CareMobi was developed using a user-centered design process that actively engaged ADS staff, family caregivers, and community stakeholders throughout its development. Iterative feedback was solicited through interviews, focus groups, and usability testing to ensure the tool was feasible to integrate into workflows, responsive to caregiver needs, and adaptable across diverse ADCs. This participatory process emphasized co-design to identify priority features (eg, streamlined messaging, daily activity summaries, and health-related observations) and reduce barriers such as duplicate documentation or technology burden. The full details of CareMobi’s design and development process, including prototype testing, have been reported elsewhere in prior research; here we focus on evaluation of its implementation, usability, and acceptability in practice [10].

We adopted a pragmatic, low-burden field usability evaluation of CareMobi in ADCs, guided by a Goal–Question–Metric (GQM/GAQM) [13] approach. Originally developed for software engineering and adapted for usability research, GQM ensures that data collection remains purposeful, relevant, and aligned with stakeholder needs. In practice, evaluators first articulate the overarching goals (eg, feasibility, usability, and acceptability), then derive key questions that operationalize those goals (eg, “Can caregivers complete core tasks without difficulty?”), and finally specify the metrics that can provide evidence (eg, task completion rates, satisfaction ratings, and interview themes). This hierarchical structure promotes rigor while keeping evaluations pragmatic and low burden, particularly in resource-constrained real-world settings such as ADCs. This allowed us to explicitly tie high-level goals (usability, acceptability, and feasibility) to specific questions and measurable metrics, while respecting constraints of under-resourced ADC settings. We aligned these questions and metrics with the GAQM framework’s principle of first defining a goal (conceptual), then attributes (usability dimensions), then operational questions, and finally metrics (quantitative and qualitative). This approach is outlined in Table 1.

The evaluation combined field (in situ) testing, self-user usability (unmoderated use in real settings), post-use questionnaire, and qualitative interviews with staff and family caregivers in ADCs located in rural and underserved communities.

We partnered with 2 ADC sites primarily serving people living with dementia that were recruited at the Annual Meeting of the National Adult Day Services Association. Within each center, paid staff and family caregivers were recruited to pilot test CareMobi in their real daily workflows (as opposed to a lab setting), over a 12-week period. Our approach was consistent with “field studies” in mobile usability literature, where tasks are executed in the natural environment, and contextual factors (interruptions, real devices, ambient conditions) are preserved [10].

Recruitment occurred over 3 months through center leadership, who distributed study information to eligible caregivers and staff; interested individuals were then contacted by the research team to confirm eligibility and obtain informed consent. Eligible ADC staff included those directly involved in the day-to-day care of people living with dementia and responsible for communicating with family caregivers. Eligible family caregivers were those who identified as the primary caregiver of people living with dementia attending one of the participating ADCs, were aged 18 years or older, and owned a smartphone or tablet capable of running the CareMobi app. Both staff and caregivers were required to speak English and be willing to provide informed consent.

Prior to deployment, participants received minimal training, an intentional design choice to evaluate how intuitive the CareMobi platform was in real-world use. ADC staff participated in a 30-minute orientation covering the study overview and app functions. Family caregivers were assisted, as needed, in downloading the app using a QR code and were then encouraged to explore it independently. For ongoing support with the app, staff and caregivers were directed to a video library developed by an instructional technologist. Staff and caregivers were also advised to reach out to the study team with any questions or regarding technical issues. This pragmatic, low-burden approach mirrored real-world implementation conditions and aligned with the GQM framework, specifically addressing goals of assessing intuitiveness and efficiency of use with minimal external support.

Staff were instructed to use their discretion and share information through CareMobi that was tailored to each person living with dementia and caregivers’ needs, while caregivers were encouraged to use the app as much or as little as they preferred to engage with the ADC. There was no minimum requirement for use in order to reflect real-world engagement patterns. CareMobi uses HIPAA (Health Insurance Portability and Accountability Act)–compliant security protocols, including encrypted data transmission and storage, authenticated user access, and restricted role-based permissions. No identifiable information was shared outside the secure platform, and all study data were deidentified prior to analysis.” Following consent, participants were instructed to download the CareMobi app from their preferred app store, and family caregivers completed a baseline survey. After the 12-week usage period, participants completed a post-use questionnaire assessing perceived usability, satisfaction, burden, and open-ended feedback. We also conducted semistructured interviews with a purposive subset of caregivers and staff, focusing on their experience, trust concerns, communication challenges, and suggestions for improvement. Staggered incentives, in the form of gift cards, were provided upon completion of the baseline survey, usability questionnaire, and interview (US $150 total).

