Introduction

The global trend of population aging is reshaping societies worldwide, demanding integrated approaches to address multifaceted challenges in healthcare, housing, transportation, and social services.

Aging is commonly accompanied by geriatric syndromes that compromise functional capacity1 and increase vulnerability, often leading to reduced quality of life2,3.

Functionality is typically assessed using hetero-applied instruments based on observed independence in basic, instrumental, and advanced activities of daily living (ADLs), as conceptualized by Katz4. Hetero-applied instruments are administered by professionals who observe and score task performance, focusing on objective independence. In contrast, self-applied tools are completed directly by older adults, capturing their self-perceived difficulty in daily activities. This distinction is important because subjective difficulty often emerges before observable loss of autonomy.However, such instruments rarely capture the subjective perception of difficulty, which can vary depending on physical and psychological conditions5 as well as social factors6. Moreover, self-administered instruments in Spanish, validated in Latin American populations, are scarce7,8,9, limiting their usefulness for culturally appropriate assessment and personalized care planning.

Reduced functionality leads to increased dependence and reduced well-being. Limitations often begin with mobility1 and can later extend to social and leisure activities10. These functional losses are associated with sedentarism and comorbidities11, while poverty and malnutrition have also been identified as major determinants12. Moreover, social exclusion further exacerbates vulnerability in later life13.In Turkey, for example, 90% of older women needed help with cleaning and shopping, 80% with cooking, and over 60% with personal hygiene and mobility11. Similar trends are observed in Latin America, where vulnerability is exacerbated by inequities in access to support services2. These functional limitations often generate emotional responses, such as fear, sadness, or frustration, especially in individuals who strive to remain independent14.

Understanding older adults’ perceptions of the difficulties they face in daily living is essential for developing interventions that are relevant, acceptable, and effective. This aligns with user-centered and co-design approaches, which emphasize active participation of older adults in identifying needs and shaping solutions15,16. Several recent studies have adapted and validated ADL measurement tools based on the ICF (International Classification of Functioning, Disability and Health) framework in non-Western contexts, reinforcing the need for culturally and contextually appropriate instruments17.

While performance-based assessments remain common in geriatric evaluations, recent studies highlight the importance of incorporating self-perceived functional ability to capture subjective challenges and individual contexts18.

Contemporary assessment tools like the ADL-Interview (ADL-I) have been developed in alignment with the ICF to assess perceived functional abilities in older adults19, supporting the need for context-sensitive and person-centered instruments.

This study addresses these gaps by developing and validating the EDAAD (Escala de Dificultad Asignada a Actividades de la Vida Diaria, in Spanish) scale. This is a self-applied instrument that captures perceived difficulty across a comprehensive set of ADLs. It was tested with a probabilistic sample of older adults in Chile, rigorously evaluating its psychometric properties. The study also explores the relationship between difficulty scores and sociodemographic variables, and proposes a classification of activities by difficulty level to support inclusive design and aging-in-place strategies.

Unlike most existing ADL assessment instruments, which focus on observable performance or levels of dependence, EDAAD specifically captures perceived difficulty as reported by independent older adults. This perspective is crucial for understanding early functional limitations that precede the loss of autonomy. Moreover, while widely-used instruments such as the Katz Index and Lawton IADL Scale have been validated in clinical settings, there is a lack of validated, self-administered tools tailored to the cultural and linguistic context of Latin America. EDAAD addresses this gap, offering a psychometrically sound tool that complements existing measures and enhances user-centered evaluation and intervention planning.

The importance of incorporating self-perception in the assessment of functional ability is well supported by established frameworks. The International Classification of Functioning, Disability and Health (ICF), developed by the World Health Organization, emphasizes the dynamic interaction between health conditions, personal factors, and environmental contexts. It highlights that clinical impairments do not solely define functioning and disability, but also how individuals experience and perceive their ability to perform daily activities in their environment20. Similarly, the Person–Environment Fit Model21 underscores the role of subjective experience in determining whether an individual can successfully manage environmental demands. These frameworks support the inclusion of self-reported perceived difficulty as a valid and necessary dimension for assessing functional status and designing supportive interventions.

