Assessment of climate and social impacts:
the case of Finnish home care services
This part introduces a novel methodology to assess the environmental, especially the climate impacts and social impacts of digital healthcare and care services. We also present the results of our quantitative and qualitative assessment concerning climate and social impacts of digital home care services in Finland, centred on medicine robot services for older clients.
We show how and under what conditions distance spanning solutions in health care and care can contribute to the green transition. We also give practical guidance for future impact assessments.
Case region and services
The empirical data on medicine robot services were collected in the region of Päijät-Häme in Southern Finland (see endnote 3). The region is rather sparsely populated with potentially long distances to services. At the time of the data collection (2022), home care services were provided by Päijät-Häme Joint Authority for Health and Wellbeing (currently called Päijät-Häme wellbeing services county; hereafter the County), the unit for elderly care services and rehabilitation. This regional organisation provides services for the more than 200,000 residents of the region with its 7,000 employees. Päijät-Häme was selected as one of the five model Nordic regions of collaboration in the healthcare and care sector (Nordic Welfare Centre, 2022).
Part of the home care services is provided remotely with the help of distance spanning solutions. The remote care and technology unit, Severi, serves regular home care clients in the region using medicine robot services and other services. Severi’s staff comprises both nurses and assistant nurses. The Päijät-Häme region has been a pioneer in remote care in Finland. In 2022, there were 257 medicine robot clients (15.3 percent of all home care clients). Medicine robot services have been provided since 2016 and are therefore not new in the region, although environmental sustainability and climate impacts have not been studied.
Medicine robots are available to the region’s home care clients free of charge. Home care professionals generally refill the robots every 1–2 weeks. In the event of potential disturbances, such as power outages, the device sounds an alarm that is directed to a care professional. The device also sounds an alarm if the client does not take the medicine offered or if there is an attempt to break into the device. The medicine robot services function independently or can complement other home care services. The brands used in the region are Evondos (from 2016) and Axitare (from 2020) (see photo 1). A medicine robot requires mains power but not a fixed internet connection, thanks to its built-in mobile modem. To access the backend, as well as the robot itself, employees have a tablet, computer or mobile phone with an internet connection, which the County acquires (itself). Apart from the devices, the County purchases the overall service including maintenance from a technology provider. One of the medicine robots uses medicine bags, while the other uses dose cups. The medicine robot services aim to replace one or more of the care professionals’ home visits, though not necessarily all of them.
 |  |
Photo 1. Medicine robots (photos: Evondos, on the left, and Axitare, on the right).
Data collection
The data collection served the quantitative assessment method, including the development of a calculation model, and the qualitative assessment. The data were collected with the help of interviews with the County staff and the two technology suppliers, as well as documents provided by the County, such as log information related to the use of the services, evaluation reports, planning documents and statistics, plus annual reports on the use and share of services in the region and the development of service usage. The main knowledge needs identified for the data collection were as follows:
- From the perspective of clients, care professionals and the service system:
- Detailed description of the service (e.g. devices used, architecture of the backend system, daily organising of the service, work environments/spaces)
- Reasons for introduction/digitalisation of the service (economic and/or related to service quality), impacts (both targeted impacts and other impacts in a broad sense)
- Information on how the services had been organised in the past and how the digitalisation of the service changed how the various participants operated
- Information on patient/client satisfaction
- From companies (technology suppliers): description of the service from the company’s perspective, matters related to manufacturing materials, production, electricity consumption and materials/parts recycling/life cycle, backend services and network usage.
The interviewees in the case of the medicine robot services:
- 9 interviewees representing the county staff (assistant nurses, a registered nurse, an immediate supervisor, management representatives)
- 2 interviewees representing the technology suppliers
The aim was to collect as broad information as possible on both direct and indirect impacts. The content of the interviews was tailored according to the role of each interviewee. Ethical standards were maintained during the study. All interviewees gave their informed consent for participation and subsequent interviews. The interviewees could leave the interview at any time. A research permit was obtained from the County and the participants’ confidentiality was observed. The interviews were conducted from January to March 2022 online as Teams interviews and were recorded and transcribed. Most of the interviews were individual interviews, although a few of them were carried out in pairs. The interviewees were selected in such a way that the most comprehensive information about the impacts was obtained. They represented the management, development and planning, supervision of client service work, employees working in client services, as well as technology trainers and technology suppliers (the companies) (see below).
The material and energy inputs required by both the ICT system and the physical service were investigated based on the interviews, statistics and other documents. For the calculation of climate impacts, quantitative data on direct and indirect factors were collected as comprehensively as possible. The interview data provided a rich basis for the qualitative assessment, including information on how people’s actions affected the impacts.
