Introduction
The increasing prevalence of dementia in aging populations has transformed the clinical landscape of acute care hospitals in the past decades (GBD, 2019). Healthcare workers face many challenges associated with providing care to hospitalized older people living with Alzheimer’s disease (AD), Lewy body dementia (LBD), vascular dementia (VaD), and other causes of dementia. Among these challenges, the management of behavioral and psychological symptoms of dementia (BPSD) stands out as one of the most pressing and complex issues faced by frontline healthcare workers (Anantapong et al., Reference Anantapong, Jiraphan, Aunjitsakul, Sathaporn, Werachattawan, Teetharatkul, Wiwattanaworaset, Davies and Sampson2025). BPSD – which encompasses a spectrum of heterogeneous neuropsychiatric manifestations such as agitation, psychosis, apathy, anxiety, and depression, alongside behavioral disturbances like wandering, resistance to care, and medication refusal – affects the vast majority of individuals with dementia over the course of their illness (Lyketsos et al., Reference Lyketsos, Lopez, Jones, Fitzpatrick, Breitner and DeKosky2002). These symptoms often become exacerbated in the unfamiliar and overstimulating environment of the acute care units, where routines are disrupted, informal caregivers are absent, and standardized workflows may fail to accommodate individual cognitive, mental, and physical vulnerabilities (Beardon et al., Reference Beardon, Patel, Davies and Ward2018).
Poorly managed BPSD during hospitalization is not only distressing for patients and families but also carries serious implications for health systems. Evidence links BPSD with prolonged lengths of stay, increased rates of in-hospital complications (e.g., delirium, falls), higher readmission rates, higher discharge rates to institutional long-term care, and considerably higher healthcare costs (Berish et al., Reference Berish, Kuzmik and Boltz2024; Kales, Reference Kales2015). In response to BPSD present in hospitalized older people, hospitals frequently resort to interventions that are misaligned with geriatric best practices, including the use of physical and chemical restraints – despite their well-documented limited benefits and substantial harms (Chou et al., Reference Chou, Hsu, Wang, Chu, Liao, Liang, Chen and Lin2020). These suboptimal responses often reflect gaps in provider training, insufficient staffing, inadequate environmental supports, and the absence of system-level strategies tailored to the needs of cognitively impaired hospitalized older people, as well as instances where regulations are insufficiently known or inconsistently applied. Importantly, staff across disciplines commonly report feeling ill-equipped and unsupported in managing BPSD, highlighting the need for institutional investment in continuous medical education, interdisciplinary care models, and clinical infrastructure that promote dementia-sensitive care (Desmarais et al., Reference Desmarais, Bleau, Bolduc, Champagne, Décary, Fortier, Fournier, Kergoat, Larochelle, Léveillé, Patry, Papillon-Ferland, Rajotte and Taillefer2022). From an administrative standpoint, the absence of robust and standardized quality indicators (QIs) specific to BPSD in acute care represents a critical barrier to improving care. QIs are essential tools for evaluating and enhancing healthcare quality (Evans et al., Reference Evans, Lowinger, Sprivulis, Copnell and Cameron2009). They allow organizations to assess adherence to best practices, identify areas of inefficiency or risk, and track the impact of interventions over time. While QIs are well established in domains such as infection prevention and surgical safety, similar frameworks for the management of BPSD remain undeveloped. Without clearly defined indicators, efforts to standardize care, allocate resources effectively, and drive quality improvement in this domain are severely hindered.
In the province of Quebec, the growing demographic of dementia illustrates the clinical pressures facing healthcare systems. In 2021, approximately 20,500 community-dwelling people aged 65 years and older received a new diagnosis of dementia, representing an incidence rate of 1.3 per cent (INESSS, 2023). That same year, nearly 92,000 individuals (5.5% of the population aged 65 years and older) were living with a diagnosis of dementia and over half (51%) of these older people with dementia visited an emergency department, with more than one-quarter (27%) experiencing two or more visits. One-third were hospitalized for short-term acute care at least once, and 15 per cent of these hospitalizations were followed by a readmission within 30 days (INESSS, 2023). These figures underscore the significant healthcare needs associated with dementia and highlight the importance of establishing robust Qis – particularly for monitoring BPSD – to guide clinical practice and optimize care in acute settings. In Quebec, these challenges arise within a policy context that explicitly recognizes the vulnerability of hospitalized older people. The Approche adaptée à la personne âgée (AAPA) serves as the provincial reference framework for the care of older people in acute care settings since 2011 and is formally endorsed by the Ministère de la Santé et des Services sociaux (MSSS) as a standard of care (MSSS, 2011). AAPA promotes a person-centred, interdisciplinary approach aimed at preventing functional decline during hospitalization by addressing core domains such as cognition and behavior, mobility, nutrition, continence, and sleep. Although AAPA is expected to be implemented across acute care hospitals, its translation into routine clinical practice remains variable, reflecting differences in local resources, organizational structures, and staff training. Furthermore, while provincial performance QIs were previously used to monitor AAPA implementation, these performance measures have been discontinued in 2020 during the COVID-19 pandemic (MSSS, 2020), and no standardized, publicly reported monitoring framework is currently in place. The absence of systematic monitoring may have contributed to unintended declines in the quality of care for hospitalized older people, including greater risk of physical deconditioning and undernutrition. This disconnect between policy intent and measurable practice highlights the need for clearly defined, feasible, and clinically meaningful QIs to support accountability, guide quality improvement efforts, and strengthen the delivery of evidence-based BPSD care in acute care settings.
