The accelerating global aging process has transformed the health management of functionally disabled older populations into one of the most pressing public health challenges of the 21st century. According to projections from the United Nations, by 2050, the number of adults aged 65 or over will be twice as big as the amount of children under the age of five and also surpass the number of adolescents aged between 15 and 24 years, with improvements in survival expected to add approximately 5 years to the life expectancy at birth for the world’s population (1). In China, this demographic transition is equally pronounced, with 264 million people aged over 60 years and 190 million people aged over 65 years recorded in 2020 (2). Among the global elderly population, disability and chronic disease are closely linked to advancing age, with substantial numbers of functionally impaired older people requiring comprehensive care support (3).

Malnutrition represents a particularly critical concern in this population, with prevalence rates varying significantly across care settings, from 3% in community environments to approximately 30% in rehabilitation and subacute care facilities (4). Functionally disabled older adults are particularly vulnerable to malnutrition due to multiple factors including swallowing difficulties, reduced appetite, medication side effects, and physical limitations that impair food preparation and consumption (5). The consequences are severe, with malnourished disabled older adults experiencing a 2.5-fold increased risk of mortality, 40% higher rates of infection complications, and significantly prolonged hospital stays compared to well-nourished counterparts (6).

The Knowledge, Attitudes, and Practices (KAP) framework serves as a fundamental diagnostic research tool that illuminates a population’s comprehension, beliefs, and behaviors regarding specific health topics, operating on the premise that knowledge positively influences attitudes, which subsequently shape behavioral practices (7). In the context of functionally disabled older adults, family caregivers play a crucial role in implementing nutritional management interventions. This is particularly significant in China, where the unique “9,073” elderly care model indicates that approximately 90% of elderly individuals live at home, 7% depend on community support, and only 3% reside in professional institutions (8). The complexity of this relationship becomes evident when considering that caregivers face multifaceted challenges, including physical harm, psychological pressure, and responsibility management burdens (9). Research has demonstrated that caregivers of moderately and severely functionally disabled older adults show significantly higher caregiving burden scores and poorer health-related quality of life compared to those caring for individuals with mild disabilities (8). Understanding caregivers’ KAP profiles is therefore crucial for developing targeted interventions that can improve both caregiver competence and care outcomes. The KAP framework has been widely used to examine health-related behaviors across diverse populations because of its intuitive logic linking knowledge, attitudes, and practices. Nevertheless, its application to caregiver populations requires nuance, since attitudes in caregiving are often operationalized not as favorable or unfavorable dispositions but as perceived burden or role appraisal. This may lead to atypical patterns, where greater knowledge can increase awareness of caregiving challenges, thereby elevating perceived burden, while burden in turn may constrain the translation of knowledge into practice. These dynamics suggest that while the KAP model provides a useful heuristic, the interpretation of associations should be contextualized within caregiving research.

However, limited research has specifically examined KAP regarding nutritional management among caregivers of functionally disabled older adults, with existing studies primarily focusing on general caregiving competence or specific disease conditions rather than comprehensive nutritional care practices (10, 11). In the present study, we interpreted attitude within the KAP model as caregiver burden, consistent with studies that conceptualize emotional strain and role appraisal as an attitudinal component of caregiving. Therefore, this study aimed to assess the knowledge, attitudes, and practices (KAP) of caregivers of functionally disabled older adults regarding nutritional management.

Initially, a total of 638 questionnaires were initially collected. After excluding 3 questionnaires from participants who did not consent to data usage, 1 questionnaire from a respondent under 18 years old, 5 questionnaires identified as outliers, 19 questionnaires where the functionally disabled older adults were under 60 years old, and 60 questionnaires with incorrect answers to trap questions, 550 valid questionnaires remained for final analysis, with an effective rate of 86.21%.

Caregivers demonstrated insufficient knowledge, moderately strained attitudes, yet generally proactive practices concerning the nutritional management of functionally disabled older adults. These findings underscore the need for targeted educational interventions to enhance caregivers’ nutritional knowledge, which may, in turn, optimize their attitudes and improve care practices in community and home-based elderly care settings.

