People v. Dendel

748 N.W.2d 859, 481 Mich. 114

• Court: Michigan Supreme Court
• Decided: May 28, 2008
• Docket: Docket 132042
• Status: Published

Opinions

CORRIGAN, J.

Defendant, an insulin-dependent diabetic, was convicted of second-degree murder for injecting the victim, her live-in partner, with a lethal dose of insulin. The Court of Appeals reversed her conviction and remanded for a new trial after concluding that defense counsel was ineffective for failing to produce an expert to refute the testimony of the prosecution’s experts that the victim died from an insulin overdose. We reverse the judgment of the Court of Appeals and remand to the Court of Appeals to consider the remaining issues presented by the parties on appeal.¹ Defense counsel was not ineffective under the test of Strickland v Washington, 466 US 668; 104 S Ct 2052; 80 L Ed 2d 674 (1984), because defendant did not prove that she was prejudiced by her counsel’s failure to produce an expert witness. The Court of Appeals erred in holding that defense counsel could have presented an expert witness who would have refuted the testimony of the prosecution’s experts to the extent that defendant’s acquittal would have been reasonably probable. Further, the trial court correctly held that, in light of the strong circumstantial evidence of defendant’s guilt, it was not reasonably probable that the outcome would have been different had a defense expert testified.

I. FACTUAL BACKGROUND

Defendant and the victim, Paul Michael Burley, were in a long-term relationship and had lived together for years. Burley had been taking numerous medications for several serious illnesses, including an infection with human immunodeficiency virus (HIV), herpes, hepatitis B and C, epilepsy, ataxia, neuropathy, chronic obstructive pulmonary disease, severely impaired vision, dementia, lymphoma, and throat cancer. Burley was not, however, diabetic. By defendant’s own account, Burley was a difficult person to care for. Defendant was solely responsible for making sure that Burley took his medications and for tending to his everyday needs.

Defendant’s relationship with Burley’s family was strained, partly by what she perceived as the family’s failure to help with Burley’s care. Before Burley’s death, defendant had told his sister that “if something happens to your brother, your family won’t know what hit you.” About one week before Burley’s death, defendant, frustrated with Burley’s demands, also told Burley’s sister, “I can’t take this” and “I feel like giving him a shot of insulin.” As an insulin-dependent diabetic, defendant had access to insulin and knew how to inject it. Defendant also knew how insulin metabolizes and that no trace of insulin would remain in Burley’s blood after an insulin injection.

Defendant had expressed her frustration with caring for Burley to a Family Independence Agency (FIA) employee. Less than a week before Burley’s death, defendant e-mailed the FIA employee to seek help with caring for Burley. Defendant told the employee that she could not manage all of Burley’s demands on her own. During a subsequent telephone conversation, defendant again stated that she was frustrated and concerned that the situation was deteriorating and that she no longer knew how to manage Burley. The FIA employee suggested that defendant have Burley evaluated at a mental-health facility or have him placed in respite or hospice care.² The FIA employee testified that defendant had never before expressed any problems with caring for Burley.

During the week leading up to Burley’s death, defendant sought help from the Department on Aging. The department representative told defendant that she did not qualify for help because both she and Burley were not 60 years old. The representative suggested that defendant instead contact hospice services. Defendant replied that hospice services would not help because Burley was not yet near death.

Defendant’s caregiving situation took another turn for the worse the day before he died. On March 14, 2002, a visiting nurse had been assigned to assist defendant and educate her in the proper methods of care. She visited five times, but, on the day before Burley’s death, the nurse terminated her services because Burley had been uncooperative. When the nurse told defendant that she was terminating her services, defendant became “quite tearful and upset.” Defendant told the nurse that she did not know how long she could continue caring for Burley.