During the study it became clear that the data needed for the quantitative assessment of climate impacts were not publicly available. Data needed in such assessments tend to be private or confidential and therefore need to be collected by service owners or through good cooperation with all the related parties, such as in this study.
Principles of the impact assessments
Quantitative assessment of climate impacts
The novel methodology that was developed for assessing the climate impacts of digital services comprises an assessment framework and a calculation model.
The assessment framework is based on life cycle assessment methodology (LCA). LCA is an ISO standardised method to study complex value chains in order to understand potential environmental impacts (International Organization for Standardization (ISO), 2006). LCA enables quantification of the potential environmental impacts of the whole product system, including upstream impacts. This is important when studying systems that are global in nature. An LCA-based approach allows for the assessment of multiple environmental impact categories but for now, the availability of data typically limits the assessment to climate impacts with an indicator of Global Warming Potential (kg CO2e), which is a standard unit for measuring carbon footprints. In addition to standardised methods, further industry-specific rules or even case-specific refinements are often needed.
Digital services typically comprise such a high number of components that an exact assessment of each component is virtually impossible. For example, the network component of a service alone may utilise hundreds of connections and processing units. Another challenge to estimating the impacts of digitalisation is that the process of digitalisation is typically gradual. It can take years to fully digitalise a service and the service itself also develops over time. Thus, achieving a clear before-and-after comparison is a rare occurrence.
Figure 1 illustrates our framework for the quantitative assessment of the climate impacts of digital services. This is the most simplified framework that still adequately describes all the digital components of the service. The production of a digital service requires user IT equipment, internet connection and servers. The consumption of a digital service requires an internet connection and an access device. The access device can be dedicated to the service (e.g. medicine robot or dedicated tablet) or not (PC, smartphone, personal tablet also used for other purposes).
This simplified framework allowed for an assessment based on the collected data. Calculations based on detailed solution architecture would have been virtually impossible. In LCA, impacts are calculated per defined functional unit. It is a measure of the performance of the product system that is studied. It provides a reference to which all inputs and outputs in the system can be related. The functional unit of a one-year service use of one client/patient was selected to facilitate comparison of the climate impacts of digital services with the climate impacts of potentially saved travel.
Figure 1. The framework for the assessment of the climate impacts of digital services (source: Melkas et al., forthcoming; adapted from Tuominen-Thuesen et al., 2022).
The assessment framework was tested using real-life cases to detect any potential shortcomings and the limits of its use. We also wanted to know what a roadmap for future assessments should look like and whether it is correct to calculate the impacts in this work in this way. The services of two case regions were chosen as the test environment for the framework in order to identify the applicability of the framework and the calculation model for assessing the climate impacts of the studied digital services. We also gained information on the further development of the framework and the calculation model. This part only focuses on one of the regions and the medicine robot services.
Qualitative assessment
The method developed in this study for qualitatively assessing social and climate impacts is based on the principles of human impact assessment (HuIA, e.g. Melkas et al., 2020; Nelimarkka & Kauppinen, 2007). An assessment of human impacts can be used to structure new perspectives and describe solution options. Impacts assessed using HuIA can be planned or unintended and can be the result of long chains or networks of impacts (Nelimarkka & Kauppinen, 2007). Thus, HuIA is comprehensive by nature: impacts are not limited beforehand, but efforts are made to comprehensively identify them and make them visible. In this study, assessing the impacts on people entails, for example, examining the chains of impacts of digitalisation that can be related to well-being, relationships between people, and changes at care work, inter alia. This approach has been used in studies of the digitalisation of healthcare and care services, which have concerned traditional technology such as safety alarm systems (e.g. Melkas, 2011), and emerging technologies, such as care robots (Melkas et al., 2020). The essence of this approach is to holistically identify the positive, negative and neutral impacts on the different people and groups of people involved.
For the qualitative assessment in this study, the interviews were analysed using content analysis. We searched for both climate impacts and social impacts. An inductive thematic analysis (Braun & Clarke, 2006) of the data was conducted. The transcribed text and notes were then thoroughly reviewed to capture all aspects of the research topic. In this study, the impacts were grouped into positive and negative climate impacts and social impacts. The positive and negative social impacts were further grouped into impacts on clients/patients, impacts on care professionals and impacts on service organisations and society (Figure 2).
Environmental, especially climate impacts: Positive and negative |
Social impacts: Positive and negative
|
Figure 2. Categorisation of the qualitative results (source: Melkas et al., forthcoming; adapted from Tuominen-Thuesen et al., 2022).
Assessment of climate and social impacts: results
Quantitative assessment of climate impacts
The climate impacts of medicine robot services were calculated for a functional unit based on annual use of a medicine robot by a client who takes medication three times a day. Figure 3 describes the basic components of the service. These are the same components as in the framework (Figure 1) with the addition of physical visits to refill the medicine robot.
