An effectiveness and cost-estimation model for deploying assistive technology solutions in elderly care

ABSTRACT Background Deployment of modern assistive technologies is one of the major trends. The main objective is the provisioning of the effectiveness and cost-estimation model for deploying assistive technology solutions in elderly care intended for evaluation and showing specificity to the costs and associated benefits of providing smart technological solutions for seniors. Methods The model uses demographic projections taken from Eurostat for EU countries and the disability incidence from the annual report of the Ministry of Labour and Social Affairs of the Czech Republic as an input. The model was implemented in the software Stella Professional dedicated to system dynamics modelling. Results In relation to the combination of five assistive devices for the elderly the optimal solution the cost savings are 37.8% or182 billion CZK, cumulatively in the simulated time period 2021–2060. The model allows adjustments regarding price levels of various countries through the caregivers’ wage. Obtained results showed robustness of the model to this uncertainty as percentage savings varied only from 37.6% to 39.8%, a mere 2.2% difference. Conclusion With respect to the ongoing demographic transition, the need to employ smart device solutions should further increase and their price could decline. The highest savings will come from countries with higher wages for carers.


Introduction
The demographic composition of many countries in the world has changed because of increasing population of the elderly as a result of advancements in lifestyle, medicines, and healthcare.This imposes additional demands on healthcare systems in terms of people and future costs.Projections by Eurostat suggest that by 2040, the number of people aged over 65 in the European Union (EU) will rise by approximately 7%, while at the same time, the overall population will decrease by 1 million [1].This disharmony introduces difficulties in providing adequate elderly care in the future.A possible remedy for the stated difficulties is to enable the elderly to remain independent and healthy for as long as possible [2] using improved technological solutions and products.
Research on the deployment of smart technological solutions is scanty, and data are limited to the potential health and socioeconomic benefits of assistive technology [3].Although assistive technology is recognized as a human right, only 5-15% of the population actually needs it [4].However, access to assistive technology can have many health and socioeconomic benefits.Assistive technology supports the desire for aging in place and independence of the elderly, enabling them and their caregivers to improve their quality of life (QoL).To achieve the accessible, compassionate, and ethical development of assistive technology, it is important that the technologies remain usable and affordable [5].In combination with other critical efforts to help the elderly and their caregivers, assistive technology can provide solutions hitherto neglected in global health and socioeconomic discussions [4].The World Health Organization (WHO) and United Nations International Children's Emergency Fund (UNICEF) have also, among other things, recommended in their latest report on assistive technology [6] that it is important to invest in data and evidence-based policy, especially outcomes in terms of human rights and QoL, their families and community or country at large, affordability and availability of assistive technology, service delivery models, financing models, cost benefits, and cost-effectiveness of assistive technology from the perspectives of users, programmes, countries, etc.Additionally, they recommend investments in research, innovation, and enabling ecosystems.
Assistive technology has so far been assigned a low priority in the research agenda because studies have been mostly small-scale and inadequate to draw serious conclusions.Therefore, governments and politicians have not considered them in development and management strategies.This study aims to provide new perspectives to relevant stakeholders on the need to extend their efforts to provide and implement assistive technology.The main objective is to provide an effective cost-estimation model for deploying assistive technology solutions in elderly care for evaluation and showing specificity to the costs and associated benefits of providing smart technological solutions for seniors.Therefore, the main contribution of this work is the model that calculates the benefits of economic improvements in quantitative terms, that is, the investment required to build up the delivery of assistive technology and the costs associated with its usage.In other words, our intention is to conduct a cost-effectiveness analysis of the deployment of smart technological solutions for seniors and then use the findings to plan and organize larger studies, encourage technology market leaders to invest, and include finding-based directions in national strategies.Additionally, by examining the effectiveness and benefits of technology usage in this context, this study contributes to the society and research community in several ways.By stimulating research in the area of assistive technology deployment in elderly care based on cost-effectiveness, this study contributes to the ultimate result, that is, the improvement of QoL of the elderly and their relatives and caregivers.Similarly, by investigating the utilization of assistive technology for seniors' care through the model from an economic viewpoint, it raises awareness and contributes to the discussions on these issues, and underlines the benefits to wider society, technology markets, healthcare systems, governments, etc.If considered, this view can contribute to a decrease in the current essential social problems, such as inadequate healthcare of seniors, shortage of medical and caregiving staff, fatigue of relatives and caregivers, inadequate wages etc.Additionally, an important benefit relates to supporting the technology industry embrace this new field to produce profitable solutions.If this industry becomes interested, the available solutions would become affordable, which will encourage governments to focus on technology development, wage-related issues, and other healthcare challenges.
To answer this question, this study uses the methodology described in the following structure: Section 2 provides a literature review, including article sources and search strategy, article selection process, article review and data extraction, and analysis of selected articles.Section 3 discusses the research methodology, study design, model description, data from Czech facilities used to verify the model, and technology selection.Sections 4 and 5 present and discuss the obtained results.Section 6 concludes the study and provides directions for future research.

