Determinants of the number of ATDs used by community-dwelling people recovering from stroke with mild cognitive impairments: A path analysis

Abstract

Objective: This study examined the associative relationships among age, cognition, anxiety, and participation in explaining the number of Assistive Technology Devices (ATDs) used by stroke survivors through path analysis.

Methods: A cross-sectional study was conducted with 196 community-dwelling stroke survivors. Data on ATD usage, cognitive function (Montreal Cognitive Assessment), anxiety (Beck Anxiety Inventory), and participation (Utrecht Scale for Evaluation of Rehabilitation-Participation) were analyzed using SPSS 22.0 and Amos 24.0.

Results: The modified path model demonstrated a good fit to the data. Age, anxiety, and participation had direct effects on the number of ATDs used, while cognition did not show a statistically significant effect. Anxiety also had an indirect effect through participation, indicating a dual role of participation in either increasing or reducing ATD reliance. Age influenced anxiety and participation both directly and indirectly.

Conclusions: This study identified the pathways through which age, anxiety, and participation influence ATD usage among stroke survivors. Given the complexity of interplay of psychological and functional factors, ATD prescriptions should adopt a user-centered approach, considering participation levels, psychological responses, and environmental factors to optimize effectiveness and long-term use.

Keywords

Assistive technology device community path analysis Stroke survivors

Introduction

Stroke is a vascular disorder that affects the brain’s blood vessels, including intracerebral hemorrhage and cerebral infarction, and is a leading cause of disability and mortality worldwide (Sacco et al., 2013). Stroke can result in partial paralysis or complex dysfunctions, such as impairments in speech, comprehension, or memory, depending on the location and severity of the injury, etc. It can also lead to chronic or long-term disability (World Stroke Organization, 2019). People with disabilities often experience limitations in daily activities and restrictions in participation (Singam et al., 2015).

A rehabilitation approach that facilitates meaningful participation for individuals with chronic disabilities is essential (Cardol et al., 2002). Rehabilitation enhances daily living skills, fosters individual autonomy, supports the establishment of social roles, and aids in adaptation to new lifestyles (Young & Forster, 2007). Such rehabilitation has positive effects on physical, cognitive, and social aspects of functioning and well-being (Hopman & Verner, 2003). However, many individuals experience persistent participation restrictions following a stroke, and their daily activities are often confined to their home environments (Ada et al., 2003).

Assistive technology (AT) encompasses both systems and services that apply knowledge and technology to devices, equipment, instruments, or software to support users in engaging in purposeful and productive activities and leading an independent life (World Health Organization, 2016). Assistive Technology Devices (ATDs) are defined as products that enhance users’ abilities and enable them to perform actions, tasks, and activities independently (LaPlante et al., 1992). However, if users’ functional needs and their willingness to actively participate are not considered, ATDs may negatively impact their capabilities, social engagement, and overall participation (Scherer & Glueckauf, 2005). Additionally, users may abandon ATDs, leading to issues regarding the appropriateness of ATD prescriptions and resource waste (Wessels et al., 2003).

Multifaceted factors including personal, environmental, contextual, psychosocial, and social elements play a role in the effective adoption and application of AT, as described in various AT frameworks. Among these, the Human Activity Assistive Technology (HAAT) model emphasizes the importance of active engagement in a given environment and highlights the need to consider personal factors, including physical, cognitive, and emotional aspects, for the optimal use of AT (Cook & Hussey, 2002). Additionally, the Matching Person and Technology (MPT) model focuses on identifying and addressing personal attributes such as ability, experience, and attitude toward AT use, emphasizing that the user’s perspective is central to AT selection (Scherer, 2002, 2005). Beyond understanding AT frameworks, it is crucial to investigate the factors influencing AT usage and the number of ATDs utilized by actual users.

It has been reported that the factors influencing the number of ATDs used are associated with personal aspects such as physical safety, emotional security, and the maintenance and promotion of independence in goal-oriented activities (Gitlin, 1995). Previous studies have shown that individuals with greater difficulties in activity and participation, such as executing tasks, performing actions, and engaging in life situations, tend to use more ATDs (Agree & Freedman, 2000). This relationship is influenced by the extent to which individuals have opportunities for autonomy and greater independence, depending on their level of activity and participation (Gjessing et al., 2018). Additionally, cognitive function has been identified as a key factor in the effective use of ATDs (Sheffield et al., 2013) and in the adoption of new and diverse ATDs (Jutai et al., 2007). Consequently, cognitive function is recognized as a significant variable that facilitates ATD use (Wessels et al., 2003). Furthermore, advanced age (Spillman, 2004) and higher levels of anxiety have been found to increase ATD use (Wielandt et al., 2006). Studies examining the relationship between activity, participation, and ATD usage have reported a strong correlation between the number of ATDs used and an individual’s occupational roles, cognitive function, and psychological factors (Byun et al., 2015). Moreover, as individuals age, they frequently encounter increased difficulties and greater dependency in performing daily activities, reinforcing the importance of age as a relevant determinant (Kylberg et al., 2013).

While participation is often described as an outcome in stroke rehabilitation, it has also been reported to mediate the relationship between intrinsic factors such as cognition and anxiety and extrinsic factors such as ATDs and environmental support (Botticello et al., 2014; Hammel et al., 2015). ATDs can enhance participation by promoting functional independence, while individuals with higher participation levels may actively seek ATDs to support their engagement in meaningful activities (Federici et al., 2016; Lenker & Paquet, 2003). Given this bidirectional relationship, further research is needed to explore the mediating role of participation in the broader framework of aging, cognition, and psychological factors.

