We conducted an evaluative study based on descriptive qualitative methods [41].

For dashboard development, we used a collaborative approach [42] at the core of learning health system approaches. Such an approach can lead to more relevant, efficient, and sustainable results than a more traditional approach in terms of development and problem-solving in the relevant communities and organizations [43].

In collaboration with the ministerial advisor and the four project managers of the QAP, we selected a format and relevant indicators for this exercise. These choices of format and indicators were then validated with members of the QAP advisory committee, composed of managers, leading clinicians, and anyone involved in the QAP implementation from regional organizations in the province.

A format of a maximum 2-sided sheet was selected (see Multimedia Appendix 1). The choice of a paper rather than digital format was made owing to its simplicity, accessibility (easily accessible to all without depending on technology or connectivity), and instantaneous visibility (without requiring complex navigation). These advantages were prioritized despite the knowledge of disadvantages such as the need for manual updates and space limitations. This support is particularly useful experimentally, even though we acknowledge that it will need to evolve into a digital modality during potential routinization in regular managerial processes.

On the first page, nine indicators were selected from provincial clinical-administrative data from the Quebec Integrated Chronic Disease Surveillance System developed by the Institut national de santé publique du Québec [44]. Indicator selection was based on a conceptual framework on the quality of care for dementia, covering a continuum of care from primary care to emergency use and hospitalizations [45]. Eight of these indicators (type of physician most regularly visited, percentage of people with at least one visit to a family doctor, average number of visits to a family doctor per person, percentage of people with at least one emergency room visit, percentage of people with at least one hospitalization, average number of days hospitalized per person, percentage of people with at least one hospitalization with an alternative care level—representing patients who are hospitalized but no longer requiring acute care and waiting for long-term placement—and average number of days in care level) were measured in 2019-2020 in the population of individuals with dementia 65 years and older, and one indicator (prevalence) was measured between 2000-2001 and 2019-2020 in the population of individuals with dementia 40 years and older. While these surveillance data are highly useful for tracking diseases, they are sometimes underused by local actors due to their excessive quantity and the lack of means for rapid analysis [44]. Data visualization, facilitated by dashboards, makes these data more accessible, facilitating decision-making.

On the second page, we presented findings of a thematic analysis of two sources: annual reports to the Ministry of Health and Social Services produced by the managers involved in the QAP between 2017 and 2019, as well as notes of meetings between the ministerial advisor and the four QAP project managers with the managers involved in the QAP in 2020. While quantitative data highlight high-level trends, qualitative data served to enrich the dashboards by providing context to the figures, offering a more comprehensive and engaging perspective [46].

Finally, we chose to present specific results for the population of each health organization and for the entire province. This choice was made to contextualize the results of each organization to those of the entire province.

In spring 2022 (April to June), the ministerial advisor and the four QAP project managers (EM, CF, JD, and CM) organized 90-minute virtual meetings with the managers and leading clinicians involved in the QAP implementation as well as other persons involved in the QAP (eg, research personnel) in each of the 24 regional health organizations. These meetings aimed to share the new directions of the QAP and explore the progress and challenges of the pandemic with each organization. The last 30 minutes of these meetings were devoted to presenting regional dashboards and facilitating group interviews by the research team. The research team had approximately 10 minutes to present the dashboards (5 minutes for quantitative aspects and 5 minutes for qualitative aspects) and approximately 20 minutes to conduct a group interview with the participants in the meetings to assess the interest and utility of the dashboards as well as to assess the needs for improvement of these dashboards. The interview guide included four open-ended questions and prompts: “Are these results surprising to you?” “How will this dashboard be useful to you?” “Are there any results missing that you would have wanted in this dashboard?” and “Do you have any questions?”

Four members of the research team alternatively facilitated the group interviews (GAL, MG, YC, and IV). One member (GAL) was present in 20 group interviews, ensuring consistency, reinforcing study fidelity and credibility [47]. Two members were always present: one to present the quantitative part of the dashboards (GAL or IV) and the other to present the qualitative part (MG or YC). In addition, one member was designated the main note-taker, while the second member supported the main note-taker and had to read and complete the notes or suggest modifications on a document shared live and online by the main note-taker.

