Addressing the critical need for accurate fall event detection due to their potentially severe impacts, this paper introduces the Spatial Channel and Pooling Enhanced You Only Look Once version 5 small (SCPE-YOLOv5s) model. Fall events pose a challenge for detection due to their varying scales and subtle pose features. To address this problem, SCPE-YOLOv5s introduces spatial attention to the Efficient Channel Attention (ECA) network, which significantly enhances the model\'s ability to extract features from spatial pose distribution. Moreover, the model integrates average pooling layers into the Spatial Pyramid Pooling (SPP) network to support the multi-scale extraction of fall poses. Meanwhile, by incorporating the ECA network into SPP, the model effectively combines global and local features to further enhance the feature extraction. This paper validates the SCPE-YOLOv5s on a public dataset, demonstrating that it achieves a mean Average Precision of 88.29 %, outperforming the You Only Look Once version 5 small by 4.87 %. Additionally, the model achieves 57.4 frames per second. Therefore, SCPE-YOLOv5s provides a novel solution for fall event detection.

Falls, which are prevalent among older adults, may not only cause severe physical injuries, but also lead to low fall self-efficacy (FSE). Low FSE is associated with restricted activity, which putatively increases risk of future falls. However, emerging studies have failed to confirm this association. Furthermore, the interplay between age, gender, and fall history with falls has not been adequately addressed in adults aged 70 years or older. The aims of this secondary analysis were to: (1) prospectively explore the association of FSE and fall events considering age, gender, and fall history, and (2) examine the characteristics of fall events and fall-related outcomes. Forty-seven community-dwelling adults over 70 years of age were followed for about 12 months. During the follow-up, 22 participants with low FSE experienced 119 fall events whereas 25 participants with high FSE reported 106 fall events. Among fallers, 72.3% (n = 34) experienced recurrent fall events. About 15.0% (n = 34) of 225 fall events resulted in injuries and 4.0% of injuries required medical care. FSE was a statistically significant predictor of future fall events (incident rate ratio = 0.96, p = .013) regardless of age, gender, and fall history. Participants with low FSE were more likely than those with high FSE to fall more frequently without noticeable prodromal symptoms and apparent reasons. These findings suggest that FSE is an important protective factor against future fall events. However, interpretation of these results requires caution given the small sample size and effect size.

Background: Falls present a major challenge to active aging, but the relationship between neighborhood factors and falls is poorly understood. This study examined the relationship between fall events and neighborhood factors, including neighborhood social cohesion (sense of belonging, trust, friendliness, and helpfulness) and physical environment (vandalism/graffiti, rubbish, vacant/deserted houses, and perceived safety walking home at night). Methods: Data were analyzed from 9259 participants over four biennial waves (2006-2012) of the Health and Retirement Study (HRS), a nationally representative sample of adults aged 65 and older in the United States. Results: In models adjusting for demographic and health-related covariates, a one-unit increase in neighborhood social cohesion was associated with 4% lower odds of experiencing a single fall (odds ratio (OR): 0.96, 95% confidence interval (CI): 0.93-0.99) and 6% lower odds of experiencing multiple falls (OR: 0.94, 95% CI: 0.90-0.98). A one-unit increase in the physical environment scale was associated with 4% lower odds of experiencing a single fall (OR: 0.96, 95% CI: 0.93-0.99) and with 5% lower odds of experiencing multiple falls (OR: 0.95, 95% CI: 0.91-1.00) in adjusted models. Conclusions: The physical and social neighborhood environment may affect fall risk among community-dwelling older adults. Findings support the ongoing need for evidence-based fall prevention programming in community and clinical settings.

OBJECTIVE: To obtain insight into what kind of monitoring technologies exist to monitor activity in-home, what the characteristics and aims of applying these technologies are, what kind of research has been conducted on their effects and what kind of outcomes are reported.
METHODS: A systematic document search was conducted within the scientific databases Pubmed, Embase, Cochrane, PsycINFO and Cinahl, complemented by Google Scholar. Documents were included in this review if they reported on monitoring technologies that detect activities of daily living (ADL) or significant events, e.g. falls, of elderly people in-home, with the aim of prolonging independent living.
RESULTS: Five main types of monitoring technologies were identified: PIR motion sensors, body-worn sensors, pressure sensors, video monitoring and sound recognition. In addition, multicomponent technologies and smart home technologies were identified. Research into the use of monitoring technologies is widespread, but in its infancy, consisting mainly of small-scale studies and including few longitudinal studies.
CONCLUSIONS: Monitoring technology is a promising field, with applications to the long-term care of elderly persons. However, monitoring technologies have to be brought to the next level, with longitudinal studies that evaluate their (cost-) effectiveness to demonstrate the potential to prolong independent living of elderly persons. [Box: see text].