Early Start of Anti-Dementia Medication Delays Transition to 24-Hour Care in Alzheimer’s Disease Patients: A Finnish Nationwide Cohort Study

Abstract

Background:

Dementia is one of the strongest predictors of admission to a 24-hour care facility among older people, and 24-hour care is the major cost of Alzheimer’s disease (AD).

Objective:

The aim of this study was to evaluate the association of early start of anti-dementia medication and other predisposing factors with 2-year risk of transition to 24-hour care in the nationwide cohort of Finnish AD patients.

Methods:

This was a retrospective, non-interventional study based on individual-level data from Finnish national health and social care registers. The incident cohort included 7,454 AD patients (ICD-10, G30) comprised of two subgroups: those living unassisted at home (n = 5,002), and those receiving professional home care (n = 2,452). The primary outcome was admission to a 24-hour care facility. Exploratory variables were early versus late anti-dementia medication start, sociodemographic variables, care intensity level, and comorbidities.

Results:

Early anti-dementia medication reduced the risk of admission to 24-hour care both in patients living unassisted at home, with a hazard ratio (HR) of 0.58 (p < 0.001), and those receiving professional home care (HR, 0.84; p = 0.039). Being unmarried (HR, 1.69; p < 0.001), having an informal caregiver (HR, 1.69; p = 0.003), or having a diagnosis of additional neurological disorder (HR, 1.68; p = 0.006) or hip fracture (HR, 1.61; p = 0.004) were associated with higher risk of admission to 24-hour care in patients living unassisted at home.

Conclusion:

To support living at home, early start of anti-dementia medication should be a high priority in newly diagnosed AD patients.

Keywords

Alzheimer’s disease cholinesterase inhibitors dementia Finland healthcare institutionalization memantine nursing homes register

INTRODUCTION

Alzheimer’s disease (AD) is the most common cause of dementia in adults aged 65 years and older and is the leading cause of disability in the elderly [1]. The prevalence of AD is growing rapidly worldwide commensurate with increases in life expectancy [2]. As a progressive neurodegenerative condition, AD causes significant deterioration in cognitive, functional, and mental capabilities, which results in varying levels of difficulties in a daily living.

Currently available pharmacological treatments alleviate AD symptoms, but there is no cure for AD and the disease is eventually fatal [2]. Acetylch-olinesterase inhibitors (AChEIs), donepezil, rivastigmine, and galantamine, are typically used to treat mild to moderate AD, whereas the N-methyl-D-aspartate (NMDA) receptor blocker memantine is used to treat moderate to severe AD. Early start of anti-dementia medication (within 3 months of diagnosis) has been suggested to reduce the risk of death in AD patients compared with late start of medication (after 3 months of diagnosis) [3].

As the disease progresses, AD patients require increasing levels of medical and social care, leading to a high burden on caregivers and significant costs to society. Dementia is one of the strongest predictors of admission to institutional 24-hour care among older people [4]. It has been estimated that approximately 25%of patients with dementia become institutionalized within three years of diagnosis and 70–90%before death [5 –8]. In Western Europe, the mean total societal cost per patient estimate for AD were € 30,122 and € 35,255 in 2010 and 2015, respectively, including direct medical (health care, medication), non-medical (social care), and informal (unpaid care) costs [2]. In France, Germany, and UK, the 18-month societal costs per patient were estimated at € 33,339, € 38,197, and € 37,899 in 2010 [9]. In a previous Finnish study, the direct medical and non-medical costs of AD in 2014 were estimated at € 30,808 [10]. Institutional care is the major cost of AD, and delaying of institutionalization has been suggested to be cost reducing for European health systems [11].

Comprehensive characterization of risk factors for institutionalization is critically important in attempts to support AD patients living at home and for developing cost-effective care. A complex framework of predisposing and protective factors, as well as cultural and social influences and available resources, affect institutionalization decisions [12]. Poorer cognition, behavioral and psychological symptoms, degree of functional impairment, and caregiver burden have been shown to be associated with an increased risk of institutionalization [12 –14]. Severe disease and older age significantly increase the risk of institutionalization, whereas the effect of many sociodemographic factors is complex and dependent on sociocultural environment [12, 14]. Although the first results of randomized clinical trials assessing the effect of AChEIs on institutionalization of AD patients showed no positive impact, more recent evidence suggests that long-term treatment with AChEIs alone or in combination with memantine delay the time for institutionalization [15 –18]. However, currently available information on the effect of anti-dementia medication on institutionalization is largely based on studies with relatively small sample size, and population-level data is scarce.