At 3:00 a.m. on the day of Burley’s death, defendant called 911, reporting that Burley had been hallucinating and running around with a butter knife. Defendant asked the police to come take Burley to a mental institution. When the police arrived, Burley was sitting calmly in a chair. He told the officers that he was fine and that there was no problem. The police decided to leave Burley at home because he was not a threat to himself or others. One officer testified that defendant was visibly upset with Burley and the police. Defendant also later admitted that she was frustrated with the officers’ decision and that she was hoping for relief because she was at her “wit’s end.”

Defendant contended that later that day she discovered Burley slumped over on the couch and unresponsive. She testified that, because Burley was cold and covered with purple blotches, she thought he might be dead. Rather than calling 911, however, she instead called a friend, who arrived and contacted 911. While the police and emergency personnel were removing Burley’s body from the house, one of Burley’s sisters telephoned. Defendant answered the phone, but quickly ended the conversation without telling her that Burley had died.

During the next several days, defendant spoke with several of Burley’s siblings. She never informed them of his death, but instead falsely told them that he had been hospitalized. One of the victim’s sisters described a 74-minute conversation with defendant two days after Burley’s death. She testified that defendant was “very upbeat” and “nonchalant” in her discussion of topics ranging from Burley’s health to antique jewelry. During this conversation, defendant laughed while describing an alleged incident when Burley had wandered away from the apartment complex and become lost. Yet defendant never mentioned Burley’s death.

Defendant wanted Burley’s body cremated without an autopsy being performed. Although an autopsy was performed despite defendant’s wishes, defendant had Burley’s body cremated before his family learned about his death. When a police detective incorrectly told defendant that the medical examiner had detected insulin in Burley’s body, defendant called him a liar and explained that insulin could not be detected in the human body after death because it breaks down and depletes naturally.

After defendant’s arrest, she told police detectives that Burley had injected himself with insulin. During a later interview with a police detective, defendant said, “That poor dear, he killed himself for me.” She told the detective that despite Burley’s severely impaired vision and problems with holding things, he could inject himself with insulin.³ Defendant also told defense counsel that Burley had killed himself by an insulin injection and that she wanted him to pursue this theory of defense at trial. Defendant also testified that Burley had mental problems and that he had “talked suicide for 10, 15 years.” She had informed two of Burley’s doctors of his suicidal intentions.

Defendant was charged with first-degree murder. The prosecution theorized at the bench trial that defendant injected the victim with a lethal dose of insulin on April 2, 2002. The prosecution presented two expert witnesses, Dr. Bernardino Pacris⁴ and Dr. Michael Evans,⁵ who testified that the evidence supported the theory that Burley had died from an insulin injection rather than from natural causes or an overdose of one of his medications.⁶ Defense counsel Joseph Filip argued that Burley had died either by injecting himself with insulin or from the side effects of numerous medications prescribed for him. Defense counsel did not present any expert testimony to rebut the testimony of the prosecution’s experts.

The trial court found defendant guilty of the lesser-included offense of second-degree murder. Defendant moved for a new trial, arguing that Filip had deprived her of a fair trial by failing to conduct a reasonable investigation into the cause of Burley’s death. The trial court denied the motion. The Court of Appeals remanded for a Ginther⁷ hearing to determine whether Filip had provided ineffective assistance.

At the Ginther hearing, appellate defense counsel called Dr. Laurence Simson,⁸ who testified that the evidence did not support the view that Burley had died from an insulin overdose. Dr. Pacris defended his trial testimony that Burley had died of hypoglycemic shock caused by insulin.⁹ The trial court rejected defendant’s claim of ineffective assistance of counsel. Instead, it found that defense counsel’s performance had been objectively reasonable. The court concluded that defendant had not been prejudiced by Filip’s failure to call an expert forensic pathologist to rebut the opinions of the prosecution’s experts. The court explained why the outcome of the trial would not have been different if the defense had offered Dr. Simson’s testimony:

And if the case was just,... the police had a dead body and you have Dr. Pacris and Dr. Simson, that would be one thing. It wasn’t that. If there was a lot of other testimony, of statements and other witnesses and other things that pointed in that direction, it would have made the testimony of Dr. Evans and Dr. Pacris not as... clear. But I don’t know that I can say that there’s a reasonable probability that the outcome would have been different. There was still — there was other evidence,... admittedly all circumstantial, but there was a lot of other evidence. I am not convinced that that has been established, that it’s reasonably probable that the outcome would have been different....

