Article Text
PDF
Abstract
Background: The heart failure (HF) virtual consultation (VC) is an eHealth tool for delivery of peer-to-peer specialist advice to general practitioners (GPs) to discuss HF diagnosis/management. We aim to investigate the impact of the VC service on onward referral rate and quality of assessment by GPs, as well as assess VC patient characteristics; Clinical Frailty Score (CSF), age and morbidity.
Methods: This prospective observational study collected VC data on: demographics, comorbidity, frailty, referral indication, the impact of VC on clinical care and the GP response to the question ‘what would you have done without the VC service’. We compared patient characteristics to a control population of patients attending the HF unit (HFU) (n=118).
Reults: Between 2015 and 2021, 1681 VC cases were discussed. The majority of cases were discussed from remote areas (75%). Rediscussion cases increased from 0% to 34%. VC patients were older (76.2 (±11.3) vs 73.1 (±12.5) years, p<0.05), more frail (CSF=3.8 (±1.7) vs 3 (±1.6), p<0.01) and multimorbid (number of comorbidities=7.1 (±3.4) vs 3.8 (±1.9), p<0.001) compared with patients attending the HFU. Without the VC, 93% of cases would have been referred to face-to-face hospital services. Instead, VC resulted in only 9% of cases being referred to hospital services. The remainder of cases were managed by the VC service, in a shared GP-specialist approach. GP use of natriuretic peptide (NP) increased from 0% in 2015–2016 to 63% in 2021 and use of TTE increased from 0% in 2015–2016 to 69% by 2021.
Conclusions: The VC service provides a platform for case discussion in particular for older, frailer patients and reduces onward hospital referrals. This may facilitate early diagnosis and management of suspected HF in the current era of long outpatient waiting times. The quality of community HF assessment improved as indicated by increased use of NP/TTE by GPs.
- HEART FAILURE
- Health Services
- Delivery of Health Care
Data availability statement
Data are available upon reasonable request.
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
Many face-to-face hospital outpatient services have long waiting times, which have been exacerbated by the COVID-19 pandemic.
Early heart failure (HF) specialist input reduces HF morbidity and mortality.
WHAT THIS STUDY ADDS
Using the virtual consultation (VC) can offer underserved populations (older, frailer, multimorbid and living more remote) early specialist input and provide a platform for continuity of shared care with the general practitioner.
The quality of HF assessment by general practitioners improved over time, as evidenced by incremental use of natriuretic peptide and transthoracic echocardiography prior to VC case discussion.
HOW THIS STUDY MIGHT AFFECT RESEARCH, PRACTICE OR POLICY
Early specialist input via the VC could facilitate early diagnosis and improved up-titration of guideline-directed medical therapy to patients with/suspected HF.
Background
With the ageing population, improved survival from ischaemic heart disease, as well as effective life prolonging heart failure (HF) therapies,1 the burden of HF is set to grow.2 This, alongside the cancellation of many non-emergency services during COVID-19, challenges the current and future capacity of specialty services to assess all referred patients in standard face-to-face clinical services in a timely manner.3 Current National Institute for Health and Care Excellence (NICE) guidelines advise for those with suspected HF to be reviewed by a specialist within 2–6 weeks.4 Unfortunately, this is poorly adhered to, with long delays in diagnosis and treatment,5 which results in worse outcomes.6
Patient age and frailty are also increasing,7 which can make it difficult for patients to travel to face-to-face services for specialist opinion. As chronic disease management appropriately transfers to the community,8 the general practitioner (GP) becomes ever more central to HF management. Providing GPs with the right tools, alongside streamlined specialist HF input, to be able to manage diagnostic challenges and implement guideline-directed medical therapy (GDMT) is key to effective community HF management.5 This is especially the case in an era of increasing HF complexity, with growing multimorbidity,9 which likely contributes to the current poor implementation of GDMT.10
The HF virtual consultation (VC) is a method previously described,11 which was established in 2015 as an eHealth tool for the delivery of specialist care and a mode to transfer HF knowledge and management into the community.12 It does this through a remote biweekly, real-time video-conferencing platform with a HF specialist and a group of GPs, who bring forward diagnostic or management issues related to their patients with HF or suspected HF. It offers both early specialist input (with a referral time of 2–6 weeks depending on clinical urgency), as well as to impart education and knowledge to each VC case discussed. In addition, it minimises the need for patient travel, an important issue in particular for frailer individuals and their carers. The VC service has already demonstrated some positive results, with GPs reporting a perceived improvement in knowledge, self-efficacy and patient-related outcomes, alongside good user acceptability of the service.11 The service has now been established for 6 years, and our hypothesis is that patients availing of the service are those less likely to be able to travel, such as older, frailer, more morbid patients.