Häggblom-Kronlöf and Sonn (2007) emphasized the importance of understanding users’ functional factors to meet the needs of potential ATD users in daily life. While previous studies have explored the relationships between age, cognition, and participation in ATD use, none have examined their pathways and influences on the number of ATDs used by community-dwelling individuals with mild cognitive impairments after a stroke.

In a study by Häggblom-Kronlöf and Sonn (2007), it was reported that a comprehensive understanding of users’ functional factors is essential to meet the needs of potential ATD users in daily life. Although previous studies have examined the relationships between age, cognition, and participation in ATD use, no studies have analyzed the pathways and influences among these factors in determining the number of ATDs used by community-dwelling people with mild cognitive impairments following a stroke. In this study, age, anxiety, and cognition were designated as exogenous variables influencing other factors, while participation was set as a mediating variable, and the number of ATDs used was defined as the final endogenous variable. The purpose of this study is to develop a path analysis model examining the relationships among demographic characteristics, cognition, anxiety, and participation in explaining the number of ATDs used by community-dwelling individuals recovering from stroke with mild cognitive impairments.

Methods

Research participants and selection criteria

This study was approved by the Institutional Review Board of Konyang University (Ref. no: KYU-2018–022-02). Participants were community-dwelling people recovering from stroke. They were informed of the study’s purpose, procedures, content, and measures to protect private information before providing informed consent. The inclusion criteria were diagnosis of stroke by a neurologist; living at home; adults with stroke (over 19 years); those with normal or mild cognitive impairment (Korean Mini-mental State Examination (Kang et al., 1997, K-MMSE) scores above 18) (Jutai et al., 2007); and understand the purpose of the research and agree to participate. Exclusion criteria were individuals with communication difficulties and non-ATD users. Exclusion criteria were those with communication difficulties. A total of 233 participants answered the survey, but 37 were excluded due to no history of stroke (n = 25), incomplete answers, or did not-use of ATDs (n = 12). Thus, the final analysis included 196 participants. According to the Maximum Likelihood Estimation (MLE) method in Structural Equation Modeling (SEM), a sample size of at least 100 to 150 is required, with 200 considered more appropriate (Hair et al., 2006). Therefore, the sample size in this study was deemed appropriate.

Research procedures

This study was conducted between September 2018 and January 2019. Using a convenience sampling method, participants were recruited from public health centers, welfare centers, and home-visiting rehabilitation service centers. Data collection was carried out by occupational therapists at each institution, and the researcher provided training on data collection procedures and evaluation items to minimize inter-rater errors. Data were collected either in a designated space within the institutions or in participants’ home environments. Demographic characteristics and the number of ATDs used were reported by participants and their family members through a semi-structured questionnaire, while data on cognition, participation, and anxiety were collected by the evaluators.

Measurements

Number of Assistive Technology Devices used

The number of ATDs used in various activities, such as activities of daily living, electronic information access, and mobility, was surveyed. The number of ATDs used was assessed based on the National Assistive Technology Center’s (NATC) classification system. Usage was categorized into 12 areas: personal medical treatment, skills training, orthoses and prostheses, and recreation (NATC, 2019), along with 111 subcategories. A checklist reflecting this classification system was used to investigate the number of ATD items used by participants.

Korean version of Montreal Cognitive assessment (K-MoCA)

We used the Montreal Cognitive Assessment (MoCA) tool, which was developed by Nasreddine et al. (2005) and translated by Lee et al. (2008). This tool is divided into spatiotemporal/executive, vocabulary, attention, sentence structure, abstraction, recall, and orientation sections. The maximum score is 30. Normal cognitive function is defined as a score above 23, and a score below 22 is classified as mild cognitive impairment (Lee et al., 2008). The reliability of the instrument in this study was 0.84.

Beck Anxiety Inventory (BAI)

We used the 21-item Beck Anxiety Inventory (BAI) tool developed by Beck et al. (1988) and translated by Yook and Kim (1996). This tool measures anxiety across the cognitive, emotional, and physical domains. The scoring is based on a Likert scale from 1 to 4, with higher scores indicating greater anxiety. In previous studies, the internal consistency reliability was reported to have a Cronbach’s α of 0.91 (Yook & Kim, 1996), and in this study, the reliability was 0.95.

Utrecht Scale for Evaluation of Rehabilitation-participation (USER-participation)

We used a measurement tool developed by Post et al. (2012) and translated by Lee et al. (2017). This tool consists of three sections: measuring frequency of participation in various activities, restriction of activity due to disease or health status, and satisfaction with activities. The scoring system for each section is different. Frequency is scored from 0 to 5, restriction is scored from 0 to 3, and satisfaction is scored from 0 to 4. The sum of each section’s score is converted to a 100-point scale, with higher scores indicating higher participation rates. In previous studies, the reliability for frequency was reported with a Cronbach’s α of 0.60–0.75, α = 0.85–0.91 for restriction, and α = 0.78–0.88 for satisfaction (Lee et al., 2017). In this study, the reliability of the measurement tool was α = 0.81 for frequency, α = 0.88 for restriction, and α = 0.84 for satisfaction.

