STATE of Oregon, Plaintiff-Respondent,
v.
Terry Dean DAVIS, Defendant-Appellant.
02C45953; A134216.
Court of Appeals of Oregon.
Argued and Submitted July 22, 2009. Decided May 12, 2010.George W. Kelly, Eugene, argued the cause and filed the briefs for appellant.
Rolf C. Moan, Acting Solicitor General, argued the cause for respondent. With him on the brief was John R. Kroger, Attorney General.
Before SERCOMBE, Presiding Judge, and BREWER, Chief Judge, and ARMSTRONG, Judge.[*]
*988 SERCOMBE, P.J.
Defendant appeals from a judgment of conviction after a jury found him guilty of one count of murder, ORS 163.115, and one count of manslaughter, ORS 163.118, for the death of his young daughter. The victim died of internal injuries to her brain and abdomen caused by physical abuse; the jury necessarily concluded that the physical abuse occurred in the hours immediately before the victim's death while she was in defendant's care. Defendant contends that the trial court erred in excluding three types of evidence that he claims are relevant to establish that the fatal internal injuries to the victim occurred earlier while in another person's care: evidence of prior injuries to the victim; a lay witness opinion that the victim had symptoms of a brain injury earlier; and testimony that the victim's mother was informed that the victim might have earlier suffered a brain injury where mother did not report that potential condition to physicians examining the victim in the days leading up to her death.
A verdict against a criminal defendant will be affirmed, notwithstanding evidentiary error, if the error did not affect a "substantial right" of the defendant. OEC 103(1) ("Evidentiary error is not presumed to be prejudicial. Error may not be predicated upon a ruling which admits or excludes evidence unless a substantial right of the party is affected, and [the objection is preserved]."). A defendant's substantial rights are not affected by evidentiary error if there is "little likelihood that the error affected the verdict." State v. Hansen, 304 Or. 169, 180-81, 743 P.2d 157 (1987). The determination whether the improperly excluded evidence could have affected a verdict requires an assessment of the "relative strength of the parties' evidence" and, given the totality of the evidence, "how significant was the erroneously [excluded] evidence?" State v. Lytsell, 187 Or.App. 169, 181, 67 P.3d 955 (2003) (internal quotation marks omitted). We conclude that it was error to exclude evidence that the victim's mother was told of a potential brain injury a few days before the victim's death and that, given the closeness of the case, the exclusion of that evidence could have affected the verdict. We therefore reverse and remand for a new trial.
Given the importance of the totality of the evidence, we set out the facts with some particularity. Defendant's daughter was 15 months old when she died on June 30, 2002. The victim lived with her mother, Ecklund. Ecklund, the victim, and Ecklund's other children vacationed in Mexico and California in the middle of June 2002. During the trip, the victim fell off a motel bed. Ecklund told police later that she "couldn't get there fast enough" and that the victim hit the "top of her head" in the fall. Ecklund testified, however, that the victim exhibited no signs of distress after the fall.
Shortly after the family's return to Oregon, on June 24, the victim had an overnight visit with defendant. She returned to Ecklund's care the following day. On June 25, Ecklund noticed that the victim was not feeling well; she was vomiting and her eyes were "glassy." Ecklund asked her friend Payne, a registered nurse, to stop by and examine the victim. Payne did so and noticed that the victim was unresponsive. The trial court excluded as hearsay Payne's testimony that she had told Ecklund that the victim's condition looked just like that of Payne's daughter when she had suffered a traumatic brain injury years before. Payne tested the victim for dehydration by pinching the victim's skin and noticed that the victim's eyes were glassy and somewhat "rolled up" in her head; Payne recommended that Ecklund have the victim examined by a doctor. During her testimony at trial, Ecklund did not recall Payne's visit or recommendation.
Ecklund took the victim to Santiam Hospital a few hours after Payne's examination. There, the victim was evaluated, treated for dehydration and a viral infection, and given intravenous fluids. Her symptoms resolved, and the victim returned home the next morning.
Defendant testified that, on June 29, around 7:00 p.m., he picked up the victim for an overnight visit. The victim was fussing and crying when defendant arrived, but showed no other signs of illness or injury. Once they arrived at defendant's home, defendant gave the victim a bath and changed *989 her diaper. He then began doing some household chores while the victim, another child of defendant, and defendant's small dog played together in the garage. When defendant returned to the garage, he found the victim lying on her side on the floor. Defendant testified that the other child told him that the small dog had knocked over the victim. Defendant picked up the victim, but decided that she seemed unhurt.