A divided Court of Appeals reversed and remanded for a new trial. The majority summarized its holding as follows:

Defense counsel’s failure to consult with and present the testimony of appropriate medical experts to address the central issue in this case, the cause of Burley’s death, was clearly deficient in light of prevailing professional norms and, but for that deficiency, there is a reasonable probability that the outcome of defendant’s trial would have been different. [People v Dendel, unpublished opinion per curiam, issued July 18, 2006 (Docket No. 247391), p 3.]

The Court of Appeals majority explained that, despite Dr. Pacris’s testimony that Burley had died from insulin shock, Filip failed to consult a forensic pathologist or Burley’s doctors regarding the cause of Burley’s death. The majority held that Filip’s failure to consult an informed expert who could have refuted Dr. Pacris’s conclusions essentially amounted to a concession that Burley had died from insulin shock. Because it was unlikely that Burley administered the insulin himself, in light of his physical limitations, the trial court was left to conclude that defendant administered the insulin that caused Burley’s death. The majority noted that the Ginther hearing had demonstrated that a qualified pathologist (Dr. Simson) would have (1) refuted Dr. Pacris’s conclusion that Burley died from insulin shock and (2) provided an alternative, noncriminal explanation for Burley’s death. The majority concluded: “Trial counsel’s failure deprived defendant of a substantial defense, and there is a reasonable probability that this would have made a difference in the outcome of the trial.” Dendel, supra at 4.

Judge WILDER dissented, rejecting the conclusion that defendant had been prejudiced by counsel’s performance. He relied on the trial court’s conclusion that even if Filip had introduced Dr. Simson’s testimony, the court would nonetheless have found defendant guilty in light of the weight of the evidence. This evidence supporting defendant’s guilt included the following: defendant had the opportunity to inject the insulin, defendant admitted being aware that no trace of insulin would be found in Burley’s blood after his death, defendant was under considerable stress in trying to care for Burley by herself, and defendant not only failed to inform Burley’s family of his death, but she apparently hid it from the family. Judge WILDER also noted that nothing established that Dr. Simson was more credible than Dr. Pacris. Moreover, Dr. Simson concededly could not rule out insulin shock as the cause of death. Judge WILDER stated that the effect of expert testimony depends on the fact-finder’s evaluation of credibility, and the fact-finder in this case had expressly determined that Dr. Simson’s testimony would not have changed the result of the trial.

The prosecution appealed, arguing that the Court of Appeals had erred in holding that defendant was entitled to a new trial on the basis of ineffective assistance of counsel. This Court heard oral argument on whether to grant the application or take other peremptory action.

II. STANDARD OF REVIEW

“Whether a person has been denied effective assistance of counsel is a mixed question of fact and constitutional law. A judge first must find the facts, and then must decide whether those facts constitute a violation of the defendant’s constitutional right to effective assistance of counsel.” People v LeBlanc, 465 Mich 575, 579; 640 NW2d 246 (2002). This Court reviews a trial court’s factual findings for clear error and reviews de novo questions of constitutional law. Id.

III. ANALYSIS

In People v Carbin, 463 Mich 590, 599-600; 623 NW2d 884 (2001), this Court explained the test for determining whether a defendant has been denied the effective assistance of counsel:

A defendant seeking a new trial on the ground that trial counsel was ineffective bears a heavy burden. To justify reversal under either the federal or state constitutions, a convicted defendant must satisfy the two-part test articulated by the United States Supreme Court in Strickland v Washington, 466 US 668; 104 S Ct 2052; 80 L Ed 2d 674 (1984). See People v Pickens, 446 Mich 298, 302-303; 521 NW2d 797 (1994). “First, the defendant must show that counsel’s performance was deficient. This requires showing that counsel made errors so serious that counsel was not performing as the ‘counsel’ guaranteed by the Sixth Amendment.” Strickland, supra at 687. In so doing, the defendant must overcome a strong presumption that counsel’s performance constituted sound trial strategy. Id. at 690. “Second, the defendant must show that the deficient performance prejudiced the defense.” Id. at 687. To demonstrate prejudice, the defendant must show the existence of a reasonable probability that, but for counsel’s error, the result of the proceeding would have been different. Id. at 694. “A reasonable probability is a probability sufficient to undermine confidence in the outcome.” Id. Because the defendant bears the burden of demonstrating both deficient performance and prejudice, the defendant necessarily bears the burden of establishing the factual predicate for his claim. See People v Hoag, 460 Mich 1, 6; 594 NW2d 57 (1999).[¹⁰]

We conclude that defendant has failed to demonstrate that she was prejudiced by Filip’s performance.¹¹

A. THE EXPERT TESTIMONY

Dr. Pacris testified at trial that he performed an autopsy on Burley on April 3, 2002. Dr. Pacris initially concluded that Burley had died from natural causes. But because a police officer told Dr. Pacris that he suspected that Burley might have been injected with insulin, which can be fatal to a nondiabetic, Dr. Pacris sent Burley’s fluids to AIT Laboratories to be tested for insulin, glucose, and C-peptide levels. The tests revealed that Burley’s glucose level was zero and that his insulin and C-peptide levels were normal. Dr. Pacris explained that, although the glucose levels in a person’s bodily fluids drop immediately after the person dies, the complete lack of glucose in Burley’s vitreous fluids was consistent with a finding that Burley had been injected with insulin.¹² He found acute tubular necrosis in the kidneys and dead cells in the proximal tubules of the brain, which are usually seen in people who have suffered hypoglycemic shock. Dr. Pacris ultimately concluded that the cause of death was complications from hypoglycemia, which can be caused by an insulin injection.¹³ In reaching this conclusion, he relied more on his anatomical findings and the circumstances surrounding the death rather than on the toxicological findings. Specifically, he relied on microscopic hypoxic¹⁴ changes in Burley’s brain in concluding that Burley must have been comatose for at least 12 hours before he died at 4:00 p.m. on April 2, 2002. He testified that hypoxic changes to the brain, including red neurons on the hippocampus, are only manifested if the person has been comatose for about 12 hours. Because this conclusion was inconsistent with defendant’s story that Burley had been alive and conscious at noon on that day, Dr. Pacris concluded that defendant’s story “doesn’t jive.”

Dr. Evans also testified at trial for the prosecution. He testified that if glucose had been present in Burley’s system, it would have disproved death by insulin injection. The lack of any glucose in Burley’s vitreous fluids supported the theory that Burley had been injected with insulin. Further, although the level of morphine in Burley’s blood was very high, it might not be lethal to someone who had built up a tolerance for it.

At the Ginther hearing, Dr. Simson disagreed with the conclusions of Dr. Pacris and Dr. Evans. He testified that Burley’s vitreous and blood glucose levels had been confused in the reports and in the testimony introduced at trial. Dr. Simson opined that the pathological and toxicological findings did not support the view that Burley had died of hypoglycemic shock caused by an insulin overdose. He opined that because a person’s vitreous glucose level can drop to zero after he dies, the lack of glucose in Burley’s vitreous fluids did not prove that he died of hypoglycemic shock. Dr. Simson further opined that the necrosis of the proximal tubules in Burley’s brain and the acute tubular necrosis in the kidneys could be attributed to postmortem changes rather than hypoglycemic shock. That is, Dr. Simson responded to Dr. Pacris by arguing that the anatomical changes observed in Burley’s body may have been attributable to decomposition, rather than an insulin overdose. Dr. Simson also testified that the normal reddish-brown color of the kidneys was inconsistent with kidneys that had undergone hypoglycemic shock. Dr. Simson opined that he would have concluded that Burley had died of a multiple-drug overdose, primarily caused by a high level of morphine. He explained that Burley’s morphine level at the autopsy was approximately three times the therapeutic limit, meaning that his morphine level would have been even higher if, as Dr. Pacris testified, Burley had been comatose for 12 hours before his death. Dr. Simson conceded, however, that he had seen cases of much higher levels of morphine in the blood.¹⁵ Dr. Simson also acknowledged that the evidence was “not inconsistent with hypoglycemic shock” and that he could not rule out the possibility that insulin overdose was the cause of death.

