Smart Devices and Better Health: Positive Associations Between Smart Devices and Health Among Older Adults

Abstract

Introduction

Smart devices hold significant value for enhancing the health of older adults by promoting healthcare utilization and supporting chronic disease prevention.

Methods

This study investigates the relationship between smart device usage and health status among older adults, along with the underlying mechanisms, using panel data from the 2018 and 2020 waves of the China Longitudinal Aging Social Survey (CLASS). The baseline analysis employs a Two-Way Fixed Effects model to estimate the relationship between smart device usage and multiple health indicators. To ensure robustness, we further conduct instrumental variable estimation and exclude specific interfering samples. Additionally, heterogeneity across demographic subgroups is examined using interaction-term-based split-sample regression.

Results

Use of smart wristbands is associated with better self-rated health (SRH), lower depression scores, and higher social adaptation. Smart assistants and audiobooks are also associated with better social adaptation. Overall, smart devices enhance health by increasing healthcare utilization, encouraging social participation, strengthening social support, and facilitating early diagnosis of chronic diseases. These effects vary notably across gender, marital status, residence, age, and education level. Moreover, smart device use is linked to a pronounced reduction in medical expenditures.

Conclusion

The findings reveal a positive association between smart device use and health outcomes among older adults, indicating the important role of smart devices in supporting health management and alleviating medical burdens among older adults. This study provides empirical evidence for promoting digital health interventions and suggests that policy efforts should consider demographic differences to enhance the equitable adoption of smart technologies.

Keywords

smart devices health status TWFE instrumental variable healthy aging

Introduction

With the Internet becoming an integral part of social life, smart devices are increasingly shaping the way individuals manage health. According to the Information and Communication Technologies (ICT) sector and the World Health Organization (WHO), smart device usage has experienced significant growth over the past decade. 92% of young adults in the United States own a smart device, while only 42% of older adults possess one.¹ In China, there has also been a rapid expansion in smart device use. Smartphone shipments reached 343 million units, marking a year-on-year increase of 15.9% in 2021. Additionally, smart wearable devices nearly reached 140 million units, experiencing a substantial growth rate of 25.4%.¹

Smart devices play an increasingly critical role in health management. Older adults often face challenges in using smart devices due to the digital divide. Despite approximately 90% of the population having access to the Internet, the situation differs among older adults. Only around 60% of older adults reported using the Internet, and less than half of them utilized it to manage their health.^1-3 On the one hand, older adults often face challenges in matching their capabilities to use the ever-changing smart devices. On the other hand, information sharing, online healthcare services, and smart health devices expanded access to valuable health resources. Unfortunately, these benefits are not enjoyed by older adults due to limited availability.

Smart devices are valuable in addressing potential physical, psychological, and social adaptation issues.^4-6 Firstly, smart devices contribute to physical health. They empower older adults to receive immediate feedback on heart rate and blood pressure. This real-time information enables them to detect potential health risks, and take necessary preventive measures.^7-9 Furthermore, self-checks and health reminders provided by smart devices contribute to overall improvements in health.¹⁰ Smart devices enable older adults to receive potential diagnoses at home, which reduces the checkup cost and minimizes health risks.^11,12 Moreover, individuals can monitor various health indicators in real time, including sleep patterns, calorie expenditure, heart rate, and sports mileage.¹³

Secondly, smart devices also have a positive effect on mental health.^4,14 Goumopoulos et al utilized a smart device to collect heart rate data and accurately predicted mental health status.¹⁵ Likewise, Sano et al leveraged smart devices to gather information on sleep quality and stress levels to prevent mental health risks.¹⁶ Yen revealed a positive effect of smartwatches on stress management.¹⁷ However, the higher smart device involvement was also associated with higher levels of depression and stress.⁴ Thirdly, smart devices also have a positive effect on social adaptation. Social support was increased after using smart devices.¹⁸ Some studies supported the positive health outcomes of smart devices, such as enhancing social support and reducing social isolation and loneliness.¹⁹

Smart devices contribute to the health of older adults through healthcare utilization, chronic disease prevention, social activities participation, and social support. (1) Healthcare utilization. Smart devices detect early signs of diseases and provide treatment recommendations by monitoring changes in all parts of the human body.⁶ Once there is an abnormal signal, smart devices remind patients to take medicine and seek medical service timely.²⁰ (2) Chronic disease prevention and control. Smart devices not only help the early detection of chronic diseases⁸ but also provide convenient and efficient diagnostic services through data sharing and remote communication during the treatment phase. (3) Social activities participation. On the one hand, smart devices broaden the channels to obtain social activity information; on the other hand, they expand the space to participate in activities, allowing us to participate in social activities free from spatial limitations. (4) Social support. Smart devices (such as smartphones, tablets, etc.) enable us to stay in touch with family and friends more conveniently. Social connectedness between smart device users and devices can improve social support.¹⁸ Figure 1 shows the channels of smart devices on health status:

Figure 1.

Mechanism of smart devices on health status

This study aims to examine the effect of smart devices on the health status of older adults by employing the China Longitudinal Aging Social Survey (CLASS). We investigated the health effects of three types of devices, including smart wristbands, smart assistants, and audiobooks. The health outcomes were measured by multidimensional indicators, such as self-reported health status, depression symptoms, and social adaptation. A series of robustness checks was conducted, including using the IV method and trimming interfering samples. Furthermore, the study delved into the underlying mechanisms and potential effect variations among subgroups.

Methods

Study Design and Setting

This study employed a longitudinal panel design using data from the 2018 and 2020 waves of the China Longitudinal Aging Social Survey (CLASS). The CLASS is a nationally representative survey of community-dwelling adults aged 60 years and older in China, conducted by Renmin University of China. It utilizes a stratified, multistage probability sampling design, covering 462 communities (villages, neighborhoods) across 28 provinces, municipalities, and autonomous regions. The first wave of data used in this study was collected in 2018, and the second wave in 2020. The panel consists of only two time points (2018 and 2020), which restricts the ability to examine dynamic or long-term health effects. This study was conducted and reported in accordance with the STROBE Statement for cohort studies.²¹ A completed EQUATOR checklist is provided as a supplemental file.

Participants

The CLASS target population includes all Chinese citizens aged 60 years or older. Eligibility criteria for this longitudinal analysis required participants to have completed interviews in both the 2018 and 2020 waves. From the initial sample, we excluded respondents who (1) had missing data on key variables of interest (smart device use and health outcomes) in either wave, (2) were identified as outliers on continuous variables, and (3) had incomplete information on any of the selected covariates. This process yielded a final balanced panel dataset comprising 11,608 observations from 5,804 unique respondents, each measured at two time points.

Study Size

The study size was determined by the available data from the two waves of the CLASS survey. After applying the exclusion criteria for missing data and outliers to ensure a clean dataset for longitudinal analysis, the final balanced sample included 5,804 respondents, providing 11,608 person-wave observations. This sample size provides sufficient statistical power to detect meaningful associations in our models.

