Hospital at home for acute medical illness: The 21st century acute medical unit for a changing population

Demand and supply in the acute medical pathway

The acute-care pathway is often conceptualized with the hospital at its core [1]. This centralization of acute care resource and expertize has led to improved outcomes across a range of conditions. Hospitals are able to absorb risk arising in other elements of the system by providing a default place of safety not restricted by opening hours. When concern arises in the community and the level of risk dictates the need for urgent further investigations or treatment there are few options other than to escalate care to the hospital setting. This model of care is threatened by sustained growth in demand in the context of relatively fixed capacity (Fig. 1a).

image

Rising acute care demand in the context of fixed bed capacity. (a) Number of emergency hospital admissions. (b) Number of acute and general hospital beds. (c) Proportion of acute and general beds occupied. Data obtained from NHS digital.

In the United Kingdom (UK), the number of emergency admissions is increasing at a rate which exceeds population growth [2]. There has been significant growth in the proportion of patients over the age of 65 years attending the Emergency Department (ED) and requiring emergency admission [3]. The complexity of emergency admissions is also increasing. Over one-third of the patients requiring emergency admissions have five or more health conditions, compared with only one in ten a decade ago [2].

Hospitals have absorbed year-on-year increases in emergency admissions despite substantial reductions in the number of acute beds (Fig. 1b) [4]. A broad match between demand and capacity has been maintained by dramatically reducing the average length of stay (LOS) associated with emergency admissions [2]. This pattern is not unique to the UK, over the last 20 years, the average number of hospitals beds per 1000 population has fallen by approximately 20% across Europe and by as much as one-third in some European countries [5]. The average LOS has fallen by 25% across the same time period [5].

The ability to manage increasing demand by pursuing further efficiencies in LOS may not provide a sustainable solution, in part because this strategy still relies on transfer from home to an increasingly congested acute care setting. In recent years, the NHS has seen significant deterioration in key metrics of performance [6]. Emergency Department (ED) overcrowding is the most visible manifestation of a model of acute care delivery under stress. Problems typically arise when restricted down-stream bed capacity prevents patients deemed to require ongoing in-patient care from being transferred to other areas of the hospital [7].

Operating close to the limits of bed capacity has important implications for the safety and quality of care beyond the ED. Bed occupancy levels above 85% have been linked to bed shortages, intermittent bed crises, and increased incidence of healthcare-acquired infections [8]. High bed occupancy rates are associated with higher risk of re-admission [9] and excess mortality [10]. Acute bed occupancy in the NHS is progressively rising and now consistently exceeds 90% (Fig. 1c).

During surges in demand association with the winter months, bed occupancy in excess of 95% is not uncommon [4]. Consistently working at the boundaries of capacity has important implications for the provision of elective care as the ability of a hospital to undertake planned surgical or diagnostic work is curtailed. Reversing trends in hospital bed numbers is an unpalatable solution to health-care providers operating under conditions of fiscal restraint. Established care models may no longer be fit for purpose under these conditions and new approaches to acute care are becoming an unavoidable necessity (Fig. 1).

A key policy response in the UK has been to advocate increased provision of same day emergency care (SDEC) [11]. SDEC is built on the premise that many emergency admissions to hospital are due to acute illnesses that can be effectively diagnosed and managed in a short time frame by providing rapid access to diagnostic tests and senior clinical decision makers, thereby reducing the need for in-patient care and overnight stays in acute hospital beds [12]. The approach selectively targets patients at low risk of clinical deterioration. The SDEC philosophy is increasingly applied to the design of acute services for older patients living with frailty [13]. However, this approach is difficult to apply in the context of severe acute illness necessitating ongoing treatment or functional impairment which precludes discharge. This is particularly relevant when trying to conceptualize an acute care system optimized for the challenges ahead. Frailty and severe acute illness tend to occur in tandem [14]. Modern services must be able to accommodate these elements simultaneously.

