Wilson v. State

764 A.2d 284 (2000) 136 Md. App. 27

Garrett Eldred WILSON
v.
STATE of Maryland.

Nos. 1892, Sept. Term, 1999.

Court of Special Appeals of Maryland.

November 2, 2000. Reconsideration Denied January 31, 2001.

John L. Kopolow, Asst. Public Defender (Stephen E. Harris, Public Defender, on the brief), Baltimore, for appellant.

Steven L. Holcomb, Asst. Atty. Gen. (J. Joseph Curran, Atty. Gen., Baltimore, and Douglas Gansler, State's Attorney for Montgomery County, Rockville, on the brief), for appellee.

Argued before MURPHY, Chief Judge, HOLLANDER and KENNEY, JJ.

KENNEY, Judge.

In July 1999, Garrett Eldred Wilson, appellant, was convicted by a jury in the Circuit Court for Montgomery County of first-degree murder in connection with the 1987 death of his infant son, Garrett Michael Wilson. He was sentenced to life in prison without possibility of parole.

Questions Presented

Appellant presents six questions for our review:

1. Did the trial judge err in admitting expert testimony which encroached on the jury's function to judge credibility and resolve contested facts?

2. Did the trial judge err in admitting statistical evidence and the expert opinions based on that evidence?

3. Did the trial judge take inadequate corrective action when, in closing argument, the State's Attorney misrepresented a statistical computation as the probability of appellant's innocence?

4. Did the trial judge err in prohibiting the defense expert from presenting a basis for his opinion?

5. Was the evidence rebutting the findings stated in the original death certificates admitted in violation of applicable statutes?

6. Did the trial judge err in admitting evidence of appellant's alleged murder of his infant daughter six years before the alleged murder in this case?

We answer these questions in the negative and shall affirm.

Factual Background

In 1976 or 1977, when appellant was twenty years old, he began a sexual relationship with Deborah Oliver Fennell, who was then 13 years old. She testified that appellant "wooed" her by buying her gifts and by leading a relatively expensive lifestyle. Ms. Fennell became pregnant five times in the next three years and, at appellant's behest, aborted the first four pregnancies. Appellant wanted Ms. Fennell to abort the fifth pregnancy, but the doctor they consulted refused because Ms. Fennell was five months pregnant. The couple then married; Ms. Fennell was 15 years old and appellant was 22.

Ms. Fennell, whose pregnancy was without complications, testified that, when she was seven months pregnant, appellant asked her if she would be "okay" if "anything ever happened to the baby." On February 25, 1981, Ms. Fennell gave birth to a daughter, Brandi Jean Wilson. According to Ms. Fennell and her parents, Jean and Kyle Oliver, Brandi was a very healthy baby.

Within four weeks of Brandi's birth, appellant, without his wife's knowledge, purchased from two separate insurance companies four days apart two life insurance policies in the amounts of $10,000 and $30,000 on Brandi's life. Appellant was the primary beneficiary for both policies. The insurance agents testified that, if appellant had told them he was getting two policies, their companies probably would not have sold them to him.

Appellant did not feed Brandi or change her diapers, and he did not get up during the night to care for her. On the night of April 30, 1981, Ms. Fennell had a cold, and appellant gave her three or four pills that he said were vitamins. After taking the pills, Ms. Fennell slept through the entire night, which she had not done since Brandi's birth. That night was the first and only time appellant took care of the baby. Mrs. Oliver had offered to babysit Brandi, as she had often done in the past, while Ms. Fennell rested, but appellant said that he would care for Brandi.

Between approximately 3:30 a.m. and 5:30 a.m. on April 30, 1981, Brandi died. At approximately 6 a.m. appellant, rather than calling 911 or waking Ms. Fennell, called the Olivers and told them that Brandi was dead and that they should come over to appellant's house. Mrs. Oliver told appellant to call 911, and she and Mr. Oliver left for appellant's house. Although the fire station was approximately halfway between the Olivers' house and appellant's house, and although the Olivers had to dress before coming, they arrived before the paramedics. Mark Cashman, the first paramedic to enter Brandi's room, testified that he could tell immediately by her stiffness and blue color that she was dead.

Ms. Fennell slept so soundly she was not awakened by the arrival of her parents or the police officers who followed. Her mother shook her awake.

Ms. Fennell testified that she put Brandi in the crib on her stomach. In the crib with Brandi were pillows, blankets, stuffed animals, and a comforter. Brandi did not have the ability to roll herself over. The paramedic, Mr. Cashman, testified that, when he arrived, Brandi was lying face down in the crib. After an autopsy, Brandi's death was labeled as a Sudden Infant Death Syndrome (SIDS) death. Ms. Fennell, however, told family members and friends that she thought appellant was "involved" in Brandi's death. Ms. Fennell left appellant four months after Brandi's death, and they later divorced.

On the same morning Brandi died, appellant called the insurance agent who sold him one of the policies on Brandi's life and informed him of Brandi's death. Later that day, appellant played pool and then went flying with a friend. Several witnesses testified that appellant's demeanor after Brandi's death, including at the funeral, reflected a lack of emotion.

Soon after Brandi's death, appellant collected the $40,000 in insurance proceeds and made large purchases, including a new Trans Am. He never informed Ms. Fennell that he had insured the baby, or that he had received the proceeds. She did not learn about the insurance policies until after Brandi's death. One of Ms. Fennell's friends, who was dating the agent who sold appellant the $10,000 policy, told her about that policy sometime after Brandi's death. Ms. Fennell was told by police investigators about the other policy more than a decade after Brandi's death.

In 1986 appellant became engaged to two women, Mary Anastasi and Elizabeth Bahlmann, during the same period of time and scheduled weddings for March and June 1986. Ms. Bahlmann testified that appellant, who was then working as a salesman at a health club, frequently gave her expensive gifts and often persuaded her to pay for the gifts. In March 1986 he married Ms. Anastasi in Maryland, and in May 1986 he and Ms. Bahlmann filed for a marriage license in Virginia. On that same day, however, Ms. Bahlmann surreptitiously looked at papers in appellant's wallet and discovered that he was already married. Ms. Bahlmann informed appellant that he owed her $3,500 for the wedding preparations she had made and for the gifts he had purchased for her with her credit card. Appellant promised to pay this amount, and began making payments toward the total.

In late 1986 appellant began socializing with still another woman, Julie Stinger, giving her expensive gifts. He also persuaded her to lend him $5,250.

Appellant also owed $1,000 to his uncle, Donald Ward.

Ms. Anastasi gave birth to appellant's son, Garrett Michael Wilson, on March 22, 1987. Ms. Anastasi testified that appellant interacted with the baby very infrequently, never feeding him or changing his diapers.

Within five weeks of Garrett Michael's birth, appellant approached two separate insurance companies and purchased two life insurance policies on the child's life in the amounts of $50,000 and $100,000. According to the insurance agents and the documents maintained by the insurance companies, appellant did not inform either company of the other insurance policy. Appellant was the primary beneficiary for both policies, and his wife was unaware of the policies until after he had purchased them.

Appellant discussed marriage with Ms. Stinger, but she eventually discovered that he was already married and had a child. She demanded that he repay the money she had lent him. Appellant told Ms. Stinger that he had not wanted a child and that Ms. Anastasi had had the baby against his wishes.

In August 1987, as appellant was preparing to go to Bethany Beach for a weekend with Ms. Anastasi and Garrett Michael, he spoke with Ms. Stinger, who was "pushing really hard" for the return of her money. Appellant told her he would have the money "soon" because he was "going to take care of it this weekend."

Ms. Anastasi had researched SIDS, because appellant told her that Brandi died of SIDS, and she thought the two babies' genetic similarities might make Garrett Michael susceptible to SIDS. On August 12, 1987, during the trip to Bethany Beach, Ms. Anastasi remarked to appellant that, as Garrett Michael was then five months old, he had made it to an age when SIDS deaths were far less common. In her words, he was "out of the woods." Appellant did not reply to her comment.

On August 13, 1987, after returning from the beach, appellant told Ms. Stinger that he would be getting the money to repay her "real soon." Garrett Michael died nine days later, at approximately 6 a.m. on August 22, 1987.

The morning of Garrett Michael's death was the first occasion appellant took care of him without Ms. Anastasi present. The couple was in bed when the baby cried. When Ms. Anastasi began to get up to feed the baby, appellant said that he would feed him. Ms. Anastasi was surprised because it was the first such offer appellant had made. When appellant went into Garrett Michael's room, Ms. Anastasi, listening via a room monitor, could hear footsteps approaching the crib, and then creaking sounds from a rocking chair in the nursery. The rocking-chair sounds continued for approximately seven minutes, and then she heard a "patting sound." She next heard footsteps approaching the crib again, and a "sigh" similar to "expelling air." This last sound concerned her, but she was not overly alarmed. She reasoned that, if appellant had a problem, he would call her. She then got up and went downstairs to feed her two cats, who had been pestering her for food. Afterwards, she went upstairs to Garrett Michael's room. Appellant was no longer there. She immediately noticed that the baby did not "feel right," and there was foam around his mouth. He was limp when she picked him up.

Ms. Anastasi ran back to her bedroom with the baby and met appellant as he was walking out of the bathroom. She screamed at appellant: "Garrett, what did you do to him?" Appellant, whose face she described as being "pale" at the time, did not respond. He silently walked away when Ms. Anastasi asked him to call 911. Ms. Anastasi called 911 herself and attempted CPR per the instructions she received from the 911 dispatcher. She also had to go downstairs herself to turn on the house lights to guide the paramedics to the house. Paramedics arrived and rushed Garrett Michael to Shady Grove Hospital. After the ambulance left the house, appellant and Ms. Anastasi went to her two-door Saab to follow the ambulance to the hospital. Before leaving, appellant removed the baby's car seat from the back seat.

Garrett Michael was pronounced dead at the hospital. While in the ambulance his heart had been fibrillating, i.e., beating at a very fast, non-sustainable rate. Dr. Charles Kokes, one of the State's expert witnesses at trial, testified that, based on the fibrillation, the baby was probably alive within the fifteen minutes before CPR began. An EKG reading obtained by the paramedics in the ambulance showed that Garrett Michael's body still contained electronic waves, indicating that he had just died or was in the process of dying. One paramedic testified that the baby's heart may have been beating regularly four minutes before the paramedics began CPR. After an autopsy, Garrett Michael's death was labeled a SIDS death.

On the day Garrett Michael died and the next day, Ms. Anastasi expressed to three family members and friends her opinion that appellant had murdered Garrett Michael for the insurance money.

Testimony was adduced concerning appellant's apparently unemotional demeanor after the baby's death. At the funeral appellant greeted his uncle, Donald Ward, with the comment that "[i]t is most unusual to have two SIDS deaths in one family." Ms. Anastasi testified that, although she initially believed appellant murdered Garrett Michael, she did not leave him because other people, including the director of a SIDS support group, assured her that any "foul play" would have been revealed by the autopsy.

Within a week of Garrett Michael's death, appellant told Ms. Stinger that, inasmuch as he had "taken insurance policies out on the baby," he would soon have money to repay her. He told Ms. Stinger he bought the policies "in case the baby died." Appellant collected the $150,000 in insurance proceeds and paid Ms. Stinger the $5,250 he owed her.

He also called Ms. Bahlmann and informed her that he had had a baby, that the baby had died, and that he would soon be receiving insurance proceeds from which he could repay the last of his debt to her. Even though he had had several conversations with Ms. Bahlmann between March and August of 1987, he had not told her about the baby.

Appellant made many expensive purchases after receiving the insurance money, including at least five expensive gifts of jewelry for Ms. Anastasi. On one occasion, appellant showed Ms. Anastasi $10,000 in cash, which he said was part of the insurance money.