Finally, a semistructured interview guide was used to explore participants’ experiences in greater depth. Interview domains included perceptions of transparency and trust, communication gaps, perceived benefits or frustrations, barriers to engagement, and suggestions for improving design or workflow alignment. Open-ended questions (eg, “Were there any parts of the app that felt confusing, frustrating, or difficult to use?”; “How, if at all, did it impact your experience with the ADC?”) elicited participants’ reactions to CareMobi, perceptions of usability, potential barriers, and impact on their relationship with the ADC as it pertained to the health of the people living with dementia.

To evaluate feasibility, we applied a priori STOP-AMEND-GO [17] criteria across five domains: retention, usability, satisfaction, engagement, and acceptability. Evaluable retention was defined as completion of the 3-month follow-up survey, excluding non–study-related attrition (death or transition to a higher level of care); thresholds were STOP if <60%, AMEND if 60-69%, and GO if ≥70%. Usability was assessed using the PSSQ, with criteria based on the proportion of participants endorsing “agree” or “strongly agree” for items on ease of use and ease of learning; thresholds were STOP if <60% on any item, AMEND if 60-74%, and GO if >75% on each. Satisfaction was measured as the percentage of caregivers and staff endorsing “agree/strongly agree” on overall satisfaction with CareMobi, with thresholds STOP if <60%, AMEND if 60-74%, and GO if >75%. Engagement was defined as the proportion of participants who created an account and used the app at least once; thresholds were STOP if <50%, AMEND if 50-69%, and GO if ≥70%. Finally, acceptability was measured as the percentage of staff and caregivers expressing willingness to continue using CareMobi beyond the study period, with thresholds STOP if <50%, AMEND if 60-74%, and GO if >75% (Table 2).

All study procedures were reviewed and approved by the Institutional Review Board at New York University (IRB-FY2023-7509). All participants provided informed consent prior to participation, and all data were deidentified to protect confidentiality. Compensation was US $150 for participation in all phases of the study.

In total, 15 caregivers completed baseline surveys (Table 3). Most participants identified as female (13/15, 86.7%), were aged 30-64 years (10/15, 66.7%), and reported some college or higher education (14/15, 93.3%). The sample was predominantly non-Hispanic White (14/15, 93.3% White; 14/15, 93.3% non-Hispanic). Most were married (11/15, 73.3%). Overall, 40% (6/15) were employed full-time and 20% (3/15) part-time; 26.7% (4/15) were retired. All respondents were family caregivers (15/15, 100%); 80% (12/15) identified as the sole caregiver. Relationship to the people living with dementia was most commonly adult child (7/15, 46.7%) or spouse or partner (5/15, 33.3%). Most had been caregiving for ≥1 year (11/15, 73.3%).

Across 15 respondents, confidence was generally moderate to high for managing common dementia-related challenges (Table 4). For agitation, 10/15 (66.7%) reported being extremely or moderately confident; for mood changes, 10/15 (66.7%); for mental status changes, 12/15 (80%). Confidence in deciding whether to call a provider was high (13/15, 86.7% extremely or moderately). Communication-related competencies were strongest: 14/15 (93.3%) felt extremely or moderately confident talking to a medical provider, and 12/15 (80%) reported they could readily gather information a provider would want. Areas with more dispersion included understanding vital signs (10/15, 66.7% extremely or moderately; 3/15, 20% slightly) and assessing pain (9/15, 60% extremely or moderately).

Care recipients were primarily in moderate to later stages of dementia: 7/15 (46.7%) had moderately severe cognitive decline, 2/15 (13.3%) severe, and 2/15 (13.3%) very severe. Within the prior 6 months, 6/15 (40%) had at least one emergency department visit, and 4/15 (26.7%) had been hospitalized. Medication burden was substantial: 10/15 (66.7%) were taking ≥5 daily medications and 4/15 (26.7%) were taking 3-5 (Table 5).

Of the 15 caregivers who volunteered for the study, 14 logged into the app at least once, demonstrating high initial engagement. Four participants did not complete the study due to their loved one leaving the center, either due to death or requiring a higher level of care. Of the remaining caregivers, 8/10 completed the usability questionnaire and 6/10 participated in the follow-up interview. Among the 3 staff who participated, all (100%) logged into the app, completed the PSSQ, and completed the follow-up interview.

