Table 3.

RUD studies with observational analyses and economic evaluation (n = 21)

Author, year, country, source number	Type of study	Setting, age PWD (age CG), dementia stage	Method / data analysis	n	Dementia severity definition	Conclusion
Michalowsky et al. (2018) Germany [72]	Cross-sectional, economic evaluation	Community-dwelling; 80.2; MMSE 22.8	OLS, cost-estimation	425 PWD, 254 dyads	MMSE	Costs of care doubled over the stages of dementia. For all cost categories, deficits in daily living activities were major cost drivers.
Darba et al. (2015) Spain [68]	Cross-sectional, economic evaluation	Community-dwelling; 76.2 (59,6); median CDR 1	GLM, cost-estimation	343	CDR	The costs of care for people with AD in Spain were substantial, with informal care accounting for the greatest part. Greater severity of the disease (CDR), increased direct medical, social care, informal care, and total costs.
Gerves et al. (2014) France [56]	Cross-sectional, economic evaluation	Community-dwelling; 79; MMSE 19	OLS, two-stage least square regression	57 dyads	MMSE	Living with the PwD, severity of dementia and hours spent on formal care were significantly associated with informal care time.
Åkerborg et al. (2016) Sweden [57]	Cross-sectional, economic evaluation	Community-dwelling; min 79; MMSE 17	GLM, cost-estimation	296	MMSE	Cost of dementia care increased with dependence and the impact of other disease indicators was mainly mediated by dependence.
Boström et al. (2007) Sweden [50]	Cross-sectional, economic evaluation	Community-dwelling and nursing home; 78; DLB and AD diagnose	Stepwise linear regression	34 DLB, 34 AD	Formal DLB and AD diagnose	Dependency in instrumental activities of daily living was strongly correlated with resource use in DLB patients.
Carter et al. (2019) Ireland [55]	Economic evaluation	Community-dwelling; 82; > 50% severe dementia g)	t-tests, cost-estimation	42	Dementia Severity Rating Scale (DSRS)g	Keeping highly dependent home-dwelling PwD is not cheap and raises questions about optimal resource allocation on the boundary of home care and residential care.
Ersek et al. (2010) Hungary [58]	Cross-sectional, economic evaluation	Community-dwelling; 77.4; MMSE 16.7	Cost-estimation	88	MMSE	Dementia related costs were much lower in Hungary compared to Western European countries. From the societal point of view, there was no remarkable difference between the costs of PwD living at home and in nursing homes.
Farre et al. (2018) Spain [59]	Cross-sectional, economic evaluation	Community-dwelling and nursing home; 83.1 (65); MMSE 15.1	Cost-estimation	287	MMSE	Cognitive impairment contributed to the cost of lost labour productivity in informal caregiver, especially in home care.
Gustavsson et al. (2011a) Cross-countrya [62]	Cross-country/cross-sectional	Community-dwelling and nursing home; 80.8; median MMSE stage was mild	OLS, cost-estimation	1222	MMSE, AD patients	ADL-ability was the most important predictor of societal costs of care in community dwellings irrespective of country and should therefore be central in the economic evaluation of Alzheimer’s disease therapies.
Gustavsson et al. (2011b) Cross-countryb [60]	Cross-sectional, economic evaluation	Nursing home; 75 (65.5); MMSE 20.7	t-tests, correlation, cost-estimation	2744	MMSE	Informal care was the most important component of costs of care in a mild-to-moderate Alzheimer clinical trial sample, and it was primarily driven by the ADL-ability.
Gustavsson et al. (2010) Cross-countryc [61]	Cross-sectional, economic evaluation	Nursing home; 76.3; MMSE 20.4	GLM with log, cost-estimation	1381	CDR, MMSE	ADL was an important determinant of care costs. Formal care service use was lower and informal care higher in Southern Europe compared to Western and Northern Europe.
