Nutritional status and associations with falls, balance, mobility and functionality during hospital admission

Abstract

Objectives

To explore associations between nutritional status, falls and selected falls risk factors amongst older hospitalized people. Lengths of stay amongst fallers and the malnourished were assessed.

Design

An observational longitudinal cohort study.

Setting

Geriatric Assessment and Rehabilitation Unit (GARU) of a tertiary teaching hospital.

Participants

Admissions to the GARU during a six-month period were included.

Measurements

Associations between nutritional status and falls during hospitalization, reported preadmission history of falls, functional status, balance and mobility during GARU admission were analysed. Associations between nutritional status or experiencing a hospital fall and length of stay were also examined.

Results

Malnutrition prevalence was 39% (75/194, 95% CI 32–46%) with odds of falling during admission being 1.49 (95%CI: 0.81, 2.75), p< 0.20). Patients assessed as malnourished were older (p<0.001) and more likely to have a poorer score on both the admission (p<0.05) and discharge (p<0.009) timed “Up and Go” test. Malnutrition was associated with reduced mobility (p<0.05). Those who fell during admission had statistically greater lengths of stay compared with non-fallers [median (range): 57.0 (7–127) vs 35.0 (5–227) days; p<0.002].

Conclusion

Evidence of reduced mobility was evident during GARU admission amongst older people assessed as malnourished. Considering the results, a larger study concerning nutritional status, functionality and falls in the hospitalized population is warranted The influence of nutritional status upon a person's physical functioning should be considered more broadly in falls research.

Introduction

Poor nutritional status has been associated with those who fall. Compared to non-fallers, fallers admitted to a geriatrics unit showed greater nutritional risk with less muscle and fat stores (1). Even if healthy, muscle mass declines with age, and muscle mass and function is also negatively influenced by malnutrition (2). Interventions likely to be beneficial for falls prevention include those that improve muscle strength and balance retention (3).

It is acknowledged that contributors to falls risk in hospital may be different from those identified in the community (4). Associations between nutritional status and falls has been documented in the hospital setting (5). Although explored in the community, ambulatory and long term care settings (6, 7, 8, 9, 10), less is known about the role of nutritional status on falls, balance, declining mobility and functional dependency within the hospital setting. It is important to identify how the presence of malnutrition on patient admission affects falls risk and patient recovery as this may impact upon patient safety both during and following hospital discharge.

Establishing if associations exist between nutritional status, mobility, function and balance will provide evidence to assist with the development of both prevention and treatment strategies. Worldwide, those aged 60 years or more are the fastest growing population segment with those over 80 years the fastest growing group (11). service provision models that identify and treat modifiable conditions that reduce falls risk are an important priority for health systems to address in order to maintain the health and independence of an increasingly ageing population. This study aims is to directly explore the associations between malnutrition on the physical function and balance of older people during their hospital admission, as this has not previously been reported in the literature.

Subjects and Method

Study design

An observational longitudinal cohort study examined the association between nutritional status, mobility, functional and balance status, self-reported history of falls during the past six months, and falls during inpatient sub-acute care on a Geriatric Assessment and Rehabilitation Unit (GARU). Associations between length of stay, nutritional status and experiencing a hospital fall were examined.

Participants

Admissions to the GARU of a tertiary teaching hospital during a six-month period were recruited. Data was collected for all adult patient admissions with the exception of individuals with complete paraplegia, tetraplegia or lower limb amputation without prosthesis as this study required administration of standing balance tests.

Measures

In-hospital falls were recorded via a computerised patient incident reporting systems and review of patient medical records. A fall was defined as “Any event where a patient unexpectedly comes to rest on the ground, the floor, or another lower level” (12).

A history of falls prior to hospital admission was collected subjectively via patient recall however if a patient proxy (spouse, career, family member) was available, they were used to verify the patient recall of falls history.

Malnutrition was assessed using the subjective Global Assessment (sGA) completed in accordance with the protocol described by Detsky et al (13). Gender and age at last birthday were recorded.