Although several ADL and IADL instruments exist, they typically emphasize observed performance and dependence, leaving little attention to self-perceived difficulty as an early marker of functional decline. In Chile, the scarcity of validated, self-administered tools in Spanish limits the cultural and contextual accuracy of current assessments, which are often adapted from North American or European settings. Existing scales therefore risk overlooking activities and barriers specific to Latin American older adults. The EDAAD addresses these gaps by integrating culturally relevant activities, using an ordinal response format to capture nuanced perceptions, and undergoing rigorous psychometric validation with a probabilistic sample of independent older adults. As such, EDAAD provides an innovative and context-sensitive complement to traditional performance-based measures, enhancing the evaluation of functional aging in Chile.

Objectives

  1. 1.

    To develop and validate an instrument so that independent older people can report the level of difficulty they have with activities of daily life (ADL).

  2. 2.

    To analyze the relationships between sociodemographic characteristics and the difficulty assigned by Older People (OP) to the ADLs.

  3. 3.

    To evaluate the generation of a taxonomic pattern of the ADLs based on difficulty levels. To propose a taxonomic pattern, using patterns from the evaluation.

Method and materials

Design

Observational, analytic design.

Procedure

The study comprised two phases:

Phase 1:

construction and validation of the Scale of Difficulty Assigned to Activities of Daily Living (EDAAD, under its Spanish acronym).

Phase 2:

data collection and analysis using the validated scale.

Phase 1:

construction of the EDAAD scale.

The EDAAD was developed based on the activities identified by de Oliveira et al.22 and Briede & Pérez23, complemented by interviews with Chilean older adults.

Analyses for validity, dimensionality, and internal consistency were conducted in Mplus 8.4 software.

Content validity

Was assessed by 12 experts in older adult health and assistance (physicians, nurses, psychologists, occupational therapists, physiotherapists, and sociologists). Each item was rated for relevance, clarity, and pertinence. The Content Validity Ratio (CVR) was calculated using Lawshe’s method; only items meeting the minimum threshold for 12 experts (CVR ≥ 0.56) were retained. In addition, the Content Validity Index (CVI) was computed at the scale level, yielding values above 0.80, which indicated strong overall agreement among experts. Based on these results, 31 items were retained, grouped into eight dimensions: Personal care, Home care, Mobility, Social activities, Communication using technology, Hobbies, Healthcare, and Care of others.

A six-point ordinal scale was employed, as prior studies have shown that instruments with more categories improve measurement sensitivity24. Participants rate each activity from 1 (Very easy) to 6 (Very difficult).

Dimensionality

Was examined with Confirmatory Factor Analysis (CFA) using the Weighted Least Squares with Means and Variances adjusted (WLSMV) estimation method, which is recommended for ordinal data. WLSMV treats observed categorical variables as linked to latent continuous constructs through thresholds and provides robust estimates even in moderately sized samples or when normality is violated25.

Because the analysis was confirmatory, no exploratory rotation (e.g., Varimax, Promax) was applied. Model specification was theory-driven, drawing on prior studies22,23 and expert input. An eight-factor structure was tested, corresponding to the domains of Personal Care, Home Care, Mobility, Social Activities, Communication using Technology, Hobbies, Healthcare, and Caring for Others. This decision was based on theoretical and empirical evidence rather than solely on the eigenvalue > 1 rule.

Model fit was evaluated with the chi-square test, Comparative Fit Index (CFI), Tucker–Lewis Index (TLI), Root Mean Square Error of Approximation (RMSEA, 90% CI), and Standardized Root Mean Square Residual (SRMR)26,27,28. Good fit was defined as χ²/df < 3, CFI > 0.95, TLI > 0.95, RMSEA < 0.06, and SRMR < 0.0829,30. Values of χ²/df < 5, CFI > 0.90, TLI > 0.90, RMSEA < 0.10, and SRMR < 0.12 were considered acceptable31,32.

Internal consistency

Was assessed with Cronbach’s α and McDonald’s ω. While α is widely used, it assumes equal item weighting and error variance. McDonald’s ω provides a less biased estimate by incorporating factor loadings and item-specific error terms33.

Phase 2:

sampling and application.

Sampling

Phase 1 (content validation) used non-probability expert sampling, while Phase 2 employed a probabilistic, stratified multi-stage sampling design. Communes and municipalities of the Biobío Region were identified using National Statistics Institute (2022) data. Some communes were selected via stratified random sampling to ensure rural and urban representation. Within communes, blocks and then households were randomly selected. Trained interviewers visited households, obtained informed consent, and administered the survey.