Dr. Pacris defended his trial testimony that Burley had died of hypoglycemic shock caused by insulin. In response to Dr. Simson’s Ginther hearing testimony, Dr. Pacris first testified that, in reaching the conclusion that Burley died from an insulin injection, he had principally relied on the changes observed in the brain and kidneys, rather than Burley’s low glucose level. Dr. Pacris then testified that the necrosis of the proximal tubules in Burley’s brain and the acute tubular necrosis in the kidneys could not be attributed to postmortem changes because there was no evidence that the body was decomposing.¹⁶ Dr. Pacris noted that the necrosis in Burley’s brain had occurred solely in the third and fourth layers of the cortex and that the remainder of the cortex had not yet decomposed. This difference indicated that the changes in the third and fourth layers of the cortex were not caused by general decomposition, as suggested by Dr. Simson, because some necrosis would have been found in the remainder of the cortex if the changes observed were due to general decomposition. Moreover, Dr. Pacris noted that there are microscopic differences between cells that are simply decomposing and cells that have been altered before death by changes due to lack of glucose in the body. According to Dr. Pacris, the microscopic changes observed in Burley’s kidneys reflected a lack of glucose in the blood before death, rather than general decay after death. In short, Dr. Pacris responded to Dr. Simson by arguing that the specific changes observed in Burley’s body were incompatible with Dr. Simson’s theory that the changes were caused simply by decomposition. Moreover, Dr. Pacris explained that the normal reddish-brown color of the kidneys, which Dr. Simson had found important, was not inconsistent with Dr, Pacris’s microscopic finding that the kidneys had acute tubular necrosis caused by hypoglycemic shock. Furthermore, Dr. Pacris testified that although Burley had a high level of morphine in his system, he could not have died from a morphine overdose. He explained that death from a morphine overdose is instantaneous. The person does not initially become comatose. A morphine overdose was inconsistent with the hypoxic changes in Burley’s brain that indicated he had been comatose for 12 hours before death. Although Burley’s morphine level was three times the therapeutic limit, this amount of morphine might not be fatal to a person who had developed a tolerance to the drug, as Burley had.

After hearing the testimony of Dr. Simson and Dr. Pacris, the trial court concluded that Dr. Simson’s testimony would not have changed the outcome of the trial. By declining to conclude that Dr. Simson’s testimony had effectively refuted the testimony of Dr. Pacris, the trial court implicitly held that Dr. Simson was not more credible than the prosecution’s experts. “[R]egard shall be given to the special opportunity of the trial court to judge the credibility of the witnesses who appeared before it.” MCR 2.613(C). We review a trial court’s determination of credibility for clear error. People v Knight, 473 Mich 324, 344; 701 NW2d 715 (2005). “A finding is clearly erroneous when, although there is evidence to support it, the reviewing court, on the whole record, is left with the definite and firm conviction that a mistake has been made.” Bynum v EASB Group, Inc, 467 Mich 280, 285; 651 NW2d 383 (2002).