Later that evening, defendant ordered a pizza, which was delivered after 10:00 p.m. The victim was uninterested in the pizza, but ate a few pieces of pineapple. The victim spat up or vomited, and defendant testified that he noticed that the victim had a blank affect and had become limp. He called Ecklund at 10:58 p.m. to tell her that something was wrong with the victim and asked Ecklund to come to his house to help. A few minutes later, when Ecklund arrived, she found the victim limp and unresponsive and told defendant to call 9-1-1. Defendant made the call and told the dispatcher that the victim had recently eaten pineapple and spat up and that he was concerned about a possible allergic reaction.
Paramedics arrived at defendant's home around 11:00 p.m. They found the victim lying on the floor with defendant, Ecklund, and defendant's other child nearby. Defendant told the paramedics that the victim had not suffered any trauma, but had been sitting on the couch and went suddenly limp. The paramedics found the victim unresponsive but breathing, with a body temperature of 92.7 degrees. The paramedics also noticed a small bruise on her forehead and some bruising on her abdomen. The paramedics testified that Ecklund did not tell them about the victim's visit to Santiam Hospital several days previously. The paramedics decided to transport the victim to Salem Hospital after determining that she rated extremely low on a scale measuring brain functioning.[1]
When the victim arrived at the hospital, she was assessed by Kelly, an emergency room physician. Defendant told Kelly that the victim had not had any recent episodes of vomiting or illness. Ecklund did not tell Kelly about the victim's recent hospital visit or treatment for dehydration. Kelly found the victim to be completely unresponsive. A computerized axial tomography (CAT) scan revealed that the victim suffered from a subdural hematoma, a bleeding under the dura, a membrane that encases the brain. The hematoma caused significant pressure within the victim's head. Kelly determined that she needed to have the pressure relieved as soon as possible or she would die. On hearing that diagnosis, Ecklund became very upset but defendant, by contrast, maintained a "flat affect," according to Kelly's testimony.
Because Salem Hospital has no pediatric neurosurgeons on staff, the victim was transported to Oregon Health & Science University Hospital (OHSU) in Portland for treatment. After she arrived at OHSU, doctors performed surgery to reduce the pressure in the victim's brain. Subsequent testing showed that the victim was brain dead. She was taken off life support on June 30.
The following day, Nelson, a deputy state medical examiner, performed an autopsy on the victim. Nelson found that, in addition to the subdural hematoma, the victim had suffered a torn mesentery, an internal abdominal injury, that had caused substantial internal bleeding. Nelson testified that both injuries were acute in nature; they had occurred within 24-36 hours before the victim's death. Nelson concluded that the brain injury was a "fatal injury" but also agreed that the abdominal injury could have been fatal in and of itself. He further concluded that neither injury was consistent with a fall or similar accident and that the cause of death was homicide.
At trial, the state presented evidence that the victim had received the injuries that killed her on June 29, 2002, while in defendant's care, and that those injuries were intentionally inflicted. Specifically, the state's witnesses testified that the victim's *990 injuries would have produced immediate and obvious symptoms that would have caused the victim to be unable to eat or have bowel movements. Nelson testified that the victim had been subjected to violent shaking or slamming, which had injured her brain without breaking her skull or causing external signs of trauma. Nelson concluded that the head injuries must have occurred after the victim ate pineapple in the late evening at defendant's home and could not have occurred earlier. Similarly, Nelson stated that the abdominal injury showed no evidence of healing, and that the microscopic evidence was not consistent with an older injury that became exacerbated. According to Nelson, the abdominal injury also occurred within a day or so of the victim's death.
Tegtmeyer, a doctor at OHSU, testified that the victim's head injury would have required a large amount of force and agreed that shaking or slamming a child "makes the brain literally bounce back and forth in the skull." He testified that a child with that type of brain injury would not be able to eat and would have no coordination. Other witnesses for the state were less definitive about the timing of the brain injury. Smolin, a radiologist from Salem hospital, testified that the victim's CAT scan indicated an "acute" or "very recent" blow to the head or "acceleration/deceleration injury" but that he was concerned there might be an additional "older, more chronic subdural" that could be "older than three weeks."
Defendant's expert witnesses agreed that that victim died from intentionally inflicted injuries to her head and abdomen, but did not agree with the prosecution about the time when those injuries had been inflicted. Ophoven, a pediatric pathologist, testified that both of the victim's fatal injuries occurred before June 29, 2002. Ophoven said that children experiencing an abdominal injury can follow a gradual sequence of blood loss, tissue degeneration, and shock. According to Ophoven, children so injured can be awake and ambulatory, and experience intense thirst; it is consistent with this type of injury for a child to drink fluids or eat fruit before the injury finally deteriorates. Ophoven also testified that early symptoms of such an injury are the same as symptoms of gastroenteritis, the diagnosis of the victim's condition that was made at Santiam Hospital a few days before her death.