The note-takers were instructed to note dynamics between the participants and to indicate the participant who put forth each observation noted. Thus, a code was assigned to the interlocutor: manager (ie, director, deputy director, coordinator, unit or service head), leading clinician (ie, territorial nurse, physician, or social worker), other participant (ie, research agent or research coordinator), or an unidentified person (when note-takers had not identified the interlocutor).

Consent was obtained verbally at the onset of meetings, and no audio or video recordings were conducted to streamline the analyses to protect confidentiality, especially given the involvement of project managers in organizing the meetings.

This study received ethical approval from the ethics committee of the Centre intégré universitaire de santé et de services sociaux de l’Estrie - Centre hospitalier universitaire de Sherbrooke (MP-31-2021-3701) to ensure that all research activities involving human participants adhered to ethical standards and guidelines. The participants provided informed consent before participating in the study, emphasizing their voluntary participation, the purpose of the research, and the confidentiality of their information. All procedures involving human subjects were conducted in accordance with the ethical standards of the ethics committee and the principles outlined in the Declaration of Helsinki. The research team prioritized participant well-being, privacy, and the responsible handling of data throughout the study.

Overall, 23 group interviews were conducted. One organization from a university/peripheral region could not participate due to a significant change in the local governance of the QAP, and two organizations from remote/isolated regions participated in the same group interview. Eighty-two individuals participated in the group interviews: 44 were managers, 32 were leading clinicians, and 7 were other participants (mainly research coordinators).

Leading clinicians who participated in the focus groups mainly came from university/peripheral or mixed organizations. Additionally, university/peripheral organizations were the only ones in which other types of participants (research agents or coordinators) took part in the focus group.

In general, managers and leading clinicians shared the same interest in the prevalence indicator. However, leading clinicians expressed needs that managers and other participants did not express, particularly on follow-up indicators in Family Medicine Groups (interdisciplinary primary care teams), reasons for consultations/visits in primary care or emergencies, and patient satisfaction.

Some managers and research agents or coordinators would have appreciated more information on how individuals with dementia were identified for quantitative indicators. Finally, only one leading clinician and the research agents or coordinators suggested using these dashboards for research purposes.

This study determined that dashboards are of interest to those responsible for implementing an Alzheimer Plan in regional health organizations, and their format and content were appreciated regardless of the region or profession of the participants [51]. The study also highlighted specific needs regarding these dashboards that transcend all regions. Participants from all regional health organizations expressed a need for data on orphan patients or those not registered in interdisciplinary primary care clinics, as well as data on indicators related to services offered by other health care and social professionals (ie, nurses, social workers, pharmacists, and occupational therapists). These needs seem to express a desire to better delineate the challenges posed by orphan patients in the organization of health care across the province and to better monitor how interprofessional collaboration takes shape in interdisciplinary primary care clinics (Family Medicine Groups in Quebec).

We also identified interests and specific needs among participants from different regions and different types of professionals. The interest in hospitalization and continuity of care expressed by participants from university/peripheral regions (more urban) can be explained by the higher population density and a diversity of health resources in these regions [31]. Additionally, urban regions often have a greater diversity of health care and social professionals, specialists, and medical technologies available to meet the population’s needs [52]. In contrast, in remote or isolated regions, resources and the number of professionals may be more limited [36,38]. This higher resource availability in urban settings could explain higher interest in continuity of care in urban settings.

Another significant specific need that varied between regions is the need for diagnostic tools and adapted training expressed by participants from remote and isolated regions. The geographical, demographic, and socioeconomic characteristics of these regions make them unique and require different approaches to meet their health needs [1]. Health care professionals in these regions often face different challenges in providing health care for people with dementia [35,36]. This specific need may arise from the observation that both in Quebec and Canada, there is a higher proportion of Indigenous communities in rural areas [53,54], requiring cultural adaptation of care delivery and supporting tools. This is especially important considering the data availability in these regions [55]. Two regions only received qualitative dashboards. Quantitative surveillance data were not available for these organizations either due to their small populations posing risks to data dissemination or due to the fee-for-service mode of physician payment, on which surveillance data rely, but which is less frequent in these regions [56]. The inclusion of qualitative data has proven to be a significant asset in addressing these limitations.

