This study aimed to examine pathways and interactions of the individual’s subjective and objective physical performance, cognition, and emotions on ADLs. The SEM analysed objective performance measures as well as subjective outcomes. Overall, the results did not confirm our hypothesis that all analysed dimensions (objective and subjective physical performance, cognition, and emotion) impact ADLs similarly. The main results showed that objective physical performance and cognition positively predicted higher ADLs. However, there was no direct association between subjective physical performance and emotion and ADLs (BI). Moreover, subjective performance was strongly correlated with emotions and objective performance. Results support the importance of cognitive and physical functioning for ADLs in multimorbid nursing home residents.

Influence of objective physical performance on ADLs

The SEM's factors describing objective physical performance were gait speed, sitting and standing balance, chair rise (as part of the SPPB), grip strength, and functional reach. All these factors are attributes of the latent variable. The indicators' grip strength and functional reach show medium effects. Our cohort showed heterogenous results of the included physical performance measures compared to normative data of other studies (cf. Table 2). Nonetheless, our results confirm other studies that have included subjects with a wide range of physical performance and have consistently found that physical performance is related to the prevalence of physical disability or incidental risks [46]. In addition, the observed differences between men and women as well as decreasing functional capacity are in line with previous research as well as normative data [47,48,49]. Moreover, several studies have already shown strong predictive validity of indicators of motor performance (gait or chair stand test, [50]; handgrip strength, [51] on decrements in ADLs (and IADLs).

More specifically, all these objective measures have been shown to relate to specific domains of the BI, e.g., climbing stairs, dressing, and eating [9, 50]. For example, slow walking speed is a widely used criterion in geriatric assessment and has become a single estimator of frailty and its consequences [52]. Overall, our results indicate the interplay and importance of gait speed, balance, and chair rise in predicting ADLs and highlight the value of these assessments as diagnostic tools. On the practical side, these results accentuate how important it is for nursing home residents to receive targeted preventive or rehabilitative care to maintain or improve physical performance and, thus, promote ADLs.

Influence of subjective performance on ADLs

In distinction to current literature, no direct association between subjective performance and functional status detected by the BI could be found in the SEM. Within the SEM, subjective performance was composed of fear or concerns of falling and the score of physical well-being (SF-12). Other studies found weak correlations between functional abilities, e.g. basic or instrumental ADLs and physical health-related quality of life in older adults with cognitive impairment [53]. It has to be noted that the concerns of falling were rated relatively low in our sample [41], and physical health was slightly lower than the norm sample [40]. One explanation for the lower concerns of falling of our nursing home residents might be their accessibility to care supplies. Avoiding falling is part of their living situation and their sedentary lifestyle. Therefore, some items of the FES-I (e.g., going up/down stairs; reaching above your head or on the ground, walking up/down a slope; going out to a social event (e.g., religious service, family gathering, club meeting)) might be irrelevant in their daily life.

Influence of cognitive function on ADLs

Our results demonstrate that the latent construct cognition was positively related to the functional status (BI) in nursing home residents. The observed variables loading on the latent variable cognition are scores on the MoCA [38] and SST during single- and dual-task situations. This positive relationship also confirms previous studies highlighting a connection between cognitive function and functional status in older adults [46]. All items of the BI require some physical functions related to either fine motor control (e.g., eating or grooming) or balance and mobility (e.g., transfer or climbing stairs). Several studies have shown that gait performance is reduced in individuals with cognitive decline or can even be a predictor of the development of dementia (for review, see [54]).

EF are crucial for lower- and higher-level motor functions and, as such, have relevance for the functional status, which is closely linked to physical performance. In addition, EFs play a significant role in facilitating ADLs [55]. This is crucial in older adults since aging may lead to cognitive decline and affect multiple other structures of sensorimotor control, including sensory (tactile or proprioceptive) organs, passive and active structures of the musculoskeletal system, and cardiovascular and respiratory systems. Due to these changes to different physiological systems, older adults deal with higher amounts of uncertainty in motor control and, thus, require an increase in executive control to, e.g., maintain posture during gait [19]. In addition to basic EF, cognitive decline may affect higher-level cognitive functions necessary for ADLs, such as planning or reasoning, for spatial orientation [11]. In sum, cognitive functioning has a small to moderate impact on the functional status, but not exclusively, via the motor control/objective performance pathway.

Influence of emotion on ADLs

Interestingly, SEM revealed no direct association between the latent variable „emotion“ (SF-12 mental, CES-D, and SWLS) and BI. Overall, we used three scales, including different emotional aspects. The mental SF-12 scale comprises pain, vitality, and mood, the SWLS deals with life expectancies and quality of life, and the depression scale integrates different symptoms of depression like sleep quality, deficits of attention as well as sadness. Our results contrast with Bürge et al. [18], who revealed that depression is a significant risk factor for ADLs in nursing home residents. Also, a recent study reported that poor self-rated health, poor life satisfaction, and depression are the most substantial risk factors for ADL disabilities [56]. Interestingly, we found no direct association between depressive symptoms and ADLs. Whether this is due to our measures of emotions remains speculative.

Other intercorrelations

SEM also revealed an association between the latent variables (1) „cognition“ and „objective physical performance“, as well as a relationship between (2) “objective” and “subjective physical performance” and “emotions” for women. (1) As discussed above, studies showed that the coexistence of physical and cognitive impairments is associated with the risk of developing dementia [57]. Montero-Odasso et al. [57] showed that frail participants had a higher prevalence of cognitive impairment than those without frailty and that the combination of slow gait and cognitive impairment posed the highest risk for dementia progression (77% vs. 54%). Moreover, Dodge et al. [58] estimated that cognition accounted for 18% to 36% loss in ADLs and 11% to 29% in IADLs in a community-based sample of older adults. The identified relationship between cognition and “objective physical performance” within our SEM might be one potential pathway for these correlations between cognition and ADLs.

