Article Text
PDF
Abstract
Objectives Physicians and nurse practitioners (NPs) play critical roles in supporting palliative and end-of-life care in the community. We examined healthcare outcomes among patients who received home visits from physicians and NPs in the 90 days before death.
Methods We conducted a retrospective cohort study using linked data of adult home care users in Ontario, Canada, who died between 1 January 2018 and 31 December 2019. Healthcare outcomes included medications for pain and symptom management, emergency department (ED) visits, hospitalisations and a community-based death. We compared the characteristics of and outcomes in decedents who received a home visit from an NP, physician and both to those who did not receive a home visit.
Results Half (56.9%) of adult decedents in Ontario did not receive a home visit from a provider in the last 90 days of life; 34.5% received at least one visit from a physician, 3.8% from an NP and 4.9% from both. Compared with those without any visits, having at least one home visit reduced the odds of hospitalisation and ED visits, and increased the odds of receiving medications for pain and symptom management and achieving a community-based death. Observed effects were larger in patients who received at least one visit from both.
Conclusions Beyond home care, receiving home visits from primary care providers near the end of life may be associated with better outcomes that are aligned with patients’ preferences—emphasising the importance of NPs and physicians’ role in supporting people near the end of life.
- Home Care
- Hospital care
- Pain
- End of life care
Data availability statement
The dataset from this study is held securely in coded form at ICES. While data sharing agreements prohibit ICES from making the dataset publicly available, access may be granted to those who meet pre-specified criteria for confidential access, available at www.ices.on.ca/DAS. The full dataset creation plan and underlying analytic code are available from the authors upon request, understanding that the computer programs may rely upon coding templates or macros that are unique to ICES and are therefore either inaccessible or may require modification.
This is an open access article distributed in accordance with the Creative Commons Attribution Non Commercial (CC BY-NC 4.0) license, which permits others to distribute, remix, adapt, build upon this work non-commercially, and license their derivative works on different terms, provided the original work is properly cited, appropriate credit is given, any changes made indicated, and the use is non-commercial. See: http://creativecommons.org/licenses/by-nc/4.0/.
Statistics from Altmetric.com
WHAT IS ALREADY KNOWN ON THIS TOPIC
Home-based end-of-life care is associated with improved end-of-life outcomes.
Limited existing evidence on the effectiveness of home-based nurse practitioner-led primary care.
WHAT THIS STUDY ADDS
Home visits from a nurse practitioner, physician or both are associated with improved end-of-life outcomes.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Results highlight the importance of primary care providers in supporting reduced acute care use at the end of life and in enabling community-based deaths.
Family physicians and nurse practitioners provide critical support for individuals residing in the community at the end of life, and more research is needed to examine the effectiveness of interprofessional and collaborative practice models involving a mix of physicians and nurse practitioners.
Introduction
Challenges with access to adequate and timely primary care have become a crisis across healthcare systems worldwide.1 The growing demand for physicians to provide care in the community to an increasingly complex patient population2 has led health system planners to consider increasing the number of interprofessional teams within the primary care setting to meet current demands.3 For example, this expansion includes integrating shared care models with nurse practitioners (NPs), who are registered nurses with additional qualifications and experience that enable them to practice as autonomous licensed primary healthcare providers.4 In Canada, NPs first began practising in the 1960s and their scope of practice includes health promotion, disease prevention, curative care, rehabilitative care, prescribing controlled substances and supportive care for people with life-limiting illness or near the end of life.5
Internationally, governments have prioritised the need to enhance palliative care delivery, especially in home and community care.6 7 Existing evidence on the provision of end-of-life home visits has predominantly focused on the role of physicians and the receipt of home care services (including services provided by registered nurses, personal support workers and allied health professionals).8 Physician home visits and palliative home care have been demonstrated to lead to better patient–provider communication about care planning, improved health-related quality of life, reduced hospitalisations and increased likelihood of dying at home.9 Despite the potential to expand primary care’s capacity and meet the needs of those near the end of life through interprofessional models involving NPs, few studies have examined the relationship between NP-led home visits and patient outcomes, especially using large population-level data sources.10
In this study, we described the provision of home visits by primary care providers at the end of life in a population-based cohort of deceased home care users. We compared across characteristics and healthcare outcomes—including hospitalisations, emergency department (ED) visits, symptom management and location of death—in those who received visits from physicians and NPs to those who did not receive any home visits from a provider near the end of life.
Methods
Setting and design
We conducted a retrospective cohort study examining the outcomes of decedent home care clients in Ontario, Canada, who did and did not receive home visits from an NP, physician or both in their last 90 days of life.