Missing Data and Attrition

Missing data were addressed through listwise deletion. Cases with missing information on any of the variables included in the analytical models were excluded from the final dataset. This approach was deemed acceptable given the relatively low proportion of missingness after merging the two waves and the large initial sample size. Loss to follow-up between 2018 and 2020 was inherent in the construction of the balanced panel, as only respondents present in both waves were retained.

Variables

Outcome Variables

Health status, the primary outcome of this study, was measured using three distinct indicators:

(1) Self-rated health (SRH) was measured by the question: “How do you think of your current physical health status?” Responses were given on a five-point Likert scale ranging from 1 (“unhealthy”) to 5 (“very healthy”), with higher scores indicating better perceived health.

(2) Depressive symptoms were evaluated using the six-item version of the Center for Epidemiologic Studies Depression Scale (CES-D-6). Respondents rated each item on a five-point Likert scale from 1 to 5 according to the severity of their symptoms over the past week: a. I felt that I was just as good as other people.b. I felt lonely. c.I felt sad. d.I felt hopeful about the future. e. I could not get going/I did not feel like eating. f.My sleep was restless. The total score ranges from 6 to 30, with higher scores reflecting greater severity of depressive symptoms.

(3) Social Adaptation: Measured using a composite index based on eight items about social life and interpersonal functioning. Respondents were asked to rate each item on a five-point scale ranging from 1 to 5, indicating the extent to which the statement applied to them. The summed score ranges from 8 to 40, with higher values denoting better social adaptation. Specific items include: a.if given the opportunity, I am willing to participate in certain tasks of the community; b.I often want to do something for society; c.I enjoy learning now; d.I think I am still a useful person to society; e.Social changes are happening too quickly, and it’s difficult for me to adapt to these changes; f.Now, more and more viewpoints are making it difficult for me to accept; g.More and more new social policies nowadays make it difficult for me to accept; h.The current social changes are increasingly unfavorable for older adults.

Main Independent Variables

The primary independent variable was the use of three common smart devices: smart wristbands, smart assistants, and audiobooks. These devices were selected as they represent accessible and prevalent technologies for older adults in China. For each wave, respondents were asked whether they used each device. Three separate dummy variables were created for each device (1 = use, 0 = no use). These three types of devices were selected not only for their prevalence among Chinese older adults but also for their conceptual alignment with distinct mechanisms of health production. Wristbands (e.g., fitness trackers) are primarily designed for continuous, objective collection of physiological and behavioral data. This function directly reflects the “information collection/processing/interaction” node in the framework. By providing real-time monitoring and long-term data trends, wristbands deliver personalized feedback and encourage users to increase physical activity and improve sleep. Additionally, data recorded by wristbands, such as blood pressure trends or atrial fibrillation notifications, can serve as evidence during medical consultations, facilitating more accurate diagnosis and treatment decisions. This enhances the quality and efficiency of healthcare utilization, reducing unnecessary visits or delayed care.

Smart assistants rely on natural language interaction, with a focus on the “interaction” aspect of the information node, extending into social dimensions. Smart assistants can support users through medication reminders, emergency assistance, and emotional companionship. These functions enhance both instrumental and emotional dimensions of social support. By simplifying access to information, smart assistants reduce barriers to social engagement, thereby increasing the frequency and ease of social activity participation and mitigating the negative health effects of social isolation.

Although audiobooks differ from conventional health devices, their role in the health production mechanism can be explained through the pathways of information interaction and cognitive stimulation. Audiobooks represent a non-visual mode of information interaction. For individuals with visual impairments, reading difficulties, or those engaged in multitasking, they offer continuous access to knowledge, cultural content, and health-related information. Shared audiobook content can serve as a conversational topic, promoting interpersonal communication and social participation. Additionally, listening to audiobooks during activities such as walking or exercising encourages outdoor activity and social contact.

Covariates

To account for potential confounding, we included a range of sociodemographic factors known to be associated with both technology adoption and health in older adults. These covariates were measured at each wave and included: age (continuous, in years), gender (female vs. male), education attainment (categorized into seven levels from illiteracy to bachelor’s degree), marital status (married vs. other), living area (rural vs. urban), individual annual income (continuous, logarithmically transformed), number of cohabitants (continuous), and pension status (having a pension vs. not).

Quantitative Variables

Continuous variables were handled as follows. Age and number of cohabitants were left in their original continuous form. Individual income exhibited a skewed distribution; therefore, it was logarithmically transformed for the analyses to improve model fit and interpretability. All other quantitative variables (SRH, CES-D-6 score, Social Adaptation score) were treated as continuous outcomes in the linear fixed effects models.

Bias

To mitigate potential bias, several strategies were employed. First, the use of a longitudinal panel with individual fixed effects helps control for time-invariant unobserved heterogeneity that could confound the association. Second, we controlled for a comprehensive set of time-varying sociodemographic covariates to address confounding by observable characteristics. Third, we explicitly examined and addressed missing data to reduce potential selection bias. Despite these efforts, we acknowledge that the findings indicate associations rather than definitive causal effects, as bias from time-varying unmeasured confounders or reverse causality cannot be entirely ruled out.

Statistical Methods

Main Analysis

To estimate the association between smart device use and health outcomes, we employed a Two-Way Fixed Effects (TWFE) linear regression model. This model was specified as:

{H e a l t h}_{i t} = α + β {S m a r t d e v i c e}_{i t} + γ X_{i t} + μ_{i} + λ_{t} + ε_{i t}

(1)

where

{H e a l t h}_{i t}

represents the health status of respondent

i

at time

t

{S m a r t d e v i c e}_{i t}

is a dummy variable indicating the use of a specific smart device, and

X_{i t}

is a vector of time-varying sociodemographic covariates. The terms

μ_{i}

and

λ_{t}

represent individual and year fixed effects, respectively, and

ε_{i t}

is the idiosyncratic error term. The inclusion of individual fixed effects

μ_{i}

controls for all time-invariant confounding, whether measured or unmeasured, by focusing on within-individual changes in smart device use and health over time. The year fixed effects

λ_{t}

control for secular trends affecting the entire sample. Standard errors were clustered at the individual level to account for repeated observations. Separate models were fitted for each of the three smart devices and each of the three health outcomes.

Sensitivity Analyses

To address potential endogeneity concerns and assess the robustness of our findings, we conducted a series of sensitivity analyses. First, we employed an instrumental variable (IV) approach to mitigate bias from unobserved confounding and reverse causality. Internet use was selected as the instrument for smart wristband adoption, based on its strong correlation with device usage while plausibly satisfying the exclusion restriction by not directly influencing individual health outcomes. This model was specified as:

{S m a r t d e v i c e}_{i t} = α_{0} + β_{1} {I n t e r n e t u s e}_{i t} + γ_{2} X_{i t} + μ_{i} + λ_{t} + ε_{i t}

(2)

{H e a l t h}_{i t} = α_{1} + β_{2} {S m a r t d e v i c e}_{i t} + γ_{3} X_{i t} + μ_{i} + λ_{t} + ε_{i t}

(3)

To further mitigate bias from self-selection, particularly given that individuals with chronic conditions may be more likely to adopt health-monitoring devices, we conducted an additional robustness check using a trimmed sample that excluded older adults with chronic diseases. All statistical analyses were performed using Stata. 18, with statistical significance set at p < 0.05.

