Purpose/Hypothesis: Falls in the older adult population are the leading cause of fatal and non-fatal injuries in America. Polypharmacy, the use of multiple medications, has been identified as a major risk factor for falls in older adults. A variety of medication screens exist that identify adverse effects of medications which can directly impact fall risk; however, current screening measures have limitations. The Quantitative Drug Index (QDI) is a new, clinically anchored index to quantify all potential adverse effects associated with drug-mediated fall risk. The purpose of this study was to validate the QDI as a fall risk screening tool. Number of Subjects: 138 adults were recruited from local senior centers and screened. Inclusion criteria: community-dwelling, age 60 to 89 years, and currently prescribed at least one medication. Exclusion criteria: progressive neurological disorders, unstable medical conditions, cognitive impairment, severe depression or anxiety, severe lower extremity impairment that would impact mobility, and severe vision impairment. Materials and Methods: Mobility and balance outcome measures related to fall risk included: 30-second chair stand test, 10-meter walk test, Timed Up and Go (TUG) and Dynamic Gait Index (DGI). Self-report measures of fall risk included fall history, Fall Risk Questionnaire (FRQ) and Activity-specific Balance Confidence scale (ABC). The QDI was derived from each participant's medications. Participants were classified as either fallers or nonfallers based on self-report history of falls within the past year. Nonparametric Spearman’s Rho correlations were used to determine relationships between faller status and measures of fall risk. A receiver operating characteristic (ROC) curve analysis determined cutoff scores for outcome measures related to faller status. Results: A fair to moderate relationship between the QDI and several physical performance and self-report measures was identified: FRQ (r=0.363), ABC (r=-0.401), DGI (r=-0.360). However, little to no relationship was found between faller status and QDI score (r=0.221). The ROC analysis determined the area under the curve for QDI was 0.63 with a cutoff score of 2.5 yielding sensitivity of 78% and specificity of 47%. Conclusions: The development of the QDI was an interdisciplinary effort between pharmacists and physical therapists to screen for fall risk in older individuals. The QDI offers a better way to quantify the adverse effects of drugs on mobility compared with simple drug counts. The QDI alone does not identify individuals at fall risk; however, the QDI is significantly correlated to several measures of fall risk, including FRQ, ABC, and DGI. The ROC Curve Analysis identified a cutoff score for fall risk for the QDI which was found to have similar sensitivity and specificity to the TUG. Clinical Relevance: The QDI could be incorporated into electronic medical records to identify patients who may be at fall risk and would be appropriate for further balance and mobility evaluation.

No description available.