Appellant and Ms. Anastasi were divorced in 1993 or 1994, at approximately the same time that Ms. Anastasi discovered that appellant had been maintaining a marriage in Texas to a woman named Vicki Lynn Wilson, with whom he had a child.

Discussion

I.

The State presented testimony from four medical experts about the deaths of Brandi and Garrett Michael: Dr. Ann Dixon, Maryland's Deputy Chief Medical Examiner, who performed the autopsy on Brandi in 1981; Dr. Charles Kokes, a forensic pathologist who worked in the Maryland Medical Examiner's Office in 1987 and performed the autopsy on Garrett Michael; Dr. John Smialek, Maryland's Chief Medical Examiner; and Dr. Linda Norton, a forensic pathologist in private practice specializing in pediatric deaths. Appellant contends that the trial judge erred by allowing these witnesses to testify about their opinions concerning the deaths of Brandi and Garrett Michael. Appellant believes that their testimony "encroached on the jury's function to judge credibility and resolve contested facts." In particular, appellant contends that the court erred by allowing the witnesses to testify that the infants died of suffocation or "probable suffocation" and that the deaths were homicides.

Maryland Rule 5-702 provides:

Expert testimony may be admitted, in the form of an opinion or otherwise, if the court determines that the testimony will assist the trier of fact to understand the evidence or to determine a fact in issue. In making that determination, the court shall determine (1) whether the witness is qualified as an expert by knowledge, skill, experience, training, or education, (2) the appropriateness of the expert testimony on the particular subject, and (3) whether a sufficient factual basis exists to support the expert testimony.

Maryland Rule 5-704 provides, in pertinent part:

(a) In general. Except as provided in section (b) of this Rule [concerning testimony about the mental state of defendants], testimony in the form of an opinion or inference otherwise admissible is not objectionable merely because it embraces an ultimate issue to be decided by the trier of fact.

"[T]he admissibility of expert testimony is a matter largely within the discretion of the trial court and its action will seldom constitute a ground for reversal." Myers v. Celotex Corp., 88 Md.App. 442, 460, 594 A.2d 1248 (1991), cert. denied, 325 Md. 249, 600 A.2d 418 (1992). "A trial judge's decision to admit or exclude expert testimony will be reversed only if it is founded on an error of law or some serious mistake, or if the judge has abused his discretion." Franch v. Ankney, 341 Md. 350, 364, 670 A.2d 951 (1996) (citation omitted).

It is quite clear that an expert's conclusion is inadmissible if it requires that witness to resolve internally conflicting material facts involved in the question, prior to rendering his opinion. The rationale for excluding conclusions based on a question having internally contested facts is that the expert is required to make a judgment which invades the jury's province. Under such circumstances, the proper way to submit a hypothetical question is to ask the witness to presume the truth of certain facts as if they were not the subject of dispute.

Kruszewski v. Holz, 265 Md. 434, 444-445, 290 A.2d 534 (1972) (citation omitted).

According to Md. Rule 5-702, which codified the modern common-law rule regarding expert testimony, a trial court must determine whether the evidence to be presented is a proper subject of expert testimony. The inquiry turns on whether the trier of fact will receive appreciable help from the expert testimony in order to understand the evidence or to determine a fact in issue. The trial court need not consider whether the trier of fact could possibly decide the issue without the expert testimony. Nor must the subject of the expert testimony be so far beyond the level of skill and comprehension of the average layperson that the trier of fact would have no understanding of the subject matter without the expert's testimony.

Sippio v. State, 350 Md. 633, 648-649, 714 A.2d 864 (1998) (citations omitted). The trial court must determine whether the three requirements of Md. Rule 5-702 have been satisfied. Id.

We now turn to the specific subjects on which expert witnesses may offer their opinion in the context of forensic pathology. Appellant argues that none of the State's experts should have been allowed to testify as to whether the death in this case was the result of a homicide, which is a finding on manner of death. Appellant argues that a finding on the manner of death is the responsibility of the jury, and he cites case law from various jurisdictions in support of this claim. We can resolve this question with Maryland law.

The recent case of Sippio v. State is instructive. In that case, the Court of Appeals considered whether a trial court erred by allowing Dr. Smialek, the Chief Medical Examiner, to opine that a woman, who died of a gunshot wound, was the victim of a "homicide" rather than an "accident." A crucial issue was whether the medical examiner could testify about the "manner" of the deceased's death in addition to the medical "cause" of her death. The Court stated that the "cause" of death refers to the disease, process, or condition that led to the death, and the "manner" of death refers to whether the death was natural or unnatural. A death deemed "unnatural" may be sub-classified as an accident, suicide, homicide, or "undetermined." Sippio, 350 Md. at 643, 714 A.2d 864.

In Maryland, statutes define the role that medical examiners perform in investigating deaths. Maryland Code (1982, 1994 Repl.Vol., 1999 Supp.), Health-General Article ("H.G."), § 5-301 et seq. H.G. § 5-311 provides, in pertinent part:

(a) (1) The Chief Medical Examiner and, as to their respective counties, each of the deputy medical examiners shall keep complete records on each medical examiner's case.

(2) The records shall ... include:

(i) The name, if known, of the deceased;

(ii) The place where the body was found;

(iii) The date, cause, and manner of death; and

(iv) All other available information about the death.

(Emphasis added.)

In Sippio, the Court of Appeals described the statutory framework as follows:

Section 5-301 et seq. of the Health-General Article establishes the State Postmortem Examiners Commission and sets forth the procedures for the medical examiner to follow where death occurs as a result of, for example, suicide, violence, etc. Where such deaths occur, § 5-309 requires the medical examiner to investigate. Section 5-311 requires the medical examiner to keep complete records of each such case. As part of the medical examiner's investigation, the medical examiner receives notice from the police or sheriff of "facts concerning the time, place, manner, and circumstances of the death." § 5-309(b).^[1] The medical examiner shall perform an autopsy if the medical examiner considers it necessary. § 5-310. If so, the autopsy report is attached to the record of the medical examiner's case pursuant to § 5-311(b). After the medical examiner's report and autopsy are completed and after performing an investigation, the medical examiner then "deliver[s] to the State's Attorney for the county where the body was found a copy of each record that relates to a death for which the medical examiner considers further investigation advisable." § 5-311(c). This record5 can be used as "competent evidence in any court in this State of the matters and facts contained in it." § 5-311(d)(2).

Before 1990, it was the practice of the medical examiner to record the manner of death on a death certificate form. It was not until a 1990 amendment to the Health-General Article, however, that the legislature specifically added "manner of death" to the list of items that a medical examiner was to record in the records of each case. See Chapter 238 of the Acts of 1990 (amending § 5-311(a)(2)(iii)). At that time, the legislature did not define manner of death, nor did it mandate how manner of death should be expressed in the medical examiner's records.

5 "Record `[m]eans the result of a view or examination of or an autopsy on a body' but `[d]oes not include a statement of a witness or other individual.'" [Citing H.G. § 5-311(d)

Sippio, 350 Md. at 645-646, 714 A.2d 864 (some footnotes omitted).

In reaching a conclusion as to the cause and manner of death, the medical examiner is to use information provided to him by the police, including "the known facts concerning the time, place, manner, and circumstances of the death." H.G. § 5-309(b) (emphasis supplied). The medical examiner may use information outside the autopsy itself to make the determinations required by law. Indeed, such outside information may be crucial to a manner of death determination.

In Schlossman v. State, 105 Md.App. 277, 659 A.2d 371 (1995), for example, this Court held that a medical examiner's opinion concerning the manner and cause of a man's death had an adequate foundation, even though it was based on information gathered from police reports. Schlossman was charged with involuntary manslaughter in connection with the death of Baldwin, whom Schlossman found unconscious and severely intoxicated on his property. Schlossman and his companions had poked Baldwin with sticks, urinated on him, and poured paint on him. When Baldwin showed signs of consciousness, Schlossman and his companions rolled Baldwin into a ditch four feet deep, threw stones and a mattress at him, and kicked trash on him. Baldwin attempted unsuccessfully to crawl out of the ditch. The next day Schlossman and his friends returned to the location and discovered that Baldwin was dead. They buried Baldwin, breaking his leg to fit him into the hole they dug. Schlossman, 105 Md.App. at 281, 659 A.2d 371.

The following year, Baldwin's body was discovered by Annapolis police officers. An assistant medical examiner, Dr. Golle, conducted an autopsy and discovered that Baldwin's coronary arteries were almost completely obstructed by atherosclerosis and that the body had several fractures and lacerations, although these may have occurred after death. Dr. Golle was initially unable to specify a cause of death. Schlossman, 105 Md.App. at 281-282, 659 A.2d 371.

After receiving copies of police reports and witnesses' statements, Dr. Golle concluded that Baldwin died of severe coronary artery disease and that the manner of death was a homicide; the doctor stated that Baldwin died of a "heart attack while involved in an altercation." Schlossman, 105 Md.App. at 282, 659 A.2d 371. Dr. Golle also testified that Baldwin had had serious health problems related to his alcoholism, including cirrhosis of the liver, alcohol liver disease, delirium tremens, seizures, and chronic obstructive pulmonary disease. In Dr. Golle's opinion, the stress Baldwin experienced while being harassed by Schlossman and his companion caused Baldwin to suffer a myocardial ischemia, similar to a heart attack. Dr. Golle testified that he formed his opinion about the cause of death "based on his review of the autopsy and information in the police reports." Schlossman, 105 Md.App. at 294, 659 A.2d 371.

Schlossman challenged his conviction, arguing that Dr. Golle's opinion lacked an adequate foundation. This Court affirmed, stating:

[Schlossman] has not stated, nor have we found, any evidence demonstrating that Dr. Golle's opinion as to the cause of the victim's death was based on facts that were not adduced at trial. Dr. Golle's responses indicate that he did not rely on the specific version of the events set forth in the police reports in concluding that Baldwin was the victim of a homicide. In other words, he did not conclude that the victim died as a result of injuries that he may have sustained after allegedly being hit with. stones, trash, paint, and urine, or as a result of any other of the specific acts mentioned in the police reports. In contrast, Dr. Golle based his conclusion on the simple fact that the victim was involved in a stressful "altercation" with [Schlossman] and others; whether the victim suffered any specific injuries as a result of the altercation was immaterial to Dr. Golle. That the victim was involved in such an altercation with [Schlossman] the day before he was found dead was established by evidence at trial, prior to Dr. Golle's testimony, through witnesses who described the events surrounding the victim's death. Under these circumstances, we conclude that Dr. Golle's testimony was supported by a sufficient factual basis and that the court did not abuse its discretion in admitting it.

Schlossman, 105 Md.App. at 295-296, 659 A.2d 371.

For a pathologist,

[d]etermining the manner of death involves correlating the circumstances that surround the death with the findings at the autopsy and any eyewitness accounts.

The medical examiner may also attempt to determine the sequence of events with the information available. Thus, the medical examiner has a wide range of information at his disposal in order to make a finding. R. Taylor, R. Bux, and D. Kirk, Forensic Pathology in Homicide Cases, 40 Am.Jur. Trials 501, 540, 541 (1990).

It is not outside the forensic pathologist's duties to develop a crime scene scenario if an acceptable theory has not already been advanced. It is possible that none of the advanced theories explain the evidence, in which case a new theory is needed. There also may be so little evidence that it is impossible to make any statements as to the circumstances of the death.

Taylor, et al., 40 Am.Jur. Trials at 581.

