Abstract
The COVID-19 pandemic restricted access to health and social care for older adults. In response, a standardized, self-report instrument, the interRAI COVID-19 Vulnerability Screener (CVS), was developed. In collaboration with a community support service organization, this project aimed to evaluate a surveillance process to identify those at risk and triage them to health- or social-care services. A convergent, mixed methods design was used. Virtual focus groups were conducted with partner staff. The clinical, social, and functional characteristics of screened clients were analyzed descriptively. The mixed methods analysis generated implementation considerations. Participants successfully utilized the CVS to identify and support vulnerable clients who may have otherwise been overlooked. Most screened clients were not experiencing COVID-19 symptoms, yet had elevated mortality risk should they become infected, and were negatively impacted by social isolation. Findings support the use of the CVS by community support services during the pandemic or other disasters to identify those at risk due to frailty and social or economic vulnerability. Implementation requirements include a knowledgeable workforce, active facilitation, and incorporation of the CVS into existing workflows. Health- and social-care integration would be enhanced by the development of updated privacy policies and shared digital infrastructure.
1. Introduction
Older adults with multiple chronic conditions have complex health- and social-care needs and require frequent interactions with all sectors of the health system [1, 2]. These sectors and their providers tend not to be coordinated or connected, contributing to poor health outcomes and experiences for older adults and their caregivers [3]. This fragmentation has been exacerbated by the ongoing COVID-19 pandemic and social-distancing measures, further threatening the health, functional, cognitive, and social status of older adults [4, 5]. During the first waves, older adults were disproportionately represented in requiring intensive care, dying from the infection, contracting COVID-19, and for those with multiple chronic conditions, having more severe cases [4, 6–9]. Social distancing measures have meant many older adults are experiencing social isolation, which is associated with depression, anxiety, and cognitive decline [10, 11]. Restrictions on community mobility placed greater demands on family caregivers and limited access to health-promoting programs [5, 12]. Additionally, food insecurity has increased during the pandemic [13].
In each wave of the pandemic, there were massive disruptions that restricted access to health-care services to only those experiencing emergent issues in acute care and reduced access to routine management of chronic conditions in primary care [14, 15]. In addition to reduced availability and delivery of home-care services, some older adults also declined services due to a fear of COVID-19 transmission [16]. Community support services (CSS) were particularly impacted during the pandemic due to reliance on older volunteers, initial lack of access to personal protective equipment and immunizations, and limited access to technology [17]. CSS organizations provide essential services to maintain older adults safely in their homes, such as meal and medication delivery, transportation, adult day programs, homemaking, home maintenance, practical in-home support, friendly visiting, community dining, social and recreational programs, exercise and fall prevention classes, and information and referral to other services.
Thus, the pandemic has created an urgent need for ongoing, proactive surveillance to identify older adults who are at risk of negative consequences due to physical or cognitive impairments and triage them to the required health- or social-care service provider. However, most surveillance tools that have been used during the pandemic only assess for COVID-19 symptoms and are lacking a way to identify and monitor the broader impact of the pandemic on older adults. In response to this limitation, the interRAI COVID-19 Vulnerability Screener (CVS) was developed by an international collaboration of researchers and care providers as a brief standardized screening instrument [18, 19]. The CVS was created based on two validated interRAI instruments (i.e., the interRAI Home Care and interRAI Check-Up) and a review of 15 international self-assessment tools for COVID-19 symptoms [18, 20–22]. The CVS is designed to be conducted virtually by a clinician or lay individual (e.g., family member, volunteer, administrative staff). It is a self-reported assessment, constructed to obtain the older adult’s perspective on their well-being, but may be completed with support from a caregiver, or by a caregiver on behalf of the older adult. The CVS has four main measurement functions: (a) tracking COVID-19 symptoms, (b) frailty measurement, (c) mortality risk estimation, and (d) measures of social and economic vulnerability [18]. Information from the CVS and its automatically generated clinical outputs can be used to identify persons in need of urgent medical attention or a comprehensive geriatric assessment by a clinician and initiate planning to monitor risks related to underlying medical issues, functional difficulties, mood problems, social isolation, caregiver distress, or financial strain [18].