Results

Demographic Characteristics

Table 1 presents the descriptive statistics and mean-difference tests between participants who used smart wristbands and those who did not. Among the total sample, 9.44% (921 out of 11,608) reported using a smart wristband. Significant differences are observed across all outcome and control variables between the two groups.

Table 1.

Summary

Variables	Non-use		Smart devices use		Mean-diff	T
Variables	N	Mean	N	Mean	Mean-diff	T
SRH	10687	3.390	921	3.611	-0.221***	-7.599
Depressive symptoms	10687	15.751	921	14.779	0.972***	9.431
Social adaptation	10687	24.423	921	25.836	-1.413***	-9.913
age	10687	71.185	921	69.799	1.386***	5.764
gender	10687	0.486	921	0.529	-0.043**	-2.491
edu	10687	3.178	921	4.002	-0.824***	-17.940
marry	10687	0.724	921	0.807	-0.082***	-5.422
livingarea	10687	0.351	921	0.093	0.257***	16.013
lnincome	10687	8.498	921	9.311	-0.814***	-15.153
company	10687	2.601	921	2.853	-0.253***	-5.966
pension	10687	0.848	921	0.971	-0.123***	-10.265

Note: SRH: Self-rated health; ^* p < 0.1, ^** p < 0.05, ^*** p < 0.01; Since the descriptive statistics for the three types of smart devices would take up too much space, only the results for the smart wristbands are presented here.

In terms of health outcomes, smart wristband users reported better self-rated health (mean SRH = 3.611) compared to non-users (mean SRH = 3.390), with a statistically significant difference of 0.221 (p<0.01). Conversely, users exhibited lower depressive symptoms (mean = 14.779) than non-users (mean = 15.751), a difference of -0.972 (p<0.01). Their social adaptation score was notably higher (mean = 25.836 for users vs. 24.423 for non-users, diff = -1.413, p<0.01).

Demographically, smart wristband users were younger (mean age = 69.80) than non-users (71.19). They also had higher educational attainment (mean edu = 4.002 vs. 3.178), were more likely to be married (proportion = 0.807 vs. 0.724), and had substantially higher log-income (mean = 9.311 vs. 8.498). Additionally, users showed a higher proportion having a pension (0.971 vs. 0.848). All differences in these characteristics are statistically significant at the 1% level, suggesting substantial socio-demographic disparities between adopters and non-adopters of smart wristbands.

Benchmark Regression Results

Table 2 presents the benchmark estimation results regarding the relationship between different smart devices and health outcomes. The use of a smart wristband shows a significant positive association with self-rated health and social adaptation, and a significant negative association with depressive symptoms. The use of a smart assistant is associated with a significant increase in SRH, but does not show a significant relationship with social adaptation. The use of audiobooks demonstrates a positive association with social adaptation and a negative association with depressive symptoms, while its link to self-rated health is not significant.

Table 2.

Benchmark Results

Variables	(1)	(2)	(3)	(4)	(5)	(6)	(7)	(8)	(9)
Variables	SRH	Depressive symptoms	Social adaptation	SRH	Depressive symptoms	Social adaptation	SRH	Depressive symptoms	Social adaptation
Smart wristband	0.271***	-0.775***	0.334*
	(0.038)	(0.169)	(0.187)
Smart assistant				0.118**	0.712***	0.125
				(0.054)	(0.232)	(0.266)
Audiobook							0.033	-0.807***	1.485***
							(0.059)	(0.274)	(0.291)
Constant	3.410*	10.625	-0.510	3.423*	7.599	0.105	3.426*	9.366	0.392
Ind-Fe	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Year-Fe	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes	Yes
Observations	6,578	5,572	5,852	6,578	5,572	5,852	6,578	5,572	5,852
R-squared	0.766	0.705	0.805	0.763	0.704	0.805	0.763	0.704	0.806

Note: SRH: Self-rated health; ^* p < 0.1, ^** p < 0.05, ^*** p < 0.01.

All specifications include individual fixed effects, year fixed effects, and a comprehensive set of socio-demographic controls. The results suggest heterogeneous effects across device types: smart wristbands are primarily linked to comprehensive health benefits, and smart assistants and audiobooks are associated with social functioning and psychological well-being.

Robustness Checks

Table 3 reports the results of robustness checks conducted to address potential endogeneity and sample selection bias. The first three columns present estimates using an instrumental variable (IV) approach, with internet use serving as the instrument. This variable is chosen because it is strongly correlated with the adoption of smart wristbands, while plausibly satisfying the exclusion restriction by not directly influencing individual health outcomes.

Table 3.

Robustness Check by IV Estimates and Trimming Samples

Variables	(1)IV	(2)IV	(3)IV	(4)Trimming samples
Variables	SRH	Depressive symptoms	Social adaptation	SRH
Smart wristband	1.052***	-7.361***	9.231***	0.650***
	(0.158)	(0.687)	(0.945)	(0.137)
Constant	4.609***	13.603***	26.809***	4.206***
	(0.105)	(0.421)	(0.581)	(0.703)
Ind-Fe	Yes	Yes	Yes	Yes
Year-Fe	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes
Observations	14,022	12,429	12,617	730

The IV estimates confirm a significant positive relationship between smart wristband use and health outcomes, with substantially larger coefficients than those in the benchmark regression. Specifically, smart wristband use is associated with a 1.052-point increase in self-rated health, a 7.361-point reduction in depressive symptoms, and a 9.231-point improvement in social adaptation. The increased magnitude of these coefficients suggests that the ordinary least squares estimates in the baseline model may have been attenuated by measurement error or unobserved confounding.

To further mitigate bias from self-selection, particularly given that individuals with chronic conditions may be more likely to adopt health-monitoring devices, the fourth column presents results from a trimmed sample that excludes older adults with chronic diseases. Even within this more homogeneous subgroup, smart wristband use remains positively associated with better health outcomes, as indicated by a 0.650-point improvement in self-rated health. This finding reinforces the robustness of the core result, suggesting that the observed health benefits are not solely driven by pre-existing health conditions or selective adoption among less healthy individuals.

Mechanism Analyses

In this section, we aim to shed light on potential mechanisms and investigate the drivers of the positive association between smart devices and health.

Table 4 examines the potential mechanism through which smart wristbands influence health outcomes, focusing on healthcare utilization behavior. The first column estimates the relationship between smart wristband adoption and self-treatment behavior. Results indicate a statistically significant positive association, suggesting that device users are more likely to engage in self-care or initial health management before seeking formal medical services.

Table 4.

Mechanism: Healthcare Utilization

Variables	(1)	(2)
Variables	Self-treat	Smart wristband
Smart wristband	0.034*
	(0.019)
serious_illness		0.060**
		(0.030)
Constant	-1.625*	-0.041
	(0.963)	(0.910)
Ind-Fe	Yes	Yes
Year-Fe	Yes	Yes
Controls	Yes	Yes
Observations	6,590	6,522
R-squared	0.705	0.628

The second column evaluates whether experiencing a serious illness influences the likelihood of adopting a smart wristband. The estimated coefficient is positive, implying that prior severe health shocks are positively associated with smart device adoption among older adults. Taken together, these results support a preventive health mechanism. The positive link with self-treatment indicates that smart wristbands may enhance users’ awareness of their health status and encourage proactive management. Thus, smart wristbands likely contribute to health improvement by facilitating early monitoring, promoting regular self-care, and fostering sustained health management practices, rather than merely serving as a response to existing severe conditions.