SDEC does not offer a panacea for hospitals struggling to cope with the volume and complexity of patients presenting to the hospital. SDEC services are no less susceptible to saturation in the face of rising demand than the hospitals they operate within. This is a design feature of an acute care pathway that consolidates the tools required to risk stratify and treat acute illness in the hospital setting. The need for some hospitals to emergently re-allocate the space dedicated to SDEC delivery in order to provide additional in-patient bed capacity highlights a potential drawback of this approach [15]. Reducing the capability to undertake SDEC may be maladaptive in the medium term, but a hospital at maximum capacity requires an immediate solution. Distributing some of the hospital functions to other parts of the acute care pathway may create a more resilient system by reducing dependence on the fixed capacity within the existing hospital infrastructure.

The hospital environment may not provide the optimal location of care for some patient groups. There is extensive commentary in the literature on the iatrogenic harms associated with hospital admission [16, 17]. This is particularly relevant to older people with frailty and multi-morbidity who are disproportionally affected by adverse events during in-patient care [18]. In-patient care can be complicated by falls [19], delirium [20], pressure sores, urinary incontinence, hospital acquired infection [21], and functional decline [22, 23]. The risk does not terminate abruptly at the point of discharge. Post-hospital syndrome describes a period of generalized risk to a range of adverse events in the immediate period following discharge [24]. The syndrome is felt to reflect stressors such as deconditioning, disturbed sleep, and nutritional deficiency which occur during the course of in-patient care which compound the physiological effects of illness and result in susceptibility to complications during recovery. A causal relationship between complications occurring during the course of in-patient care and the hospital environment is difficult to prove unequivocally. This should not preclude a search for potentially safer alternatives.

Introducing hospital at home as a model of assessment and intervention

Hospital at home (HaH) provides short-term, targeted interventions equivalent to that delivered within an acute hospital, but within an individuals’ usual place of residence. HaH is delineated from other community-based services by its role in managing acute conditions at a level of severity or complexity that would invariably require escalation to the hospital if the HaH option was unavailable.

This definition of HaH is important and frequently debated. The HaH term has been used in association with disparate models of care with different characteristics, many of which do not replicate hospital bed-based care. The need to establish a clear HaH identity is important given the plethora of services which operate in the community healthcare space. Some community services share features with HaH which can generate confusion. HaH services administer intravenous medications at home, but HaH is not primarily a home antibiotic service (commonly referred to as an outpatient parenteral antimicrobial therapy (OPAT) service in the UK) [25].

HaH services are multidisciplinary and give functional recovery high priority, but they are not reablement services [26] or restorative services [27] which provide short-term interventions to improve functional recovery following resolution of an acute illness. Establishing a common language is important to ensure patients, clinicians, commissioners, and policy makers have a clear understanding of the remit of the model.

HaH models are primarily distinguished from other services by the acuity and complexity of the patients they care for, frequently defined conceptually as patient group that would otherwise be in an acute hospital bed. The threshold for hospital admission, though, is ill-defined and whether the patient would otherwise have required hospital assessment is dependent on a degree of counterfactual thinking. Focusing instead on the processes of care that are delivered allows for a clearer understanding of whether ‘hospital level care’ is in fact being delivered in the home or care home.

The management of acutely unwell patients at home requires access to diagnostic tests, access to hospital level interventions and access to clinical decision makers all within a timeframe consistent with the clinical urgency of the problem. These key features have been summarized by the UK Hospital at Home Society (see Box 1). The exact specification of any individual HaH service is likely to be influenced by local need and existing infrastructure. The extent to which HaH care substitutes for the hospital may vary. A common form of HaH selects appropriate patients within the ED and provides ongoing care at home as if they were an in-patient. Conceptually, this can be imagined as the patient being transferred to a hospital ward, but the hospital ward is located in the patient's usual residence and contains only one bed. The treatment regimen, clinical review, and monitoring are provided by the HaH team and when resolution is achieved the patient is discharged back to the care of their general practitioner. BOX 1 UK Hospital at home society: Key features of hospital at home The acuity and complexity of the patient condition differentiates hospital at home from other community services. It provides urgent access to hospital-level diagnostics (such as endoscopy, radiology, or cardiology) and may include bedside tests such as point of care (POC) blood tests and point of care ultrasound (POCUS). It provides hospital level interventions (such as access to intravenous fluids, therapy, and oxygen). It requires daily input from a multidisciplinary team and sometimes multiple visits and provisions for 24 h cover with the ability to respond to urgent visits. It requires secondary care level specialist leadership and clear lines of clinical responsibility. Defined inclusion and exclusion criteria, with defined target population for example for over 18 or over 65. These programs deliver a time limited short-term intervention of 1–14 days. Hospital at home patients have equity of access to other specialty advice as though an in-patient.