The target sample size was calculated with a 95% confidence level, a 6.3% margin of error, and a proportion of 50% to maximize variability. Based on this, the required sample was set at 201 participants. For the 31-item, 8-factor CFA, this number corresponds to an item-to-participant ratio of approximately 6.5:1, which falls within the widely cited guideline of 5–10 participants per item in psychometric research. This adequacy is further supported by the use of WLSMV estimation, suitable for ordinal items with 5–7 categories and factors defined by at least three indicators. While the sample size was deemed sufficient for parameter estimation and model testing, we acknowledge that larger samples would enhance stability and generalizability of the findings9.

Data collection

Surveys were conducted face-to-face between October 2022 and February 202334. Interviewers, recruited from a pool of university students aged 28–35, underwent 11.5 h of training on study objectives, question comprehension, standardized administration, ethical considerations (informed consent, confidentiality, access to results), and strategies to minimize interviewer bias (age, gender, education, etc.)8. Each interview lasted 50–60 min. Interviewers read the questions aloud and recorded the answers on standardized forms.

Inclusion criteria

Chilean men and women aged 65 or older, residents of the Province of Concepción (urban or rural) for the past 12 months, and classified as independent with an EFAM-A score ≥ 4335.

Exclusion criteria

Individuals institutionalized, classified as dependent (EFAM < 43), or with mental health disorders affecting informed consent (e.g., dementia, schizophrenia).

Instruments

The EDAAD was designed to be culturally relevant, multidimensional, and tailored to daily activities of Chilean older adults. Unlike previous scales, it integrates all identified domains and employs an ordinal format to capture perceived difficulty.

Statistical analysis

Because scale scores did not follow a normal distribution (Kolmogorov–Smirnov test), non-parametric tests were applied. Mann–Whitney U was used for dichotomous variables, Kruskal–Wallis for multi-category variables, and Spearman’s rho for ordinal or continuous predictors. Analyses were performed in Stata SE 16.0, with significance set at p < 0.05.

Results

Phase 1: construction and validation of the EDAAD tool

To assess the factorial structure of the EDAAD tool, a confirmatory factor analysis (CFA) was conducted on the 31 items distributed across eight theoretically proposed factors (Model A). The fit indices for this model were largely outside acceptable thresholds, except for the SRMR, which was within an acceptable range8,29,30,31.

Subsequently, modification indices were examined. The highest modification index was observed between items 20 and 21 (MI = 273.837), leading to the specification of Model B, which included a correlated error term between these items. Model B showed acceptable values for SRMR and CFI, and the TLI was close to the recommended cutoff.

Further analysis revealed another high modification index between items 27 and 24 (MI = 137.518), prompting the construction of Model C, which included both correlated error terms (20–21 and 27–24). However, Model C did not substantially improve the overall fit compared to Model B (see Table 1).

Given that Model B already demonstrated an adequate fit (χ²/df < 3, CFI > 0.90, RMSEA < 0.08) and maintained the hypothesized eight-factor structure, it was chosen over Model C based on the principle of parsimony and theoretical consistency. While Model C showed a slightly better fit (CFI = 0.912 vs. 0.905 in Model B), the gain was minimal and came at the cost of adding an additional error correlation without strong theoretical justification. Such modifications increase the risk of overfitting and reduce model generalizability. Therefore, Model B was considered the most appropriate and interpretable solution.

Table 1 Fit indicators of the scale of difficulty assigned to activities of daily living for older adults (EDAAD).
Full size table

In addition, the dimensions (Table 2) that would be worked with in the analysis were defined.

Table 2 Dimensions of the Study.
Full size table

As this was a confirmatory factor analysis, no exploratory rotation (e.g., Varimax, Promax) was applied. The eight-factor structure was theory-driven, based on prior ADL research and expert input, rather than solely on eigenvalue criteria. Standardized factor loadings for the 31 items ranged from 0.34 to 0.97 (all p < 0.001) and were consistent with the predefined domains: Personal Care, Home Care, Mobility, Social Activities, Communication using Technology, Hobbies, Healthcare, and Caring for Others. Detailed loadings are reported in Supplementary Table S1 and illustrated in Fig. 1.

