Background: Although geriatric depression is prevalent, diagnosis using self-reporting instruments has limitations when measuring the depressed mood of older adults in a community setting. Ecological momentary assessment (EMA) by using wearable devices could be used to collect data to classify older adults into depression groups. Objective: The objective of this study was to develop a machine learning algorithm to predict the classification of depression groups among older adults living alone. We focused on utilizing diverse data collected through a survey, an Actiwatch, and an EMA report related to depression. Methods: The prediction model using machine learning was developed in 4 steps: (1) data collection, (2) data processing and representation, (3) data modeling (feature engineering and selection), and (4) training and validation to test the prediction model. Older adults (N=47), living alone in community settings, completed an EMA to report depressed moods 4 times a day for 2 weeks between May 2017 and January 2018. Participants wore an Actiwatch that measured their activity and ambient light exposure every 30 seconds for 2 weeks. At baseline and the end of the 2-week observation, depressive symptoms were assessed using the Korean versions of the Short Geriatric Depression Scale (SGDS-K) and the Hamilton Depression Rating Scale (K-HDRS). Conventional classification based on binary logistic regression was built and compared with 4 machine learning models (the logit, decision tree, boosted trees, and random forest models). Results: On the basis of the SGDS-K and K-HDRS, 38% (18/47) of the participants were classified into the probable depression group. They reported significantly lower scores of normal mood and physical activity and higher levels of white and red, green, and blue (RGB) light exposures at different degrees of various 4-hour time frames (all P<.05). Sleep efficiency was chosen for modeling through feature selection. Comparing diverse combinations of the selected variables, daily mean EMA score, daily mean activity level, white and RGB light at 4:00 pm to 8:00 pm exposure, and daily sleep efficiency were selected for modeling. Conventional classification based on binary logistic regression had a good model fit (accuracy: 0.705; precision: 0.770; specificity: 0.859; and area under receiver operating characteristic curve or AUC: 0.754). Among the 4 machine learning models, the logit model had the best fit compared with the others (accuracy: 0.910; precision: 0.929; specificity: 0.940; and AUC: 0.960). Conclusions: This study provides preliminary evidence for developing a machine learning program to predict the classification of depression groups in older adults living alone. Clinicians should consider using this method to identify underdiagnosed subgroups and monitor daily progression regarding treatment or therapeutic intervention in the community setting. Furthermore, more efforts are needed for researchers and clinicians to diversify data collection methods by using a survey, EMA, and a sensor.

Demographic aging of the world population contributes to an increase in the number of persons diagnosed with dementia (PWD), with corresponding increases in health care expenditures. In addition, fewer family members are available to care for these individuals. Most care for PWD occurs in the home, and family members caring for PWD frequently suffer negative outcomes related to the stress and burden of observing their loved one's progressive memory and functional decline. Decreases in cognition and self-care also necessitate that the caregiver takes on new roles and responsibilities in care provision. Smart technologies are being developed to support family caregivers of PWD in a variety of ways, including provision of information and support resources online, wayfinding technology to support independent mobility of the PWD, monitoring systems to alert caregivers to changes in the PWD and their environment, navigation devices to track PWD experiencing wandering, and telemedicine and e-health services linking caregivers and PWD with health care providers. This paper will review current uses of these advancing technologies to support care of PWD. Challenges unique to widespread acceptance of technology will be addressed and future directions explored.

Background: Internet-based health resources can support informal caregivers who are caring for children or adolescents with health care needs. However, few studies discriminate informal caregivers’ needs from those of their care recipients or those of people caring for adults. Objective: This study reviews the literature of health-related Internet use among informal caregivers of children and adolescents. Methods: A total of 17 studies were selected from literature searches conducted in 6 electronic databases: PubMed, Cochrane, CINAHL, PsycINFO, ERIC, and EMBASE. All databases searches were limited to articles published in the years 2004 to 2014 in peer-reviewed publications. Search terms consisted of “health-related Internet use,” “eHealth,” “Internet use for health-related purpose(s),” “Web-based resource(s),” and “online resources,” combined with informal caregiver (or “parents”) of “child,” “adolescent,” “student,” “youth,” and “teen.” The age range of the children receiving care was limited to younger than 22 years. Their informal caregivers were defined as persons (parents) who provided unpaid care or assistance to a child or an adolescent with health problems. Results: Among 17 empirical studies, the majority of informal caregivers of children with medical issues were the parents. Quantitative studies (14/17, 77%) reported prevalence and predictors of health-related Internet use, while mixed-methods and qualitative studies (3/17, 24%) investigated informal caregiver perceptions of helpful health-related Internet use and barriers of use. The prevalence of health-related Internet use varied (11%-90%) dependent upon how health-related Internet use was operationalized and measured. Disease-specific information was used for decision making about treatment, while social support via virtual communities and email were used for informal caregiver emotional needs. A digital divide of Internet access was identified in lower educated minorities. Most studies had methodological challenges resulting from convenience sampling, cross-sectional surveys, lack of theoretical frameworks, or no clear definitions of health-related Internet use. Conclusions: This study provides an important understanding of how family members use Internet-based information and support systems during child caregiving. Healthcare providers and policy makers should integrate family needs into their current practices and policies. Further rigorous research is required to design efficient and effective nursing interventions.

