How to coordinate care for people with long term conditions or other complex needs is a growing challenge in the UK, says Lara Sonola, but Canada provides a possible answer
There is an increasing number of people suffering one or more chronic conditions and they often experience fragmented care across health and social care settings.
‘Growing demand for nursing and personal care has pushed up the average cost per patient’
Across the world, countries have taken different approaches to developing care coordination strategies. Canada and the UK face similar problems integrating care for people with complex needs including fragmented finances and poor collaboration between acute and community or primary care providers.
Growing demand for nursing and personal care has pushed up the average cost per patient, while local areas have sought to implement programmes of integrated care for older people to improve patient experience and quality of care while controlling costs.
Positive results
Some small, evaluated pilot programmes across Canada have demonstrated promising results. In Quebec, the preferred reporting items for systematic reviews and meta-analyses (PRIMSA) approach was developed in 1994 to improve continuity of care for older people.
The model uses an integrated service delivery network of health and social care providers alongside case management delivered by multidisciplinary teams. The key clinical features are:
- coordination between services;
- a single point of entry;
- case management;
- an assessment tool;
- a personalised care plan; and
- an information tool.
The PRISMA model
The PRISMA model has also implemented a governance model operating at strategic, tactical and clinical levels. The governing body responsible for the project and strategic planning included key decision makers from health and social care, clinicians and local community agencies.
‘The model significantly reduced both patient functional decline and visits to the emergency department’
At the tactical level, service managers formed an operational committee, while at the clinical level multidisciplinary teams developed care plans and managed care needs.
A four year evaluation comparing patient outcomes of the project in three pilot areas against three control areas found that the model significantly reduced both patient functional decline and visits to the emergency department.
Levels of patient satisfaction and empowerment also increased and yet the PRISMA model did not appear to reduce costs or significantly effect hospital admissions.
Identifying design features
These findings correlate to the key markers for success identified in the King’s Fund’s comparison of care coordination models.
In the two year research project at the King’s Fund, supported by Aetna and the Aetna Foundation, five UK examples of care coordination were studied.
The aim was to identify the main design features of the coordination process and draw out key lessons to implement care coordination in different contexts and settings.
‘There was a lack of robust evaluation and routine measurement of clinical outcomes or patient feedback’
The five sites were based in Midhurst, Surrey; Oxleas Foundation Trust in Greenwich and Bexley; Sandwell in Birmingham; South Devon and Torbay; and Pembrokeshire.
Our cross-cutting analysis identified a series of important design features including the need for multiple referrals into a single entry point, the value of case management delivered by a named care coordinator to provide personal continuity, the production of a tailored, holistic care plan that identifies the needs of the patient and carer, and the value of targeting these approaches effectively at service users.
Across the sites there was a lack of robust evaluation and routine measurement of clinical outcomes or patient feedback. All the programmes were in the process of developing strategies to measure impact and could demonstrate some positive results, particularly in terms of improving clinical outcomes and quality of life.
Overall, the quality of the evidence was fairly weak with little ability to attribute positive outcomes to the care coordination process. This differs from the Canadian experience using the PRISMA model, where outcomes for patients in the intervention group were compared to patients in three control groups over four years.
Gathering evidence
Although these types of evaluation are timely and expensive, failing to gather rigorous evidence is a significant weakness of many current models in the UK.
The main success factors identified from the PRISMA model were that leaders from local health and social care providers, provincial government and regional health bodies worked with researchers to form a long term collaborative partnership that supported the design, implementation and evaluation of the initiative over 15 years.
‘Care coordination models flourished at the neighbourhood level, where they benefit from close engagement with and knowledge of the local community’
These features echo the markers for success we identified in the King’s Fund’s research. We observed that successful care coordination programmes require time to mature, develop robust evaluation methods and build legitimacy to be accepted by professionals and patients.
This is facilitated by consistent leadership from key individuals who have nurtured alliances within the local health and social care economy over years, a feature evident in the South Devon and Torbay case study. In addition, our case study sites continuously worked to refine and improve their programmes.
Big picture
Unlike many small scale care projects in the UK, the Quebecois PRISMA approach was not a short lived pilot. Following the evaluation, it was rolled out across the province as part of a wider integrated care policy.
When the Ministry of Health adopted the model, flexibility was built in to enable each region to tailor the programme to their population.
This was also observed in UK research. We found that care coordination models flourished at the neighbourhood level, where they benefit from close engagement with and knowledge of the local community.
‘Care coordination models flourished at the neighbourhood level, where they benefit from close engagement with and knowledge of the local community’
In Pembrokeshire, the implementation of integrated health and social care teams, called community resource teams, differed across the county to account for local differences and relationships.
Some of the challenges observed in the UK case study sites are mirrored in the PRISMA model such as GP engagement, which was variable across our sample of care coordination programmes.
In Oxleas’ advanced dementia service this contributed to slower than anticipated progress in identifying patients for referrals and in their participation in the care coordination process.
GP engagement
South Devon and Torbay achieved far greater levels of engagement, with GPs hosting the virtual ward model, building on existing relationships with commissioners and using financial incentives to encourage and acknowledge the time required to deliver care coordination.
In Quebec, PRISMA achieved 73 per cent participation from primary care physicians during the first four years, although they did not attend multidisciplinary meetings to discuss patients with case managers directly.
As in Quebec, we found that programmes of care coordination need to be built from the bottom up to ensure effective partnerships, rather than introducing new top-down models of care, no matter how well they may have worked in other settings.
‘Programmes of care coordination need to be built from the bottom up to ensure effective partnerships’
In Midhurst, following the sudden closure of an inpatient palliative unit, staff worked with commissioners and Macmillan Cancer Support to conduct a consultation exercise with user groups, local GPs and neighbouring hospices.
This facilitated involvement from local stakeholders and the establishment of an innovative service tailored to local needs. However, this is not to underestimate the importance of securing wider political, organisational and professional support.
Integrated care pioneers
The most stable programme in our sample, at South Devon and Torbay, benefited from active support and leadership from health and social care commissioners for integrated models of care over a period of many years.
Integrated care has come to the fore in the health policy landscape with the announcement of 14 integrated care pioneers and the better care fund.
In Quebec, the PRISMA approach did not demonstrate a positive impact until the third year of the evaluation. It should be remembered that new approaches need time as well as support to demonstrate real improvements in care for patients.
Lara Sonola is senior researcher of health policy at the King’s Fund
1 Readers' comment