The aim of this study was to design, implement, and evaluate a surveillance process utilizing the interRAI CVS to identify older adults at risk and connect them to the required care, leveraging support from a CSS organization. CSS providers are ideally positioned to conduct surveillance during the pandemic given they support many vulnerable older adults. Lessons learned from the program implementation and evaluation can inform how integrated health and social care can be successfully accomplished in the community, in addition to identifying some facilitating practices and policies.
2. Research Methods
This project used a convergent, mixed methods research design, informed by the integrated Promoting Action on Research Implementation in Health Services (iPARIHS) framework [23–25]. The iPARIHS is an organizing framework that helps to explain why the implementation of evidence-informed interventions is successful through an examination of the interacting constructs of the innovation, recipients, inner and outer contexts, and the process of facilitation [23]. The overarching research question was as follows: What are the key components of an implementation process for a social-care organization to conduct surveillance of older adults utilizing the interRAI CVS?.
A convergent design implements both quantitative and qualitative strands at the same time in the research process and then uses information learned in both strands to answer the study’s research question [24]. Mixed methods designs provide some special advantages when utilized in implementation research. In particular, they allow researchers to account for the unique context and how the intervention was implemented, which are essential components in the success or failure of an intervention [23, 26]. Through a combination of population analysis and purposive sampling, mixed methods blend statistical data with the experiential knowledge of key informants to unpack the problematic “black box” of implementation research [p. 318, 27]. The use of a mixed methods design also facilitates the recognition and identification of unintended consequences that arise during the implementation—something that is often missed in solely quantitative assessments [27].
3. Implementation Setting and Sample Recruitment
The partnering CSS is a nonprofit, charitable organization of 30 staff and 120 volunteers serving primarily rural townships in Ontario, Canada. Programs are offered to adult clients aged 18 and older and living in the community although the majority served are older adults.
Eligible participants included any staff member of the partnering CSS who had been involved in using the interRAI CVS. The collaborating executive director of the CSS used a standardized e-mail script to provide a brief overview of the study to CSS staff and invite them to participate in one focus group about their experiences using the interRAI CVS in their day-to-day work. Interested individuals were directed to contact the first author by e-mail, and a letter of information was then shared. Quantitative data collection (i.e., interRAI CVS) was completed by members of the collaborating organization for all current and new clients, and thus, no direct recruitment of CSS clients was made by the research team. A working group was convened including the executive director and the other members of the research team. This group met three times over the six-month project to review the study’s progress and findings.
4. Data Gathering
4.1. Qualitative Data Collection
Two virtual focus groups with partner staff were conducted online through Microsoft Teams by the first author in November 2020, both lasted approximately 1.5 hours. Sessions were audio-digitally recorded and professionally transcribed. Focus group questions explored staff experiences related to receiving training about the CVS, using the CVS with their clients, strengths and challenges that arose during the implementation process, and changes that could be made to improve the process in the future (refer to Table 1 for the focus group guide).
4.2. Quantitative Sample
The CVS was completed over the phone or in-person with all of the partner organization’s clients and directly entered into the Raisoft Ltd. web-based platform by trained partner staff. When staff were in the community without access to wifi, they documented the person’s responses in hardcopy and entered the data on the platform when back in the office.
The deidentified data for CSS clients screened between June and November of 2020 were exported from the Raisoft platform to a secure server hosted by the University of Waterloo. The CVS contains the following clinical outputs: COVID-19 Symptom Count, COVID-19 Emergency Care Flag, Major Comorbidity Count (MCC), presence of chronic conditions, Assessment Urgency Algorithm (AUA), Self-Reliance Index (SRI), count of impaired ADLs, access to medications and groceries, self-rated health, loneliness, caregiver status, financial concerns, and health service use over last 90 days (refer to Table 2 for a description of these outputs).