Context
Under the Canada Health Act, all medically necessary hospital, diagnostic and physician services in Canada are funded through an interlocking federal, provincial and territorial system. The extent of coverage for non-medical services (eg, home care services and long-term care) vary across provinces and territories. Services not covered by most federal and provincial healthcare plans include vision and dental care, outpatient prescription drugs for populations under the age of 65, and most outpatient rehabilitative and restorative care. Those deemed ineligible for the provincial healthcare plan (eg, the Ontario Health Insurance Plan (OHIP)) can appeal provincially, apply for federal health coverage, pay for healthcare out-of-pocket or purchase private health insurance. Eligibility for provincial and territorial healthcare insurance programmes are unique to each region and based on length of time spent in residence, as well as citizenship or immigration and refugee designation.
In Ontario, people who require support to remain at home may be referred for home care, which includes publicly funded rehabilitative care, nursing care (including NP home visits), personal care and general support with instrumental activities of daily living (ADL). Referrals can be made by healthcare professionals, family members or patients themselves. Services are provided based on patients’ needs, which are routinely assessed using the interRAI Home Care (interRAI HC) Assessment Instrument,11 and funded by the Ontario Ministry of Health. Across Canada, InterRAI instruments are licensed and routinely used for home care, long-term care and in some complex continuing care settings for the assessment of care needs. Home visits provided by physicians are covered by OHIP.
Data sources
We used population-based, person-level health administrative data housed at ICES (formerly known as the Institute for Clinical Evaluative Sciences) to derive the study population and outcomes of interest (see detailed database descriptions in online supplemental appendix I). Data held at ICES contain administrative information (eg, date of healthcare encounter, services provided) associated with healthcare encounters at the individual level and services covered under the provincial health insurance plan.
Supplemental material
Briefly, we identified decedents and derived their baseline demographic information and vital statistics from the Registered Person’s Database. Home visits were identified using the Home Care Database (HCD) and OHIP claims for NPs and physicians, respectively. We used the Canadian Institute for Health Information’s Discharge Abstract Database, the Continuing Care Reporting System database, and the National Ambulatory Care Reporting System (NACRS) to identify active chronic conditions, individuals’ acute care utilisation, and death within an institution (see online supplemental appendix I for variables and corresponding database). Additional clinical characteristics and needs of a subset of patients were derived from the interRAI HC database for those who had an assessment within 3 months prior to death. The Ontario Drug Benefit claims database (ODB) was used to ascertain claims for medications related to pain and symptom management. These datasets were linked using unique encoded identifiers and analysed at ICES.
Study population
We identified all decedents who had a home care record in the last 90 days of life and died between 1 January 2018 and 31 December, 2019. We used home care clients to ensure the entire cohort was eligible for NP home visits, which were identified through the HCD. We excluded individuals aged <19 years or >115 years, those not eligible for OHIP at any point in the last year of life, and those living in a long-term care home in the last 90 days of life, as institutionalised individuals and those without OHIP coverage would be ineligible for home-based services.
Exposure
We compared decedents who did not receive any home visits to those who had received at least one home visit delivered by only an NP, only a physician (included all practising physician specialties), or both an NP and physician during the last 90 days of life. Physician-delivered home visits were captured using OHIP billing codes (online supplemental appendix II). NP-delivered palliative home visits were captured through a HCD record, as visits delivered by NPs outside of the provincial home care programme are not captured in health administrative data.
Outcomes
Our outcomes were at least one ED visit, at least one hospital admission, at least one medication claim related to pain and symptom management occurring during the last 3 weeks of life and death in the community. We captured symptom management using claims made to the ODB programme, which covers most of the drug costs for those receiving end-of-life home care services and those aged 65 years or above. Symptom management claims were summarised according to pharmacological treatments for three common end-of-life symptoms: pain management, delirium or agitation, terminal secretions or nausea (full list in online supplemental appendix III). We defined community death as deaths that were not captured within institutional records, which could include home or a hospice setting (see online supplemental appendix I for databases used).
