当症状不足时：大语言模型精神科筛查中的证据加权模式
When Symptoms Are Not Enough: Evidence-Weighting Patterns in Large Language Model Psychiatric Screening
AI总结：
本研究引入SCID锚定基准，评估五个大语言模型在精神科筛查中的表现，发现模型在焦虑症、抑郁症和创伤后应激障碍分类中，当存在功能保留或保护性背景时倾向于低估症状证据，导致假阴性错误。
摘要：
随着心理健康护理需求超过临床医生提供的评估，对可扩展筛查工具的需求日益增加。大语言模型（LLMs）可能从患者叙述中识别精神科风险，但其在不同诊断、人口统计亚组和证据使用模式中的可靠性仍不确定。我们引入了一个基于SCID的基准，包含555个半结构化体验访谈，并配有焦虑症、重度抑郁症、创伤后应激障碍和任何当前心理健康障碍的诊断参考标签。使用零样本任务特定提示，我们评估了五个最先进的LLM，并检查假阴性错误是否反映了遗漏的精神科证据或对症状、功能损害和保护性背景线索的差异化加权。不同任务和模型的表现各异，准确率从0.49到0.86，马修斯相关系数从0.16到0.38。GPT-4.1 Mini和GPT-5 Mini显示出最一致的疾病特异性准确率。亚组分析发现，男性参与者的抑郁症分类准确率高于女性，没有一致的年龄相关模式，种族阶层间存在适度的非均匀变异。证据整合分析显示，假阴性的焦虑症和PTSD分类通常包含明确的症状证据，但伴有功能保留、应对能力或社会支持。功能损害证据使模型输出偏向阳性分类，而保护性背景证据则使输出偏离。这些发现表明，LLMs可能支持可扩展的精神科筛查，但它们在功能保留或保护性背景下低估症状证据的倾向需要在临床部署前进行仔细验证。
英文摘要：
As demand for mental health care outpaces clinician-delivered assessment, scalable screening tools are increasingly needed. Large language models (LLMs) may identify psychiatric risk from patient narratives, but their reliability across diagnoses, demographic subgroups, and evidence-use patterns remains uncertain. We introduce a SCID-anchored benchmark of 555 semi-structured experiential interviews paired with diagnostic reference labels for anxiety disorder, major depressive disorder, post-traumatic stress disorder, and any current mental health disorder. Using zero-shot task-specific prompting, we evaluated five state-of-the-art LLMs and examined whether false-negative errors reflected missed psychiatric evidence or differential weighting of symptom, functional-impairment, and protective-context cues. Performance varied across tasks and models, with accuracy ranging from 0.49 to 0.86 and Matthews correlation coefficients from 0.16 to 0.38. GPT-4.1 Mini and GPT-5 Mini showed the most consistent disorder-specific accuracy. Subgroup analyses found higher depression-classification accuracy among male than female participants, no consistent age-related pattern, and modest non-uniform variation across race strata. Evidence-integration analyses showed that false-negative anxiety and PTSD classifications often contained explicit symptom evidence but were accompanied by preserved functioning, coping ability, or social support. Functional-impairment evidence shifted model outputs toward positive classifications, whereas protective-context evidence shifted outputs away. These findings suggest that LLMs may support scalable psychiatric screening, but their tendency to discount symptom evidence in the presence of preserved functioning or protective context requires careful validation before clinical deployment.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
