Background
Shu Feng Hsu was caring for her eleven-month-old niece, Chloe Chen, in the family’s Quincy home when the child became unresponsive. Evidence showed that Chloe had appeared healthy that morning. A home recording later captured a baby crying, eleven distinct thuds, and coughing or gagging sounds shortly before Hsu called 911. A responding paramedic found Chloe unresponsive, with her eyes rolled back and a visible bruise on her head. She died after suffering extensive intracranial injuries.
Prosecutors charged Hsu with first-degree murder. Medical witnesses for the Commonwealth attributed Chloe’s death to recent, repeated blunt-force head injuries. One expert concluded that the injuries were “consistent with” abusive head trauma—a clinical diagnosis covering brain or skull injuries inflicted through blunt impact, violent shaking, or both—and were inconsistent with an accidental short fall. Defense experts offered a competing explanation: chronic subdural bleeding that could have re-bled after a minor bump or fall. The jury rejected murder but convicted Hsu of involuntary manslaughter on theories of an unlawful battery endangering life and wanton or reckless conduct.
Before trial, Superior Court Judge Douglas H. Wilkins held a Daubert-Lanigan hearing, Massachusetts’s procedure for testing the reliability of scientific expert evidence. He found abusive head trauma generally accepted as a medical diagnosis, but imposed careful limits. Experts could say the injuries were “consistent with” the diagnosis, not that abusive head trauma caused Chloe’s death. They also could not claim that short falls can never produce such injuries. Hsu appealed both the sufficiency of the evidence and the admission of the expert testimony; the Supreme Judicial Court transferred the case on its own initiative.
The Court’s Holding
The SJC affirmed. On sufficiency, Justice Gaziano explained that a conviction may rest entirely on circumstantial evidence and that competing reasonable inferences belong to the jury. Chloe was healthy before being left as Hsu’s sole charge; the recording captured crying and repeated thuds; her condition deteriorated immediately afterward; and the medical evidence supported recent, nonaccidental blunt-force trauma. Although experts disagreed about whether an older injury could have re-bled, the jury was entitled to credit the Commonwealth’s witnesses. That proof supported findings both that Hsu committed a battery she knew or should have known endangered life and that her conduct created a high likelihood of substantial harm.
The court also upheld the expert ruling. Under Daubert-Lanigan, general acceptance within the relevant scientific community can establish reliability even without unanimity. The trial judge had relied on statements from numerous medical organizations and peer-reviewed literature recognizing abusive head trauma as a valid diagnosis reached through clinical history, examination, laboratory findings, and imaging. Criticism focused largely on whether shaking alone can cause the associated injuries, while the broader diagnosis encompasses multiple mechanisms, including blunt impact. The existence of scientific disagreement therefore did not make the diagnosis inadmissible.
Nor did the word “abusive” improperly decide guilt for the jury. The pediatric expert addressed injury mechanism and said the findings were consistent with abusive head trauma; she did not identify Hsu as the perpetrator or opine on guilt. The phrase carried prejudice risk because jurors might treat it as a proxy for culpability, but the trial judge reasonably managed that risk by limiting the testimony, allowing cross-examination and competing experts, and instructing jurors to reach their own conclusions. Finally, testimony that Chloe’s particular injuries were inconsistent with a short fall did not violate the order barring the broader assertion that short falls can never cause abusive head trauma.
Key Takeaways
- Massachusetts courts may treat abusive head trauma as a generally accepted medical diagnosis, notwithstanding an active minority scientific critique and contrary authority elsewhere.
- Experts may say injuries are “consistent with” abusive head trauma when the opinion explains medical mechanism without assigning responsibility or declaring the defendant guilty.
- A judge can reduce the prejudice associated with the emotionally charged word “abusive” through phrasing limits, adversarial testing, competing expert evidence, and a clear jury instruction.
- A case-specific opinion that particular injuries do not fit a short fall is different from the categorical and scientifically disputed assertion that short falls can never cause such injuries.
Why It Matters
Hsu gives Massachusetts trial judges and criminal practitioners a detailed framework for litigating abusive-head-trauma evidence. The SJC did not endorse an automatic pathway to admission. It approved a record-based reliability determination coupled with restrictions that preserved the jury’s role. Prosecutors still must establish the medical foundation and connect expert opinions to the individual injuries; defense counsel remains free to expose methodological disputes, present competing explanations, and argue that the diagnosis does not establish who caused an injury.
The decision is also important beyond child-abuse prosecutions. It illustrates how Massachusetts courts handle contested science: general acceptance does not require consensus, and disagreement generally affects weight rather than admissibility when the trial judge has a sufficient reliability record. At the same time, terminology that carries an implicit moral judgment demands attention under the rule excluding evidence whose unfair prejudice substantially outweighs its value. Here, carefully drawn limits—“consistent with,” not “caused by”—allowed the jury to hear useful medical evidence without handing an expert the ultimate decision.