Table 5 explores the mechanism through which smart wristbands influence health by examining the relationship between smart wristbands use and the diagnosis of chronic diseases. The results indicate that the use of smart wristbands is significantly associated with an increase in the total number of diagnosed chronic conditions, suggesting that device users are more likely to be aware of multiple health issues.

Table 5.

Mechanism: Disease Diagnosis

Variables	(1)	(2)	(3)	(4)
Variables	Total diseases	Hypertension	Heart disease	Diabetes
Smart wristband	0.249***	0.126***	0.026	0.043**
	(0.071)	(0.024)	(0.016)	(0.018)
Constant	2.749	2.204*	-0.002	0.656
	(3.711)	(1.238)	(0.830)	(0.927)
Ind-Fe	Yes	Yes	Yes	Yes
Year-Fe	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes
Observations	6,590	6,590	6,590	6,590
R-squared	0.761	0.727	0.791	0.710

Specifically, smart wristband adoption shows a strong positive correlation with the diagnosis of hypertension and a significant association with diabetes diagnosis. These findings collectively support a diagnosis facilitation mechanism: rather than causing illness, smart wristbands, through features such as continuous physiological monitoring (e.g., heart rate, activity tracking) and health data feedback, likely enhance users’ awareness of underlying health abnormalities. This increased awareness can prompt earlier and more frequent clinical consultations, thereby improving the detection rates of conditions such as hypertension and diabetes.

Table 6 examines the mechanism of social activities participation to explain the positive relationship between smart wristbands and health outcomes. The results demonstrate that smart wristband use is significantly and positively associated with increased participation across all four measured social activities: watching television, singing, playing mahjong, and square dancing.

Table 6.

Mechanism: Social Activities Participation

Variables	(1)	(2)	(3)	(4)
Variables	TV	Sing	Mahjong	Squaredance
Smart wristband	0.224**	0.480***	0.605***	0.501***
	(0.098)	(0.143)	(0.148)	(0.146)
Constant	21.969***	15.594**	17.950**	18.875**
	(5.095)	(7.422)	(7.672)	(7.578)
Ind-Fe	Yes	Yes	Yes	Yes
Year-Fe	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes
Observations	6,590	6,590	6,590	6,590
R-squared	0.689	0.554	0.555	0.555

These findings indicate that smart wristbands do not merely function as passive health monitors but actively promote social engagement among older adults. The devices may facilitate social participation through features such as activity reminders, progress sharing, or by fostering a sense of community among users. Notably, the coefficients are largest for group-based and interactive activities such as mahjong and square dancing, suggesting a stronger effect on activities involving social interaction and physical coordination.

Table 7 investigates the mechanism of social support, presenting the relationship between smart wristband use and three dimensions of social interaction. Smart wristband adoption is associated with a significant increase in the frequency of meeting friends and a substantial increase in the propensity to receive help from friends.

Table 7.

Mechanism: Social Support

Variables	(1)	(2)	(3)
Variables	Meet friends	Talk with friends	Help from friends
Smart wristband	0.250***	-0.076	0.354***
	(0.054)	(0.050)	(0.056)
Constant	-4.354	-5.826**	-3.687
	(2.831)	(2.587)	(2.891)
Ind-Fe	Yes	Yes	Yes
Year-Fe	Yes	Yes	Yes
Controls	Yes	Yes	Yes
Observations	6,590	6,590	6,590
R-squared	0.763	0.753	0.738

This pattern suggests that smart wristbands primarily facilitate tangible and instrumental forms of social support, such as face-to-face meetings and practical assistance. The devices may encourage more organized social gatherings or cooperative activities, thereby strengthening real-world social bonds and reciprocal support networks. Overall, the positive effects on concrete supportive behaviors align with the observed improvements in social adaptation and mental health. By enhancing direct social engagement and mutual aid, smart wristbands may help users build stronger, more reliable support systems, which are crucial for mitigating stress, fostering a sense of belonging, and ultimately contributing to better psychological and social well-being.

Subgroup Analyses

In this section, we conducted subgroup analyses to examine whether the relationship varies across different population characteristics, including gender, marital status, living area, age, and education level. Interaction terms were used to estimate heterogeneous effects, with results for self-reported health (SRH) presented in Table 8, depression in Table 9, and social adaptation in Table 10.

Table 8.

Subgroup Analyses: The Effect of Smart Wristband on Self-Reported Health Status

Variables	(1)	(2)	(3)	(4)	(5)
Variables	SRH	SRH	SRH	SRH	SRH
Smart wristband #gender	-0.063
	(0.074)
Smart wristband #marry		-0.177**
		(0.089)
Smart wristband #livingarea			0.324**
			(0.141)
Smart wristband #dummyage				0.033
				(0.074)
Smart wristband #dummyedu					-0.094
					(0.132)
Constant	3.426*	3.300*	3.500*	3.657*	2.863
	(1.982)	(1.981)	(1.980)	(1.990)	(2.026)
Ind-Fe	Yes	Yes	Yes	Yes	Yes
Year-Fe	Yes	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes	Yes
Observations	6,578	6,578	6,578	6,578	6,578
R-squared	0.766	0.767	0.767	0.767	0.767

Note: dummyage: Dummy variable (1=older group, i.e., age above sample mean; 0=younger group).

dummyedu: Dummy variable (1=higher-educated group, i.e., years of education above the sample mean; 0=lower-educated group).

Table 9.

Subgroup Analyses: The Effect of Smart Wristband on Depression

Variables	(1)	(2)	(3)	(4)	(5)
Variables	Depressive symptoms	Depressive symptoms	Depressive symptoms	Depressive symptoms	Depressive symptoms
Smart wristband #gender	-0.026
	(0.329)
Smart wristband #marry		0.705*
		(0.395)
Smart wristband #livingarea			-2.127***
			(0.644)
Smart wristband #dummyage				0.289
				(0.328)
Smart wristband #dummyedu					-0.953
					(0.600)
Constant	10.663	11.178	9.721	9.616	6.189
	(11.535)	(11.522)	(11.505)	(11.564)	(11.639)
Ind-Fe	Yes	Yes	Yes	Yes	Yes
Year-Fe	Yes	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes	Yes
Observations	5,572	5,572	5,572	5,572	5,572
R-squared	0.705	0.706	0.706	0.706	0.706

Note: dummyage: Dummy variable (1=older group, i.e., age above sample mean; 0=younger group).

dummyedu: Dummy variable (1=higher-educated group, i.e., years of education above the sample mean; 0=lower-educated group).

Table 10.

