numerous homes for both litigation and research purposes.
The Wallaces emphasize that Dr. Godish published in a peer-reviewed journal an article on the Berge equation which stated, "the Berge equation is a reliable predictor of formaldehyde levels measured at one set of environmental conditions and standardized to another." Thad Godish Jerome Rouch, An Assessment of the Berge Equation Applied to Formaldehyde Measurements Under Controlled Conditions of Temperature and Humidity in a Mobile Home, 35 J. Air Pollution Control Ass'n 1186, 1187 (1985). They also point to the fact that his study has not been challenged and that the scientific community accepts his use of the Berge equation.
In this case, the court made the following findings with regard to the Berge equation:
21. Dr. Godish made no appreciable attempt to stabilize or "precondition" ambient air temperature or humidity in the Wallace manufactured home prior to obtaining these ambient air measurements, and obtained only one set of measurements on one occasion.
22. In contrast, Dr. Godish's own published work and the HUD standards employ protocol requiring preconditioning of ambient air temperature and humidity for a period of seven days prior to testing of the regulated components, plywood and particleboard. Godish and Rauch, "An Assessment of the Berge Equation Applied to Formaldehyde Measurements under Controlled Conditions of Temperature and Humidity in a Mobile Home," 35 Journal of the Air Pollution Control Association 1186 (1985); 24 C.F.R. § 3280.502; 49 Fed. Reg. 32001 (1984).
23. HUD standards also require a 0.5 air change per hour during the preconditioning and testing process. 24 C.F.R. § 3280.406. (ASTM E-13333-90 standards except as modified by § 3280.406); See also, 49 Fed. Reg. 32001 (1984).
24. In Dr. Godish's own published work he also obtained multiple measurements for nine different combinations of ambient air temperature and humidity in multiple manufactured homes over a period of seven hours per day for a period of seven days after preconditioning, in order to report average measurements. Godish and Rauch, "An Assessment . . .," supra.
25. In this case Dr. Godish then seeks to apply what he refers to as the Berge equation to "correct" from the actual ambient air formaldehyde concentration levels he found at 71-1/2° F in the Wallace manufactured home to the levels that would be expected at 78° F, which is within the large chamber temperature standard (77° F +/- 2° F) set by HUD for the laboratory testing of component plywood and particleboard materials used to construct manufactured houses.
26. After such "correction" by the Berge equation, Dr. Godish testified that the ambient air formaldehyde concentration level in the Wallace manufactured home on October 8, 1994 when he conducted his testing would have been 0.170 ppm, if the temperature inside the home had been 78° F.
27. Dr. Godish testified, and has written, that the Berge equation by itself has a rate of error of a minimum of +/-12% Godish and Rauch, "An Assessment . . .," supra.
Record at 374-76. Based on these findings, the court concluded that Dr. Godish's use of the Berge equation was unreliable. We agree.
The court based its conclusion on several factors. First, there is no general acceptance in the scientific community for use of the Berge equation to "correct" ambient air formaldehyde level measurements in a manufactured home and certainly not when only one sample is analyzed. Second, his theory cannot be empirically tested because he relies on numerous variables which were not preconditioned. Third, his theory and methodology lack substantial peer review. Finally, his theory and methodology have a substantial and unquantifiable rate of error.
A review of the record indicates that the evidence supports the trial court's findings and conclusion to exclude Dr. Godish's testimony regarding the Berge equation. First, there is no general acceptance within the scientific community supporting the use of the Berge equation to correct singular ambient air formaldehyde levels as Dr. Godish did here. Dr. Godish's testimony supports this finding. At the Daubert hearing, Dr. Godish was asked:
Now, are you aware of anybody else who's taken the Berge equation and who's put it to the same test that you did in this article? In other words, are you aware of anybody else who's used the Berge equation, which is used to measure particleboard in a chamber test, and has applied that to manufactured homes?
Record at 659. Dr. Godish replied, "I don't know, no. I'm not aware." Id. Further, William M. Pierce, a defense expert, testified that:
[The Berge equation] is an equation describing some data. . . . [The Berge equation]'s an analogy to something that it's a real physical property of some things, but it does not you can get any number you want. If that were not so, how can we explain how two years earlier the Department of Health got 0.05? In fact, if you look during the day, there's what we call diurnal variation. It matters what time of day you measure. So if you were to tell me I measured multiple times in this same building over a long period of time under, you know, controlled and understood conditions, that would be one thing, and you could define a relationship over time with that building. But for you to take one measurement and apply an average analysis, I disagree with that.
Record at 210.
