Background Acute illness may catalyse changes in care preferences, but the influences on preferences are incompletely understood in frail older people following acute illness.

Aim To explore influences on the care preferences of frail older people following acute illness.

Methods Prospective cohort study utilising a concurrent dominant status (quan + QUAL) mixed-methods design with six month follow up. Inclusion criteria were: age ≥65; Rockwood clinical frailty score ≥5; recent acute illness requiring secondary care assessment or admission. Of a study sample of 90 participants, 18 purposively sampled participants contributed both survey responses about their preferred care outcomes and serial in-depth qualitative interviews with a topic guide exploring influences on preferences. Patterns of preference stability identified in the survey responses of these 18 participants were explored during thematic analysis of interviews.

Results Median patient age 84 (inter-quartile range (IQR) 81 – 87), 53% female. Median frailty score 6 (IQR 5 – 7). Evolving awareness of deteriorating or uncertain health trajectories tended to influence preferences away from ‘extending life’, towards ‘staying out of hospital’, ‘being comfortable’, or ‘improving quality of life’. Care experiences influenced preferences more prominently than time during the study, particularly affecting place-of-care preferences. Health status changes did influence preference stability, but those with close support from family reported that family considerations were more important than health status changes as an influence on preferences, as family gave them ‘something to live for’. The presence of prominent longstanding personal values, often relating to independence, supported stability of preferences.

Conclusions The stability of care preferences is influenced by experience, evolving awareness and family considerations, but is less influenced by time or health changes in this population. By exploring health awareness and care experiences with patients, clinicians may be better able to elicit preferences and deliver care responsive to preferences following acute illness.