Functional outcome measures were assessed and recorded on admission as well as during routine discharge assessment. Physical function outcome measures collected were the Functional Independence Measure (FIM) motor subsection (14), the step test (number of steps able to be completed onto a 7.5 cm step in 15 seconds, average of two trials, one with the left foot, one with the right) (15), the timed “Up and Go” test (the time to walk 3 metres, turn around and walk 3 metres back, starting and finishing in a seated position) (16), the Functional Reach Test (distance able to lean forward when standing before overbalancing) (17), and the timed static standing test with eyes closed (sum of three repetitions of up to 30 seconds) (18, 19).

Procedure

Malnutrition assessment was conducted by hospital dietitians within 72 hours of patient admission to the GARU ward. Physical function was assessed by hospital occupational therapists (motor FIM) and physiotherapists (all other measures). Falls history prior to admission was collected by hospital physiotherapists while in-hospital falls were recorded by all hospital staff. De-identified data was collated from patient medical histories by hospital physiotherapists who forwarded de-identified data to project investigators for data entry and analysis. Ethical approval for this study was granted by both the hospital and university ethics committee.

Data analysis

Due to the low number of patients assessed as SGA C, this category (severely malnourished) was combined with SGA B (moderately malnourished) for these analyses. Logistic regression analysis was employed to examine whether malnutrition was associated with a history of falls over the previous six months and also whether a patient became a faller or not during their inpatient stay. Negative binomial regression was used to examine whether malnutrition was significantly associated with experiencing a higher rate of falls during the inpatient stay.

Associations between nutritional status and physiological outcomes were assessed using independent t-tests for normally distributed data (motor FIM) and Mann-Whitney U tests for non-normally distributed data (all other outcomes). statistical analysis was performed using the statistical Package for the Social Sciences (Release 11, SPSS Inc, Chicago, II, 2003). A statistical significance level of p<0.05 was used.

Results

Data from 194 patients were collected during the study with 39% (75/194, 95% CI 32-46%) assessed as malnourished. Table 1 summarises descriptive characteristics. Those assessed as moderately or severely malnourished were older (p<0.001) and more likely to have a poorer score on both the admission (p<0.05) and discharge (p<0.009) timed “Up and Go” test. There was not a statistically significant association between the odds of the self-reported history of falls upon admission and being assessed as moderately or severely malnourished during the hospital admission [Odds Ratio (95%CI) = 1.57 (0.86, 2.86), p <0.15).

Table 1.

Admission demographics and physical function outcome measures of study patients assessed to be malnourished and well-nourished (n=194)

	n	Well-nourished	n	Moderately (n=68) or severely malnourished (n=7)	p value
Age (years)	119	71.5 (14.3)	75	80.9 (9.5)	P<0.001
Gender Male % (n)	119	36.1% (43/119)	75	42.7% (32/75)	P<0.37
Motor FIM	118	56.9 (16.5)	71	54.2 (15.1)	P <0.25
Step test average∗	119	0 (0, 4)	75	0 (0, 3)	P <0.25
Functional reach∗	118	6.5 (0, 18.25)	74	1.1 (0, 16.25)	P <0.27
Static stance∗	118	3.4 (0, 90)	75	0 (0, 90)	P <0.23
TUG∗	62	25.7 (17.3, 40.5)	46	32.9 (23.8, 51.6)	P<0.05

Data presented are median (IQR) and comparison by Mann-Whitney U test if indicated by

^∗otherwise they are mean (sd) and compared by unpaired t-test.

Those who fell during admission had statistically greater lengths of stay compared with non-fallers [median (range): 57.0 (5-227) days; p<0.002]. There was no significant difference in length of stay whether self-reporting a fall or not during the six months prior to admission [40.5 (5-227) vs 36 (5-128) days; p<0.49] or being assessed as well-nourished or malnourished [45.0 (8-128) vs 41.5 (5-104) days; p<0.64).

Of the patients assessed as well-nourished (SGA A), 29% (32/110) became fallers during their inpatient admission compared to 33% (23/68) of those assessed as moderately malnourished and 43% (3/7) of those assessed as severely malnourished. However, the association between malnutrition and whether a patient became a faller during their inpatient stay was not statistically significant at the given sample size [Odds Ratio (95%CI) = 1.49 (0.81, 2.75), p< 0.20).