Subgroup Analyses: The Effect of Smart Wristband on Social Adaptation

Variables	(1)	(2)	(3)	(4)	(5)
Variables	Social adaptation	Social adaptation	Social adaptation	Social adaptation	Social adaptation
Smart wristband #gender	0.438
	(0.362)
Smart wristband #marry		0.309
		(0.438)
Smart wristband #livingarea			3.386***
			(0.716)
Smart wristband #dummyage				-0.207
				(0.361)
Smart wristband #dummyedu					2.354***
					(0.651)
Constant	-0.781	-0.618	-1.862	0.185	-1.000
	(11.535)	(11.536)	(11.496)	(11.577)	(11.521)
Ind-Fe	Yes	Yes	Yes	Yes	Yes
Year-Fe	Yes	Yes	Yes	Yes	Yes
Controls	Yes	Yes	Yes	Yes	Yes
Observations	5,852	5,852	5,852	5,852	5,852
R-squared	0.805	0.805	0.806	0.805	0.806

Note: dummyage: Dummy variable (1=older group, i.e., age above sample mean; 0=younger group).

dummyedu: Dummy variable (1=higher-educated group, i.e., years of education above the sample mean; 0=lower-educated group).

The analysis indicates significant heterogeneity in health outcomes. Regarding SRH (Table 8), married individuals show a negative association with self-reported health among married users of smart wristbands. Similarly, those living in rural areas showed a significant positive interaction, indicating better SRH in this subgroup. No significant moderating effects were found for gender, age, or education on SRH.

For depression (Table 9), living in rural areas was associated with a substantial reduction in depressive symptoms, while being married showed a significant increase. Gender, age and education did not yield significant interactions.

In terms of social adaptation (Table 10), living in rural areas and higher education levels were associated with significantly improved adaptation. Other subgroups did not show statistically significant effects.

Overall, these findings highlight that the relationship between smart wristbands and health is highly heterogeneous across different demographic groups. The devices appear particularly beneficial for rural residents in reducing depression and enhancing social adaptation. Such variation underscores the importance of tailored interventions when implementing digital health technologies.

Medical Expenditures Decrease

Table 11 presents the relationship between three types of smart devices (smart wristbands, smart assistants, and audiobooks) and medical expenditures. The results indicate that the use of smart wristbands is associated with a statistically significant reduction in medical spending. Similarly, audiobook usage shows a significant negative association with expenditures, estimated at –0.089. In contrast, while smart assistants also exhibit a negative coefficient, the effect is not statistically significant. These findings suggest that smart wristbands and audiobooks may effectively help lower medical costs. The overall evidence supports the potential of certain digital health tools in reducing healthcare expenditures.

Table 11.

The Effect of Smart Devices on Medical Expenditures

Variables	(1)	(2)	(3)
Variables	lnexp	lnexp	lnexp
Smart wristband	-0.057***
	(0.021)
Smart assistant		-0.022
		(0.028)
audiobook			-0.089***
			(0.032)
Constant	-0.047	-0.125	-0.186
	(1.542)	(1.544)	(1.542)
Ind-Fe	Yes	Yes	Yes
Year-Fe	Yes	Yes	Yes
Controls	Yes	Yes	Yes
Observations	5,698	5,698	5,698
R-squared	0.770	0.770	0.770

Note: lnexp: Natural log of medical expenditure.

Determinants of Smart Device Use

Based on the regression results presented in Table 12, this section analyzes the socio-economic determinants of adopting three distinct types of smart devices among older adults. The findings reveal a nuanced pattern of stratified adoption, significantly influenced by education, residence, and household structure.

Table 12.

Determinants of Smart Device Use

Variables	(1)	(2)	(3)
Variables	Smart wristband	Smart assistant	Audiobook
age	0.003	0.003	0.007
	(0.013)	(0.009)	(0.008)
gender	-0.079	0.139	0.003
	(0.134)	(0.094)	(0.087)
edu	-0.021	-0.049*	-0.048*
	(0.042)	(0.030)	(0.027)
marry	-0.073	-0.027	-0.007
	(0.045)	(0.032)	(0.029)
livingarea	0.011	-0.053***	0.048***
	(0.020)	(0.014)	(0.013)
lnincome	0.003	0.002	0.003
	(0.005)	(0.004)	(0.003)
company	0.023***	0.017***	0.004
	(0.005)	(0.004)	(0.003)
pension	-0.045	-0.035	-0.001
	(0.033)	(0.023)	(0.021)
Constant	0.014	-0.068	-0.316
	(0.907)	(0.640)	(0.588)
Ind-Fe	Yes	Yes	Yes
Year-Fe	Yes	Yes	Yes
Observations	6,590	6,590	6,590
R-squared	0.627	0.715	0.720

The coefficient estimates indicate a significantly negative relationship between educational attainment and the use of smart assistants and audiobooks. This suggests that individuals with higher education levels are less likely to adopt these specific devices. A plausible interpretation is that more highly educated seniors may rely on more versatile digital tools (e.g., smartphones, tablets, or computers) for information and entertainment, perceiving lower marginal utility from dedicated devices like all-in-one machines or audiobook platforms. The effect of education on smart wristband adoption is not statistically significant.

A stark urban-rural digital divide is evident. Residence in a rural area is strongly and negatively associated with the use of a smart assistant but strongly and positively associated with the use of audiobooks. This divergence highlights differential adoption patterns based on geographic context: rural older adults are significantly less likely to use complex, integrated devices, potentially due to constraints in digital infrastructure, literacy, or support networks. Conversely, they show a higher propensity to use audiobooks, a device format that is often less technically demanding and provides auditory companionship, making it a more accessible technology in rural settings.

Household size exhibits a strong positive correlation with the adoption of both smart wristbands and smart assistants. This result supports the family support or household digital ecology hypothesis. A larger number of co-residents, likely including younger family members, facilitates device adoption through mechanisms such as financial co-purchase, technical assistance, instructional support, and the creation of a shared environment conducive to technology use. The effect on audiobook adoption is positive but not statistically significant, suggesting its use may be more individualized and less dependent on household dynamics.

Discussion

This study provides empirical evidence on the positive role of smart devices in promoting the health of older adults in China. Based on a nationally representative panel dataset and a two-way fixed effects modeling strategy, we find that the use of smart wristbands is significantly associated with better self-rated health, reduced depressive symptoms, and improved social adaptation. Smart assistants and audiobooks are also associated with better social adaptation. Importantly, as emphasized throughout this paper, these findings indicate positive associations rather than definitive causal effects.

Based on existing literature and theoretical frameworks, we propose several plausible pathways through which these associations may operate, including healthcare utilization, social activities, social support, and management of chronic conditions. The findings support positive associations between smart device use and increased healthcare utilization, greater participation in social activities, strengthened social support, and earlier diagnosis and management of chronic conditions. It is important to emphasize that while these mechanisms are theoretically grounded, they represent potential explanations rather than empirically tested causality in the current study. Furthermore, these associations are heterogeneous across subgroups, being more pronounced among certain demographic profiles such as rural residents and those with higher education. Additionally, we find that smart device adoption is associated with a significant reduction in medical expenditures, suggesting a potential economic benefit that warrants further investigation.