A good example of the steps a medical examiner takes in developing an opinion is contained in Sippio. In that case, the defendant admitted that he had fired the shot that killed the deceased. His defense was that the shot was accidental. The medical examiner, Dr. Smialek, testified, in pertinent part, as follows:

[Dr. Smialek:] I had information that [Sippio] had told the police that he had shot Ms. Branch.

[Defense Counsel:] And did that aid you in coming up with the conclusion that it was not a natural, accidental, suicidal or undetermined cause of death?

[Dr. Smialek:] I considered that information together with the physical findings on the body, the fact that the wound was not a typical contact gunshot wound such as I would see in a suicide.

So that the form from the investigation together with my findings at the autopsy allowed me to reach a conclusion that this was a homicide, which means that someone else fired a weapon to kill Ms. Branch." (Emphasis added).

* * *

[Defense Counsel:] If a shooting is an accidental shooting and you examined the body of that accident victim not knowing whether it is an accident or not, and the cause of death is a gunshot wound to the head, would you use the block homicide to check off your findings?

[Dr. Smialek:] I'm not sure I understand your question.

If the information available to me indicates that a gunshot wound was the result of an accident such as a gun falling onto a floor and discharging, I would call [it] an accident.

Is that what the form from the investigation to go with my examination led me to believe? I would not call an accident a homicide. I wouldn't call a homicide an accident.

* * * [Defense Counsel:] And homicide has a lot of different categories, does it not, sir?

[Dr. Smialek:] Legally there are categories for homicides, yes.

[Defense Counsel:] What would those categories include, if you know?

* * *

[Dr. Smialek:] There are categories that include self defense, categories that allow a homicide that's caused in the course of say police action to be excusable.

Those are some of the categories. Self defense is a categorization of homicide.

[Defense Counsel:] All of that would be included under your check mark of homicide, correct?

[Dr. Smialek:] I don't consider those particular factors.

What leads to the homicide, whether it was intentional or unintentional in reaching my conclusion, those are legal issues.

[Defense Counsel:] So intent, what caused that person to be on your table, is not part of your conclusion in this report, is that correct?

[Dr. Smialek:] That's right. The intent of the person who pulls the trigger isn't something that I can consider.

Sippio, 350 Md. at 650-662, 714 A.2d 864 (emphasis in original).

The definition of "homicide" is central to an understanding of this issue, and, indeed, appellant argues that a finding of homicide on its own somehow inevitably leads to a conclusion of appellant's guilt in this case. Homicide is the "killing of one person by another." Black's Law Dictionary 739 (7th ed.1999). The sixth edition of BLACK'S explained:

Homicide is not necessarily a crime. It is a necessary ingredient of the crimes of murder and manslaughter, but there are other cases in which homicide may be committed without criminal intent and without criminal consequences.... The term "homicide" is neutral; while it describes the act, it pronounces no judgment on its moral or legal quality.

Black's Law Dictionary 734 (6th ed.1990). See Sippio, 350 Md. at 654, 714 A.2d 864. The seventh edition states:

"The legal term for killing a man, whether lawfully or unlawfully, is `homicide.' There is no crime of `homicide.' Unlawful homicide at common law comprises the two crimes of murder and manslaughter. Other forms of unlawful homicide have been created by statute: certain new forms of manslaughter (homicide with diminished responsibility, and suicide pacts), infanticide, and causing death by dangerous driving."

Black's Law Dictionary 739 (7th ed.1999)' (quoting Glanville Williams, Textbook of Criminal Law 204 (1978)) (emphasis in original).

We return to the question of whether the pathologists' testimony in this case about the manner of death constituted a finding on an ultimate issue. In deciding this issue, the Sippio Court pursued a two-part analysis. First, the Court addressed the defendant's argument that it was improper for the medical examiner to testify that the manner of death was homicide because that testimony, "when juxtaposed with such concepts as accident, suicide, and natural causes ... clearly takes on a criminal connotation," and reached a legal conclusion reserved for the jury. Sippio, 350 Md. at 643, 714 A.2d 864. The Court reasoned that, in light of the Health-General Article's requirements concerning medical examiners, to prohibit medical examiners from testifying about their findings as to manner of death, which those witnesses are "required by law to denote and record for possible use at trial ... would be akin to holding that medical examiners are not qualified to determine manner of death, or that medical examiners' findings are generally unreliable evidence in a court of law." Sippio, 350 Md. at 647, 714 A.2d 864. The Court held that the legislature's inclusion of the phrase "manner of death" in the 1990 statutory amendment referenced above "made it abundantly clear that the legislature intended to bring the determination of manner of death into the province of the medical examiner's duties." Sippio, 350 Md. at 647, 714 A.2d 864.

In Benjamin v. Woodring, 268 Md. 593, 605, 303 A.2d 779 (1973), the Court of Appeals affirmed a trial court's decision not to admit into evidence a death certificate classifying the decedent's death as a suicide, because doing so would have supported the argument of a party seeking to show that the decedent had a reduced testamentary capacity before his death. In Sippio, the Court distinguished Benjamin and quoted approvingly Terry v. State, 34 Md.App. 99, 366 A.2d 65 (1976), in which this Court also distinguished Benjamin. In Terry, this Court explained:

In spite of the suggestion by the Court [in Benjamin ] that the investigative duties of medical examiners are limited by law to `essential facts concerning the medical causes of death,' we cannot conceive that this precludes calling the medical examiner as an expert witness to express his opinion in a case. Once called, testifying under oath, subject to the requirement that he state the basis for his conclusion and be subject to cross-examination, an entirely different situation exists than an effort to introduce that opinion into evidence solely on the basis of a death certificate.

Terry, 34 Md.App. at 107-108, 366 A.2d 65. Thus, in Sippio, as in Terry, the reviewing court concluded that there is no per se prohibition on receiving testimony from medical examiners about the manner in which a decedent died. Sippio, 350 Md. at 648, 714 A.2d 864.

In Sippio, the Court of Appeals then analyzed the medical examiner's specific testimony in light of the general legal standard for expert testimony. The Court held that the medical examiner was qualified as an expert, thus satisfying Md. Rule 5-702(1); that expert testimony was appropriate for the subject about which the medical examiner was testifying, i.e., the results of a gunshot wound on a human body, thus satisfying 5-702(2); and that a legally sufficient factual basis existed to support the expert's testimony. Rule 5-702(3). Sippio, 350 Md. at 649-653, 714 A.2d 864.

The Court of Appeals in Sippio considered and rejected the defendant's argument that the medical examiner's testimony was improper because it was a legal conclusion "appropriately committed to the judgment of the trier of fact." Sippio, 350 Md. at 653, 714 A.2d 864. The Court of Appeals rejected this contention for two reasons: first, Md. Rule 5-704(a) specifically provides that opinion testimony "otherwise admissible is not objectionable merely because it embraces an ultimate issue to be decided by the trier of fact." If the opinion testimony of an expert embraces an ultimate issue to be decided by the jury, the trial court determines, under Md. Rule 5-702, whether the testimony will be helpful to the trier of fact. Sippio, 350 Md. at 654, 714 A.2d 864. The exception to Rule 5-704(a), as provided in 5-704(b), is that an expert is precluded from testifying as to whether the defendant possessed the requisite mental state that constitutes "an element of the crime charged." The Court determined that the medical examiner's testimony did not trigger this one exception to Rule 5-704(a), and it was thus permissible. Sippio, 350 Md. at 654, 714 A.2d 864.

The Court also rejected Sippio's argument because "whether [the decedent's] death was a homicide was not the ultimate issue in this case, and [the medical examiner's] opinion on [the decedent's] manner of death was not a legal conclusion reserved for the trier of fact." Sippio, 350 Md. at 654, 714 A.2d 864. After explaining "homicide" does not itself necessarily denote a crime, the Court observed: "To secure a conviction of first- or second-degree murder, the State bore the burden to prove beyond a reasonable doubt that Sippio intentionally shot [the decedent]. Only the jury could decide whether Sippio possessed the requisite mental state for such a conviction." Sippio, 350 Md. at 655, 714 A.2d 864. The jury, before assessing criminal culpability, first had to decide whether a homicide had even occurred. The Court noted that "[the medical examiner] did not testify to Sippio's intent, but rather merely testified that [the decedent's] death occurred as a result of a homicide." Sippio, 350 Md. at 655, 714 A.2d 864. The Court held:

[H]omicide in itself does not equate to criminal culpability; instead it is the killing of one human being by another..., regardless of the intent of the party who commits the act. We, thus, find no merit to Sippio's contention that [the medical examiner's] testimony invaded the province of the jury.

Sippio, 350 Md. at 655, 714 A.2d 864.

In the present case, the essence of a SIDS diagnosis was a central but not the ultimate issue. According to the testimony of the State's expert witnesses, SIDS is not itself a specific cause of death. Rather, SIDS is a classification given to infant deaths that match certain criteria but have no discernible cause. In other words, it is a diagnosis of exclusion. SIDS diagnoses are made when babies who appear to have been in good health suddenly stop breathing with no apparent physical explanation. Fewer deaths were diagnosed as "SIDS deaths" in the United States in the 1990s than in the 1970s or 1980s. This decrease coincides with the "Back to Sleep" publicity campaign, which encourages parents to put their infants to sleep on their backs, rather than on their stomachs. The decrease in the number of SIDS diagnoses also could be attributed to the increased ability to identify the exact cause of death.

The State's medical experts testified that deaths labeled as SIDS deaths are generally not viewed as being caused by genetics, although appellant's medical expert testified that some infants may have genetic predispositions that render them more apt to suffer a SIDS-type death. The experts testified that studies show that deaths labeled as SIDS deaths occur more commonly under certain circumstances: in non-Caucasian families, in homes with tobacco smokers, and when babies are put to sleep face down. Of course, if a specific, recognizable cause of death is identified for a particular deceased infant, such as a reaction to tobacco smoke or suffocation caused by a blanket blocking the infant's airways, that death would no longer be properly classified as a SIDS death.

The State's experts noted that many deaths that were attributed to SIDS in the past may have actually had other causes. In this context, the investigative report takes on particular significance, because it can give the medical examiner clues to explain the death that the autopsy did not. In this case, the investigation reports proved especially important.

Prior to addressing the substance of appellant's arguments with respect to each of the State's expert witnesses, we must dispose of the State's argument that appellant failed to preserve this issue for appeal. We note that, prior to the testimony of the experts, appellant requested and received a standing objection to "any medical witnesses ... to testify to either manner or cause of death because the State has not finish[ed] its amendment to a vital record procedures [sic] as set out under Health General Section 4-214 to modify the death certificate." Generally, grounds for objection not raised at trial and then brought up on appellate review are not preserved. Fearnow v. Chesapeake & Potomac Telephone Co. of Maryland, 342 Md. 363, 379, 676 A.2d 65 (1996) (citing Black v. Leatherwood Motor Coach Corp., 92 Md.App. 27, 33-34, 606 A.2d 295, cert. denied, 327 Md. 626, 612 A.2d 257 (1992)).

Nevertheless, appellant did raise the grounds below that he has raised on appeal, although he did not make the objection until after Dr. Koke and Dr. Dixon had testified. The objection took place prior to the testimony of Drs. Smialek and Norton:

I again renew my motion, and I want to—that he should not be allowed to testify to any of the facts that would contravene or go against the original autopsy findings, without having followed the proper procedure.

I am going to ask you at this time to strike the testimony of Dr. Kokes and Dr. Dixon on the basis that their testimony was not expert testimony.

And the same is going to apply for Dr. Norton later. I will ask you not to receive hers and, if you do, to strike it, because these witnesses all are, as I think I figured out—are rendering opinions that are solely within the jury's province.

They are usurping the jury's position, because if I understand it, there are no changes in any medical testimony. They are all using either statistics or extraneous—extraneous—other stuff, like life insurance, as a basis. Those are things that are solely within the province of the jury.