5. Data Analysis
5.1. Qualitative Analysis
A qualitative descriptive approach was taken for the qualitative strand [32, 33]. A qualitative description is an analysis approach that purposely positions the research conclusions as close as possible to the experiences reported by participants [32]. Directed content analysis was used to analyze the deidentified focus group transcripts [34]. This involved the systematic classification, coding, and interpretation of text data with the goal of identifying key themes or patterns. It is appropriate for use with an existing theory or framework, which can be used to structure the coding process and the interpretation. Codes for the analysis were generated from the characteristics of the four iPARIHS constructs [23], with an additional “other” code for transcript sections that did not fit any of the pre-existing codes. All interviews were coded independently by the first two authors. The first author reviewed the coding for areas of convergence and divergence before both researchers met to discuss. Consensus on final codes and themes was reached through discussion, and no new categories were identified.
5.2. Quantitative Analysis
Descriptive data analysis was conducted on the deidentified client assessment data drawn from the CVS. IBM SPSS Statistics for Windows, Version 23.0 was used for the descriptive statistical analysis. Percentages and frequencies were used to describe the categorical data (e.g., gender), and means were used to describe the central tendency of the continuous data (e.g., age) [35].
5.3. Mixed Methods Analysis and Interpretation
After completion of the qualitative and quantitative analysis components, the results from the two strands were compared with a joint display to generate the mixed methods interpretation [24]. The use of a joint display facilitates the direct comparison of results across the qualitative and quantitative strands to produce an overall interpretation that is broader than the individual components [24, 36]. The domains of the iPARIHS framework were used in the joint display to organize the results [23]. The results from each strand are evaluated for convergence, where the findings confirm the results of the other, or divergence, where the results provide different or complementary aspects of the implementation process [24].
To ensure trustworthiness, results from the mixed methods interpretation were shared with the organization’s executive director during the final working group meeting. The feedback provided was positive and indicated that the results of the interpretation aligned with her experiences and the feedback she had received from their staff. Additional details regarding the facilitation of the implementation were shared and incorporated into the interpretation.
5.4. Ethical Considerations
This study received ethics approval from the University of Waterloo Research Ethics Committees (42470). The use of the University of Waterloo server for secondary data analysis of interRAI instrument information has also obtained ethics approval (30173). Informed consent was obtained from all qualitative study participants prior to their participation in the focus group. Digital audio files and transcripts were stored on a password-protected server behind the university’s firewall. Transcripts were anonymized for identifying information.
6. Results
6.1. Qualitative Results
Two focus groups (FG) were conducted with 5 participants (P) in total. Three participants participated in the first focus group, and two participants in the second. The participants held different positions in the organization, including care provision and leadership roles. Their experiences and learnings implementing the CVS are summarized by the constructs of the iPARIHS framework [23].
6.2. Innovation: interRAI CVS
The CVS largely fits within their existing practices as a “really easy-to-use” instrument (FG2, P1). Participants shared that they found the CVS advantageous over an unstructured phone or in-person check-in and felt the standardized approach help them find clients in need of support. The CVS provided an opportunity for clients to be frank about issues they were facing during the pandemic. A participant shared this experience:
One situation that came to mind for me is in calling and checking in with a client who otherwise was quite reserved, and he was always put together and wasn’t super emotional. When I asked him the question about loneliness, he actually started crying and broke down. So that was a question that otherwise he wouldn’t really come out with unless you asked. So that opened the door for us to be able to chat with him a little bit more and link him to friendly visiting and other programs too. (FG1, P2)
They noted that they were able to identify clients with many different informational, social, and health service needs to which they were able to respond through system navigation and care coordination. Critical to the success of the implementation of the CVS was the creation of a process for postscreening of those most in need of support, as this participant explained:
The clients that flagged three and up [on the AUA], those people we were following up with them directly right away, making sure that those supports were put in place. We put them on that monthly cycle if nothing had changed and they seem to have all of their supports in place, then we can move to the three months (follow-up). (FG2, P2)
Participants did experience some challenges. For example, participants found the completion of the CVS over the phone difficult for some clients as they did not have facial expressions to draw on to determine whether the client understood the question. They felt the self-report instrument was not a good fit with their clients living with dementia and accessing the adult day program because the clients may not recall details such as their health history unless a caregiver was available to participate. Staff also noted that some of the CVS questions were culturally incongruent with clients’ beliefs, where, for example, it is not appropriate to be asked directly about personal hygiene or bathing.