Fig. 1
figure 1

Confirmatory factorial structure for the Scale of difficulty assigned to activities of daily living for older adults (model B). Circles: Latent variables; Squares: Observed variables; Numbers next to squares: Standardized estimated coefficient (Standard error); Numbers next to curve line: Correlated error (Standard error).

Full size image

Description of the difficulty assigned to activities of daily living

Considering the 8-factor model, its reliability was calculated finding that, according to Cronbach’s α, it was excellent in two factors (> 0.9), good in one factor (> 0.8), adequate in three factors (> 0.7), poor in one factor (> 0.6), and questionable in the other one (> 0.5). However, using MacDonald’s ω, it was excellent in three factors (> 0.9), good in one factor (> 0.8), adequate in two factors (> 0.7), poor in one factor (> 0.6), and questionable in the final one (> 0.5) (Table 3).

Upon calculating the average scores for each factor, it was found that older adults assign greater difficulty to communicating using technologies (M = 3.71), and performing hobbies (M = 3.43), and less difficulty to personal care activities (M = 1.96) and mobility (M = 2.70).

Before conducting the bivariate analyses, the distribution of the eight perceived difficulty dimensions was assessed. Descriptive indicators included skewness and kurtosis values, as well as the Kolmogorov–Smirnov test with Lilliefors correction. Although some distributions showed moderate deviations (e.g., Personal care: skewness = 1.31; kurtosis = 3.25), all variables significantly deviated from normality (p < 0.001). Given this, and considering the ordinal nature of the scales, non-parametric statistical tests were applied, Mann–Whitney U for two-group comparisons and Kruskal–Wallis H for comparisons with more than two groups. These robust procedures ensured valid inference without the need for data transformation and preserved the interpretability of the original scale scores.

Table 3 Reliability and descriptions of the factors of the scale of difficulty assigned to activities of daily living for older adults.
Full size table

Phase 2: sampling and application of EDAAD

A sample of 201 participants was obtained, aged between 60 and 99, with an average age of 72.84 (SD = 7.06). Their sociodemographic characteristics are presented in Table 4.

Table 4 Characterization of the older adults.
Full size table

Overall, 36.3% (n = 73) of participants reported at least one ADL with a score of ≥ 4 (“moderately difficult” or higher) on the EDAAD scale, indicating the presence of self-perceived difficulty. This threshold was established to differentiate between activities perceived as easy (scores 1–3) and those perceived as difficult (scores 4–6). The results showed that men assign greater difficulty than women to activities related to home care (p < 0.05), communication using technology (p < 0.05), hobbies (p < 0.05), and caring for others (p < 0.05).

Regarding age, those who were older, assigned greater difficulty to the activities related to personal care (p < 0.05), home care (p < 0.001), and caring for others (p < 0.001), as well as for social activities (p < 0.05), mobility (p < 0.001), and communication using technology (p < 0.001).

Meanwhile, older adults with a higher educational level, assigned less difficulty to activities such as home care (p < 0.01), healthcare (p > 0.01), and caring for others (p < 0.01), as well as mobility (p < 0.001), communication using technology (p < 0.001), and hobbies (p < 0.001).

Those who reported having a higher income, whether individual or family-based, assigned less difficulty to activities related to personal care (p < 0.001), home care (p < 0.001), healthcare (p < 0.001), and caring for others (p < 0.001), as well as having less difficulty with mobility (p < 0.001), communicating using technology (p < 0.001), and with their hobbies (p < 0.001).

Older adults who had paid work assigned less difficulty to activities involving personal care (p < 0.05), home care (p < 0.05), mobility (p < 0.01), social activities (p < 0.05), and communication using technologies (p < 0.01).

There was no difference in the difficulty assigned to daily life activities for those who had or did not have children (p > 0.05).

However, there were differences between those who took part in a group for older people and those who did not. As such, those who took part in these groups, assigned less difficulty to activities related to home care (p < 0.01), mobility (p < 0.001), communication with technology (p < 0.01), hobbies (p < 0.001), healthcare (p < 0.05), and taking care of others (p < 0.01).

Finally, the older adults who reported having better health, assigned less difficulties to the activities related to personal care (p < 0.001), home care (p < 0.001), healthcare (p < 0.001), and caring for others (p < 0.001), as well as less difficulty for mobility (p < 0.001), communication using technologies (p < 0.001), and with hobbies (p < 0.001).