Article sources and search strategy
An article search was undertaken in August 2022 to identify peer-reviewed articles published in English in the last five years (2018-2022).The databases searched included Web of Science (WoS), PubMed, and Google Scholar.Article extraction was performed using the following search phrases alone or in various combinations using logical operators of 'AND' and 'OR': 'cost estimation', 'cost effective analysis', 'cost benefit analysis', 'cost utility analysis', 'cost analysis', 'economic evaluation', 'economic assessment', 'assistive technology', 'assistive product', 'assistive device', 'elderly care', 'seniors care', 'older adults care'.

Article selection process
Titles and abstracts were analyzed by two independent researchers to select articles that met the inclusion criteria.They compared the titles and abstracts with the inclusion and exclusion criteria to achieve a common opinion.The inclusion criteria were as follows: (i) articles that included any type of economic evaluation of assistive technology solutions in elderly care; (ii) articles that were peer-reviewed and published in English within the last five years (2018-2022); and (iii) articles that used same research methodology.The exclusion criteria were as follows: (i) non-English articles; (ii) articles that werewithout an abstract; (iii) articles published before 2018; (iv) research published in conference proceedings, books, book chapters, and master's and doctoral theses; (v) research that did not include any type of economic evaluation of assistive technology for elderly care; (vi) research focussing on industry and production of assistive technology solutions; and (vii) articles with inadequate information for article categorization.
When consensus regarding abstract suitability could not be reached, the article was further analyzed by reviewing the full texts of the selected articles.Following data abstraction from the selected articles, a new researcher independently reviewed 20% of the randomly chosen articles.In case of lack of consensus on the suitability of articles, opinion was sought from a second new researcheran interdisciplinary expert who gave a final decision on whether or not to include the article in the literature review.

Article review and data extraction
An inductive approach was adopted for this review.A summary of each article is provided below; throughout this process, the following items were systematically extracted: first author, journal and impact factor, year of publication, country, objective, application area of assistive technology, economic evaluation, main findings, and influence of assistive technology.The extracted items provided input data that justified the proposal of a cost-estimation model for deploying assistive technology solutions in elderly care.