Finally, organizations that span a larger territory (often from remote/isolated or mixed regions) expressed the need for a more granular analysis [42]. Several participants wanted more precise data for smaller territories. However, an ethical constraint prevented us from producing dashboards for smaller territories. Producing a single dashboard for all 24 regional health organizations remains a challenge to explore in future work.

The collaborative approach among researchers, scientific advisors from the National Institute of Public Health of Quebec, QAP project managers, and the QAP ministerial advisor is a major strength of the study [42]. This approach ensured a good understanding of the QAP implementation and formed an alliance between different stakeholders. The research and scientific advisor team could identify the most relevant data, while project managers—as the true points of contact with leading clinicians and managers of different regional health organizations—ensured that the messages were meaningful and well understood by all. The ministerial advisor and the four project managers responsible of the QAP also ensured that the collected and presented data allowed all stakeholders to align with the QAP orientations.

The study also has limitations, particularly in terms of participant acquiescence biases during focus groups. Although the meetings were organized for guidance purposes, managers and leading clinicians from regional health organizations may have felt the need to demonstrate their progress to the ministerial advisor and project managers, as well as to researchers, due to their apparent proximity to the decision-makers and in front of their colleagues. To counter this acquiescence bias, we reiterated the independence of the research team from decision-makers at the beginning of each meeting. Another limitation is the categorization of different regions. Such categorizations are often arbitrary and inconsistent and do not consider the diversity of each region, especially for larger territories, which are often more remote [57]. However, this categorization aligns with that of the Ministry of Health and Social Services and reflects the organization of resources across the different regions of Quebec [50]. Finally, the relatively short time for conducting focus groups (30 minutes) could be considered a limitation. However, in addition to achieving data saturation during focus groups, we conducted these focus groups across all health organizations involved in the QAP and involved several stakeholders (at least one manager was always present, leading clinicians participated in several focus groups, and even research professionals participated in a few), ensuring good representativity of the results.

Considering that the use of dashboards for an Alzheimer Plan is not documented in the scientific literature, future research should focus on the adoption of this tool in different regional health organizations in Quebec and at the governmental level. Studies on the use of dashboards exist for the evaluation of major neurocognitive disorders and the improvement of care offered to these individuals [20-22], but no study specifically mentions the use of a dashboard for an Alzheimer Plan. This will facilitate the transition from an experimental dashboard to a regular tool for managing or monitoring the implementation of an Alzheimer Plan. The composition of focus group participants was different for each region, which may have influenced the results and conclusions. The presence of more leading clinicians in university/peripheral regions may have led to discussions more focused on clinical aspects, while the predominance of managers in mixed or remote/isolated regions may have led to discussions more focused on logistical and organizational challenges related to service delivery in these regions. Other demographic characteristics of participants, such as gender and sex, could have influenced discussions and analyses. It would be interesting for future studies to analyze these differences. Furthermore, no patient or caregiver was part of the focus group, a major element of a learning health system. With increasing incentives to include citizens in health innovations, it will be a great opportunity to study the impact of their perspectives on developing and using dashboards.

In conclusion, dashboards are part of a learning health system and are a very useful tool for reporting on the challenges and issues related to the implementation of an Alzheimer Plan. However, it is important to consider the differences in the utility and information needs of various regions and types of professionals when developing dashboards to enable an adapted, efficient, and equitable implementation of an Alzheimer Plan that extends to a diverse set of organizations with varied resources. Taking these differences into account in the development of dashboards supporting the implementation of an Alzheimer Plan allows for better meeting the needs of all individuals with major neurocognitive disorders and providing optimal and equitable care, regardless of their region of residence.