Statistical analysis

Demographic characteristics were analyzed using frequency distributions, and normality was assessed through skewness and kurtosis. Pearson’s correlation analysis was conducted to evaluate reliability. Analyses were performed using the statistical software packages SPSS 22.0 and AMOS 24.0 (IBM Corp, Armonk, NY, US).

Path analysis

SEM was used to test the hypothesized associative relationships among latent and observed variables (Kline, 2010). Path analysis used a parallel mediation model to analyze the independent effects of multiple mediators on ATD usage, assessing their direct and indirect influences. In the path model, all variables are measured simultaneously, making directionality of relationships unclear. Correlations analysis was used tointerpret these relationships, and non-correlated variables can be linked based on previous studies (Moon, 2009).

Path analysis was performed to verify the hypothetical model of the number of ATDs used and to analyze the relationship. Maximum Likelihood Estimation (MLE) was used for parameter estimation of the model representing the associations in the path analysis. In order to evaluate the goodness of fit of the model, the chi-square, Goodness of Fit Index (GFI), Tucker-Lewis Index (TLI), Normed Fit Index (NFI), Comparative Fit index (CFI), Root Mean-squared Residual (RMSEA) and Standard RMR (SRMR) was checked.

Results

Demographic characteristics

Participants were predominantly male (58.2%), aged 70–79 years (32.7%), high school educated (38.3%), and 6–10 years post-stroke (44.9%). Almost all participants (95.9%) were married and were not working (82.7%). They lived in two-person households (41.3%) and in small cities (53.6%). Some (57.1%) received home care services, and most had a monthly household income of ≤ ₩ 990,000 (KRW) (Table 1).

Table 1.

Demographic Characteristics of People Recovering From Stroke (n = 196).

Characteristics	Subjects (n)	Percentage (%)
Sex
Male	114	58.2
Female	82	41.8
Age
≤59	43	21.9
60 – 69	62	31.6
70 – 79	64	32.7
≥80	27	13.8
Education level
Uneducated	6	3.1
Elementary school	42	21.4
middle School	40	20.4
High school	75	38.3
≥ University	33	16.8
Time since onset (year)
≤5	84	42.9
6 – 10	88	44.9
≥11	24	12.2
Marital status
Married	188	95.9
Not married	8	4.1
Vocational status
Has a vocation	34	17.3
None	162	82.7
Family member
One-person	36	18.4
Two-person	81	41.3
Three/Four-person	65	33.2
≥ Five-person	14	7.1
Region of residence
Metropolis	19	9.7
Small-medium cities	105	53.6
Small town	72	36.7
Care services
Receives a care services	112	57.1
None	84	42.9
Monthly income (KRW-Won, ₩)
≤990,000	98	50.0
1,000,000 – 1,990,000	54	27.6
2,000,000 – 2,990,000	22	11.2
≥3,000,000	22	11.2

The descriptive statistics for variables

The average values of the selected exogenous variables were age 67.68, cognition 21.68, and anxiety 41.04. Participation, used as a parameter variable, was 40.33, and the average number of ATDs used, the final endogenous variable, was 9.54. All absolute values of kurtosis and skewness for the measurement variables were less than 2.3, indicating that the data satisfied the conditions for normal distribution (Lei & Lomax, 2005). The goodness of fit of the model was tested using Maximum Likelihood Estimation (MLE) in the Structural Equation Modeling (SEM) framework (Benson & Fleishman, 1994) (Table 2).

Table 2.

Descriptive Statistics for Variables (n = 196).

Variables		M±SD	Skewness	Kurtosis
Exogenous variables	Age	67.68 ± 10.63	−.448	.322
	Cognition	21.68 ± 2.63	.733	−.046
	Anxiety	41.04 ± 13.58	.647	−.195
Parameter variables	Participation	40.33 ± 11.87	−.126	−.791
Final endogenous variables	Using assistive device	9.54 ± 3.58	.001	−.714

Number of ATDs used by people recovering from stroke

ATDs were most commonly used in personal care and protection (386, 20.6%), including devices for dressing and undressing, protecting the body, and assisting with hygiene maintenance. The second most frequently used ATDs were those for furnishings and home adaptations, such as handrails and posture-assisting devices, accounting for 335 (17.9%) of the devices. The third most frequently used ATDs were 316 (16.9%) devices related to personal mobility, such as walkers and wheelchairs (Table 3).

Table 3.

Number of ATDs Used of People Recovering From Stroke (n = 196).

Types of assistive device	Examples of assistive device in use	Number of ATDs	Percentage (%)
For personal medical treatment	Aspirators, respiration trainers, supports for standing, and pressure sore mattress	223	11.9
For training in skills	Incontinence alarms, AAC, mouse control, joystick, and typing	47	2.5
Orthosis and prostheses	Upper limb orthoses, wrist-hand orthoses, shoulder orthoses, and ankle-foot orthoses	228	12.2
For personal care and protection	Bathing wear, button hooks, commode chairs, and non-slip bathmats	386	20.6
For personal mobility	Canes, walking chairs, wheelchairs, powered wheelchairs, and mobile hoists	316	16.9
For housekeeping	Assistive products for cutting, chopping, cutlery, and chopsticks and straws	17	0.9
Furnishings and adaptations to homes and other premises	Reading tables, back supports, bed lifts, and supporting handrails	335	17.9
For communication and information	In-the-ear hearing aids, audible computer displays, a magnification	191	10.2
For handing objects and devices	Tube-squeezing keys, timer switches, and environmental control systems	58	3.1
For environmental improvement and assessment	Capacity meter, electronic scale, and lighting control	48	2.6
For employment and vocational training	Work desks, work chairs, baggage trolleys, and non-slip pads	5	0.3
For recreation	Handicraft tools, gardening, and lawn care for private use	16	0.9
Total		1870	100

Abbreviations: AAC: Augmentative and alternative communication control of input units.