The trial court was thus on notice that appellant was raising objections on this ground. Not specifically designated as a continuing objection with respect to Drs. Smialek and Norton, appellant continued to object to the experts' testimony.

Although not a model of preservation, we believe that we may nevertheless address the substance of these issues. All of the experts discussed below were qualified as experts under Md. Rule 5-702(1) and were testifying on subjects appropriate to their area of expertise—forensic pathology. Md. Rule 5-702(2). Thus, our discussion will focus on whether each expert's opinion was supported by a legally sufficient factual basis. Md. Rule 5-702 (3). With respect to the latter issue, all of the information from the investigative reports the experts had relied on in formulating their opinions and findings had been entered into evidence prior to the experts' testimony.

Dr. Dixon

Dr. Dixon testified that during Brandi's autopsy in 1981 she found no apparent signs of injury or disease. There was no swelling of the brain. Because she was unaware of any suspicious circumstances concerning the death, in 1981 Dr. Dixon labeled the cause of death as SIDS and the manner of death as "natural." She identified certain elements of the autopsy findings, such as pinpoint hemorrhages on the thymus gland, in front of the heart, and on the surfaces of the lungs and heart, which were consistent with suffocation but were not enough, by themselves, to lead her to opine that suffocation was the cause of death.

In 1998, detectives contacted Dr. Dixon and asked her to analyze Brandi's death again. Based upon new information provided to her, including witnesses' statements and information about the insurance policies, she changed her conclusion regarding the cause of Brandi's death to "probable suffocation" and the manner of the death to "undetermined." She defined suffocation as the cutting off of the baby's air supply by "some external means." Dr. Dixon opined that Brandi did not die naturally, but she could not definitively conclude that the death was a homicide.

Appellant argues that, because Dr. Dixon stated that the police reports contained "useful information," for her "the insurance was a factor in [her] ultimate conclusions and [she] drew an incriminating inference from it.... [R]esolution of such contested facts is for the jury, not the experts." Regarding Dr. Dixon and the other expert witnesses, appellant contends that their testimony was erroneously admitted because it "encroached on the jury's function to judge the credibility of the witnesses and weigh their testimony and on the jury's function to resolve contested facts."

We conclude that Dr. Dixon's testimony was properly admitted. She used information in police reports as well as her autopsy to come to her conclusions on the cause and manner of death. Dr. Dixon did not, as appellant contends in his first theory of error, judge the credibility of other witnesses. She did read witnesses' statements contained in the police reports, but she, like all of the experts, conceded that if those statements were wrong or false her opinion could change:

Q [by defense counsel]: ... So you listened to [the detective]. She tells you some stuff. And based on the stuff she tells, you accept it as the Gospel.

And based on accepting that as the Gospel, you, as a medical examiner, change your opinion—not based on a scientific fact but based on an investigation, true?

A: I do not accept it as the Gospel, and I do not accept it as fact as such. But I accept it as useful information which coupled with my essentially negative autopsy, which is the same essentially negative autopsy that was in 1981.

Now with this new information, yes, I think it is throws [sic] a different light on the whole matter. And, yes, I changed the manner and cause of death appropriately.

* * *

We are not dealing with just one piece of information here. I am dealing with the whole amount of information that was furnished to me. I am not taking an isolated fact.

Regarding the second of appellant's two theories, that Dr. Dixon testified about an ultimate issue in the case, we first note that, like the medical examiner in Sippio, Dr. Dixon's testimony did not concern the ultimate issue at trial. At appellant's trial the ultimate issue was whether he murdered Garrett Michael. Dr. Dixon's testimony concerned whether Brandi's death was due to natural causes, an accident, or the actions of another human being, and not appellant's mens rea. Moreover, Dr. Dixon did not opine that Brandi definitely had been suffocated by another human being. She merely classified the cause of her death as a "probable suffocation," without a specified cause, and the manner of her death as "undetermined." The trial court did not err in admitting Dr. Dixon's testimony.

Dr. Kokes

The State also presented testimony from Dr. Charles Kokes, a forensic pathologist who had worked in the Maryland Medical Examiner's Office in 1987 and performed the autopsy on Garrett Michael. In performing the autopsy, Dr. Kokes discovered no evidence of external or internal trauma or abnormalities that would explain Garrett Michael's death. The baby's brain was swollen, but he believed that could have been due to resuscitative efforts made by medical personnel. The baby had also had ventricular fibrillation, an irregular heartbeat, during the ambulance ride to the hospital. At the time of the autopsy, Dr. Kokes labeled the cause of death as SIDS and the manner of death as natural. No police investigation was conducted at the time of Garrett Michael's death, so Dr. Kokes relied solely on information generated by medical personnel and his own autopsy.

In 1997, Maryland prosecutors brought Dr. Kokes more information regarding Garrett Michael's death, including statements from Ms. Anastasi, information about Brandi Jean's death, and the insurance information. After considering all of the information now available to him, Dr. Kokes determined that the cause of Garrett Michael's death was smothering and the manner of the death was homicide.

Dr. Kokes testified that, apart from the medical findings and investigatory reports, one of the reasons he now believed that the manner of death was homicide was because of the rarity of two or more SIDS deaths occurring in the same family. He stated that it is "common knowledge in the general field of forensic pathology" that SIDS deaths rarely reoccur within the same family.

Dr. Kokes also provided more specific statistics obtained from death certificates within the United States from 1979. Despite the age of the statistics, Dr. Kokes stated that the data remained consistent through 1987. For every 1,000 live births of Caucasian infants, 1 or 2 deaths occurred that were labeled as SIDS deaths.^[2] Brain swelling (edema) was present in only 1 out of every 100 deaths attributed to SIDS. Dr. Kokes testified that the probability of Garrett Michael dying from SIDS, with edema, was therefore 1 in 100,000 (i.e., 1 in 1,000 live births, multiplied by 1 in 100 SIDS cases with edema). Dr. Kokes explained that, in light of that statistic, the probability that a second baby in the same family would also die from SIDS would be 1 out of 100,000,000 (i.e., 1 in 100,000 multiplied by 1 in 1,000). He characterized this figure as "a very rough estimate," but he said that the odds of such a coincidence were "so low [as] to make it impossible."

Dr. Kokes stated that, in his opinion, SIDS is not caused by genetic factors. He agreed that SIDS deaths occur more often in homes with tobacco smokers and also occur more often when babies are put face down in their beds. Dr. Kokes testified that he believed an obstruction was placed over Garrett Michael's airways, causing his death, "either a hand or something akin to a towel or pillow," or perhaps a blanket. Dr. Kokes added that, because it takes less force to smother an infant than it does an adult, markings that might indicate smothering are often absent.

In evaluating the testimony of Dr. Kokes, we engage in the same analysis as we did in evaluating Dr. Dixon's testimony. We hold that Dr. Kokes's testimony satisfied the three-pronged test of Md. Rule 5-702. Like Dr. Dixon's testimony, Dr. Kokes's testimony did not address the ultimate issue of whether appellant murdered his infant son. Rather, Dr. Kokes's testimony addressed the threshold question, the cause and manner of Garrett Michael's death:

Q [by the State]: And so it would be accurate to say that [a second infant dying in a family that had a SIDS death with edema] would occur in 1 in 100,000,000 live births?

A: Well, that is going to be a roughly— a very rough estimate. These numbers serve to illustrate the basis for my change in opinion regarding, first of all, the manner of death.

And that is how I have approached review of this particular case—by looking first of all at the investigative information that was brought forth and then looking at the anatomic pathology finding[s] which have not changed since 1987, and re-examining those, reviewing those in light of this new investigative information.

He did not identify the person who may have been responsible for the baby's death.

As was the case in Sippio, and indeed with all of the medical experts who testified for the State at appellant's trial, Dr. Kokes was subject to vigorous cross-examination.^[3] Appellant was able to demonstrate to the jury elements of Dr. Kokes's reasoning process, particularly concerning the statistical evidence and Dr. Kokes's unfamiliarity with some of the sources of his knowledge, which might cause the jury to question Dr. Kokes's opinions.

The trial court did not err in admitting Dr. Kokes's testimony. His testimony concerned a finding based on factors that he could consider and that were in evidence.

Dr. Smialek

Dr. John Smialek, Maryland's Chief Medical Examiner, explained how the medical examiner's office goes about evaluating cause and manner of death in various cases:

A: Well, the first element of a determination of cause and manner of death, as set out in the statute, is that we investigate the circumstance of the death.

And that includes information regarding where that person was found, any information that family members might provide about that particular death, information regarding the person's medical condition.

And then the other element is the autopsy examination, the physical examination of the body. That includes an external and internal examination of the organs, then a microscopic examination of the internal structures, a toxicological examination, and any other tests that might be indicated in that particular death.

All of that information is then evaluated, and we are responsible for reaching conclusions as to the cause and manner of death after that has been completed.

Q [by the State]: Why is it important to gather investigatory information about the circumstances of the death in order for you to make a determination of the cause and manner of death?

A: Well, most importantly, the investigation assists a medical examiner in determining the manner of death.

For example, a cause of death may be determined to be a gunshot wound to the head. That would certainly explain why that person died.

But the circumstances might indicate that that person committed suicide, fired that weapon themselves and caused their own death.

Or there might be evidence to indicate that the gun fell to the floor, went off accidentally, and struck that person in the head. That would make the death accidental.

Or there might be information to indicate that someone else fired the weapon, and that would make the manner of death a homicide. All with the same cause of death.

So, it is an essential part of the function that a medical examiner carries out to review evidence regarding the circumstances of the death.

Dr. Smialek went on to testify about SIDS deaths in particular. He explained that babies die from SIDS when a central nervous system abnormality interrupts heart activity and breathing. He stated that in 1989 a commission sponsored by the National Institutes of Maternal Health altered the recommended criteria for a SIDS diagnosis, lowering the upper age limit from two years to one and suggesting that a complete examination, including an autopsy, an evaluation of the scene where the baby died, and a review of medical records, should be conducted before a SIDS diagnosis was made. Although Dr. Smialek did not review autopsy results of either child at the time they died, he reviewed both autopsy reports in 1997 in addition to information about the insurance policies and witness statements. Dr. Smialek testified, however, that, in making his conclusions about the deaths, he did not rely on statistical probabilities or on information about the insurance policies. He did rely on information about objects in both infants' cribs and on witnesses' statements that he evaluated in the context of this case and in light of his experience as a medical examiner.^[4]

He concluded that Brandi was probably suffocated, although he did not suggest a particular manner in which her death occurred. He amended the autopsy report on her death to list the manner of death as "undetermined."

Dr. Smialek testified that the swelling present in Garrett Michael's brain was not consistent with SIDS and suggested that his air supply may have been cut off for several minutes. Furthermore, Garrett Michael's lungs were "expanded," which also suggested suffocation and not SIDS. Dr. Smialek saw no independent physical evidence of an injury that would have cut off Garrett Michael's air supply, and he was aware that, according to Ms. Anastasi, there was nothing in the crib that could have accidentally obstructed the baby's airway. Dr. Smialek testified that the most probable cause of the obstruction was the physical act of another person, and he therefore concluded that Garrett Michael's death was a homicide. The autopsy report on Garrett Michael's death was changed to reflect this.

The trial court did not err in admitting Dr. Smialek's testimony. His testimony concerned a finding based on factors that he could consider and that were in evidence.