Effect size estimation

To assess the magnitude of the observed differences, the rank-biserial correlation (r) for Mann–Whitney U tests, and η² for Kruskal–Wallis tests, were calculated. The results, including test statistics, significance levels, and effect size estimates, are presented in Table 5.

Table 5 Non-parametric test results across predictors and dimensions of perceived difficulty in daily activities. Includes statistical test used, test statistic, p-value, and effect size estimates (η², r).
Full size table

Regarding group participation, older adults who participated in senior groups reported less difficulty in home care (Z = 2.698, p = 0.007, η² = 0.038), mobility (Z = 3.805, p < 0.001, η² = 0.075), technology (Z = 2.690, p = 0.007, η² = 0.037), leisure (Z = 4.598, p < 0.001, η² = 0.110), healthcare (Z = 2.335, p = 0.020, η² = 0.028), and caring for others (Z = 2.852, p = 0.004, η² = 0.042), ranging from minor to moderate effects.

Education level and income level, tested using Kruskal–Wallis H, also showed significant associations with multiple dimensions. For instance, educational level was associated with mobility difficulties (p < 0.001, η² = 0.115), technology use (p < 0.001, η² = 0.243), and healthcare (p < 0.001, η² = 0.253), with moderate to major effects. Similarly, higher income was associated with lower difficulty in all domains, with η² values ranging from 0.07 to 0.50.

Significant differences were also observed according to participants’ self-perceived health status. The Kruskal–Wallis test showed that older adults with better perceived health reported significantly lower difficulty in all domains. Specifically, these differences were significant in personal care (H = 8.719, p = 0.033, η² = 0.029), home care (H = 18.258, p < 0.001, η² = 0.078), mobility (H = 26.968, p < 0.001, η² = 0.122), technology use (H = 38.847, p < 0.001, η² = 0.183), hobbies (H = 47.450, p < 0.001, η² = 0.227), healthcare (H = 24.419, p < 0.001, η² = 0.109), and caring for others (H = 30.989, p < 0.001, η² = 0.143). These results indicate that the effect sizes ranged from small to large, suggesting that self-assessed health is an important factor in perceived difficulty with daily life activities.

Fig. 2
figure 2

Source: Preparation by the authors.

Activities considered in the study.

Full size image
Fig. 3
figure 3

Sociodemographic variables, degree of difficulty, and relationship with the variables considered in the study.

Full size image

The activities identified as relevant (Fig. 2) connect the sociodemographic variables with the degree of difficulty, as shown in Fig. 3. In this sense:

Gender (♀): Females report higher difficulty with “Communication using technologies,” “Care of others,” and “Home care.” There is no significant gender difference in the other activities.

Age (↑↓): Older age is associated with increased difficulty in “Communication using technologies,” “Social activities,” “Mobility,” and “Personal care.” Conversely, older age correlates with decreased difficulty in “Hobbies,” “Care of others,” “Home care,” and “Healthcare.”

Education (↓): Higher education levels correlate with lower perceived difficulty across all activities except “Social activities” and “Children,” where there is no clear relationship.

Personal Income (↓): Similar to education, higher personal income correlates with lower perceived difficulty across all activities except “Social activities,” “Children,” and “Paid job,” which show no clear relationship.

Family Income (↓): The trend mirrors personal income, with higher family income correlating with lower perceived difficulty across most activities. Exceptions are “Social activities,” “Paid job,” and “Children.”

Paid Job (X): Having a paid job does not consistently correlate with difficulty across the activities. The “X” likely indicates no statistically significant correlation.

Children (—): Having children does not show a consistent relationship with difficulty across the activities. The “—” likely indicates no statistically significant correlation.

Old Adults Group (—): Belonging to an “old adults’ group” does not have a consistent relationship with perceived difficulty across activities. The “—” likely indicates no statistically significant correlation. This is somewhat surprising, given the age correlations, and warrants further investigation.

Health Perception (↓): Better health perception correlates with lower perceived difficulty across all activities except “Paid job” and “Children.”