Correlation analysis of variables

Age of the exogenous variable showed a negative correlation with cognition (r = −.267, p < .001) and a positive correlation with anxiety (r = .183, p < .05). The correlation between anxiety and cognition was not statistically significant (r = −.066, p > .05). Participation as a parameter showed a negative correlation with age (r = −0.356, p < .001) and anxiety (r = −0.345, p < .001), and a positive correlation between cognition (r = .217, p < .001). The final endogenous variable, the number of ATDs used, showed a statistically significant positive correlation between age (r = .325, p < .001) and anxiety (r = −.351, p < .001). Cognition (r = −.108, p > .05) and participation (r = −.039, p > .05) did not show any significant correlation with the number of ATDs used (Table 4).

Table 4.

Correlation of Variables Related to Number of ATDs Used (n = 196).

Predictor variables	a	b	c	d	e
a. Age	1
b. Cognition	−0.267***	1
c. Anxiety	0.183*	−.066	1
d. Participation	−0.356***	.217**	−.345^***	1
e. Number of ATDs used	.325***	−.108	.351***	−.039	1

*p < .05, **p < .01, ***p < .001.

Measurement of hypothetical path model

Chi-square statistic and goodness-of-fit index were used to verify the model fit. The larger the chi-square, the greater the difference between the covariance matrices, which means that the model does not fit the data. The p-value of the chi-square statistic is inversely related to the chi-square, and if the p-value is larger than 0.05, the null hypothesis of the proposed model will not be rejected (Figure 1).

Figure 1.

Hypothetical path model diagram.

The goodness-of-fit index for the hypothetical model of this study was chi-square = 23.719 and the p-value was .000, which was less than .05. Other goodness-of-fit indices showed good levels above .90 for CFI (.907) and (GFI .963), while TLI (.649) and NFI (.895) showed indices lower than .90. Additionally, both RMSEA (.159) and SRMR (.103) exceeded the 0.1 cutoff, indicating that they did not meet the established statistical standards and required further adjustments (Table 5).

Table 5.

Test of Model Fit.

Goodness	χ² (p)	df	TLI	NFI	CFI	GFI	RMSEA	SRMR
Hypothetical path model	23.719 (.000)	4	.649	.895	.907	.963	.159	.103
Modified path model	3.724 (.155)	2	.922	.969	.984	.992	.066	.032

Measurement of modified path model

To improve model fit, modification indices (M.I.) were examined, identifying potential adjustments that could enhance statistical suitability. The results suggested that adding covariance paths between age and cognition (M.I. = 13.851, Par Change = −7.418) and age and anxiety (M.I. = 6.510, Par Change = 26.243) would significantly reduce model misfit. These adjustments were incorporated to enhance model fit and better capture the underlying relationships between variables (Kline, 2010).

After incorporating these adjustments, the theoretical validity of the revised model was reviewed. Prior research confirmed relationships were age to anxiety (Zamzam et al., 2020) and age to cognition (Lee & Kahng, 2011). In the modified path model, Age, Anxiety, and Cognition were designated as exogenous variables. Participation was set as a mediating variable that links the effects of Age and Anxiety to the endogenous variable, the Number of ATDs used. Cognition was modeled as an exogenous variable with a direct path to the Number of ATDs used.

The result of the fitness of the modified model diagram showed a chi-square of 3.724 and had a p-value of .155, exceeding .05. Other indices were TLI (.922), NFI (.969), CFI (.984), and GFI (.922) which all exceed .90, meeting the model fit. The RMSEA (≤.08 good model) was .066 and SRMR (<.05 good model) was .032, and the goodness-of-fit index for the modified model was generally good (Table 5).

Standardized coefficients (β), which represents the degree of influence of independent variables on dependent variables in the modified path model, are presented in Figure 2 and Table 6. Factors directly affecting the number of ATDs used were participation (β = .210, p < .01), anxiety (β = .361, p < .001), and age (β = .322, p < .001). Factors directly affecting participation were age (β = −.304, p < .001), anxiety (β = −.290, p < .001). Age was found to be a factor that directly affects anxiety (β = .183, p < .01) and cognition (β = −.267, p < .001) (Figure 2).

Figure 2.

Modified path model diagram.

Table 6.

Path Coefficients of the Number of ATDs Used Model.