Ophoven opined that the autopsy photos and tissue samples showed that the victim's tissue had begun to heal from the injuries to her abdomen. Specifically, Ophoven testified that the victim's abdominal injury was more than three or four days old when the victim died. Ophoven identified granulation tissue on slides taken from the victim's abdominal injury and testified that granulation tissue forms many days after an injury occurs. Ophoven concluded that it was impossible that the victim's abdominal injury had been inflicted the day she died. Ophoven had a similar opinion about the victim's head injury. She testified that there was evidence of older clotting and trauma to the head.
Wittwer, an emergency room physician, also testified for the defense. Wittwer testified that a child with a subdural hematoma can go up to two weeks without noticeable effects and that only a CAT scan would reveal the brain injury at that early stage. Wittwer also testified that a child with a severe abdominal injury might revive and appear normal after receiving intravenous fluids, as the victim did a few days before she died. According to Wittwer, a child with a mesentery injury might not be sensitive in the front of the abdomen. Additionally, Leestma, a specialist in neuropathology, also found granulation tissue, indicating days of healing, on the victim's slides. Leestma testified that the victim's brain injury was at least five days old.
At trial, defendant sought to introduce evidence of (1) a prior leg fracture and rib injuries the victim suffered that may have been caused by child abuse; (2) testimony by Payne that Payne believed that the victim appeared to be suffering from a brain injury several days before her death; and (3) Payne's testimony that she told Ecklund on June 25 that the victim's symptoms resembled those of Payne's daughter when she suffered a traumatic brain injury. The trial court excluded that evidence, and defendant assigns error to each of those rulings. We reject without further discussion defendant's *991 challenges to the trial court rulings on admission of evidence of the victim's prior unrelated injuries and Payne's lay opinion regarding a possible brain injury.
Defendant's third assignment of error that the court erred in excluding evidence of what Payne told Ecklund about the victim's condition on June 25requires more particular analysis. In defendant's offer of proof, Payne testified that she told Ecklund, four days before the victim died, that the victim's condition could be a brain injury. Specifically, Payne testified, "I do believe I told [Ecklund] that [the victim's condition] reminded me of [my daughter], but mymy thing is that I feel I told her [the victim] needed to be checked out." Payne confirmed that she had "shared with [Ecklund] over time that [Payne's daughter] had had a motor vehicle accident, she had been injured brainwise, she's had these multiple hospitalizations," and that Ecklund would have known what Payne meant when she said that the victim reminded her of her daughter.
The trial court excluded that testimony as hearsay. Defendant argues, the state concedes, and we agree, that the testimony was not hearsay. OEC 801(3) defines "hearsay" as "a statement, other than one made by the declarant while testifying at the trial or hearing, offered in evidence to prove the truth of the matter asserted." The statements were not offered to show the truth of the matter asserted, that the victim had similar symptoms to Payne's daughter. Rather, defendant sought to admit the evidence to show that Ecklund was told that victim's condition might be of the same nature as those of a brain-damaged child who had needed extensive treatment and hospitalization and that Ecklund had not disclosed that information to the paramedics or to any of the victim's treatment providers at Santiam Hospital, Salem Hospital, or OHSU in the four days between Payne's visit and the victim's death. Defendant argues that his defense was based on the theory that the victim's injuries had occurred earlier, when the victim was out of defendant's care, and that the jury could infer from Payne's excluded testimony that Ecklund knew something about when and how the injuries occurred and had hidden that information. Defendant asserts that the evidence is admissible and its exclusion was prejudicial.
Although the statements are not hearsay, the state argues that they were nonetheless inadmissible because they were irrelevant and that, therefore, their exclusion was not reversible error. There is no ground for reversal of a judgment where a trial court makes a correct ruling admitting or excluding evidence, but articulates an erroneous reason for the ruling. State v. Nielsen, 316 Or. 611, 629 n. 13, 853 P.2d 256 (1993). OEC 401 defines "relevant evidence" as "evidence having any tendency to make the existence of any fact that is of consequence to the determination of the action more probable or less probable than it would be without the evidence." Under OEC 402, relevant evidence is generally admissible and irrelevant evidence is not.
Payne's testimony about Ecklund's knowledge of the victim's potential brain injury was relevant under OEC 401. That fact had a tendency to make more probable the existence of deliberate nondisclosures by Ecklund of that condition to treatment providers. Those deliberate nondisclosures, in turn, were of consequence to the defense theory that the victim's injuries had been previously incurred. Thus, the proffered evidence assists in establishing the relevance of other evidence, namely, the testimony of the paramedics and emergency room doctors who said that Ecklund never told them about the possibility of an earlier brain injury. That evidence supports the inference that Ecklund was hiding her knowledge about the victim's prior injuries. The evidence is relevant, and the court erred in ruling that it should be excluded.