Study Variables
We characterised patients according to their age, sex, geography, neighbourhood income and the local health integration network (LHIN) (ie, regional health authority) within which they resided. We only captured income quintiles for urban areas since using rural postal codes representing much larger geographical areas can result in considerable income misclassification.12 We captured the prevalence and count of 17 health conditions, which were determined using all healthcare utilisation records and diagnosis codes within 2 years prior to the index date, based on previously developed and validated methods.13–26 We collapsed all cardiovascular-related diseases (ie, acute myocardial infarction, congestive heart failure, coronary heart disease and hypertension) and respiratory diseases (ie, asthma and chronic obstructive pulmonary disease), as these are associated with similar care trajectories. We captured patients’ care characteristics during their last 90 days of life, including the types of formal home care services received (eg, personal support or nursing), average hours of home care services received, and level of any inpatient palliative care involvement during a hospital stay.27 We captured the primary care model to which patients belonged, namely whether patients were rostered through a blended capitation model, blended fee-for-service model or received services through a traditional fee-for-service model. Blended capitation and blended fee-for-service models are remuneration models used by over half of Ontario-based family physicians, in which physicians receive payments for providing a basket of services to enrolled patients versus a non-blended traditional fee-for-service model where remuneration is from direct billing of services and is used by all other physician specialties. We stratified home care hours based on those deemed to be in their last few months of life vs not, since this end-of-life designation is associated with higher service provision, including equipment and home-based hospice supports.28
Statistical analysis
We summarised demographic and clinical characteristics using descriptive statistics, stratified by receipt of home visits and provider type, using χ2 and analysis of variance (ANOVA) tests. Logistic regression models examined the independent relationship between the receipt of one or more home visit and each outcome, defined dichotomously. We modelled acute care use and symptom management occurring during the last 3 weeks of life, controlling for age, sex, rurality, local health region, chronic conditions, formal home care hours occurring 90–22 days before death, previous hospitalisations and previous inpatient palliative care encounters occurring prior to the last 90 days of life, primary care model and number of days spent in the community during the last 90 days of life and before the outcome occurred. Restricting formal home care hours to a timeframe of 22–90 days before death, ensured they occurred prior to outcomes. We modelled death in the community, controlling for all the same covariates as above occurring anytime during the last 90 days of life.
This reporting of methods and results in this paper adheres to the Reporting of Studies Conducted using Observational Routinely-collected Data (RECORD) statement and reporting guideline.29
Sensitivity analyses
We conducted sensitivity analyses using a subset of those with an interRAI HC assessment completed 90–22 days before death to address possible confounding by differences in clinical severity and care needs which can influence outcomes and to reduce selection bias of those with better support in the community having a higher likelihood of community death. We estimated logistic regression models with all previously included covariates and additional clinical and social variables in a nested analysis: first, we included clinical variables pertaining to patients’ function (using the ADL Self-Performance Scale), overall health stability (using the Changes in Health, End-Stage Disease and Signs and Symptoms Scale) and cognition (using the Cognitive Performance Scale); then, we added covariates pertaining to the availability of social support, including the presence of a live-in caregiver and caregiver distress. This nested analysis allowed us to compare the relative contribution of additional clinical as well as social characteristics to the model fit and their association with the outcomes of interest.
Results
Descriptive results
The cohort included 103 664 Ontario eligible decedents (a cohort creation flow diagram can be found in online supplemental appendix IV). The mean age was 77.8 (SD: 13.1) years, with over 80% of individuals aged 65 years or older (table 1). During the last 90 days of life, one-third (34.5%) of the cohort received at least one home visit from a physician, 3.8% received at least one home visit from an NP and 4.9% received at least one home visit from both an NP and a physician. Significant differences in characteristics of decedents visited by an NP only compared with those visited by a physician only included being younger (mean age: 74.4 SD±13.2 vs 78.6 SD±13.2), male (53.1% vs 48.3%) and living in rural areas (19.5% vs 10.9%). Furthermore, greater proportions of those visited by an NP only lived in rural and northern geographical health regions, which also had the lowest proportion of home visits delivered (figure 1—LHINs 13 and 14) compared with more central locations. A high proportion of decedents rostered within capitation-based models, including family health teams, received NP home visits (table 2).
Percentage of home care patients who received a home visit from a primary care provider (NP and/or physician) during their last 3 months of life across local health integration networks (LHINs) who died between 1 January 2018 and 31 December 2019. NP, nurse practitioner.