Path	β	B	SE	Critical ratio
Age → Anxiety	.183	.233	.090	2.595^**
Age → Cognition	−.267	−.066	.017	−3.861^***
Age → Participation	−.304	−.339	.072	−4.684^***
Anxiety → Participation	−.290	−.253	.057	−4.468^***
Age → Number of ATDs used	.322	.109	.024	4.629^***
Cognition → Number of ATDs used	−.043	−.059	.089	−.664
Anxiety → Number of ATDs used	.361	.095	.018	5.385^***
Participation → Number of ATDs used	.210	.063	.021	2.969^**

**p < .01, ***p < .001.

Direct, indirect, and total effects of factors

Anxiety (β = .361, p < .001), participation (β = .210, p < .01), and age (β = .322, p < .001) had a statistically significant direct effect on the number of ATDs used. For the total effect, age (β = .325, p < .01) was the most important, followed by anxiety (β = .301, p < .01) and participation (β = .201, p < .01). The factors that indirectly affected the number of ATDs used were anxiety (β = −.061, p < .01). Cognition was not significant in both direct (β = −.043, p > .05) and total effects (β = −.043, p > .05) of the number of ATDs used. Age had a direct effect on anxiety (β = .183, p < .01), participation (β = .304, p < .001), and cognition (β = −.267, p < .001), and anxiety had a significant direct effect on participation (β = −.290, p < .001) (Table 7).

Table 7.

Path Analysis of Standardized Direct, Indirect, and Total Effects.

	Path of variable	Direct effects	Indirect effects	Total effects
Age	→ Cognition	−.267^***		−.267^**
	→ Anxiety	.183^**		.183^**
	→ Participation	−.304^***	−.053^**	−.356^**
	→ Number of ATDs used	.322^***	.003	.325^**
Cognition	→ Number of ATDs used	−.043		−.043
Anxiety	→ Participation	−.290^***		−.290^**
Anxiety	→ Number of ATDs used	.361^***	−.061^**	.301^**
Participation	→ Number of ATDs used	.210^**		.210^**

**p < .01, ***p < .001

Discussion

The level of participation of people recovering from a stroke varies from person to person (Van Der Zee et al., 2013). Path analysis showed that participation positively influences ATDs use, indicating that higher participation levels lead to greater ATD adoption. This aligns with previous research suggesting ATDs compensate for functional limitations and promote independence (Agree & Freedman, 2000; Gjessing et al., 2018).

Anxiety negatively affected participation, reducing engagement in daily activities, but directly increased ATD use, indicating greater reliance on assistive technology. This aligns with previous research indicating that ATDs compensate for functional limitations and promote independence (Agree & Freedman, 2000; Gjessing et al., 2018). These findings highlight participation as a direct determinant of ATD use, while also being influenced by anxiety. Anxiety directly increased ATD usage, while also influencing participation, which was associated with ATD adoption. Participation’s impact on ATD usage may vary depending on individual psychological and social contexts (Federici et al., 2016). While participation is associated with greater ATD use, its relationship with anxiety and ATD reliance requires further investigation. These findings suggest that participation plays a significant role in shaping ATD utilization patterns while being influenced by broader psychological and functional factors. Future research should explore how participation, in conjunction with psychological and social influences, contributes to assistive technology use in stroke rehabilitation.

Anxiety had a direct positive effect on both participation and the number of ATDs used, but its influence on ATD usage was moderated by participation. While individuals with higher anxiety levels tended to use more ATDs, those with higher participation levels were less dependent on ATDs despite experiencing high anxiety. These results align with a previous study that found anxiety to be a factor influencing ATD usage (Wielandt et al., 2006). People recovering from stroke have difficulty in performing daily activities and forming social relationships due to anxiety (Astrom, 1996). Such enhanced anxieties are known to limit patients’ perception and learning (Knowles, 1981). Wielandt et al. (2006) reported that ATD usage increased when anxiety levels decreased, which contrasts with the findings of the present study, where higher anxiety levels were associated with greater ATD usage. However, our results suggest that participation plays a compensatory role, enabling individuals to rely less on ATDs even when experiencing high anxiety levels. A study by Song (2003) found that psychological and social anxiety decreased as individuals improved their ability to carry out daily activities. Therefore, to alleviate psychological anxiety among stroke survivors, it is necessary to promote participation, provide appropriate ATDs, and enhance their utilization.

Age had a direct effect on the number of ATDs used and an indirect effect through participation. The findings indicate that older individuals used a greater number of ATDs. Additionally, age indirectly influenced ATD usage through participation, suggesting that older individuals who engage in higher levels of participation may experience greater benefits from ATDs. This result aligns with a previous study showing that age influences ATD usage (Edwards & Jones, 1998). As individuals age, they experience both physical and psychological decline (Ritsatakis, 2008), and the speed of recovery from symptoms is slower than that of younger people with disabilities (Marini et al., 2004). ATDs play an even more critical role for older stroke survivors by compensating for health decline and functional limitations, ultimately promoting independence (Nasr et al., 2016).

On the other hand, cognition was not found to be a significant factor in the direct path affecting the number of ATDs used. In this study, to enhance the accuracy of interviews and self-reported questionnaires, only participants with a K-MMSE score of 18 or higher were included in the analysis. This suggests that within the range of normal to mild cognitive, cognitive function may not be a key determinant in explaining the number of ATDs used. Previous studies have reported that cognitive ability influences ATDs usage by affecting users’ capacity to operate the devices and recall training methods for their use (Wielandt et al., 2006). Additionally, clinical characteristics of individuals recovering from a stroke indicate that while cognitive function plays a role in the effective utilization of ATDs, its impact is more pronounced when mobility impairments are present, even if physical function is compromised (Jutai et al., 2007). These findings suggest that once cognitive function reaches a certain range, its effect on the number of ATDs used may be limited. Therefore, future research should include individuals with varying degrees of cognitive impairment to provide a more comprehensive analysis of the relationship between cognitive function and ATD usage.

