Order Michigan Supreme Court
Lansing, Michigan
December 22, 2010 Marilyn Kelly,
Chief Justice
Michael F. Cavanagh
Maura D. Corrigan
139860 Robert P. Young, Jr.
Stephen J. Markman
Diane M. Hathaway
RAQUEL ROBELIN, Conservator, for Alton Thomas Davis,
Justices
TEIJA McCALL,
Plaintiff-Appellee,
v SC: 139860
COA: 279780
Kent CC: 04-010444-NH
SPECTRUM HEALTH HOSPITALS, d/b/a
SPECTRUM HEALTH-EAST CAMPUS,
JOHN HARTMANN, M.D., and
ADVANTAGE HEALTH PHYSICIANS, P.C.,
Defendants-Appellants.
_________________________________________/
On November 4, 2010, the Court heard oral argument on the application for leave
to appeal the September 10, 2009 judgment of the Court of Appeals. On order of the
Court, the application is again considered, and it is DENIED, there being no majority in
favor of granting leave to appeal or taking other action.
KELLY, C.J. (concurring).
I concur with the order denying leave to appeal I do not believe that the lower
courts misapplied the legal standards governing the admissibility of expert testimony or
that allowing the testimony of plaintiff’s expert was improper.
FACTS AND PROCEEDINGS
Plaintiff’s daughter, McCall, was born at defendant Spectrum Hospital in 2001.
Shortly after her birth, she suffered a neonatal stroke. As a result, she is unable to walk
without assistance and has had significant developmental delays in language and speech.
Plaintiff filed this medical malpractice action in 2004, alleging that McCall
suffered hypoxia, which caused the stroke, which in turn caused neurological disabilities.
The parties dispute when the stroke occurred and whether the health care professionals
missed indicators of it.
2
In 2008, defendants moved to strike the testimony of Dr. Ronald Gabriel,
plaintiff’s causation expert. The trial court held a Daubert1 hearing on the issue but
ultimately denied the motion. The Court of Appeals denied interlocutory leave to appeal
and denied defendants’ motion for a peremptory reversal. Defendants appealed here, and
we remanded the case to the Court of Appeals for consideration as on leave granted.2
In 2009, on remand, the Court of Appeals affirmed the trial court’s ruling denying
defendants’ motion to exclude Dr. Gabriel’s testimony.3 The panel ruled that the trial
court had been correct in concluding that Dr. Gabriel simply used the process of
elimination, a time-honored and well-tested means of reaching a conclusion.
In 2010, defendants again sought leave to appeal in this Court. We heard oral
argument on the application.4
ANALYSIS
MCL 600.2955(1) and MRE 702 require that an expert’s opinion be “reliable,”
“assist the trier of fact,” and be “the product of reliable principles and methods.”5 The
Court of Appeals has held that, as long as the basic methodology and principles utilized
by an expert are “sound and create a trustworthy foundation for the conclusion reached,
the expert testimony is admissible no matter how novel.”6
In this case, the trial court did not use the magic word “reliable” or the phrase
“will assist the trier of fact.” However, the trial court conducted a lengthy two-day
Daubert hearing to evaluate Dr. Gabriel’s proposed testimony. The court concluded that
the testimony has “a foundation in fact and in science, and is based, at least in reasonable
degree, upon demonstrable data.”
In that ruling, the court discussed its “gatekeeping function” under Daubert and
addressed each of the factors in MCL 600.2955(1). It certainly does not appear to me, as
1
Daubert v Merrell Dow Pharmaceuticals Inc, 509 US 579; 113 S Ct 2786 (1993).
2
Robelin v Spectrum Health Hospitals, 482 Mich 985 (2008).
3
Robelin v Spectrum Health Hospitals, unpublished opinion per curiam of the Court of
Appeals, issued September 10, 2009 (Docket No. 279780).
4
Robelin v Spectrum Health Hospitals, 486 Mich 851 (2010).
5
MCL 600.2955(1); MRE 702.
6
Nelson v American Sterilizer Co (On Remand), 223 Mich App 485, 492 (1997), citing
Daubert, 509 US at 596.
3
it does to the dissenters, that the trial court’s ruling involved an “oversight” as to what
was required of it.
The dissenters also criticize the trial court’s application to Dr. Gabriel’s testimony
of the factors set forth in MCL 600.2955(1)(a), (b), (e), and (g). Contrary to the
dissenters, I believe that the trial court did conclude that Dr. Gabriel’s theory had been
subject to testing under MCL 600.2955(1)(a). The court noted that it would be
impracticable to do prospective studies of this sort, so “replication” was impossible.
