Dataset Information

Actigraphy features for predicting mobility disability in older adults.

ABSTRACT: Actigraphy has attracted much attention for assessing physical activity in the past decade. Many algorithms have been developed to automate the analysis process, but none has targeted a general model to discover related features for detecting or predicting mobility function, or more specifically, mobility impairment and major mobility disability (MMD). Men (N??=??357) and women (N??=??778) aged 70-89 years wore a tri-axial accelerometer (Actigraph GT3X) on the right hip during free-living conditions for 8.4??±??3.0 d. One-second epoch data were summarized into 67 features. Several machine learning techniques were used to select features from the free-living condition to predict mobility impairment, defined as 400 m walking speed??<0.80 m s-1. Selected features were also included in a model to predict the first occurrence of MMD-defined as the loss in the ability to walk 400 m. Each method yielded a similar estimate of 400 m walking speed with a root mean square error of ~0.07 m s-1 and an R-squared values ranging from 0.37-0.41. Sensitivity and specificity of identifying slow walkers was approximately 70% and 80% for all methods, respectively. The top five features, which were related to movement pace and amount (activity counts and steps), length in activity engagement (bout length), accumulation patterns of activity, and movement variability significantly improved the prediction of MMD beyond that found with common covariates (age, diseases, anthropometry, etc). This study identified a subset of actigraphy features collected in free-living conditions that are moderately accurate in identifying persons with clinically-assessed mobility impaired and significantly improve the prediction of MMD. These findings suggest that the combination of features as opposed to a specific feature is important to consider when choosing features and/or combinations of features for prediction of mobility phenotypes in older adults.

SUBMITTER: Kheirkhahan M

PROVIDER: S-EPMC5360536 | biostudies-literature | 2016 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Actigraphy features for predicting mobility disability in older adults.

Kheirkhahan Matin M Tudor-Locke Catrine C Axtell Robert R Buman Matthew P MP Fielding Roger A RA Glynn Nancy W NW Guralnik Jack M JM King Abby C AC White Daniel K DK Miller Michael E ME Siddique Juned J Brubaker Peter P Rejeski W Jack WJ Ranshous Stephen S Pahor Marco M Ranka Sanjay S Manini Todd M TM

Physiological measurement 20160921 10

Actigraphy has attracted much attention for assessing physical activity in the past decade. Many algorithms have been developed to automate the analysis process, but none has targeted a general model to discover related features for detecting or predicting mobility function, or more specifically, mobility impairment and major mobility disability (MMD). Men (N = 357) and women (N = 778) aged 70-89 years wore a tri-axial accelerometer (Actigraph GT3X) on the right hip during free-living condit ...[more]

PMID: 27653966

Similar Datasets

Project description:BackgroundThe ability to walk at various speeds is essential to independence for older adults. Maintaining fast walking requires changes in spatial-temporal measures, increasing step length and/or decreasing step time. It is unknown how mobility affects the parameters that change between preferred and fast walking.Research questionHow does preferred walking performance and measures of strength and mobility relate to the approach (decreasing step time or increasing step length) older adults at risk for mobility disability use to maintain fast walking speeds?.MethodsPeak isokinetic dynamometry of knee and ankle and several mobility evaluations, including the Timed Up-and-Go, Short Physical Performance Battery, and Dynamic Gait Index, assessed mobility and strength in 57 participants, aged 65-80. Biomechanical gait analysis was used to analyze step length, step time, gait speed at preferred and fast gait speeds and ground reaction force during preferred walking. A score combining the differences between step length and time at fast and preferred speeds (Length-Time Difference) separated participants into two groups: (1) Length, representing a predominant increase in step length to walk fast and (2) Time, a predominant decrease in step time.ResultsThose who decreased step time to produce increased speed performed worse during repeated chair stands (p = .006) with no difference in isokinetic strength (p ≥ .15). During preferred walking, the Time group displayed increased propulsive impulse compared to the Length group (p = .007), despite no differences in preferred speed, step length, or time (p ≥ .50).SignificanceWhile kinetics of preferred walking differed between groups separated by Length-Time Difference, basic spatial-temporals of preferred walking did not in this homogenous population. Length-Time Difference relates to a common mobility assessment and could be easily calculated by clinicians to provide a quantitative and more sensitive measure of ambulatory performance.