Dr. Norton

The fourth medical expert to testify for the State was Dr. Linda Norton, a forensic pathologist in private practice who specializes in pediatric deaths. She had previously been a medical examiner in Texas and in Alabama. She had written one article about SIDS and frequently lectured to social workers, law enforcement officers, and prosecutors about SIDS, child abuse, and child deaths. After reviewing the prosecution's files on the case, which included autopsy reports for both babies, medical records, photographs of the bodies and the death scenes, and witnesses' statements, she concluded that the cause of Brandi's death was suffocation and the manner of death was homicide.

Photographs of Brandi lying face down in her crib showed a lividity pattern caused by blood settling down toward the front of the face, but also showed that the forehead and nose were lighter in color. From these photos Dr. Norton concluded that the baby's face had been pressed against something, such as the mattress, thus preventing blood from collecting in the nose and forehead. She testified that the position in which Brandi was found, with her face pointing straight down into the mattress, is an unnatural sleeping position not normally adopted by humans. Brandi also had froth around her nose, indicating that she had attempted to breathe against an obstruction. In reaching her conclusions about Brandi's manner of death, Dr. Norton considered the fact that Garrett Michael had died in similar circumstances several years later.

Dr. Norton also concluded that Garrett Michael was suffocated and that the manner of his death was a homicide. She reached her conclusion after considering all of the circumstances surrounding the boy's death, including the insurance policies and Ms. Anastasi's statements about listening to appellant in the baby's room through the monitor.

Dr. Norton testified that SIDS deaths occur for approximately 1 baby out of every 2,000 live births. This statistic was from information collected from medical examiners' offices in the United States that employ the minimum criteria for adequately identifying deaths as SIDS deaths. She concluded that the probability of two SIDS deaths occurring in the same family is therefore 1 in 4,000,000 (i.e, 1 in 2,000 multiplied by 1 in 2,000). She testified that fewer deaths were diagnosed as SIDS deaths in the late 1990s than during the 1980s, and that part of this change might be attributed to the actions of parents, who have been encouraged by child-care experts to put babies to sleep on their backs instead of on their stomachs. Dr. Norton also stated that SIDS is not genetic. Although she admitted that research was still being conducted in this area, she doubted that any new links between SIDS and genetics would be uncovered.

Appellant argues that the trial court erred by allowing Dr. Norton to testify to her opinion that the deaths were homicides. As with the other experts, appellant contends that Dr. Norton's opinions invaded the province of the jury to judge the credibility of the witnesses and to resolve contested facts.

We note, first, that Dr. Norton did not testify concerning the credibility of other witnesses except insofar as defense counsel forced her to make such an assessment between Ms. Anastasti's statement and Megan Churchill's statement. Dr. Norton recognized that Churchill's statement was hearsay and admitted that Ms. Anastasi's might hold more sway because she was directly involved. She did clarify, however, by noting that Churchill's statement also conflicted with evidence from the emergency technicians as to when they arrived on the scene and the fact that there were still electrical impulses in the heart. This was not a judgment on credibility, on who was telling the truth, and, if it were, appellant could hardly complain, since he posed the question. Klauenberg, 355 Md. at 543, 735 A.2d 1061.

Second, we reject appellant's contention that Dr. Norton's testimony invaded the province of the jury to resolve contested facts. Dr. Norton specifically stated that she developed her opinion "based on the totality of the records, and... not based on presuming that any one single person is telling the absolute truth or that any one single person is telling all lies." She looked at the scene of death photographs for Brandi, witnesses' statements, police reports, the fact that appellant was the last person to be with both children after having previously shown little interest in the babies, and the fact that there had been insurance policies on both children. Dr. Norton recognized that people may insure their children for a variety of reasons, but she also noted that it provides a motive, particularly when viewed in light of the other evidence in the case.

This is not a case like Bohnert v. State, 312 Md. 266, 539 A.2d 657 (1988), where the defendant was convicted of sexual child abuse after Temple, a social worker, testified that, based on the statements of the victim, the victim's mother, and another woman, she believed that the victim had, in fact, been sexually abused. The Court of Appeals reversed the conviction, stating:

The opinion of Temple that Alicia in fact was sexually abused was tantamount to a declaration by her that the child was telling the truth and that Bohnert was lying. In the circumstances here, the opinion could only be reached if the child's testimony were believed and Bohnert's testimony disbelieved. The import of the opinion was clear— Alicia was credible and Bohnert was not. Also, the opinion could only be reached by a resolution of contested facts—Alicia's allegations and Bohnert's denials.

Thus, the opinion was inadmissible as a matter of law because it invaded the province of the jury in two ways. It encroached on the jury's function to judge the credibility of the witnesses and weigh their testimony and on the jury's function to resolve contested facts. Inasmuch as the opinion was inadmissible as a matter of law, it was beyond the range of an exercise of discretion.

Bohnert, 312 Md. at 278-279, 539 A.2d 657.

In the present case, in contrast, Dr. Norton's opinion testimony, in addition to not judging the credibility of other witnesses, could be reached without a resolution of contested facts. That she considered statements contained in the prosecution's files did not reflect her acceptance of the statements as credible; the credibility of those statements were separate, individual issues and any successful attack by appellant against the credibility of the statements would render Dr. Norton's opinion less weighty. In other words, Dr. Norton's opinion was reached in the context of the medical findings. By contrast, the social worker in Bohnert reached her opinion despite the medical findings, concluding that no sexual abuse had taken place. Bohnert, 312 Md. at 270-271, 539 A.2d 657.

We conclude that the trial court did not err by allowing Dr. Norton, or indeed any of the State's expert witnesses, to render opinions regarding the causes and manners of death of appellant's children. Unlike the social worker's testimony in Bohnert, 312 Md. at 278-279, 539 A.2d 657, neither Dr. Norton nor the other expert witnesses presented by the State attempted to resolve contested factual credibility issues. Rather, the information presented provided the backdrop against which they examined or reexamined their medical findings. Certain facts, though probably harmful to appellant, were not seriously contested. They included the circumstances surrounding Brandi's death and their similarities to Garrett Michael's death.

Appellant suggests that it will always be inappropriate for a medical examiner to testify that a death was a homicide, because that term is essentially a label to be placed only by the fact-finder. This position fails to recognize that a determination of homicide is not a determination of a criminal act and, thus, not a determination of a defendant's criminal agency. The statutes governing the conduct of medical examiners clearly instruct them to determine whether deaths are homicides. H.G. §§ 5-309, 5-311. If medical examiners must, by statute, make this determination and report it on the death certificate, they should be able to testify as to how they reached a particular conclusion. See Sippio, 350 Md. at 647, 714 A.2d 864. What is essential is that there is a logical relation between the physical or medical observations and the extrinsic circumstances considered in determining the manner of death.

We conclude further that the trial court did not err by allowing the testimony of Dr. Norton. If medical examiners employed by the State may testify regarding their findings as to the cause and manner of death, it follows that a qualified forensic medical expert would not be precluded from testifying and considering the same information as was considered by the medical examiner. Although the expert in question, Dr. Norton, was testifying for the State, there would be many times when the defense would want the right to call its own witness to counter the State's position, as did the defendant in this case. If independent medical experts were restricted from testifying in regard to issues addressed by the State's experts, the defense would be unfairly prejudiced. Thus, it is more logical to conclude that the legislative determination permitting a medical examiner to consider cause and manner of death would permit a qualified expert to counter or support that determination.

Appellant was able to cross-examine the experts and could and did present evidence, including his own expert witness, challenging their theories. The jury was able to weigh these contesting presentations and factor them into its resolution of appellant's guilt or innocence.

II.

Appellant's second argument is that the trial court erred by admitting statistical evidence concerning the probabilities of two deaths occurring from SIDS in one "family." The two infants involved in this case, of course, were not in a single two-parent biological "family." They had a father in common, but different mothers. This factor immediately modifies a statistical analysis based upon probabilities of events occurring in the same biological family.

The experts, as noted above, employed the "product rule." The Court of Appeals has said that "[s]tated generally, the product rule means that the probability of two events occurring together is equal to the probability that event one will occur multiplied by the probability that event two will occur." Armstead v. State, 342 Md. 38, 69-70, 673 A.2d 221 (1996) (citation omitted). The Court of Appeals went on to explain:

The classic illustration [of the product rule] is coin tossing; the probability of finding "heads" on two successive coin tosses is equal to the probability of heads on the first toss, 50%, times the probability of heads on the second toss, 50%, equaling 25%.

The product rule is valid if the individual events are independent, i.e., if the outcome of the first event does not impact the outcome of the second event. In the coin toss example, this means that the outcome of the first coin toss does not affect the outcome of the second coin toss, which is a valid assumption. By comparison, assume we wish to calculate the probability of having both a checking account and a loan from a particular bank. This is an example of non-independent or linked events. We can not calculate the probability of having both a loan and a checking account at the same bank by multiplying together the individual probabilities under the product rule because a person is more likely to obtain a loan from the bank where he maintains a checking account. To illustrate nonindependence as it applies to human characteristics (although not genetic characteristics) ... men who have beards are probably more likely than others to also have moustaches.

Id. at 70, 673 A.2d 221 (citations omitted).

Appellant argues that SIDS deaths are not unrelated or completely independent events, and he points to the fact that in the United States "SIDS" deaths occur more often in some populations than in others, i.e., in families with lower incomes, in families with smokers, and in ethnic minority families. We note that Dr. Kokes corrected for some of this by looking at SIDS statistics for Caucasian babies.

Where "the presence or absence of one characteristic has no effect on the probability that another characteristic will be present," those characteristics are statistically independent. John M. Kobayaski, Proof of Causation in Toxic Torts and Related Cases, C607 ALI-ABA 1173, 1191 (1991). In the scientific literature, the various risk factors for SIDS, such as prone sleeping position, smoking during pregnancy, etc., are "independent risk factors." John Kattwinket, et al., Changing Concepts of Sudden Infant Death Syndrome: Implications for Infant Sleeping Environment and Sleep Position, 105 Pediatrics 650 (Mar. 1, 2000), available in 2000 WL 10704223 at *2. Although various factors have been present in infants who died of SIDS, however, at least one study has "concluded" that not one or even a combination of these so-called "risk factors" were powerful enough to be predictive of future SIDS victims. Robert M. Reece, Fatal Child Abuse and Sudden Infant Death Syndrome; A Critical Diagnostic Decision, 91 Pediatrics 423 (1993).

In any event, a myriad of factors are involved in the discussion of so-called "SIDS deaths," and those factors may render SIDS rates higher in some populations than in others. For instance, tobacco smoke may have a deleterious effect on infants, and may contribute to their death in a way that is not currently understood. Second, infants in lower-income families may receive less medical care, contributing to more deaths. Finally, there may be an increased effort to determine a specific cause of death, rather than simply labeling a death as a SIDS death, for children whose families have higher incomes. We do not suggest that any of these hypotheses are necessarily accurate; we merely note that appellant's argument that SIDS deaths are interrelated is not totally accurate. It is accurate to assert that the deaths labeled as SIDS deaths may indeed have a connection. The inaccuracy arises once that connection is discovered, because the deaths should no longer be identified as SIDS deaths. We must keep in mind that SIDS is a diagnosis of exclusion that is subject to change if an actual cause is uncovered. The scientific literature has shown that, from what is currently known, SIDS risk factors are not interrelated but are independent. This is particularly true of multiple SIDS deaths in one family, where the likelihood of recurrence is less than one percent. Catherine L. Goldenberg, Sudden Infant Death Syndrome as a Mask for Murder: Investigating and Prosecuting Infanticide, 28 Sw. U.L.Rev. 599, 606 (1999).