Digital healthcare monitoring uses multidisciplinary sensing techniques to track diverse human data and behaviors. Machine learning can promote an individual's well-being through more efficient and accurate health status monitoring. However, challenges hinder precise monitoring, such as privacy concerns, varied subjects, diverse sensors, and different objectives. To help address these challenges, this thesis explores projects spanning various healthcare domains. Explainable and robust machine-learning solutions are proposed and tested, which include novel signal processing guidelines, innovative feature engineering methods, and pioneering deep-learning networks. These solutions contribute to the state-of-the-art in their respective healthcare domains. The first project addressed the challenge of assessing fall risk among individuals with varying levels of mobility using inertial sensors. Machine-learning models were developed and evaluated using datasets from stroke survivors and community-dwelling elders with participants of varying levels of mobility. Risk indicators were obtained through kinematics simplification that are both explainable and modifiable. These indicators considerably enhance fall risk classification performance compared to existing approaches and the conclusions align with available biomechanical evidence. In the second project, a new machine-learning architecture was created for fall detection and classification using multistatic radar sensing. This new approach (called eMSFRNet) solved the common problem of weak and varied Doppler signatures caused by line-of-sight restrictions. It is the first method that can classify among fall types using radar sensing, and yielded state-of-the-art accuracy for both fall detection (99.3%) and seven fall types classification (76.8%) tasks. In the third project, a novel combination of signal processing and a machine learning framework (named MIND) was designed to detect and forecast motor restricted and repetitive behaviors (RRBs) among children with autism spectrum disorder (ASD), using data from multiple wearable sensors. Contrary to prior beliefs that such detection or forecasting was unattainable, the novel MIND AI framework offers a comprehensive and generalizable approach. Transition behaviors were first defined and then identified, suggesting the potential to detect behavioral shifts preceding motor RRBs. The new signal monitoring quantification (MQ) guidelines minimize the impacts of inconsistent data caused by individualized sensor placements. MIND achieved 100% accuracy in detecting motor RRBs on new subjects with unfamiliar behavior types and 92.2% accuracy in forecasting motor RRBs. In conclusion, the work in this thesis showcases the pivotal contributions of robust and explainable machine learning solutions tailored for specific healthcare challenges. These contributions either solve longstanding problems in different healthcare fields or guide new research directions. The new methodologies introduced – including the MQ guidelines, modifiable fall risk indicators, and innovative deep learning models – all help to advance healthcare machine learning applications by merging accuracy with explainability. / Doctor of Philosophy / Digital healthcare monitoring uses advanced techniques to monitor a person's health and behavior. With the help of machine learning (think of this as teaching computers to think and learn), it is possible to improve health monitoring to be faster and more accurate. Still, there are important challenges to overcome, including concerns regarding personal privacy, the variety of ways in which data can be collected, and the diverse goals of each monitoring tool. This research addressed these challenges by creating and evaluating new machine learning methods for application to multiple healthcare areas. New, understandable, and powerful machine learning methods were developed, pushing the boundaries of how best to use varied technologies for monitoring. A few highlights of the research include the following. First, a method was developed to better determine if an older adult is at a higher risk of falling. The ability of the method to estimate falling risk was very strong, and superior to previously-reported methods. This new method can also explain why an individual might be at a higher risk of falling and offers suggestions on how to walk more stably. Second, a technique was created to process radar signals to detect falls and to determine the type of fall that occurred. This technique solves a long-standing problem with radar, specifically that this sensing technology often provides unclear and unstable signals. Third, a machine-learning method was constructed to identify repetitive (self-injurious) behaviors among children with autism spectrum disorder using signals from wearable sensors. This novel method can detect behaviors quite accurately, even in challenging scenarios. One notable finding was changes in normal behavior can be identified shortly before the repetitive behaviors occur. Overall, this research contributes substantially new and effective methods for healthcare and understandable machine learning solutions. These contributions help to solve challenging, ongoing problems and pave the way for future innovations. Methods such as those developed promise a future where technology can better assist in healthcare, making it more precise and understandable for everyone.

No description available.

Intraocular lenses (IOLs) provide distance and near refraction and are becoming the standard for cataract surgery. Multifocal glasses increase the variability of toe clearance in older adults navigating stairs and increase fall risk; however, little is known about the biomechanics of stair navigation in individuals with multifocal IOLs. This study compared clearance while ascending and descending stairs in individuals with monofocal versus multifocal IOLs. Eight participants with multifocal IOLs (4 men, 4 women; mean age = 66.5 yr, standard deviation [SD] = 6.26) and fifteen male participants with monofocal IOLs (mean age = 69.9 yr, SD = 6.9) underwent vision and mobility testing. Motion analysis recorded kinematic and custom software-calculated clearances in three-dimensional space. No significant differences were found between groups on minimum clearance or variability. Clearance differed for ascending versus descending stairs: the first step onto the stair had the greatest toe clearance during ascent, whereas the final step to the floor had the greatest heel clearance during descent. This preliminary study indicates that multifocal IOLs have similar biomechanic characteristics to monofocal IOLs. Given that step characteristics are related to fall risk, we can speculate that multifocal IOLs carry no additional fall risk.

Residents in nursing facilities are more prone to falls than those living in the community. Injuries resulting from falls impact residents, their families, and healthcare costs. The gap in nursing practice was the lack of a comprehensive fall-prevention program in a long-term care facility that had experienced high fall rates among residents. This project addressed whether an educational program using the American Medical Directors Association's clinical practice guideline and the Centers for Disease Control and Prevention's STEADI (Stopping Elderly Accidents, Deaths, & Injuries) toolkit for fall- prevention improved the self-efficacy of direct-care staff in preventing falls among residents in a long-term care facility. The practice-focused question focused on whether education on the use of an integrated multifactorial fall-prevention guideline would increase confidence of long-term care staff in reducing falls in long-term care residents. The evaluation used the 11-item Self-Efficacy for Preventing Falls-Nurse scale for 5 licensed nursing staff and the 8-item Self-Efficacy for Preventing Falls-Assistant scale for 21 nursing assistants. The positive change in self-efficacy scores of nurses and nursing assistants after the education program was greatest for face-to-face team communication regarding fall risk and individual resident prevention plans. The use of best-practice guidelines that improve fall risk-assessment and use of fall precautions to decrease the number of falls and falls with injury has the potential to bring about positive social change by improving the nursing care of nursing home residents, resulting in improved resident safety and quality of life.