(2) The association between “objective physical performance” and “emotions” might relate to study findings that showed that an increase in depressive symptoms in cognitively unimpaired nursing home residents [27] was associated with a decline in physical performance. Furthermore, as reported above, reduced physical fitness and independence led to increased depressive symptoms in nursing home residents [59], which in turn might affect the intrinsic motivation to be physically active or increase sadness or anger according to their own disabilities. Also, Verghese et al. [60] suggested that the underlying processes of mood, cognition, and fitness should be observed in concert to explain performance in ADLs in older adults living independently. Concerning our SEM, one might suggest that these interdependencies could also be true for nursing home residents in long-term care. However, our findings also showed a reciprocal correlation, which means that the interdependencies obtain in both directions.

In addition to the interaction of emotion and objective physical performance, there was a strong connection between emotional variables and subjective physical performance in women. Previous studies examining community-dwelling older adults found more fall-related concerns and anxiety for women than men [61]. The SEM of our study revealed a strong positive relationship between subjective motor performance and emotions, indicating a more indirect effect of emotions on ADLs. In other populations needing care, studies reported an association between the prevalence of fear of falling and activity restrictions [62] which could lead to reduced ADL abilities in women. Despite the evidence of the relationship between physical performance and cognition, subjective physical performance might be a key to the sedentary behavior of nursing home residents who objectively have physical and cognitive resources. They can perform ADLs, but the emotional state combined with the subjective physical performance could be a more significant barrier for activities than the objective physical and cognitive state. Moreover, the behavior might be influenced by the nursing staff's behavior in the environment [63]. Evidence suggests that caregivers fear the residents' potential falls or pain when performing ADLs leading to fear avoidance behavior and activity restrictions [64]. Maybe this relationship is underestimated in daily practice in nursing home settings.

Implications for future interventions to promote resources for nursing home residents

This cross-sectional study analysed the pathways and interactions of nursing home residents’ health and physical performance, cognition, and emotion regarding their influence on ADLs. The most significant influence was found in this cohort on objective and subjective physical performance and cognition. Therefore, resources for nursing home residents related to physical performance and cognition should be targeted, for example, through exercise programs. In addition, effective interventions to strengthen health resources, maintain independence in ADLs, and prevent or delay disabilities due to functional decline are highly prioritized in healthcare research [65]. Moreover, for nursing home residents, independent of their frailty status, exercise interventions with resistance, mobility, and balance training [66] have been shown to affect independence and the ability to perform ADLs positively. Considering the SEM results, multicomponent exercise interventions (e.g., a combination of strength, endurance, balance, coordination, and task-specific training on ADLs) combined with cognitive exercises [67] might be a solution. Combined cognitive-motor interventions with dual-task exercise can promote cognitive function in community-dwelling older adults and people with dementia [68] and might simultaneously significantly improve motor and cognitive function. Finally, but especially importantly, more attention must be drawn to the perceived motor performance in nursing home residents.

Considering the investigated associations between depressive symptoms, satisfaction with life, and mental health status with fear of falling shown by SEM, future interventions should also adapt to the environment and target individual functional resources of life-space mobility of nursing home residents. Increasing these individual resources for life-space mobility and other ADLs with targeted exercise interventions, including self-efficacy measures, might also positively affect satisfaction with life and emotional well-being. In addition, social participation significantly influences satisfaction with life [5]. Therefore, exercise interventions should be conducted in groups within the living environment.

The study results might encourage caregivers, clinicians, and policymakers to include tailored exercise interventions for nursing home residents to prevent further decline in functional performance and maintain independence in ADLs.

Strengths and limitations

The study has several strengths but also some limitations. First, it must be noted that the data collection was standardized in a multicenter study within 47 nursing homes all over Germany, in urban and rural areas with high and low social status, and with a large sample size (n = 448). To the best of our knowledge, this is the first study covering various nursing homes with different SES. Moreover, the measurements included allow us to gain a holistic view of functional, cognitive, and emotional health and fitness. However, the nurses selected the participants according to their subjective estimation of the eligibility criteria. This might have led to a selection bias.

Unfortunately, we did not integrate measurements of self-efficacy and motivation for being active or mobile. There is some evidence that maintaining self-efficacy and confidence or optimism is particularly difficult under the conditions of multimorbidity and the need for care. Those in need of care often cannot reduce the unpleasant consequences of their diseases or threatening events, such as the increasing loss of physical or cognitive performance. This can lead to changed beliefs about control on a cognitive and emotional level and may have consequences on a motivational level and, in turn, on mobility [5]. Emotionally, the inability to act leads to anger or sadness, up to the intensification of depressive symptoms. Both aspects then contribute to the fact that motivation to be physically active is reduced. Due to the cross-sectional character of this analysis, these relationships need to be confirmed in future longitudinal studies.

Of course, the participants of this study were a vulnerable group, often with a multimorbid status. Moreover, the examined population integrates not an equal number of men and women. This leads to difficulties in controlling for all possible confounders and could lead to substantial heterogeneity within some subscales of the integrated measures. However, gender-related differences were only observed for the functional performance measures. Nevertheless, the overall prediction of factors affecting ADLs in this study cohort can be considered good regarding statistical data from the model fit with comparable effect sizes for both genders.