Paramedics assessing patients with complex comorbidities in community settings: results from the CARPE study

Objectives The aim for this study was to provide information about how community paramedicine home visit programs best “navigate” their role delivering preventative care to frequent 9-1-1 users by describing demographic and clinical characteristics of their patients and comparing them to existing community care populations. Methods Our study used secondary data from standardized assessment instruments used in the delivery of home care, community support services, and community paramedicine home visit programs in Ontario. Identical assessment items from each instrument enabled comparisons of demographic, clinical, and social characteristics of community-dwelling older adults using descriptive statistics and z-tests. Results Data were analyzed for 29,938 home care clients, 13,782 community support services clients, and 136 community paramedicine patients. Differences were observed in proportions of individuals living alone between community paramedicine patients versus home care clients and community support clients (47.8%, 33.8%, and 59.9% respectively). We found higher proportions of community paramedicine patients with multiple chronic disease (87%, compared to 63% and 42%) and mental health-related conditions (43.4%, compared to 26.2% and 18.8% for depression, as an example). Conclusion When using existing community care populations as a reference group, it appears that patients seen in community paramedicine home visit programs are a distinct sub-group of the community-dwelling older adult population with more complex comorbidities, possibly exacerbated by mental illness and social isolation from living alone. Community paramedicine programs may serve as a sentinel support opportunity for patients whose health conditions are not being addressed through timely access to other existing care providers. Protocol registration ISRCTN 58273216. Supplementary Information The online version contains supplementary material available at 10.1007/s43678-021-00153-4.


What is known about the topic?
Community paramedicine programs are designed to improve access to care for vulnerable patient groups but case finding remains a challenge.

What did this study ask?
We investigated who community paramedicine patients are and how they compare to clients receiving community care using identical assessment items.

What did this study find?
Community paramedicine patients have higher proportions of multiple chronic disease and mental healthrelated conditions than others who receive community care.

Introduction
Community paramedicine programs address barriers to care faced by community-dwelling older adults (≥ 65 years of age) or other vulnerable patient populations who may otherwise resort to calling an ambulance or visiting an emergency department (ED) [1][2][3][4]. Community paramedicine home visit programs have improved access to care for frequent callers through collaboration between primary care providers and community home care and support services agencies [4][5][6][7][8][9] resulting in patients avoiding ED visits upwards of 78% of the time and higher admission rates when visits are unavoidable [10]. Frequent callers use paramedic services for reasons beyond acute medical emergencies including to address personal or social care needs (such as loneliness, food insecurity, or other deficits in quality of life), chronic conditions (such as pain, disease, or ongoing management of mental health), or functional and mobility difficulties related to advanced age [11][12][13][14][15][16]. Across Canada, expansion of community paramedicine from pilot projects to province-wide programs [17][18][19][20] has been supported by a growing evidence base [5-7, 9, 21-24]. Community paramedicine programs are attempting to shift from "reactive responses" towards better management of chronic conditions with fewer exacerbations [1] by targeting frequent callers who represent up to 20% of ED visits [12].
Whether community paramedicine home visit programs represent a duplication of community-based services requires further exploration [14,25,26]. Studies have found that home care nursing visits are associated with same-day ED visits [27], that home care clients use paramedic services for transportation to the ED for such visits [16], and that paramedic referrals are associated with increased utilization of home care services [28]. If home care clients present with lower acuity levels at an ED visit and are not admitted to hospital [27], more information is needed to determine how community paramedics could better "navigate" their role in the delivery of integrated care [25,26]. While community paramedicine home visit programs often incorporate collaborations with other outof-hospital community care programs [5], an expanded description about the demographic and clinical characteristics of community paramedicine patients that includes comparisons with existing community care populations would demonstrate the role community paramedicine programs play in case-finding individuals for the delivery of out-of-hospital community care and prevention of ED visits. Implementation of the PERIL rule [29] to inform paramedic referrals to home care services demonstrated how paramedic screening at the time of a 9-1-1 call increased appropriate provision of home care services [28]. Assessments by community paramedics capturing a wide breadth of clinical observations could demonstrate similar utility in community paramedicine home visit programs, even where patient enrollment is determined by local program design [30], and guide further coordination between primary care providers, home care, and community support services [31].
Our study proposed to identify characteristics of existing community paramedicine home visit patients across multiple jurisdictions and compare them to clients from other community-based care providers. We hypothesized that patients in community paramedicine home visit programs represent a distinct subset of community-dwelling older adults with complex needs and a limited social support structure that contributes to their enrolment in these programs.

Overview
Our study used routinely collected de-identified secondary data about individuals assessed for their eligibility of home care services or as part of the delivery of community support services or community paramedicine programs across Ontario. We used identical variables from each data set to compare the home care and community support services client populations to those enrolled in community paramedicine home visit programs. This study was approved by Hamilton Integrated Research Ethics Board (#1650D).

