Hospital at Home: A Literature Review

Fig 1. Summary of Research Into Hospital at Home

Since 2000 UK governments have advocated home based care as a reform that can shrink expensive hospital capacity.  

The National Beds Enquiry (2000) recognised ambulatory and intermediate care facilities as a feature of systems which relied less on hospital beds.  This led to the incorporation into the NHS plan (DoH 2000) of a £900 million investment by 2003/4 in Health and Social Care ‘to promote independence and improve quality of care for older people’. 

Whilst the Government recognised that ‘intermediate care cannot be the responsibility of only one professional group or agency’ (DoH, 2001) the programme was positioned as an alternative to hospital care ie to ‘avoid admissions’ rather than as a synergistic collaboration between hospital and community providers along a single pathway.

Evidence of Efficiency

Proponents of integrated models of care assert that “integration of care components in newer health systems will maximize patient benefits and organisational efficiency” (Boult and Pacala, 1999).  Indeed, a review of thirteen integrated care programmes from the US, UK, Sweden and the Netherlands (Ouwens et al, 2005 cited by Ramsey and Fulop 2008) reported several beneficial themes:

1. reduced hospitalisation
   (reported in 6 programmes, 3 with significant trends)
2. increased process capability
   (reported in 5 programmes, one of them with a significant trend)
3. Increase in functional status and health outcomes
   (reported in 9 programmes, one of them with a significant trend)
4. Increase in patient satisfaction
   (reported in 3 programmes)
5. Increase in quality outcomes
   (reported in 6 programmes)
6. Decrease in costs
   (reported in 4 programmes)

However, UK evidence of efficiency is less compelling.

The leading US Health Management Organisation (Kaiser Permanente) was highlighted as using far fewer hospital beds per head of population than the NHS (Feacham et al 2002). This triggered a series of UK pilots testing the prevention and case management elements of the Kaiser model. 

Whilst some success was reported in terms of fewer admissions and shorter stays, the results were in line with national patterns so could not be attributed to the new models of care.  Nor was there any explicit evaluation of cost efficiency (Ham 2005).

Whilst the rhetoric of integration was used, the pilots were commissioner-led and focussed on reducing hospital admissions by providing accessible alternatives ie they sought to circumvent rather than integrate with an essential step in the care pathway. 

Evidence of Clinical Appropriateness

There is evidence to suggest that community supported models can provide equivalent care for the right patients.

A UK study comparing outcomes for in-patient versus home-based care at three months concluded that there were no significant differences between the two methods of care except where patients had complex health needs (Shepperd, et al 1998) in which case the in-patient model was superior.

A US study of 455 patients provided with acute-level care at home (including treatment of cellulitis, pneumonia and exacerbation of COPD) also concluded that outcomes for appropriately selected patients matched those of a hospital care model (Leff et al 2005). 

Whilst, this form of care is not right for all patients (Shepperd, et al 1998, Burns & Pauly, 2002) it can benefit “segments of the patient population such as the elderly, the frail elderly” (Burns & Pauly 2002) although other research warns that readmissions from home-care are higher in elderly patients needing complex medical care (Leff et al 2005). 

Balancing hospital care for elderly patients is not easy.  Their admitting complaint is typically an exacerbation of one aspect of a more complex condition and they often need social support as well.  These patients can dwell in hospital beds beyond their need for acute care because of poor coordination and collaboration between hospital and community/social infrastructure (Kao & Walter 2009). 

Extended hospital stays also can adversely affect the function and health of patients, particularly older patients (Kao & Walter 2009) implying a greater need for ongoing care.   Such delays have been attributed to “efforts on the part of individuals, departments and services to make themselves efficient without regard for the resulting organisational consequences.” (Scott and Hawkins, 2008). 

This lends credence to The National Service Framework for Older People (DoH, 2001) which emphasises the need to integrate health and social care services and specifically envisages a coordinated team including general practitioners, hospital doctors, nurses, physio therapists, occupational therapists and allied health professionals.

Evidence of Patient Preference

UK evidence (Shepperd, Doll and Broad 2009) shows home-based care is popular with patients; and US evidence (Burton 1998) shows home-based care is more comfortable and puts less burden on families. 

Further, patients’ carers appear to prefer domiciliary care with perceptions being reinforced by experience (Ojoo, Moon & McGlone 2002).  This view is qualified however by the See-saw Report (Harvey et al 2008) that care delivered closer to home whilst preferable to patients may result in the “burden [falling] on informal carers”.

Evidence of Cost Effectiveness

One UK study found that hospital-at-home can deliver care at similar or lower costs than an equivalent admission to an acute hospital (Jones, Wilson, Parker 1999). 