6.3. Recipients: Staff, Clients, and Caregivers
The strong and collaborative practice of the CSS team facilitated embedding the CVS into their daily work during the pandemic when clients were experiencing many needs related to social-distancing directives. Participants were very knowledgeable of both health- and social-care supports and had strong existing networks across many sectors, as this participant explained:
A lot of these [CVSs] resulted in us navigating or connecting with the [home and community care], police and also our food banks, [municipal] cleaning services and [municipal] family resources. We end up working with a lot of folks who are struggling with dealing with mental health. And so the CMHA [Canadian Mental Health Association], we were able to connect with them on some stuff. (FG1, P1).
Thus, they were able to expertly connect their clients to both their own and others’ services. Another participant also shared, “It built bridges with a lot of people, 110%, because we had people say, “Well, what can you do for me?” As an organization, we do meals, we have driven. Generally, we had a solution for what they needed.” (FG1, P3). They did not share the CVS report outside of their organization, and only relayed what they felt were important details for that specific referral. The CSS organization did not have formal information-sharing processes in place for back-and-forth communication about shared clients with home- or primary-care providers, which this participant noted as a gap:
I wouldn’t say it was a whole lot, but I would call the family physician. In a lot of cases, we would connect with their family or their emergency contact. We can be one-sided in that way, in terms of what we can call and tell them our concerns, but they can't tell us the information without signing consents for the client saying that they can communicate with us. (FG1, P2).
Participants shared that their clients appreciated the contact, knowing that someone cared, and receiving the connections to other services. In some cases, clients were uncomfortable disclosing personal information over the phone, but this was mitigated by clearly stating up front who was calling and the purpose and, on some occasions, providing a number for the client to call them back so they could confirm it was their trusted provider on the phone. In situations where caregivers did not participate in the screening, participants found that some clients, due to cognitive impairment, were not aware of how overwhelmed their caregivers were by their health needs. The participants due to their ongoing communication with caregivers were aware of the impact of the pandemic on caregivers: “We had a lot of clients from day program who went to respite on a regular basis whose family members lost that respite. As for those caregivers, they were burnt out and the clients didn’t know.” (FG2, P1).
6.4. Local and External Health System Context
Participants embedded the CVS into their daily practices with both existing and newly referred clients using a clear process. They shared that their history of innovation and flexibility as an organization with a strong focus on independent living at home allowed them to provide services differently during the pandemic. This participant explained how they adapted as community health-care services were closed or providing virtual care:
The problem was we were basing our knowledge on how it was before COVID, on how this would work. Everybody was sort of jumping and trying to figure out, “Well, how are we going to get through this? What are we going to do?” So, we just didn’t have that information right in front of us, so we sort of had to figure out, “Well, who’s doing what and how is it going to happen?” (FG2, P2)
They also creatively responded to external policy changes that had impacted their clients’ receipt of home and community care services:
A lot of system navigation came out of it because we realized the PSWs weren’t showing up from the [home care agency] and the [care coordinator] didn’t know that because the client didn't want to call them or hurt them. They didn’t want the PSW mad. So, we got to be the bad guy. We’d realized that accessing food was an issue and then we’d link them with the food bank. (FG1, P1)
6.5. Active on Ongoing Facilitation
Participants shared that leaders in the organization employed a variety of strategies to successfully implement the CVS into practice, beginning with taking a team approach to screening: “When the rubber met the road on this project, we had almost all hands on deck. Everybody, really, we really pulled together as a team” (FG1, P1).
Other strategies included setting target goals and dates for CVS completion, auditing the achievement of goals, daily check-ins and support, and redistributing workload as needed. Participants also felt supported by the leadership team in keeping them apprised of pandemic-related service changes, “I will give a toot of the horn to our superiors. We were well-prepared that we knew who our community partners were out there. We’re pretty well linked out here and we know who we’re supposed to go to.” (FG2, P2). Facilitation also included taking time to celebrate their successes by providing lunch and sharing stories in team meetings of where they had identified and helped a person in real need.