As noted earlier, the exclusion of relevant evidence will not require reversal if there is little likelihood that the error affected the verdict. That determination requires an assessment of the strength of the parties' evidence and the significance of the excluded evidence given the totality of the record. Thus, where there is convincing evidence of the defendant's guilt in the record as a whole, and little, if any, likelihood that the error affected the jury's verdict, then the *992 error is harmless. State v. Walton, 311 Or. 223, 231, 809 P.2d 81 (1991) (inadmissible evidence that the defendant denied knowing another person involved in the crime and that the defendant gave a false name was harmless because of the convincing evidence in the record as whole); Hansen, 304 Or. at 180, 743 P.2d 157 (a substantial right of a criminal defendant is not affected if there is substantial and convincing evidence of guilt and little likelihood that the error affected the verdict); State v. Miller, 300 Or. 203, 220-21, 709 P.2d 225 (1985), cert. den., 475 U.S. 1141, 106 S.Ct. 1793, 90 L.Ed.2d 339 (1986) ("[A]ppellate courts are required to affirm the trial court, notwithstanding evidential error, whenever there is * * * substantial and convincing evidence of guilt in a criminal case.").
The state argues that the evidence establishing an inference of Ecklund's deliberate nondisclosure of the victim's condition is not significant because Ecklund did, in fact, take the victim to physicians for diagnosis and treatment. Ecklund took the victim to the hospital for medical assistance shortly after speaking with Payne on June 25 and immediately directed defendant to seek medical care for the victim on June 29. Defendant, on the other hand, asserts that the case against him rested entirely on circumstantial evidence that the victim's injuries occurred on June 29 while she was under his care. The expert witnesses were divided on the timing of the victim's brain and abdominal injuries. Thus, defendant argues that any evidence supporting the existence of prior injuries to the victim's brain was crucial to his defense.
We agree with defendant. As described above, the state and the defense presented substantial expert testimony about the causation and age of the victim's injuries. The experts disagreed completely about when the injuries likely occurred. There was expert testimony that the injuries that killed the victim could not have been inflicted before the victim ate pineapple on the night she died. There was also expert testimony that the abdominal injuries must have been more than a few days old to show the healing structures that were visible at autopsy and that evidence of clotting showed a predicate, earlier trauma to the victim's head. The record, in short, did not contain evidence of guilt that was so substantial and convincing that evidentiary error could be presumed to be harmless. Instead, the jury was required to discern whether the state's evidence concerning the timing of the victim's injuries constituted proof beyond a reasonable doubt of defendant's guilt, in light of the contradictory scientific explanation offered by defendant's experts. Any evidence of prior injury to the victim's brain might have tipped the balance in defendant's favor. The evidence that Ecklund hid information suggesting an earlier injury was significant.
Moreover, defendant's proposed inferences from Ecklund's conduct concerning the victim's medical treatment are not dissimilar from the conclusions that the state argued the jury should draw from defendant's conduct concerning that treatment. The state emphasized at trial that a parent's response to a child's injury and the parent's conduct in seeking or delaying medical treatment can be a significant indicator for child abuse and encouraged the jury make an inference from defendant's conduct that he was the victim's abuser. The state argued that defendant responded inappropriately to the victim's symptoms and delayed treatment by calling Ecklund rather than seeking emergency aid. The prosecutor argued to the jury in closing argument that
"people that abuse kids don't want to get caught, and the first step in them getting caught is when they go and get medical care. They don't want to face what they have done and they don't want to get caught.
"* * * * *
"* * * [T]his is * * * a cover-up kind of response, which is he knows he did it, he know what he did was wrong, and now he's trying to act like it's not as big of a deal as what it really is."
The excluded evidence at issue in this case may have led the jury to infer that, although Ecklund sought medical treatment for the victim, she did not give doctors all the information they might need. Although Payne's excluded testimony is not as weighty as the evidence the state discussed at closing argument, *993 it is nonetheless true that parental response to information about a child's condition can give rise to inferences about that parent's involvement in causing the condition. We therefore conclude that the trial court's error was not harmless.
Reversed and remanded.
NOTES
[*] Brewer, C.J., vice Edmonds, P.J.
[1] The paramedics and an emergency room doctor testified that the victim rated a 4 on the Glasgow Coma Scale, a measure that evaluates brain function. The scale has three categories of symptoms, and a person receives a score from 1 to 5 in each category. A person with normal functioning would rate a 15 score; a dead person would score a 3.