Patients managed in the HaH model are typically considered as equivalent to patients receiving care in a hospital bed allowing privileged access to more sophisticated hospital level diagnostics, such as cross-sectional imaging or endoscopy. Care can still be escalated to an acute-hospital bed if the prevailing clinical condition dictates. HaH models may be particularly well suited to meet the care needs of older patients living with frailty. Frailty and comorbid conditions such as cognitive impairment and functional dependence are associated with prolonged LOS in patients admitted acutely to hospital [28, 29]. The potential operational advantages of a care delivery model that preferentially targets medically complex patients with above average LOS are self-evident.

Optimal medical care of acute frailty syndromes is not always reliant on investigations and treatments that need to be delivered in a hospital setting. The provision of comprehensive geriatric assessment, a multi-disciplinary process designed to address patient-orientated objectives with a focus on function, is as important as medical intervention when considering outcome following acute-illness [30]. Diagnostics and interventions routinely consist of common blood tests, plain X-rays, intravenous fluid, antibiotics, and diuretics while pausing potentially harmful medications. It is much easier to imagine this level of hospital care being emulated in the home setting. The hospital infrastructure required to support more aggressive medical and surgical interventions, such as cross-sectional imaging and intensive care, may not always be critical components of high-quality care in this context. The acute-care pathway for older people with frailty does not need to be built around proximity to these resources, although access to them may be required.

The potential to reduce the incidence of delirium is a key argument in favor of the HaH care. Delirium is a common presenting feature of acute illness in older people and frequently develops during the course of the in-patient care [31]. Delirium is an independent risk factor for mortality [32], prolonged LOS [33], and care home placement [34] in the 12-month period following acute hospital admission. Delirium can precipitate the onset of dementia and accelerate its progression [35]. Delirium is a multifactorial condition, and the relative contribution of environmental factors has not been established empirically. However, the importance of strategies to increase orientation in the management of delirium suggests the incidence and severity of delirium may be ameliorated by providing care in more familiar surroundings [36]. Providing care at home may also reduce deconditioning and associated functional decline which typically accompanies acute illness in older patients living with frailty.

HaH, envisaged as model of care tailored to the needs of older patients with frailty must be mindful of the balance between escalation of treatment in the hope of recovery and the risk of exposing patients to the unnecessary burden of active treatment without realistic chance of success. The need for an emergency admission is an important adverse prognostic sign in older patients and can be the antecedent of future decline or represent transition to a terminal phase of illness. Mortality in patients over the age of 85 years admitted to the hospital with acute illness approaches 50% at 1 year, five times higher than patients under 60 years of age [37].

The need to engage patients in conversation about their care preferences and record advance care plans (ACP) is a vitally important component of care which is frequently overlooked. Acute illness is often complicated by impaired capacity to participate in decision making which makes prior knowledge of a patient's preferences invaluable. The prevalence of ACP among patients admitted to UK hospitals with medical emergencies is low, even in specific patient groups who are well recognized as being at high risk of death within an year.[38] An effective ACP should ideally be concise, comprehensive, and universally recognized across care interfaces [39]. It is common for people to express the wish to die at home when given the opportunity to declare a preference [40, 41]. The objective is more likely to be achieved in patients that have also expressed a preference only to receive symptomatic management [42]. In a system that confines the resources required to manage acute illness in the hospital, a preference for active treatment and a preference to die at home are almost mutually exclusive.

