How best to fix a broken hip

Objectives: To develop evidence-based guidelines for the treatment of proxi mal femoral fractures to optimise functional outcome while minimising lengt h of stay in hospital. Data sources: Systematic literature search of MEDLINE and CINAHL computer d atabases, bibliographies, and current contents of key journals for 1966-199 5. Study selection: English-language randomised controlled trials of all aspec ts of acute-care hospital treatment of proximal femoral fracture among subj ects aged 50 years and over with proximal femoral fractures not due to meta static disease. Data extraction: Two independent reviewers, blinded to authors, institution and study results, followed a standard Cochrane Collaboration protocol and assessed study quality and treatment conclusions. When necessary, a third review was performed to reach consensus. Results: Of the 120 articles published between 1966 and December 1995, 97 m et the inclusion criteria. Fifteen clinical interventions were reviewed. Fi ve were supported by National Health and Medical Research Council (NHMRC) l evel I evidence (prophylactic anticoagulants, prophylactic antibiotics, reg ional anaesthesia, pressure-relieving mattresses, and internal surgical fix ation), two had no supporting randomised controlled trial evidence (time to surgery, time to mobilisation after surgery) and the remainder were classi fied as having Level II evidence. A review of current practice (1993-94) id entified wide variability in these interventions across five acute-care hos pitals in the Northern Sydney Area Health Service. Conclusions: Randomised controlled trial evidence (NHMRC Levels I and II) e xists for many, but not all, aspects of hip fracture treatment. There is a need for changes to be made to some aspects of practice in accordance with evidence-based guidelines.