Commentary

Bernard is first discharged after improvement, although he has not fully recovered. He can walk with help and participate in brief conversation, but needs prompting for meals, dressing and medicines. The handover makes those needs explicit. Rosa's agreement to help is accompanied by organised support and a route for reassessment; it is not treated as consent to provide unlimited care.

A later increase in drowsiness is compared with his recent post-discharge state as well as his earlier baseline. The clinician finds new fever and hypoxia and arranges hospital assessment. This is a new deterioration within a longer recovery, rather than inevitable fluctuation that should simply be tolerated [G01].

Cole and colleagues found that worse recovery status after discharge was associated with more subsequent adverse events [P42]. That study does not prove that every partially recovered patient will be readmitted, or that all such admissions are preventable. It supports taking persistent symptoms and care needs seriously. The family narrative in S40 describes the practical demands of sustained supervision, without establishing the cause of that published patient's illness.

Bernard's second recovery is slower. The final page shows a review of support based on actual night-time needs, mobility and cognition. Moving to a care home is a considered response to present needs, not proof that cognitive improvement has ended. At four months he remains substantially impaired. Both further assessment and opportunities for ordinary enjoyment continue, without a promised return to his previous independence.

An anonymised composite case that does not describe one identifiable person and is not research evidence.