Background: Informal caregivers of persons with dementia experience higher levels of chronic stress in the caregiving trajectory. The Internet provides diverse types of caregiver resources that may help ameliorate their stress and relevant negative outcomes. However, there is limited information about the prevalence and factors of using Internet-based resources for health- and caregiving-related purposes in informal caregivers of persons with dementia. Objective: Specific aims of this study were (1) to determine the prevalence and factors of caregiver’s health-related Internet use and (2) to compare sociodemographic and caregiving-related characteristics between health-related Internet users and non–health-related Internet users among informal caregivers of persons with dementia. Methods: This quantitative investigation was a descriptive correlational design using a secondary data analysis. Primary data were collected via a survey conducted in 2009 by the National Alliance for Caregiving and the American Association of Retired Persons. Telephone interviews utilizing standardized questionnaires were used to collect self-reported information about sociodemographics and caregiving-related history (N=450). Descriptive statistics and a hierarchical binary logistic regression analysis were completed based on the stress process model. Results: Approximately 59% (265/450) of dementia caregivers were identified as health-related Internet users. Caregivers’ sociodemographics and their subjective responses of caregiving stress were the most significant factors to identify health-related Internet users followed by workload assisting in instrumental activities of daily living of persons with dementia. There were significant differences for caregiver’s age, levels of education and income, hours spent caregiving, and the relationship to persons with dementia between health-related Internet users and non–health-related Internet users ( P <.05 for all). After controlling for confounding effects, younger age of persons with dementia (OR 0.278, 95% CI 0.085-0.906), higher education levels of caregivers (OR 3.348, 95% CI 2.019-5.552), shorter caregiving time spent per week (OR 0.452, 95% CI 0.243-0.840), higher levels of caregiver’s emotional stress (OR 1.249, 95% CI 1.004-1.555), and financial hardship (OR 4.61, 95% CI 1.416-14.978) were identified as newly emerging factors of health-related Internet use. Conclusions: Although the Internet provided useful resources for caregivers of persons with dementia, dementia caregivers reported lower levels of health-related Internet use compared to the general public. Our findings confirmed the impact of age, education levels, and/or income on Internet use reported in previous studies. However, the predictive value of subjective responses of caregiving stress for health-related Internet use was a new addition. These findings will assist health care providers, researchers, and policy makers in identifying who is the least likely to access Internet-based resources and how Internet-based strategies can best be designed, implemented, and distributed to meet the needs of this group of users. [Interact J Med Res 2015;4(1):e1]

Background: The need for digital literacy in aging populations is increasing in the digitalizing society. Digital literacy involves the identification, evaluation, and communication of information through various digital devices or relevant programs. Objective: The aims of this study were to develop an Everyday Digital Literacy Questionnaire (EDLQ), a digital literacy assessment scale, and subsequently evaluate its psychometric properties using a population of community-dwelling older adults in South Korea. Methods: The EDLQ was developed using an instrument development design. A nationwide survey was conducted, and the study included 1016 community-dwelling older adults (age ≥60 years). To evaluate the psychometric properties, the participants were randomly divided into 2 groups (n=508 each), and the internal consistency (Cronbach α and McDonald ω), structural validity (exploratory factor analysis and confirmatory factor analysis), hypothesis-testing construct validity using the eHealth Literacy Scale (eHEALS), and measurement invariance were analyzed. Results: Among the initial 30 items of the EDLQ, 22 items with a 3-factor solution had a total explained variance of 77%. The domains included “information and communication” (9 items), “content creation and management” (4 items), and “safety and security” (9 items). Confirmatory factor analysis was conducted with this 3-factor solution (χ2206=345.1; normed χ2206=1.7; comparative fit index=0.997; Tucker-Lewis index=0.997; root-mean-square error of approximation=0.036; standardized root-mean-square residual=0.050; composite reliability=0.903-0.959; average variance extracted=0.699-0.724; R2=0.616-0.773). Hypothesis-testing construct validity with the eHEALS revealed a strong correlation (r=0.75). Cronbach α and McDonald ω coefficients were .98 and 0.98, respectively. The fit indices for measurement invariance, including the configural, metric, and scalar invariance models, demonstrated a satisfactory fit to the data. Our findings suggest that the psychometric properties of the 22-item EDLQ are valid and reliable for assessing digital literacy among older Korean adults. Conclusions: In this study, we developed a digital literacy measure with strong psychometric properties that made it suitable for assessing the digital literacy of community-dwelling older adults in Korea. To broaden its applicability, however, further assessment of its feasibility for use with different languages and cultures is necessary. Moreover, more empirical research on digital literacy and related factors in older adults can facilitate the development of personalized digital health care services and educational interventions in the digital society.