Analysis of selected article
An analysis of the selected articles is summarized in Table 1.The intention was to identify studies that considered the economic evaluation of assistive technology solutions for elderly care in high-impact journals.The considered period was limited to five years to make the review manageable and focus on the latest relevant studies.To achieve diversity and inclusion, we reviewed studies performed in various countries, such as Australia, Canada, the Netherlands, Spain, Sweden, the United Kingdom (UK), and the United States of America (USA).The objectives of these studies are summarized, indicating various assistive technology solutions as the most innovative developments.It can be assumed that in recent years, the most frequently researched technologies have been the most important areas for researchers and research funders.Tun et al. [7] identified 10 areas where assistive technology devices can be used in elderly care: aged care monitoring, chronic patient healthcare monitoring, clinical applications, emergency conditions, mental health, movement disorders, rehabilitation, preventive measures, accessibility to healthcare services, and accessibility for caregivers.Therefore, we mapped selected studies in terms of application areas, providing indications about those that appeared promising for further research, that is, mental health/dementia [8][9][10][11] and rehabilitation [12][13][14].
The included studies reported the main findings related to the economic evaluation of assistive technology solutions in elderly care.Economic evaluation in elderly care can be performed using several methods [19], such as cost-effectiveness analysis (CEA), costutility analysis (CUA), and cost-benefit analysis (CBA).CEA utilizes a single natural parameter to indicate the effects of health intervention (e.g.number of prevented strokes), whereas CUA combines different aspects in a virtual parameter (e.g.quality-adjusted life year (QALY)).Another option to combine the intervention's effects is CBA, which values effects based on the estimation of willingness to pay (WTP).Albala et al. [20] emphasized that the most important economic evaluations were based on CEA and CBA.Therefore, the economic evaluations employed in the selected studies included four CEAs [8,[10][11][12]18] and three CBAs [9,16,17].However, these studies have reported inconsistent results.Howard et al. [8,10] found that assistive technology was neither effective nor cost-effective.In contrast, de Batlle et al. [17] reported that the implementation of various assistive technologies reduced health costs and proved to be cost-effective.
Compared to CEA [21], multi-criteria decision analysis (MCDA) allows local-level decision makers to use a wider measurement of effectiveness by including the preferences of multiple stakeholders and outcomes beyond health and longevity [22,23].This additional information helps in decision-making in elderly care.However, meaningful results on its effectiveness cannot be reported because of shortage of high-quality studies [24].There are several reasons for this finding; one is that only a few studies have focussed on health economic evaluations that compare the benefits of technologies with costs.Another reason is the immaturity of technologies that are often not implemented in real-life environments.This makes it difficult to precisely estimate the future cost of a technology because it is expected that future costs will be lower than current costs [24].
Information about the influence of assisted technology was extracted from the considered studies because, traditionally, they will not be considered important in comparison to the cost savings or will be reported as part of the main study findings.The influence of assistive technology refers to the consideration of how assistive technology interacts with the user, family, or social care/healthcare.This influence can be measured at various conceptual levels, ranging from functional performance to QoL [25].The economic evaluations performed by the analyzed studies were mainly focussed on assessing how assistive technology reduced the burden on caregivers [11,14] and improved individual QoL [8,[10][11][12][13][15][16][17].This is in line with the findings from [26] that indicate that many assistive technologies currently proposed are targeted to prolong and support the independence of older adults and provide help for formal and informal caregivers.
The effectiveness of assistive technology is crucial when planning and implementing elderly care [27].To achieve effectiveness, assistive technology needs to contribute to various aspects of seniors' lives [28], such as material living conditions, health, education, productive and valued activities, governance and basic rights, leisure and social interaction, natural and living environments, economic and physical safety, and the overall experience of life.In other words, assistive technologies should be designed to be personalized and satisfy multiple smart aging components (i.e.behaviour and social aspect, technologies, and medical care according to Song et al. [29]).These, in combination, contribute to multiple QoL dimensions, and jointly accomplish the goal of improving seniors' QoL.Assistive technologies should be incorporated into the idea of multidimensional and multidisciplinary personalized approaches for senior individuals, which lead to the improved adoption of assistive technology [30].
In addition to this QoL viewpoint on the effectiveness of assistive technologies, which is inclusive per se as QoL covers multiple aspects (e.g.life-long learning [31]), it is important to understand the challenges for assistive technology development.For instance, Zhang et al. [32] summarized the challenges of China's smart homes for elderly care.Additionally, Huter et al. [19] identified challenges and recommendations related to the economic evaluation of digital nursing technologies.These challenges are comparable to the challenges of assistive technologies in elderly care, indicating that conducting an economic evaluation of such technologies is important but complex.

Study design
This study presents the results of the effectiveness assessment of a smart technological (assistive technology) solution for seniors, which will help optimize the limited human resources in the field of care for the elderly.The model setup was as follows: . Specification of activities with which seniors are normally assisted using the characteristics of ADL and IADL .Specification of existing technological solutions and in connection with IADL .Characteristics of the population in developed countries .Finding evidence of costs related to IADL .Setting the links of the model to determine the conditions suitable for effective technology deployment Different scenarios were created in the model to answer the following questions: . Which price covers the costs associated with new technology solutions? .Where is the break-even point for chosen solutions in the market?