To enhance the interpretability of our findings, we translated the regression coefficients into percentage changes relative to the sample means of health outcomes. The estimates indicate that smart wristband use was associated with an 8.1% improvement in self-rated health, a 4.9% reduction in depressive symptoms, and a 1.4% increase in social adaptation.² These effect sizes are not trivial when considered in the context of population health. The 8.1% gain in self-rated health is comparable to the improvement observed after participating in structured physical activity programs for six months. The 4.9% decrease in depressive symptoms approaches the minimal clinically important difference reported for brief psychosocial interventions in community-dwelling older adults. Although the 1.4% increase in social adaptation is modest, it reflects a meaningful enhancement in social engagement that could mitigate social isolation.

From a policy perspective, the scalability and low cost of smart wristbands make these incremental benefits particularly valuable. If the observed associations are causal, widespread adoption of such devices could yield substantial public health gains. For instance, a population-level reduction in depressive symptoms of 4.9% would translate into millions fewer individuals experiencing clinically significant distress, potentially reducing the demand for mental health services and lowering associated societal costs. Similarly, improvements in self-rated health and social adaptation may contribute to longer independent living and reduced healthcare utilization among aging populations. Thus, even modest individual-level effects can accumulate into significant societal benefits.

Compared to prior studies,^22-24 this study contributes to the existing literature in several incremental ways. First, while previous research has largely focused on single devices or aggregated measures of technology use, we extend this topic by simultaneously examining three distinct categories of smart devices, smart wristbands for health monitoring, smart assistants for information processing, and audiobooks for entertainment, and their respective associations with multidimensional health outcomes among older adults. This approach allows for a more nuanced understanding of how different types of digital tools may relate to different domains of health.

Second, building on prior studies that have documented positive associations between smart devices and health using cross-sectional or simple regression designs, we add to the evidence base by employing a two-way fixed effects panel model combined with multiple robustness checks, including instrumental variable estimation and sample trimming. These methodological enhancements help address concerns about unobserved heterogeneity and endogeneity, thereby providing more credible estimates of the associations.

Third, we extend the geographical and contextual scope of the literature by focusing on a large, nationally representative sample of older adults in China, a rapidly aging society where digital technology adoption is accelerating but empirical evidence remains limited. Drawing on this unique context, we integrate theoretical perspectives from health economics, social epidemiology, and gerontology to offer a comprehensive discussion of potential mechanisms linking smart device use to health, including healthcare utilization, social participation, social support, and chronic disease prevention. While these mechanisms are not empirically tested for causality in the current study, our theoretical synthesis provides a useful framework for future research to investigate mediating pathways in the Chinese context and beyond.

Collectively, these contributions offer an incremental yet meaningful advancement in understanding the relationship between smart device use and health among older adults, particularly in non-Western aging populations where such evidence is still emerging.

Consistent with prior research, our findings align with studies reporting positive associations between smart devices and health outcomes in China. However, we note important methodological distinctions. Wei and Guo’s study,²⁵ which included 1,110 smartphone users and employed a multivariate ordered logit model without addressing endogeneity, may have produced biased estimates due to omitted variable concerns. More rigorous evidence comes from randomized controlled trials: Yen and colleagues designed an RCT to explore the effects of smart wearable devices on healthy lifestyles and quality of life, revealing significant improvements in exercise, self-actualization, and stress management among smartwatch users.¹⁷ Liang et al and Wang et al further documented associations between smartphone use and older adults’ health.^26,27 Conversely, Harwood et al found that higher smart-device involvement was associated with higher levels of depression and stress,⁴ highlighting that the direction of associations may depend on usage patterns, contexts, or populations.

Drawing on existing theoretical frameworks and empirical evidence from prior studies, we propose four potential channels through which smart devices may contribute to health promotion. It is important to reiterate that these mechanisms, while plausible and grounded in the literature, are not tested as causal pathways in our empirical analysis and should be interpreted as theoretical interpretations rather than empirically validated pathways.

Healthcare utilization

Prior research suggests that smart devices can detect early signs of diseases and provide treatment recommendations by monitoring abnormal changes.⁶ Kang et al argued that smart devices can collect real-time patient data and transfer information for assessment to healthcare providers.⁷ Extending this logic to our findings, one plausible interpretation is that the observed health associations may be partially mediated by improved healthcare access and utilization, though this hypothesis requires formal testing in future research.

Chronic Disease Diagnosis and Management

Smart devices provide fast, cost-effective, and reliable health monitoring services for older adults, potentially facilitating early detection of chronic conditions.⁸ For example, prior studies have documented applications in heart failure and asthma management.^9,28 Built-in sensors and dedicated health apps provide users with continuous, personalized health data, potentially reducing information asymmetry regarding one’s own health status. In the context of our findings, these features may contribute to the observed health associations by enabling earlier detection of abnormalities and prompting timely professional consultation, a mechanism that remains speculative.

Physical Activity and Health Behaviors

Previous research indicates that smart devices can promote exercise and stress management.^17,29 Studies have shown that using smart wristbands increases exercise motivation³⁰ and plays a crucial role in increasing physical activity participation through real-time activity monitoring.^31-33 Consistent with this literature, one plausible interpretation of our results is that the observed health associations may operate through more social and physical activities.

Social Support and Connectedness

Prior studies suggest that smart devices can improve social support for users,¹⁸ with some research reporting positive outcomes related to social support, isolation, and loneliness.¹⁹ Instant messaging apps, social media platforms, and video call functions allow older adults to maintain frequent contact with geographically dispersed family and friends. Online communities and interest-based groups facilitate the formation of new weak-tie connections, providing novel sources of informational and emotional support. These pathways are well-documented in the literature and offer a compelling explanation for the relationship between smart device use and health outcomes.

Medical Expenditures Decrease

Our finding that smart device use is associated with significant reductions in medical expenditures suggests potential economic benefits that merit further investigation. Drawing on health-economic frameworks, we theorize that this association may operate through several mechanisms. Smart devices may empower users with continuous health data, reducing information asymmetry and fostering greater health awareness, potentially leading to earlier, less costly interventions. For elderly populations with chronic conditions, smart devices may facilitate better disease management, reducing acute exacerbations and hospitalizations. Additionally, smart devices may reduce transaction costs associated with navigating the healthcare system through online appointment booking, telemedicine, and access to reliable health information.

Heterogeneous Associations

We observed notable variations in the associations between smart device use and health outcomes across population subgroups. Urban-rural disparities emerged as a contributing factor, consistent with prior research.³⁴ One theoretical interpretation is that in urban settings, older adults often have access to a wider array of alternative resources for health management and social connectivity, making smart devices a complementary tool. In contrast, in rural areas where traditional resources and infrastructure are relatively scarce, smart devices may serve as a primary substitute, providing critical access to health information, remote social connections, and telehealth services. Consequently, the marginal benefit derived from adopting the technology may be substantially higher for rural residents. Regarding marital status, previous studies documented a longevity advantage for the married,³⁵ but we found a higher likelihood of poor self-rated health among married older adults. Educational attainment showed a protective association with health,¹⁹ consistent with our findings.