This study explored the KAP of caregivers regarding nutritional management for functionally disabled older adults in China, revealing a set of interrelated but uneven patterns. While most caregivers acknowledged the importance of nutritional care, as indicated by the dominant portion identifying it as a critical aspect of caregiving, this recognition did not consistently translate into adequate knowledge or uniformly supportive attitudes. The gap between perceived importance and actual knowledge may reflect a broader issue of knowledge recognition without access to relevant, practical information, a phenomenon previously noted in similar caregiving populations. For example, in a study on family caregivers of patients with cerebral infarction in China, the mean knowledge score was 6.67 ± 1.73, while attitudes and practices were relatively higher at 32.95 ± 2.46 and 28.64 ± 4.39, respectively, with strong positive effects observed from knowledge to both attitudes and practices (10). In contrast, our study revealed a higher knowledge range (0–24) but comparatively moderate knowledge levels and negatively correlated attitudes, highlighting contextual and content-specific challenges in elderly nutritional care. Similarly, Shahin and Hussien (2021) reported that caregivers of children with epilepsy initially exhibited low knowledge, attitude, and practice scores, which significantly improved after a structured educational intervention. Their findings underscore that without systematic and targeted support, caregivers tend to struggle in translating recognition into actionable practice, reinforcing our observation of a mismatch between perceived importance and operational knowledge (18).

The SEM findings offer additional insight into these discrepancies. Knowledge showed a strong association with caregiving practice, both directly and indirectly through attitude. However, the negative association between knowledge and attitude suggests a more complex dynamic. As caregivers acquire more nutritional knowledge, they may become increasingly aware of the limitations in their caregiving environments or personal capacity, which can, in turn, increase perceived stress or emotional burden. This finding indicates that the traditional KAP pathway may not fully capture the complexity of caregiving contexts. Alternative frameworks, such as stress-appraisal-coping models or role theory, may also help explain why greater knowledge is linked to heightened burden and why burden may constrain the application of knowledge in practice. Alternative explanations should also be considered. On the one hand, this discrepancy is explained by the fact that attitude was measured using the Zarit Burden Interview Short Form (ZBI-12), which reflects caregiver burden rather than a positive attitudinal construct. Accordingly, higher knowledge may help reduce burden, while greater burden may hinder good practices, thereby yielding negative correlations. Another possibility relates to sample characteristics, as participants were primarily recruited from hospital settings and may not represent the full spectrum of caregiving experiences. Contextual influences, such as limited community support and the predominance of family-based care in China, may also exacerbate the sense of burden even among more knowledgeable caregivers. Taken together, these findings suggest that the KAP framework, while informative, should be interpreted flexibly in caregiving research, with recognition of burden as a negative attitudinal construct. This interpretation aligns with prior observations in caregiving research, particularly in settings where structural support is limited and the caregiving role is assumed with minimal preparation or training (19, 20). However, given the cross-sectional design of our study, these findings should be interpreted as associations rather than causal effects. It should also be acknowledged that the use of the ZBI-12 to represent the attitudinal component is a conceptual limitation, as this instrument measures caregiver burden rather than attitudes specifically related to nutritional management.

Attitudinal responses indeed reflected considerable emotional strain. A substantial proportion of caregivers reported feeling overwhelmed by the demands of caregiving, with concerns extending to time constraints, social withdrawal, and uncertainty in task execution. These expressions are consistent with patterns documented in other healthcare systems, where caregivers of individuals with high dependency levels often experience a tension between duty and capacity, particularly in the absence of external resources or shared responsibilities (11, 19). It is worth noting that these burdens were not uniformly distributed. For example, caregivers of individuals in poorer health conditions tended to report more negative attitudes.

In contrast, caregiving practices were generally proactive and stable. Many respondents indicated consistent engagement in dietary planning, hydration monitoring, and adaptation of food texture based on swallowing ability. These practices demonstrated strong positive correlations with knowledge scores and inverse associations with burdened attitudes, reinforcing the pathway identified in the structural equation model. However, areas requiring more specialized or adaptive responses—such as nasogastric feeding and tracking weight changes—were less frequently implemented. This pattern suggests that while caregivers may perform well on routine tasks, the application of more nuanced or technical strategies remains limited. Such findings mirror reports from comparable caregiving settings where practice quality improves with knowledge exposure but plateaus when caregivers lack ongoing support or advanced guidance (11, 21).

Socioeconomic factors further stratified these outcomes. Caregivers with prior training consistently demonstrated higher scores across all three dimensions, indicating the efficacy of even limited structured instruction. The association between lower income and higher knowledge and practice scores, although initially counterintuitive, may reflect more hands-on caregiving involvement among households unable to afford formal care services. Conversely, wealthier respondents may delegate daily care, resulting in limited personal familiarity with nutrition management protocols. These findings highlight how household context influences not only caregiving responsibilities but also the accumulation of caregiving competencies (22, 23).