When faced with potentially reversible acute deterioration, clinicians practicing in the community are faced with a dichotomous decision between escalation of care to a hospital, or inferior active treatment at home. HaH provides a third option, by facilitating more aggressive active management at home while being well positioned to support the transition to a purely symptomatic approach in the absence of a positive response to treatment. Figure 2 highlights the essential processes of care of an acute medical HaH.

image

Processes of care in HaH.

Established models of HaH and the current evidence base

The HaH model is not an entirely new concept and services focusing on older patients have been described in Europe [43-45], North America [46, 47], and Australia [48]. The HaH care is understood and practiced in a manner which reflects the local demand and existing healthcare infrastructure, as a result the models described are diverse in terms of organization and clinical processes. The extent to which HaH has been adopted at the national scale is difficult to determine with clarity as the HaH literature is formed primarily of studies investigating individual services covering specific geographical areas. International comparisons of HaH models are also challenging. HaH is specifically targeted at patients who would otherwise require admission to hospital, but admission thresholds are ill-defined and the product of various clinical norms and cultural factors.

In the UK, access to the HaH care is not universal, and the capabilities of individual HaH services vary considerably. The HaH model in Scotland is relatively mature, and supported by a number of governmental policy documents and guidelines to support a more consistent approach to service design and delivery [49]. This approach is less evident in other parts of the UK. A recent survey of acute hospitals in the UK suggested approximately half of the hospitals were able to refer directly to an HaH service [50]. The majority of HaH services described in the survey did not have the capability to provide an assessment by a physician at home or access to point-of-care diagnostics. This infers the services described were not designed to manage acute illness at levels of acuity that would typically require in-patient admission.

HaH has been investigated in multiple randomized controlled trials and has been the subject of several well-conducted systematic reviews and meta-analyses. The literature can be broadly summarized into studies investigating HaH models in specific conditions, such as decompensated heart failure [51] and exacerbations of obstructive airways disease [52] and studies which offer a more general appraisal of the approach [53, 54]. The HaH model consistently demonstrates equivalent or favorable outcomes in comparison with the hospital bed-based care. Reduced incidence of delirium is a frequent, but not universal, finding in studies investigating HaH models in older people [46, 55-57]. A recent UK multi-center randomized controlled trial of a geriatrician led HaH services for older patients with acute medical illness demonstrated no difference in mortality, but lower rates of delirium and lower requirements for long-term residential care in the patient group that received HaH [58].

The HaH literature is characterized by trials with relatively small-sample sizes and interventions which vary in both patient selection, clinical processes, and operational design [59]. The absence of a universally accepted definition of HaH makes interpretation of the outcomes reported from meta-analyses difficult, as shown by the variation in studies in Table 1. The potential for significant differences in the clinical processes which characterize each individual HaH model make estimates of the overall effect on outcome opaque. Meta-analyses which restrict study selection to HaH interventions that substitute hospital bed-based care for a substantial proportion of the acute care episode have demonstrated a statistically significant reduction in mortality [54, 60].

