The Enhanced Services for Avoiding Unplanned Admissionsi aims to ‘identify patients who are at high risk of avoidable unplanned admissions, establish a minimum two per cent case management register and proactively manage these patients’. The focus is on improving services for vulnerable patients and those with complex physical or mental health needs, who are at high risk of hospital admission or re-admission.
In Europe a 2013 consensus study states that “The European Union has placed specific importance on defining frailty, as frail persons are high users of community resources, hospitalization, and nursing homes. It is assumed that early intervention with frail persons will improve quality of life and reduce costs of care.”ii
Identifying patients means finding them and this can be difficult. The Enhanced Service Specification suggests the use of ‘a risk-stratification tool or an alternative method to identify vulnerable older people, high risk patients and patients needing end-of-life care who are at risk of unplanned admission to hospital’. Risk stratification tools in general use in the UK have included patients at risk of rehospitalisation (PARR) model, PARR++ and the Combined Model which to varying degrees make use of information which is readily available (e.g. presence of long term conditions, A&E admissions and outpatient appointments).
Limitations of risk stratification
The problem is the data that is readily available is often insufficient. It cannot know about those who may be at high risk but have not been in recent contact with the NHS. Often too few patients are identified, they are already well known and receiving extensive services. The group is too small for preventative actions to have much impact on unplanned admissions. Several studiesiii show that in order to significantly impact admissions intervention needs to be able to reach the larger cohorts that have a lower average risk.
Even where there is contact, useful data may not be recorded. For example our experience suggests that the GP record is unlikely to have an up to date measurement of mobility, let alone daily activities and social and emotional factors. However a patient’s daily activities and social and emotional state can provide very useful clues to someone’s level of risk. In a study (n=131)iv of a series of acutely admitted patients (in which overall 33% of admissions were identified as avoidable or somewhat avoidable) it was found that admissions due to poor mobility, falls or confusion were more likely to be avoidable or somewhat avoidable.
Another very relevant, everyday example that is difficult to model and manage data is the risk of bereavement. An elderly couple that compensate for each others’ functional difficulties may be low risk whilst both are alive but individually high risk in the event of the death of one of them.
The idea of Community Frailty Checks
In 2013 Intelesant started building its Howz platform, initially for Trafford CCG where Intelesant has enabled care homes to record and share Advance Plans with GP’s and ultimately to an Electronic Palliative Care Information Co-ordination System (EPaCCS). The platform was then extended to provide a way in which patients, families, care providers, social services and the NHS to share non-clinical information securely over the N3 network.
As we developed this system we realised that those in regular contact with an elderly person could, with the right tools, provide a stream of valuable observations to provide up to date intelligence on at risk patient groups in the gaps between GP visits or other contacts with the NHS. These observations would of course need a consistent structure. After extensive research and discussion with clinicians at University Hospital South Manchester (UHSM) Intelesant selected the Groningen Frailty Indicatorv (GFI) for this purpose as the GFI has a self-assessed version, it covers health, daily living and social and emotional domains and allows for grades of frailty to be identified.
South Manchester Project
The Living Longer Living Better Initiative of the Manchester Health and Wellbeing Board identified the frail elderly as a key cohort for innovation in more integrated care. This provided an opportunity to try out the idea of Community Frailty Checks and Intelesant were commissioned by University Hospital South Manchester on behalf of South Manchester CCG and working with Manchester City Council to do this. Intelesant also worked with Manchester Carers Forum to develop a training program and website for the community volunteers doing the Checks. The educational work was part funded by Health Education NorthWest.
The GFI question set was augmented with additional questions to provide more context. Governance is provided by an Integrated Steering Group of UHSM, Manchester City Council and South Manchester CCG. Completed Checks can be transmitted electronically to local GP’s over the N3 network. The pathway from this point can include referral to a multi-disciplinary team with the consent of the patient.
It is too early to say whether Community Frailty Checks can reduce unplanned admissions (though this is our goal). How might they do this? The most obvious way would be by providing the GP with a short list of names of people that might benefit from a clinical assessment leading onto formal early interventions by either the NHS or social services.
However as we have been doing this work we have seen other benefits that may be just as significant. The process of asking and answering the questions can help identify concerns the person could not explain or see the root causes of problems that they had missed or covered up. This can lead to getting help more quickly. The Community group doing the Checks gets a more systematic understanding of the needs of the people it is seeking to help.
Screening for frailty
One question that is often asked is whether it is appropriate to ‘screen for frailty’. There seem to be two reasons that this comes up. One is cost – which is a major consideration if this is to be done with clinical resources. The second is the ethical issue to do with telling someone that they have a condition that they were otherwise unaware of and for which ‘nothing can be done.’
Another is to do with the GFI tool we have chosen and whether this can be sufficiently sensitive for a diagnosis of frailty without, for example, a timed walk.
Our response is that a core idea of the Community Frailty Check is to provide valuable intelligence at low cost for a more detailed clinical review if needed. We do not tell people that they are frail or give them a score and neither is it our job to diagnose frailty. The feedback that they get from using the GFI is identifying the domains that they and their carers should focus on and pointers to local sources of help.
Intelesant has also started planning to link the responses to the Community Frailty Checks to local service directories. This could allow health and social commissioners to nudge residents to services designed to reduce the pressure on GP’s and A&E that would otherwise be difficult for them to find.
Intelesant is extending the Howz platform to include sensor data from the home, using low cost energy monitoring to detect the usage of everyday objects and systems that are a vital part of daily living for cooking, drinking and heating. Via a smartphone app the data is fused with reports of visits from formal and informal carers to create an up to date picture of the progression of frailty.
ii Frailty consensus: a call to action. – PubMed – NCBI . 2014. Frailty consensus: a call to action. – PubMed – NCBI . [ONLINE] Available at:http://www.ncbi.nlm.nih.gov/pubmed/23764209. [Accessed 02 December 2014].
iii Reducing emergency admissions: are we on the right track? | The BMJ. 2014. Reducing emergency admissions: are we on the right track? | The BMJ. [ONLINE] Available at: http://www.bmj.com/content/345/bmj.e6017. [Accessed 18 November 2014].
iv Avoidable acute hospital admissions in older people | British Journal of Healthcare Management 2012
v Measurement properties of the Groningen Frailty Indicator in home-d… – PubMed – NCBI . 2014. Measurement properties of the Groningen Frailty Indicator in home-d… – PubMed – NCBI . [ONLINE] Available at:http://www.ncbi.nlm.nih.gov/pubmed/22579590. [Accessed 18 November 2014].