Search

As global demographics change, ageing is a global phenomenon which is increasingly of interest in our modern and rapidly changing society. Thus, the application of proper prognostic indices in clinical decisions regarding mortality prediction has assumed a significant importance for personalized risk management (i.e., identifying patients who are at high or low risk of death) and to help ensure effective healthcare services to patients. Consequently, prognostic modelling expressed as all-cause mortality prediction is an important step for effective patient management. Machine learning has the potential to transform prognostic modelling. In this paper, results on the development of machine learning models for all-cause mortality prediction in a cohort of healthy older adults are reported. The models are based on features covering anthropometric variables, physical and lab examinations, questionnaires, and lifestyles, as well as wearable data collected in free-living settings, obtained for the “Healthy Ageing Initiative” study conducted on 2291 recruited participants. Several machine learning techniques including feature engineering, feature selection, data augmentation and resampling were investigated for this purpose. A detailed empirical comparison of the impact of the different techniques is presented and discussed. The achieved performances were also compared with a standard epidemiological model. This investigation showed that, for the dataset under consideration, the best results were achieved with Random UnderSampling in conjunction with Random Forest (either with or without probability calibration). However, while including probability calibration slightly reduced the average performance, it increased the model robustness, as indicated by the lower 95% confidence intervals. The analysis showed that machine learning models could provide comparable results to standard epidemiological models while being completely data-driven and disease-agnostic, thus demonstrating the opportunity for building machine learning models on health records data for research and clinical practice. However, further testing is required to significantly improve the model performance and its robustness.

The increased use of sensor technology has been crucial in releasing the potential for remote rehabilitation. However, it is vital that human factors, that have potential to affect real-world use, are fully considered before sensors are adopted into remote rehabilitation practice. The smart sensor devices for rehabilitation and connected health (SENDoc) project assesses the human factors associated with sensors for remote rehabilitation of elders in the Northern Periphery of Europe. This article conducts a literature review of human factors and puts forward an objective scoring system to evaluate the feasibility of balance assessment technology for adaption into remote rehabilitation settings. The main factors that must be considered are: Deployment constraints, usability, comfort and accuracy. This article shows that improving accuracy, reliability and validity is the main goal of research focusing on developing novel balance assessment technology. However, other aspects of usability related to human factors such as practicality, comfort and ease of use need further consideration by researchers to help advance the technology to a state where it can be applied in remote rehabilitation settings.

Background Older adults may use wearable devices for various reasons, ranging from monitoring clinically relevant health metrics or detecting falls to monitoring physical activity. Little is known about how this population engages with wearable devices, and no qualitative synthesis exists to describe their shared experiences with long-term use. Objective This study aims to synthesize qualitative studies of user experience after a multi-day trial with a wearable device to understand user experience and the factors that contribute to the acceptance and use of wearable devices. Methods We conducted a systematic search in CINAHL, APA PsycINFO, PubMed, and Embase (2015-2020; English) with fixed search terms relating to older adults and wearable devices. A meta-synthesis methodology was used. We extracted themes from primary studies, identified key concepts, and applied reciprocal and refutational translation techniques; findings were synthesized into third-order interpretations, and finally, a “line-of-argument” was developed. Our overall goal was theory development, higher-level abstraction, and generalizability for making this group of qualitative findings more accessible. Results In total, we reviewed 20 papers; 2 evaluated fall detection devices, 1 tested an ankle-worn step counter, and the remaining 17 tested activity trackers. The duration of wearing ranged from 3 days to 24 months. The views of 349 participants (age: range 51-94 years) were synthesized. Four key concepts were identified and outlined: motivation for device use, user characteristics (openness to engage and functional ability), integration into daily life, and device features. Motivation for device use is intrinsic and extrinsic, encompassing many aspects of the user experience, and appears to be as, if not more, important than the actual device features. To overcome usability barriers, an older adult must be motivated by the useful purpose of the device. A device that serves its intended purpose adds value to the user’s life. The user’s needs and the support structure around the device—aspects that are often overlooked—seem to play a crucial role in long-term adoption. Our “line-of-argument” model describes how motivation, ease of use, and device purpose determine whether a device is perceived to add value to the user’s life, which subsequently predicts whether the device will be integrated into the user’s life. Conclusions The added value of a wearable device is the resulting balance of motivators (or lack thereof), device features (and their accuracy), ease of use, device purpose, and user experience. The added value contributes to the successful integration of the device into the daily life of the user. Useful device features alone do not lead to continued use. A support structure should be placed around the user to foster motivation, encourage peer engagement, and adapt to the user’s preferences.

Background Older adults with diabetes take fewer steps per day than those without diabetes. The purpose of the present study was to investigate the association of daily step count with incident diabetes in community-dwelling 70-year-olds. Methods This prospective cohort study included N = 3055 community-dwelling 70-year-olds (52% women) who participated in a health examination in Umeå, Sweden during 2012–2017, and who were free from diabetes at baseline. Daily step count was measured for 1 week using Actigraph GT3X+ accelerometers. Cases of diabetes were collected from the Swedish National Patient Register. The dose-response association was evaluated graphically using a flexible parametric model, and hazard ratios (HR) with 95% confidence intervals (CI) were calculated using Cox regressions. Results During a mean follow-up of 2.6 years, diabetes was diagnosed in 81 participants. There was an inverse nonlinear dose-response association between daily step count and incident diabetes, with a steep decline in risk of diabetes from a higher daily step count until around 6000 steps/day. From there, the risk decreased at a slower rate until it leveled off at around 8000 steps/day. A threshold of 4500 steps/day was found to best distinguish participants with the lowest risk of diabetes, where those taking ≥ 4500 steps/day, had 59% lower risk of diabetes, compared to those taking fewer steps (HR, 0.41, 95% CI, 0.25–0.66). Adjusting for visceral adipose tissue (VAT) attenuated the association (HR, 0.64, 95% CI, 0.38–1.06), which was marginally altered after further adjusting for sedentary time, education and other cardiometabolic risk factors and diseases (HR, 0.58, 95% CI, 0.32–1.05). Conclusions A higher daily step count is associated with lower risk of incident diabetes in community-dwelling 70-year-olds. The greatest benefits occur at the lower end of the activity range, and much earlier than 10,000 steps/day. With the limitation of being an observational study, these findings suggest that promoting even a modest increase in daily step count may help to reduce the risk of diabetes in older adults. Because VAT appears to partly mediate the association, lifestyle interventions targeting diabetes should apart from promoting physical activity also aim to prevent and reduce central obesity.