Limitations and future research

A limitation of this study is the lack of measurement of exogenous variables that may influence the number of ATDs used, such as demographic and socioeconomic factors, stroke-related functional impairment, and residential environment conditions. Since ATD usage is shaped by multidimensional factors beyond individual psychological and behavioral aspects, future research should incorporate these variables for a more comprehensive analysis. Additionally, while the hypothetical path model in this study was supported by previous research, unmeasured confounding variables may still exist, potentially affecting the pathways identified. Refining the model by integrating additional variables and conducting longitudinal analyses would enhance its predictive validity. Moreover, as this study employed a cross-sectional design, it cannot establish a directional relationship between ATD usage and participation. While participation was hypothesized as a mediator, the relationship may be bidirectional, with ATD usage enhancing participation and higher participation levels driving ATD use. This dynamic may also vary across rehabilitation stages and personal contexts, requiring future longitudinal studies to determine its potential directional relationship. Furthermore, to minimize measurement bias in self-reported data, only participants with mild cognitive impairment were included, which limits the generalizability of the findings to individuals with more severe impairment. Future studies should consider a broader range of cognitive abilities to better understand the role of cognition in ATD usage. Furthermore, various mediation models, such as parallel or moderated mediation, should be explored to refine the understanding of the pathways influencing ATD usage identified in this study and provide a more comprehensive analysis.

Conclusion

The overall fitness of the path model for the number of ATDs used appears to be robust. Anxiety, Age, and participation were found to have both direct and indirect effects on the number of ATDs used, while Cognition did not show a statistically significant effect. These findings confirm the associations among key factors influencing ATD usage among community-dwelling individuals recovering from stroke with normal to mild cognitive impairment. Rather than focusing on the provision of multiple ATDs, prescribing should prioritize individualized user needs, psychological factors, and participation levels. Clinicians should assess both psychological and functional aspects, such as anxiety and participation, to ensure appropriate device selection and effective utilization. Given the aging population and rising disability rates, a strategic approach is needed to optimize ATD allocation and improve accessibility. Future research should explore the personalized prescription frameworks and strategies to enhance long-term ATD adoption.

Key findings

For community-dwelling individuals recovering from stroke with normal to mild cognitive impairment, the number of ATDs used was influenced by age, anxiety, and participation.

Prescribing ATDs for people recovering from stroke should consider multiple factors, including participation level, psychological response, and choices of activities in community living.

The prescription of individual ATDs should be based on a thorough review of the user’s actual needs, functional abilities, and environmental context to ensure appropriate and effective utilization.

Footnotes

Declaration of conflicting interests

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

Funding

The author(s) disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by a grant from the National Research Foundation of Korea (NRF), funded by the government of Korea (MSIP; Ministry of Science, ICT, & Future Planning) (NRF-2017R1C1B5017510).

Ethics statement

ORCID iD

Young-Myoung Lim

References

Ada

Dean

C. M.

Hall

J. M.

Bampton

Crompton

(2003). A treadmill and overground walking program improves walking in persons residing in the community after stroke: A placebo-controlled, randomized trial. Archives of physical medicine and rehabilitation, 84(10), 1486–1491. https://doi.org/10.1016/S0003-9993(03)00349-6

Agree

E. M.

Freedman

V. A.

(2000). Incorporating assistive devices into community-based long-term care: An analysis of the potential for substitution and supplementation. Journal of Aging and Health, 12(3), 426–450. https://doi.org/10.1177/089826430001200307

Astrom

(1996). Generalized anxiety disorder in stroke patients: A 3-year longitudinal study. Stroke, 27(2), 270–275. https://doi.org/10.1161/01.str.27.2.270

Beck

A. T.

Epstein

Brown

Steer

R. A.

(1988). An inventory for measuring clinical anxiety: Psychometric properties. Journal of Consulting and Clinical Psychology, 56(6), 893–897. https://doi.org/10.1037/0022-006X.56.6.893

Benson

Fleishman

J. A.

(1994). The robustness of maximum likelihood and distribution-free estimators to non-normality in confirmatory factor analysis. Quality and Quantity, 28(2), 117–136. https://doi.org/10.1007/bf01102757

Botticello

A. L.

Rohrbach

Cobbold

(2014). Disability and the built environment: An investigation of community and neighborhood land uses and participation for physically impaired adults. Annals of Epidemiology, 24(7), 545–550. https://doi.org/10.1016/j.annepidem.2014.05.003

Byun

E. J.

Kim

S. M.

Park

J. H.

(2015). A Comparative study on the participation of the elderly and its affecting factors between rural and urban areas. Journal of Korean Society of Occupational Therapy, 23(3), 67–77. https://doi.org/10.14519/jksot.2015.23.3.06

Cardol

de Jong

B. A.

van den Bos

G. A.

Beelem

de Groot

I. J.

de Haan

R. J.

(2002). Beyond disability: Perceived participation in people with a chronic disabling condition. Clinical Rehabilitation, 16(1), 27–35. https://doi.org/10.1191/0269215502cr464oa

Cook

Hussey

(2002). Assistive technologies: Principles and practice: Mosby Elsevier.

10.

Edwards

N. I.

Jones

D. A.

(1998). Ownership and use of assistive devices amongst older people in the community. Age and Ageing, 27(4), 463–468. https://doi.org/10.1093/ageing/27.4.463

11.