In Finland, approximately 92,000 people (1.7%) of the entire Finnish population suffer from dementia, and roughly 14,500 new diagnoses are being made annually [19]. Approximately 70–80%of all dementia patients have AD [2]. The Finnish elderly care system has a goal of keeping the elderly living at their homes, and 93%of AD patients diagnosed between 2011–2014 lived at home at the time of diagnosis [3]. To support living at home, elderly persons may be comprehensively evaluated for the need to receive professional home care. The care of patients with dementia is provided by tax-based public health and social care, and follows uniform practices across the country, based on national care guidelines [20]. Decision about AD medication should be made for each patient soon after the diagnosis, but the actual initiation of medication may be delayed due to administrative reasons such as delays in reimbursement decisions. Approximately 85%of AD patients start medication within 3 months of diagnosis, and almost all individuals with dementia receive intensive professional care at the end of life, regardless of their sociodemographic or economic resources [3, 8]. Comprehensive patient and treatment data are routinely recorded in the nationwide social and health care registers, which allows evaluation of care practices and outcomes, and forms the basis for development of future management strategies.

The primary objective of the study was to evaluate the impact of early start (within 3 months of diagnosis) of anti-dementia medication versus late (anti-dementia medication started after 3 months of diagnosis or no medication during the follow-up) on 2-year risk of transition to 24-hour care. Secondary objectives included analyses of the effects of indiv-idual medications (donepezil, rivastigmine, galantamine, or memantine) and other predisposing factors (sociodemographic factors, comorbidities, level of home care) on 2-year risk of transition to 24-hour care.

MATERIALS AND METHODS

Data sources and study population

This was a retrospective, non-interventional registry study based on comprehensive individual-level data from the Finnish national registries. The study was based on a cohort of 316,470 individuals who were≥74 years of age in 2011–2014, from several regions in Finland [3].

Data on diagnoses (The International Classification of Diseases, 10th revision, ICD-10 codes), age at the time of first diagnosis, gender, health and social care utilization, and housing status were obtained from the Care Registers for Health and Social Care by the Finnish Institute for Health and Welfare. Data on taxable income were obtained from the Tax Registry by the Finnish Tax Administration. Date of death was obtained from the Cause of Death Registry by Statistics Finland. Data on purchases of reimbursed drugs were obtained from the Register of Reimbursable Drugs by the Social Insurance Institution of Finland.

A date of first AD diagnosis (any ICD-10 code in the class G30^*) in 2012 defined the index date for an incident cohort of 9,204 AD patients (Fig. 1). For each patient, the AD diagnosis was verified to be the first by requiring two years of history without recorded AD diagnoses. In addition, patients who had prior admission to a 24-hour care facility (n = 1,646) or admission to a 24-hour care facility/death by the index date (n = 104) were excluded from the primary study cohort (n = 7,454). The cohort was followed up for 4 years: 2 years before the index date and a maximum of 2 years after the index date or until admission to a 24-hour care facility or death, whichever occurred first; an exception being Fig. 2, in which the death after transition was also reported. Due to the fact that persons need to be evaluated to receive professional home care and may hence differ from those living unassisted at home (i.e., without professional care), the primary study cohort was categorized based on their status at the index date: those living unassisted at home (no professional home care, n = 5,002), and those receiving professional home care (referred to as patients receiving home care; n = 2,452). In the primary analysis for the total population, this categorization was used as an adjusting variable. In the secondary analyses, these categories were analyzed as two separate subgroups.

Fig. 1

Population flowchart.

Fig. 2

Changes in housing status during the follow-up. The proportion of patients by housing status±2 years of index date.

AD patients who had started anti-dementia medication (donepezil, rivastigmine, galantamine, or memantine) within three months from the AD diagnosis were defined as early anti-dementia medication starters (n = 6,630 at baseline). Patients who started anti-dementia medication later than 3 months from diagnosis (n = 483 at baseline) as well as those who did not start anti-dementia medication at any point during the follow-up period (n = 341 at baseline) were classified as late anti-dementia medication starters (n = 824 in total). Individual anti-dementia medications were analyzed based on the first medication started after the diagnosis.

Outcome measures and explanatory variables

The primary outcome was an admission to admission to a 24-hour care facility, which included 24-hour care in a primary care health center (admission to ward care in a primary care health center for a period of more than 90 days), 24-hour care in a nursing home, and long-term 24-hour sheltered housing (admission to a facility offering 24-hour sheltered housing). The primary outcome was analyzed in the total cohort (primary analysis) and in the two subgroups of patients living unassisted at home and in homecare (secondary analyses).

The following sociodemographic and care variables were investigated as potential predisposing factors for transition to 24-hour care: age (> 85 versus≤85 years); patient’s income (categorized according to belonging into the highest income decile based on patients’ yearly income in the index year 2012); marital status (being unmarried versus married in 2012); having a family caregiver receiving financial support for providing informal home care (informal caregiver, yes versus no); level of professional home care (classified into care intensity levels 1, 2–3, 4–5) based on the burden of care categories by the Finnish Institution of Health and Welfare Registry for Home Care Patients (Supplementary Table 1). Patients living unassisted at home were categorized as care intensity level 1. Both patients living unassisted at home and those receiving professional home care include patients who have a family caregiver receiving financial support for providing informal home care (referred to as having an informal caregiver). In addition, the prevalence of the following comorbid conditions, based on data two years prior to the index date, were investigated as potential predisposing factors: diabetes (ICD-10: E10–E14), cardiovascular diseases (ICD-10: I20–I25, I48, I50, I60–I69), chronic bronchial/lung diseases (ICD-10: J40–J47), pneumonia (ICD-10: J15–J18), cancer (ICD-10: C^*), hip fracture (ICD-10: S72), mental diseases (ICD-10: F04–F09), other neurological disorder (ICD-10: G20, G35, G40) (^*refers to any ICD-10 code underneath the designated parent code). Diagnoses with the most clinical relevance were included in the assessment of individual comorbidities.