Falls and fall related injuries are a major health issue to the elderly population of Sweden. The number of accidents in the community caused by falls is increasing. It is mainly the elderly who need to be hospitalized after a fall. Patients with stroke, cognitive disorders, or hip fracture, have a particularly high risk of falling. A common consequence of falls is femur fractures. To prevent and reduce the occurrences of falls in hospitals, an individual care plan aimed at the patient, is required, as well as an individual and feasible nursing intervention. It also require that health professionals use their knowledge and the patient’s and the hospital's resources, to implement the care plan. Since 2008 SKL runs a national effort to reduce preventable harm and improve patient safety, where fall related injuries are an area of focus. Aim: The aim with this study is to examine and describe health professionals' attitudes towards falls and fall prevention, and whether the nursing staff considers themselves to be proficient in falls and fall prevention. Method: This was a quantitative study and Mann Whitney U test were conducted in order to analyse and summarize the answers from the questionnaires. Results: The respondents have a positive attitude towards fall prevention. The perception among the nursing staff is that they feel well trained in fall prevention, and have sufficient knowledge in fall prevention. Conclusion: The conclusion of this study is that nurses and nursing staff take falls among patients very seriously and have a positive attitude towards fall prevention. They also consider themselves to have a good knowledge about falls and fall prevention. / Fall och fallskador är ett stort folkhälsoproblem för den äldre befolkningen i Sverige. Antalet olyckor i samhället orsakade av fall ökar. Det är de äldre som oftast behöver sjukhusvård efter fallolyckor. Patienter med stroke, kognitiva störningar eller höftfrakturer har en särskilt hög risk att råka ut för fall och fallskador. En vanligt förekommande konsekvens av fall är höftfrakturer. För att förebygga och minska att fall sker krävs en individuell vårdplan för patienten och individuella och genomförbara omvårdnadsåtgärder. Det krävs att vårdpersonal använder sina kunskaper och patientens och sjukhusets resurser för att genomföra vårdplanen. Sedan 2008 driver SKL en nationell satsning för att minska antalet vårdskador och öka patientsäkerheten. Ett område som omfattas av denna nationella satsning är fall och skador till följd av fall. Syfte: Syftet med den här studien är att undersöka och beskriva vårdpersonalens attityder till fall och fallprevention och om vårdpersonal anser sig ha goda kunskaper om fall och fallprevention. Metod: Studien har använt sig av en kvantitativ metod i form av en enkätundersökning. Beskrivande statistik och Mann Whitney U test användes för att bearbeta enkätsvaren. Resultat: Resultatet visar att vårdpersonal till övervägande del har en positiv uppfattning till fallprevention och ser allvarligt på fallskador. Enkäten visar att vårdpersonalen upplever och känner att de är välutbildade i fallprevention och har kunskaper för att ge och skapa en god fallprevention i sitt arbete. Slutsats: Slutsatsen i denna studie är att vårdpersonal tar fall och fallskador på allvar och har en positiv uppfattning till fallprevention. De anser sig också ha goda kunskaper inom fall och fallprevention.