Most caregivers reported managing elderly individuals with multiple chronic conditions—primarily cardiovascular and metabolic diseases—followed by mental health issues and physical injuries. These health challenges often require specific dietary interventions and close monitoring, yet such conditions were not matched by correspondingly high knowledge scores in the relevant areas. Table 2 reveals persistent uncertainty around topics such as micronutrient needs, seasonal diet adjustments, and safety precautions during feeding, particularly for nasogastric procedures. Inconsistent familiarity with these aspects points to structural weaknesses in information dissemination and caregiver preparation, a limitation previously observed in non-institutional elder care environments (23, 24).

Emotional burden, as captured in the attitudinal dimension, appeared to be shaped by the absence of institutional scaffolding. Caregivers frequently reported wishing to relinquish their responsibilities or feeling that their own lives had been significantly disrupted. Although such sentiments are not uncommon in long-term care contexts, the degree to which they co-occur with knowledge gaps and weak support systems warrants greater policy attention. These findings echo broader regional challenges in elder care where informal caregivers carry disproportionate responsibilities without sufficient access to training or psychosocial support (22, 25).

Our findings indicate that increasing caregiver knowledge alone may inadvertently elevate perceived burden. Therefore, a more systematic training infrastructure is urgently needed, integrating both procedural knowledge and strategies to mitigate stress. Caregiver education programs should focus not only on general nutritional principles but also on procedural knowledge specific to clinical conditions and feeding modalities. Training should be delivered through accessible formats, such as short video modules or community workshops, and adapted to varied literacy levels. Content should cover areas identified in this study as especially weak, including safe feeding techniques, the physiological impact of common chronic diseases, and signs of malnutrition. Evidence from related intervention studies suggests that even brief but targeted educational input can lead to marked improvements in caregiving outcomes, provided the material is practically oriented and culturally appropriate (26, 27).

Alongside technical training, the emotional dimensions of caregiving require greater institutional acknowledgment. Strategies to mitigate caregiver stress could include peer support networks facilitated by community health centers, access to professional dietary consultation, and respite care services. Interventions of this kind have been shown to reduce caregiver burnout and improve both care quality and caregiver retention in other settings, particularly when implemented within a broader system of elder care coordination (28, 29).

These proposed solutions must be embedded within local systems of care and tailored to the heterogeneity of caregivers identified in this study. Professional caregivers, family members, and volunteers differ not only in their baseline competencies but also in their motivations and access to resources (30, 31). Designing interventions that respond to these differences may enhance engagement and long-term effectiveness. For example, family caregivers may benefit from flexible learning modules, while volunteers could be integrated into structured community health teams with ongoing supervision. Ultimately, improving nutritional care for functionally disabled older adults requires a shift in how caregiving roles are supported and understood. Isolated interventions are unlikely to yield sustainable improvements unless they are situated within a more coordinated framework that includes training, supervision, emotional support, and institutional recognition of caregivers’ contributions (32, 33).

This study has several limitations that should be considered when interpreting the findings. First, the cross-sectional design precludes any inference of causality among knowledge, attitudes, and practices. Second, data were collected through self-reported online questionnaires, which may be subject to recall bias and social desirability bias. Third, the attitudinal component was operationalized using the ZBI-12, which measures caregiver burden rather than attitudes specifically related to nutritional management; this conceptual mismatch should be taken into account when interpreting the findings. Fourth, the study sample was recruited through convenience sampling in hospital settings, which may introduce selection bias. Caregivers without hospital contact, such as those providing care entirely at home, may differ in knowledge, burden, and practices. Fifth, the study was limited to a specific geographic region, which may further restrict the generalizability of the results to other populations or caregiving settings. Additionally, this study did not account for potential unmeasured variables (e.g., access to health services, social support), which may influence the SEM results. Although our sample size (N = 550) exceeds commonly recommended thresholds for SEM, the number of estimated parameters in the model may still affect the stability of some parameter estimates., larger samples would further improve the stability and generalizability of the parameter estimates.

In conclusion, caregivers of functionally disabled older adults demonstrated limited nutritional knowledge, accompanied by moderately negative attitudes, yet relatively active practices, suggesting a disconnect between awareness and behavioral execution in nutritional care. Targeted educational interventions that enhance caregivers’ nutritional knowledge may help foster more positive attitudes and, in turn, further improve caregiving practices in clinical and community settings. Given that greater knowledge was associated with increased caregiver burden in our study, educational programs should be designed not only to enhance nutritional knowledge but also to mitigate potential stress. For example, training sessions could integrate modules on coping strategies, stress management, and problem-solving skills alongside nutrition education. In addition, linking caregivers to community resources and peer support networks may help reduce the emotional load associated with caregiving. Tailored, interactive, and context-specific training formats, such as workshops or online programs with practical case simulations, could further improve both the accessibility and the effectiveness of interventions. By combining knowledge transfer with psychosocial support, such programs may optimize both caregiver well-being and care outcomes.