Do physical fitness and cognitive function mediate the relationship between basic activities of daily living and quality of life in older adults with dementia?

Dementia is a major public health issue worldwide that results from diseases or injuries that affect the brain [1]. This syndrome is primarily characterized by a progressive deterioration in cognitive functions, in which its related symptoms are severe enough to interfere with independence in activities of daily living (ADLs) [2]. To date, the most recommended treatments are drug therapies and non-pharmacologic interventions that can temporarily slow the worsening of related symptoms, ameliorate functional capacity, and improve quality of life (QoL) [2].

QoL is a multidimensional construct that encompasses subjective and objective elements. Among its contributing factors, cognitive and functional abilities have often been highlighted [3‐5]. Independence in ADLs is one of the cornerstones of every QoL framework in dementia, as functional impairment negatively impacts the QoL of people with dementia [6]. Complimentarily, the dysfunction in cognitive abilities is related to decreased QoL [7]; however, the available evidence is still not convincing as this relationship seems to be associated with behavioral and psychological symptoms such as depression [8, 9]. This inconsistency may also be attributed to the discrepancies in ratings of QoL domains between individuals with dementia and their carers [10] or to different QoL frameworks.

The ability to perform ADL is related to both physical fitness and cognitive function in individuals with dementia [11, 12]. Such association needs to be better established and likely requires a comprehensive understanding of the temporal changes in cognitive and physical function, as the deterioration in both often coincide [13]. While consistent evidence supports that cognitive impairment is a more robust predictor of physical decline than vice-versa [14, 15], it is also known that cognitive function deteriorates considerably faster succeeding physical disability [16, 17]. To date, few studies have demonstrated that frailty status, physical impairments and basic ADL disability may be noteworthy predictors of dementia onset [18‐20]. Conversely, increasing evidence suggests that cognitive deterioration seems to precede and predict future functional decline in cases of mild dementia [21‐23].

Cognitive decline plays a major role in the loss of autonomy and independence as there is a clear relationship between cognitive function and the ability to perform ADLs [24], mainly due to the progressive dysfunction of executive functions [25]. Indeed, as performance in instrumental ADLs is sensitive to cognitive decline, impairment in these ADLs tends to happen in the early stages of cognitive impairment [26] and is considered a clinical hallmark of dementia. In contrast, basic ADLs decline tends to occur in more advanced stages of dementia and may happen due to the combined effect of cognitive deterioration and (or be related to) motor changes and neurological manifestations [27].

Functional impairment and disability increase the care burden on caregivers and impact the QoL of individuals with dementia [28]. While physical performance is associated with QoL in individuals with dementia [29], the contribution of cognitive function to QoL perception seems erratic [30]. However, physical and cognitive functions are often examined separately, reinforcing the lack of information about their direct contribution to the QoL or their indirect effect on ADL performance. This study aims to examine the mediating effect of physical fitness and cognitive function in the relationship between independence in basic ADLs and QoL among older adults with dementia.

Participants’ sociodemographic (e.g., age, years of education) and clinical information (e.g., dementia subtype, pharmacological treatment) were gathered via a structured questionnaire.

A total of 112 older adults were eligible for this study. Of those, five were excluded due to missing data in QoL-AD. The final sample comprised 107 older adults with ages ranging between 61 and 94 years (74.8% women; mean age 78.21 ± 7.70 years). The most common dementia subtypes were due to Alzheimer’s disease (37.4%) or unspecified causes (34.6%); most situations (69.1%) were diagnosed for a median of 3 [1–4] years. Sample characteristics are presented in Table 1.

Partial correlations controlling for age and gender are presented in Table 2. Overall, ADL was positively correlated with QoL (r = 0.310, p = 0.001), lower body strength (r = 0.479, p < 0.001), aerobic endurance (r = 0.495, p < 0.001) and functional mobility (r = 0.541, p < 0.001), while being negatively correlated with cognitive function assessed by ADAS-Cog (r = − 0.203, p = 0.038).

All measures presented an acceptable to excellent internal consistency with the exception of MMSE (α = 0.686; Table 3).