This study’s primary objectives were to identify currently available measures and indicators in acute care for the monitoring of best practices with regards to BPSD in hospitalized older people and to report the perspectives of healthcare providers on measures that should be prioritized by healthcare institutions to ensure quality of care. The secondary objectives were to identify the perceived facilitators and obstacles to the monitoring of best practices in acute care.
Materials and methods
Study design and context
This observational cross-sectional study utilized an electronic survey design with a deductive qualitative analysis (Fife & Gossner, Reference Fife and Gossner2024), and we followed the Standards for Reporting Qualitative Research (SRQR) (O’Brien et al., Reference O’Brien, Harris, Beckman, Reed and Cook2014) as well as the Checklist for Reporting Results of Internet E-Surveys (CHERRIES) (Eysenbach, Reference Eysenbach2004). The Comité d’éthique de la recherche en sciences et en santé (CERSES) of the University of Montreal was contacted for ethical approval. As a quality care improvement initiative, the project fell under Article 2.5 of the Canadian Tri-Council policy statement, and ethics approval was determined not to be required (CIHR, 2022).
The BPSD in Acute Care Workgroup is a multidisciplinary team composed of clinicians, researchers, and healthcare administrators with expertise in providing optimal care to older people across the continuum of care. Members of the team include geriatricians, specialized registered nurses, pharmacists, academic professors, and healthcare administrators with experience in acute care settings, as well as experts in qualitative and quantitative research, quality of care, and BPSD. Notably, team members bring extensive experience in the clinical management of BPSD, training and supporting frontline healthcare providers, promoting adherence to standards of care, and advocating for improved resources for older person care. This collective expertise provided a deep insider perspective on the systemic challenges of BPSD care. To ensure trustworthiness, the team engaged in ongoing reflexivity, recognizing that their advocacy roles could influence interpretation of the data; this was mitigated through independent double coding, member-checking, and workgroup debriefing. The BPSD in Acute Care Workgroup initiatives have received the support of the Chief Nursing Offices and Directions of the Support Program for the Autonomy of Seniors (SAPA) of healthcare institutions as well as the support of the Direction générale des aînés et des proches aidants and of the Direction générale des affaires universitaires, médicales, infirmières et pharmaceutiques of the Ministère de la Santé et des Services sociaux (MSSS) of the province of Quebec, Canada.
Participant recruitment
A convenience sample of physicians, nurses, and healthcare administrators working within acute care units (i.e., medicine, surgery, geriatric units, and emergency departments) was recruited across general and specialized care hospitals of the province of Quebec. Participation was voluntary, without incentives, and anonymous. Invitations to participate, which included an information letter about the purpose of the study and the mission of the BPSD in Acute Care Workgroup, were sent by email to healthcare institutions’ program directors and chiefs of service in early January 2025. Program directors and Head of Care Units were instructed to identify within their institution potential participants to whom forward the invitation to partake in the survey. Potential participants had to be healthcare professionals working on medical, surgical, mixed, geriatric wards or in the emergency department, to have experience with the implementation and monitoring of best practices with regard to providing care to hospitalized older people (i.e., person ≥65 years old), and to be involved directly or indirectly with the management of BPSD or similar problematics in the context of delirium. Reminders to participate were sent by email in early February 2025.
Data collection
The survey was conducted online on the Survey Monkey© platform. Participants were asked to complete a semi-structured survey that was developed by the multidisciplinary team based on a recently published scoping review of the literature (Desmarais et al., Reference Desmarais, Bolduc, Bergeron, Desforges, Gauthier, Décary, Patry and Kergoat2025) and pilot tested on healthcare providers with varying functions to ensure clarity, readability, and content validity. The survey was in French, composed of five consecutives sections with fixed (nonadaptative) questions (i.e., 1. Purpose of the survey; 2. Definitions and examples of healthcare indicators; 3. Respondent’s characteristics; 4. Currently available measures and quality indicators for BPSD; and 5. Quality indicators that should be used) and took between 30 and 45 minutes to complete (see Supplementary File). A total of 20 close and open-ended questions were developed to capture measures and potential QIs falling within five a priori major categories of indicators, in keeping with the multimodal considerations of BPSD and the team’s experience: i. Screening and reporting; ii. Assessment; iii. Management; iv. Impacts and hospitalization outcomes; v. Investments and resources. Participants were also asked to report perceived facilitators and obstacles to the monitoring of best practices about BPSD in acute care. The online survey was opened for six weeks, from January 13, 2025, to February 21, 2025.