The Court of Appeals stated that Dr. Simson’s testimony would have “refuted [Dr. Pacris’s] conclusions that Burley died as a result of an insulin overdose____” Dendel, supra at 4. Hence, the Court of Appeals implicitly concluded that the trial court had committed clear error by failing to find Dr. Simson more credible than Dr. Paeris and Dr. Evans. However, unlike the Court of Appeals panel, we see no reason to disturb the trial court’s implicit finding on the credibility of the expert witnesses; the testimony does not clearly demonstrate that one expert witness was more credible than another. Although Dr. Simson opined that the pathological and toxicological findings did not support the view that Burley had died of hypoglycemic shock, Dr. Paeris defended his position at the Ginther hearing by offering legitimate reasons for his findings and for discounting Dr. Simson’s theory that the relevant changes in Burley’s body were due simply to general decomposition. Dr. Simson did not respond to Dr. Pacris’s rebuttal of his testimony. It is also significant that Dr. Simson conceded the possibility that Burley had died from insulin overdose. Thus, Dr. Simson did not conclusively refute Dr. Pacris’s testimony that Burley had died of an insulin overdose. We are not “left with the definite and firm conviction” that the trial court erred in finding that Dr. Simson was not more credible than Dr. Paeris and Dr. Evans.

Further, defendant’s own statements supported the theory of the prosecution’s experts regarding the cause of Burley’s death. After defendant’s arrest, she told both police detectives and defense counsel that Burley had injected himself with insulin. These statements were inconsistent with Dr. Simson’s theory of death, but were consistent with the testimony of the prosecution’s experts that Burley had died of an insulin overdose.¹⁷

For all these reasons, we have no cause to believe that if Dr. Simson had testified at trial, the trial court would have given more weight to his testimony than that of the prosecution’s experts. We conclude that defendant did not establish a “reasonable probability” that the outcome of the trial would have been different had Dr. Simson testified. Strickland, supra at 694 (emphasis added).¹⁸

B. OTHER CIRCUMSTANTIAL EVIDENCE

We also conclude that the trial court did not err when it held at the Ginther hearing that, even if Filip had called an expert to rebut the testimony of Dr. Pacris and Dr. Evans, “there was a lot of other evidence” supporting defendant’s conviction and that the outcome of the trial would have been the same. Even if Dr. Simson had testified, the strong circumstantial evidence supported the theory that defendant had given Burley an insulin injection.

Burley was difficult to care for because of his multiple health problems, which included dementia. Defendant was under a great deal of stress as Burley’s sole caregiver.¹⁹ Frustrated by Burley’s demands, defendant had considered giving him a shot of insulin, which she knew could be lethal and would be difficult to detect in a deceased person. When her caregiving situation became worse, defendant unsuccessfully attempted to obtain assistance in caring for Burley from several sources. Less than 24 hours before Burley’s death, defendant became “quite tearful and upset” when the nurse assisting defendant terminated her services because Burley had been uncooperative. Defendant admitted that she was at her “wit’s end” in the middle of that night when the police declined to take Burley away after he caused a disturbance. In light of the facts leading up to Burley’s death, the trier of fact could reasonably conclude that this nighttime incident caused defendant to finally snap and follow through with her idea to inject Burley with insulin. This finding would be consistent with Dr. Pacris’s testimony that hypoxic changes in Burley’s brain indicated that he had fallen into a coma from insulin-induced hypoglycemic shock at about 4:00 a.m., shortly after the police left.

The trier of fact could also infer that defendant’s actions after Burley’s death demonstrated her guilty state of mind and her attempt to cover up the crime. Defendant testified that when she suspected that Burley might be dead, she did not contact 911, but instead called a friend to come over. Defendant lied to Burley’s family about his condition and hid his death from the only persons who might have questioned the cause of death and recalled her threat to inject him with insulin.²⁰ Moreover, defendant managed to have Burley’s body cremated before Burley’s family could question the cause of death. She had also wanted Burley’s body cremated without an autopsy being performed,²¹ but was unable to prevent the autopsy. This circumstantial evidence regarding defendant’s state of mind further supports the prosecution’s theory that defendant murdered Burley.

