Elsevier

Clinical Nutrition

Volume 35, Issue 4, August 2016, Pages 900-906
Clinical Nutrition

The impact of weight loss and low BMI on mortality of nursing home residents – Results from the nutritionDay in nursing homes

https://doi.org/10.1016/j.clnu.2015.06.003Get rights and content

Summary

Background

Weight loss and low body-mass-index (BMI) are well-known risk factors for mortality among older persons. Both items represent a reduced nutritional state but their nature is different. It is, however, unclear which of these factors is more important for prognosis and if there is interference or interaction between them.

Objective

To measure the prevalence of low BMI and weight loss in nursing home residents and to analyze their impact on mortality.

Methods

The nutritionDay in nursing homes is an international annual one-day cross sectional survey evaluating malnutrition and 6-months mortality. The data collected from 2007 to 2012 were analyzed. The mortality risk due to a BMI <20 kg/m2 and weight loss >5 kg was calculated by logistic regression analysis with adjustment for confounding factors.

Results

In total, 10,298 residents with a mean age of 85 years (78% female) from 191 nursing homes in 13 countries were included in the analysis. Eighteen percent (17.7%) had a low BMI < 20 kg/m2, and 11.3% lost > 5 kg of body weight in the previous year. Low BMI <20 kg/m2 and weight loss >5 kg were both independent and significant risk factors for 6-months mortality (OR 1.7 and 1.5; p < 0.001 in the adjusted model). In comparison to the effect size of low BMI and weight loss alone, the interaction of both factors was substantial (OR = 1.4; p = 0.056). The average 6-months mortality was 13.8%. Subjects with BMI ≥20 kg/m2 without weight loss >5 kg presented the lowest mortality of 11.2%. In subjects with one of the two features, weight loss >5 kg or low BMI <20 kg/m2, mortality was significantly increased to 17.4% and 19.8%, respectively. However, if both features were present simultaneously the mortality increased disproportionally to 35.7% (OR 3.5; p < 0.001). In addition, this study presents an overlap and a strong dose-effect relationship of both items with regard to the 6-month mortality.

Conclusion

A BMI <20 kg/m2 and weight loss >5 kg in one year are both independent and equally relevant risk factors for the 6-months mortality of nursing home residents aged 65 years or older. For this reason weight loss should be avoided and residents with a low BMI and weight loss should receive particular attention and nutritional care.

Introduction

Weight loss and low body-mass-index (BMI) are well-known risk factors for poor outcomes and increased mortality of older subjects [1], [2], [3], [4], [5], [6], [7], [8], [9], [10], [11], [12], [13], [14], [15], [16], [17], [18]. Both phenomena represent a reduced nutritional state, but their nature is substantially different. Weight loss in older persons is predominantly associated with disease and frailty. A low BMI however, might be a permanent characteristic of a person or a consequence of previous weight loss. To date, it is unclear which of these two factors, weight loss or low BMI, is more important for the prognosis of older subjects and if there is an interference or interaction between them. The fact that subjects with a low BMI are more likely to have lost weight in the past than subjects with a high BMI raises the question of interference between these two factors. In other words, it is possible that weight loss is much more important than a low BMI, and that a low BMI is an overestimated statistical risk factor because numerous subjects with a low BMI may have lost weight before. In fact, there are several studies in older persons indicating that weight loss might be a much more important risk factor than a low BMI [10], [14], [15], [16], [18].

To our knowledge there are few studies that compare the relevance of weight loss and a low BMI in older persons within a single data set [1], [3], [9], [11], [13]. None of these studies analyzed the possible interference and interaction of these two risk factors and some have conflicting results with regard to the prognostic impact of both factors. The nutritionDay in nursing homes (NH) study includes information about weight loss, BMI and mortality. Thus, it is suitable to answer these questions for the population of older nursing home residents.

Section snippets

Methods

The nutritionDay project evaluates disease-related malnutrition in the form of an international, voluntary, annual one-day cross-sectional survey. The survey was first performed in hospitals [4], [19]. Since 2007, the project has been extended to nursing homes [20]. The following analysis is based on all data collected from 2007 to 2012 for the nutritionDay in nursing homes project. With the exception of Japan, only Western developed countries took part. The data from Japan were not used for

Results

Between 2007 and 2012, 369 nursing homes with 26,651 nursing home residents from 17 countries took part in the survey. After the exclusion of participants with incomplete data, especially missing outcomes, and those with age below 65 or repeated participation, 10,298 data sets from 191 nursing homes in 13 countries could be analyzed. Please see Fig. 1 for details about the number and reasons for exclusion of participants and Table 1 for the number of included residents in different years and

Discussion

In our study, both weight loss >5 kg and low BMI <20 kg/m2 were independently associated with increased mortality and showed a risk of comparable magnitude. Even in the adjusted analysis this association with 6-month mortality was robust. However, presenting both features simultaneously increased the mortality risk disproportionally (OR 3.5), due to an interaction between BMI and weight loss. With this finding, our data extend the current knowledge about nutritional risk factors and their

Conclusion

A BMI <20 kg/m2 and weight loss >5 kg in one year are both independent and equally relevant risk factors for the 6-months mortality of nursing home residents aged 65 years or older. If both features are simultaneously present, the mortality risk is disproportionally increased due to an interaction between both features. A low BMI, as well as weight loss, show a strong dose-effect relationship with regard to the 6-month mortality of nursing home residents. On the basis of previous interventional

Financial support

This study was supported by an ESPEN research grant.

Statement of authorship

RW, CS and DV designed the study. RW, CK, CS, UT and DV carried out the analysis. DV, MS, CCS and MH managed the annual survey in nursing homes and the data base. RW and DV wrote the manuscript. All authors read and reviewed the manuscript and approved the final version.

Conflict of interest

None.

Acknowledgments

We would like to thank all participating nursing homes for supporting the nutritionDay.

References (26)

  • S. Kimyagarov et al.

    Body mass index (BMI), body composition and mortality of nursing home elderly residents

    Arch Gerontol Geriatr

    (2010)
  • A.L. Cawood et al.

    Systematic review and meta-analysis of the effects of high protein oral nutritional supplements

    Ageing Res Rev

    (2012)
  • J.L. Locher et al.

    Body mass index, weight loss, and mortality in community-dwelling older adults

    J Gerontol A Biol Sci Med Sci

    (2007)
  • Cited by (54)

    • Sarcopenic Dysphagia Is Associated With Mortality in Institutionalized Older Adults

      2022, Journal of the American Medical Directors Association
    • Hungry for more: Low resident social engagement is indirectly associated with poor energy intake and mealtime experience in long-term care homes

      2021, Appetite
      Citation Excerpt :

      Despite the recognized importance of meals to nutritional intake in LTC settings, the risk of malnutrition also continues to be a common issue, with prevalence estimates ranging from 40% to 80% of residents at risk or malnourished (Donini et al., 2013; Keller et al., 2017a; Muurinen et al., 2015). Malnutrition and poor food intake have been associated with higher levels of cognitive impairment, functional impairment, lower body mass index (BMI), dysphagia, eating challenges, and higher rates of mortality (Keller et al., 2017a; Wirth et al., 2016). Malnutrition risk is also attributed to multi-level interactions between residents and the larger psychosocial (Keller et al., 2017a; Watkins et al., 2017) and socio-political environments (Lowndes et al., 2015) within LTC homes.

    View all citing articles on Scopus
    View full text