Federici

Meloni

Borsci

(2016). The abandonment of assistive technology in Italy: A survey of national health service users. European Journal of Physical and Rehabilitation Medicine, 52(4), 516–526. https://doi.org/10.1016/j.apmr.2014.12.008

12.

Gitlin

L. N.

(1995). Why older people accept or reject assistive technology. Generations: Journal of the American Society on Aging, 19(1), 41–46. https://www.jstor.org/stable/44877289.

13.

Gjessing

Jahnsen

R. B.

Strand

L. I.

Natvik

(2018). Adaptation for participation! Disability and Rehabilitation: Assistive Technology, 13(8), 803–808. https://doi.org/10.1080/17483107.2017.1384075

14.

Häggblom-Kronlöf

Sonn

(2007). Use of assistive devices–a reality full of contradictions in elderly persons' everyday life. Disability and Rehabilitation: Assistive Technology, 2(6), 335–345. https://doi.org/10.1080/17483100701701672

15.

Hair

J. F.

Black

W. C.

Babin

B. J.

Anderson

R. E.

Tatham

R. L.

(2006). Multivariate data analysis (6th ed.): Pearson Prentice Hall.

16.

Hammel

Magasi

Heinemann

Gray

D. B.

Stark

Kisala

Carlozzi

N. E.

Tulsky

Garcia

S. F.

Hahn

E. A.

Hahn

E. A.

(2015). Environmental barriers and supports to everyday participation: A qualitative insider perspective from people with disabilities. Archives of physical medicine and rehabilitation, 96(4), 578–588. https://doi.org/10.1016/j.apmr.2014.12.008

17.

Hopman

W. M.

Verner

(2003). Quality of life during and after inpatient stroke rehabilitation. Stroke, 34(3), 801–805. https://doi.org/10.1161/01.STR.0000057978.15397.6F

18.

Jutai

Coulson

Teasell

Bayley

Garland

Mayo

Wood-Dauphinee

(2007). Mobility assistive device utilization in a prospective study of patients with first-ever stroke. Archives of physical medicine and rehabilitation, 88(10), 1268–1275. https://doi.org/10.1016/j.apmr.2007.06.773

19.

Kang

D. L.

Hahn

(1997). A validity study on the Korean Mini-Mental State Examination (K-MMSE) in dementia patients. Journal of the Korean Neurological Association, 15(2), 300–308.

20.

Kline

R. B.

(2010). Promise and pitfalls of structural equation modeling in gifted research. In Thompson

Subotnik

R. F.

(Eds.), Methodologies for conducting research on giftedness (pp. 147–169): American Psychological Association.

21.

Knowles

R. D.

(1981). Dealing with feelings. Managing anxiety. American Journal of Nursing, 81(1), 110–111. https://pubmed.ncbi.nlm.nih.gov/6906117/.

22.

Kylberg

Löfqvist

Horstmann

Iwarsson

(2013). The use of assistive devices and change in use during the ageing process among very old Swedish people. Disability and Rehabilitation: Assistive Technology, 8(1), 58–66. https://doi.org/10.3109/17483107.2012.699585

23.

LaPlante

M. P.

Hendershot

G. E.

Moss

A. J.

(1992). Assistive technology devices and home accessibility features: Prevalence, payment, need, and trends (No. 217). U.S. Department of Health and Human Services, Public Health Service, Centers for Disease Control and Prevention, National Center for Health Statistics.

24.

Lee

H. J.

Kahng

S. K.

(2011). Age and gender differences in cognitive functioning among elderly. Ment Health Soc Work, 37(37), 255–278.

25.

Lee

J. H.

Park

J. H.

Kim

Y. J.

Lee

S. H.

Post

M. W.

Park

H. Y.

(2017). Validity and reliability of the Korean version of the utrecht scale for evaluation of rehabilitation-participation. Occupational Therapy International, 2017(1), 9452051–9452055. https://doi.org/10.1155/2017/9452051

26.

Lee

J. Y.

Lee

D. W.

Cho

S. J.

D. L.

Jeon

H. J.

Kim

S. K.

You Ra Lee Youn

J. H.

Kwon

Lee

J. H.

Maeng Je Cho Cho

M. J.

(2008). Brief screening for mild cognitive impairment in elderly outpatient clinic: Validation of the Korean version of the Montreal Cognitive Assessment. Journal of Geriatric Psychiatry and Neurology, 21(2), 104–110. https://doi.org/10.1177/0891988708316855

27.

Lei

Lomax

R. G.

(2005). The effect of varying degrees of nonnormality in structural equation modeling. Structural Equation Modeling, 12(1), 1–27. https://doi.org/10.1207/s15328007sem1201_1

28.

Lenker

J. A.

Paquet

V. L.

(2003). A review of conceptual models for assistive technology outcomes research and practice. Assistive Technology, 15(1), 1–15. https://doi.org/10.1080/10400435.2003.10131885

29.

Marini

Baldassarre

Russo

De Santis

Sacco

Ciancarelli

Carolei

(2004). Burden of first-ever ischemic stroke in the oldest old: Evidence from a population-based study. Neurology, 62(1), 77–81. https://doi.org/10.1212/01.WNL.0000101461.61501.65

30.

Moon

S. B.

(2009). Basic concepts and applications of structural equation modeling: With AMOS17.0: Hakjisa.

31.