Data analysis
Data analysis was performed following Creswell and Poth’s data analysis model (Reference Creswell and Poth2018) via manual coding and thematic development, without the aid of a qualitative analysis software. Firstly, received responses from all participants were transferred into a single document to facilitate the team’s familiarization with the data, analysis, and classification of responses. The survey platform’s log file was systematically reviewed to detect and prevent duplicate entries by cross-checking time stamps, IP addresses, and session identifiers, thereby ensuring that each participant was represented only once in the data set. Secondly, received responses were read by the members of the workgroup. Participants who did not provide any responses to the questions in sections 4 and 5, which were the focus of the present analysis, were excluded to ensure that only complete and analysable data were retained. Thirdly, for each participant, two reviewers independently read their survey’s responses, transcribed them for clarity and uniformity into themes, and classified them manually according to our five major categories. We also further classified identified themes manually into three main types of indicators according to the framework of Donabedian: (1) Structural QIs; (2) Process QIs; (3) Outcome QIs (Donabedian, Reference Donabedian1988) (see Figure 1). Two leads of the workgroup (AB, PD) were responsible for accuracy and reliability of data analysis, which was ensured through performing validation of responses’ transcription into themes and classification of these within the five major categories and three main types of QIs. Conflictual responses were resolved by the main team lead (PD) after having revisited the raw data (e.g., participant’s responses verbatim) and consensus discussion between all reviewers who conducted coding. Fourthly, the two leads of the workgroup (AB, PD) synthesized and interpreted the themes into distinct coherent overall themes. Fifthly and lastly, findings were summarized into tables according to the deductive framework.
Conceptual framework of quality indicators in acute.
Figure 1. Long description
The diagram is organized into two main sections.
At the top, five gray rectangular boxes are arranged horizontally, connected by small black bars to indicate a sequence. From left to right, they are labeled:
i. Screening and reporting.
ii. Assessment.
iii. Management.
iv. Impacts and hospitalization outcomes.
v. Investissement and resources.
Below this sequence are three large, light-gray horizontal blocks stacked vertically, representing the types of indicators.
1. Structural indicators: Measures indicating the availability of structures and resources that enable the delivery of care.
2. Process indicators: Measures reporting that specific activities and actions are performed.
3. Outcome indicators: Measures reporting the end results or effects of interventions and care provided.
Results
In total, 38 personalized invitations to participate were sent to program directors and heads of service of 26 different healthcare institutions, all of whom agreed to forward the invitation to key stakeholders within their respective institutions. The online survey was accessed 114 times, and 55 participants provided sufficient responses to be analysed (completion rate of 48%). Participants included Heads of Care Unit (40%), supervisors of the specialized approach to senior care (29.1%), physicians (12.7%), clinical administrators (9%), and registered nurses (3.6%), with varied years of experience in providing care to older people (Table 1). Participants worked in different institutions, acute care units, and settings, including university-affiliated hospitals (40%) and urban setting (63.6%). Overall, responses of stakeholders working in 26 different institutions, from 12 different sociosanitary regions of the province of Quebec, were obtained.
Survey respondents
Table 1. Long description
The table consists of two columns: Characteristics and Respondents N = 55.
Main function/occupation:
- Head of care unit: 22 (40%)
- Supervisor of the specialized approach to senior care (A A P A): 16 (29.1%)
- Physician: 7 (12.7%)
- Clinical advisor/administrator: 5 (9%)
- Registered nurse: 2 (3.6%)
- Not specified: 3 (5.4%)
Years of experience:
- Less than 3 years: 20 (36.4%)
- 3 to 5 years: 17 (30.9%)
- 6 to 8 years: 5 (9%)
- 9 to 11 years: 1 (1.8%)
- 12 to 14 years: 6 (10.9%)
- 15 years plus: 6 (10.9%)
Main acute care unit (respondents could report multiple):
- Geriatric medicine: 12 (21.8%)
- Medicine: 12 (21.8%)
- Surgery: 8 (14.5%)
- Emergency department: 5 (9.1%)
- Across the institution (in charge of the A A P A): 27 (49.1%)
Type of institution:
- University hospital: 22 (40%)
- Nonuniversity hospital: 31 (56.3%)
- Not specified: 2 (3.6%)
Setting:
- Urban: 35 (63.6%)
- Suburban or rural: 18 (32.7%)
- Not specified: 2 (3.6%)
* Some respondents reported working on several different acute care units.
Abbreviation: AAPA, Approche adaptée à la personne âgée (specialized approach to senior care).
Participants provided a total of 677 responses, of which 392 responses (57.9%) pertained to currently available measures and indicators in acute care for the monitoring of best practices with regard to BPSD in hospitalized older people and 285 responses (42.1%) pertained to measures that should be prioritized by healthcare institutions to ensure quality of care. Most responses were categorized as potential QIs for monitoring specifically the management of BPSD (185 responses; 27.3%), followed by QIs for monitoring institutional investments and resources (120 responses; 17.7%), for the assessment (119 responses; 17.6%), for the impacts and hospitalization outcomes (113 responses; 16.7%), and for the screening and reporting (93 responses; 13.7%).
