Examining trajectories of hospital readmission in older adults hospitalised with hip fracture from residential aged care and the community.

Abstract

To examine hospital service use trajectories of older adults who were hospitalised for hip fracture and living in a residential aged care facility (RACF) or the community, and to identify factors predictive of trajectory group membership. These findings may inform future programmes aimed at reducing unexpected hospitalisations and subsequently reduce health care costs. A group-based trajectory analysis of hospitalisations was conducted for adults aged ≥ 65years hospitalised for hip fracture during 2008-2009 in New South Wales, Australia. Linked hospitalisation and RACF data were examined for a 5-year period. Group-based trajectory models were derived for RACF and community-dwelling older adults based on the number of subsequent hospital admissions following the index hip fracture. Multinomial logistic regression examined predictors of trajectory group membership for subsequent hospital admissions. There were 5752 hip fracture hospitalisations, with two-thirds of hip fractures occurring in community-dwellers. Key predictors of trajectory group membership for both RACF residents and community-dwellers were age group, sex, hospital length of stay and cognitive impairment. Assistance with activities of daily living and complex health care needs were also predictive of group membership in RACF residents. Location of residence and time to move to a RACF were additional predictors of group membership for community-dwellers. Health service use trajectories differed for RACF residents and community-dwellers; however, there were similar patient characteristics that defined trajectory group membership. Low users of hospital services living in RACFs or the community included older adults with generally unfavourable health conditions, potentially indicating that palliative care or advanced care directives and community-care initiatives, respectively, have played a part in the lowered frequency of rehospitalisation.