In addition to deciding whether the product rule was properly applied in this case, we must also look at the admissibility of this evidence in the context of appellant's right to a presumption of innocence. There are no Maryland cases on point in this area, and the bulk of the cases located have examined the applicability of statistics in the context of establishing paternity. Most recently the Sixth Circuit Court of Appeals noted the following:

[I]t is "susceptible to debate" as to whether the probability of paternity statistics violate the presumption of innocence, inasmuch as courts have allowed the use of these statistics without finding error. See State v. Hartman, 145 Wis. 2d 1, 426 N.W.2d 320, 328 (Wis.1988) (Steinmetz, J., dissenting) (collecting cases from jurisdictions that allow the use of the statistics in question); see also Coe v. Bell, 161 F.3d 320, 353 (6th Cir.1998) (finding that a result is not dictated by precedent "when courts have reached divergent results on an issue before resolution by the Supreme Court"); Cain [v. Redman], 947 F.2d [817,] 821 [(6th Cir.1991)] (noting the fact that the jury instructions found to be unconstitutional were routinely given without challenge rendered their correctness "susceptible to debate among reasonable minds").

Lyons v. Stovall, 188 F.3d 327, 341 (6th Cir.1999), cert. denied, ___ U.S.____, 120 S. Ct. 2197, 147 L. Ed. 2d 233 (2000).

State courts are split on this issue. The Supreme Court of Minnesota, for example, has long cautioned that the admission of statistical evidence runs the risk of "undermin[ing] the presumption of innocence." State v. Boyd, 331 N.W.2d 480, 483 (Minn.1983). However, in the more recent past, particularly in the context of DNA evidence, other courts have allowed admission of statistical evidence after conducting a balancing test. Padgett v.. State, 668 So. 2d 78, 85-86, cert. denied 668 So. 2d 88 (Ala.1995). Other states have found no error at all in the admission of statistics, Griffith v. State, 976 S.W.2d 241, 247 (Tex. App.1998), or no error so long as the statistical approach is accepted in the scientific field for which it is applicable. People v. Venegas, 18 Cal. 4th 47, 74 Cal. Rptr. 2d 262, 954 P.2d 525, 549 (1998).

The Court of Appeals, however, has held that admission of statistical evidence does not violate a defendant's due process rights. Armstead, 342 Md. at 88, 673 A.2d 221. Of course, the presumption of innocence is one part of the larger right to due process. Owens v. State, 352 Md. 663, 695, 724 A.2d 43, cert. denied, 527 U.S. 1012, 119 S. Ct. 2354, 144 L. Ed. 2d 250 (1999); Stanley v. State, 43 Md.App. 651, 655-656, 406 A.2d 693 (1979). In Armstead, there was expert testimony about DNA matches and the probability that a match was random due to laboratory error. The jury was informed of the laboratory error rate, and the defendant was able fully to address the issue of statistics on cross-examination. Armstead, 342 Md. at 88, 673 A.2d 221.

Like the defendant in Armstead, appellant was able fully to explore the issue of statistics through cross-examination. Dr. Kokes was thoroughly cross-examined on the accuracy and origin of the statistics he quoted on SIDS deaths. Appellant also effectively cross-examined Dr. Norton on her statistics, since they were different from those cited by Dr. Kokes. As stated above, Dr. Kokes derived his figure of 1 or 2 SIDS death for every 1,000 Caucasian live births from a review of 1979 death certificates in the United States. Dr. Norton, on the other hand, testified that the rate of SIDS deaths was 1 in every 2,000 live births. Dr. Norton was cross-examined quite effectively by defense counsel as to the accuracy of her figures and those of Dr. Kokes.

In addition, the trial court in this case provided instructions to the jury advising it how it was to use the statistical evidence in this case:

During this trial, you have heard testimony regarding statistical probabilities. Certain experts in rendering their opinions relied in part on the statistical probabilities of a SIDS death occurring twice within the same family.

You may consider this testimony only in evaluating the weight to be given to those opinions. The weight of the evidence does not depend on the number of witnesses on either side.

This instruction clarified to the jury how it was to make use of the statistical evidence presented in the trial. The jury is presumed to understand and follow such instructions. Wilson v. State, 261 Md. 551, 570, 276 A.2d 214 (1971); Brooks v. State, 85 Md.App. 355, 360, 584 A.2d 82 (1991). Thus, we assume that the jury did not rely solely on that evidence in rendering its verdict and used that evidence, to the extent that it did, as directed in the instruction.

We find no error in the trial court's decision to allow statistical evidence and expert testimony using the product rule.

III.

Appellant argues that the trial court erred by failing to take adequate corrective action when the State, during rebuttal closing argument, mentioned a statistical probability as the probability that appellant was innocent. Appellant argues that the trial court should have either granted his motion for mistrial or, in the alternative, instructed the jury in a manner that would forbid it to "use statistics and compare that to the burden of proof beyond a reasonable doubt."

Prior to closing arguments, in a bench conference concerning jury instructions, the trial court asked whether a proposed instruction, which stated in part that the jury could consider statistical evidence "solely in evaluating the weight to be given to the opinions" of the witnesses, would give the jury "a subliminal message that there is something really special about statistics?" The State replied:

Well, it does precisely the other thing. It says that these—the only relevance of statistics in this case is to the doctors, not to us.

I mean, we shouldn't stand up—I shouldn't be able—or [the Assistant State's Attorney] shouldn't be able to stand up and say there is a one-in-200 million chance, which is the biggest number we have right now-a one-in-200 million chance that this man is innocent. We shouldn't be able to say that.

Thus, it is perplexing and disturbing that, during the State's rebuttal closing argument, the State argued to the jury as follows concerning the testimony of Miles Jones, M.D., a pathologist who had testified for the defense:

Dr. Jones said, "It is not SIDS," so it could not have been natural causes. We know it is not suicide. There is no evidence whatsoever that this was an accident.

So, basically what he was telling you was that it was a homicide. You shouldn't be surprised about that because [the Assistant State's Attorney] had the chart out. He said, "What is your philosophy on how you decide, because we all know that SIDS is not genetic? When do you decide that it is a homicide?"

He said, "Well, on the third death. The first one could be SIDS. The second one, I am not going to do SIDS. I will make it undetermined, and the third is a homicide."

Well, the problem was, he wasn't paid $550 an hour to come here and tell you that the second death in this case was a homicide. So, he stuck with undetermined. But of course, he didn't have any information.

But what he also told you in terms of the statistics we have talked about, the doctors relying on the statistics of SIDS, he told you that in his numbers that it was 3 in 1,000 certain SIDS deaths.

The second time, the death could be attributed to SIDS. There is 3 in 1,000 live births that would be a SIDS. Well, let us use his numbers and be conservative.

Assuming it is 3 in 1,000 for the first. Then, he also told you that less than 1 percent of SIDS deaths had the brain swelling, the edema.

If you multiply his numbers, instead of 1 in 4 million, you get 1 in 10 million. So, what he told you basically, that if you take out accident and suicide, which he did—he took out SIDS and he took out any natural causes, 1 in 10 million that the man sitting here is innocent. That was what a doctor, their expert, told you. (Emphasis supplied.)

After the State concluded its rebuttal closing argument, appellant moved for a mistrial based on the State's comment about the "1 in 10 million" chance of innocence. The court denied the motion, and appellant requested that the court give the jury a curative instruction: "I would ask you to ... tell them that you can disregard it [the statistics]. You can never ever, ever, use statistics and compare that to the burden of proof or reasonable doubt. They have no place in this case, and that is what I am asking." The court ruled that, rather than issue a new instruction, it would repeat its earlier instruction on statistics. It then stated to the jury:

Members of the jury, I think [it] is appropriate to review with you one instruction which I read to you earlier and [which] will be contained in your packet.

By doing this, I am not trying to highlight this instruction above all others. I want you to consider this instruction in conjunction with all of the other instructions you have received from the Court.

During this trial you have heard testimony regarding statistical probabilities. Certain experts in rendering their opinions relied in part on the statistical probabilities of a SIDS death occurring twice in the same family.

You may consider this testimony only in evaluating the weight to be given to those opinions and not for any other purposes.

Appellant contends the trial court's action was inadequate to rectify the prejudice caused by the State's comment.

When presenting closing arguments to the jury, attorneys are afforded great leeway. Henry v. State, 324 Md. 204, 230, 596 A.2d 1024 (1991). "The prosecutor is allowed liberal freedom of speech and may make any comment that is warranted by the evidence or inferences reasonably drawn therefrom." Degren v. State, 352 Md. 400, 429, 722 A.2d 887 (1999) (citation omitted).

Generally, ... the prosecuting attorney is as free to comment legitimately and to speak fully, although harshly, on the accused's action and conduct if the evidence supports his comments, as is accused's counsel to comment on the nature of the evidence and the character of witnesses which the [prosecution] produces.

While arguments of counsel are required to be confined to the issues in the cases on trial, the evidence and fair and reasonable deductions therefrom, and to arguments of opposing counsel, generally speaking, liberal freedom of speech should be allowed. There are no hard-and-fast limitations within which the argument of earnest counsel must be confined—no well-defined bounds beyond which the eloquence of an advocate shall not soar. He may discuss the facts proved or admitted in the pleadings, assess the conduct of the parties, and attack the credibility of witnesses. He may indulge in oratorical conceit or flourish and in illustrations and metaphorical allusions.

Wilhelm v. State, 272 Md. 404, 412-413, 326 A.2d 707 (1974) (citations omitted).

It has long been recognized that, due to the nature of criminal trials, "`in the heat of argument, counsel do occasionally make remarks that are not justified by the testimony, and which are, or may be, prejudicial to the accused.'" United States v. Young, 470 U.S. 1, 10, 105 S. Ct. 1038, 1043, 84 L. Ed. 2d 1 (1985) (citing Dunlop v. United States, 165 U.S. 486, 498, 17 S. Ct. 375, 379, 41 L. Ed. 799 (1897)).

Nevertheless, a criminal conviction is not to be lightly overturned on the basis of a prosecutor's comments standing alone, for the statements or conduct must be viewed in context; only by so doing can it be determined whether the prosecutor's conduct affected the fairness of the trial.

Young, 470 U.S. at 11, 105 S. Ct. at 1044. The Court of Appeals has recognized this reluctance to grant a new trial based on improper remarks in closing argument:

Despite the wide latitude afforded attorneys in closing arguments, there are limits in place to protect a defendant's right to a fair trial. Not every improper remark, however, necessarily mandates reversal, and "[w]hat exceeds the limits of permissible comment depends on the facts in each case." We have said that "reversal is only required where it appears that the remarks of the prosecutor actually misled the jury or were likely to have misled or influenced the jury to the prejudice of the accused." This determination of whether the prosecutor's comments were prejudicial or simply rhetorical flourish lies within the sound discretion of the trial court. On review, an appellate court should not reverse the trial court unless that court clearly abused the exercise of its discretion and prejudiced the accused.

Degren, 352 Md. at 430-431, 722 A.2d 887 (citations omitted). The prejudice to the defendant must be so great that the remarks "effectively undermined the bedrock principle that a defendant is presumed innocent." Degren, 352 Md. at 431, 722 A.2d 887.

In light of the foregoing, then, we must determine whether the State's remark in the instant case was so prejudicial as to require the first requested remedy, a mistrial. We note first that allegedly prejudicial remarks must be viewed by looking at them within the context of the facts of the particular case. Wilhelm, 272 Md. at 415, 326 A.2d 707; Young, 470 U.S. at 12, 105 S. Ct. at 1044. Wilhelm provides a guide as to the factors that should be considered when reviewing a trial court's denial of a motion for mistrial under these circumstances:

When in the first instance the remarks of the State's Attorney do appear to have been prejudicial, a significant factor in determining whether the jury were actually misled or were likely to have been misled or influenced to the prejudice of the accused is whether or not the trial court took any appropriate action, as the exigencies of the situation may have appeared to require, to overcome the likelihood of prejudice, such as informing the jury that the remark was improper, striking the remark and admonishing the jury to disregard it.