Available measures and indicators in acute care for BPSD
Participants’ responses were synthesized and interpreted into 59 distinct recurrent themes, which are summarized in Table 2 (Structural QIs), Table 3 (Process QIs), and Table 4 (Outcome QIs). For measures pertaining to screening and reporting, most responses touched on monitoring healthcare workers practices of systematically screening for delirium and appropriate documentation of patient’s cognitive and behavioral status. The Confusion Assessment Method (CAM) was the most frequently cited tool for screening delirium by responders. As well, participants mentioned monitoring the number of patients with a diagnosis of delirium, agitation, or dementia in the ED or on the wards during a specific period. For measures pertaining to assessment, most responses touched on monitoring metrics related to the involvement of specialized healthcare workers such as mobile team dedicated to BPSD, specialized nurses, geriatric medicine, psychiatry, and geriatric psychiatry, and utilization rate of in-house protocols, algorithms, and standardized prescriptions for the assessment of older hospitalized people with delirium or BPSD. For measures pertaining to management, respondents chiefly mentioned the monitoring of metrics related to the use of physical restraints and the prescription of pharmacological interventions, such as benzodiazepines and antipsychotics. Participants also mentioned metrics related to the practice of involving specific healthcare professionals and adherence to patient personalized care plan or program. For measures pertaining to the impacts and hospitalization outcomes, respondents recurrently mentioned the length of stay (68 mentions) as a fundamental outcome indicator. Other themes included the number of incidents involving an agitated patient (e.g., aggression towards other patients or healthcare workers) and the cost of constant observation practices. For measures pertaining to investments and resources, responses touched on monitoring the availability within the institution of tools, material, protocols, algorithms, and standardized prescriptions for delirium and BPSD as well as the availability within the institution of specialized healthcare workers with an expertise in BPSD and of training programs for healthcare workers.
Reported available and priority structural quality indicators in acute care for the monitoring of best practices for BPSD
Table 2. Long description
The table contains a single column titled Investments and resources. The list includes the following bulleted items, where an asterisk indicates a priority indicator:
* Availability of protocols, algorithms, and standardized prescriptions for the prevention and management of delirium*
* Availability of protocols, algorithms, and standardized prescriptions for the prevention and management of B P S D*
* Availability of a mobile team dedicated to B P S D*
* Availability of nurses specialized in geriatrics and of clinical advisors with expertise in B P S D*
* Availability of specialists (geriatric medicine, psychiatry, geriatric psychiatry)*
* Availability of a special care unit for B P S D
* Efficient staffing with nurse-to-patient ratios that consider the complexity of caring for patients with B P S D*
* Availability of training programs on delirium, dementia, and B P S D*
* Availability of material for the management of wandering (e.g., electronic monitoring bracelet)
* Availability of material for the development and implementation of nonpharmacological interventions for B P S D*
* Dedicated awareness days for delirium, dementia, and B P S D*
* Availability of information systems and technology allowing the monitoring of quality indicators
* Availability of information systems and technology allowing protocols, algorithms, and standardized tools to be readily available to healthcare workers*
* Educational material for healthcare workers
* Educational material for patients, families, and caregivers
* Environmental adaptation of wards for older patients*
B P S D stands for behavioral and psychological symptoms of dementia.
* Structural quality indicators that are deemed priority by respondents.
Abbreviation: BPSD, behavioral and psychological symptoms of dementia.
Reported available and priority process quality indicators in acute care for the monitoring of best practices for BPSD
Table 3. Long description
The table is organized into three primary sections.
1. Screening and reporting:
- Number of patients screened for delirium in the Emergency Department.
- Number of patients screened for risk of postoperative delirium.
- Utilization rate of standardized screening tools such as C A M, R A D A R, 4 A T, A M T 4, P R I S M A-7, Edmonton scale, and I S A R in the Emergency Department and on the wards.
- Documentation of the patient’s baseline level of attention in medical charts.
- Documentation of the patient’s risk factors for delirium in the medical charts.
- Documentation of delirium, dementia, and B P S D in discharge notes or medical summary.
- Documentation of complications during hospital stay in discharge notes or medical summary.
2. Assessment:
- Number of consultations to the mobile team dedicated to B P S D.
- Number of consultations to the specialized nurse or clinical advisor.
- Number of consultations to medical specialists including geriatric medicine, psychiatry, and geriatric psychiatry.
- Utilization rate of protocols, algorithms, and standardized prescriptions for the assessment of delirium.
- Utilization rate of protocols, algorithms, and standardized prescriptions for the assessment of B P S D.
- Utilization rate of standardized tools for the characterization of B P S D using validated tools like C M A I, N P I, and behavioral observation grids.
- Adherence and completeness of the assessments at admission, such as geriatric comprehensive assessment, A I N E E S signs, pain assessment, patient’s life story, and frailty.
- Quality of medication review performed.
3. Management:
- Number of consultations to experts and advisors external to the institution.
- Number of consultations to second and third line specialized experts.
- Adherence and completeness of standardized tools for the longitudinal monitoring of B P S D.
- Adherence to the Hospital Elder Life program, H E L P.
- Adherence to the nursing care plan.
- Re-assessment and management of pain.
- Involvement of family members and informal caregivers.
- Documentation of implemented interventions to mitigate B P S D in medical charts.
* Process quality indicators that are deemed priority by respondents.