Model description
The model was implemented in Stella Professional, version 1.9, developed by the Independent School Entrance Exam (ISEE) Systems, with the aim of modelling the dynamic behaviour of the market associated with the increasing adoption of new solutions for ADL support.Owing to the lack of reliable data, this part was eventually skipped, and the prices of assistive devices were exogenous variables along with the rest of the input data.The complete equation dump is included in the Appendix.
In its current version, the model evaluates the most economical combination of available assistive devices for each age group and disability class (together referred to as category) based on their price and capability before computing the overall expenses for provided care at the national level and potential savings by adopting the suggested solution.
The model uses demographic projections derived from Eurostat, with age cohorts ranging from 0 to 100 + by one year.These data were then partially summed into age groups 0-17, 18-59, 60-69, 70-79, 80-89, and 90 + .Population data for the Czech Republic and other EU countries were available as inputs.
From the annual reports of the Ministry of Labour and Social Affairs of the Czech Republic, the total number of people receiving disability compensation was obtained for the above age groups and four disability classes (I-IV).Compensations are provided to people who, due to enduring health problems, require the assistance of another person in performing basic ADLs.Based on the number of activities where aid is required, the disability classes are defined by law to determine the value of the compensation; there are four levels of dependence: grade I (slight dependence), grade II (medium-heavy dependence), grade III (heavy dependence), and grade IV (total dependence).
The relative fractions (percentage) of people for each disability class and age group were considered constant throughout the simulation.Multiplying them by the population data for a given year, we computed the total number of people receiving care in each category (Table 2).
The required hours of care per person for each category and the ADLs of interest (toilet, hygiene, food, movement indoor/outdoor) were estimated by averaging the data provided by the home-care services (description in Chapter 3.3.1).This yields a 3D array indexed by age group, disability class, and ADL, which is then used to calculate the total hours of care and overall expenses by means of built-in array functions, including element-wise arithmetic and partial sums (along one or more dimensions).
Assistive devices can replace a portion of the overall hours provided by caregivers.Each device is described by a set of coefficients between zero and one, representing its efficiency (the approximate fraction of care hours it can handle) for a given ADL.The economic parameters of the device are price and lifetime (in total hours of service).
If multiple devices have a non-zero value for the same ADL, the maximum of these values is used.Owing to the overlapping capabilities of the devices, the real value of a device significantly drops when a device with similar characteristics is already part of the combination.
The number of replaceable care hours and the corresponding financial savings are computed for all combinations, yielding a winner in each category.Break-even costs were evaluated as an auxiliary value.Replaced_Hours_per_ADL (3D) [Age_Group; Dis-ability_Class; ADL] -Hours of human care per person and ADL that can be saved using the winning combination of SDs.
Selected_Combo_Yearly_Costs -The actual yearly costs of the combination of devices selected by the user via the interface window.
Total_Costs_without_SDs -The overall yearly costs of care provided by human caregivers.
Total_Savings -The overall yearly savings when deploying the most economical combination of SDs in each category.
Percentage_Savings -The relative savings over the entire simulation period.

Sector description Demography Sector:
The population data from Eurostat projections were summed into six groups.The result was multiplied by the disability incidence, yielding the overall number of people for each age group and disability class that was used as an input in the care sector (Figure 1).
Based on projections for the Czech Republic, the total number of people in the working age (cohort 18-59) is expected to drop from 5.98 million to 5.03 million by 2050, whereas the 60 + cohort is expected to grow from 2.88 million to 3.81 million during the same period.The total number of people receiving disability benefits is expected to grow from 385k to 609k.These preliminary results depend on long-term population changes and the actual percentage of people receiving care (Figure 2).
SD_Evaluation Sector: Based on user input, all possible combinations of available SDs were examined.The combination efficiency for a given ADL is considered as the maximum of the individual efficiencies of the comprising SDs.Multiplied by the yearly care hours per person in each category, this provides the total number of replaceable care hours for the considered combination.The associated yearly operating costs are then compared with the corresponding wages of a human caregiver, yielding break-even yearly costs and potential savings.The winning combination differs between categories, as more expensive solutions become viable with increasing hours of care required.Yearly savings in terms of both hours and expenses are used as inputs in the care sector (Figure 3).Source: Annual report of Ministry of Labour and Social Affairs for 2019 [33].