Statistical analyses

Survival time was defined as the time from the date of diagnosis to the date of transition to 24-hour care, date of death, or the end of two years post-diagnosis, whichever occurred first. The Fine-Gray (Fine JP, Gray RJ) proportional hazards model was used to estimate subdistribution hazard ratios (HRs) and 95%confidence intervals (CI) for 2-year risk of transition to 24-hour care. Death was considered as a competing event. Standard error was clustered at the municipality level to account for possible autocorrelation, especially that resulting from the inter-municipal variation in care system structures, such as nursing home bed capacity. Municipalities with less than 50 individuals in the study population were clustered together.

Cumulative incidence function estimates were used to assess differences in the rate of transition to 24-hour care between subgroups of AD patients living unassisted at home with no regular care services and those receiving professional home care, as well as differences between early users of individual anti-dementia medications and all late users. Stata version 15.0 was used in the estimation.

Ethical considerations

This study was approved by The Finnish Institute for Health and Welfare (THL/517/5.05.00/2014). No ethics approval or informed consent from the cohort was required by Finnish legislation, as the persons were not contacted, the study did not affect the treatment of the patients, and only pseudonymized data were used. The study was conducted in accordance with the Helsinki Declaration of 1975, Good Pharmacoepidemiology Practices (GPP), and Data Protection Directive (DPD).

RESULTS

Cohort characteristics

Of the AD patients in the total study cohort, 67.1%(n = 5,002) lived unassisted at home and 32.9%(n = 2,452) received professional home care at the index date (referred to as home care group) (Fig. 1, Table 1). The mean follow-up time was similar in patients living unassisted at home (19.2 months; SD, 8.2) and in patients receiving home care (18.8 months; SD, 7.4) (Table 1). The majority of patients in both groups were female (63.4%and 71.0%in patients living unassisted at home and receiving home care, respectively). At the index date, the mean age was higher for patients receiving home care (83.7 years; SD, 5.1) than for patients living unassisted at home (81.7 years; SD, 5.3). Altogether, 29.9%of patients living unassisted at home and 45.2%of patients receiving home care were over 85 years of age. A lower proportion of patients living unassisted at home (2.1%) versus receiving homecare (15.5%) had a family caregiver who received support for informal care. Of patients receiving home care, most (66.3%) were categorized at level 2–3 of care.

Table 1

Descriptive statistics of the cohort at baseline

	Unassisted at home Total n = 5,002 (67.1%)		Professional Home Care Total n = 2,452 (32.9%)		All Total n = 7,454
Basic demographics	mean	SD	mean	SD	mean	SD
Age (y), mean (SD)	81.7	5.3	83.7	5.1	82.4	5.4
Follow-up time (months), mean (SD)	19.2	8.2	18.8	7.4	19.0	7.9
	n	%	n	%	n	%
Age over 85 y, n (%)	1,498	29.9%	1,109	45.2%	2,607	35.0%
Died (no transition), n (%)	511	10.2%	255	10.4%	766	10.3 %
Died (with or without transition), n (%)	782	15.6%	404	16.5%	1,186	15.9%
Female, n (%)	3,177	63.4%	1,742	71.0%	4,919	65.6%
Highest income decile, n (%)	549	11.0%	174	7.1%	723	9.7%
Housing status	n	%	n	%	n	%
Transition to a 24-hour care facility	1,279	25.6%	924	37.7%	2,203	29.6%
Unmarried	2,681	53.6%	1,748	71.3%	4,429	59.4%
Informal caregiver	106	2.1%	309	15.5%	486	6.5%
Care intensity level 1	5,002¹	100.0%	131	5.3%	5,133	68.9%
Care intensity levels 2-3	n/a	n/a	1,625	66.3%	1,625	21.8%
Care intensity levels 4-5	n/a	n/a	696	28.3%	696	9.3%
Anti-dementia medication	n	%	n	%	n	%
Anti-dementia medication started within 3 months of index date	4,427	88.5%	2,203	89.9%	6,630	88.9%
No anti-dementia medication within 3 months of index date²	575	11.5%	249	10.1%	824	11.1%
No anti-dementia medication during the follow-up	259	5.2%	82	3.3%	341	4.6%
Galantamine³	239	4.8%	136	5.5%	375	5.0%
Memantine³	469	9.4%	343	14.0%	812	10.9%
Donepezil³	3,024	60.4%	1,359	55.4%	4,383	55.8%
Rivastigmine³	695	13.9%	365	14.9%	1,060	14.2%
Comorbidities	n	%	n	%	n	%
Diabetes	580	11.6%	419	17.1%	999	13.4%
Cardiovascular disease	2,372	47.4%	1,351	55.1%	3,723	49.9%
COPD	238	4.8%	131	5.3%	369	5.0%
Pneumonia	416	8.3%	247	10.1%	663	8.9%
Cancer	755	15.1%	361	14.7%	1,116	15.0%
Hip fracture	130	2.6%	104	4.2%	234	3.1%
Neurological disorder	127	2.5%	76	3.1%	203	2.7%
Mental disorder	241	4.8%	96	3.9%	337	4.5%