Abbreviations: 4AT, Screening for delirium and cognitive impairment; AINEES signs, Autonomie, intégrité, nutrition, élimination, état cognitif, sommeil; AMT4, Abbreviated Mental Test 4; BPSD, behavioral and psychological symptoms of dementia; CAM, confusion assessment method; CMAI, Cohen-Mansfield agitation inventory; HELP, Hospital Elder Life Program; ISAR, identification of seniors at risk; NPI, neuropsychiatric inventory; PRISMA-7, Program of research to integrate the services for the maintenance of autonomy; RADAR, recognizing acute delirium as part of your routine.
Reported available and priority outcome quality indicators in acute care for the monitoring of best practices for BPSD
Table 4. Long description
The table is organized into three main sections.
Section 1. Screening and reporting. This section includes six indicators:
• Number of patients diagnosed with delirium in the Emergency Department on the wards.
• Number of patients hospitalized for delirium.
• Number of Emergency Department visits for B P S D or dementia-related issues.
• Number of patients hospitalized mainly for the management of B P S D or dementia-related issues.
• Number of patients hospitalized with a diagnosis of dementia as a comorbidity.
• Number of patients with agitation in the Emergency Department on the wards.
Section 2. Management. This section includes four indicators:
• Number of patients who received physical restraints and duration of physical restraints use.
• Number of patients who received a pharmacological intervention for agitation as needed.
• Number of patients who received a pharmacological intervention regular prescription for agitation.
• Number of patients who received a behavioral or nonpharmacological intervention to mitigate B P S D.
Section 3. Impacts and hospitalization outcomes. This section includes 14 indicators:
• Number of incidents and accidents such as aggression towards other patients or healthcare workers.
• Number of code white and involvement of security agents.
• Number of ward or room transfers for B P S D or delirium.
• Number of older people with a stay in the Emergency Department longer than 48 hours or 72 hours.
• Number, duration, and cost of constant observation practices.
• Number of adverse drug reactions.
• Number of healthcare workers mobilized.
• Number of care hours provided by private help.
• Length of hospital stay, length of stay in the Emergency Department, and hospitalization costs.
• Percentage of inpatients with alternate level of care status.
• Re-admission for delirium or B P S D.
• Relocation following hospitalization, including relocation to a special care unit for B P S D.
• Level of healthcare workers satisfaction with working environment conditions.
• Level of patient and caregiver satisfaction with care received.
* Outcome quality indicators that are deemed priority by respondents.
Abbreviation: BPSD, behavioral and psychological symptoms of dementia.
Priority measures and indicators in acute care for BPSD
Of the 59 initial distinct recurrent themes identified, 44 themes were mentioned by participants as measures that should be prioritized by healthcare institutions (flagged in Tables 2–4). For screening and reporting, participants reiterated the importance of monitoring the process of screening for delirium and use of standardized tools as well as the adequate documentation in medical notes. For assessment, participants’ responses highlighted the importance of monitoring the involvement of specialized professionals and the use of standardized protocols, algorithms, tools, and prescriptions. For management, participants’ responses touched more on outcomes indicators than process indicators, emphasizing the monitoring the requirement of physical restraints and pharmacological interventions. For impacts and hospitalization outcomes, participants emphasized the importance of monitoring adverse events related to BPSD and delirium, including incidents and accidents, code white interventions (i.e., responses to actual or potential violence), and the use and costs of constant observation practices. In addition, participants underscored the need to track hospital performance metrics such as the length of stay in hospital and in the Emergency Department, hospitalization costs, relocation following hospitalization – including relocation to specialized care units for BPSD – as well as re-admission rates. Respondents also pointed out that the perspectives of both healthcare workers and patients/caregivers should be incorporated through measures of satisfaction, feelings of competency, and burnout, reflecting the human and organizational impact of BPSD management in acute care. For investments and resources, participants prioritized monitoring the institutional capacity to deliver best practices through the availability of protocols, algorithms, and standardized prescriptions for both delirium and BPSD. They also stressed the importance of resource availability, including specialized personnel (mobile teams, specialized nurses, geriatric medicine, psychiatry, geriatric psychiatry), training programs, adequate nurse-to-patient ratios, and access to nonpharmacological materials and interventions. Structural factors such as environmental adaptation of units for hospitalized older people and the availability of information systems to support clinical decision-making and quality monitoring were also identified as priority measures.