However, it pointed to retrospective studies analyzing data that “does seem to indicate
some causal nexus between hypoxia in the prenatal state and a neonatal stroke.”
Regarding factor (b), the court explicitly found that journal articles in the record
seemingly demonstrated the causal nexus that Dr. Gabriel asserted and had been subject
to peer review. Hence, his opinion on causation and its basis have been subject to peer
review. It appears that the dissent reads MCL 600.2955(1)(b) as requiring that a peer-
reviewed article mention Dr. Gabriel by name and attribute the opinion specifically to
him. This is an overly restrictive view of factor (b). The only requirement in MCL
600.2955(1)(b) is that the expert opinion and its basis have been subjected to peer review.
Regarding factor (g), the trial court specifically found that “the articles referenced
certainly are outside the context of litigation.” Simply because Dr. Gabriel’s opinion is
retrospective in nature does not preclude it or his methodology from being used outside
the context of litigation.
Regarding factor (e), I agree with the dissent that the trial court should have
considered a “relevant expert community” expanding beyond plaintiff’s and defendants’
experts. This factor directs trial courts to consider whether an opinion and its basis are
generally accepted by “individuals who are knowledgeable in the field of study and are
gainfully employed applying that knowledge on the free market.” Thus, the trial court
should have considered the broader expert community when evaluating this factor.
Nonetheless, because the trial court otherwise fulfilled its gatekeeper function, I
do not believe that its misapplication of one of many evaluative criteria warrants reversal.
Moreover, the trial court also applied the three other factors from MCL 600.2955(1), and
the dissent does not take issue with its application of those factors.
The dissenters similarly err in their criticisms of the Court of Appeals decision.
Their criticisms are premised on its conclusion that Dr. Gabriel’s retrospective use of the
process of elimination is flawed because there are no known predictors of neonatal
strokes. Therefore, supposedly, defendants could not have acted prospectively to prevent
McCall’s stroke. However, Dr. Gabriel’s proposed testimony was that evidence arising
before and after McCall’s birth demonstrated that she experienced hypoxia and a
decreased heart output, leading to a blood clot and stroke. The medical literature in the
4
record establishes that perinatal asphyxia is considered the most common cause of
neonatal stroke. If the treating physicians could have discerned that McCall suffered
hypoxia, it was foreseeable that she could suffer a neonatal stroke. Thus, Chapin v A & L
Parts Inc,7 is not distinguishable and Dr. Gabriel’s testimony was properly admitted.
Moreover, the dissenters assert that Dr. Gabriel’s process of elimination
methodology fails to establish that whatever remains is a proximate cause of the alleged
injury. They are mistaken. It is certainly true that “correlation does not equal
causation.”8 However, this argument has nothing to do with the reliability of Dr.
Gabriel’s opinion. Instead, what the dissenters seem to be arguing is that plaintiff’s claim
must fail as a matter of law for failure to show proximate cause. As we are not reviewing
a motion for summary disposition, this argument is outside the scope of this Court’s
review.
CONCLUSION
The trial court adequately performed its gatekeeper function under Daubert, MCL
600.2955, and MRE 702. Therefore, it did not “shirk[] its responsibility,” as the dissent
contends, nor was its decision to admit Dr. Gabriel’s testimony an abuse of discretion.
Thus, I concur in the order denying defendants’ application for leave to appeal.
CAVANAGH and HATHAWAY, JJ., would affirm the judgment of the Court of
Appeals.
CORRIGAN, J. (dissenting).
I would reverse the judgment of the Court of Appeals and remand for entry of an
order granting defendants’ motion to strike the testimony of plaintiff’s proffered
causation expert. By denying leave to appeal today, the Court fails to correct the
misapplication of the legal standards that govern the admissibility of expert testimony.
Accordingly, I respectfully dissent.
I. FACTUAL AND PROCEDURAL BACKGROUND
In May 2001, plaintiff’s daughter Teija McCall suffered a neonatal stroke, which
resulted in the now nine-year-old child experiencing severe developmental disabilities.
On October 25, 2004, plaintiff filed this medical malpractice action, alleging that the
stroke could have been prevented if defendants timely diagnosed and treated the fetal
distress and immediately delivered McCall via caesarian section, rather than permitting
plaintiff’s vaginal delivery to proceed. Defendants disputed the timing and the cause of
the stroke. Defendants also moved to strike the testimony of plaintiff’s proffered
7
274 Mich App 122 (2007).
8
Post at ___.