Nasr

Leon

Mountain

Nijenhuis

S. M.

Prange

Sale

Amirabdollahian

(2016). The experience of living with stroke and using technology: Opportunities to engage and co-design with end users. Disability and Rehabilitation: Assistive Technology, 11(8), 653–660. https://doi.org/10.3109/17483107.2015.1036469

32.

Nasreddine

Z. S.

Phillips

N. A.

Bédirian

Charbonneau

Whitehead

Collin

Cummings

J. L.

Chertkow

(2005). The Montreal Cognitive assessment, MoCA: A brief screening tool for mild cognitive impairment. Journal of the American Geriatrics Society, 53(4), 695–699. https://doi.org/10.1111/j.1532-5415.2005.53221.x

33.

National Assistive Technology Center . (2019). Public notice on designation of assistive device (No. 2016-15): National Rehabilitation Center. Resource document. https://knat.go.kr/knw/home/board/brd_vew.php

34.

Post

M. W.

van der Zee

C. H.

Hennink

Schafrat

C. G.

Visser-Meily

J. M.

van Berlekom

S. B.

(2012). Validity of the utrecht scale for evaluation of rehabilitation-participation. Disability & Rehabilitation, 34(6), 478–485. https://doi.org/10.3109/09638288.2011.608148

35.

Ritsatakis

(2008). Demystifying the myths of ageing: World Health Organization Regional Office for Europe.

36.

Sacco

R. L.

Kasner

S. E.

Broderick

J. P.

Caplan

L. R.

Connors

J. J.

Culebras

Elkind

M. S. V.

George

M. G.

Hamdan

A. D.

Higashida

R. T.

Hoh

B. L.

Janis

L. S.

Kase

C. S.

Kleindorfer

D. O.

Lee

J. M.

Moseley

M. E.

Peterson

E. D.

Turan

T. N.

Valderrama

A. L.

… Vinters

H. V.

(2013). An updated definition of stroke for the 21st century: A statement for healthcare professionals from the American heart association/American stroke association. Stroke, 44(7), 2064–2089. https://doi.org/10.1161/STR.0b013e318296aeca

37.

Scherer

M. J.

(2002). The change in emphasis from people to person: Introduction to the special issue on assistive technology. Disability & Rehabilitation, 24(1-3), 1–4. https://doi.org/10.1080/09638280110066262

38.

Scherer

M. J.

(2005). Living in the state of stuck: How assistive technology impacts the lives of people with disabilities (4th ed.): Brookline Books.

39.

Scherer

M. J.

Glueckauf

(2005). Assessing the benefits of assistive technologies for activities and participation. Rehabilitation Psychology, 50(2), 132–141. https://doi.org/10.1037/0090-5550.50.2.132

40.

Sheffield

Smith

C. A.

Becker

(2013). Evaluation of an agency-based occupational therapy intervention to facilitate aging in place. The Gerontologist, 53(6), 907–918. https://doi.org/10.1093/geront/gns145

41.

Singam

Ytterberg

Tham

von Koch

(2015). Participation in complex and social everyday activities six years after stroke: Predictors for return to pre-stroke level. PLoS One, 10(12), Article e0144344. https://doi.org/10.1371/journal.pone.0144344

42.

Song

Y. H.

(2003). Srudy of stroke patient's anxiety followed by daily behavior level and family supporting level (master's thesis): Pochon CHA University.

43.

Spillman

B. C.

(2004). Changes in elderly disability rates and the implications for health care utilization and cost. The Milbank Quarterly, 82(1), 157–194. https://doi.org/10.1111/j.0887-378x.2004.00305.x

44.

Van Der Zee

C. H.

Visser-Meily

J. M.

Lindeman

Jaap Kappelle

Post

M. W.

(2013). Participation in the chronic phase of stroke. Topics in Stroke Rehabilitation, 20(1), 52–61. https://doi.org/10.1310/tsr2001-52

45.

Wessels

Dijcks

Soede

Gelderblom

G. J.

De Witte

(2003). Non-use of provided assistive technology devices, A literature overview. Technology and Disability, 15(4), 231–238. https://doi.org/10.3233/TAD-2003-15404

46.

Wielandt

McKenna

Tooth

Strong

(2006). Factors that predict the post-discharge use of recommended assistive technology (AT). Disability and Rehabilitation: Assistive Technology, 1(1-2), 29–40. https://doi.org/10.1080/09638280500167159

47.

World Health Organization . (2016). Priority assistive products list: Improving access to assistive technology for everyone, everywhere (No. WHO/EMP/PHI/2016.01). World Health.

48.

World Stroke Organization . (2019). Facts and figures about stroke. Resource document. [Website]. https://www.world-stroke.org/component/content/article/16-forpatients/84-facts-and-figures-about-stroke

49.

Yook

S. P.

Kim

Z. S.

(1996). A clinical study on the Korean version of Beck Anxiety Inventory : Comparative study of patient and non-patient. Korean Journal of Clinical Psychology, 16(1), 185–197.

50.

Young

Forster

(2007). Review of stroke rehabilitation. BMJ, 334(7584), 86–90. https://doi.org/10.1136/bmj.39059.456794.68

51.

Zamzam

Tamtomo

D. G.

Widyaningsih

(2020). Biopsychosocial factors affecting quality of life in post-stroke patients: A path analysis evidence from surakarta hospital, central java. Journal of Epidemiology and Public Health, 5(1), 1–14. https://doi.org/10.26911/jepublichealth.2020.05.01.01