Figure 3. Medicine robot service components for the impact assessment (source: Melkas et al., forthcoming; adapted from Tuominen-Thuesen et al., 2022).
The climate impacts were calculated for both types of medicine robot. The overall results were quite similar. An illustration of the aggregated results can be found in Figure 4. Manufacturing of the robot dominates the climate impacts. Refilling and energy used by the robot together represent less than half of the climate impacts compared to the climate impacts of manufacturing. The backend and mobile data made only a small contribution to the overall climate impacts. Medicine cups – used in one of the robots – were responsible for more climate impacts than the energy used by the robot.
Figure 4. Climate impacts of a medicine robot (aggregated) (source: Melkas et al., forthcoming; adapted from Tuominen-Thuesen et al., 2022).
A medicine robot replaces travel. Clients without a robot typically need two visits per day, while the studied robots need to be refilled every two weeks. Figure 5 compares the climate impacts of the medicine robot services to alternative means of transport to fulfil daily medication needs.
In short, the studied medicine robots are a climate-friendly option when the distance to a client by car (even an electric car) is more than one kilometre (two kilometres by bike).
Figure 5. Comparing the climate impacts of medicine robot services to the climate impacts of avoided travel (source: Melkas et al., forthcoming; adapted from Tuominen-Thuesen et al., 2022).
To conclude, a medicine robot has good potential from the perspective of reducing climate impacts. It is particularly beneficial in rural areas and where client visits necessitate the use of a car.
Technology and service providers are key to reducing carbon footprints. The manufacturing phase of the studied medicine robots is responsible for over 50 percent of the assessed climate impacts. The carbon footprint of the manufacturing depends on the design of the robot and the use of materials, as well as on its lifespan. Operational energy and the robot’s data usage should also be optimised but have less impact on its carbon footprint. Requiring technology and service providers to submit carbon footprint calculations would be a good way to ensure green services.
Qualitative assessment of climate impacts
Positive climate impacts
The qualitative assessment provided in-depth information about the service context and revealed a number of additional impacts. Medicine robot services enable a reduction in the number of kilometres driven by care professionals, meaning the number of home visits could be reduced from as many as 60 visits per month per client (for administering morning and evening medicine) to just two visits (to refill the robot). Usually, however, only some of the visits are replaced by a robot. One of the aims is to also optimise routes and visits to clients’ homes. In the County, the distance from a home care office to a client’s home can be tens of kilometres. A care professional noted:
“I personally find them [medicine robots] useful. They have been well received in the work community. Suitable clients are suggested from the field [by the care professionals]. It's frustrating to drive 50 minutes just to administer morning medicine.”
Medicine robot services provide the opportunity to save on the protective equipment (disinfectants, gloves, masks) used by care professionals, which is beneficial for both the environment and the economy. Also, medicine robot services may reduce medicine waste. When only the correct dose of medicine is dispensed, there is no requirement for large packets of medicine that may expire if the medicine is no longer needed.
The life cycle of medicine robots is quite long: 7–8 years. The robots are considered durable and can be repaired by replacing worn parts. They are passed on from one client to the next, and their components are recycled.
Negative climate impacts
The negative climate impacts of medicine robot services have been previously described. However, the qualitative assessment showed that unnecessary additional driving sometimes occurs and could be avoided with better planning. This is related to the guidance given to the clients. When the client receives the robot for the first time, guidance is always given by a care professional and may need to be given several times in the beginning. The technical expertise of the care professional is essential so that they can help if there are any problems when the client is using the device. It is also essential from a more general perspective, as mentioned by a technology supplier:
”Thorough training and support produces [positive] environmental impacts by ensuring that the device is not left unused or that there is no need to return to administering the medicine on site.”
Even though a medicine robot generally reduces the amount of driving needed, error messages or alarms from the device sometimes mean a home visit by a care professional is necessary. Most of these situations can be dealt with by a care professional over the phone, such as when the device sounds an alarm because medicine has not been taken. The interviewees mentioned that sometimes a technician from the manufacturer needed to be called to repair the device, although this was rare. As to workspaces (offices), medicine robot services do not affect the number of workspaces, as the services were previously provided in the clients’ homes.
Qualitative assessment of social impacts
Social impacts were divided into both positive and negative impacts on different levels – clients, employees and organisations and society (Table 1). Social impacts may be planned (so that they are in line with the aim of the digitalisation actions, from a social perspective) or unexpected. The importance of including social impacts in comprehensive impact assessments was reinforced by the fact that the study revealed various intertwined and multi-directional impacts.