Considering all this strong circumstantial evidence of defendant’s guilt, we hold that the trial court did not err in concluding that defendant would have been convicted of second-degree murder even if Dr. Simson had challenged the conclusions of the expert witnesses for the prosecution.²²

rv CONCLUSION

Defense counsel was not ineffective for failing to produce an expert at trial who would rebut the testimony of the prosecution’s experts that Burley died from an insulin overdose. Defendant was not prejudiced by Filip’s failure to produce an expert witness because there is no indication that the trial court would have accepted the testimony of defendant’s expert over that of the prosecution’s experts and there was other strong circumstantial evidence to support defendant’s guilt. Therefore, we reverse the judgment of the Court of Appeals and remand to the Court of Appeals to consider the remaining issues.

Taylor, C.J., and Weaver, Young, and Markman, JJ., concurred with CORRIGAN, J.

1 Because the Court of Appeals held that defendant received ineffective assistance of counsel at trial, it declined to address the remainder of the issues presented by defendant’s appeal and the prosecutor’s cross-appeal.

2 The FIA employee testified that as long as Burley was competent, the FLA could not compel defendant to put Burley in a nursing home.

3 Defendant also suggested the unlikely scenario that if Burley had not injected the insulin himself, perhaps someone had broken into her apartment, found her insulin and syringe, and given Burley the shot.

4 Dr. Pacris is an Oakland County medical examiner and a former Jackson County forensic pathologist who has been qualified as an expert witness in more than 100 trials.

5 Dr. Evans is the president and chief executive officer of AIT Laboratories, the former state toxicologist for Indiana, and a professor of toxicology who has testified as an expert in 35 states.

6 We discuss Dr. Pacris’s and Dr. Evans’s trial testimony in detail in part 111(A) of this opinion.

7 People v Ginther, 390 Mich 436; 212 NW2d 922 (1973).

8 Dr. Simson is a forensic pathology consultant and a former professor of pathology, an Ingham County pathologist, and a national consultant in forensic pathology to the Surgeon General of the United States Air Force.

9 We discuss Dr. Simson’s and Dr. Pacris’s Ginther hearing testimony in detail in part 111(A) of this opinion.

10 The dissent accuses us of misunderstanding defendant’s burden under the prejudice prong of Strickland. Yet, ironically, it is the dissent, not us, that applies the wrong standard. The dissent states: “Because defendant has shown that her trial counsel’s performance deprived her of a substantial defense, she has met her burden of showing prejudice, unless other evidence rendered this defense unbelievable.” Post at 147. The dissent fails to recognize that to demonstrate prejudice, a defendant “must show the existence of a reasonable probability that, but for counsel’s error, the result of the proceeding would have been different.” Carbin, supra at 600. Instead, the dissent erroneously suggests that prejudice is presumed if defendant was deprived of one of several theories of defense. Contrary to the dissent’s assertion, that a defense attorney performed deficiently in presenting a viable defense does not automatically require the conclusion that there is a reasonable probability that the result of the proceeding would have been different absent counsel’s deficient performance. The dissent does not explain why there is a reasonable probability that she would have been acquitted had defense counsel presented expert testimony to support the theory that Burley died of a morphine or multiple-drug overdose. Justice Kelly also mentions repeatedly that she thinks that defendant might be innocent. But the guilt or innocence of the accused is a matter to be decided by the fact-finder, not the appellate courts. Defendant is not entitled to relief unless she satisfies Strickland’s test for prejudice.

11 The dissent argues that Filip’s performance was deficient because he failed to present an expert to challenge the prosecution’s theory regarding the cause of death. The dissent’s argument is misplaced. The majority does not conclude that defendant failed to show that counsel’s performance was deficient. Rather, the majority concludes only that defendant was not denied the effective assistance of counsel, because she failed to show that she was prejudiced by counsel’s performance. This aspect of the dissent’s argument appears directed at the concurrence, not the majority opinion.

12 An insulin injection causes a nondiabetic’s glucose level to drop to a dangerous level, depriving the brain of necessary glucose. The person’s brain will then shut down, and the person will become comatose.

