Structure in the midst of uncertainty: the Gold Standards Framework and Amber Care Bundle
The British Geriatrics Society recently began working closely with the Gold Standards Framework on their Acute Hospital Accreditation Programme and the Amber Care Bundle, with their ongoing work to improve end-of-life care.
Both groups promise to improve the quality and consistency of care for patients as they approach the end-of-life, including for patients with uncertain prognosis, where recovery is possible but not guaranteed. Here we present a real-life case from a recent morbidity and mortality meeting and ask the GSF/Amber Care Bundle team to consider where they might help.
It was mid-November when the patient was initially admitted. He presented with a non-specific decline in mobility but was soon identified as having bilateral leg ulceration and was treated as having cellulitis. He had a background of alcohol dependency and recurrent falls. He had previously been seen by a geriatrician for hyponatraemia and chronic anaemia and no cause for either had been found. In the year leading up to admission, he had developed gradually increasing dependency. He had declined a care package but, at the point of admission, was struggling to get out-of-doors and had become dehydrated and malnourished. His house was in an awful state and upon his admission, his GP had notified the local environmental health team of hazardous conditions in his home.
He was admitted under a general physician and treated with antibiotics. After 21 days he moved to a slow-stream rehabilitation facility where a consultant geriatrician’s review highlighted the additional issues of subclinical hypothyroidism and hypovitaminosis D. He was commenced on a programme of strength and balance training, with rehabilitation goals around safe independent transfer in and out of bed.
Twenty days later, he fell whilst transferring out of bed and developed left hip and arm pain. He moved back to the acute trust site, where the orthopaedic surgeons diagnosed fractures of his left acetabulum and humerus. After four days of conservative management, with mild opioids administered orally, he was referred back to the rehabilitation team. They, however, declined his care as he was too confused and they were worried that he had a delirium, the source of which had not been clearly identified.
Seventeen days later, after treatment for a urinary tract infection, he was transferred back to the slow-stream rehabilitation facility. This time around, the consultant geriatrician review identified issues with polypharmacy (he had accumulated a complex combination of weak and strong opioids), worsening functional dependency (he was now bedbound), and worsening cognition (his MMSE had declined from 27 to 12/30).
Eight days later he was transferred back to the acute hospital facility, where he was again admitted under a general physician. His confusion had worsened further. He had become pyrexial. He was diagnosed as having sepsis secondary to urinary tract infection on the basis of monoclonal growth of E-coli from urine and an associated E-coli bacteraemia. He was commenced on intravenous antibiotics and fluids.
Seventeen days later, he was still bedbound and still suffering from marked cognitive impairment when he was moved to a “medical outlying” bed on an elective specialty ward, pending transfer back to slow stream rehabilitation. He remained there, without demonstrating functional or cognitive recovery, prior to transfer back to the slow stream care facility after a further 27 days.
After a further 4 days at the rehabilitation facility, he was moved, again back to the acute hospital, again under a general physician. He was once again treated for sepsis and once again showed improvement in his inflammatory markers and observations but failed to show any improvement in his cognition or functional status. His treatment was delivered on a general medicine ward. After 16 days he was deemed “medically fit for discharge”, to have “no rehabilitation potential” and was listed for transfer to a care home. The discharge co-ordination team, unhappy with his progress, asked for a geriatrician’s opinion.
The geriatrician opinion suggested that he had persistent delirium and identified a gradual worsening of inflammatory markers, which the base-ward team had failed to identify, since stopping antibiotics. Over a further 17 days he was recommenced on antibiotics, had an NG tube inserted, a central line inserted and underwent paracentesis for emergent ascites. On the 17th day, now in late April, he was once again reviewed by the consultant geriatrician, who identified that no progress was being made and the patient was close to death.
A meeting was convened with the patient’s next-of-kin. He, by now, had no capacity to participate in end-of-life decisions, and it was agreed that he should be commenced on the end-of-life care pathway, with plans for him to have palliative care delivered in his own home as part of a fast-track discharge scheme. The patient died, 21 days later, in hospital whilst awaiting this.
Our patient spent 173 days in hospital across 8 wards and 3 hospital campuses. He was under the care of 10 separate consultants over that period. This is far from gold standard care. It will, sadly, be all too familiar to those of us that work in acute trusts.
There are many aspects of this case that readers will identify where things might have been done better. There were, however, many points in his stay where simple day-to-day care, including symptom control, were approached in a poorly co-ordinated way. Whichever way the case is considered, this was the patient’s last year of life and his quality of care - and quality of life - were inadequate. We asked proponents of the gold-standards framework and Amber bundle how these schemes might have mitigated against this, even whilst continuing to actively pursue treatment?
Gold Standards Framework (click here to read the principles of the GSF)
Before admission to hospital, the patient might have been flagged up by the GP for inclusion on their QOF/GSF/ DES register, with team meeting discussions to prevent avoidable admissions. This might have led to the original admission being averted. If a person resides in a GSF trained care home, similar GSF coding and measures helps prevent admission (GSF care homes have halved admission and hospital death rates).