While our empirical analysis focuses on older adults in China, we speculate that the core findings may hold relevance for broader contexts. The associations are expected to be most directly generalizable to other middle-income and developing countries undergoing similar dual transitions of digitalization and population aging (e.g., Brazil, India, or Thailand). In these settings, smart devices often serve as a primary, low-cost gateway to health information and social connectivity, potentially leapfrogging traditional infrastructure constraints. However, the extent to which our findings generalize to other populations remains an empirical question requiring cross-country comparative research.

Regarding other age groups, we speculate that the direction of the association may be similar, but the magnitude and dominant channels could differ. For younger and working-age populations, who are typically digital natives, the marginal association of device adoption with basic social connectivity might be smaller. However, the health production function may shift: for them, specific features like fitness tracking, mental wellness apps, and management of work-life balance stress via digital tools could become more salient pathways. Crucially, the negative externalities of excessive use might also be more pronounced among younger cohorts, suggesting a potential non-linear or inverted-U relationship. These speculations, while grounded in theoretical reasoning, require empirical validation in future studies targeting different age groups.

Conclusion

This study demonstrates that smart devices are positively associated with the health of older adults in China. Based on nationally representative panel data and a Two-Way Fixed Effects model, the findings reveal that using smart wristbands is associated with better self-rated health, fewer depressive symptoms, and better social adaptation. Smart assistants and audiobooks also indicate a positive association with social adaptation. The findings reveal a positive association between smart device use and health outcomes among older adults; however, these results should not be interpreted as causal effects. The health benefits are primarily realized through the following mechanisms: increased healthcare utilization, more participation in social activities, enhanced social support, and earlier diagnosis and timely management of chronic diseases. Notably, the relationships exhibit clear heterogeneity across different groups, with more pronounced outcomes among rural residents and those with higher education levels. Furthermore, the adoption of smart devices is associated with a significant reduction in medical expenditures, highlighting their economic value beyond health promotion. In summary, smart devices can serve as comprehensive management tools to address the health challenges of an aging population.

Limitations

This study has several limitations that should be acknowledged. First, it is important to note that the findings of this study indicate associations rather than definitive causal effects. And, the panel data consists of only two time points (2018 and 2020), which restricts the ability to examine dynamic or long-term health effects. Second, due to the constraints of the large-scale survey data used, it was not possible to investigate a more detailed range of smart device types. We have, however, made efforts to include and distinguish between three prevalent categories: smart wristbands, smart assistants, and audiobooks. Third, the health outcomes examined in this paper are primarily based on the World Health Organization’s (WHO) multidimensional definition of health—encompassing physical, mental, and social well-being—and thus do not extend to other important dimensions such as cognitive function and life satisfaction. These areas represent valuable directions for future research to further explore the holistic impact of smart devices on aging populations. Future research should collect more detailed data on device usage, employ more rigorous research designs to strengthen causal inference, and incorporate dimensions such as cognitive function and life satisfaction into the health outcome framework for further investigation.

Supplemental Material

Supplemental Material - Smart Devices and Better Health: Positive Associations Between Smart Devices and Health Among Older Adults

Supplemental Material for Smart Devices and Better Health: Positive Associations Between Smart Devices and Health Among Older Adults by Yuanyang Wu, Shuyu Dong, Huimin Zhou, Xiaodong Li and Conghui Ma in Inquiry: The Journal of Health Care Organization, Provision, and Financing.

Footnotes

Acknowledgments

We would like to express our gratitude to the China Longitudinal Aging Social Survey (CLASS) research team and field team for their contributions to data collection. We would also like to thank all the interviewees who voluntarily participated in the CLASS.

ORCID iD

Yuanyang Wu

Ethical Considerations

The China Longitudinal Aging Social Survey (CLASS) was conducted in accordance with the ethical standards outlined in the Declaration of Helsinki. As a sociological research project focused on the socio-economic status of older adults, the CLASS survey was specifically designed to avoid any adverse effects on respondents’ mental health and did not involve human experimentation as defined by the Declaration. The survey protocol was approved by the academic committee of the School of Population and Health at Renmin University of China. The survey was conducted within the legal framework governed by the Statistics Law of the People’s Republic of China (Chapter I, Article 9).

Consent to Participate

All participants provided informed consent prior to participation. Specifically, verbal informed consent was obtained from all individual participants included in the survey. Additionally, interviewers documented detailed information regarding the informed consent process, including whether participants agreed to participate, the time of consent, and, where applicable, the reasons for refusal. All information collected was used exclusively for sociological research purposes, with strict measures implemented to ensure confidentiality and anonymity. The details of the informed consent process are maintained by the Institute of Gerontology at Renmin University of China and the National Survey Research Center.

Author Contributions

Y.W. was a prominent contributor to the manuscript. Y.W. conceptualized and designed the study. Y.W. and H.Z. contributed to the literature search, figures, study design, data analyses, data curation, data interpretation, and writing–the original draft. S.D. and X.L. contributed to the methodology and writing review & editing. C.M. and Y.W. supervised the whole study and reviewed the manuscript draft. Y.W., S.D., H.Z., and C.M. proofread the manuscript. All authors reviewed the final version of the manuscript.

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the National Natural Science Foundation of China (grant number: 72274066).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Data Availability Statement

The datasets analyzed during the current study are available from the China Longitudinal Aging Social Survey (CLASS) repository upon reasonable request and successful application through the Institute of Gerontology, Renmin University of China at . Researchers interested in accessing the data must submit a data use application describing the purpose of the study and research plan, specifying the data required. Applicants are also required to sign a data use agreement provided by the data distributor. The signed and stamped agreement, along with the research plan, should be scanned and sent to the official CLASS email address: class_ruc@163.com. Upon review and approval of both documents, the requested data will be sent to the applicant’s email within seven working days.

Exemption From Further Ethical Review

According to Article 32, Items 1 and 2 of China’s “Ethical Review Measures for Life Science and Medical Research Involving Human Subjects” (issued February 18, 2023), research involving human subjects that uses only de-identified data from public databases and does not involve direct information related to individuals may be exempt from ethical review. Furthermore, if the research data originate from a project that has obtained informed consent and the use of the data conforms to the scope of the original informed consent, it may also be considered exempt from further ethical review. As the present study utilized de-identified secondary data from the CLASS public database and adhered to the original informed consent scope, it met the criteria for exemption from additional ethical review.

Supplemental Material

Supplemental material for this article is available online.

Notes

References

Anderson

Perrin

. Technology use among seniors. Pew Res Cent Rep. 2017. https://www.pewinternet.org/2017/05/17/technolgy-use-among-seniors/

Levine

Lipsitz

Linder

. Trends in Seniors’ Use of Digital Health Technology in the United States, 2011-2014. JAMA. 2016;316(5):538-540. doi:10.1001/jama.2016.9124.

Ryan

. The US Census Bureau Reports: Computer and internet use in the United States. Census Bur Rep; 2017. https://www.census.gov/content/dam/Census/library/publications/2017/acs/acs-37.pdf

Harwood

Dooley

Scott

Joiner

. Constantly connected – The effects of smart-devices on mental health. Comput Hum Behav. 2014;34:267-272. doi:10.1016/j.chb.2014.02.006.