COPD, chronic obstructive pulmonary disease; SD, standard deviation. ¹Patients living unassisted at home were categorized in as care intensity level 1, ²Including patients who did not start anti-dementia medication during the follow-up, ³First medication started after the diagnosis.

The proportion of early anti-dementia medication starters was similar in patients living unassisted at home (88.5%) and patients receiving home care (89.9%) (Table 1). During the follow-up, 5.2%of patients living unassisted at home and 3.3%of patients receiving home care did not start any anti-dementia medication. Donepezil was the most frequently used first medication among the patients who started medication within three months of diagnosis, used by 60.4%of patients living unassisted at home and by 55.4%of patients receiving home care. The use of rivastigmine and galantamine was similar in both groups, while a higher proportion of patients receiving home care (14.0%) started with memantine compared to patients living unassisted at home (9.4%). A total of 542 patients started more than one medication within 3 months of diagnosis.

Cardiovascular disease was the most common comorbidity in both groups, diagnosed in 47.4%of patients living unassisted at home and in 55.1%of patients receiving home care. Diabetes, COPD, pneumonia, hip fracture, and neurological diseases were slightly more common in patients receiving home care, while cancer and mental diseases were more common in patients living unassisted at home (Table 1).

Changes in housing status

A total of 2,203 (29.6%) patients were transitioned to 24-hour care facility during the 2-year follow-up. The proportion of patients transitioned to 24-hour care facility was lower in patients living unassisted at home (25.6%, n = 1,279) compared to patients receiving home care (37.7%, n = 924). Mortality rate was similar in both groups (15.6%and 16.5%in patients living unassisted at home and receiving home care, respectively). Of those alive two years after the index date, 39.4%(n = 2,472) of patients lived unassisted at home, 32.1%(n = 2,013) received home care, and 28.4%(n = 1,783) were transitioned to 24-hour care facility (Fig. 2).

Anti-dementia medication and 2-year risk of transition to 24-hour care

Early anti-dementia medication use (within 3 months of diagnosis) was associated with lower risk of transition to 24-hour care in the total cohort (HR, 0.647; 95%CI, 0.562–0.745) (Fig. 3). The effect of early anti-dementia medication start was stronger in patients living unassisted at home (HR, 0.584; 95%CI, 0.466–0.732) compared with patients receiving home care (HR, 0.843; 95%CI, 0.717–0.992) (Figs. 4 5). Of individual anti-dementia medication, early galantamine start was associated with the lowest risk of transition to 24-hour care (HR, 0.498; 95%CI, 0.417–0.595) for all patients, followed by donepezil (HR, 0.609; 95%CI, 0.522–0.711), rivastigmine (HR, 0.739; 95%CI, 0.622–0.877), and memantine (HR, 0.781; 95%CI, 0.671–0.910) (Fig. 3). All AChEIs significantly decreased the risk of transition to 24-hour care in both patients living unassisted at home and patients receiving home care, whereas the reduced risk for memantine was observed only in total population (Figs. 3 –5).

Fig. 3

Factors associated with 2-year risk of transition to 24-hour care facility in the total cohort. The figure shows subdistribution hazard ratios with 95%confidence intervals and p-values, estimated using the Fine-Gray competing risk regression model. Early starters (started drug within 3 months from diagnosis), total (any drug), or individual medications versus late starters total (started drug after 3 months from diagnosis/no medication). COPD, chronic obstructive pulmonary disease.

Fig. 4

Factors associated with 2-year risk of transition to 24-hour care facility in AD-patients living unassisted at home. The figure shows subdistribution hazard ratios with 95%confidence intervals and p-values, estimated using the Fine-Gray competing risk regression model. Early starters (started drug within 3 months from diagnosis) total (any drug) or individual medications versus late starters total (started drug after 3 months from diagnosis/no medication). COPD, chronic obstructive pulmonary disease.