Home care clients
Information about home care clients included all individuals assessed using the interRAI Home Care (HC) assessment [32], between April 1, 2018 and March 31, 2019 in one health region in Ontario, Canada. The interRAI-HC is a mandatory standardized assessment instrument for individuals that are expected to receive home care services for 60-days or more [33]. The Canadian Institute for Health Information's Home Care Reporting System is a repository of interRAI-HC data used for epidemiologic research and reporting on quality measures [33].

Community support services clients
Information about clients receiving community support services was obtained from individuals who had been assessed using the interRAI Community Health Assessment (CHA) assessment instrument [34], between January 1, 2017 and December 31, 2017 (the most recent year for which data were available) in multiple jurisdictions across Ontario. The interRAI-CHA includes the same assessment domains as the interRAI-HC but uses a modular design [34]. For example, assessors could be alerted to the need for a more detailed mental health assessment for some clients thereby completing these assessment items only on those where it was indicated and not others. The interRAI-CHA is used by community support services to assess individuals who receive services like homemaking, friendly-visiting, or adult day programs [33]. Agencies establish their own parameters for use of the interRAI-CHA and share data with the inter-RAI Canada Repository [35]. Both the interRAI-HC and interRAI-CHA include decision support scales and screeners that have undergone extensive testing with demonstrated validity and reliability [33,36].

Community paramedicine home visit program patients
Data about individuals enrolled in community paramedicine home visit programs were obtained from paramedic services that implemented a standardized assessment instrument as part of the Common Assessments for Repeated Paramedic Encounters (CARPE) study (ISRCTN 58273216). Several paramedic services participated in development of the CARPE assessment instrument through a process including literature review [5], expert panel consultation [30], and an environmental scan of community paramedicine assessment practices [37].
Six paramedic services implemented the CARPE assessment instrument voluntarily as part of a quality improvement process within existing community paramedicine home visit programs. All paramedic services had similar patient enrollment criteria: diagnoses of Congestive Heart Failure (CHF), Chronic Obstructive Pulmonary Disease (COPD), or diabetes, and health system utilization that included at least three 9-1-1 calls, two ED visits, or one hospital admission in the preceding year [17,38]. Community paramedics participated in a 4-hour training session about assessment practices (delivered by a member of the research team in collaboration with local paramedic service educators), received an assessment instrument training manual, and could attend teleconference sessions to clarify any remaining questions. The CARPE assessment instrument was embedded in electronic medical record software and data were provided to the research team for the period from April 1, 2018 to March 31, 2019. The CARPE assessment instrument was used to assess any patient enrolled in a community paramedicine home visit program (herein called community paramedicine patients), either during their enrolment or as part of regular reassessments.

Analysis
Data were analyzed using descriptive statistics (including calculation of standard error) for each identical item from the respective assessment instruments (see supplemental file Table S1 for variable list). For reporting purposes, items were grouped by domain and collapsed into dichotomous variables to identify presence of disease, health deficits, or indicators of impairment (according to the nature of the respective assessment item). Comparative analysis for each assessment item tested proportions of responses using z-test (with ∝ = 0.05) to investigate differences between the community paramedicine patients and the other cohorts of community-dwelling older adults according to identical fields from the respective assessment instruments. Analysis was completed using SAS 9.4 (SAS Institute Inc, Cary, NC) and excluded incomplete or partial assessments. Table 1 provides the demographic characteristics, living conditions, and health system utilization data for each group; 29,938 individuals assessed with the interRAI HC, 13,782 individuals assessed with interRAI CHA, and 136 individuals assessed with the CARPE assessment instrument. Mean ages were 78.8 (SD ± 13.5), 78.2 (SD ± 13.7), and 75.7 (SD ± 14.2) for home care clients, community support services clients, and community paramedicine patients respectively. The proportions of female patients-60.3%, 68.2%, and 64.0%-suggested scant evidence of differences in gender representation between groups. Differences in proportions of individuals living alone was evident, with more community support clients (59.9%) and fewer home care clients (33.8%) when compared to community paramedicine patients (47.8%). The proportion of patients admitted to hospital in the past 90 days was not significantly different when comparing community paramedicine patients to home care clients-47.1% and 41.9% respectively-higher than the proportion observed in community support services clients, 13.6%. clients and 90.5% in community support services clients) and a higher proportion who were not adherent with their prescription medications (40.4% compared to 13.6% in home care clients). Statistically significant differences between the community paramedicine patients and the other groups were observed in all of these comparisons.