INTERNATIONAL JOURNAL OF HEALTHCARE MANAGEMENT
Care Sector: Multiplying the average care hours per person and ADL by the total number of people in the respective category, we obtain the overall hours of unassisted care per category, and subsequently, its cost.Applying the best available combination per category, as given by the SD_evaluation sector, the modified care hours along with the associated costs are computed.The results are then summed over all categories, yielding the total yearly costs for both assisted and unassisted care, and finally, the perceptual savings throughout the simulation period (Figure 4).

Data
Data from selected social care facilities in the Czech Republic and selected technologies with a direct link to monitored ADL actions were used for the specific setting and verification of the functionality of the model.

Data from selected social care facilities in the Czech Republic
In 2020, data were requested on the care for the elderly in their home environment.The data were provided by two facilities with a total of 609 clients (194 men and 415 women).The information provided includes age, gender, degree of dependence (I-IV) according to MPSV [33] legislation in the Czech Republic, and diseases/restrictions.Furthermore, the services provided were recorded, including purchases, routine cleaning and maintenance of the household, water delivery, accompanying adults, one-time import/ removal, assistance and support in serving food and drink, assistance in dressing, assistance in using the toilet, assistance in orientation in the space, assistance in moving to a bed or wheelchair, assistance in preparing and serving food, assistance in personal hygiene, assistance with basic hair and nail care, washing and ironing bed or personal laundry, preparation and serving of food and drink, preparation and serving food  individually, renting compensatory aids (mechanical wheelchair, walker, toilet chair, etc.), regular cleaning, large cleaning, heating in the stove, massages, and pedicure (irrelevant to us) (Table 3).
The parameters by which these actions were categorized and subsequently financially valued were: the number of hours, number of actions, number of visits, amount of payment per year, and distance travelled.
In terms of the prepared model, the above information is used in four main categories of activities, which are described by the sum of the hourly subsidies for the corresponding actions: o personal hygienehelp with personal hygiene, help with basic hair and nail care o Toilethelp with using the toilet o Movementhelp with orientation in space, help with moving to a bed or wheelchair, one-time import/removal, shopping, and errands o food (cooking and serving food)

Rate of replacement of the caregiver with a smart device
To link information related to caregivers' time spent on individual tasks and the functionalities of technologies for seniors, a non-standardized interview was conducted with seven people in the field of social care.The main task was to determine the extent to which they considered the given technology to be beneficial for the given operation (Table 4); this was expressed in percentages.The interviews were conducted in February 2021.The technology was introduced using a short video of the product and interviews were conducted.Table 4 presents the averages of all the responses for each cell type.
Five technologies with a direct link to ADL activities were selected for the case study; thus, it was possible to express the link to the observed variables in the model.A description of these technologies is provided below.
A four-wheeled upward walker is a mobility and standing support roller, a four-wheeled platform that supports seniors at the arm level during walking and can be folded to allow for comfortable sitting and rest when needed.Robotic arms are among the most useful ADL wares currently available, and are especially useful for feeding assistance; they have been studied to improve the psychosocial ability of users [34].In the realm of intelligent hygiene management, the robotic shower system performs care functions of assisting patients with bathing.The HSR is a wheeled platform with an arm that can perform several features, including communication with relatives and caretakers on an attached display, picking, lifting, etc. [35,36].The hip-attached battery-powered walking assistive device is lightweight and supports the user at the knee and hip levels, partially replacing the burden on the legs to facilitate safer and faster walking.All selected devices have commercial models or are currently in development, and their true names are concealed to avoid advertising references in scientific articles.

Model assumptions
The presented model simulation results were obtained using the following parameter values.For caregivers' wages, we assumed a gross wage rate corresponding to the Czech Republic conditions of 250 CZK/h.Regarding device costs, we had commercial pricing data only for the four-wheeled upward walker and hip-attached battery-powered walking assistive device; for other devices, we assumed their costs along with device longevity (assumed to be five years of usage for all devices).The relevant data are listed in Table 5.
The rest of the parameters used for the simulation were presented previously, as they are part of the model structure and modelling approach; for example, ADL replacement rates in Table 4.

