Improving patient and project outcomes using interorganisational innovation, collaboration and co-design

Key messages

• Collaborative multi-agency teamwork that includes primary care and mental health services can deliver improvements in health and wellbeing for whole populations.

• Delivering health improvements to whole populations can be facilitated by a combination of quality improvement methodology strongly linked to stakeholder participation in co-design of projects.

Why this matters to me

I know that healthcare services need to change the way they work together to better support people with common mental health problems and particularly people who have traditionally been excluded from services. To do this, we need to work together across traditional organisational barriers. I want to identify ways that patients and the community can work with healthcare professionals as equal partners to improve access and quality in our services and make a real difference to the health of individuals and whole communities.

Abstract

Background Common mental disorders (CMDs) are a leading cause of disability. The Department of Health has launched a large-scale initiative to improve access to evidence-based psychological treatments, such as cognitive behavioural therapy (CBT), through the Improving Access to Psychological Therapy (IAPT) programme. Access to IAPT services by black and minority ethnic (BME) communities is lower than for other groups.

Setting The London Borough of Ealing in west London; a diverse borough with areas of high BME population and relatively high deprivation.

Aim To compare the outcomes of two linked quality improvement (QI) projects undertaken by Ealing Mental Health and Wellbeing Service (MHWBS), both with the same aim of increasing access to talking therapies for BME communities.

Methods Application of QI methodologies supported by the NIHR CLAHRC for northwest London in two different settings in Ealing. One, the ‘Southall project’, was set within a wider initiative for collaborative improvements and shared learning (the Southall Initiative for Integrated Care) in an ethnically diverse area of Ealing; it was undertaken between April 2010 and September 2011. The second, ‘the Ealing project’, operated in the two other Ealing localities that did not have the advantage of a broader initiative for collaborative improvements; it was undertaken between April 2011 and September 2012.

Results Comparison of the monthly referral rates of BME patients (standardised per 10 000 general practitioner (GP)-registered patients) show that the Southall project was more effective in increasing referrals from BME communities than the Ealing project.

Conclusion Broad local participation and ownership in the project design of the Southall project may explain why it was more effective in achieving its aims than the Ealing project which lacked these ownership-creating mechanisms.

Introduction

This paper compares two approaches to increasing referrals of people of black and minority ethnic (BME) origin to the Improving Access to Psychological Therapies (IAPT) service within Ealing's Mental Health and Wellbeing Service (MHWBS). A sister paper (Improving access to primary mental health services: are link workers the answer?) explains the background: a complex intervention in Southall (2009–2012) aimed to develop the capacity of a community of primary care practices to improve services, in collaboration with multiple other organisations; one of its strands of work produced the Southall link worker project. The sister paper showed that general practices with a link worker from the MHWBS had a marked increase in referrals of BME patients to IAPT compared with those that did not.

At the time of the projects, the MHWB IAPT service employed cognitive behavioural therapists, wellbeing advisors, counsellors, mental health advocates, a physical activities co-ordinator and community development workers. The service was delivered by three teams covering three districts in Ealing: Southall, Northolt and Greenford (NAG) and Acton and Central Ealing (ACE). The MHWBS received referrals of patients of black and minority ethnic (BME) background from all three districts. This paper compares data from two different projects, both of which aimed to increase referral rates from people of BME background. One operated in Southall and was called ‘the Southall project’. The second operated in ACE and NAG and was called ‘the Ealing project’.

The Southall project emerged from a wider initiative for collaborative service improvements – the Southall Initiative for Integrated Care (Box 1). It operated between April 2009 and April 2012 and placed mental health link workers within six of the 23 Southall general practices. This project received a grant, training and project management support from the Collaboration for Leadership in Applied Health Research and Care for Northwest London (CLAHRC NWL), and it used NIHR quality improvement (QI) methodologies and ongoing analytical support for data collection, analysis and evaluation.¹

Box 1. The Southall Initiative for Integrated Care.

The Southall Initiative for Integrated Care operated between 2009 and 2012.² It included five mechanisms designed to enable multiple-way collaboration for on-going system-wide improvements:

1. An annual cycle of collective reflection and co-ordinated action. A sequence of events, scheduled long in advance, allowed stakeholders from different organisations to agree shared vision and priorities for improvement, devise co-ordinated improvement projects, and learn from these projects by putting in place synchronised improvements.

2. Interorganisational shared leadership teams engaged people from their own organisations and disciplines in the improvement projects that emerged from the annual cycle of activity.

3. Overt expectation that membership of shared leadership teams would change each year but four main sectors would continue to contribute to ongoing annual cycles of collective reflection and co-ordinated action – mental health, acute care, social care, and community care.

4. Clustering of general practices into convenient geographical groupings, with different practices leading development in different areas on behalf of the whole cluster.

5. Routinely gathered data about cost and quality of care within the GP clusters, to observe in real-time the impact of the collaborative activity.

The Southall Initiative therefore provided a systematic way for practitioners and managers that served the same geographic area to work with many different agencies to improve whole systems of care. The link worker project was one of the projects that emerged from the Southall Initiative.

The success of the partnership resulted in a second CLAHRC grant to apply QI methodology in the other two Ealing districts (ACE and NAG) – known as the ‘Ealing project’. The Ealing project ran from April 2011 to September 2012. The Ealing project used the learning and experience gained from the previous work in Southall but did not benefit from the wider initiative for collaborative working that characterised the Southall project. This presented a natural experiment – one project used CLAHRC QI methodology alone (ACE and NAG), whereas the other (Southall) nested the CLAHRC QI methodology within a broader initiative for whole-community participation in collaborative service improvements. This paper asks the question: which of these approaches was better at referring BME patients to the IAPT service?