Positive social impacts
Detailed social impacts always depend on the type of service and technology being offered. Our qualitative assessment of medicine robot services revealed several positive social impacts, such as the preservation of client activity and independence, better regional equality in access to services regardless of place of residence, rationalisation of employees' work, increased work flexibility due to a reduction in the amount of time that employees spent travelling, and better allocation of societal resources (see also Table 1).
Medicine robot services (like the use of home care technologies in general) have led to savings in client care fees. The clients felt that the devices were easy to use; they are relatively automated and reliable and do not require clients to have technical skills.
A care professional noted:
”The attitude of clients and their relatives varies. Often, they resist using it [the robot] at first, but when they try and learn and then realise how useful it is, their attitude is generally positive and they start using the service.”
The use of medicine robots also led to the positive impacts of better-quality medicational care and fewer medication errors. The studied medicine robots will always administer the medicine at the specified time, which improves the accuracy of the medication. If a care professional administers the medicine, timing may differ. The robot also relieves the pressure of having to remember to take the medicine. The robot can also give other reminders. In addition, the service provides a sense of participation and accomplishment. Some clients do not like receiving visits from a care professional, so a medicine robot enables a client to maintain a sense of independence, while ensuring access to care.
Negative social impacts
A number of negative social impacts were also identified, such as issues related to client inequality (digital skills, or the service in question being unsuitable for the client and problems resulting from this), employee workload while learning a new way of working, and the increase in management challenges and level of complexity (see also Table 1).
The use of a medicine robot requires different types of skills, such as refilling the robot. New skills are also required in order to assess whether a client has a need for such a service; care professionals must know how to assess which clients the devices would be suitable for. It takes time for a care professional to process changes in the medication dosage, especially if the change is supposed to take effect immediately. The processing of the change depends on the device. In general, however, it is recommended that the change takes place from the next refilling in order to minimise the number of errors.
Familiarising a client with a device and visits due to alarms have to be conducted together with or in addition to other work, because no time is allocated in the work schedule for such activities. Typical error messages concern a device being unplugged. The client may also turn the medicine robot upside down, causing the medicines to become mixed up. Sometimes the medicines are installed incorrectly, or the bag roll gets stuck. In such cases, a care professional has to visit the client’s home in order to rectify the problem. Technical problems are also negative impacts, even though such problems are typically caused by network load issues. However, they are relatively rare.
The perceived unsuitability of the devices in the home environment was described as a negative social impact from the client's perspective. The large size of medicine robots may be a surprise and a client may consider it unattractive and inappropriate for use in their home. Also, a medicine robot does not provide social interaction, unlike a visit or a video call from a care professional. It is important to consider the suitability of the technology to the client, especially in the case of people with memory and/or mental illnesses, as they may have delusions or suspicions, as pointed out by a care professional:
”Some clients with specific illnesses may be very suspicious of such devices. They may think that the device is being used to spy on them or secretly photograph them. …We only try the devices if the client is suitable. But it is also possible to discontinue the service and return to a care professional’s visits."
Table 1. Positive and negative social impacts of medicine robot services (source: Melkas et al., forthcoming; adapted from Tuominen-Thuesen et al., 2022).
Impacts on… | Positive impacts | Negative impacts |
Clients/patients | Savings on service fees, improved accuracy in the timing of medication and care visits, less medication errors, maintaining a sense of independence and autonomy | Not suitable for everyone (e.g. a client potentially being suspicious); perceived suitability of the devices (size, appearance) in the home environment |
Employees | Reduction in the time required for travelling and technical tasks, easier planning of time use (more time for actual care work rather than, for example, changing protective equipment and disinfecting between client visits) | Change of work, new tasks (e.g. responsibility for technology, assessment of service needs) Change requires learning, which can be overwhelming No reduction in total workload because the number of clients keeps increasing; easy visits have decreased, while challenging visits remain |
Service organisations and society | More rational allocation of resources (face-to-face visits for those clients who really need them), possible increase in general appreciation and attractiveness of care work | Challenges related to work culture and in incorporating the technology into service processes Management challenges and complexity; new and old ways of working collide (e.g. engaging and assisting care personnel, procurement expertise, support services) |
Home care quality may be perceived as worse when technology is used in such services. Thus, the initial reaction of clients and their loved ones to the introduction of technology in home care is often negative. The implementation of the studied medicine robot services always starts with a two-week trial period, after which there is an option to stop using it. Most clients are satisfied after the trial period and want to continue using the device. The clients themselves were not interviewed in this study so their experiences were described by the care professionals.
Although the technology can be useful, the overall workload in home care services has not decreased because the number of clients keeps increasing. The interviewees stated that the easy visits have decreased due to the technology while the challenging visits have remained.