Cost development analysis
Given the current demographic trends in the Czech Republic, the expected costs for elderly care (work of caregivers) in disability categories I-IV are expected to rise from the current 8.31 billion CZK to 16.6 billion CZK (nearly doubling) by 2060not including costs for medication or other than the wage costs of institutions providing care.
First, we analyzed the savings potential of each device separately.Widespread use of the four-wheeled upward walker is estimated in the disabled elderly population, with a total estimated cost savings of 29%, or 139 billion CZK, for the total cumulative savings.The robotic arm also has significant potential, expressed by a 14.2% savings potential, or 68.5 billion CZK of cumulative savings.The HSR is also significant because of a wide range of possible addressable ADL needs, with the result of 8.9% savings potential, or 43 billion CZK total savings.The hip-attached battery-powered walking assistive device results are weaker owing to its high specialization across ADL dimensions and relatively high assumed price, with the result of 5.1% of estimated savings, or 24.6 billion CZK, cumulatively.Robotic showers and smart devices capable of addressing hygiene ADL needs results were the worst because of their high specialization and price, making them suitable for institutional use only (results were 2.5% of total savings potential, or 12.1 billion CZK, cumulatively).
Full-scale solution: In the following analysis, we focus on the optimal solution from the perspective of the cost savings in the system.All presented devices are used for optimization (without considering possible synergies between employed smart devices), and each device can be separately attributed to each age cohort and disability class of the elderly (Table 6).
The maximum realizable cost savings are achieved with the predominant usage of the four-wheeled upward walker and robotic arm.A robotic showera smart device capable of addressing elderly ADL hygiene needswas employed in the optimal solution only in the oldest age cohort with the highest dependability.
The usage of a concrete combination of devices is given by the break-even price, which is the price of work of the caregiver given by his/her wage and the combination of ADL needs in a given age group and disability class that he/she is required to serve.The break-even prices thus calculated are shown in Figure 5.
The unique combination of devices depicted in Table 6 needs to satisfy two conditions: the price of their usage (purchasing price/lifetime) has to be lower than the work of the caregiver; and they have to supply the needed ADL replacement (that is why four-wheeled upward walker is typically not used in the first disability class despite its low cost).
The highest cost (49.2k,first line in Figure 5) has a combination of four-wheeled upward walker-robotic arm-robotic shower, which is used only once in the model.A combination of four-wheeled upward walker-robotic arm is used 8x times in total (see Table 6) with a total annual price of 6.2k (second line in Figure 5).The four-wheeled upward walker is employed 8x times by the model, owing to its low yearly price of usage, that is, 1.2k.
Optimal solution realizable cost savings are, 37.8% of total (or, 182 bil.CZK [173 bil.-191 bil.CZK for ±5% change in proportion of disability categories incidence in each age group, or ]), cumulatively in the simulated time period of 2021-2060, their expected time development is in Figure 6.

Sensitivity analysis
As we could not obtain pricing data regarding robotic showers and HSR, our estimates could be questioned.Thus, we employed a sensitivity analysis where the prices of the aforementioned devices were varied in the range of ±50% (42.5k to 107.5k CZK for robotic shower, 25k to 75k CZK for HSR).
The obtained results showed the robustness of the model to this uncertainty, as percentage savings varied only from 37.6% to 39.8%a mere 2.2% difference.Figure 7 shows this in absolute terms.
The discrepancies were caused solely by the robotic shower, which was employed twice at its lowest cost.Thus, we can conclude that in the context of the presented model, robotic showers and HSR can be sufficiently replaced by different, cheaper devices, or their usage by the elderly at home will probably never be economical.