13 Although Dr. Pacris did not find a needle mark on Burley’s body, he explained that insulin is injected by means of a hypodermic needle, which normally does not leave a visible mark on the body.

14 “Hypoxic” is defined as “[djenoting or characterized by hypoxia.” Stedman’s Medical Dictionary (26th ed). “Hypoxia” refers to a “[d]e-crease below normal levels of oxygen in inspired gases, arterial blood, or tissue....” Id.

15 The therapeutic level for morphine is 30 to 100 nanograms per milliliter of blood. The laboratory report stated that Burley had a morphine level of 328 nanograms per milliliter. Dr. Simson testified that he had seen cases as high as 800 to 900 nanograms of morphine per milliliter. The laboratory report listed the lethal level of morphine at 200 to 2,300 nanograms per milliliter, indicating that there have been cases of morphine levels up to 2,300 nanograms per milliliter.

16 Dr. Pacris referred to the decomposition of the body tissues as “autolysis.”

17 The dissent argues that the defense theory that Burley killed himself by an insulin injection is “a highly unlikely occurrence given his debilitated physical condition ....” Post at 146. Although this may or may not be true, defendant herself, who presumably knew Burley’s physical capabilities better than anyone else, told the police detectives that Burley was physically able to inject himself with insulin and had in her opinion done so. Thus, it was reasonable for defense counsel to argue that Burley had injected himself with insulin.

18 As discussed, the trial court, which was the finder of fact at the bench trial, stated at the Ginther hearing that the outcome of the trial would not have changed if Dr. Simson had testified. Because we review de novo the trial court’s determination of prejudice, however, the fact-finder’s determination on that issue at the Ginther hearing is not binding on the appellate courts. We underscore that the test for prejudice is an objective test and that appellate courts should not simply defer to the trial court’s judgment regarding prejudice, even if the trial court was the fact-finder at the original trial, as in this case.

19 The dissent supports its assertion that defendant does not have “the behavioral profile of a cold-blooded killer,” post at 148, by stating that defendant financially supported Burley while he was ill. The dissent mischaracterizes the couple’s financial situation. In fact, defendant received $730 monthly from the FIA to care for Burley and Burley’s social security disability benefits of $530 monthly. As Burley’s caregiver, she was entitled to live in government-subsidized housing. Although Burley’s family and the FIA urged defendant to place Burley into a nursing home, hospice care, or some other program that would furnish Burley with better medical care, defendant declined to do so, explaining to Burley’s sister that if she were to put Burley into a nursing home, she would lose her housing, the FIA benefits, and Burley’s income. On the other hand, if Burley were to die, defendant would gain some financial security: defendant was the sole beneficiary of six life insurance policies that she had taken out on Burley, worth approximately '$25,000 at the time of Burley’s death. Thus, the evidence suggests that defendant may have had financial motivations for rendering care to Burley.

20 The dissent suggests that defendant did not tell Burley’s family about his death because she, not Burley’s family members, had cared for Burley toward the end of his life. Although we cannot know with certainty defendant’s motives, defendant’s failure to inform Burley’s family of his death was sufficiently unusual to support an inference that defendant acted with a guilty state of mind.

21 We do not disagree with the dissent’s assertion that a decision to cremate a loved one, by itself, is not unusual. But the decision to have a loved one cremated before the victim’s family knows about the death and before an autopsy can be performed supports an inference of a guilty state of mind.

22 The dissent states:

[H]ad defense counsel challenged the cause of death, the finder of fact would have been left with two reasonable alternatives: (1) to decide that the evidence showed that defendant killed Burley or (2) to conclude that Burley killed himself, intentionally or accidentally, possibly to spare his loving companion of nearly 30 years the burden of his continuing care. [Post at 150.]

Yet Filip’s decision not to present an expert witness challenging the conclusions of the prosecution’s expert witnesses regarding the cause of death left the fact-finder with the same reasonable alternatives. The only difference is that Filip chose to argue that Burley killed himself with insulin, not morphine or some other drug. This was a viable defense that Filip energetically pursued.