Outcome following fractured neck of femur--variation in acute hospital care or case mix?

BACKGROUND

This study examined the quality of care given to patients admitted to hospital with a fractured neck of femur by assessing the link between outcome, case severity and resource use. Fractured neck of femur was chosen for this study as it is a common condition amongst elderly people which causes considerable morbidity and mortality, uses a high proportion of acute hospital resources and is a condition where virtually all new cases will come under the care of the hospital service.

METHODS

Three hospitals which had different case fatality rates and costs were included in the study. These were an inner-city teaching hospital (Hospital 1), an inner-city associated teaching hospital (Hospital 2) and an associated teaching hospital in an urban location (Hospital 3). Patients were recruited for this study over a 12-month period. Details on case severity and basic demographic date were collected on admission, and information on the process of care was collected during the hospital stay. Four outcome measures were addressed: activities of daily living (ADL) before discharge and at three months post-fracture; mortality up to 12 months post-fracture; complications occurring after admission to hospital; and destination on discharge.

RESULTS

A total of 492 patients were recruited into the study, with a male to female ratio of 1:4 and an age range of 60-101 years. Patients admitted to the three hospitals showed no difference with respect to the presence of co-morbidities, medication, pre-fracture ADL, mental state, age and sex. There were some differences observed in pre-fractured place of residence. Hospital 1 had the highest proportion of patients admitted from sheltered housing and other hospital. Hospital 2 the highest proportion from residential homes, and Hospital 3 the highest proportion admitted from their own homes. Hospital 3 discharged patients at an earlier stage of recovery in that a higher proportion were discharged with a poor ADL index. This hospital also had more orthopaedic complications but fewer medical complications; however, the outcome in terms of ADL at three months post-fracture and mortality at 12 months was similar in all three hospitals. The severity variables which predicted poor outcome were co-morbidities, impaired mental state, impaired ADL pre-fracture, increasing age and an extracapsular fracture. After controlling for severity variables, the resource variables had not further impact on mortality, either in hospital or within one year. An epidural anaesthetic was related to a poor ADL at three months and more orthopaedic complications but fewer medical complications. There was also a hospital effect in that Hospital 3, which performed the most epidurals, had the highest proportion of orthopaedic complications but the lowest proportion of medical complications. When the operating surgeon was a consultant, there were more orthopaedic complications, but this was not related to these patients having a worse case severity on admission. However, among the cases operated on by consultants, there were no hospital deaths. No other resource variables were related to ADL at three months, or orthopaedic or medical complications.

CONCLUSIONS

The results show that a poor outcome following a fractured neck of femur was related to increased case severity at the time of fracture. The resource variables had very little impact on the outcome.