5
causation expert, Dr. Ronald Gabriel, arguing that Dr. Gabriel’s proposed testimony did
not meet the criteria set forth in MCL 600.2955 and MRE 702.
After conducting a two-day Daubert9 hearing, the trial court issued an opinion
from the bench. The court denied defendants’ motion to strike, opining that it would
admit Dr. Gabriel’s testimony and “let the chips fall where they may.” On remand from
this Court, the Court of Appeals affirmed the trial court’s decision.10 The Court of
Appeals concluded that the trial court did not abuse its discretion in admitting Dr.
Gabriel’s testimony, stating in part:
The trial court correctly concluded that Dr. Gabriel simply used
process of elimination, a sufficiently tested and time-honored way to arrive
at an answer that it is even the quintessential Sherlock Holmes
methodology. Stated in various ways in various books, once all
impossibilities are filtered out, whatever remains, irrespective of its
improbability, must be the truth.10
____________________________________________________________
10
In The Hound of the Baskervilles, chapter 4, Sherlock Holmes observed
that “we balance probabilities and choose the most likely. It is the scientific
use of the imagination, but we have always some material basis on which to
start our speculation.”
____________________________________________________________
Defendants applied for leave to appeal in this Court. We heard oral argument on the
application.
II. ANALYSIS
An abuse of discretion occurs when the trial court’s decision to admit or exclude
evidence, including expert testimony, falls outside the range of principled outcomes.11
When a court admits legally inadmissible evidence, it necessarily abuses its discretion.12
In this case, the trial court abused its discretion when the court denied defendants’ motion
9
Daubert v Merrell Dow Pharmaceuticals Inc, 509 US 579 (1993).
10
Robelin v Spectrum Health Hospitals, unpublished opinion per curiam of the Court of
Appeals, issued September 10, 2009 (Docket No. 279780).
11
Woodard v Custer, 476 Mich 545, 557 (2006).
12
Craig v Oakwood Hosp, 471 Mich 67, 76 (2004).
6
to strike the testimony of plaintiff’s proffered causation expert because Dr. Gabriel’s
testimony fails to satisfy the exacting criteria of MCL 600.295513 and MRE 702.14
13
MCL 600.2955 provides in pertinent part:
(1) In an action for the death of a person or for injury to a person or property, a
scientific opinion rendered by an otherwise qualified expert is not admissible unless the
court determines that the opinion is reliable and will assist the trier of fact. In making that
determination, the court shall examine the opinion and the basis for the opinion, which
basis includes the facts, technique, methodology, and reasoning relied on by the expert,
and shall consider all of the following factors:
(a) Whether the opinion and its basis have been subjected to scientific testing and
replication.
(b) Whether the opinion and its basis have been subjected to peer review
publication.
(c) The existence and maintenance of generally accepted standards governing the
application and interpretation of a methodology or technique and whether the opinion and
its basis are consistent with those standards.
(d) The known or potential error rate of the opinion and its basis.
(e) The degree to which the opinion and its basis are generally accepted within the
relevant expert community. As used in this subdivision, “relevant expert community”
means individuals who are knowledgeable in the field of study and are gainfully
employed applying that knowledge on the free market.
(f) Whether the basis for the opinion is reliable and whether experts in that field
would rely on the same basis to reach the type of opinion being proffered.
(g) Whether the opinion or methodology is relied upon by experts outside of the
context of litigation.
14
MRE 702 provides:
If the court determines that 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 if (1) the testimony is based on
sufficient facts or data, (2) the testimony is the product of reliable principles and
7
The trial court erred when it failed to expressly determine that Dr. Gabriel’s
proposed testimony is reliable and that his causation opinion would assist the fact-finder.
MCL 600.2955(1) states that “[i]n an action for the death of a person or for injury to a
person or property, a scientific opinion rendered by an otherwise qualified expert is not
admissible unless the court determines that the opinion is reliable and will assist the trier
of fact.” The language of MRE 702 underscores the importance of admitting reliable
expert testimony that will assist the fact-finder. MRE 702 provides in pertinent part:
If the court determines that 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 . . . may testify
thereto in the form of an opinion or otherwise if (1) the testimony is based
on 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.
