The development of the assessment tool was in preparation of the upcoming randomized controlled intervention trial examining the effectiveness of a digitally supported care management intervention for family caregivers (German title: Gesund Angehörige Pflegen [GAIN]). The trial was supported by the Federal Joint Committee (G-BA [Gemeinsamer Bundesausschuss/Innovationsausschuss]). The funding code (FKZ) is 01VSF18030).

We involved individuals from 3 stakeholder groups, namely, family dementia caregivers, health care professionals (general practitioners, neurologists, psychiatrists, psychologists, and nurses), and Alzheimer Society representatives. We approached HCPs of a medical university and memory clinic staff, as well as Alzheimer Society representatives, and asked for their feedback (phase 1). Family dementia caregivers who were visiting a memory clinic with their relative or friend with dementia were asked whether they would like to provide feedback on the paper-based version of the assessment instrument regarding the content, clarity, and comprehensiveness (phase 2) as well as the usability of the digital version of the assessment tool as a self-completion assessment instrument on a PC tablet (phase 3). No further demographic data were collected.

We used an iterative user-centered participatory approach with 3 phases. Previously used questionnaires [20,23,24] assessing unmet needs and care interventions from previous dementia care management studies [4,25-31] in Germany were used as starting point.

All data collected through written feedback, protocols, think-aloud protocols, and interviews were analyzed thematically following the steps by Braun and Clarke [51]. The six phases involved were (1) familiarizing ourselves with the data, (2) generating initial codes, (3) constructing themes, (4) reviewing themes, (5) defining and naming themes, and (6) producing the report [51].

The study is conducted in accordance with the criteria (valid at present) of the Declaration of Helsinki, the ICH-guidelines for Good Clinical Practice, the Memorandum for Safeguarding Good Scientific Practice (German Research Foundation), and the International Ethical Guidelines for Biomedical Research Involving Human Subjects. Ethical approval has been obtained from the Ethical Committee of the University Medicine Greifswald (BB120/2019) and the Ethical Committee of the University Medicine Rostock (A2020/0013). The trial is registered at ClinicalTrials.gov (NCT04037501).

A total of 27 family caregivers of people with dementia (n=18), HCPs (n=7), and Alzheimer Society representatives (n=2) participated. Among the HCPs were general practitioners, neurologists, psychologists, psychiatrists, and nurses. The mean age of the 27 family caregivers was 57 (SD 13.78; range 35‐88) years and 9 (50%) were women. The HCPs’ and representatives’ mean age was 41 (SD 8.05; range 28‐50) and 6 out of 9 (67%) were women. As this was a participatory approach, no further demographic data were collected.

After the input from the advisory board including neurologists, psychiatrists, psychologists, nurses, and Alzheimer Society representatives, the final structure of the assessment instrument was selected. It was the one with the highest percentage of matches between problem-centered needs and the associated interventions (69/72, 96%; version 3; see Table 1).

The unmet needs assessment tool comprised several validated questionnaires with a total of 57 questions examining a person’s self-reported demographic information, use of medical and nonmedical services, unmet needs (CANE), health-related quality of life (EQ-5D-5L), caregiver burden (Zarit Burden Interview [ZBI]), and social support (Lubben Social Network Scale [LSNS]). It can be completed using a digital version or a paper-based version. In total, the assessment set covers 51 health- and support-related aspects.

Domains covered by the unmet needs assessment tool are the following: (1) health and care, (2) employment, (3) information and knowledge, (4) emotional support, (5) social support, and (6) caregiver burden.

The structure of the unmet needs assessment tool can be seen in Table 2.

The number of unmet needs addressed the participants’ medical needs, home care needs, psychosocial needs, and needs connected to the caregiver role. This needs assessment instrument included selected parts of the CANE [47].

Health-related quality of life was assessed using the EQ-5D-5L [41]. This instrument comprises 5 dimensions: mobility, self-care, usual activities, pain or discomfort, and anxiety or depression. Each dimension has 5 levels varying from no problems to extreme problems. Each level corresponds to a 1-digit number that expresses the level selected for that dimension ranging from 1 to 5 with higher numbers indicating more severe problems. The digits for the 5 dimensions can be converted into an overall index by an not publicly accessible algorithm. The index describes the participant’s health-related quality of life. It also contains a vertical visual analogue scale (EQ VAS), with endpoints that are labelled “Best imaginable health state” and “Worst imaginable health state.” The VAS reflects the patient’s own judgement and can be used as a quantitative measure of health outcome.

Social support was assessed using LSNS-6 [44,45]. This scale is a self-report measure of social engagement including contact to and interaction with family and friends on a 6-item scale. Total scores range from 0 to 30 with an equally weighted sum of the 6 items. The family and friends subscales include questions regarding the number of friends and family one has regular contact with as well as their availability for help and support in private matters. High scores indicate strong social networks.

Informal caregiver burden was assessed using the 7-item version of the Zarit-Burden Interview (ZBI-7). The short version ZBI is a caregiver self-report measure to examine burden, which is associated with functional and behavioral impairments in the social, psychological and physiological context, and home care situation [42,43]. It contains 7 items using a 5-point scale. Response options range from 0 (Never) to 4 (Nearly to Always). Total scores range from 0 indicating no burden to 28 indicating severe burden.