Transversal themes
Participants were prompted to specify in their responses how QIs were measured (i.e., operationalized), whether their institution had specific objectives (i.e., thresholds) regarding these QIs, who was responsible for collecting these measures, and who was accountable of results. Respondents reported three general approaches for data collection to support the calculation of QIs for the monitoring of BPSD-related care. Firstly, several indicators are collected manually at predefined intervals (e.g., daily or by fiscal period) through various sources, such as unit managers, assistant head nurses, nursing advisors, sector heads, staff nurses, and medical records departments. Secondly, annual or biannual audits are conducted either on samples of medical records or on all patient records meeting the target criteria on a designated day (once or twice per year). Thirdly, specific events – including incidents and accidents, code white events, and the use of restraints – must be reported to a provincial registry. In some institutions, portions of the data collection process are computerized. For example, a Microsoft Power BI Delirium module is used to measure the incidence of delirium based on discharge summary data; however, respondents noted that this approach is complex, with potential limitations related to incomplete summaries, underdiagnosis, and variable data reliability. A Microsoft Power BI module is also used to monitor indicators related to restraint use, with data generated automatically from the electronic restraint reporting form. Other institutions rely on healthcare information systems to extract the number of hospitalizations where delirium is coded as the primary diagnosis, with data collection performed by the medical records department. Interventions carried out by specialized teams, such as AAPA (Approche adaptée à la personne âgée – Specialized approach to senior care), are recorded through Engagement, Governance, Access, and Protection (EGAP) intervention-entry software. Completion of online training modules is monitored through local or provincial electronic learning management systems (ENA) that are linked to the employee’s personnel file. Participants frequently emphasized that, within their institutions, explicit objectives or performance thresholds for QIs specific to BPSD were generally lacking; instead, institutional priorities were often oriented towards broader operational targets such as occupancy rates and length of stay in the emergency department and inpatient units, suggesting that BPSD-related indicators may not yet be fully integrated into institutional performance frameworks.
Facilitators and barriers to the monitoring of quality of care
Respondents’ perceptions of facilitators and obstacles to the monitoring of QIs in acute care are reported in Table 5. Participants highlighted several facilitators that can enable successful monitoring of BPSD in acute care. Notably, participants emphasized the critical role of leadership accountability and commitment, pointing to the value of designated ‘champions’, managerial engagement, and clear lines of responsibility. Participants underscored that without explicit prioritization of BPSD care from hospital leadership, sustained implementation of QIs is unlikely. Participant #21 stated: ‘Facilitators include adopting a non-judgmental approach, conducting awareness-raising activities, and having an implementation program that is meaningful. It is important to value the practice of caring for older people, making it everyone’s responsibility rather than solely that of the Directorate for Older Adult Support’. Training also emerged as a recurring theme; staff called for structured and ongoing education – not only on clinical aspects of BPSD but also on the use and relevance of indicators themselves. Participant #43 wrote: ‘The continuous training of healthcare teams on the assessment and management of delirium and BPSD is important, as is on the appropriate use of quality indicators’. Importantly, many viewed interdisciplinary collaboration and support from specialized geriatric and psychiatric teams as necessary to bridge practice gaps while also identifying opportunities to embed indicator use into existing organizational routines, such as huddles and audits. These insights reflect a desire for practical integration of best practices within clinical workflow rather than the creation of parallel, burdensome systems.
Perceived facilitators and obstacles to the monitoring of quality indicators in acute care
Table 5. Long description
The table is divided into two main sections: Facilitators and Obstacles.
Facilitators section includes:
* Leadership and accountability: Designate a responsible person skilled in B P S D care; clear commitment from nursing directors and administrators; appoint project champions; active managerial engagement.
* Training and skills development: Initial and ongoing training on delirium and B P S D; awareness programs; involvement of clinical nurse consultants.
* Clinical and organizational support: Specialized mobile teams like geriatricians and psycho-educators; B P S D and A A P A committee; interdisciplinary partnerships.
* Organizational practices: Clear communication; systematic processes for monitoring; supportive culture; integration into daily workflow like staff huddles.
* Human and material resources: Trained, stable staff; adjusted staffing ratios; addition of volunteers.
* Physical and environmental factors: Adapted environments with noise reduction; facilitating family presence.
* Technology and digitalization: Use of technology to reduce double data entry; E H R access; simple electronic tools; automation of reporting.
* Continuous monitoring and quality improvement: Regular audits; monitoring results; use of simple indicators.
Obstacles section includes:
* Leadership, priorities, and institutional culture: Dilution of responsibility; institutional ageism; frequent protocol changes; lack of alignment with ministry priorities; resistance to indicators.
* Human resources: Staff shortages; high turnover; lack of trained personnel; staff instability.
* Training and knowledge gaps: Lack of knowledge about geriatric syndromes; misunderstanding of B P S D versus mental illness; lack of mandatory training.
* Time constraints and workload: Severe time constraints; excessive workload; time-consuming manual data entry.
* Financial and material resources: Lack of funding; restricted budgets; limited administrative support.
* Physical and environmental limitations: Inadequate physical environment like shared rooms; overstimulating settings.
* Technology and data issues: Non-computerized data; difficulty extracting data; lack of reliable data; complex tools.
* Change management and implementation barriers: Resistance to change; lack of ownership; competing priorities where geriatric care is sidelined; lack of systematic follow-up.
Abbreviations: AAPA, Approche adaptée à la personne âgée (specialized approach to senior care); BPSD, behavioral and psychological symptoms of dementia; EHR, electronic health record.