Both the statute and the evidentiary rule emphasize the court’s obligation to ensure that
any expert testimony is reliable and will assist the fact-finder. However, the record
reflects that the trial court did not make either determination. The record also reflects
that the trial court failed to reference MRE 702 or determine that Dr. Gabriel’s testimony:
“(1) . . . is based on sufficient facts or data, (2) . . . 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.” Contrary to Chief Justice Kelly’s view, the absence of these critical
determinations on the record before us should not be minimized as a failure to “use the
magic word” or sidestepped by characterizing the two-day Daubert hearing as
“lengthy.”15 Rather, these apparent oversights demonstrate an inadequate exercise of the
court’s gatekeeping role and bolster the conclusion that the trial court abused its
discretion when it denied defendants’ motion to strike Dr. Gabriel’s testimony.16
Although the trial court addressed the criteria set forth in MCL 600.2955(1)(a)-(g),
the court erred when it failed to apply each criterion with diligence or precision. For
methods, and (3) the witness has applied the principles and methods reliably to the facts
of the case.
15
Ante at __.
16
See Gilbert v DaimlerChrysler Corp, 470 Mich 749, 780 (2004), quoting Kumho Tire
Co Ltd v Carmichael, 526 US 137, 158-159 (1999) (Scalia, J., concurring) (“While the
exercise of this gatekeeper role is within a court’s discretion, a trial judge may neither
‘abandon’ this obligation nor ‘perform the function inadequately.’”).
8
example, the trial court acknowledged that Dr. Gabriel’s opinion had not been “subjected
to scientific testing and replication,” contrary to § 2955(1)(a). Yet the court concluded
that plaintiff met this criterion, even though it expressly found that plaintiff did not, and
could not, satisfy § 2955(1)(a). The trial court also determined that plaintiff satisfied §
2955(1)(b) without identifying any publication that had subjected Dr. Gabriel’s opinion
to peer review. In effect, the court inverted § 2955(1)(b), which instructs it to consider
“[w]hether the opinion and its basis have been subjected to peer review publication,” and
not whether Dr. Gabriel can extrapolate any modicum of support for his opinion from the
inconclusive medical literature published to date. Additionally, the court concluded that
plaintiff satisfied § 2955(1)(e), which directs the trial court to consider “[t]he degree to
which the opinion and its basis are generally accepted within the relevant expert
community.” This finding is puzzling because the court limited its discussion of the
“relevant expert community,” which § 2955(1)(e) defines as “individuals who are
knowledgeable in the field of study and are gainfully employed applying that knowledge
on the free market,” to the two causation experts who testified. Also puzzling is the trial
court’s failure to identify any member of the relevant expert community besides Dr.
Gabriel who “generally accepted” his opinion.17 Further, the trial court determined that
plaintiff satisfied § 2955(1)(g) although the admittedly retrospective nature of Dr.
Gabriel’s opinion precludes it from being “relied upon by experts outside of the context
of litigation.” That is, Dr. Gabriel’s opinion has no practical value whatsoever for those
interested in predicting and preventing neonatal strokes because it relies entirely on a
retrospective analysis of case-specific factors in the context of litigation. The dearth of
findings that actually satisfy the criteria of § 2955(1) exposes the trial court’s abuse of
discretion.
The Court of Appeals essentially glossed over the trial court’s flawed analysis and
concluded that Dr. Gabriel’s testimony should reach a jury because Dr. Gabriel derived
his opinion from the process of elimination or “the quintessential Sherlock Holmes
methodology.” However, the Court of Appeals offhand observation ignores the
unambiguous language of the statute and the evidentiary rule. No provision in MCL
600.2955 or MRE 702 excuses the trial court from fulfilling its obligation as gatekeeper
when an expert claims that the methodology underlying his opinion derives from the
process of elimination. And referring to “the quintessential Sherlock Holmes
methodology” does not negate the fact that the trial court’s role as gatekeeper “applies to
all stages of expert analysis.”18 Indeed, the “[c]areful vetting of all aspects of expert
17
The other causation expert identified by the court, Dr. Michael Johnston, testified that
“[t]here’s no scientific evidence for what [Dr. Gabriel] says[,]” and that “scientifically,
what has been said, on paper and by deposition by Dr. Gabriel, is completely wrong.”
18
Gilbert, 470 Mich at 782 (emphasis in original).
9
testimony is especially important when an expert provides testimony about causation.”19
The trial court was obligated to thoroughly vet Dr. Gabriel’s testimony under the criteria
set forth in MCL 600.2955 and MRE 702. Quite simply, the court shirked its
responsibility. The Court of Appeals clearly erred when it affirmed the trial court’s
misapplication of the legal standards governing the admissibility of expert testimony.