Patient characteristics of all decedents who died between 1 January 2018 and 31 December 2019 who received formal home care services in their last 90 days of life (captured 90 days before death)
Care characteristics during the last 90 days of life of all decedents who died between 1 January 2018 and 31 December 2019 and who received formal home care services in their last 90 days of life
Home visits associated with end-of-life outcomes
Compared with not receiving a home visit, receiving at least one home visit from an NP, a physician or both an NP and a physician was associated with reduced odds of being hospitalised at least once in the last 3 weeks of life (OR 0.54, 95% CI 0.50 to 0.59; OR 0.47, 95% CI 0.46 to 0.49; OR 0.31, 95% CI 0.28 to 0.34), respectively (figure 2). Similarly, receiving at least one home visit from an NP, a physician or both an NP and physician was associated with reduced odds of visiting the ED at least once in the last 3 weeks of life (OR 0.75, 95% CI 0.67 to 0.83; OR 0.72, 95% CI 0.68 to 0.75; OR 0.50, 95% CI 0.43 to 0.58), respectively. Receiving a home visit from an NP, a physician or both was also associated with increased odds of receiving medication for symptom management in the last 3 weeks of life (OR 2.54, 95% CI 2.33 to 2.76; OR 2.39 95% CI 2.30 to 2.48; OR 3.48, 95% CI 3.14 to 3.84), respectively. Finally, having at least one home visit from an NP, a physician or both was associated with increased odds of dying in the community (OR 2.98, 95 %CI 2.75 to 3.24; OR 4.58, 95% CI 4.42 to 4.74; OR 7.20, 95% CI 6.57 to 7.89), respectively.
Crude and adjusted results of regression analyses of association between hospitalisation, emergency department visits, symptom management claims and a community death and receipt of a home visit from only a nurse practitioner (NP), physician or both compared with no home visit.
Sensitivity analyses
Sensitivity analyses were conducted on a subset of 16 307 individuals (15.7% of cohort) who had completed an interRAI HC assessment in the 90–22 days before death. This subgroup had similar baseline characteristics to the overall cohort (online supplemental appendix V). Aside from the odds of ED visits becoming insignificant, associations observed in the main results were not significantly changed by adding clinical or social support characteristics to the models, including for caregiver distress (online supplemental appendix V).
Interpretation
In this analysis, we observed home visits provided near the end of life were associated with reduced the odds of acute care use, increased the odds of medication prescription for symptom management, and increased the odds of dying in the community. Furthermore, the magnitude of the association between each outcome and home visits was largest when NPs and physicians both provided at least one home visit to the patient as they approached the end of life.
There is a growing body of literature on the role and benefits of integrating NPs into home and community primary care.10 30 31 Our results support and contribute to this evidence base by examining the receipt of palliative home visits from NPs and describing its correlation with healthcare use and outcomes at a population level. Specifically, our findings are aligned with existing data indicating that end-of-life home visits from primary care providers are uncommon, yet associated with improved outcomes.32 With a rapid need to respond to the current crisis in primary care provision, considerations for funding and establishing more robust evidence on NPs and integrated interdisciplinary teams are paramount, especially for vulnerable patients near the end of life.33 Capitation-based remuneration models are offered to physician groups and often are affiliated with interprofessional teams in Ontario.34 The results that greater proportions of decedents belonging to capitation-based models had a home visit from both an NP and physician may suggest these models support community-based NPs and physicians working alongside one another.
Geographical differences in visit provision observed in this study is aligned with international literature which has highlighted several challenges to providing consistent services across regions.35 Variations observed may be partially due to differences in regionally funded programmes across Ontario, driven by the priorities of local policy and decision-makers or measures to augment service provision in traditionally underserved areas (eg, rural and remote areas). Previous research has highlighted the importance of NPs practising in rural and underserviced areas for increasing the reach of primary care and alleviating some of the existing pressure and growing demand for community care.36 The number of practising NPs in Canada has increased to more than 7000 over the past few decades, with the largest workforce of NPs in Ontario, and the creation of additional positions is expected in the near future.37 Our finding that a higher proportion of rural decedents were receiving a home visit from NPs may be related to the growth in this profession and suggest that the NP workforce might be able to help address regional disparities—n increasingly important aspect to highlight, since providing home visits increases patients’ odds of receiving care and dying at home, a desired place of care during the end-of-life period for many patients.38
Despite growing evidence that interprofessional primary care is valuable and effective,3 notable barriers exist to successful NP integration, including physicians’ lack of knowledge of NPs’ scope of practice and role.39 Existing care models can influence care provision and practice patterns. For example, future research and policy could focus on identifying and describing interprofessional primary care teams and, where appropriate, supporting the integration of NPs into primary care. This could be done through identifying which models are best suited to collaboration or shared care and providing guidance on role definition within an integrated team.