Conversely, participants reported a range of systemic and operational obstacles that limit the feasibility of BPSD monitoring in real-world settings. Barriers extended beyond individual knowledge or motivation to include broader structural and cultural challenges. Institutional ageism and a lack of leadership prioritization for geriatric care were perceived as major deterrents, compounded by staffing shortages, high turnover, and lack of specialized training in BPSD. Participant #33 wrote: ‘The barriers are multiple and include staffing shortages, a lack of commitment from decision-makers to prioritize the evaluation of practices, and the absence of training programs to help teams use indicators effectively’. These workforce issues were further aggravated by excessive workloads and time constraints, making additional documentation or data entry for indicators feel unrealistic. Participant #53 stated: ‘There is a lack of human and financial resources, a lack of time… the clinical complexity of BPSD cases is significant… and as with all organizational projects and changes, change management is complex’. Respondents also expressed frustration with technological fragmentation – citing difficulties with data extraction, absence of automated tools, and a general lack of integration with electronic health record (EHR) systems. Many voiced concerns that current indicators fail to capture the clinical complexity of BPSD and that implementation efforts often lack sustained follow-up or engagement. Overall, findings suggest that while healthcare workers recognize the value of BPSD monitoring, its success hinges on coordinated institutional investment, resource allocation, and alignment with the realities of acute care practice.
Discussion
This study aimed to report the currently available measures and QIs in acute care for the monitoring of best practices with regard to BPSD in hospitalized older people as well as to report the perspectives of stakeholders on measures that should be prioritized. Building on our prior scoping review (Desmarais et al., Reference Desmarais, Bolduc, Bergeron, Desforges, Gauthier, Décary, Patry and Kergoat2025), which identified 88 BPSD-related QIs reported in the conventional and grey literature, the present study confirms substantial alignment between published indicators and those identified as relevant by clinicians and healthcare managers in real-world acute care settings within the province of Quebec. Importantly, while the scoping review provided a comprehensive mapping of existing measures and potential indicators, the current study extends this work by characterizing these QIs in greater depth, including their perceived clinical relevance, priority, feasibility of measurement, and current use across institutions. This study demonstrates that a number of QIs are currently available, in use, or could be readily measured by healthcare institutions, with most considered high priority for monitoring the screening, assessment, and management of BPSD in older people with dementia during acute care hospitalizations. In contrast to the literature-based scoping review, stakeholder perspectives highlighted practical considerations that are often underreported in published studies, including the importance of feasibility, the potential for automated extraction from EHR, and the influence of institutional culture and local resources on implementation. According to the healthcare managers and clinicians consulted, both the methods for data collection and the ways these data are used lack standardization. This variability in practice reduces comparability across institutions and may limit the utility of existing indicators for benchmarking or quality improvement purposes (Sinsky et al., Reference Sinsky, Bavafa, Roberts and Beasley2021).
Stakeholders expressed a strong preference for QIs that are straightforward to compile, specific to BPSD, and, wherever possible, automatically generated by EHR or other automated measurement tools, rather than collected manually. Manual compilation is not only time-consuming but also susceptible to transcription errors, inconsistency in interpretation, and delays in reporting (Yin et al., Reference Yin, Guo, Sholle, Rajan, Alshak, Choi, Goyal, Jabri, Li, Pinheiro, Wehmeyer, Weiner, Safford, Campion and Cole2021). In contrast, automated extraction from structured EHR fields could facilitate real-time monitoring, reduce staff workload, and improve the timeliness of quality improvement interventions (Haerian et al., Reference Haerian, Varn, Vaidya, Ena, Chase and Friedman2012).
According to the Canadian Institute for Health Information (CIHI), emerging digital health solutions bring both opportunities and challenges for data quality (CIHI, 2016). On one hand, newer systems can incorporate standardized definitions, decision-support prompts, and built-in quality checks, potentially transforming workflows and significantly reducing the burden of data collection. They can also enable integration of multiple data sources, such as prescribed medication, implemented nonpharmacological interventions, laboratory results, imaging, and clinical assessments, into a single analytic framework. On the other hand, these systems may introduce new complexities, including the coexistence of multiple, sometimes conflicting, data sources or the presentation of information in nonchronological formats, issues that can complicate tasks such as determining the temporal relationship between comorbidities and hospital complications (CIHI, 2016). Furthermore, differences in system configuration, interoperability, and user training across healthcare organizations can lead to inconsistencies in how data are recorded and interpreted. To maximize the benefits of automated data capture, organizations responsible for care quality should collaborate closely with data providers, EHR vendors, and clinical teams. Such collaboration should aim to ensure that automated BPSD-related indicators are not only feasible to generate but also clinically meaningful, consistently defined and operationalized, and aligned with established reporting frameworks. This approach could enhance comparability across sites, support large-scale surveillance and research, and ultimately improve the quality of care for older people with dementia in acute care settings.
In this context, the interRAI suite of standardized clinical assessment instruments offers a unique opportunity to support quality measurement across the continuum of care for older people with dementia (Hirdes et al., Reference Hirdes, Ljunggren, Morris, Frijters, Finne Soveri, Gray, Björkgren and Gilgen2008). interRAI is widely implemented in Canada and internationally in continuing care settings, including long-term care and home care, where standardized assessments underpin validated QIs related to cognition, mood and behavior, restraint use, and potentially inappropriate antipsychotic prescribing. These indicators are routinely used for benchmarking, accountability, and quality improvement at organizational and system levels. In contrast, despite the availability of interRAI instruments designed for acute care (Gray et al., Reference Gray, Beattie, Boscart, Henderson, Hornby-Turner, Hubbard, Wood and Peel2018), their uptake and use for systematic quality monitoring in hospital settings remain limited, resulting in a fragmentation of quality measurement at key transition points. The BPSD-related QIs identified in the present study could help bridge this gap by aligning acute care measurement with established interRAI constructs, thereby facilitating continuity in the assessment and monitoring of behavioral symptoms across settings. Conversely, interRAI’s standardized definitions, assessment items, and risk-adjustment methodologies could inform the operationalization and implementation of acute care–specific BPSD QIs, supporting harmonized reporting, improved comparability across institutions, and more coherent monitoring of care quality for people living with dementia throughout their healthcare trajectory.