Discussion

Findings from this study underscore the critical need for policies and interventions tailored to the specific challenges faced by older adults. By leveraging co-design approaches and fostering interdisciplinary collaborations, these results may provide actionable pathways to improve independence and quality of life. Addressing digital and social gaps appears pivotal to ensuring accessible and inclusive services for aging populations.

This discussion is organized as follows: First, comments are made on the psychometric validation of EDAAD. Second, the results of the application of EDAAD are addressed from a taxonomic perspective. Third, the implications of these findings are explored.

Regarding the psychometric validation of the EDAAD (Escala de Dificultad Asignada a Actividades de la Vida Diaria), the results of the factor analysis support the 8-factor model B. By examining the existing tools to evaluate activities of daily living7,9,36,37,38,39, it becomes clear that most rely on Katz’s (1983) classification and assess either the ability to perform tasks or the degree of dependence. These frameworks, while clinically valuable, often fail to account for the subjective experience of difficulty as perceived by older adults. In contrast, EDAAD captures perceived difficulty, a construct shaped by personal, social, and contextual factors. This shift from evaluating objective performance to capturing individual perception addresses a critical methodological gap in gerontology, where psychosocial factors increasingly influence aging trajectories. Furthermore, in human-centered design, understanding how difficult a task feels—rather than whether it can be performed—is key to developing empathetic, contextually grounded solutions. Thus, EDAAD contributes a novel perspective that complements and extends traditional assessment approaches. This shift from evaluating objective performance to capturing individual perception addresses a critical methodological gap in gerontology, as subjective indicators of ADL/IADL difficulty have been shown to predict future decline and quality of life40,41. Furthermore, in human-centered design, understanding how difficult a task feels — rather than whether it can be performed — is key to developing empathetic, contextually grounded solutions. Recent systematic reviews emphasize that UCD and HCD methods must incorporate older adults’ subjective experiences and perceived barriers to ensure user relevance42. Studies on voice-interface design confirm that perceived difficulty and emotional response are critical in shaping usable solutions.

Reliability analysis indicated good internal consistency in six out of eight factors. However, the dimensions related to “communication using technologies” and “hobbies” showed lower reliability, suggesting the need to reassess these dimensions. It may be necessary to expand the number of items in each factor to more accurately reflect the complexity of these constructs.

When analyzing the reported difficulties by sociodemographic profile, activities can be categorized along a continuum from low to high difficulty. This classification was based on the mean scores of each factor: activities with mean values < 2.5 were classified as “low difficulty,” those between 2.5 and 3.5 as “medium difficulty,” and those > 3.5 as “high difficulty.” These thresholds were selected to reflect the ordinal nature of the response scale and to enable consistent interpretation across domains. The order observed in the sample was as follows: personal care, home care, mobility, social activities, healthcare, care of others, communication technologies, and hobbies.

Activities with less perceived difficulty

Personal care activities—such as dressing or eating—showed the least difficulty. They were inversely associated with income and self-assessed health, but positively correlated with age. These findings align with previous research5,6,13,43. Notably, tasks within personal care vary in complexity: bathing tends to be more difficult than eating, possibly due to physical and environmental demands44,45.

Home care activities, such as cooking or cleaning, were found to be moderately difficult and were also inversely related to education, income, and health perception, and were more frequently reported by older males. These gender differences may reflect traditional caregiving roles in Latin America46,47. Variability in reported difficulty across studies could stem from health status and regional differences48.

Activities with intermediate difficulty

Mobility-related ADLs had a medium level of perceived difficulty and were influenced by age, education, work status, and income. Limited access to assistive products or services may exacerbate these challenges49. Similar trends have been reported in studies from different contexts44,45.

Social and healthcare activities also presented a medium level of difficulty. Declining cognitive reserve and the lack of recent social or work engagement may explain this50,51. Although some studies report low difficulty in social engagement, the variance may be attributable to context and accessibility44.

Activities with higher perceived difficulty

Communication technologies and hobbies were perceived as the most challenging, especially by women and those with lower levels of education or income. This may reflect generational and gendered differences in digital exposure52. Digital literacy and affordability are key barriers53. Leisure activities also differ by gender, with women often preferring manual tasks that may become harder with age due to pain or fine motor decline54.

Implications

Self-perceived difficulty reveals unmet needs and can inform the design of targeted services and solutions. Interestingly, the most difficult tasks identified were related to “use of objects” rather than physical exertion, pointing to new areas for design innovation.