Mental health related conditions and psycho-social well-being
Higher proportions of anxiety disorder and depression were found in community paramedicine patients (44.1 and 43.4% respectively compared to 18.3%, and 26.2% in home care clients and 15.8% and 18.8% in community services support clients) and they displayed higher proportions of associated symptoms (See Fig. 2, included in Appendix)-all of which were statistically significant differences. A higher proportion of community paramedicine patients would meet the criteria for further assessment of their mental health, 29.4% when compared to 12.4% in community support services clients. More community paramedicine patients indicated that they had experienced a major life stressor or a decline in social activities in the past 90 days (52.2% compared to 22.5% in home care clients and 25.7% in community support services clients).

Communication and functional abilities
When comparing community paramedicine patients to community support services clients, little evidence of difference was observed in the proportions of individuals who had difficulty communicating or functional deficits for some Activities of Daily Living and Instrumental Activities of Daily Living (specifically personal hygiene, dressing lower body, transportation, and phone use) (see Fig. 3, included in Appendix). But the home care client group displayed statistically significant differences, with higher proportions of dependence for all functional items when compared to community paramedicine patients.

Discussion
We found that patients in community paramedicine home visit programs likely represent a distinct sub-group of the community-dwelling older adult population because of numerous differences observed between our cohort groups. Our data suggests that the proportion of individuals with mental health needs, complex co-morbidities, and ongoing health conditions or symptoms are often different between community paramedicine programs and home care and community support services agency populations. Higher proportions of health needs in community paramedicine patients suggests they are a complex patient group who could benefit from more integrated care that includes an interface between multiple care providers-reinforcing a characteristic of many community paramedicine home visit programs [5,17]. By illustrating differences between community paramedicine home visit patients and other community-dwelling older adults, efforts can support case-finding by all care providers to improve patient care access and reduce unnecessary utilization of 9-1-1 or EDs by these individuals.

Strengths and weaknesses of the study
Duplication of services with other existing communitybased health care services is a critique of community paramedicine programs [25] but, to our knowledge, our study is the first to compare the characteristics of the older adults that are receiving these services. While community paramedicine programs looking to serve frail older adults may have targeted enrollment efforts favouring these individuals, the differences we found suggests duplication of services is unlikely because complex comorbidities, likely exacerbated by mental illness, appeared less common amongst individuals receiving community support services or home care programs. Even if other programs or services are providing care to such patients, community paramedicine programs are likely providing a necessary additional level of support to the existing supports patients may be receiving from community-based care providers. For example, remote patient monitoring programs (broadly implemented in Ontario) support chronic disease management by identifying patients' worsening health symptoms prior to exacerbations that require a 9-1-1 response while complementing existing care from other community-based care providers [17,38]. The inferences drawn from our comparisons should be made cautiously because they are based on a small convenience sample of community paramedicine patients assessed using a prototype assessment instrument. Community paramedicine programs remain relatively small in comparison to the number of clients seen through other established programs delivering community-based care. While the sizes of the sample cohorts present a limitation to the inferences, they are reflective of the differences in sizes of the patient populations and were large enough to power statistical analyses. To further strengthen our analysis, we excluded small counts (< 10) of observations from the community paramedicine cohort.

Implications for clinicians and health policy
Opportunity exists for further collaboration between community-based support services agencies and home care providers, community paramedicine home visit programs, and other parts of the healthcare continuum-particularly primary care providers-to improve coordination of care to medically complex community-dwelling older adults [3,10,39]. For example, a risk scale used to determine the likelihood of an ED visit in home care clients is a likely predictor for use of paramedic services for transportation to the ED [16]. Shared case-finding to identify at-risk patients could support greater coordination between hospitals, home care providers, community support services agencies and community paramedicine programs and lead to improved patient safety and reduce unnecessary ED visits and 9-1-1 utilization.

Implications for future research
Anonymized data were obtained for our study meaning that analysis of cross-membership between cohorts was not possible. It is possible that a handful of patients could have been represented in all three groups and questions remain about the likelihood of this. High levels of cross-membership between groups would have lessened the likelihood of observing differences in our analysis. Future research should aim to provide a complete analysis of health system utilization amongst community dwelling older adults.
High proportions of mental health-related conditions were identified in community paramedicine patients. Other research has demonstrated that mental health and social isolation can contribute to repeated 9-1-1 use [11,14,15,40]. While we provided a comparison to other cohorts of community-dwelling older adults, further comparisons are needed with additional community and geriatric mental health populations. Community paramedicine programs should explore further integration with local community support services agencies and home care providers as part of their program design and community paramedics may benefit from greater education about addressing mental health needs, particularly amongst older adults.

Conclusion
Our analysis showed that community-dwelling older adults in community paramedicine home visit programs may represent a distinct patient group with a greater proportion of mood symptoms, ongoing health conditions, and complex comorbidities than comparable patient populations that receive home care or community support services. Enrolment into a community paramedicine home visit program may be indicative of a combination of inadequate social support structures or clinical instability and decline of a patient's condition. Community paramedicine home visit programs may provide a sentinel support opportunity for community-dwelling older patients whose health conditions are not otherwise being addressed through timely access to other existing care providers.