Article Text

other Versions

Cohort study
Link between length of hospital stay and mortality among hip fracture patients varies across healthcare systems
  1. Søren Paaske Johnsen1,
  2. Pia Kjær Kristensen1,2
  1. 1Department of Clinical Epidemiology, Aarhus University Hospital, Aarhus, Denmark
  2. 2Department of Orthopedic Surgery, Regional Hospital Horsens, Horsens, Denmark
  1. Correspondence to: Dr Søren Paaske Johnsen, Department of Clinical Epidemiology, Aarhus University Hospital, Olof Palmes Allé 43-45, Aarhus 8200, Denmark; spj{at}

Statistics from

Commentary on:

Implications for practice and research

  • The link between length of stay and mortality is complex and appears to differ across healthcare settings.

  • More insight into the potential causal mechanisms linking length of stay and mortality is needed.


Hip fracture is the most serious complication related to osteoporosis both from a patient and societal perspective due to the high case-fatality rate and healthcare costs.1 Since length of stay plays a pivotal role for the healthcare costs, reducing the length of stay has become an objective across healthcare systems. However, the aim of lowering the costs of hospital-based care must be balanced against the clinical needs of the patients. It is currently uncertain whether length of hospital stay plays an independent role for the subsequent mortality of patients with hip fracture, but a recent Swedish registry study reported length of stay below 10 days to be associated with a higher mortality.2


The aim of the study was to investigate whether length of hospital stay was associated with mortality within 30 days after discharge for hip fracture patients aged 50 years or older. The study used The New York Statewide Planning and Research Cooperative System (SPARCS) database to identify 188 938 hip fracture patients with and without surgical treatment between 2000 and 2011. SPARCS holds data on demographics, in-hospital diagnoses, procedures and charges. Length of stay was categorised as 1–5, 6–10, 11–14 and >14 days. The 30-day mortality was determined with a linkage to the New York State Department of Vital Statistics. The authors computed survival estimates using the Kaplan-Meier method and compared mortality risk using logistic regression analysis with adjustment for covariates. Furthermore, several sensitivity and subgroup analyses were performed to assess the robustness of the results of the primary analyses.


Length of stay of 11–14 days was associated with increased odds of 30-day mortality after discharge compared to length of stay of 1–5 days corresponding to an adjusted OR of 1.32 (95% CI 1.19 to 1.47). The association was even more pronounced for the patients with a length of stay above 14 days (adjusted OR=2.03 (95% CI 1.84 to 2.24). The positive association between length of stay and mortality remained even after excluding patients discharged to a hospice facility and those who left the hospital against medical advice.


The majority of the evidence published so far has indicated that shorter length of hospital stay is not associated with a higher risk for adverse clinical outcomes.3 An important exception is the Swedish study by Nordstrom et al,2 which reported shorter length of stay in hospital after hip fracture to be associated with increased risk of death after discharge.

The study by Nikkel et al had many features in common with the Swedish study, including the population-based design, the use of registries and the substantial size of the study population. Furthermore, the key elements of the analytical strategy also resembled the study by Nordstrom et al. It is therefore remarkable that the results of the US study are so different from those of the Swedish study, and the need for a better understanding of the reason for the seemingly contradictory findings is evident. The fundamental differences in the organisation of the healthcare systems in the USA and Sweden are immediately drawn to attention. In the USA, 90% of patients with hip fracture are discharged to rehabilitation facilities, which serve a transitional role. In contrast, a high proportion of patients in European countries are discharged home. In addition, notable differences exist in length of stay between the two studies, with financial incentives driving a shorter length of stay in the USA compared with Sweden. These structural differences could potentially explain the contradictory results. However, it is important to notice that the mechanisms underlying the association between length of stay and mortality still remains to be clarified and caution is needed before making firm conclusions about causality, that is, the existing data do not allow us to conclude that further reducing the length of stay in the USA or increasing the length of stay in Sweden would translate into a lower mortality.


View Abstract


  • Competing interests None declared.

  • Provenance and peer review Commissioned; internally peer reviewed.

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.