Discussion

For the first time, associations were identified between nutritional status upon hospital admission and the functional outcome measure of the timed “Up and Go”. Although some differences in other functional outcome measures were not significant due to lack of study power, there was a tendency for better functionality scores amongst those assessed as well-nourished.

A more complete picture for nutrition and intermediate falls risk factors such as physical functioning is starting to emerge in the hospital environment. Malnutrition exacerbates muscle loss (2) and many nutrients that influence muscle function are lower in older people admitted to hospital after a fall (1, 10). Malnutrition has been associated with poorer physical function (20) (current study) and hospital falls (5). Amongst malnourished older hospitalized people, nutritional supplementation increases protein synthesis (21) and when combined with resistance training, best maintained strength in malnourished older people hospitalized after a fall (22). However resistance training, in the absence of concurrent nutritional support, exacerbated nutritional risk through weight loss (22). Consequently, the need to ensure optimized nutritional intakes when physical therapy programs are undertaken is emphasized.

The impact of optimised nutrition care upon the Cochrane recommendation (3) of “muscle strengthening and balance retraining” (an intermediate outcome) to reduce falls (a clinical outcome) (23) is yet to be documented in the older hospitalized population. In community settings, nutritional supplementation has improved energy, nutrient and fluid intake and reduced falls (24, 25). Further adequately powered studies are required to confirm our initial hospital findings and to further elucidate the impact of optimal nutrition status and support on both functionality and falls risk in the hospital setting.

Greater lengths of stay were also evident amongst those who fell in hospital during this study. Active nutritional support has been shown to positively impact on hospital length of stay (24 vs 40 days), and on deaths and complications (44% vs 87%) when admitted after a fall with fractured neck of femur (26). However, the impact of optimal nutritional support upon the length of stay of those who fall in hospital has not yet been ascertained.

Malnutrition is under-recognised and under-diagnosed in the rehabilitative care setting (27, 28). The prevalence of those at nutritional risk in this study were similar to figures from other hospitalised populations (29, 30, 31, 32). Malnutrition prevalence varies according to the criteria used, but is estimated to be around 10-60% of people in hospital and longer term care settings such as nursing homes (28, 32, 33). Additionally, it has also been shown that nutritional status can deteriorate during a hospital admission (34, 35) and that active nutritional support can positively influence such trends (27).

Nutritional status needs to be more widely incorporated into multifactorial explorations of falls risk factors amongst older people. For over a decade, poorer anthropometry has been associated with fallers in both the community (9, 24) and hospital setting (1, 10) However, malnutrition has not been included in multifactorial explorations of falls risk parameters within either community (36) or hospital settings (37, 38, 39). The current study findings that malnutrition was associated with poorer functionality as assessed through the timed “Up and Go” indicate that optimising nutritional status may be an important consideration as a component of comprehensive geriatric rehabilitation programs.

The main limitation is that the present study sample size was moderate, indicating that we may have committed a type II statistical error (erroneously accepting no difference between the groups) when considering the relationship between malnutrition and the occurrence of falls in hospital. Although this association was found to not be statistically significant in the present study, the odds ratio point estimate (1.49) would arguably be of clinical importance if the association was found to be significant indicating that more research in this area utilizing larger sample sizes should be undertaken. Missing data resulted from performing the study within the constraints of actual clinical practice. Results are potentially conservative for those assessed as moderately malnourished and severely malnourished were combined. A larger study enabling the severely malnourished to be independently assessed may provide further evidence of associations between nutritional status, physical function and balance.

In conclusion, some evidence of reduced mobility during admission was evident amongst older hospitalised people with poorer nutritional status. Including nutritional status when exploring falls risk parameters and optimising nutritional status during care should be further explored as essential components of the multifactorial approaches to falls prevention strategies in hospital.

Acknowledgements: The support and the co-operation of all ward staff in the completion of this study is acknowledged and appreciated. We additionally thank Steven McPhail for his assistance during the completion of this study.