Course of activities of daily living in nursing home residents with dementia from admission to 36-month follow-up

Newly published data from the United Nations (UN), state that the number of people aged 65 years or older in Europe, has increased from 8% in 1950 to 19% in 2019, and it is expected that the proportion of the elderly in the society will continue to increase [1]. The elderly population often has more comorbidities and polypharmacy [2, 3], leading to more frailty, and thus, the growing proportion of the elderly population in the society may impose a great burden to the next of kin and the health care system.

In Norway, unlike many other countries, the health care services are public [4, 5]. The primary health care service is, by law, managed by the local municipalities [6]. Services provided include social services (such as housing and home services), domiciliary care, and institutional care, mainly in nursing homes (NH). In many countries, including Norway, the NHs are institutions that provide fulltime nursing care for people when informal care and/or domiciliary care are not sufficient to fulfil the needs of care, due to the severity of illness and/or disability. Functional impairment, age, dementia, and psychosis are the main reasons for NH admission [7], with dementia being one of the factors strongest associated with admission to NH [8‐10]. Approximately 80% of the long-term care residents in Norwegian NHs have dementia according to previous studies [11‐13].

Dementia is a syndrome caused by a variety of brain disorders, usually of a chronic progressive nature. A progressive decline in cognitive function is one of the main characteristics of dementia. The primary risk factor for developing dementia is age, and therefore, the rising life expectancies in the society will increase the number of people who develop dementia [14]. The leading type of dementia is Alzheimer’s disease (66%) [14]. Other common causes of dementia are vascular dementia, Lewy body dementia/Parkinson disease with dementia and frontotemporal dementia [15]. According to the ICD-10 criteria for dementia, there must be a decline in memory and other cognitive abilities, such as judgement and thinking, and a decline in emotional control or motivation, without the consciousness being affected [16]. People with dementia utilize health care services more frequently [17], and dementia is not just affecting the person with the disease, but also the family members and the professional health care staff supporting the person with dementia and/or family [18‐20]. Individuals with dementia often have neuropsychiatric symptoms (NPS) [21‐24]. NPS include psychiatric and behavioural symptoms such as agitation, depression, hallucinations, aggression and apathy. In a descriptive study from 2018, which included the same participants as the present study, 62% of the residents with dementia had at least one clinically significant NPS at inclusion to the NH [25]. Dementia is also often associated with a reduced ability to perform activities of daily living [11, 26‐31]. Several studies have found a correlation between the brain atrophy in individuals with dementia and their dysfunction in activities of daily living [32, 33]. The association between dementia and activities of daily living is demonstrated in the ICD-10 criteria for dementia, with one of the criteria being that there is a decline in cognitive abilities which causes impaired performance in everyday activities [16].

Activities of daily living (ADL) describes the ability to perform practical everyday tasks necessary for basic, and more complex self-care. It consists of instrumental ADL (I-ADL), with complex, higher order skills, and personal ADL (P-ADL), with self-maintenance skills, such as dressing, eating, bathing and toileting. A decline in P-ADL causes individuals to become more dependent of the caregivers and professional support [17, 34, 35].

The association between dementia and declined P-ADL function has been assessed in several studies and settings [11, 26‐28, 31, 36‐39], and some of the studies have explored the association in NH settings [11, 26, 28, 36‐38]. However, most of these studies used data from various constellations of ADL-items from the mandated Minimum data set (MDS) assessment instrument used in NHs in the USA [26, 36‐38]. Very few studies have used internationally accepted and validated P-ADL instruments, using a sum score of predefined aspects of P-ADL [11, 28]. Furthermore, most of the previous studies, have not adjusted for additional factors potentially important for P-ADL [11, 26, 36‐38]. These factors include for example NPS, physical comorbidity, age and medication [28, 40‐42].

To our knowledge, there are only two studies that have described the course of P-ADL, adjusted for variables known to have an influence on P-ADL function [28, 31]. One of the studies included older people receiving domiciliary care [31], and the other included NH residents only [28]. These studies had follow-up assessments after 18 and 12 months, respectively. More frequent assessments would give a more detailed understanding of the association between dementia and P-ADL function. Furthermore, in the study of NH residents, the residents were not included consecutively after admission to the NH [28]. To increase the understanding of the association between dementia and P-ADL function, we wanted to include the participants at the time of admission to the NH. We believed that this would give us the opportunity to explore a possible decline in P-ADL as a consequence of the NH admission, in other words, to assess whether the time after the admission had an impact on the decline in P-ADL.

The aim of this study was to examine if and how P-ADL assessed biannually over 36 months changed over time in NH residents. The primary aim was to explore the association between the severity of dementia and P-ADL function over time in NH residents. The secondary aim was to assess if and how additional factors such as neuropsychiatric symptoms, type of NH unit and use of medication were associated with the outcome.