Second, when Dr. Godish tested the Wallace home, he did not control variables that affect formaldehyde results. Record at 741-42. In his published article on the Berge equation, Dr. Godish preconditioned the home's environmental conditions including humidity and ventilation. However, when Godish tested the Wallace home, he did not precondition those variables. Consequently, the Wallace results cannot be empirically tested.
Third, Dr. Godish's methodology, using the Berge equation to correct ambient air formaldehyde levels in a manufactured home, has not been subject to peer review. While Dr. Godish's article on the Berge equation was published in a peer-reviewed journal, Dr. Godish did not apply the Berge equation to the Wallace home as he presented it in his article. In his article, he tested the same mobile home at several different measurements over a period of seven days. Record at 655. To reach his conclusions in this case, however, he took one test at the Wallaces' home on one day. Therefore, the methodology presented for peer review in Dr. Godish's article was different from that used by him in testing the Wallace home.
Finally, the Berge equation has a substantial and unquantifiable rate of error. Dr. Godish testified as follows:
Q: And didn't you testify in your deposition that the error rate for the Berge equation, as you determined it to be, was plus or minus fifteen percent (15%) at a minimum?
A: Plus or minus twelve percent (12%) at a minimum.
Q: At a minimum, right?
A: Well, that's within a ninety-five percent (95%) confidence interval, yes. So, ninety-five percent (95%) of the time you would expect it to be anywhere from zero plus twelve percent (12%), zero minus twelve percent (12%).
Q: But you used the term, the error rate as being at a minimum plus or minus. In fact, [in] your deposition, [you indicated] a plus or minus fifteen percent (15%), didn't you?
A: If I did, it's twelve percent (12%) in the paper.
Q: But that, if I understand your testimony correctly, whether it's twelve percent (12%) or fifteen percent (15%), it could be higher in an individual test setting?
A: It could be, yes.
Record at 751.
In this case, the trial court performed its function as "gatekeeper" by excluding the testimony regarding the Berge equation. As the Berge equation was found to be an unreliable method for determining formaldehyde levels, it was properly excluded.
IV. Extrapolation
The Wallaces argue further that the trial court abused its discretion when it excluded Dr. Godish's testimony regarding the extrapolation methodology. Dr. Godish was prepared to testify regarding the levels of formaldehyde in March of 1989, the date of purchase, using this method. The Wallaces contend that the trial court placed unreasonable standards for scientific certainty on Dr. Godish's testimony. They also contend that Dr. Godish's research, knowledge, scientific literature, and information demonstrated that he was qualified to present his opinion and made his testimony reliable.
The trial court made the following findings regarding Dr. Godish's use of the extrapolation method:
28. Dr. Godish then seeks to use the October 8, 1994 formaldehyde concentration levels, "corrected" by the Berge equation, as the basis to extrapolate backward in time to determine the ambient air formaldehyde concentrations in the Wallace manufactured home when purchased in 1989.
29. In order to accomplish this extrapolation backward in time, Dr. Godish utilizes an estimated half-life or decay rate value for formaldehyde.
30. According to Dr. Godish, the half-life of formaldehyde is the time period that it takes for measured ambient air formaldehyde level to decrease by one-half.
31. In Dr. Godish's opinion, formaldehyde has an estimated half-life value of 21 months and a second half-life value between two and four years.
32. Dr. Godish also admits that "half-life" is a confusing misnomer, and that the concept of a decay rate is more accurate.
33. Dr. Godish admits that his extrapolation methodology represents a worst case living scenario for the approximate time when the Wallaces moved into their manufactured home in 1989.
34. While Dr. Godish uses estimated half-life or decay rate values to arrive at his opinions in this case, he admits that formaldehyde has no definite half-life or decay rate and that there are no scientific studies which establish a definite half-life or decay rate for formaldehyde. See, Godish, "Residential Formaldehyde Contamination: Sources and Levels," 2 Comments Toxicology 115, 126-132.
35. Dr. Godish admits that he is unable to identify any peer-reviewed articles or journal articles reporting an established half-life or definite decay rate for formaldehyde.
36. Dr. Godish admits that there is no recognized, standard testing method for evaluating the alleged half-life or decay rate for formaldehyde. Id.
Record at 376-77. Based on these findings, the court concluded that Dr. Godish's testimony concerning the extrapolation process was also unreliable. Specifically, the trial court concluded that Dr. Godish's extrapolation of ambient air formaldehyde levels at earlier points in time through use of an estimated half-life or decay rate was not accepted within the relevant scientific community.