Two models were constructed to examine the mediating effect of cognitive function and physical fitness in the relationship between physical independence in ADLs and QoL. We did not find statistically significant results when testing cognitive function as a mediator between BI and QoL (β = 0.009, p = 0.345). Namely, BI did not exert a significant direct effect on cognitive function (β = 0.072, p = 0.361) nor on QoL (β = 0.118, p = 0.297). Additionally, BI directly affected QoL (β = 0.277, p = 0.002) (Fig. 1).

The fit of the model was met according to the following parameters: X²(1) = 0.303, p = 0.582, SRMR = 0.0162, RMSEA < 0.001, GFI = 0.999, NFI = 0.997 and CFI = 1.000.

As measured by the BI, independence in basic ADL had a positive significant direct effect on physical fitness (β = 0.672, p = 0.004), while this exerted a direct effect on QoL (β = 0.360, p = 0.011) (Fig. 1). Independence in basic ADL positively affected QoL mediated by physical fitness (β = 0.242, p = 0.011). No direct effect was observed between independence in basic ADL and QoL (β = 0.043, p = 0.733) (Fig. 2).

The fit of the model was met according to the following parameters: X²(4) = 0.124, p = 0.998, SRMR = 0.0037, RMSEA < 0.001, GFI = 1.000, NFI = 0.999 and CFI = 1.000. Based on R² values, the final model accounted for 45% of the variance in physical fitness and 15% in QoL.

Independence in basic ADL was significantly associated with QoL and this relationship seems to be mediated by physical fitness; the same, however, was not observed for cognitive function. Our results demonstrated an absence of statistically significant results when testing cognitive function as a mediator between BI and QoL. This could be somewhat potentially explained by behavioral and psychological symptoms commonly observed across the dementia spectrum rather than by cognitive function per se. Indeed, apathy, aberrant motor behavior and appetite disturbances are associated with higher rate of basic ADL impairments [50]. Also, Banerjee et al. [8] found that behavioral and psychological symptoms in dementia were associated with QoL, whereas cognitive function was not. In accordance with our findings, Clemmensen et al. [24] found no correlation between cognitive function and basic ADLs. Some studies even suggest that rather than global cognitive capacity, some specific cognitive functions (i.e., visual constructive or executive functions) seem to affect the performance of ADL [51]. Also, another study examining the relationship between cognitive function, demographic and disease variables on ADL functionality in individuals with mild to moderate Alzheimer’s disease showed that non-cognitive variables explained 18% of variance. However, the model explained 39% of the variance by adding neuropsychological factors [52]. The authors pointed out that constructional cognitive abilities, figural and verbal memory, longer disease duration and depression were significant predictors of declines in ADL functionality [52]. Longitudinal data demonstrated that dementia severity predicted the decline in basic ADL functioning over time [22, 53], and Vellas et al. [54] also confirmed that a severe impairment in ADAS-cog (e.g., an increase of 7 points or more) was associated with a decline in basic ADLs in a long-term follow-up (18 months). In this sense, it is crucial to further understand whether the deterioration in basic ADL is only driven by physical function or cognitive decline, as basic ADL dependency tends to occur in more severe stages of dementia [27].

As for the mediation role of physical fitness in the relationship between basic ADL independence and QoL, the evidence on cross-sectional and longitudinal studies is consistent, suggesting that specific components of physical fitness such as lower extremity muscle strength and balance are crucial for older adults to maintain a standing posture and independently perform their basic ADL. Indeed, low scores of performance tests such as TUG or chair stand test are associated with future worsening of basic ADL in older adults [55, 56]. The relationship between physical fitness and QoL is less consistent as only a few cross-sectional studies found significant associations between strength or mobility and QoL in older adults [57, 58], whereas to our knowledge no longitudinal study has analyzed this association.