xiii, 109 p. : ill. / Falls are a significant source of physical, social, and psychological suffering among elderly adults. Falls lead to morbidity and even mortality. Over one-third of adults over the age of 65 years will fall within a calendar year, with almost 10,000 deaths per year attributed to falls. The direct cost of falls exceeds $10 billion a year in the United States. Fall incidents have been linked to multiple risk factors, including cognitive function, muscle strength, and balance control. The ability to properly identify balance impairment is a tremendous challenge to the medical community, with accurate assessment of fall risk lacking. Therefore, the purpose of this study was to assess balance control during gait among young adults, elderly adults, and elderly fallers; determine which biomechanical measures can best identify fallers retrospectively; demonstrate longitudinal changes in elderly adults and prospectively assess fall risk; and provide a method for mapping clinical variables to sensitive balance control measures using artificial neural networks. The interaction of the whole body center of mass (CoM) in relation to the base of support (BoS) assessed static and dynamic balance control throughout gait. Elderly fallers demonstrated reduced balance control ability, specifically a decreased time to contact with the boundary of the BoS, when compared to young adults at heel strike. This decreased time might predispose older adults to additional falls due to an inability to properly respond to perturbations or slips. Inclusion of these balance control measures along with the Berg Balance Scale and spatiotemporal measures demonstrated sensitivity and specificity values of up to 90% when identifying 98 elderly fallers and non-fallers, respectively. Additionally, 27 older adults were followed longitudinally over a period of one year, with only the interaction of the CoM with the BoS demonstrating an ability to differentiate fallers and non-fallers prospectively. As the collection and analysis of these biomechanics measures can be time consuming and expensive, an artificial neural network demonstrated that clinical measures can accurately predict balance control during ambulation. This model approached a solution quickly and provides a means for assessing longitudinal changes, intervention effects, and future fall risk. This dissertation includes both previously published and unpublished co-authored material. / Committee in charge: Dr. Li-Shan Chou, Chair; Dr. Andrew Karduna, Member; Dr. Marjorie Woollacott, Member; Dr. Ronald Stock, Member; Dr. Arthur Farley, Outside Member

[pt] Com o aumento da popularidade e acessibilidade de sistemas de realidade virtual (RV) de alta qualidade, tem-se levantado preocupações com relação a tendência dos sistemas de realidade virtual em provocar perda de equilíbrio. O equilíbrio é essencial para o uso seguro da realidade virtual e a perda do mesmo pode causar lesões graves. O objetivo deste trabalho é criar um sistema para avaliar o impacto da realidade virtual no equilíbrio humano. Neste trabalho, propomos e conduzimos um experimento usando o Oculus Rift e o MS Kinect Sensor. Nesse experimento, foi possível observar, quantificar e comparar o efeito de diferentes cenas de RV no equilíbrio dos usuários, bem como o efeito de avisos visuais e sonoros sobre perda de equilíbrio. / [en] With the increasing popularity and accessibility of high-quality Virtual Reality (VR) systems, concerns have been raised about the propensity of VR to induce balance loss. Balance is essential for safe use of VR experience and its loss can result in severe injury. This project is set to create a system able to measure the impact of VR in the human balance system. In this work, we design and conduct an experiment making use of the Oculus Rift VR headset and MS Kinect Sensor. In this experiment, we are able to visualize, quantify, and compare the effect of different VR scenes on the balance of the experiment subjects as well as the effect of visual and auditory warnings of balance loss.

One and a half million people are currently living in residential care facilities; as the baby boomer generation ages, this number will increase to 3 million. Approximately 3 out of 4 residents of these facilities fall each year, and 10% to 20% of those falls result in serious injuries such as fractures, disability, and a decreased quality of living. The BOUNCE Back fall initiative is a multifactorial program that uses a systematic approach starting on admission and to re-evaluate a resident following a fall. Nursing and therapy uses the Morse Fall Scale and the Elderly Mobility Scale to assess and categorize the resident's risk for falls. Guided by Lewin's theory of change, this project was designed to assess the effectiveness of the fall initiative as a quality improvement 60-day (August 2016- September 2016) pilot study in a skilled nursing and rehabilitation facility as a potential means to reduce the number of resident falls. Sixty residents (aged 64 to 98, mean age 81) were assessed at a minimum 2 time points to determine their level of fall risk and needed intervention, within 60 minutes of admission to the facility and 7 days postadmission. De-identified pre- and post-implementation data were provided from the corporate quality measure database, entered into a spreadsheet, and numbers were compared. As a result of the fall prevention pilot, for August 2016, 5 falls occurred with no repeat fallers; September 2016, 3 falls with 1 repeat faller which is a significant decrease from 14-22 falls occurring per month for 2 consecutive years. Following implementation, the facility scored 3%-5% for the number of falls, which is below the 7% threshold set forth by the pilot facility's corporate office. Prior to the implementation of the initiative, the facility had not met the 7% fall threshold in 2 years