6.6. Quantitative Results
A total of 594 individuals were screened between June 2020 and November 2020 (see Table 3). Clients were predominantly female (68.4%) with a mean age of 78.6 years (SD = 11.3). They were mostly living alone (48.3%) or with relatives (49.3%; e.g., spouse, children).
Most of the clients screened were not experiencing COVID-19 symptoms (88.7%) or requiring emergency care (96.5%) but due to pre-existing comorbid conditions, many (45.3%) were at elevated risk of mortality should they be infected with COVID-19. Additionally, 15% of clients had the highest need and urgency for a comprehensive, health-care assessment (i.e., a 5 or 6 on the AUA), and 23.4% had a moderate level of urgency (i.e., 3 or 4). Also, a third (34.2%) of individuals reported feeling lonely.
A smaller number of clients were experiencing financial difficulties (5.9%) and were unable to obtain both their medications and groceries (7.9%) due to the pandemic. Only 4.2% of clients identified their caregivers as feeling overwhelmed. In the past 90 days of screening, most clients had not sought care in the emergency department (89.4%), had an inpatient hospital stay (94.3%), or seen their primary-care provider (59.9%).
6.7. Mixed Methods Interpretation
The implementation recommendations were informed by an analysis of the convergent and divergent findings along with the results from the qualitative and quantitative components (refer to Table 4 for an overview of the implementation strategies for a surveillance process utilizing the interRAI CVS to identify older adults at risk and connect them to the required care during the pandemic).
Regarding the innovation, the qualitative and quantitative data converged as evidenced by the rapid collection of a large number of CVSs in routine practice at the organization and the tangible benefits they observed in locating clients in need during the pandemic. The interRAI CVS deployed through a web-based platform fit into CSS practice and provided an advantage over COVID-19 symptom-only screeners. The availability of wifi in rural communities was a challenge in collecting the CVS and generating the outputs in real time.
The clients’ social and medical complexity reported by staff was also evident in the CVS outputs. However, staff noted a higher level of caregiver burden than was captured by the CVS. Thus, when implementing a self-report instrument in settings, such as an adult day program, where the clients are living with dementia, it is important to build a mechanism to have a caregiver present to help with screening.
Organizationally, the implementation processes the CSS developed was based on previous strategies they found effective in other initiatives. These included practicing with the CVS to build confidence prior to implementation, starting a daily huddle to touch base and answer questions, developing a script to prepare clients for the purpose of the CVS, and continually communicating. The expertise of the CSS in information provision and referrals positioned them well for conducting screening in challenging conditions, such as the pandemic, and knowing what services to provide and who to connect with on behalf of their clients (e.g., mental health-care providers).
The implementation process of the CVS was actively facilitated by leaders at the organization. The importance of screening and added value to clients was clearly communicated to all staff, organizational champions were appointed, and processes were established to make the use of the CVS part of everyday practice (e.g., use the CVS at the time of meal delivery). This included plans for who was doing screening and when, clear postscreening processes based on AUA scale level, and ongoing follow-up plans for clients identified as high risk. Another successful facilitation strategy was to consistently highlight the benefits of screening, which included supporting good client care and successful care transitions.
7. Discussion
In this mixed methods study, we found that CSS organizations can successfully implement and use a standardized self-report interRAI instrument to provide integrated health and social care during a pandemic. Clients of the CSS were experiencing multiple vulnerabilities due to the pandemic that may not have been identified without active surveillance. In this study’s context, many key implementation considerations are realized, including a knowledgeable workforce and the use of best practice facilitation strategies, yet some barriers remain in facilitating communication between the health- and social-care sectors.
The findings from this study demonstrated a clear need for surveillance instruments to screen beyond COVID-19 symptoms given that few CSS clients had COVID-19 symptoms, but many had unmet needs that were exacerbated because of the pandemic measures. This is especially important for older adults and persons with disabilities who have unmet health- and social-care needs that are not known or addressed until a crisis point [19, 37]. The CVS has high potential for use to actively surveil vulnerable populations not only in future waves of the COVID-19 pandemic, but beyond, such as to identify isolated persons in the aftermath of environmental disasters relating to climate change [38, 39].
