Simplicity over Complexity: Limited Added Value of Diffusion Analysis for Smartphone-Based Fall-Risk Classification in Older Adults

Background/Objectives: Smartphone accelerometry enables accessible postural assessment, but whether complex temporal measures add value is unclear. We examined whether smartphone global postural sway measures discriminate older adults with and without previous falls in relation to force-platform measures, and whether Stabilogram Diffusion Analysis (SDA) adds value. Methods: Forty-eight community-dwelling adults aged ≥60 years were allocated by fall history (≥1 fall in the previous year) to Falls (n = 20) or Control (n = 28) groups. Quiet standing (eyes open/closed) was recorded simultaneously by a force platform and a lower-back smartphone accelerometer. Variables were compared with mixed analyses of variance (false discovery rate [FDR]-corrected) and between-device correlations; discrimination used cross-validated ridge logistic regression. Results: All 12 global variables differed between groups (p_FDR ≤ 0.035; 10 out of 12 in the original subsample of 40). Among SDA parameters, only amplitude-related measures differed. Cross-validated areas under the curve were 0.66–0.77 for global models; adding SDA never improved discrimination (ΔAUC −0.06 to 0.03; all 95% confidence intervals including zero), including in sensitivity analyses. Group differences were inseparable from Timed Up and Go performance. Conclusions: In this exploratory sample, smartphone global measures discriminated fall-history groups as well as SDA-based models; prospective, external validation is required before screening use.

Authors

Institutions

Publication Details

Journal
Applied Sciences
Published
2026-11-13
DOI
https://doi.org/10.3390/app162211247
Primary Topic
Balance, Gait, and Falls Prevention
Type
article
Field-Weighted Citation Impact
0.00

Funders

Controls
|||
ALL TIME
JAN
FEB
MAR
APR
MAY
JUN
JUL
AUG
SEP
OCT
NOV
article

Simplicity over Complexity: Limited Added Value of Diffusion Analysis for Smartphone-Based Fall-Risk Classification in Older Adults

Manuela Brito Duarte, Bianca Callegari, Luis Carlos Pereira Monteiro, Szymon Kuliś et al.
2 links
Applied Sciences
Balance, Gait, and Falls Prevention
article

Simplicity over Complexity: Limited Added Value of Diffusion Analysis for Smartphone-Based Fall-Risk Classification in Older Adults

Manuela Brito Duarte, Bianca Callegari, Luis Carlos Pereira Monteiro, Szymon Kuliś, Jan Gajewski, Aymee Lobato Brito, Nathalya Ingrid Cardoso do Nascimento, Gizele Cristina da Silva Almeida, Givago da Silva Souza, Gabriela Souza de Vasconcelos, Jéssica Cristina Santos de ASSIS
article en

Abstract

Background/Objectives: Smartphone accelerometry enables accessible postural assessment, but whether complex temporal measures add value is unclear. We examined whether smartphone global postural sway measures discriminate older adults with and without previous falls in relation to force-platform measures, and whether Stabilogram Diffusion Analysis (SDA) adds value. Methods: Forty-eight community-dwelling adults aged ≥60 years were allocated by fall history (≥1 fall in the previous year) to Falls (n = 20) or Control (n = 28) groups. Quiet standing (eyes open/closed) was recorded simultaneously by a force platform and a lower-back smartphone accelerometer. Variables were compared with mixed analyses of variance (false discovery rate [FDR]-corrected) and between-device correlations; discrimination used cross-validated ridge logistic regression. Results: All 12 global variables differed between groups (p_FDR ≤ 0.035; 10 out of 12 in the original subsample of 40). Among SDA parameters, only amplitude-related measures differed. Cross-validated areas under the curve were 0.66–0.77 for global models; adding SDA never improved discrimination (ΔAUC −0.06 to 0.03; all 95% confidence intervals including zero), including in sensitivity analyses. Group differences were inseparable from Timed Up and Go performance. Conclusions: In this exploratory sample, smartphone global measures discriminated fall-history groups as well as SDA-based models; prospective, external validation is required before screening use.

Applied SciencesVol. 16(22)
Universidade do Estado do Pará (BR), Universidade Federal do Pará (BR), Universidade Federal de Goiás (BR), Józef Piłsudski University of Physical Education in Warsaw (PL)
Conselho Nacional de Desenvolvimento Científico e Tecnológico, Fundação Amparo e Desenvolvimento da Pesquisa
Good health and well-being
Balance, Gait, and Falls Prevention
AI Navigator

Ask Laika to Summarize, Analyze, and Connect papers live on the map.

Summarize Papers & Methodologies

Extract key findings, datasets, and comparative methods across publications.

Benchmark Rankings & Visual Analytics

Rank top research institutions, authors, funders, topics, and journals by Field-Weighted Citation Impact (FWCI) and paper volume with instant charts.

Connect Distant Disciplines

Bridge topological clusters on the map to find hidden collaborative intersections.

Laika access coming soon