Limitations
This study is the first to describe, at a population level, the provision of home visits from NPs to those at the end of life—by leveraging the rich health administrative data in Ontario, the most populous province in Canada. However, as with all administrative data-driven research, there are limitations. First, this study identified NP visits using data from the publicly funded home care programme. While this captures many NPs, most practising in Ontario are salaried without recording individual service codes, with data from 2020 showing 34.8% worked in hospitals, 27.0% in community health, 2.9% in long-term care and 31.7% in other places of work.37 Although this is a limitation of our study, it suggests that we have likely underestimated the involvement of NPs in end-of-life care, and those working in other primary care settings may have contributed to the positive care outcomes observed across all and any exposure groups. Second, we found a higher proportion of decedents received a physician home visit (34.5%) in our home care cohort than previous research on all Ontario decedents nearing the end of life (20.4%).32 This reduces generalisability of these results to formal home care recipients, a characteristic which may increase individuals’ likelihood of receiving home visits, due to referrals or connections through the home care system, yet still represents a majority (60%)40 of Ontario decedents. Third, our data only captures if a patient was visited by both providers and cannot be used to infer a collaboration between providers.
Conclusions
Receiving a home visit at the end of life from an NP, physician or both is associated with reduced acute care use, increased symptom management medications, and more community deaths. These results support evidence on the benefits of integrating NPs into primary care, particularly supporting their involvement in end-of-life, community-based care.
Data availability statement
The dataset from this study is held securely in coded form at ICES. While data sharing agreements prohibit ICES from making the dataset publicly available, access may be granted to those who meet pre-specified criteria for confidential access, available at www.ices.on.ca/DAS. The full dataset creation plan and underlying analytic code are available from the authors upon request, understanding that the computer programs may rely upon coding templates or macros that are unique to ICES and are therefore either inaccessible or may require modification.
Ethics statements
Patient consent for publication
Ethics approval
The use of the data in this project is authorised under section 45 of Ontario’s Personal Health Information Protection Act (PHIPA) and thus did not require review by a Research Ethics Board.
Acknowledgments
This study was supported by ICES, which is funded by an annual grant from the Ontario Ministry of Health (MOH) and the Ministry of Long-Term Care (MLTC). This project has been funded in part by a contribution from Health Canada, Health Care Policy and Strategies Program. This document used data adapted from the Statistics Canada Postal CodeOM Conversion File, which is based on data licensed from Canada Post Corporation, and/or data adapted from the Ontario Ministry of Health Postal Code Conversion File, which contains data copied under licence from Canada Post Corporation and Statistics Canada. Parts of this material are based on data and information compiled and provided by: CIHI, the Ontario Ministry of Health, and Ontario Registrar General (ORG) information on deaths, the original source of which is Service Ontario. We thank IQVIA Solutions Canada Inc. for use of their Drug Information File.
References
Supplementary materials
Supplementary Data
This web only file has been produced by the BMJ Publishing Group from an electronic file supplied by the author(s) and has not been edited for content.
Footnotes
MMS and AR are joint first authors.
Twitter @sarina_isenberg, @mhoward101, @DocSandyB, @CarrieHeer, @amytmhsu
Contributors MS and AR were the lead authors, ABE and RM conducted the analysis, and all the other authors (SRI, CW, MM, MH, CEK, CAK, DM, AS, DP, SB, CB, PT, SD, KM, CH, DC, DM, ATH) were involved with all stages including design, conception, interpretation of the data, drafting or revising for intellectual content and final approval of the version submitted for publication. ATH is responsible for the overall content as guarantor and accepts full responsibility for the finished work as well as the conduct of the study, had access to the data, and controlled the decision to publish.
Funding This project has been funded in part by a contribution from Health Canada, Health Care Policy and Strategies Program. This work was also supported by the Canadian Institutes of Health Research (funding reference numbers PO4—187020 and PJT—185966).
Disclaimer The analyses, conclusions, opinions, statements and views expressed therein are those of the author and do not necessarily reflect those of ORG or the Ministry of Government Services or CIHI. The views expressed herein do not necessarily represent the views of Health Canada.
Map disclaimer The depiction of boundaries on this map does not imply the expression of any opinion whatsoever on the part of BMJ (or any member of its group) concerning the legal status of any country, territory, jurisdiction or area or of its authorities. This map is provided without any warranty of any kind, either express or implied.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.
Supplemental material This content has been supplied by the author(s). It has not been vetted by BMJ Publishing Group Limited (BMJ) and may not have been peer-reviewed. Any opinions or recommendations discussed are solely those of the author(s) and are not endorsed by BMJ. BMJ disclaims all liability and responsibility arising from any reliance placed on the content. Where the content includes any translated material, BMJ does not warrant the accuracy and reliability of the translations (including but not limited to local regulations, clinical guidelines, terminology, drug names and drug dosages), and is not responsible for any error and/or omissions arising from translation and adaptation or otherwise.