CSS providers and leaders are able to expertly navigate a wide array of health and support services including meal service, transportation service, adult day programs, and volunteer visiting [40]. They are in an optimal position to coordinate care for clients and intervene when necessary, as well as understand and support the needs of caregivers. However, CSS faced challenges during the pandemic, notably a marked reduction in their volunteer-staff base [17]. Funding for both home care and CSS has not kept up with demand and growing client complexity, particularly in light of the pandemic’s exacerbation of food insecurity and family caregiver burden [12, 13, 17, 41]. Previous qualitative research has also found that primary-care providers do not have formal information-sharing processes in place for continuous communication about shared clients with CSS providers and tend to have stronger linkages with health services [42–44].
Importantly, there is a need for an investment in infrastructure to support CSS with digital technology capacity and access [17]. Standardized comprehensive instruments, such as the interRAI CVS, have great potential to facilitate integrated care by providing a common language for older adults, health- and social-care providers and a mechanism to monitor quality at both the program and system level [45]. A key feature of an integrated health- and social-care system is a shared data platform to enable information sharing and communication [3, 45].
7.1. Study Strengths and Limitations
Several strengths are present in this work. First, by partnering with an existing CSS in a strongly embedded way, this research was able to identify real-world challenges and facilitators. The participation of the executive director of the CSS in the working group helped ensure the reported findings reported are accurate and reflective of the context in which they were gathered.
Second, the use of a mixed methods approach, aided by the iPARIHS theoretical framework, facilitated a more holistic understanding of the implementation process.
There are a few limitations to this study. We did not collect quantitative data on service disposition postscreening so we cannot describe the number and/or types of connections made by the CSS providers. The study was done at one CSS organization, which may not be reflective of the sector norm. In particular, this CSS provides multiple services and had pre-existing knowledge of health and social-care supports and referral relationships, and thus, other single-service agencies (i.e., transportation) may require additional learning regarding services to connect to based on screened clients’ needs.
8. Conclusion
The standardized, self-report screening instrument, the interRAI CVS, was successfully implemented by a CSS organization as a surveillance tool to identify and triage vulnerable clients to health- and social-care services. The CVS fits well into an existing practice of CSS as an easy-to-use instrument that goes beyond symptom assessment to help identify at-risk persons during the pandemic and potentially other natural disasters. Strengthening relationships and communication between CSS and the health-care system can be facilitated by the use of common language afforded by interRAI instruments. However, realizing this integration will require investment in the CSS sector to support their contributions to supporting older adults to age well at home.
Data Availability
The de-aggregated quantitative data and qualitative transcripts cannot be shared in order to protect patients’ and participants’ identities.
Additional Points
What is known about this topic. (i) Older adults with multiple chronic conditions have complex care needs and receive community support services to maintain wellness and safety at home. (ii) The pandemic created a need for proactive surveillance to identify older adults at risk due to health- and social-care service disruptions. What this Paper Adds. (iii) The interRAI COVID-19 vulnerability screener (CVS) instrument is a feasible and helpful tool to identify at-risk persons and can be effectively deployed by community support services during the pandemic and future disasters. (iv) Strengthening relationships and communication between community support services and the health-care system is facilitated using common language afforded by interRAI instruments and supports integrated care.
Conflicts of Interest
The authors declare that they have no conflicts of interest.
Acknowledgments
The authors are most grateful for the funding provided by the Mitacs Accelerate Program to conduct this research. The authors also thank, Raisoft Ltd, a Finland-based software vendor that provided the CVS software at no cost during the pandemic. Thank you to Ms. Aimun Shah for her thoughtful review of a near-final draft of this manuscript. Thank you to the staff of the partner CSS organization for your support and participation in this project. This study was funded by MITACS Accelerate. Salary support for Dr. Heckman was provided by Schlegel Research Chair in Geriatric Medicine and for Dr. Northwood by Canadian Institutes for Health Research and the SE Research Centre during a postdoctoral fellowship.