This study’s primary objective was to assess the impact of the VC service on onward face-to-face hospital referrals and compare this to ‘what the GP would have done without the VC service’. Secondary objectives were to compare Clinical Frailty Score (CSF), age, morbidity and number of medications between patients reviewed in the VC service and compare them to patients reviewed in the face-to-face HF unit (HFU). Finally, we wanted to investigate the use of natriuretic peptide (NP) and transthoracic echocardiography (TTE) by GPs at the point of referral to the VC.
Methods
This project was a prospective, descriptive observational study with an additional comparative control arm to assess patient characteristics between two services. From the inception of the VC in 2015, data were prospectively collected on each VC case. The structure of the VC required the GP to provide the following information before the case discussion: GP location, patient demographics, management before the VC (including whether NP and/or TTE had been performed) and the specific question to be addressed. These data were presented in slide format during the VC to facilitate discussion. During the VC, we asked the GP presenting the case; ‘what would you have done without the VC service’. This was recorded as: referral to hospital for face-to-face review (HFU/general cardiology/acute medical assessment unit (AMAU)/emergency department (ED)/other specialty), referral to other community-based service, manage the patient by themselves or other. Following completion of the VC, the decision taken was recorded and transmitted electronically to the GP surgery and filed.
In addition, we prospectively collected data from the St Vincent’s Healthcare Group HFU to compare CFS, comorbidity, number of medications and the age of patients reviewed in the HFU versus the VC service. The HFU is an ambulatory face-to-face tertiary hospital service, where patients with HF or suspected HF can be reviewed regarding: a new diagnosis of HF, up-titration of GDMT, review post hospitalisation or for hospital admission avoidance purposes. HFU patients were recruited consecutively until the calculated sample size was reached to reduce the risk of selection bias. This project was a supported pilot service initiated by the Health Service Executive (Irish Healthcare Service) as part of the Sláintecare programme which is an overarching strategy to redefine structures for chronic disease management. Patients or the public were not involved in the design, or conduct, or reporting, or dissemination plans of our research.
Statistical analysis
Results were collated and analysed using SPSS V.27. Parametric data were expressed as their mean ±SD, while non-parametric data were expressed by their median and IQR. Categorical data were expressed as frequencies and percentages. Comparison of CSF, comorbidity, medications, age and gender between patient groups were conducted using independent samples t-test or χ2 analyses. Covariate analyses were undertaken using Analysis of Covariance (ANCOVA) and Analysis of Variance (ANOVA) for age and gender adjustments. Listwise deletion was used for VC missing data. All statistical tests were two-tailed with a p value of 0.05 defining statistical significance.
Sample size calculation
To address the primary objective in this descriptive observational study, we abstracted data from all available VC cases from 2015 to 2021. For the secondary objective, which was a comparative arm assessing patient characteristics, we assumed there would be at least a one point difference in the CFS, in patients who were able to travel to the HFU compared with patients who were discussed through the VC. Previous unpublished data from our HFU demonstrates an average CFS of 3 (±2). With an alpha of 0.05 and a beta of 0.90, enrolment ratio of 1:1, this resulted in at least 84 patients to be recruited from the HFU to demonstrate a difference.
Results
The VC from 2015 to 2021
Over 6 years, a total of 1681 VC cases were discussed, with an increment from 53 cases in 2015 to 310 cases in 2021. Table 1 compares VC and HFU baseline characteristics.
