McGowan JG, Martin GP, Krapohl GL, et al. What are the features of high-performing quality improvement collaboratives? A qualitative case study of a state-wide collaboratives programme. BMJ Open 2023;13:e076648. doi: 10.1136/bmjopen-2023-076648
Why it matters
Quality improvement collaboratives bring together healthcare professionals from multiple organisations to work together on shared improvement goals. Though their structure and methods vary, they usually aim to accelerate the adoption of better ways of providing care across a group of clinical settings.
Despite their widespread use as an improvement strategy, collaborative improvement programmes have had mixed impacts in clinical services, and the features of these programmes that drive high performance are uncertain. Improving clarity about ‘what good looks like’ in how collaborative programmes are designed and delivered is crucial to improve the consistency with which they improve care.
The state-wide Michigan Collaborative Quality Initiatives (CQIs) programme is a rare example of a collaborative model that has achieved success in a range of clinical specialties over a long period. We aimed to identify the distinctive features of the programme that drive high performance, to help us understand “what good looks like” in the design and delivery of collaborative improvement programmes.
The setting for the study was the Michigan Collaborative Quality Initiatives (CQIs) programme, which is made up of multiple state-wide improvement collaboratives across a range of medical and surgical specialties.
We carried out a qualitative case study, which involved interviews with clinicians and managers, observation of a quality improvement meeting of the Michigan Surgical Quality Collaborative (the largest and longest-standing CQI), and analysis of programme documents.
What we found
We identified five drivers of high performance in the Michigan collaboratives programme:
- Learning from positive deviance
The collaboratives didn’t endorse any named quality improvement methodology, but instead, in keeping with the principles of positive deviance, focused on finding interventions that worked to address specific quality problems from within the community, and making them easy to share and implement widely.
- High-quality coordination
The organisation and actions of coordinating centres were foundational to the success of the collaboratives. Well-resourced, clinically-led teams with quality improvement expertise offloaded burdens associated with participation from local sites, and connected members from different hospitals to address shared quality and safety challenges
- High-quality measurement and comparative performance feedback
High-quality measurement and data were a crucial part of successful collaborative quality improvement. The Michigan collaboratives optimised audit and feedback as an improvement strategy; they understood that, without credible data, clinicians couldn’t reliably identify quality challenges or track the impact of interventions.
- Careful use of motivational levers
The collaboratives used a range of motivational levers to drive behaviour change, including competitiveness, peer pressure, and professional esteem. Healthy competition was seen as a key driver of improvement – clinicians were keen to be recognised by their peers as performing well.
- Mobilising professional leadership and building community.
The collaboratives gave clinicians a common purpose which supported their underlying motivation to improve care for patients as a community of professionals.
To do this, they had to create an environment of trust. This was achieved by setting boundaries around data use, insulating improvement activities from the influence of insurance companies, building on pre-existing social networks, and protecting clinicians’ autonomy.
Rigorous quality measurement, professional leadership, cultivating a collaborative culture, creating accountability for quality, and taking the burdens associated with participation in the programme away from hospitals all contributed to the high performance of the Michigan collaborative programme. Our study offers valuable learning for optimising the design, resourcing, and delivery of multi-organisational improvement programmes in healthcare.