Fig. 5

Factors associated with 2-year risk of transition to 24-hour care facility in AD-patients receiving professional homecare. The figure shows subdistribution hazard ratios with 95%confidence intervals and p-values, estimated using the Fine-Gray competing risk regression model. Early starters (started drug within 3 months from diagnosis) total (any drug) or individual medications versus late starters total (started drug after 3 months from diagnosis/no medication). COPD, chronic obstructive pulmonary disease.

Two years after the index date, 27%of early compared with 38%of late anti-dementia medication starters were transitioned to 24-hour care facility (Fig. 6). Of early starters, 15%of patients using AD patients using galantamine, 22%of patients using rivastigmine, 23%of patients using donepezil, and 23%of patients using memantine were transitioned to 24-hour care facility.

Fig. 6

2-year estimates for the cumulative incidence of transition to 24-hour care facility in the total cohort of AD patients. In the estimation, death was modelled as a competing risk event. Early starters (started drug within 3 months from diagnosis) total (any drug) or individual medications versus late starters total (started drug after 3 months from diagnosis/no medication).

Other factors associated with 2-year risk of transition to 24-hour care

A female sex (Hazard Ratio, HR, 1.099; 95%Confidence Interval, CI, 1.016–1.189), age≥85 (HR, 1.675; 95%CI, 1.528–1.837), being unmarried (HR, 1.517; 95%CI, 1.394–1.652), living unassisted at home (HR, 1.399; 95%CI, 1.046–1.872), and having informal caregiver (HR, 1.141; 95%CI, 1.005–1.297) were associated with a higher risk of transition to 24-hour care in the total AD population (Fig. 3). In the secondary analyses performed in subgroups, being unmarried (HR, 1.692; 95%Cl, 1.484–1.929) and having an informal caregiver (HR, 1.694; 95%Cl, 1.202–2.389) were associated with a higher risk of transition to 24-hour care in patients living unassisted at home, whereas higher care intensity level was associated with an increased risk of transition to 24-hour care in patients receiving home care (HR, 1.632; 95%CI, 1.181–2.259 for care level 2–3, and HR, 2.716; 95%CI, 1.992–3.702 for care level 4–5). Neurological disorder (HR, 1.680; 95%CI, 1.162–2.433) and hip fracture (HR, 1.610; 95%CI, 1.161–2.232) were associated with increased risk of transition to 24-hour care in patients living unassisted at home but not in patients receiving home care (Figs. 4 5).

DISCUSSION

In this retrospective, nationwide registry study based on 316,470 elderly (≥74 years of age) Finnish individuals, we assessed the predictors of 2-year risk of transition to 24-hour care in patients with AD. We found that 30%of AD patients were transitioned to 24-hour care facility within 2-years of diagnosis. Early start of anti-dementia medication (within 3 months of diagnosis) significantly reduced the 2-year risk of transition to 24-hour care (HR, 0.65; p < 0.001) compared to late anti-dementia medication start. The reduced risk was observed with all AChEIs and memantine started within 3 months of diagnosis. The other predisposing factors for transition to 24-hour care differed by housing status: being unmarried, having an informal caregiver, or having a diagnosis of neurological disorder or hip fracture were associated with higher risk of transition to 24-hour care in patients living unassisted at home but not in patients receiving professional home care.

Memory diseases are major predictors of institutionalization among the elderly [21 –23]. Based on a registry study in Finnish community-living individuals aged 65 and over, dementia is the strongest determinant of institutionalization from chronic conditions [4]. Transition from home to a 24-hour care facility presents remarkable econo-mic costs for society, but is also a burden for affected individuals, who lose their social and environ-mental networks, as well as to their family members [12, 24]. Previous studies suggest that delaying institutionalization improves the quality of life (QoL) in the elderly and may increase life expectancy in patients with dementia [25, 26]. Well-timed institutionalization is thus a mutual aim from both societal and individual pers-pectives.

In our incident cohort of AD patients diagnosed between 2011–2014, 26%of AD patients living unassisted at home and 38%receiving home care at the time of diagnosis were transitioned to a 24-hour care facility during the 2-year follow-up. In comparison, a meta-analysis of 42 studies from various countries showed that the institutionalization rate of persons with dementia is around 20%in the first year after diagnosis of dementia, increasing to around 50%after 5 years [12]. A previous Finnish registry study indicated that 70%of women and 55%of men with dementia were institutionalized during the 5.5-year follow-up, compared to 15%and 10%of total population [4].

Transition from home to institutional 24-hour care facility is determined by a complex network of clinical, sociodemographic, and relationship characteristics of patients with AD and caregivers, as well as primary and secondary stressors and resources available [12]. In addition, social, cultural, and economic environment influence institutionalization decisions. One of the main aims of AD treatment is to stabilize disease course and delay institutionalization. Although there is no cure for stopping disease progression, anti-dementia medication may temporarily reduce symptoms and help the patients cope everyday life and live at home for longer periods [27].