Boyle et al. [59] study suggested that motor performance accounted for a high proportion of the variance in basic ADLs in people with dementia, whereas cognitive function was not significantly associated. These findings support a new evidence that found a correlation between mobility-related functional parameters and ADL performance in people with dementia [60]. Mobility is an important predictor of changes in the QoL of older adults [61] and tends to get impaired as dementia progresses, affecting everyday function and QoL [62]. TUG performance (a marker of functional mobility) is influenced by lower limb strength and balance [63], reinforcing the potential indirect role of physical fitness in QoL parameters. Specific physical fitness components such as strength, flexibility, agility/dynamic balance, and aerobic endurance are related to cognitive function, basic ADL and QoL in people with dementia [12]. Also, sarcopenia-related factors such as lower-extremity function and skeletal muscle mass seem to be associated with QoL [29, 64]. In fact, sarcopenia is a risk factor for poor QoL in older adults and is associated with an increased risk of cognitive impairment [65, 66]. A recent longitudinal study comprising older adults found that poor muscle function, but not reduced lean muscle mass, is the driver of the association between sarcopenia with the incident of dementia and mild cognitive impairment [67]. In this sense, individuals’ functional parameters should be considered when analyzing the QoL of people with dementia.

Although there is clear evidence suggesting that higher independence in basic ADL is associated with better QoL in people with dementia [68], some methodological details seem to blur this relationship. This can be attributed to several factors, of which we highlight the challenge in assessing ADLs or QoL in people with dementia, and the complexity around basic ADL deterioration across the dementia spectrum. Measuring the functional ability of people with dementia is challenging as self-reported measures in this population may raise concerns over reliability [69]. Frequently, proxy/carer/informant reports or performance measures are used because they seem to be more reliable in this population, but assessment instruments vary greatly across studies due to the lack of a standard measure [69]. We used BI to assess independence on basic ADL. BI is a generic instrument that presents acceptable psychometric proprieties in the geriatric population [70, 71]. However, several reports assume that cognitive abilities influence its scores [51]. Recent evidence even suggests that BI is not appropriate to assess ADL independence in people with dementia directly [72]. To address this issue, our data was obtained by proxy report. In what concerns the second highlighted factor, it is known that ADL performance deteriorates differently for every basic activity and that it is dependent on the disease severity [73]. There may be a hierarchy in the functional decline of ADLs, since participants tend to lose the ability to bathe independently before losing other skills, such as dressing, using the toilet, and transferring [74]. Also, the ability to eat autonomously seems well-preserved as cognition declines [27]. This hierarchy framework is supported by well-known longitudinal data showing that older adults tend to lose the ability to perform activities that require lower extremity strength earlier than upper extremity strength [75]. Basic ADL performance has a different effect on QoL across dementia stages [76], possibly due to the carers’ perception of the impact of basic ADL dependency on QoL and their lives, particularly in severe stages [73].

In our work, we outlined the importance of physical fitness, particularly lower extremity tests, when examining the relationship between basic ADL independence and QoL. However, further longitudinal studies with large samples size are also needed to explore the potential bidirectional relationship between cognitive function, disability in basic ADL and QoL. Determining other potential mediators of these relationships such as behavioral and psychological symptoms is also to be considered.

Some limitations must be acknowledged in this study, and the results should thus be interpreted with caution. Primarily, our results might not be generalizable due to our small sample size, the sample’s specific characteristics (i.e., high levels of independence in basic ADL, unequal distribution of participants from both genders and dementia subtypes), and lack of data on disease severity. In addition, both independence on ADL and QoL questionnaires relied on carers’ reports, and evidence suggests an apparent discrepancy between self-reports and caregiver reports [27], possibly induced by carer’s burden; therefore, we cannot exclude potential bias.

Our study findings present contributions to the available knowledge about the mediating effect of physical fitness (i.e., lower body function, aerobic capacity, and mobility) on the relation between basic ADL independence and QoL of older adults with dementia, reinforcing the need to address, measure and control physical capacities, and not only direct dementia-related factors like cognitive function, neuropsychiatric symptoms, and disease duration. Future longitudinal studies would be important to explore the temporal relationship between physical and cognitive function and its contribution to basic ADL dependency and QoL in individuals with dementia.