Among the 8 caregivers who completed the application usability questionnaire, perceived usability and satisfaction were high. At least three-quarters “agreed/strongly agreed” that the app was easy to use (6/8, 75%) and easy to learn (7/8, 87.5%); 6/8 (75%) liked the interface and found information well organized. Most felt comfortable using the app in social settings (6/8, 75%) and endorsed appropriate time demands (6/8, 75%). Willingness to use the app beyond the study period was high (6/8, 75%), and overall satisfaction was favorable (6/8, 75% “agree/strongly agree”).

Communication and access benefits were frequently endorsed: 7/8 (87.5%) reported that the app improved access to their loved one at the center, 7/8 (87.5%) said it made communication with the ADC convenient, and 6/8 (75%) felt comfortable communicating with the center via the app. Navigation consistency was also rated positively (7/8, 87.5%) (Table 6).

Staff reported highly favorable usability ratings, with median scores at or near the ceiling across nearly all items (median=7, “strongly agree”). All 3 participants indicated that the CareMobi app was easy to use, easy to learn, and well organized (median=7). Satisfaction was uniformly high (all median=7): staff strongly agreed they would use the app again, that it contained the functions and capabilities expected, and that it provided an acceptable way to communicate with caregivers.

Communication-related items also demonstrated the strongest endorsement. Staff unanimously reported that the app improved access to caregivers, made communication more convenient, and increased opportunities to interact with caregivers (median=7 across items). They also expressed confidence that any information sent would be reliably received (median=7).

Minor variability was observed on a small number of items. For example, one staff member reported difficulty recovering from mistakes, producing a median score of 6 (“agree”) on that item. Another staff member selected “neither agree nor disagree” on certain workflow-related items (eg, time fit and error recovery), though the median across sites still remained 7 (Table 7).

Consistent with the high ratings for convenient communication and improved access (75-88%), interviews with caregivers and staff, and open-ended written comments, revealed eight major themes that described CareMobi’s value and limitations in daily practice.

In general, caregivers expressed strong support for CareMobi in their open-ended survey responses: “CareMobi is very helpful to me.” “It is great.” “Keep it up :)” “I found the app to be user friendly and easy to communicate to others.”

This study evaluated the acceptability, demand, and practicality of CareMobi, a mobile platform designed to strengthen communication between ADCs and family caregivers. We posit that by making ADC-based care more transparent to families (through timely updates and role-specific, low-burden interactions), CareMobi, if feasible and acceptable, can reduce uncertainty and build trust among caregivers. This, in turn, can support sustained participation in ADCs and bolster engagement and connectedness for people living with dementia who are at risk of loneliness and social isolation. Our findings provide preliminary support for this proposition.

CareMobi demonstrated high acceptability among both caregivers and staff, meeting or exceeding all prespecified STOP-AMEND-GO thresholds [17]. Specifically, usability scores surpassed the >75% “agree/strongly agree” criterion across all items, satisfaction levels exceeded benchmarks, and more than 70% of caregivers engaged with the app at least once. Staff usability scores reached ceiling levels, with median ratings of 7 (strongly agree) across nearly all items. Acceptability was also met, with more than three-quarters of participants expressing interest in using CareMobi beyond the study. Caregivers emphasized the platform was “user friendly,” “very helpful,” and reassured them by being “a tap away from knowing the daily report of their loved ones.” Together, these findings provide strong evidence of acceptability and support CareMobi’s potential as a low-burden, scalable tool for use in ADCs, and provide robust evidence for moving forward with broader implementation trials.

Technology adoption in dementia caregiving populations has historically been low, with many studies reporting attrition, underuse, or abandonment of digital tools due to steep learning curves, lack of personalization, or misalignment with caregivers’ daily realities [18,19]. In contrast, CareMobi achieved high engagement among study volunteers: 93% of caregivers logged into the app at least once, and 75% of caregivers and 100% of staff expressed willingness to continue use beyond the study. This success may be attributed to its co-designed, user-centered development process, which emphasized simplicity. Staff specifically highlighted reduced “phone tag,” noting that streamlined messaging saved time and allowed them to redirect effort toward direct care. This represents an important advantage over other health care technologies that inadvertently add to staff burden rather than reducing it [12]. Staff observed that bidirectional communication via the app was more likely to be achieved with caregivers who were new to the ADC, as they relied on CareMobi as their primary means of communication with staff. In contrast, established caregivers, who were more accustomed to phone-based communication, tended to use the app more passively to track updates. This suggests that caregivers who have developed routines with staff may perceive less added value in switching.