User-centered design approaches have proven effective in developing interventions tailored to the motivations and challenges of older adults, especially when grounded in behavioral theory, such as the Health Belief Model (HBM), which emphasizes perceived susceptibility, benefits, barriers, and self-efficacy in adopting health behaviors55.“These approaches align with the shift toward participatory design and the recognition of older adults as experts in their lived experiences.

A phased co-design intervention is proposed in Fig. 4, starting with low-difficulty ADLs to ensure early success and user engagement, followed by medium and high-difficulty activities12,56. This approach promotes user-centered design, allowing for the gradual exploration of more complex barriers while building trust and insights into users’ preferences57,58.

Fig. 4
figure 4

Source: Preparation by the authors.

Intervention strategy based on a codesign approach.

Full size image

Co-design, which involves collaboration between both older adults and professionals, helps translate functional challenges into actionable solutions15,59. Several experiences support the feasibility of co-design with older adults in areas ranging from food preparation to digital inclusion60,61,62. Recent evidence highlights that involving older adults in co-design processes enhances both the acceptability and effectiveness of health-related interventions63. This reinforces the value of the participatory approach proposed in the intervention strategy outlined in this paper. Figure 4 presents a co-creation model addressing challenges in daily living for older adults. Box (a) shows the main participants—older users, designers, and health professionals—working collaboratively. Box (b) outlines an iterative process of telling, enacting, and making that leads to tailored solutions. Box (c) classifies daily activities by difficulty level, guiding the focus of co-designed interventions.

EDAAD offers a scalable and adaptable framework to guide these processes by classifying tasks along a difficulty continuum and linking them to relevant sociodemographic variables. Moreover, the instrument can be integrated at multiple phases of the design process: during ideation, it helps prioritize user-reported difficulties to focus on; during prototyping, it guides the development of targeted solutions for high-difficulty domains; and during evaluation, it provides a standardized outcome measure to assess whether co-designed interventions effectively reduce perceived difficulty.

Limitations

This study has several limitations that should be acknowledged. First, the sample was limited to 201 older adults from a single Chilean region (Biobío), which may restrict the generalizability of the results to other cultural or geographic contexts. Second, although this sample size corresponds to an item-to-participant ratio of approximately 6.5:1—considered acceptable for CFA with ordinal items estimated via WLSMV and with ≥ 3 indicators per factor—a larger sample would further strengthen precision, allow more robust testing of measurement invariance, and improve generalizability. Third, although the psychometric analyses yielded strong indicators of internal consistency and factor structure, construct validity could be further examined using convergent and discriminant methods or test–retest reliability in future studies. For example, convergent validity could be assessed by comparing EDAAD scores with established measures such as the Katz ADL, Lawton–Brody IADL, WHODAS 2.0, or the Late-Life Function and Disability Instrument (LLFDI), as well as objective mobility tests like the Timed Up and Go (TUG) or the Short Physical Performance Battery (SPPB). Discriminant validity could be explored through measures theoretically unrelated to functional difficulty, such as social desirability, personality traits, or loneliness scales. These approaches would allow for more robust construct validation and strengthen the interpretability of EDAAD in diverse contexts. Fourth, while the instrument captures perceived difficulty, it does not directly measure functional performance or environmental barriers, which are also important dimensions of functional assessment. Future research could extend the EDAAD by incorporating environmental factors such as housing layout, accessibility of public spaces, or availability of assistive technologies, given their known influence on the ability to perform ADLs. Considering these contextual variables would enhance the ecological validity of the tool and provide a more holistic understanding of functional challenges. Finally, the cross-sectional design does not allow assessing the scale’s responsiveness to change over time or in response to interventions.

Conclusions

This study introduces EDAAD, a novel and psychometrically validated self-administered instrument that captures perceived difficulty in activities of daily living among older adults. By shifting focus from observed performance to subjective difficulty, EDAAD fills a methodological and practical gap in functional assessment. The tool demonstrated strong reliability and an interpretable eight-factor structure, while also revealing associations with key sociodemographic variables such as gender, age, education, and income. Beyond measurement, EDAAD provides researchers, clinicians, and designers with a practical and user-centered resource to guide inclusive interventions, inform policy, and support autonomy and aging in place.