Regarding the assessment’s tablet usability, informal caregivers, HCPs, and Alzheimer Society representatives identified 2 subthemes, namely, simple layout and handling with slightly different subtheme categories.

This study describes the phases of an iterative, participatory process we used to develop a digital tablet-based assessment tool designed to comprehensively identify the needs of informal caregivers of people with dementia in clinical practice. Through the collaboration with family dementia caregivers, health care professionals, and Alzheimer Society representatives, we developed a digital unmet needs assessment instrument that consists of a set of standardized instruments but also includes additional questions and adaptations that were incorporated along the iterative development process. The unmet needs assessment tool focuses on needs that can arise from two sources: either from the personal needs of the family caregiver or from the care responsibilities for people with dementia. With the contribution of informal caregivers and other stakeholders, we designed an assessment tool that covers a comprehensive range of needs considered important. At the same time, with the input of the stakeholders, the time to complete the assessment procedure was substantially reduced from the initial 37 minutes to 18 minutes. This indicates that we reached our goal of developing a comprehensive, feasible, and time-efficient assessment device.

With respect to the content of the assessment instrument, we identified important aspects regarding clarity, assessment structure, the reduction of completion time as well as content areas important to informal caregivers. We took the feedback in multiple rounds and incorporated changes along the way. Similar approaches have been conducted successfully by other research teams developing assistive technologies to support self-management of people with dementia [52].

Regarding the digital tablet-based assessment tool, we could incorporate crucial aspects regarding the layout and the handling of the tablet. The difference between our first digital assessment version compared with the final digital assessment version illustrates the importance of involving multiple perspectives, especially of those for whom the assessment instrument or technology is intended. In the literature, there is an increase of studies involving patients and stakeholders in the development or testing of technology [52-54].

The participatory approach allowed us to incorporate aspects into the development of the digital assessment tool that we, as researchers, would have missed otherwise, as can be seen in the first digital tablet-based assessment version in Figure 1 when compared with the final version in Figure 2. The input of informal caregivers, health care professionals, and Alzheimer Society representatives has broadened our perspective and has given us important hints and advice towards aspects that are important for a digital, self-completion unmet needs assessment tool. They identified questions that were unclear, suggested possible answer options, and gave us ideas on how to simplify the navigation throughout the assessment procedure. Some researchers have used similar procedures for the development of instruments to assess patients’ needs [24,47] or to facilitate implementation [19]. A review by Fischer et al [55] reports on the importance of user involvement, particularly the involvement of older adults in technology design. In their review of 40 empirical studies published between 2014 and 2018, these authors examined the consequences of involving older adults in technology design and stated that learning, adjusted design, and improved sense of participation were outcomes frequently stated in studies [55]. Not only did they report that involvement facilitated learning about the needs of older adults but it also led to the generation of new technology ideas such as a companion robot, for instance, which was not in the researchers’ mind before involving older adults [55]. Newly gained insights then lead to iterative adjustments of the prototype design [55]. This was also the process that we followed making adjustments to the unmet needs assessment tool. With respect to the quality of technology developed with user involvement, Kopeć and et al [56] noticed that the overall quality of the technology could be improved. They argued that teams who involved users to develop a mobile app were rated higher by an independent jury in a competition than teams who did not. Older adults appreciated the sense of participation and felt that they were being treated as experts on their own lives and equal partners [57]. Researchers reported that involved users enjoyed their experience and some users described their involvement as “happy memories” [58-60].

In contrast to already existing unmet needs assessment instruments, our digital assessment tool identifies a informal caregiver’s self-reported unmet needs and ties them directly to interventions most of which are offered within the German health care system. Through the matching process of possible unmet needs with interventions in phase 1 of our assessment development process, we aimed to overcome the pitfalls of identifying problem-centered needs without having support services to meet them at hand. In addition, the digital assessment tool acknowledges the intertwined nature of the dyad, namely the informal caregiver and the person with dementia focusing on not only the health and support needs of the caregiver but also the caregiver’s care responsibilities. Although we did not include the CNA-D, we did compare the content and ensured that no important aspect was missed in the assessment procedure.

The stakeholders who contributed to the development of the assessment tool were a convenience sample. We have asked informal caregivers, health care professionals and Alzheimer Society representatives who either visited a memory clinic with their significant other, worked at the memory clinic or who were already known to us due to previous collaborations. Furthermore, the sample is limited to people living in a specific region in Northern Germany (Mecklenburg-Western Pomerania). In addition, our level of involvement was limited to stakeholders sharing their feedback on the assessment tool, which is only 1 of the 3 levels of involvement [61]. The tendency to involve stakeholders only in some rather than all levels of involvement was also highlighted by Fischer et al [55] arguing for further research on the actual practices of user involvement. For future projects, a higher level of involvement is sought, especially since current results indicate that an elaborated, theory-based participatory approach including people with dementia, informal caregivers and regional stakeholders might raise the chances of successful implementation in routine care [19].

Our study has shown the feasibility and success of user involvement in the development of an assessment tool for identifying the unmet needs of people with dementia and their informal caregivers. By presenting our approach, we hope to motivate researchers worldwide to expand stakeholder involvement in the early stages of studies. To facilitate this, the questionnaire will be made available to researchers on the homepage of the Network for Translational Dementia Care Research. Further evaluations, such as the acceptance of the tool from the perspective of the study nurses and GPs, are currently being evaluated and will be published separately.