Vossius et al. 2019 Norway [49]	Cross-sectional, longitudinal, economic evaluation	Community-dwelling; 81.5; CDR-SoB 6.4h	GLM, cost-estimation	257	MMSE, CDR	There is no potential cost-saving effect of day care designed for people with dementia. The use of day care did neither result in a reduced use of care nor in a delay of nursing home admission.
Wubker et al. 2015 Cross-countryd [63]	Cross-country, economic evaluation	Community-dwelling; 83.3; SMMSE mildi	OLS, cost-estimation	1661	MMSE	Transition into nursing home, increased total costs of dementia care from a societal perspective.
Handels et al. 2018 Cross-countrye [64]	Cross-sectional, economic evaluation	Community-dwelling; 78 (66); MMSE 19	OLS, cost-estimation	451	MMSE	The study found varying relationships between unmet needs and quality of life, and no association between unmet needs and care costs, although the results were sensitive to various factors.
Costa et al. (2018) Cross-countryf [65]	Cross-sectional, economic evaluation	Community-dwellig, nursing home; 83.2 (63); MMSE moderate-severe	Cost-regression	1446	Formal diagnose, MMSE	Agitation symptoms had a substantial impact on informal care costs in the community care setting.
Buylova et al. (2020) UK [73]	Cross-sectional, economic evaluation	Nursing home; (62) FAST; 6e and above	OLS, cost-estimation	79	DSM-IV	Agitation was a key driver of costs in people with advanced dementia presenting complex challenges for symptom management, service planners, and providers.
Vandepitte et al. (2020a) Belgium [67]	Cross-sectional, economic evaluation	Community-dwelling; 78.7 (67.4); median GDS stage moderate-severe	OLS, cost-estimation	355	GDS	Characteristics of the caregiver and the PwD were associated with the monthly costs of care from a third-party payer and a societal perspective.
Vandepitte et al. (2020b) Belgium [54]	Cross-sectional, economic evaluation	Community-dwelling; 78.7 (67.4); median GDS stage moderate-severe	Cost-effectiveness analysis based on modelling	355	GDS	In-home respite care program in addition to standard community-based dementia care was a cost-effective approach compared with standard community-based dementia care.
Hojman et al. (2017) Chile [53]	Cross-sectional, economic evaluation	Community-dwelling; median age range 61–80 (60.7); Mean ADL; 62.5	GLM regression, cost estimation	330	SS-IQCODE	Lower socio-economic status was associated with higher costs due to informal care and, possibly, symptom severity.
Nakabe et al. (2019) Japan [51]	Cross-sectional, online survey	Community-dwelling; 81.8 (52.2); median care-need level: 2	χ2 automatic interaction detection (CHAID) analysis	1383	Own estimation of care-need levels based on function	Informal care costs were related to caregivers’ employment and cohabitation status rather to the situations of people with dementia. Out-of-pocket payments for long-term care services were related to care-need levels and family economic status.
Nakabe et al. (2018) Japan [52]	Cross-sectional, Online survey	Community-dwelling, nursing home; 82.5 (51.9); median care-need level: 2	Descriptive analyses	3841	Own estimation of care-need levels based on function	The inclusion of informal care costs reduced the differences in total personal costs among the residence types.

^a Sweden, Spain, UK, US, ^b Australia, France, HK, Italy, Netherlands, NZ, Singapore, US, ^c Sweden, Denmark, UK, Belgium, France, Germany, Switzerland, The Netherlands, Italy, Spain, Greece, Romania, ^d Estonia, Finland, France, Germany, the Netherlands, Spain, Sweden and the UK, ^e Germany, Ireland, Italy, the Netherlands, Norway, Portugal, Sweden, UK, ^f Estonia, Finland, France, Germany, Netherlands, Spain, Sweden and England, ^g Carter et al. [55], page 5, ^h SoB - sum of boxes. Vossius et al. [49], page 6, Table 1, ⁱ SMMSE standardized mini-mental state examination, Wübker et al. [63], page 696