Strengths and limitations
Our study has several strengths worth highlighting. The recruitment strategies, purposeful and intentionally designed, were successful in engaging participants from a wide range of healthcare institutions across the province of Quebec. We obtained the viewpoints of stakeholders from the most populous sociosanitary regions and from integrated health and social services centres (CISSS) and integrated university health and social services centres (CIUSSS) with the population-dense catchment areas in the province. This approach allowed us to gather the perspectives of a diverse group of stakeholders, resulting in information-rich responses that reflect varied organizational contexts. The trustworthiness of our findings is further reinforced by rigorous qualitative methodology, including team-based consensus review through regular workgroup meetings at multiple stages of analysis, specifically at the transcription, classification, synthesis, and summarization of findings stages, which enhanced credibility and minimized interpretive bias. Nonetheless, certain limitations should be acknowledged. While all respondents were healthcare providers involved directly or indirectly in the care of older people, we did not capture the perspectives of medical archivists or professionals involved in health data governance and valorization, groups that may have been aware of other relevant, but less visible, QIs. Similarly, most respondents held leadership or supervisory roles, which may have resulted in perspectives that reflect managerial or consultative viewpoints rather than those of frontline staff; consequently, the QIs identified and prioritized could be perceived by bedside clinicians as less applicable or less desirable in daily practice, potentially limiting their transferability and affecting real-world implementation. Furthermore, we did not specifically recruit stakeholders from intensive care units or psychiatric wards, as these settings typically operate under different clinical objectives, patient profiles, and resource structures. Similarly, although specialized BPSD units provide care specifically to individuals with BPSD, stakeholders from these units were not recruited in the present study, as these units function within distinct care models, programs, and organizational contexts; consequently, the QIs used in such settings may differ from those identified here. In addition, because recruitment relied on program directors who agreed to forward the invitation, the number of eligible stakeholders who received the survey is unknown, precluding calculation of response rates by institution or profession and raising the possibility that some relevant perspectives were not captured. Additionally, the study did not include participants from Quebec’s northern sociosanitary regions; however, this is partly due to the fact that these regions primarily rely on multimission health centres rather than standalone acute care hospitals. This may limit the transferability of our findings to those acute care contexts. As well, while our study provides a comprehensive view of the Quebec healthcare landscape, the findings may not be fully representative of other provinces or international contexts where the integration of social and health services (e.g., the CISSS/CIUSSS model) differs significantly. Additionally, our study focused on healthcare stakeholders and did not incorporate the perspectives of patients or informal caregivers, whose lived experiences could have provided complementary insights into the relevance and feasibility of the identified QIs. Finally, as in most survey-based studies, participation was voluntary, raising the possibility of response bias whereby individuals with greater interest in BPSD care or quality measurement may have been more likely to respond.
Conclusion
This study provides a comprehensive overview of current practices and priorities regarding QIs for the monitoring of best practices of managing BPSD in older people with dementia in acute care. The indicators identified through this qualitative work should be considered candidate QIs, reflecting stakeholder perspectives on relevance and feasibility rather than a finalized or standardized set. While several indicators are already in use and widely regarded as high-priority, their collection and application remain variable across institutions, with limited standardization, operational definitions, and implementation strategies. To address these gaps, the BPSD in Acute Care Workgroup will undertake a structured Delphi process with a multidisciplinary panel of clinical, methodological, and health system experts. This next step aims to systematically prioritize candidate QIs and advance their operationalization, standardization, and potential validation towards a core set of indicators suitable for implementation in Quebec and, ultimately, Canadian acute care settings. Stakeholders expressed a strong preference for simple, clinically meaningful indicators that can be automatically generated from EHR data, thereby reducing workload and enhancing timeliness. Moving forward, coordinated efforts among clinicians, quality-of-care agencies, data providers, and EHR developers will be essential to embed these consensus-based indicators into routine data systems, enabling reliable monitoring, benchmarking, and targeted quality improvement and fostering more consistent, person-centred care for older people with dementia in acute care.
Supplementary material
The supplementary material for this article can be found at http://doi.org/10.1017/S0714980826100622.
Data availability statement
The data that support the findings of this study will be kept for a minimum of 5 years after collection and are available upon reasonable request at the discretion of the corresponding author.
Acknowledgements
The authors would like to acknowledge Sylvain G. Bélisle for his assistance with creating the tables and figure.
Author contribution
P.D., A.B., and M.C. designed the study. P.D., A.B., and M.C. analysed the data. P.D. and A.B. drafted the manuscript. All authors contributed significantly to the content, critically reviewed, and approved the final manuscript for publication.
Financial support
The authors declare none.
Competing interests
The authors declare none.