This study of 582 NH residents with established dementia, followed with biannual assessments from admission to 36 months, showed that the degree of dementia (assessed with the CDR-SoB) at baseline, and the course of dementia at the follow-up assessments, were associated with the degree of P-ADL function. More severe dementia both at baseline and at the follow-up assessments was associated with lower P-ADL function, with the association being stable over time.

When looking at the P-ADL function adjusted for dementia assessed simultaneously with P-ADL, there was a decline in P-ADL function over time, both in unadjusted and adjusted models. This is in line with previous studies, both among NH residents and community dwelling elderly with dementia [26, 28, 31, 36‐39]. However, none of these studies assessing older adults with dementia in NHs included the residents consecutively after admission to the NH. By including the residents at the time of admission, as was done in the present study, the changes that might occur due to the time spent in the NH, will be the same for all the participants, and thus the study sample will be more homogenous when assessed over time.

Many studies have shown that more severe dementia is associated with lower P-ADL function in NH residents [26, 28, 36‐38]. In the present study, we also found that the association between P-ADL function and degree of dementia was stable over time. In a 52-month follow-up study of NH residents from Norway, the rate of decline in P-ADL function explained by the degree of dementia, decreased during follow-up [28], which differs from the present study. The follow-up time in that study was considerably longer than in the present study, and it is not clear whether a 52-month follow-up with biannual assessments of the present study would show the same results. Furthermore, the NH study mentioned above had few follow-ups (3 in total), and the NH residents were not included consecutively after admission to the NH. This may possibly explain the differences in the results between the 52-month study and the present study.

It could be questioned whether the P-ADL of an individual with dementia under 65 years could be compared to the P-ADL of an older person with dementia. Our clinical experience is that those being admitted to a Norwegian nursing home with dementia are in a state that makes municipal domiciliary care insufficient, due to dementia, P-ADL function and/or NPS, independent of age. With the PSMS scale assessing very basic activities of daily living, the measured P-ADL are comparable even if the age range of those admitted to the nursing home are broad. Nevertheless, age adjustments in the analysis should be performed, as was done in this study. However, in the present nursing home study where all participants were included at admission to the nursing home, we did not find an association between age and P-ADL or decline in P-ADL. In line with this finding, a Japanese study of persons with dementia living at home found no difference in P-ADL decline in those of 75 years or older compared to younger people [59]. However, a previous Norwegian Nursing home study from 2015 found higher age associated with lower P-ADL function [31].

In several previous studies, both among participants with and without dementia, NPS have been found to be associated with P-ADL function [28, 31, 40, 41, 60‐62]. In the present study, we found that more apathy symptoms, and higher agitation and affective sub-syndrome scores were associated with a lower P-ADL function. Even though we found an association between NPS-syndromes and P-ADL function, it remains unclear whether the worsening of NPS gives a lower P-ADL function, or if it is the other way around. However, an explanation for these results, can be that there is a common underlying reason, that gives both more NPS, and a lower P-ADL function. Several structural changes in the brain have been associated with problems in P-ADL function [32, 33, 63]. Changes in some of the same areas have been associated with having NPS [64]. This suggests that there could be common underlying structural changes in the brain that lead to both more NPS and problems in P-ADL function.

In the present study, the use of anti-dementia medication was associated with a better P-ADL function, which is in line with previous studies [28, 31, 34, 65]. This association remained significant even when controlling for additional factors, suggesting that clinicians may consider using these medications with the intention of improving the residents’ P-ADL function, and not only their cognitive function. It is well known that there are a variety of structural changes in the brain in the different types of dementia. Studies have shown that anti-dementia medication can delay these structural changes [66]. Many of the same areas are affected in individuals with a low P-ADL function, and our results could thereby be explained by the fact that anti-dementia medication also delays the changes in the brain associated with the loss of P-ADL function. To our knowledge, however, this association has not yet been explored, and further research is therefore required to explore the impact anti-dementia medication has on the brain, as well as on P-ADL function. However, another explanation for the results in the present study, can be that the residents with the lowest P-ADL function may not be able to utilize the possibly low effect of the anti-dementia medication. Thus, it may be a lower indication for giving these residents anti-dementia medication, and there will therefore be fewer people with low P-ADL function among the residents who uses these medications.

As expected, there was an association between P-ADL function and GMHR. In addition, it was found that belonging to a SCU at a NH was associated with better P-ADL function, compared to those who belonged to a RU. We do not know if the reasons for getting these results are due to differences in the structure and staffing between RU and SCU, or other factors, such as rest confounding.

This 36-months longitudinally study of NH residents with dementia, followed with biannual assessments (N = 582 at baseline, and N = 166 at 36-months), found that more severe dementia at baseline and at the follow-up assessments was associated with lower P-ADL function. This association was stable over time. The health care planners and clinicians should therefore pay attention to degree of dementia, and other factors associated with P-ADL function in NH residents, in order to improve the quality of care for the residents with dementia, and to help maintain their level of function as long as possible.