A comparison of VC and HFU case baseline characteristics
The VC population demonstrated almost equal proportions of males and females. The mean VC age was 76.1 years (±11.2), older than those attending the HFU (73.1 years (±12.5), p<0.05). A total of 333 GPs used the VC service, ranging from presenting 1 case discussion to 40 case discussions (median=2 (IQR 1–4)). Many VC cases were discussed with GPs who were geographically remote, with the majority of cases being discussed from outside of Greater Dublin (75%), which is where the VC hub and HFU are based. VC patients had a mean CFS of 3.8 (±1.7), greater than the mean CFS (3.0 (±1.6), p<0.01) of those attending the HFU (n=118). VC patients were also multimorbid, with a mean number of comorbidities of 7.1 (±3.4) compared with 3.8 (±1.9), (p<0.001) in those attending the HFU. Polypharmacy was prevalent, with a mean number of prescribed medications of 9.6 (±4.5) but no difference compared with patients attending the HFU (10.0 (±4.0), p=0.417). After adjusting for age and gender, there remained a statistically significant difference (p<0.05) between the VC and HFU patients in terms of frailty and comorbidity. The presenting symptoms of VC patients were predominately dyspnoea (45%), peripheral oedema (24%) and fatigue (17%). The dominant reason for consultation was assistance with diagnosis (70%) and the remainder was related to management queries or interpretation of cardiac investigations.
Referral patterns from VC
Of 1669 VC cases, only 9.0% (n=150) were referred to onward face-to-face hospital services. This included general cardiology outpatients (n=98, 5.9%) and HFU (n=52, 3.1%). All other cases were managed by the GP and further VC discussion as needed.
Figure 1 demonstrates a comparison of actual VC referrals versus what the GP would have done without access to the VC service (years captured were 2017–2021, n=1286). Without the VC service, there would have been 92.8% onward face-to-face referrals to the hospital. This would have included referral to outpatient clinics (n=641, 50.0%), HFU (n=441, 34.3%), acute medical assessment unit (AMAU) (n=92, 7.1%) and the ED (n=18, 1.4%). Rediscussion of VC cases increased from 2015 to 2021, initially with no rediscussion cases in 2015 or 2016, to more than a third of cases (n=105, 34%) cases in 2021.
A comparison of actual (green) referral outcomes from the GP after using the heart failure virtual consultation versus what the GP would have done (blue) if they did not have access to the virtual consultation platform. Referral outcomes include elective referrals (OPD/HFU), urgent or emergency outcomes (AMAU/ED), GP only care and GP-shared care with VC (this is the concept of the GP and heart failure specialist sharing the care of a patient using the VC platform). AMAU, acute medical assessment unit; ED, emergency department; GP, general practitioner; HFU, heart failure unit; OPD, outpatient; VC, virtual consultation.
Transfer of initial investigations to the community
Figure 2 demonstrates GP requested NP and TTE prior to VC case discussion. Initially, there were no patients who had NP screening or TTE prior to their cases being discussed in VC between 2015 and 2016. Following on from this, the presence of NP and TTE results prior to VC case discussion subsequently improved. In 2017, 17% of cases had an NP test prior to VC discussion, this steadily increased from 43% in 2018 to 63% in 2021. This can similarly be seen in GP request of transthoracic echocardiography prior to the VC, which steadily increased from 0% in 2015–2016 to 27% in 2017 and up to 69% by 2021.
The percentage of VC patients each year who had GP requested natriuretic peptide testing (A) and transthoracic echocardiography (B) prior to case discussion in VC. Number in brackets is the number of VC patients who had the investigation each year. GP, general practitioner; VC, virtual consultation.
Discussion
Following on from the initial preliminary report on the HFVC service,7 the data presented here provide a more robust analysis of the referral patterns and value of this platform of care for patients with HF. The data show that over a period of 6 years there was incremental growth in the use of the service from 53 cases in 2015 to 310 cases in 2021. The service effectively reduced onward referral to face-to-face outpatient clinics or acute/emergency services from 93% of cases to 9% and was used for continued review and assessment as indicated by an increasing number of review consultations over the years.
As anticipated, the HF population who availed from the VC service were older, multimorbid and more frail compared to patients attending face-to-face HF services. This is an important, under-served HF subgroup, as older patients with a higher burden of frailty are significantly less likely to achieve optimal GDMT, which can contribute to their higher risk of HF hospitalisation and death.13 It is also possible that comorbidity and frailty burden are associated with a reduction in the likelihood of referral for specialist opinion by standard routes, which our data supports. Harnessing the VC offers a unique opportunity for specialist input for early diagnostic purposes, diuretic adjustment to reduce likelihood of acute decompensation, as well as optimisation of GDMT in older, frailer patients, thereby potentially improving outcomes.