This study demonstrated that early anti-dementia medication starters (medication started within 3 months of diagnosis) had a significantly lower 2-year risk of transition to 24-hour care compared to late medication starters (medication started after 3 months of diagnosis or no medication). During the 2-year follow-up, 27%of early versus 38%of late anti-dementia medication starters were transitioned to 24-hour care facility. Lower risk for early versus late starters was observed in both patients living unassisted at home (HR, 0.584, p < 0.001) and in patients receiving home care at the time of diagnosis (HR, 0.843, p = 0.039).

Previously, inconsistent results have been published on the effect of anti-dementia medication on the risk of transition to 24-hour care. The first clinical trials showed no significant benefits with donepezil compared with placebo in 3-year risk of institutionalization [17]. In turn, a more recent study indicated that long-term treatment with galantamine and other AChEIs was associated with approximately 30%lower yearly risk of nursing home placement in patients with AD and AD with CVD [15]. The DOMINO-AD (Donepezil and Memantine for Moderate-to-Severe Alzheimer’s Disease) trial suggested that continued treatment with donepezil delays nursing home placement for patients with moderate-to-severe AD during 12 months of treatment, but make no difference during 3 years of follow-up [16]. However, none of the previous studies have assessed the effect of early versus late anti-dementia medication on the risk of institutionalization in AD patients.

Of individual anti-dementia medications, we found that galantamine was associated with the lowest risk of transitioned to 24-hour care facility, although all AChEIs significantly reduced the risk in both patients living unassisted at home and receiving home care. The result is in line with a re-contact follow-up study of randomized control trial indicating that of the AChEIs, galantamine was associated with slightly lower risk of nursing home placement compared to other AChEIs [15]. Interestingly, memantine was found to have an effect on the risk of transition to a 24-hour care facility only in the total population, but not in either subgroup alone. In comparison, previous evidence from observational studies showed that combining memantine with AChEIs significantly delays the nursing home admission in patients with AD and reduced the costs associated with institutionalization [18, 28]. Our finding may be partly explained by the fact that AChEIs are the first line treatment for mild to moderate AD patients, whereas the NMDA receptor blocker memantine is used to treat moderate to severe AD either alone or in combination with AChEIs. Thus, patients starting memantine presumably had more severe disease at the index date than AChEI starters, which is an independent predictor of institutionalization [12, 14]. The observation that memantine was more common in patients receiving home care (used by 14%) compared with patients living unassisted at home (9%) supports this supposition.

In contrast to time of anti-dementia medication start and age, other sociodemographic factors associated with transition to 24-hour care were different for AD patients living unassisted at home and receiving home care. For patients living unassisted at home, being unmarried and having an informal caregiver were associated with increased risk of transition to 24-hour care, which were not significant for patients receiving home care. This suggests that patients living unassisted at home are more dependent on support provided by their living partner, as previously reported [29]. Having an informal caregiver may in turn reflect the need for more intense home care and higher caregiver burden, which has been shown to predict institutionalization [30]. Consistently, the care intensity level (classified in three levels), was the strongest predictor of transition to 24-hour care for patients receiving home care. Income level (belonging to the highest income decile) had no effect on the risk of transition to 24-hour care in either group, likely reflecting the equal availability of tax-based public social and health care services for all Finnish citizens independent of incomes.

Interestingly, we found an association between comorbidities and the risk of transition to 24-hour care only in patients living unassisted at home, whereas none of the studied comorbidities predicted transition to 24-hour care in AD patients receiving home care. Previous studies have indicated that in addition to dementia, Parkinson’s disease, stroke, depressive symptoms, and other mental health problems, hip fracture, and diabetes are strongly associated with increased risk of long-term institutionalization independent of socio-demographic confounders and the presence of other chronic conditions [4]. In this study, neurological disorders (including Parkinson disease, multiple sclerosis, and epileptic syndromes), and hip fracture were associated with significantly increased risk of transition to 24-hour care only in patients living unassisted at home. One explanation might be that the treatment and follow-up of comorbid conditions is managed better in patients receiving home care. The finding suggests that more focus should be put on the diagnosis and care of comorbidities especially in patients living unassisted at home.

The major benefits of this registry study included large cohort size and high nationwide coverage of the cohort (approximately two-thirds of all Finnish individuals at≥74 years of age) as well as the detailed individual-level data available in the Finnish health and social care registers. This allowed us to study the effects of a broad range of variables on the risk of transition to 24-hour care, and the results are well generalized to the population level. In addition, as all Finnish citizens have equal access to high standards of public health and social care services, the bias resulting from differences in sociodemographic background variables or selection criteria in interventional studies is minimal.