Discussion
Knowledge is crucial to raise awareness about assistive technology in elderly care and to allocate funding to improve its effectiveness [6].To this end, this study proposes a model to evaluate the costs and associated benefits of providing assistive technological solutions for seniors.Establishing such a cost-estimation model can support decision-making across sectors and countries, providing the basis for differentiation from other studies.Management in healthcare should seriously start thinking about the abundant deployment of technologies.Because their contribution, even in view of the labour shortage and the growing number of seniors, is demonstrable.The key question that needs to be thought about and decided upon is: where and what technologies to deploy first (taking into account their prices, the ability to use themmore in residential facilities than at home).Deeper analyzes in this direction should indeed follow.Funding-related issues should be addressed immediately (especially in public health facilities) as the amount of funds allocated in a given country for a given period in relation to the facility or the care provided.These mentioned points should be answered promptly.Compared to relevant studies, this study has three main advantages.First, although qualitative studies are gaining momentum (e.g.[19,27]), this study is characterized as a quantitative study that uses standardized measures and analyzes to evaluate the effectiveness and cost-effectiveness of assistive technology deployment in elderly care.Qualitative studies have their own value, but the quantitative approach allows for generality.Second, the cost-estimation model proposed in this study has a broad range of applicability, as it is independent of assistive technology and its application area, in contrast to related studies that are limited to a particular application area (e.g.[11,12]).Third, this study provides a detailed description of the proposed cost-estimation model in terms of input and computed parameters and the evaluation process that highlights its broad practical applicability and easy scalability, as well as proper consideration of economic evaluation.
Furthermore, the results of this study indicate that smart assistive technology devices need to be considered as a means to mitigate the challenges and issues of the aging population in this particular  country, the EU, and worldwide.This is not surprising given that they have been recognized in numerous strategic plans and projects [4,6,37] because the challenges faced by the senior population are ubiquitous.The second point to discuss is that one must identify a means to solve two parallel social trends and challenges with assistive technologies worldwide: an aging population and a shortage of healthcare workers.Smart assistive technologies should be accepted and supported, not because of low prices and cost savingsgiven that prices might gradually decreasebut because of labour shortage [38,39] and rise in the care-requiring elderly population [6].The benefits of applying smart assistive technologies in senior centres, as compared to the home environment, would be more profound, especially in the context of Covid-19.
Moving from global to particular cases, and having analyzed the cost-saving attributes of different ADL devices, it is clear that a few devices may remain unaffordable for a certain section of the elderly population or disabled persons in the Czech Republic.The implication is that demand will likely increase for cheaper devices such as the four-wheeled upward walker and assistive robotic arms.
In this study, we solely focussed on homecare for the elderly, without considering the institutional setup of care.We used several devices: from easily available and cheaper mechanical devices to under-development robotic devices.
Their simulated deployment was dated to start in 2021 and end in 2060.In the model, after setting a combination of suitable devices, their prices, and population structure, which depends mainly on the size of the target group, a calculation was performed showing the possibility of cost-efficient replacement of caregivers' time spent with seniors.The calculation shows the possibility of potential savings and costs with respect to caregivers' salary.
Specific settings for the combination of the five devices listed in this case study (Table 6), their ability to replace some of the caregivers' actions (Table 4), and their price settings yielded total savings of 37.8% (or 182 billion CZK), cumulatively, in the simulated period of 2021-2060.
It is obvious that the use of devices such as robotic showers or HSR would be much more cost-efficient in institutional settings.However, we lack data related to the degree of replacement ability of employee/caregiver services in the institutional context; thus, our model does not consider this.Therefore, the results should be interpreted with this limitation in mind, as even higher cost savings can be achieved in the institutional set-up.
Another limitation is the lack of data on the production/sales costs for many devices in our selection.Even with currently sold devices, such as a hip-attached battery-powered walking assistive device, where the price is known, it is likely that the price will substantially drop in the case of a successful market launch and increased scale of production.In this case, the model results can be viewed as pessimistic, as they reflect current prices (or their estimates) only.
The model allows adjustments to the price levels of various countries through caregivers' wages.The approximate difference between Czech caregivers' wages and wages in Germany, Norway, and Britain is approximately 30%.With higher caregivers' wages, the model yields higher estimated cost savings than in the Czech Republic, with increased usage of smart devices for elderly assistance.
The model currently uses Table 2 of the elderly population ADL needs for the Czech Republic, while allowing the import of population data and price levels (through adjustable caregivers' wages) for another country, which might distort results.Although it would be trivial to turn the ADL needs table into yet another input quantity, the system of disability classes is rather unique for the Czech Republic, which might require additional pre-processing of the respective data to fit into the model.
Generally, future research should focus on the collection and implementation of international data.Market prices from producers should be considered, and their use in the institutional environment should be evaluated.