Wilhelm, 272 Md. at 423-424, 326 A.2d 707.

Other factors to consider are:

(1) whether the [improper remark] constituted a material factor in the conviction; (2) whether the comments related to a matter outside the record; (3) whether the evidence concerning appellant's guilt was not "close," but was rather "overwhelming," (4) whether there was evidence that the proceeding was "dominated by prejudice and passion," [although] ultimately, what exceeds the limits of permissible comment depends upon the facts in each case.

Walker v. State, 121 Md.App. 364, 376, 709 A.2d 177 (1998) (citation omitted). In addition, in some circumstances, a prosecutor's argument during rebuttal may be justified because it is in response to comments made by the defense during its closing. Degren, 352 Md. at 431, 722 A.2d 887. In Degren, the State remarked during rebuttal that "[t]he number one reason why you should not believe what [the defendant] says is nobody, nobody in this country has more reason to lie than a defendant in a criminal trial," and "this defendant has every reason to lie. She is a defendant." Degren, 352 Md. at 431, 722 A.2d 887. This comment was apparently made in response to defense counsel's assertions during closing that the State's witnesses had varying motives to lie.

In looking at the totality of the circumstances of the case sub judice, we do not believe that the trial court abused its discretion in refusing to grant a mistrial. The prejudice engendered by the State's improper remark was not so egregious as to have undermined the presumption that appellant was innocent. This is particularly true in light of the presumption of innocence instruction given by the trial court:

The defendant is presumed to be innocent of the charges, and this presumption remains with the defendant throughout every stage of the trial and is not overcome unless you are convinced beyond a reasonable doubt that the defendant is guilty.

As discussed, statistical evidence in and of itself does not undermine the presumption of innocence. In this case, we do not believe that the State's comment on statistical evidence adduced constituted a material factor in the conviction. There was much conflicting evidence on the various statistics put forward in this case. Three experts testified about statistics, and all three had different figures.^[5] The muddled statistical evidence may have provided assistance to the jury on the rare occurrences of SIDS death, particularly within the same family, but the statistics were not even necessary for the experts to testify to that scientifically accepted fact. Moreover, the use of the statistical testimony had been limited by the trial court. In addition, there was a great deal of other evidence, albeit circumstantial, that, if believed, could have lead to a finding of guilt.

In addition, the State's comment related to facts that were in evidence in this case, as the State explained in its brief:

Dr. Jones [appellant's expert] testified... that fewer than one percent of SIDS cases are associated with cerebral edema. Dr. Jones then said that the risk of a family having a second child die from SIDS was between 3.5 and 5.5 per 1,000....

Thus, the prosecutor took the lowest number Dr. Jones testified to, three (3), and used that in making his argument from the evidence that the testimony supported a determination that SIDS occurs in 3 out of 1,000 births. This figure is equivalent to 1 out of 333 1/3 births. Thus, applying the fact that 1 out of 100 SIDS deaths will exhibit cerebral edema, the prosecutor extrapolated from Dr. Kokes's analysis that 1 out of 33,333 SIDS deaths [probably, "live births"] will exhibit [a SIDS death] and this brain swelling. Applying the product rule ... it was appropriate to multiply these odds times the chance of a second SIDS death of 3 in 1,000, resulting in approximately 1 in 11,088,900, which the prosecutor evened out to 1 in 10 million. This was a reasonable inference to make from all the facts in the case....

The State's comment appears to be a response to defense counsel's attempts to bolster Dr. Jones's testimony during his closing argument. When viewed in this context, the remark appears to be more geared towards discrediting Dr. Jones's testimony.

Finally, this trial was not "dominated by prejudice and passion," Walker, 121 Md.App. at 376, 709 A.2d 177, and we note that the remarks complained of were the only remarks that appellant has complained about. In the context of the entire trial, which began on Monday, July 19, 1999, and ended on Thursday July 29, 1999, after two hours of deliberation, this isolated comment did not warrant a mistrial.

Our holding that a mistrial was not warranted in this case is reinforced by the fact that the trial court issued a curative instruction after the remark was made. In addition to this instruction, we note that the court had already cautioned the jury about the use of statistical evidence in the case and had instructed it to "[p]lease remember that the opening statements and the closing arguments of the attorneys are not evidence in this case. They are intended only to help you understand the evidence and then to apply the law."

We believe that the trial court's curative instruction in this case was sufficient to overcome any prejudice generated by the State's remark. We do not believe that the alternative remedy requested by appellant in this case, which appears to be a request to instruct the jury that it could not use the statistical evidence at all, was necessary. The jury is presumed to understand and follow instructions. Wilson, 261 Md. at 570, 276 A.2d 214; Brooks, 85 Md.App. at 360, 584 A.2d 82. The jury is therefore presumed to have understood the limited use to which the statistical evidence concerning the probabilities of different infants' deaths labeled as SIDS could be put, and that the State's Attorney's comment represented merely the prosecutor's assertion that, based on the testimony of appellant's own expert, it was improbable that the second death was a SIDS death and, in light of the totality of all the evidence presented, that appellant was not involved in Garrett Michael's death. There is little difference between that analysis of the prosecutor's comment and a prosecutor stating to jury members that, based on the evidence presented, they should find a defendant guilty. Degren, 352 Md. at 436, 722 A.2d 887.

We hold that the trial court's corrective action in response to the improper comment made by the State in its rebuttal closing argument was sufficient.

IV.

Appellant's fourth argument is that the trial judge erred in prohibiting a defense expert, Dr. Miles Jones, from "presenting a basis for his opinion." The opinion in question concerned the percentage of parents in the United States who purchase life insurance on the lives of their children.

Appellant proffered that his medical expert, Dr. Jones, could testify as to how many parents buy such insurance. The State objected, and the trial court ruled that the witness was not an expert on insurance and the information could only come in through other means. Appellant never presented other testimony regarding these percentages.

As previously noted, "the admissibility of expert testimony is a matter largely within the discretion of the trial court and its action will seldom constitute a ground for reversal." Celotex Corp., 88 Md.App. at 460, 594 A.2d 1248. "A trial judge's decision to admit or exclude expert testimony will be reversed only if it is founded on an error of law or some serious mistake, or if the judge has abused his discretion." Franch, 341 Md. at 364, 670 A.2d 951 (citation omitted). We discern no abuse of discretion in the trial court's ruling. As required by Md. Rule 5-702(1), there was no indication that the witness was qualified to testify as an expert concerning insurance policies held by parents in the United States. Dr. Jones's "knowledge, skill, experience, training, [and] education" is in the fields of medicine and pathology, not insurance.

We note that the trial court's ruling only prevented appellant from introducing the testimony through Dr. Jones. Appellant certainly could have sought to introduce the testimony through other means, but he did not do so.

V.

Appellant contends that the trial court erred by allowing the State to present evidence rebutting the findings of the two infants'"original" death certificates, i.e., those completed immediately after their deaths. "New" death certificates differing from the original death certificates were not entered into evidence by either party. The State made no attempt to introduce them and conceded at a bench conference that applications to amend the original death certificates were still in progress.

According to appellant, the Vital Statistics and Records Subtitle of the H.G. §§ 4-201 et seq., prohibits the introduction of evidence that casts doubt on the conclusions expressed in the original death certificates regarding the cause of death of the two infants. Appellant argues that it was impermissible for the trial court to admit evidence refuting any information contained on the original death certificates. We disagree.

In Benjamin, 268 Md. at 608, 303 A.2d 779, the Court of Appeals held that the trial court was correct in refusing to admit into evidence a death certificate, because the party seeking the certificate's admission sought to use the medical examiner's notation on the certificate that the decedent had committed suicide to bolster that party's contention that the decedent lacked testamentary capacity before his death. In Sippio, 350 Md. at 645-646, 714 A.2d 864, the Court of Appeals distinguished Benjamin from cases in which a contention is sought to be supported by the testimony of the actual medical examiner, rather than merely through a document completed by the medical examiner.

Health General § 4-201(n) provides that a death certificate is a "vital record." Section 4-212(b)(2)(ii) requires that it set forth the cause of death but not the manner of death.^[6] Under H.G. § 4-223(a), the "original or a certified copy of the certificate is prima facie evidence of the facts stated in it."

The trial court stated:

[The death certificate process] is a civil regulatory process, and I certainly do not think it bars the State's ability to present the actual experts who formulated the opinions which would underlie their ultimate conclusions in a death certificate which may not be admissible before a trier of fact anyway.

* * *

... I do not think the State is at all barred from bringing this prosecution even in advance of the amendment of the death certificate if they have the actual people who performed or who signed off [on] the autopsies....

Health General § 4-223(c) provides that a trial court has authority to determine the evidentiary value of an amended death certificate. Section 4-214(a) provides that a death certificate "may be amended only in accordance with this subtitle and any rules and regulations that the Secretary [of Health and Mental Hygiene] adopts to protect the integrity and accuracy of vital records." Section 4-214(b)(7) provides that "[a]ny amendments to death certificates requested beyond 3 years or more after the death shall require a court order."

Appellant, citing H.G. § 4-214(b)(7), argues that, without a court order, an amendment made to a death certificate more than three years after the certificate's creation is ineffectual. Appellant also contends that, construing § 4-214(b)(7) with § 4-223(a), "it is clear that only effective death certificates can have evidentiary value." Appellant argues:

The State did not ... comply with the legislatively mandated process for amending the original death certificates. Contrary to the view of the trial judge, it is not a mere "civil regulatory procedure" without evidentiary consequences. In this case the legally recognized manner of death remains natural causes. Until the State follows the prescribed procedure, it may not attempt to prove otherwise.

Appellant argues that because new certificates were not introduced into evidence, no evidence could be introduced at trial concerning causes of death other than those listed on the forms. This elevates form over substance; the original death certificates, which were introduced at trial, represented the opinions of the medical examiners involved in the autopsies at the time that the certificates were first made. Pursuant to the statute, a death certificate constitutes "prima facie evidence," i.e., "evidence that will establish a fact or sustain a judgment unless contradictory evidence is produced." Black's Law Dictionary 579 (7th ed.1999) (emphasis supplied).

At trial the court allowed those medical examiners to testify that their opinions had changed and to explain why. In addition, the court allowed other medical witnesses to testify about their opinions, which also differed from the original certificates.

As noted above, medical examiners in Maryland are required to investigate certain deaths, H.G. § 5-309(a)(1), and record the cause of those deaths on the certificates of death. H.G. § 4-212(b)(2). The medical examiners are required to maintain, for an indefinite period, records detailing, inter alia, the cause and manner of death of deceased persons. H.G. § 5-311(a)(2). The clearly worded intent of these provisions is that medical examiners accurately determine the cause and manner of death of deceased persons and that that determination be faithfully recorded.

We reject appellant's suggestion that evidence relating to the cause of death of deceased persons that is different from the original death certificate is inadmissible absent compliance with the provisions of the Health-General Article § 4-214(b)(7). Accordingly, we conclude that evidence rebutting the findings set forth on the original death certificates was properly admitted.

VI.

Appellant contends that the trial court erred by allowing the State to introduce evidence about Brandi's death.

Maryland Rule 5-404(b) provides:

(b) Other Crimes, Wrongs or Acts. Evidence of other crimes, wrongs, or acts is not admissible to prove the character of a person in order to show action in conformity therewith. It may, however, be admissible for other purposes, such as proof of motive, opportunity, intent, preparation, common scheme or plan, knowledge, identity, or absence of mistake or accident.