Some limitations apply to the study. Detailed clinical information of AD patients (e.g., cognitive, functional, or behavioral status) and certain confounders (e.g., smoking and alcohol use) were not available in the national registries. Thus, we were not able to assess the effect of disease severity on the risk of transition to 24-hour care. In addition, all patients living unassisted at home were categorized in the lowest care intensity level 1 in the analyses (referring to milder disease), which may simplify the real situation. Another limitation is related to medications; analyses on individual medications were based on purchases of prescribed drugs (the first medication started after the diagnosis), and data on the actual use or dosing of medications were not available. We also did not have AD severity directly available and used care intensity level as a proxy-adjusting variable. However, there may remain residual confounding, especially for memantine that is indicated for patients with severe AD. Consequently, the protective effects of memantine on transition to 24-hour care may be estimated too weak. Finally, there may be underestimation of AD occurrence in the Finnish registers [31]. Those not captured are likely cases with mild cognitive impairment, and hence the generalizability of the results to this group may be limited.

CONCLUSIONS

This study brings new evidence on the critical role of early anti-dementia medication start in delaying institutionalization in AD patients. Starting any of available anti-dementia medications within 3 months of diagnosis significantly reduced the risk of transition to 24-hour care, supporting the previous evidence of the benefits of AChEIs and memantine in delaying institutionalization in AD patients. Thus, early start of anti-dementia medication should be a high priority in newly diagnosed AD patients. The results also indicated that predisposing factors for transition to 24-hour care differ between AD patients living unassisted at home and patients receiving professional care, implying that different, individualized strategies are needed to support living at home. Constructing models that take into account a complex framework of individual predisposing and protective factors is crucial in developing cost-effective and patient-oriented care for an increasing number of AD patients. Future studies should focus on detailed evaluation of the health and social care use and costs in this cohort, allowing the economic comparison of different housing options at different disease stages.

Footnotes

ACKNOWLEDGMENTS

The study was conducted in the ELSE research consortium, financed by the Association of Finnish Local authorities, Social Insurance Institution and Ministry of Welfare and Health. Additional financing was received from the Academy of Finland’s IMPRO project. Harlan Barker from MedEngine Oy is acknowledged for language review.

Authors’ disclosures available online ().

The supplementary material is available in the electronic version of this article: .

References

Wimo

, Jönsson

, Karlsson

, Winblad

(1998) Health economics of dementia. John Wiley & Sons, Chichester.

Prince

, Wimo

, Guerchet

, Ali

, Wu

Y PA

(2015) World Alzheimer Report 2015. The global impact of dementia: An analysis of prevalence, incidence, costs and trends. Alzheimer’s Disease International, London.

Linna

, Vuoti

, Silander

, Hörhammer

, Halminen

, Mikkola

, Koivuranta-Vaara

, Virta

, Koivusalo

, Ylisaukko-Oja

(2019) Impact of anti-dementia medication on the risk of death and causes of death in Alzheimer’s disease. J Alzheimers Dis 71, 1297–1308.

Nihtilä

, Martikainen

, Koskinen

SVP

, Reunanen

, Noro

, Häkkinen

(2007) Chronic conditions and the risk of long-term institutionalization among older people. Eur J Public Health 18, 77–84.

Brodaty

, Connors

, Xu

, Woodward

, Ames

(2014) Predictors of institutionalization in dementia: A three year longitudinal study. J Alzheimers Dis 40, 221–226.

Smith

, Kokmen

, O’Brien

(2000) Risk factors for nursing home placement in a population-based dementia cohort. J Am Geriatr Soc 48, 519–525.

Eska

, Graessel

, Donath

, Schwarzkopf

, Lauterberg

, Holle

(2013) Predictors of institutionalization of dementia patients in mild and moderate stages: A 4-year prospective analysis. Dement Geriatr Cogn Dis Extra 3, 426–445.

Korhonen

, Einiö

, Leinonen

, Tarkiainen

, Martikainen

(2018) Time-varying effects of socio-demographic and economic factors on the use of institutional long-term care before dementia-related death: A Finnish register-based study. PLoS One 13, e0199551.

Reed

, Happich

, Argimon

, Haro

, Wimo

, Bruno

, Dodel

, Jones

, Vellas

, Belger

(2017) What drives country differences in cost of Alzheimer’s disease? An explanation from resource use in the GERAS study. J Alzheimers Dis 57, 797–812.

10.

Linna

, Silander

, Hörhammer

, Koivuranta

, Mikkola

, Virta

, Mikkola

(2019) Iäkkäiden muistisairaiden sosiaali- ja terveyspalvelujen käyttö ja kustannukset. Suomen Kuntaliitto.

11.

Wübker

, Zwakhalen

SMG

, Challis

, Suhonen

, Karlsson

, Zabalegui

, Soto

, Saks

, Sauerland

(2015) Costs of care for people with dementia just before and after nursing home placement: Primary data from eight European countries. Eur J Health Econ 16, 689–707.

12.

Luppa

, Luck

, Brähler

, König

H-H

, Riedel-Heller

(2008) Prediction of institutionalisation in dementia. Dement Geriatr Cogn Disord 26, 65–78.

13.

Toot

, Swinson

, Devine

, Challis

, Orrell

(2017) Causes of nursing home placement for older people with dementia: A systematic review and meta-analysis. Int Psychogeriatrics 29, 195–208.

14.