It is worth noting that our sample was composed of caregivers who reported relatively high baseline confidence in managing their loved ones’ needs, which may have facilitated their willingness to engage with a new tool. Many were also sole caregivers, relying on the ADC as an extension of their kinship network. This close relational tie to the center may have further motivated consistent use of CareMobi, as research suggests that engagement is supported not only by the design of the technology but also by the social context in which it was deployed [20].

ADC staff were able to integrate CareMobi into their workflows, but several refinements were identified. Both caregivers and staff requested more structured guidance on which observations should be logged (eg, hydration, bathroom trips, and mood changes) and templates such as “Today at a Glance” or “Medication Updates” to enhance efficiency. Staff expressed demand for a desktop or web version since most entered data from a computer, as opposed to a smartphone or tablet. Research shows that community-based settings, especially those with limited resources, are often excluded from technological innovation and struggle because of a fragmented digital infrastructure [21,22]. By focusing on low-cost, high-utility design, and implementing requested refinements, CareMobi can help address these structural inequities, ensuring that centers with fewer resources are not left behind in digital health advancement.

Another critical finding was CareMobi’s impact on caregiver confidence and trust. Families reported using real-time updates to interpret behaviors at home, guide conversations with medical providers, and even advocate for deprescribing when daily logs showed adequate eating and drinking. One caregiver shared: “We actually got him off one of his meds, because I could prove he was eating and drinking fine there.” This echoes existing evidence that improved caregiver communication can enhance self-efficacy, reduce decisional conflict, and promote timely care coordination [23,24]. By strengthening transparency, CareMobi appears to foster not only trust in ADCs but also caregiver empowerment in broader health care decision-making. As another caregiver said, “The better you know how things are going throughout the day, the better the day is for you both.”

Although findings were promising, the sample was predominantly Non-Hispanic White, female, and English-speaking. This does not reflect the broader demographics of ADC families nationally, where approximately 58% of clients identify as racial or ethnic minorities [4]. Language accessibility was a major limitation; CareMobi is not available in Spanish or other languages. This restricted participation for centers serving diverse populations. This represents both a limitation of the current evaluation and a critical direction for future work. Addressing language and cultural tailoring is essential for scaling CareMobi equitably and ensuring its benefits extend to the diverse populations that rely most on adult day services.

Collectively, these findings suggest that CareMobi can address longstanding barriers to ADC use. By improving “day-visibility,” simplifying communication, and providing structured yet flexible updates, the app has the potential to stabilize attendance and support sustained participation, barriers that have historically undermined ADCs’ ability to demonstrate value in long-term care [25,26]. This is particularly timely given the underfunding of adult day services and their critical role in providing cost-effective, community-based care for people with dementia [7].

While this pilot yielded encouraging results, this study had several limitations. First, although appropriate for a feasibility design, the sample size was small and drawn from 2 centers, limiting generalizability. The sample was also demographically homogenous (ie, primarily Non-Hispanic White, female, and English-speaking) and did not reflect the diversity of families served by ADCs nationally. As participation required English proficiency and smartphone ownership, findings may overrepresent caregivers who were more comfortable with technology and positively predisposed to digital tools. Second, the study duration was relatively short, constraining our ability to draw conclusions about long-term engagement or patient outcomes. Third, while we evaluated usability, satisfaction, and acceptability, we did not measure downstream effects like caregiver burden or health care use. These outcomes will be the focus of future trials.

The findings from this feasibility study will directly inform the next phase of research, which will involve a larger, multisite hybrid effectiveness–implementation trial. Future work will assess whether improved communication and transparency translate into measurable outcomes such as reductions in caregiver burden, improved people living with dementia engagement and quality of life, more consistent ADC attendance, and reduced acute-care use. Recruitment will intentionally prioritize racially, ethnically, and linguistically diverse caregivers and centers, including development of multilingual platform versions and community-engaged recruitment strategies.”

By coupling digital transparency with low-burden communication, CareMobi offers a promising approach to modernizing ADCs in ways that respect resource constraints and caregiver realities. In doing so, it reduces caregivers’ uncertainty about sending people living with dementia to the ADC, in turn, providing the people living with dementia an opportunity to socialize and engage. To fully realize this potential, future work must prioritize equity, especially linguistic inclusion, infrastructure support, and engagement in under-resourced settings. Ultimately, if scaled thoughtfully, tools like CareMobi may enrich the social fabric of dementia care, increase accessibility of ADCs, and deepen the connection between caregivers, centers, and people living with dementia.