Project description:ObjectiveTo determine whether a multicomponent intervention based on physical activity with technological support and nutritional counselling prevents mobility disability in older adults with physical frailty and sarcopenia.DesignEvaluator blinded, randomised controlled trial.Setting16 clinical sites across 11 European countries, January 2016 to 31 October 2019.Participants1519 community dwelling men and women aged 70 years or older with physical frailty and sarcopenia, operationalised as the co-occurrence of low functional status, defined as a short physical performance battery (SPPB) score of 3 to 9, low appendicular lean mass, and ability to independently walk 400 m. 760 participants were randomised to a multicomponent intervention and 759 received education on healthy ageing (controls).InterventionsThe multicomponent intervention comprised moderate intensity physical activity twice weekly at a centre and up to four times weekly at home. Actimetry data were used to tailor the intervention. Participants also received personalised nutritional counselling. Control participants received education on healthy ageing once a month. Interventions and follow-up lasted for up to 36 months.Main outcome measuresThe primary outcome was mobility disability (inability to independently walk 400 m in <15 minutes). Persistent mobility disability (inability to walk 400 m on two consecutive occasions) and changes from baseline to 24 and 36 months in physical performance, muscle strength, and appendicular lean mass were analysed as pre-planned secondary outcomes. Primary comparisons were conducted in participants with baseline SPPB scores of 3-7 (n=1205). Those with SPPB scores of 8 or 9 (n=314) were analysed separately for exploratory purposes.ResultsMean age of the 1519 participants (1088 women) was 78.9 (standard deviation 5.8) years. The average follow-up was 26.4 (SD 9.5) months. Among participants with SPPB scores of 3-7, mobility disability occurred in 283/605 (46.8%) assigned to the multicomponent intervention and 316/600 (52.7%) controls (hazard ratio 0.78, 95% confidence interval 0.67 to 0.92; P=0.005). Persistent mobility disability occurred in 127/605 (21.0%) participants assigned to the multicomponent intervention and 150/600 (25.0%) controls (0.79, 0.62 to 1.01; P=0.06). The between group difference in SPPB score was 0.8 points (95% confidence interval 0.5 to 1.1 points; P<0.001) and 1.0 point (95% confidence interval 0.5 to 1.6 points; P<0.001) in favour of the multicomponent intervention at 24 and 36 months, respectively. The decline in handgrip strength at 24 months was smaller in women assigned to the multicomponent intervention than to control (0.9 kg, 95% confidence interval 0.1 to 1.6 kg; P=0.028). Women in the multicomponent intervention arm lost 0.24 kg and 0.49 kg less appendicular lean mass than controls at 24 months (95% confidence interval 0.10 to 0.39 kg; P<0.001) and 36 months (0.26 to 0.73 kg; P<0.001), respectively. Serious adverse events occurred in 237/605 (39.2%) participants assigned to the multicomponent intervention and 216/600 (36.0%) controls (risk ratio 1.09, 95% confidence interval 0.94 to 1.26). In participants with SPPB scores of 8 or 9, mobility disability occurred in 46/155 (29.7%) in the multicomponent intervention and 38/159 (23.9%) controls (hazard ratio 1.25, 95% confidence interval 0.79 to 1.95; P=0.34).ConclusionsA multicomponent intervention was associated with a reduction in the incidence of mobility disability in older adults with physical frailty and sarcopenia and SPPB scores of 3-7. Physical frailty and sarcopenia may be targeted to preserve mobility in vulnerable older people.Trial registrationClinicalTrials.gov NCT02582138.

Dataset Information

Actigraphy features for predicting mobility disability in older adults.

Publications

Actigraphy features for predicting mobility disability in older adults.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure

Tweets