In the event of the patient being admitted, he would be identified, included on a register, and a needs-based coding would have been used to estimate his stage of decline. It is likely he would have been coded blue/ green on admission (see box for explanation of coding system). His condition and code would have been reviewed at a MDT discussion, and a proactive GSF Core Care plan would have been developed appropriate to his stage. A key worker would be allocated.
Clinical Assessment would proceed, including full geriatric assessment and use of other appropriate clinical tools. He would have been fully assessed for reversible conditions.
Personal assessment and Advance Care Planning (ACP). An initial ACP discussion with the patient and his family would have explored their understanding of his condition and future wishes and preferences. It is likely a preference for home care would have been expressed and staff would have taken measures to discharge him either home or to a care home, thereby preventing further transfers.
Regular review. If he was still in hospital, on deterioration his coding would have been reviewed at board rounds or at the MDT reviews and he might have been coded green/amber, with his needs being addressed, following the core care plan. If he still wished to be discharged home, a Rapid Discharge process would have been introduced.
In the event of further decline he would have been reviewed and coded amber/red with appropriate actions taken as above. With GSF leading to a reduced length of stay, it is more likely that care would have been given in a community bed when the patient was deemed medically fit.
Care in line with preferences. The ACP would have highlighted his and his family’s preferences, and any intervention the patient did not want. There would have been a focus on comfort care and palliation rather than the aggressive treatment that resulted in his increased risk of polypharmacy and falls, confusion with multiple moves and dying on a hospital ward. If he lacked capacity a best interest discussion would have been initiated.
Care in the final days (GSF Code red) follows the Five Priorities of care including a full discussion with the family, and this might have led to rapid discharge home or to a care home.
After-death care suggested by GSF includes support for his family, pointing to bereavement support and the giving of the death certificate. For the staff his care would have been reviewed and included in the regular ward audit.
Amber Care Bundle (click here for more detail on the principles of the Amber Care Bundle)
There were a number of opportunities both to involve the patient and/or relative in the treatment planning and to achieve greater coherence between the clinical teams during the prolonged spell in hospital. The decision to start the end-of-life pathway was taken very late and at a time when the patient was unable to participate.
We would have started the AMBER care bundle at the time the patient was transferred back to the acute site for the second time i.e. “Eight days later ..", which is about two and a half months before he died. It is possible that it should have been started earlier than this, when he became functionally dependent and developed delirium following a hospital fall.
Recognising the patient’s recovery was uncertain would have triggered a process whereby patient, family and the clinical team proactively identified the goals and appropriateness of treatments at a stage when the patient may have been able to participate. This would have included a discussion about the possibility of going home. The daily review using the ACT process would have meant that the initial decisions were reviewed with the patient and family on an ongoing basis rather than as a “one-off” planning meeting at the end-of-life care stage. The case example describes little about the patient, carers or nursing staff role during the admissions and whilst the AMBER care bundle may not have changed the interventions (such as NG tube feeding or central line use), it would have resulted in there being clear, shared goals.
Geriatric Medicine is a key speciality for using the AMBER care bundle as we frequently manage patients where there is poor prognosis and uncertainty about outcome. Use of the AMBER care bundle means that decision making is patient centred, proactive and shared with all of the team treating the patient (box 3). It doesn’t, on its own, solve the problem of the patient being treated in the wrong ward, by the wrong team without clear goals of treatment but it is a useful start.
GSF and the AMBER care bundle - the wardrobe and the tools
Hospital patients who are facing an uncertain recovery and who are at risk of dying in the next one to two months are helped to receive systematic care through the AMBER care bundle.
The care bundle is used to support care by promoting better teamwork and involvement of patients and/or carers in decisions about their care during a period of clinical instability and uncertainty. This ensures plans are agreed and co-ordinated across the clinical team.
The GSF is a framework that can be described as a wardrobe in which to hang other tools. It uses a four-colour needs-based coding scheme using anticipated prognosis.
The AMBER care bundle is a decision support tool that can fit within any care pathway or phase when a patient’s potential for recovery is uncertain. This period of clinical instability and uncertainty may occur in any of the GSF stages. The AMBER care bundle therefore complements existing tools identified in the Gold Standards Framework to support better patient centred planning and decision making during any phase of uncertainty in recovery.
We would like to thank the following people for contributing to the hypothetical alternative treatment paths of the patient journey described in the beginning of this article: Keri Thomas, Julie Armstrong Wilson and Chris Elger (GSF); and Susanna Shouls (AMBER Care Bundle)
References for AMBER Care Bundle: http://www.ambercarebundle.org/homepage.aspx
 Gott M, Ingleton C, Bennett M, et al. Transitions to palliative care in acute hospitals in England: qualitative study. BMJ 2011;342(d1773).