Meyer

Heuten

Boll

. No effects but useful? long term use of smart health devices. In: Proceedings of the 2016 ACM International Joint Conference on Pervasive and Ubiquitous Computing: Adjunct. ACM; 2016:516-521. doi:10.1145/2968219.2968314.

Zhang

Xie

, et al. Wearable Health Devices in Health Care: Narrative Systematic Review. JMIR MHealth UHealth. 2020;8(11):e18907. doi:10.2196/18907.

Kang

Park

Cho

Lee

. Recent Patient Health Monitoring Platforms Incorporating Internet of Things-Enabled Smart Devices. Int Neurourol J. 2018;22(Suppl 2):S76-S82. doi:10.5213/inj.1836144.072.

Sujith

AVLN

Sajja

Mahalakshmi

Nuhmani

Prasanalakshmi

. Systematic review of smart health monitoring using deep learning and Artificial intelligence. Neurosci Inform. 2022;2(3):100028. doi:10.1016/j.neuri.2021.100028.

Kim

Lee

. Smart Devices for Older Adults Managing Chronic Disease: A Scoping Review. JMIR MHealth UHealth. 2017;5(5):e69. doi:10.2196/mhealth.7141.

10.

Vizeshfar

Ghelbash

. Effect of a self‐care training program using smartphones on general health, nutrition status, and sleep quality in the elderly. Psychogeriatrics. 2021;21(6):910-919. doi:10.1111/psyg.12766.

11.

Abdullah

Ismael

Rashid

Abou-Elnour

Tarique

. Real Time Wireless Health Monitoring Application Using Mobile Devices. Int J Comput Netw Commun. 2015;7(3):13-30. doi:10.5121/ijcnc.2015.7302.

12.

McCann

Donnelly

O’Reilly

. Living arrangements, relationship to people in the household and admission to care homes for older people. Age Ageing. 2011;40(3):358-363. doi:10.1093/ageing/afr031.

13.

Gao

Shi

. Examining individuals’ adoption of healthcare wearable devices: An empirical study from privacy calculus perspective. Int J Med Inf. 2016;88:8-17. doi:10.1016/j.ijmedinf.2015.12.010.

14.

Long

Lei

Peng

Mao

. A scoping review on monitoring mental health using smart wearable devices. Math Biosci Eng. 2022;19(8):7899-7919. doi:10.3934/mbe.2022369.

15.

Goumopoulos

Potha

. Mental fatigue detection using a wearable commodity device and machine learning. J Ambient Intell Humaniz Comput. 2023;14(8):10103-10121. doi:10.1007/s12652-021-03674-z.

16.

Sano

Phillips

, et al. Recognizing academic performance, sleep quality, stress level, and mental health using personality traits, wearable sensors and mobile phones. In: 2015 IEEE 12th International Conference on Wearable and Implantable Body Sensor Networks (BSN). IEEE; 2015:1-6. doi:10.1109/BSN.2015.7299420.

17.

Yen

. Smart wearable devices as a psychological intervention for healthy lifestyle and quality of life: a randomized controlled trial. Qual Life Res. 2021;30(3):791-802. doi:10.1007/s11136-020-02680-6.

18.

Lee

Kwon

Lee

Kim

. Companionship with smart home devices: The impact of social connectedness and interaction types on perceived social support and companionship in smart homes. Comput Hum Behav. 2017;75:922-934. doi:10.1016/j.chb.2017.06.031.

19.

Morris

Adair

Ozanne

, et al. Smart technologies to enhance social connectedness in older people who live at home. Australas J Ageing. 2014;33(3):142-152. doi:10.1111/ajag.12154.

20.

Al-Ghamdi

. Popularity and impact of using smart devices in medicine: experiences in Saudi Arabia. BMC Public Health. 2018;18(1):531. doi:10.1186/s12889-018-5465-y.

21.

Vandenbroucke

Poole

Schlesselman

Egger

. Strengthening the Reporting of Observational Studies in Epidemiology (STROBE): Explanation and Elaboration. PLoS Med. 2007;4(10).

22.

Vavekanand

Ali Laghari

Kumar

. Applications and limitations of large language models to integrate medical context: a comprehensive review. Iran J Comput Sci. 2026;9(12):1-24.

23.

Vavekanand

. A Comprehensive Review of Multimodal Large Language Models for Medical Imaging and Omics Data. Arch Comput Methods Eng. 2026(1):1-23.

24.

Vavekanand

Kumar

. Data augmentation of ultrasound imaging for non-invasive white blood cell in vitro peritoneal dialysis. Biomed Eng Commun. 2024;3(4):1-7.

25.

Wei

Guo

. Impact of smart device use on objective and subjective health of older adults: findings from four provinces in China. Front Public Health. 2023;11:1118207. doi:10.3389/fpubh.2023.1118207.

26.

Liang

Xiong

Xie

. The effect of smartphones on the self-rated health levels of the elderly. BMC Public Health. 2022;22(1):508. doi:10.1186/s12889-022-12952-0.

27.

Wang

Liang

. Impact of Internet Use on Elderly Health: Empirical Study Based on Chinese General Social Survey (CGSS) Data. Healthcare. 2020;8(4):482. doi:10.3390/healthcare8040482.

28.

Basatneh

Najafi

Armstrong

. Health Sensors, Smart Home Devices, and the Internet of Medical Things: An Opportunity for Dramatic Improvement in Care for the Lower Extremity Complications of Diabetes. J Diabetes Sci Technol. 2018;12(3):577-586. doi:10.1177/1932296818768618.

29.

Lyons

Swartz

Lewis

Martinez

Jennings

. Feasibility and Acceptability of a Wearable Technology Physical Activity Intervention With Telephone Counseling for Mid-Aged and Older Adults: A Randomized Controlled Pilot Trial. JMIR MHealth UHealth. 2017;5(3):e28. doi:10.2196/mhealth.6967.

30.

Lao

Wang

Chang

. The Combined Effects of Sports Smart Bracelet and Multi-Component Exercise Program on Exercise Motivation among the Elderly in Macau. Medicina (Mex). 2021;57(1):34. doi:10.3390/medicina57010034.

31.

Lin

Hsu

Lay

Chiu

Chao

. Wearable device for real-time monitoring of human falls. Measurement. 2007;40(9-10):831-840. doi:10.1016/j.measurement.2007.04.001.

32.

Ehmen

Haesner

Steinke

Dorn

Gövercin

Steinhagen-Thiessen

. Comparison of four different mobile devices for measuring heart rate and ECG with respect to aspects of usability and acceptance by older people. Appl Ergon. 2012;43(3):582-587. doi:10.1016/j.apergo.2011.09.003.

33.

Lee

Chung

. Wireless sensor network based wearable smart shirt for ubiquitous health and activity monitoring. Sens Actuators B Chem. 2009;140(2):390-395. doi:10.1016/j.snb.2009.04.040.

34.

Ran

. Urban-Rural Disparity,Digital Divide and Health Inequality of the Elderly. Popul J. 2022;44(3).

35.

Goldman

Korenman

Weinstein

. Marital status and health among the elderly. Soc Sci Med. 1995;40(12):1717-1730. doi:10.1016/0277-9536(94)00281-W.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.47 MB

0.00 MB