Elsevier

Gait & Posture

Volume 31, Issue 3, March 2010, Pages 317-321
Gait & Posture

Falls prediction in elderly people: A 1-year prospective study

https://doi.org/10.1016/j.gaitpost.2009.11.013Get rights and content

Abstract

The aim of the present study was to determine whether force plate variables in single- and dual-task situations are able to predict the risk of multiple falls in a community-dwelling elderly population. Two hundred and seventy elderly persons (225 females, 45 males; age, 73 ± 7 years) performed balance assessment with and without vision. Seven force plate variables were assessed to predict the risk of multiple falls; maximum displacement in the anteroposterior and medial–lateral directions (Max-AP, Max-ML), mean displacement in the medial–lateral direction (MML), the root mean square amplitude in anteroposterior and medial–lateral directions (RMS-AP, RMS-ML), the average speed of displacement (V), and the area of the 95th percentile ellipse (AoE). Falls were prospectively recorded during the following year. A total of 437 registered falls occurred during monitoring period. The force plate variable RMS-ML in the single-task condition (odds ratio, 21.8) predicted multiple falls together with the following covariables: history of multiple falls (odds ratio, 5.6), use of medications (fall-risk medications or multiple medicine use; odds ratio, 2.3), and gender (odds ratio, 0.34). Multiple fallers had a narrower stance width than non-fallers.

Introduction

More than one-third of adults aged 65 years or older sustain at least one fall each year [1], [2]. Injuries resulting from falls are one of the leading causes of death among the elderly population [3]. Many risk factors for falling in both community-dwelling and institutionalised elderly populations have been reported [4], [5]. The ability to maintain postural stability is one of the most relevant factors for fall prediction [6]. One approach for assessing postural stability is to use quantitative posturography, in which the position of the centre of mass and its displacement within the base of support (BOS) are quantified. This displacement is expressed as a centre of pressure and is defined as the point of application of the ground reaction forces under the feet [7].

The ability to discriminate between elderly fallers and non-fallers with the help of force plate-derived variables is inconclusive [8]. One possible reason for these inconclusive results concerns the testing situations used, which generally do not mimic real-life circumstances that cause falls. In many test situations, the participants have ample opportunity to compensate for their deficits by shifting toward other control strategies, e.g., single-task versus dual-task testing. Therefore, measurement protocols should challenge the participant in more complex conditions to avoid the use of these compensatory strategies [9]. In their review, Zijlstra et al. present two studies providing some evidence that measurements with a dual-task protocol could add value to the prediction of falls [10].

More studies are needed with a prospective design that record falls as the primary outcome. Additionally, specific subgroups should be investigated, e.g., elderly (75 years of age or older). All force plate variables with a capacity to predict the risk of multiple falls should be examined and they should be done so in realistic, dual-task situations [11].

The aim of the present study was to determine whether force plate variables are able to predict the risk of multiple falls under different testing situations in a community-dwelling elderly population.

Section snippets

Participants

This study was conducted at the University Hospital Zurich, Switzerland. Participants were recruited from the greater Zurich area. A local patient organisation distributed 2000 envelopes containing an information letter concerning the study and an application form for participation in the study. These envelopes were made available to their members. An additional 500 envelopes were also displayed in the outpatient department of the hospital. A total of 315 elderly mailed in an application form,

Participants

The baseline measurements were completed for 303 participants. A total of 270 elderly subjects finished the study by sending in all 12 monthly calendar pages. Baseline population characteristics are summarised in Table 2. The dropout rate was 11%; 25 participants did not even begin the calendar, 6 were not willing to continue participating after sending in only a few monthly calendar pages, and 2 participants died during follow-up. The mean age was 73.7 ± 7 years. In the 60–74 years subgroup (n = 

Discussion

This prospective study showed that force plate-derived variables are of relevance in predicting multiple falls in community-dwelling elderly individuals, confirming previous findings [11]. The amplitude of medial–lateral movements in single-task conditions was a significant independent predictor of multiple falls, along with the covariates female gender, a history of multiple falls and the use of fall-risk medications or multiple medications.

Two previous studies reported that dual-task, rather

Conclusion

The amplitude of medial–lateral movements as measured with a force plate during single-task balance tests is an independent predictor of multiple falls in community-dwelling individuals. During testing, multiple fallers of the subgroup aged 60–75 years positioned themselves with a narrower stance width than non-fallers.

Acknowledgements

The authors wish to thank Kathrin Favero and André Schmidt for performing the measurements and Nicole Thomson of the Swiss League against Rheumatism (Rheumaliga Schweiz), René Willi of the Zurich League against Rheumatism (Rheumaliga Zurich), Physio Suisse and Marlies Dorfschmid for their assistance in recruiting. We thank Leanne Pobjoy for her help in editing the manuscript, and Dr. Anne F. Mannion who made a critical revision of the whole manuscript.
Conflict of interest statement

There is no

References (22)

  • A. Zijlstra et al.

    Do dual tasks have an added value over single tasks for balance assessment in fall prevention programs? A mini-review

    Gerontology

    (2008)
  • Cited by (116)

    • Assessing Free-Living Postural Sway in Persons with Multiple Sclerosis

      2024, IEEE Transactions on Neural Systems and Rehabilitation Engineering
    View all citing articles on Scopus
    View full text