This exclusionary rule strives to prevent juries from using other crimes evidence to infer that, because a defendant appears to have a criminal disposition, he is more likely to have committed the crime for which he is on trial. State v. Taylor, 347 Md. 363, 368-369, 701 A.2d 389 (1997) (citing Straughn v. State, 297 Md. 329, 333, 465 A.2d 1166 (1983)). Evidence of other bad acts committed by a defendant is usually not admissible unless it has special relevance to a contested issue in the case and is not offered simply to prove criminal character. Harris v. State, 324 Md. 490, 500, 597 A.2d 956 (1991). Its probative force must also substantially outweigh its potential for unfairly prejudicing the jury against the defendant. Harris, 324 Md. at 500, 597 A.2d 956.

In Whittlesey v. State, 340 Md. 30, 59, 665 A.2d 223 (1995), the Court of Appeals stated:

First, the trial court must find that the evidence "is relevant to the offense charged on some basis other than mere propensity to commit crime." Second, the court must find by clear and convincing evidence that the defendant participated in the alleged acts. Third, the court must determine that the probative value of the evidence substantially outweighs its potential for unfair prejudice. (Citations omitted.)

In this case, when appellant objected to the admission of evidence concerning Brandi's death, the trial court analyzed the relevance of that information and made the following findings:

Defendant's participation established by clear and convincing evidence

The facts which the Defendant concedes and are therefore proved beyond the standard of clear and convincing evidence, establish the substantial relevancy of this evidence to these issues.[^[7]]

The Defendant was the victim's biological father. The victim [Brandi] was an infant less than six months old. After the infant was born, the Defendant purchased two insurance policies from separate companies on the infant's life. The policies were for $10,000 and $30,000. The Defendant was the primary beneficiary for each policy. The Defendant collected the proceeds of each policy after the death.

All of these facts, present in the circumstances surrounding the 1981 death of Brandi Wilson, are present in the circumstances surrounding the death of Garrett Michael Wilson in 1987.

Given the striking similarity between these events, the State has established the substantial relevancy of this evidence to its burden of proving the identity of the killer of Garrett Michael Wilson.

In addition, the evidence has substantial relevance to proving that the Defendant had a motive to commit the crime, demonstrating both intent and identity. The State can demonstrate that on a prior occasion the Defendant collected $40,000 following the death of an infant child. This clearly establishes a base of knowledge from which the State could argue a motive on the part of Defendant to murder an infant child for $150,000 in insurance proceeds.

Further, the evidence has substantial relevancy to the State's burden to prove beyond a reasonable doubt that Garrett Michael Wilson did not die from natural or accidental causes. The evidence aids the State in proving the actus reus of the crime itself.

The trial court acknowledged that the information was prejudicial to appellant, but found that its probative value outweighed any potential for unfair prejudice:

The Court has considered whether the probative value of the evidence is sufficient to outweigh its prejudicial affect [sic]. Clearly, admission of this evidence will carry a high level of prejudice for the Defendant. However, the State's need and the probative value of this evidence is also very high. In determining the level of the State's need, the Court looks not only to the State's burden to produce a prima facie case, but also to the ultimate burden of persuasion, that of proof beyond a reasonable doubt.... The enormity of the State's need for this evidence [of Brandi's death] is beyond debate when considered in light of the following:

1. The case is eleven years old.

2. The case involves the alleged smothering death of a five-month-old infant with little to no physical evidence.

3. There was no contemporaneous police investigation.

4. The death was originally ruled a SIDS death.

5. The revised opinions of the Medical Examiner are based upon a combined review of the facts of both cases.

The Medical Examiners originally assessed each death as caused by SIDS. Their opinions have now changed to death by asphyxia due to airway obstruction, probably by smothering. The reasons for their change of opinion gleaned from the State's exhibits are the investigation of the deaths of both babies and the re-evaluation of the physical evidence in the autopsy reports in light of both cases.

If the State is unable to place before the fact-finder the circumstances surrounding the death of Brandi Wilson in 1981, the experts will be prevented from testifying to their reasons for changing their opinions from death by natural causes to homicide.

* * *

The proposed other crimes, wrongs, or acts evidence has an extremely high level of probative value. The admitted facts establish the similarity and close connection between the two incidents.

A trial court's decision to admit evidence will not be disturbed on appeal unless it has abused its discretion. Merzbacher v. State, 346 Md. 391, 404-405, 697 A.2d 432 (1997); White v. State, 324 Md. 626, 637, 598 A.2d 187 (1991). The fundamental rule regarding the admission of evidence of other crimes is that "evidence of a defendant's prior criminal acts may not be introduced to prove that he is guilty of the offense for which he is on trial." Straughn v. State, 297 Md. at 333, 465 A.2d 1166 (citations omitted). "[T]here remains only one purpose for which other crimes evidence, in and of itself, may not be admitted, that is, to `prove guilt of the offense for which the defendant is on trial.' " Burral v. State, 118 Md.App. 288, 297, 702 A.2d 781 (1997), aff'd, 352 Md. 707, 724 A.2d 65, cert. denied, ___ U.S. ___, 120 S. Ct. 89, 145 L. Ed. 2d 75 (1999) (quoting Ayers v. State, 335 Md. 602, 630, 645 A.2d 22 (1994)).

When prior bad acts are highly probative to the crime alleged, their value to the fact-finder can outweigh the prejudice inherent in their introduction. Morgan v. Foretich, 846 F.2d 941, 945 (4th Cir.1988). This is particularly true when the crime for which the defendant faces trial is one which can be perpetrated with little evidence created.

In United States v. Woods, 484 F.2d 127 (4th Cir.1973), cited by the State, the U.S. Court of Appeals for the Fourth Circuit pointed out the unique difficulties inherent in infanticide cases, especially when suffocation was the cause of death, and the relevancy of prior behavior in such cases. The Court stated:

We think also that when the crime is one of infanticide or child abuse, evidence of repeated incidents is especially relevant because it may be the only evidence to prove the crime. A child of the age of Paul and of the others about whom evidence was received is a helpless, defenseless unit of human life. Such a child is too young, if he survives, to relate the facts concerning the attempt on his life, and too young, if he does not survive, to have exerted enough resistance that the marks of his cause of death will survive him. Absent the fortuitous presence of an eyewitness, infanticide or child abuse by suffocation would largely go unpunished.

* * *

Indeed, the evidence is so persuasive and so necessary in case of infanticide or other child abuse by suffocation if the wrongdoer is to be apprehended, that we think that its relevance clearly outweighs its prejudicial effect on the jury. We reject defendant's argument that the proof was not so clear and convincing that its admissibility should not be sustained. As we stated at the outset, if the evidence with regard to each child is considered separately, it is true that some of the incidents are less conclusive than others; but we think the incidents must be considered collectively, and when they are, an unmistakable pattern emerges.

Woods, 484 F.2d at 133, 135 (citation omitted).

In the instant case, the trial court found that there was enough commonality between Brandi's death and that of Garrett Michael "to disprove an innocent explanation for the 1987 death of Garrett Michael." The trial court then proceeded to the balancing test and made very specific findings as to why this evidence, which was clearly prejudicial to appellant, was admissible in the case.

We agree with the trial court that the evidence related to Brandi's death had special relevance in this case, and we hold that the trial court did not abuse its discretion in admitting it.

JUDGMENT AFFIRMED.

COSTS TO BE PAID BY APPELLANT.

NOTES

[1] In addition, H.G. § 5-309(c) provides:

(c) Immediately on notification that a medical examiner's case has occurred, the medical examiner or an investigator of the medical examiner shall go to and take charge of the body. The medical examiner or the investigator shall investigate fully the essential facts concerning the medical cause of death and, before leaving the premises, reduce these facts and the names and addresses of witnesses to writing, which shall be filed in the medical examiner's office.

[2] Brandi and Garrett Michael were Caucasians.

[3] We note that one of the errors complained of in this case, that Dr. Kokes found Ms. Anastasi's statements more credible than that of another witness in the case, Megan Churchill, occurred during cross-examination. Ms. Churchill lived next door to appellant and Ms. Anastasi at the time of Garrett Michael's death. She provided the police with a statement that either Ms. Anastasi or appellant, she could not remember who, told her that an hour had passed between the time appellant had fed Garrett Michael and Ms. Anastasi found him lifeless in the crib. The statement itself was not admitted into evidence at trial, and Ms. Churchill was never asked about the contents of the statement. It was, however, provided to the State's expert witnesses as part of the investigative reports, and some of the experts were asked about the differences between Ms. Churchill's statement and Ms. Anastasi's as to how much time had passed between the feeding and her finding the baby, apparently in an effort to have the experts testify that they credited one statement over the other. In any event, since the testimony from Dr. Kokes on this subject was elicited by appellant on cross-examination, it was invited error, and appellant cannot now complain about these statements. Klauenberg v. State, 355 Md. 528, 543, 735 A.2d 1061 (1999).

[4] We note that appellant complains that Dr. Smialek "claim[ed] to have a special ability to make [credibility] assessments" by reading the witnesses' statements. Again, appellant brought this information out on cross-examination, and any error was invited. He cannot now complain about it. Klauenberg, 355 Md. at 543, 735 A.2d 1061.

[5] Dr. Kokes testified that the statistical probability of more than one SIDS death occurring within the same family is 1 in 100,000,000 live births. Dr. Norton testified that the statistical probability is 1 in 4,000,000, and appellant's expert Dr. Jones testified that the statistical probability is between 3.5 and 5.5 per 1,000.

[6] H.G. § 4-212 provides, in pertinent part:

(b) ... (1) A certificate of death regardless of age of decedent shall be filled out and signed by:

(i) The medical examiner, if the medical examiner takes charge of the body; or

(ii) if the medical examiner does not take charge of the body, the physician who last attended the deceased.

(2) The medical examiner or physician shall fill in only the following information on the certificate of death:

(i) The name of the deceased.

(ii) The cause of death and medical certification.

(iii) The date and hour of death.

(iv) The place where death occurred.

(c) ... Each individual concerned with carrying out this subtitle promptly shall notify the medical examiner if:

(2) The cause of death is unknown; or

(3) The individual considers any of the following conditions to be the cause of death or to have contributed to the death:

(i) An accident, including a fall with a fracture or other injury.

(ii) Homicide.

(iii) Suicide.

(iv) Other external manner of death.

* * *

(d) ... (1) if, within 24 hours after taking charge of a body, the medical examiner has not determined the cause of death, the medical examiner shall enter "investigation pending" in the cause of death section of the death certificate.

(2) As soon as the medical examiner determines the cause of death, the medical examiner shall send to the Secretary [of Health and Mental Hygiene] a report of the cause of death, for entry on the certificate.

(Emphasis supplied.)

[7] The uncontested facts, as agreed to by appellant, are as follows:

In the fall of 1980, Debbie Oliver became pregnant by the Defendant. The two married and on February 25, 1981, Brandi Jean Wilson was born. Sometime during the night of April 30, 1981, the baby died.

Prior to the death of Brandi, the Defendant purchased two life insurance policies. One policy was for $10,000 and the other was for $30,000. The Defendant was the primary beneficiary and collected on each of the two policies.

The original cause of Brandi's death was determined to be Sudden Infant Death Syndrome (SIDS), and the manner of death was Natural.

The Defendant married Missy Anastasi in March of 1996. Missy became pregnant and on March 22, 1987 Garrett Michael Wilson was born. On Saturday, August 22, 1987 during the early morning hours, the baby died.

Following the birth of Garrett Michael Wilson, the Defendant purchased two life insurance policies on the infant. One policy was in the amount of $50,000 and the other policy was in the amount of $100,000. The Defendant was the primary beneficiary for each policy. The Defendant filed for and collected the proceeds from each of these policies.

The original cause of Garrett Michael's death was determined to be SIDS.