Gaugler

, Yu

, Krichbaum

, Wyman

(2009) Predictors of nursing home admission for persons with dementia. Med Care 47, 191–198.

15.

Feldman

, Pirttila

, Dartigues

, Everitt

, Van Baelen

, Schwalen

, Kavanagh

(2009) Treatment with galantamine and time to nursing home placement in Alzheimer’s disease patients with and without cerebrovascular disease. Int J Geriatr Psychiatry 24, 479–488.

16.

Howard

, McShane

, Lindesay

, Ritchie

, Baldwin

, Barber

, Burns

, Dening

, Findlay

, Holmes

, Jones

, McKeith

, Macharouthu

, O’Brien

, Sheehan

, Juszczak

, Katona

, Hills

, Knapp

, Ballard

, Brown

, Banerjee

, Adams

, Johnson

, Bentham

, Phillips

PPJ

(2015) Nursing home placement in the Donepezil and Memantine in Moderate to Severe Alzheimer’s Disease (DOMINO-AD) trial: Secondary andpost-hoc analyses. Lancet Neurol 14, 1171–1181.

17.

Courtney

, Farrell

, Gray

, Hills

, Lynch

, Sellwood

, Edwards

, Hardyman

, Raftery

, Crome

, Lendon

, Shaw

, Bentham

(2004) Long-term donepezil treatment in 565 patients with Alzheimer’s disease (AD2000): Randomised double-blind trial. Lancet 363, 2105–2115.

18.

Lopez

, Becker

, Wahed

, Saxton

, Sweet

, Wolk

, Klunk

, Dekosky

(2009) Long-term effects of the concomitant use of memantine with cholinesterase inhibition in Alzheimer disease. J Neurol Neurosurg Psychiatry 80, 600–607.

19.

Alzheimer Europe. 2013: The prevalence of dementia in Europe (Finland). Last Updated: Monday 24 February 2014. Accessible online at: https://www.alzheimer-europe.org/Policy/Country-comparisons/2013-The-prevalence-of-dementia-in-Europe/Finland.

20.

Helsinki: The Finnish Medical Society Duodecim, Dementia. Current Care Guidelines. Working group set up by the Finnish Medical Society Duodecim, Finnish Neurological Society, Finnish Psychogeriatric Society, and Finnish Association for General Practice.

21.

Jagger

, Andersen

, Breteler

, Copeland

, Helmer

, Baldereschi

, Fratiglioni

, Lobo

, Soininen

, Hofman

, Launer

(2000) Prognosis with dementia in Europe: A collaborative study of population-based cohorts. Neurologic Diseases in the Elderly Research Group. Neurology 54, S16–20.

22.

Eaker

, Vierkant

, Mickel

(2002) Predictors of nursing home admission and/or death in incident Alzheimer’s disease and other dementia cases compared to controls: A population-based study. J Clin Epidemiol 55, 462–468.

23.

Banaszak-Holl

, Fendrick

, Foster

, Herzog

, Kabeto

, Kent

, Straus

, Langa

(2004) Predicting nursing home admission: Estimates from a 7-year follow-up of a nationally representative sample of older Americans. Alzheimer Dis Assoc Disord 18, 83–89.

24.

Wimo

, Jönsson

, Bond

, Prince

, Winblad

, Alzheimer’s Disease International (2013) The worldwide economic impact of dementia 2010. Alzheimers Dement 9, 1–11.e3.

25.

de Medeiros

MMD

, Carletti

, Magno

, Maia

, Cavalcanti

, Rodrigues-Garcia

RCM

(2020) Does the institutionalization influence elderly’s quality of life? A systematic review and meta-analysis. BMC Geriatr 20, 44.

26.

McClendon

, Smyth

, Neundorfer

(2006) Long-term-care placement and survival of persons with Alzheimer’s disease. J Gerontol B Psychol Sci Soc Sci 61, P220–227.

27.

Yiannopoulou

, Papageorgiou

(2020) Current and future treatments in Alzheimer disease: An update. J Cent Nerv Syst Dis 12, 1179573520907397.

28.

Lachaine

, Beauchemin

, Legault

, Bineau

(2011) Economic evaluation of the impact of memantine on time to nursing home admission in the treatment of Alzheimer disease. Can J Psychiatry 56, 596–604.

29.

Yaffe

, Fox

, Newcomer

, Sands

, Lindquist

, Dane

, Covinsky

(2002) Patient and caregiver characteristics and nursing home placement in patients with dementia. JAMA 287, 2090–2097.

30.

Wattmo

, Londos

, Minthon

(2014) Solitary living in Alzheimer’s disease over 3 years: Association between cognitive and functional impairment and community-based services. Clin Interv Aging 9, 1951–1962.

31.

Solomon

, Ngandu

, Soininen

, Hallikainen

, Kivipelto

, Laatikainen

(2014) . Validity of dementia and Alzheimer’s disease diagnoses in Finnish national registers. Alzheimers Dement 10, 303–309.