Another UK meta-study reviewed evidence for the transfer of specialists to primary care and joint working with acute care delivered equivalent outcomes however hospital capacity was not reduced and so overall demand increased. (Sibbald, McDonald and Roland, 2007)

In a meta-study, Ramsey and Fulop (2008) reviewed three modes for achieving integration.

1. Organisational: citing King et al (2001) and their review of children’s services during the 1990s - concluding that integrated working does not require organisational integration, and that organisational integration can result in an imbalance of power between hospital and community interests.
2. Integrated Multidisciplinary Teams: citing the work of Goodwin et al (2004) and Hamilton et al (2005) as recording improvements in working practices and even care outcomes but no cost advantages.
3. Contractual Coordination of Care Delivery: citing Ahgren (2003) as reporting no significant change to systems or services and resistance from front-line staff; and Ahgren and Axelsson (2007) which reports better outcomes where the focus is on outcomes rather than structure and led by local respected staff delivering the care.

On balance then, available evidence of reduction in cost of care resulting from home-based models is at best it is contingent, if not conflicting.

Evidence of Ancillary Benefits

Ramsey and Fulop (2008) report that integrated care can “break down barriers between primary and secondary health care, as well as health and social care” setting a platform for trust and closer collaboration.  There is also evidence that “more integrated care could have a number of benefits, including: responding more effectively to the needs of people…, improving the effectiveness and safety of care…” (Ham et all 2008)

Enthoven and Tollen (2004) record reduced management overhead in Health Management Organisations (HMOs) where payer and provider are integrated and monitor demand and plan care together.  This view is supported by Shih et al (2009) who point to information systems saving “operating expenses in managing health care supply chains.”

However, Burns and Pauly (2002) are not so supportive of this opportunity, arguing that where HMOs work properly they do so because of special not easily replicable circumstances including few competitors, an established multidisciplinary team and an already strong patient service culture. 

Again, the evidence suggests potential but the results appear ambivalent or contingent on local circumstances.

Guidance on Achieving Integration

Little practical guidance is offered.

Six layers of healthcare integration have been suggested, namely: organisational, service, clinical, functional, normative and systematic.  (Fulop et al, 2005 adapting Cantandriopoulos et al, 2001).  And where integration has achieved some success, academics have credited both foundational work and enabling facilities.  This includes:

1. The development of common, standardised activities and new organisational structures (Burns and Pauly 2002) and
2. Clinicians being able to “…effectively access and share information in a timely manner” (Shih et al 2009).  This assertion is invariably linked to the use of information technology (Shih et al, 2009; Ham, 2005).
3. “Voluntary collaboration” of partners is “vital to achieving sizable supply chain transformation.” (Shih et al 2009 citing Schneller 2006; Ahgren and Axelsson, 2007)
4. The “pivotal role” of case managers in linking health and social care (Johri et al 2003 as cited by Ramsay and Fulop 2008)

Barriers to Collaboration

One of the biggest challenges to sustained collaboration is the differing priorities of health and social care.  Health is free at the point of need while ASC has to means test and recover funding from patients and their families.

The boundary between hospital and social care is recognised as particularly challenging.  The complex mix of different funding models, reporting lines, and priorities is seen as underpinning failures of ‘inter-professional’ collaboration (Glasby 2003).

The NHS Plan 2000 and the Health & Social Care Act (2001) introduced the idea of Care Trusts as a means of Health and Social Care collaborating by pooling budgets and commissioning health and social care jointly.  However, whilst these organisations have made it easier for front line teams to coordinate care, they have not proved transformational and there remain concerns about their financial efficiency (Glasby and Peck, 2005) and the extent to which social workers can retain the scope and values underpinning their profession (Limbrey 2006).

Conclusions

Pressure to reduce the cost of national health provision is undeniable and policy frames the solution to that challenge as reducing hospital capacity by relocating care to a community setting.  However, evaluation of the performance of home-based models of care has been inconsistent and the results whilst suggesting potential are ambivalent on the questions of safety and value for money. 

Where home-based models of care have been explored, patients appear to prefer the non-hospital setting but evidence of clinical efficacy is qualified with cautions that this form of care is not always appropriate or safe.

Older people emerge as a group that may benefit from a home-based care but this same cohort, improperly managed, experience readmission.  It is also clear that any model serving elderly patients requires support from social care which presents difficult relationship challenges particularly at the boundary between hospital and home. 

The evidence of financial savings resulting from home based care is at best contingent if not conflicting.  Overseas work provides better support that UK-based research which is limited and inconclusive, not least because these attempts have seen commissioners seek to buy from ‘elsewhere’ ie to substitute rather than integrate with hospital provision; and to specify activity rather than care outcomes.