Current standard outpatient services are unable to keep up with the increasing burden of disease.3 14 A wider array of strategies to provide care and disseminate knowledge is critical, especially as challenges of increasing prevalence, frailty and complexity of care develop.9 These data indicate that VC can be a significant contributor to the modern structure of HF care, addressing many of the present-day challenges of chronic disease management. In addition to possible benefits (e.g. access to specialist input, but without the travel, time or cost) to individual patients and their carers, the use of this eHealth platform facilitates specialist delivery to remote/rural areas; thereby improving standardisation of HF access and increasing equitability of HF care.
Our experience with VC underlines the capacity of this platform to be part of an evolving outpatient strategy. Addressing the question in a real-time fashion with the GP also disseminates knowledge, which translates into greater confidence among GPs in HF care.11 This could reduce specialist need in subsequent cases. Over 6 years, there was increasing use of NP and TTE by GPs prior to VC case discussion. This indicates an improvement in the quality of HF assessment by GPs. This has occurred alongside improvements in accessibility of these tests in the community, which is also an essential step towards facilitating streamlined management of HF care in the community. A limitation of the study is accounting for other factors that could influence the quality of HF assessment, such as training/teaching of GPs external to the service. However, it is likely that the VC service played a significant role in the improvement of HF assessment, as demonstrated by our previous report assessing GP confidence, knowledge and perception of patient outcomes, when using the VC service.11 Despite many outpatient/elective services being significantly affected during the COVID-19 pandemic, figure 2 demonstrates a sustained use of GP requested TTE and NP between 2019 and 2021. This highlights that even during the pandemic, where there would have been competing influences and causes of shortness of breath, the quality of HF assessment was preserved.
Our data also indicate an increasing use of the platform for ongoing care as opposed to ‘once-off’ advice. This demonstrates the flexibility of the platform and is of particular use in patients who are older, multimorbid and frail. In other circumstances, the VC platform can be used where urgent or semiurgent advice is needed in emerging decompensation, or to encourage speeding up titration of GDMT. In addition to the different aspects of HF management, the VC platform may be of value in other chronic diseases or even in other community settings such as nursing home residents.
Finally, two key areas that warrant further research in the VC arena are: the cost-effectiveness of using a VC strategy versus usual face-to-face care and understanding patient perspectives of shared GP/specialist care versus referral to face-to-face hospital appointments.
In conclusion, we observed that the VC platform which offers early access to HF specialist input is durable and effective and may be particularly effective in remote, older, multimorbid or frailer patients, who may be compromised in terms of ease of access to specialist care. Over the course of 6 years, GP use of NP and TTE improved, indicating an improvement in the quality of community HF assessment.
Data availability statement
Data are available upon reasonable request.
Ethics statements
Patient consent for publication
Ethics approval
Formal ethical committee approval was not sought; however, we complied where possible with key principles set out by the Declaration of Helsinki.
Acknowledgments
Dr Carmel Halley RIP who was a leader in the VC project but sadly passed away before the paper was written. This project was a supported pilot service initiated by the Health Service Executive (Irish Healthcare Service) as part of the Sláintecare programme, which is an overarching strategy to redefine structures for chronic disease management.
Footnotes
Twitter @DrBethWong
Contributors BW: Data collection, analysis and drafting manuscript. JC: Data collection, analysis and drafting manuscript. MB: Senior clinical review and drafting manuscript. ML: Data analysis, senior clinical review and drafting manuscript. LM: Data collection. RF: Senior clinical review and manuscript review. JG: Senior clinical review and manuscript review. KM: concept/project manager, guarantor, senior clinical review and manuscript review.
Funding This work was performed within the Irish Clinical Academic Training Programme, supported by the Wellcome Trust and the Health Research Board (Grant Number 203930/B/16/Z), the Health Service Executive, National Doctors Training and Planning and the Health and Social Care, Research and Development Division, Northern Ireland and has been supported by the Sláintecare project within the Health Service Executives, Ireland.
Competing interests None declared.
Provenance and peer review Not commissioned; externally peer reviewed.