Older adults increasingly use health wearables, yet often cannot inspect the properties that matter for reliance. Through 31 semi-structured interviews in China, we examined how participants judged whether wearable outputs were reliable enough for everyday use. Participants relied on brand and price, visible interface activity, lived interaction experience, and comparison with bodily sensation. These cues supported conditional trust, but did not reveal sensor validity, data continuity, or failure conditions. We describe this mismatch as an observability gap and outline design directions for showing signal quality, reliability by context, human-system fit, and alert provenance.

翻译：暂无翻译