Elsevier

Clinical Nutrition

Volume 32, Issue 5, October 2013, Pages 737-745
Clinical Nutrition

Original article
Malnutrition and poor food intake are associated with prolonged hospital stay, frequent readmissions, and greater in-hospital mortality: Results from the Nutrition Care Day Survey 2010

https://doi.org/10.1016/j.clnu.2012.11.021Get rights and content

Summary

Background & aims

The Australasian Nutrition Care Day Survey (ANCDS) ascertained if malnutrition and poor food intake are independent risk factors for health-related outcomes in Australian and New Zealand hospital patients.

Methods

Phase 1 recorded nutritional status (Subjective Global Assessment) and 24-h food intake (0, 25, 50, 75, 100% intake). Outcomes data (Phase 2) were collected 90-days post-Phase 1 and included length of hospital stay (LOS), readmissions and in-hospital mortality.

Results

Of 3122 participants (47% females, 65 ± 18 years) from 56 hospitals, 32% were malnourished and 23% consumed ≤ 25% of the offered food. Malnourished patients had greater median LOS (15 days vs. 10 days, p < 0.0001) and readmissions rates (36% vs. 30%, p = 0.001). Median LOS for patients consuming ≤ 25% of the food was higher than those consuming ≤ 50% (13 vs. 11 days, p < 0.0001). The odds of 90-day in-hospital mortality were twice greater for malnourished patients (CI: 1.09–3.34, p = 0.023) and those consuming ≤ 25% of the offered food (CI: 1.13–3.51, p = 0.017), respectively.

Conclusion

The ANCDS establishes that malnutrition and poor food intake are independently associated with in-hospital mortality in the Australian and New Zealand acute care setting.

Introduction

The Australasian Nutrition Care Day Survey (ANCDS) is the largest multicentre study in the Australasian region, reporting the prevalence of malnutrition and poor food intake in 3122 patients across 56 Australian and New Zealand hospitals.1 With one-in-three patients malnourished; and two-in-three patients not consuming all of the offered hospital food, it was evident that malnutrition and poor food intake are a common occurrence in Australian and New Zealand hospitals.1

Numerous studies have suggested that in comparison to well-nourished patients, malnourished patients experience worse outcomes such as prolonged length of stay (LOS) in hospital, increased readmissions, and mortality.2, 3, 4, 5, 6 There is documented evidence to suggest that malnourished patients incur greater hospitalisation costs,7 related to longer LOS, readmissions, and greater utilisation of hospital resources.2, 5

The ANCDS found that one-in-three malnourished patients (n = 305, 30%), and one-in-five well-nourished patients (n = 371, 18%) consumed nothing or up to 25% of the food offered during the 24-h data collection period.1 Since continued sub-optimal food intake can eventually lead to deterioration of nutritional status, it is important to evaluate the effect of poor food intake on health-related outcomes. Two studies have reported the link between poor food intake during hospitalisation and mortality,6, 8 however, there is no published evidence regarding the association between poor food intake and readmissions and/or LOS.

Although previous studies have investigated associations between malnutrition and patient outcomes, issues such as heterogeneity in patient populations; study design; methods of evaluating nutritional status, food intake and/or outcomes; prevent the results from these studies being generalised throughout the acute care population. Factors such as type and severity of disease are major causes of malnutrition,9 poor food intake,10 and patient outcomes, and yet they have rarely been controlled for. Without accounting for the confounding effect of disease type and severity most studies fail to distinguish the association between the effect of disease, nutritional issues, and other factors (such as age, gender), and patient outcomes. Therefore, there is a risk of underestimating the independent effects of disease, and overestimating the independent effects of nutritional issues. The aim of this study was to take into account disease type and severity and explore associations between: (1) nutritional status; (2) food intake; and health-related outcomes (LOS, mortality, and readmissions) in participants from the ANCDS.

Section snippets

Methods

The ANCDS was conducted in two phases. Participants were recruited in Phase 1 of the study and the episode of admission was referred to as “index hospitalisation”.

In Phase 1 data were collected by dietitians from participating hospitals.1 Data included demographic, nutritional status, and 24-h food intake information for each participant.1 Participants' Body Mass Index (BMI) were calculated based on their recorded weight and height.1 To evaluate nutritional status, each participant was screened

Results

Outcomes data were available for 3017 of the total 3122 participants (97%). After data cleaning (as previously outlined), data analyses for LOS and mortality included 2982 participants (95%), and readmissions data were analysed for 2942 participants (94%).

Table 1 depicts admission-related characteristics of the participants. Malnutrition was significantly associated with age ≥ 65 years, emergency admissions, admissions other than surgical or medical, certain MDCs, severe/catastrophic PCCL

Discussion

The ANCDS is the first multicentre study in acute care hospitals across Australia and New Zealand to report the association between patients' nutritional status, food intake and health-related outcomes. The study found that patients who were malnourished or consumed ≤ 25% of the hospital offered food had significantly longer LOS and higher in-hospital mortality rates. Malnourished patients also had significantly higher readmissions rates than well-nourished patients.

Considering there are

Authors' contributions to manuscript

EA designed and coordinated the study; acquired, analysed and interpreted the data; and wrote the manuscript. MF, MB and EI provided significant advice on the study design. MBatterham provided statistical advice. All authors participated in editing and final revisions of the manuscript. All authors have read and approved the final manuscript.

Conflict of Interest

EA, MBatterham, JB, and SC have no conflict of interest to declare. MF, MB and EI are employed by Queensland Health, Australia.

Acknowledgements

The authors would like to thank (1) Participating sites for their time and effort in collecting the data for this study; (2) AuSPEN for its support in organising the webinars for training dietitians involved with data collection; and the small research grant awarded to EA in 2010; (3) Members of the AuSPEN Steering Committee for their valuable feedback on the project plan in the initial stages of the project; (4) Queensland Health for funding Queensland hospitals to recruit additional

References (31)

  • J. Kandiah et al.

    Visual plate waste in hospitalized patients: length of stay and diet order

    J Am Diet Assoc

    (2006)
  • Y. Dupertuis et al.

    Food intake in 1707 hospitalised patients: a prospective comprehensive hospital survey

    Clin Nutr

    (2003)
  • A. Perkins et al.

    Common comorbidity scales were similar in their ability to predict healthcare costs and mortality

    J Clin Epidemiol

    (2004)
  • M. Correia et al.

    The impact of malnutrition on morbidity, mortality, length of hospital stay and costs evaluated through multivariate model analysis

    Clin Nutr

    (2003)
  • D. Sullivan et al.

    Protein-energy undernutrition among elderly hospitalized patients: a prospective study

    J Am Med Assoc

    (1999)
  • Cited by (347)

    View all citing articles on Scopus

    Conference Presentation: This paper was presented at the 34th ESPEN Congress on Clinical Nutrition and Metabolism in Barcelona, Spain (8–11 September 2012).

    View full text