A review of the record indicates that the evidence supports the trial court's findings and conclusions regarding the extrapolation process used by Dr. Godish. First, the record shows that Dr. Godish's extrapolation methodology has not been subjected to peer review. Record at 214, 323-24. In particular, Kenneth Troutman, an industrial hygienist, testified that, "[Dr. Godish] uses — in his deposition he gives a[n] explanation of half-life, which is not well accepted in industrial hygiene, and he makes a — he doesn't even make a calculation . . . ." Record at 214. While this alone may not be dispositive, peer review is a relevant consideration to assess scientific validity. See Daubert, 509 U.S. at 593-94. Second, Dr. Godish concedes that there is no quantifiable rate of error for the extrapolation method used in his calculations. Dr. Godish testified, "Well, the error rate in the — there are error rates in the scientific literature, but I think that obviously the way that I have made my estimate, the error is, the error is unquantifiable and as a consequence I cannot give you an error rate." Record at 310. Finally, while the scientific community acknowledges that formaldehyde decays over time, the exact rate of decay is disputed and ultimately unknown. Record at 194, 209. During William Pierce's testimony, the following exchange occurred:
Q: Formaldehyde has a half-life?
A: No. I think that that is it's an attempt to explain what happens over time. And you're talking when you say that formaldehyde has a half-life . . . you mean that the concentration that we find in a building decays with a certain rate?
Q: That's what I understand.
A: Not that well, it's not a predictable rate.
Record at 209. The studies examining the decay rate of formaldehyde have used various methods and have produced different results. Record at 602. Dr. Godish testified to the following:
Q: There have been several studies done of decay, correct?
A: That's correct.
Q: And as far as the scientific community is concerned, how would you describe these results?
A: The results are quite variable . . . the rate of decay depends on the kind of structures that have been evaluated and the particular environments. . . . So, it there is a spectrum of results.
Record at 602.
Essentially, the extrapolation method used by Dr. Godish relies on the questionable assumption that the rate of formaldehyde's decay has been properly calculated. This assumption is not supported by any reliable scientific methods or the reliable application of any valid theory. In addition, the facts from the record support the trial court's findings and ultimate conclusion that Dr. Godish's extrapolation procedure is unreliable and, therefore, excludable under Evid. R. 702.
To the extent that the Wallaces argue that other findings by the trial court establish an abuse of discretion, we find that these findings are irrelevant to our conclusion that the expert's testimony as a whole is unreliable and, consequently, excludable. Even if the trial court erred in considering the HUD regulations, failed to note that Dr. Godish's findings were clearly within the scientifically acceptable range for error and mistakenly found that the testing was not conducted under normal living conditions, the testimony regarding the Berge equation and extrapolation method was properly excluded.
Our decision is consistent with a Third Circuit Court of Appeals decision to exclude testimony of an industrial hygienist who relied on a back-extrapolation methodology in Heller v. Shaw Indus. , Inc., 167 F.3d 146 (3rd Cir. 1999). In Heller, expert testimony was excluded because it was based on a speculative decay rate which was not supported by reliable scientific method. Id. at 162. Similar to the expert testimony in Heller, Dr. Godish's extrapolation procedure is not supported by a scientifically reliable decay rate. Accordingly, the trial court properly excluded it.
Lastly, we address the Wallaces' argument that Dr. Godish's expertise in the field of formaldehyde should be given great weight when determining if the tests he used were reliable. Being an expert in any given field, as Dr. Godish appears to be in the field of formaldehyde research, does not necessarily render the theory and methodology employed by the expert reliable. An expert's theory and methodology must provide trustworthy and reliable information that will help the trier of fact. While it is clearly helpful that Dr. Godish has previously researched and published in this area for litigation and research purposes, this factor alone is not determinative. To be admissible, an expert's theory and methodology must be supported by reliable principles independent of his expertise in the field. See Lytle v. Ford Motor Co., 696 N.E.2d 465, 474 (Ind.Ct.App. 1998) (finding that while consideration of an expert's exceptional qualifications in determining reliability of his scientific testimony is not improper, credentials alone are not conclusive evidence of reliability).
Here, the Berge equation and the extrapolation method fail to provide the reliability required by Evid. R. 702(b). As a result, they also fail to "assist the trier of fact to understand the evidence or to determine a fact in issue . . ." as required by Evid. R. 702(a). Accordingly, the trial court properly excluded Dr. Godish's testimony regarding the formaldehyde levels reached using the Berge equation and the extrapolation method.
Judgment affirmed.
FRIEDLANDER, J., and DARDEN, J., concur.