Conclusion
This study aimed to determine the cost-effectiveness of smart assistive technology for seniors.For this purpose, we developed a model to calculate the effectiveness of applying this technology and conducted a case study with five assistive devices in the Czech Republic.The contributions of this study are threefold: first, we propose a model that may be used in other countries as well for similar research activities that, as discussed earlier, are not sufficiently represented.Second, we recognize the need for further studies to treat the effects of assistive technology implementation at both the individual and national levels.The third is the conclusion that assistive technologies cannot be considered only in terms of financial effectiveness, but in a much broader sense, given that their benefits can be short-and long-term and direct or indirect.
The benefits of assistive technologies are reflected in health improvement, social inclusion, and economic returns.Assistive technologies improve the health and well-being of the elderly and also increase access to wider healthcare services.This improves the inclusion of the elderly in society by lowering the risk of loneliness and encouraging a sense of belonging.Providing assistive technology to the elderly will improve employment and productivity by helping the users overcome aging-related challenges, thereby enabling them to fully work until retirement.Thus, including smart assistive technologies in everyday life should improve QoL directly for actual users, and indirectly for caregivers by reducing their burden.
Assistive technologies are not only cost-efficient in terms of actual care costs, but their long-term impacts on users' QoL, along with their potential to deal with population aging and the associated healthcare provider shortage, are highly beneficial to society as a whole.To achieve some of the benefits, countries should improve access to assistive technologies by implementing specific actions, such as public support, policy and legislation changes, and research and development support.These activities will require the inclusion of all stakeholders (e.g.public sector, private sector, donators, etc.) and various research methods to facilitate a comprehensive vision and resource prioritization.

3. 2 . 1 .
Input fields Population (2D) [Simulation_Year; Age] -Eurostat population projection starting from 2019 by 1-year cohorts without sex distinction.Disability_Incidence (2D) [Age_Group; Disabil-ity_Class] -Relative fraction of people in each disability class for each age group.SD_Efficiency (2D) [SD; ADL]efficiency of available smart devices.Values between 0 and 1 represent the extent to which the given SD can replace a human caregiver for the respective ADLs.SD_Lifetime (1D) [SD] -Expected lifetime in years for a given SDs.SD_Purchase_Cost (1D) [SD] -The purchase costs of given SDs.Caretaker_Hourly_Wage -The average hourly wage of caretakers for the country in question.SD_Availability (1D) [SD] -Boolean array provided by the user, indicating which SDs are available for deployment.SDs with zero values were excluded from the processed combinations.Hours_of_Care_per_Person_and_ADL (3D) [Age_Group; Disability_Class; ADL] -Typical hours of care for each ADL required by a person in a given age group and disability class.

Figure 2 .
Figure 2. People in given age cohorts (left Y axis, millions) and the total amount of people receiving disability benefits (right Y axis, thousands).

Figure 4 .
Figure 4. Care costs and savings.

Figure 5 .
Figure 5. Breakeven Estimated Yearly cost for selected assistive device solution and population group [Age group; Disability class (I-IV)].

Figure 6 .
Figure 6.Cost saving with the optimal device assignment across cohorts and disability classes, billions of CZK.

Table 1 .
Summary of studies including in review and comparative analysis.

Table 2 .
Number of people per age group and disability class (in thousands) for the Czech Republic.

Table 3 .
Average hours of care per disability class and ADL (age group 18-90+).
Age group [years] Disability class membership [-]Time needed for daily activities assistance [hours per year] Source: own research in home-care services.

Table 4 .
Individual SD efficiencies for considered ADLs.

Table 5 .
Price estimate of devices.

Table 6 .
Smart device assignment in respective age cohorts and disability classes, where X denotes device usage in an optimised model scenario for respective age group and disability class of the elderly.