Methods

Common aims

Both the Southall project and the Ealing project aimed to:

• increase referral rates from general practitioners (GPs) and self-referral from individuals;

• break down cultural barriers and reduce stigma; and

• increase the level and quality of information available to the general population, community and faith groups, and health and social care professionals about mental health and the services available in the borough.

The same activities were prioritised in the Southall and the Ealing projects:

• placing MHWBS link workers (IAPT clinicians) in GP practices;

• developing multilingual materials and services and culturally specific publicity – poster campaigns, business cards, fliers and leaflets;

• offering therapies in different languages;

• opening and promoting a self-referral line;

• reviewing therapeutic practices from a cultural perspective and making changes as required to accommodate cultural norms and expectations;

• working closely with local community and faith leaders;

• attending established community events, e.g. London Mela, to promote the service and raise awareness;

• outreach work with local community and faith groups to promote good mental health and raise awareness; and

• newspaper articles and local and national community radio shows.

Differences in BME populations

The percentage of BME population for each of the three districts in Ealing is: Southall, 71%; ACE, 35%; and NAG, 54% (based on 2011 Office for National Statistics census data).

Results

In the Southall project, six link workers were placed in six GP practices. The Ealing project was unsuccessful in placing any link workers in GP practices.

Figure 1 shows the referral rates to the MHWBS per 10 000 populations for all patients (not merely BME) for a 20-month period, including four months of baseline data and 16 months of project implementation for both projects; baseline dates were December 2009 to March 2010 for the Southall project and December 2010 to March 2011 for the Ealing project. Implementation dates were April 2010 to July 2011 for the Southall project and April 2011 to July 2012 for the Ealing project. The referral rates for the Ealing project are similar throughout the period of analysis, but were much higher than for Southall. Southall started out with very low overall referral and showed a steady increase throughout the duration of the project.

Figure 1.

Referral rates per 10 000 population for (A) the Southall project and (B) the Ealing project

Figure 2 shows the referral rates per 10 000 population of BME patients for the two projects. The BME referral rates per 10 000 for the Southall project increased throughout the duration of the project. The BME referral rates for the Ealing project remained unchanged throughout the study.

Figure 2.

BME referral rates per 10 000 population for (A) the Southall project and (B) the Ealing project

Discussion

We have presented here two linked projects, both of which aimed to improve access to talking therapies for patients of BME background. The two projects used the same approach, except that the Southall project was nested within a broader initiative for collaborative change. The BME populations of each district were different with Southall having 71% BME population and ACE and NAG having 34% and 54%, respectively.

Of the two projects, the Ealing project had the higher initial referral rate (from ACE and NAG) for all patients (approximately ten per 10 000 compared with four per 10 000 in Southall) and this remained unchanged throughout the period of data gathering. One possible interpretation for this lack of improvement is that ACE and NAG were already referring the optimal number of patients (and the intervention was not needed). This interpretation is supported by the observation that Southall referrals peaked at ten per 10 000 – the same level as ACE and NAG. But if ten per 10 000 is the ‘natural’ rate of referral we would expect a proportionate number of BME referrals (i.e. 7.1 in Southall, 3.4 in ACE and 5.4 in NAG). Yet the actual BME referral rate was two per 10 000 in both ACE and NAG, and this level was unchanged throughout the period of data gathering. In contrast, the referral rate in Southall started at two per 10 000 and ended at six per 10 000 – three times higher (but still short of the expected rate of 7.1 per 10 000). We can conclude from this that the Southall project was better at improving BME referrals than the Ealing project.

It is conceivable that the improvements in Southall were solely caused by link workers – the sister paper shows that the improvements in Southall were primarily in those practices with a link worker. The Southall project placed six link workers; the Ealing project was not able to place any. The strategic implication of this would be to create more link worker posts. Having ourselves witnessed the grassroots effect of these projects we believe that the collaborationcreating mechanisms are of greater, or at least of equal, significance. It was these processes that developed a sense of ownership and cultural change that made it possible to locate the link workers in the first instance.

This natural experiment reminds us of an old truth – targeted service improvement projects on their own are not enough to facilitate cultural change. Instead, evidence-based improvement tools need to be embedded within a broader collaborative, culture-changing effort for improvement.³ Clinical commissioning groups can do this by facilitating annual cycles of interorganisational reflection and action for change from which improvement projects emerge (as happened in the Southall Initiative for Integrated Care). The Southall project included local practitioners at every stage, including project design and choice of improvement measures – and it achieved its goals. Conversely the Ealing project did not enjoy these locally based ownership-creating mechanisms and it did not achieve its goals. Participation is the key: QI work is more likely to succeed when it is embedded within wider collaborative movements for change.

AUTHORS' CONTRIBUTION

KS led the quality improvement work; LE, SG and PT conceived the study; SG and FM performed the analysis. LE and CH wrote the first draft; all authors contributed to the revision of the manuscript.

ETHICAL APPROVAL

NHS Ealing provided overview of the project and considered ethical issues. Research Ethics Committee approval was not required as this was part of a locally led evaluated service improvement project.

FUNDING

The quality improvement work was supported by the NIHR CLAHRC for northwest London and NHS Ealing. The views expressed in this publication are those of the author(s) and not necessarily those of the NHS, the NIHR, the HS&DR programme, or the Department of Health.

CONFLICT OF INTEREST

None.