The Court of Appeals also clearly erred when it endorsed Dr. Gabriel’s opinion,
which purports to establish causation based on his retrospective analysis of several
factors. As the Court of Appeals acknowledged, “Dr. Gabriel did not rely on any of the
medical evidence for its prospective, predictive value, but rather as evidence upon which
to retrospectively exclude impossible scenarios.” The notion that Dr. Gabriel can testify
as a causation expert based on his retrospective use of the “process of elimination” is
flawed for three reasons.
First, it is undisputed that there are no known predictors of neonatal strokes. As
the medical literature relied upon by Dr. Gabriel establishes, because neonatal strokes
occur unexpectedly and without any known indicia, there is no foreseeable basis from
which defendants could have acted to prospectively prevent McCall’s stroke.20
Critically, proximate causation requires foreseeability, yet Dr. Gabriel—plaintiff’s
proffered causation expert—cannot establish that McCall’s stroke could have been
predicted and avoided.
Second, the mere act of eliminating several possibilities through the process of
elimination does not necessarily establish that whatever remains is a proximate cause of
the injury alleged.21 It is axiomatic that correlation does not equal causation.
19
Id.
20
See e.g., Jeffery M. Perlman et al., Neonatal Stroke: Clinical Characteristics and
Cerebral Blood Flow Velocity Measurements, Pediatric Neurology Vol. 11 No. 4, at 281
(1994) (“The data indicate that infants who develop neonatal stroke cannot be
distinguished from infants who do not develop the lesion by current markers of perinatal
distress. Because neonatal stroke frequently occurs as an unanticipated event,
prevention may not be possible.”).
21
For the process of elimination to apply in the first instance, there must be some discrete
universe of possibilities that have been established through the prerequisite scientific
testing. Once the probabilities have been narrowed to some certain and definite number,
it is logical to apply the process of elimination to establish proximate cause. In this case,
it is undisputed that there are no known predictors of neonatal strokes. Because of the
lack of any certain and definite number of predictors, indeed any known predictors at all,
deductive reasoning through the process of elimination is simply inapt. This does not
mean that I question the value of deductive reasoning and the process of elimination in
general, only that it has been applied here in circumstances where it cannot logically
produce the conclusion drawn.
10
Nonetheless, Dr. Gabriel seeks to testify that a correlating event was the causal event
because there are no other known causes. Even a cursory reading of MCL 600.2955 and
MRE 702 establishes that the act of “ruling out” possibilities does not, and should not,
automatically “rule in” the causation sought to be established by plaintiff’s expert here.
Third, the lone case cited by the Court of Appeals to support its endorsement of
Dr. Gabriel’s retrospective use of the process of elimination, Chapin v A & L Parts Inc,
274 Mich App 122 (2007), is distinguishable from this case. In Chapin, the lack of
epidemiological studies establishing a link between the plaintiff’s occupational exposure
to brake dust and mesothelioma was not outcome determinative because the link between
mesothelioma and asbestos exposure was already established. As a result, deductive
logic provided an appropriate means of concluding that the plaintiff’s occupational
exposure to asbestos caused his mesothelioma. In this case, however, there is no
established link between any known predictor and neonatal strokes, as Dr. Gabriel
admitted by stating that “you cannot use fetal monitoring in a prospective or predictive
manner because there is none.” At best, Dr. Gabriel’s opinion amounts to an exercise in
retrospective speculation, rather than viable deductive logic. His opinion fails to satisfy
the legal standards governing the admissibility of expert testimony, and the Court of
Appeals clearly erred when it reached a contrary conclusion.
III. CONCLUSION
Because the trial court failed to carefully consider the testimony of plaintiff’s
proffered causation expert under MCL 600.2955 and MRE 702, its decision to admit Dr.
Gabriel’s testimony fell outside the range of principled outcomes. The Court of Appeals
clearly erred when it failed to correct the trial court’s misapplication of the legal
standards that govern the admissibility of expert testimony. I would reverse the judgment
of the Court of Appeals and remand for entry of an order granting defendants’ motion to
strike. Consequently, I dissent from the Court’s denial of leave to appeal.
YOUNG and MARKMAN, JJ., join the statement of CORRIGAN, J.
DAVIS, J., not participating. I recuse myself and am not participating because I
was on the Court of Appeals panel in this case. See MCR 2.003(B).
I, Corbin R. Davis, Clerk of the Michigan Supreme Court, certify that the
foregoing is a true and complete copy of the order entered at the direction of the Court.
December 22, 2010 _________________________________________
d1221 Clerk