Table 1. Variation of HaH interventions in published studies Author Year/Location Study type Participants Intervention Outcome Shepperd et al. 2021 UK RCT n = 1055 Geriatrician led service guided by principles of comprehensive geriatric assess. Daily virtual rounds. Compared with hospital bed-based care measured at 1 month, 6 months, and 1 year • 65 years or older Most services provided intravenous medications and home oxygen • No difference in mortality. • Medical illness requiring hospital admission Assessment on average with 2 h of referral. • Decrease in proportion living long-term residential care 6 and 12 months. • Physiologically stable Direct access to acute-bed based diagnostics • Decrease in delirium at 1 month. Recruited at home or from short term medical ward Increased transfer to hospital 1 month follow-up. Shepperd et al. 2016 SR and n = 1814 Studies investigating ‘admission avoidance’ hospital at home. Defined operationally as ‘providing active treatments by health-care professionals in the patients own home for conditions at would otherwise require hospital in-patient care, and always for a time limited period’ Compared with hospital bed-based care measured at 6 months Individual patient level meta-analysis 16 trials The clinical processes employed within each study were not described. • Little or no difference in mortality • Acute medical conditions primarily affecting the elderly (n = 6) May increase likelihood of living at home • COPD (n = 3) • Stroke (n = 2) • Febrile neutropoenia (n = 1) • Cellulitis (n = 1) • Neuromuscular disease (n = 1) • Community acquired pneumonia (n =1) • Dementia (n =1) Levine et al. 2021 RCT n = 91 Daily visit from an attending physician and 2 daily visits from a registered nurse. Compared with hospital bed-based care measured at 30 days. • 18 years or older Intravenous medications, home oxygen, remote monitoring, video communication, and point-of-care testing were available • Lower mean adjusted cost • Recruited from ED • Reduced readmission • Specified range of primary diagnoses • Increased levels of physical activity Included infection, heart failure, COPD and asthma Equivalent rates of delirium Qaddoura et al. 2015 SR 6 studies Included services designed to manage heart failure by providing: No significant difference in all-cause mortality (3 RCTs) Decompensated heart failure • Hospital level treatment at home to avoid in-patient care • No difference in readmission (1 RCT) 3 RCT (n = 203) • Early hospital discharge facilitated by active hospital-level treatment at home • Increased time to readmission (2 RCTs) 3 observational studies (n = 329) All included studies recruited from the ED. Studies varied in the intensity of monitoring and available intervention. • Reduced readmission rate (3 observational) Excluded Reduced ED attendance rate (3 observational) • Primary diagnosis other than decompensated heart failure • Did not offer ‘hospital level’ care Goncalves Bradley et al. 2017 SR n = 4746 Studies investigating ‘early discharge’ hospital at home. Clinical processes heterogenous. Typically provided co-ordinated rehabilitation at home Compared with hospital bed-based care measured at 3 and 6 months 32 trials • No difference in mortality hospital outreach (n = 17) • No difference in readmission community-based services (n =11) • Reduced risk of living in long-term residential care hospital-based stroke team (n = 4) • Uncertain effect on cost Caplan et al. 2012 SR n = 6992 Hospital at home services that ‘significantly substitute’ of in-patient time. Significant substitution was defined ‘as the intervention group spending significantly less time in hospital, the duration of hospital at home care being either > 7 days or > 25% of the average length of stay for the control hospital admission. • Reduced mortality 61 trials Included trials investigating hospital at home interventions with medical, surgical, psychiatric indications and interventions aimed primarily at rehabilitation. • Reduced readmission rates Clinical processes employed within each intervention not defined. • Reduced cost Harris et al. 2005 RCT n = 285 Rehabilitation focus. Daily nursing review with support from a geriatrician. Provision of a 24-hour live-in carer when necessary. Point-of-care diagnostics and parenteral treatment not provided. Compared with hospital bed-based care at 90 days • 55 years or older • No difference in personal activities of daily living or cognitive function • Acute medical illness Increased cost per patient Recruited from ED or short stay ward within 36 hours of admission. Ram 2004 SR n = 754 Services designed to manage patients with acute exacerbations of COPD recruited from the ED. Randomized within 72 h of presentation. Care delivered by a specialist respiratory nurse. Medical interventions provided not specifically defined. Compared with hospital bed-based care Acute exacerbation of COPD • No difference in mortality (6 RCTs) 7 trials • No difference in readmission (7 RCTs) Excluded Reduced cost (4 RCTs) • Acute confusion or impaired consciousness • pH < 7.35 acute radiological or electrocardiographic changes Tibaldi et al. 2009 Italy RCT n = 101 Analysis of patients with heart failure treated within the ‘Geriatric Home Specialization service’. Compared with hospital admission measured at 6 months • 75 years or older Intervention included nurse and physicians’ visits. Parenteral diuretics could be administered. • No difference in mortality • Acute decompensated heart failure • No difference in readmission Established diagnosis of heart failure Increased mean time to first readmission

It is clear that patients that receive HaH within clinical trials tend to have better experiences of their care when compared with patients that receive routine hospital bed-base

留言 (0)

沒有登入
gif