Welcome to the event blog for the latest IS4 Leaders workshop (7-9 February 2016) held at The Landing, MediaCityUK, Salford Quays. Dr Cameron Whytock, Haelo’s Improvement Science Fellow and Junior Doctor, reviews the three-day event as a delegate.
IS4 is Haelo’s flagship improvement programme, developed and delivered in partnership with the Manchester Academic Health Science Centre (MAHSC), NHS Quest and Manchester Cancer (now part of Greater Manchester Cancer), a bespoke team based training programme in improvement science that includes a blend of web-based learning, improvement coaching and site visits.
— NHS Quest (@NHSQuest) February 7, 2017
Today we welcome 11 teams from cohort three of the IS4L programme, based on the Breakthrough Series Collaborative model. We join participants for workshop three, following the second action period in the 12-month programme. The theme for workshop three is planned experimentation. To join the conversation on twitter use #IS4L
— Tazeem Shah (@ShahTazeem) February 7, 2017
At Improvement Science for Leaders @_Haelo looking forward to seeing how this might influence health research into patient impact.
— Mark Taylor (@MTaylorOxford) February 7, 2017
Kurt Bramfitt, Senior Improvement Advisor at Haelo and IS4L Course Director opens the workshop with a recap of the journey so far, including theory and plans for the next three days.
To read more about Deming and to refresh your knowledge, we strongly recommend you take a look at Maxine’s blog: ‘What is improvement science anyway?’ You can access information and resources from workshop two in the event blog (4-6 October 2016) or remind yourself of Maxine’s blog following workshop one (3-5 May 2016).
Guest speaker, Dai Roberts, IS4 Alumnus and Programme Development Lead at GM AHSN is the first up to present, My Improvement Journey.
His case study is regarding poly pharmacy in a hospice setting. The project was to reduce the number of tablets/capsules taken per day in St Anne’s Hospice over an eight month time period. Dai talks the room through his driver diagram, process mapping, tests of change and the results.
In particular he highlighted the importance of testing through small real-time PDSA cycles before scaling up. Sometimes when exciting work is occurring the scale-up can happen itself as others often want to get involved!
“Small tests of change are key. We started to design this long complicated questionnaire, then realised we just needed to ask one or two questions and build tests of change around those… We kept one ward ‘naïve’ however quite soon after the first ward began the work filtered to other wards.”
Dai continues: “My background is in health services research and making the change to improvement science was initially challenging. We couldn’t have done it without understanding Deming. It wasn’t just quality improvement, it was improvement science.”
09:50am This week, Haelo are delighted to host Lloyd Provost, API associate, member of IS4 expert faculty and internationally renowned expert in the education and application of improvement science. For those who don’t know, Lloyd literally wrote the book, as one of the authors of the Improvement Guide.
Lloyd will be teaching and coaching teams over the next three days, before they embark on their final action period ahead of the summit event in May 2017.
Lloyd begins his first session on Improvement Science in Practice by demonstrating the ability to apply improvement methodology to a small project from start to finish.
Using the Model for Improvement, Lloyd sets teams up for the Coin Slide exercise. Teams are tasked with improving the process of accurately sliding the coin to the correct distance within the confines of the regulations, with Lloyd warning “coin sliding is a heavily regulated industry”! However the process is changed within these through the use of PDSA cycles. This mirrors improvement efforts within health and social care.
— Tazeem Shah (@ShahTazeem) February 7, 2017
— Haelo (@_Haelo) February 7, 2017
— Helen Cottrill (@_HelenBarrett_) February 7, 2017
A start and finish line are taped on each table and teams are asked to record the slide distance in relation to the target. The teams are on their feet as they run PDSA cycles, learning which modifications to the process are successful and which are not.
There is a discussion around what they have learned about the process allowing the sharing of improvement ideas across the ‘collaborative’ of teams.
11:10pm After the break Lloyd continues with Improvement Science in Practice, looking at developing our change ideas, running and documenting those ideas in PDSA cycles and measuring what happens over time. The data from these tests are transformed into charts (Xbar and S charts) in Excel.
Teams were encouraged to use the ‘6 hats’ method to generate ideas. In turn the teams put on a different ‘hat’ – White hat to think objectively, yellow hat to think logically, red hat to think emotionally, green hat to think creatively, black hat to think critically and blue hat to control the discussion. This allowed the problem to be viewed from different perspectives and ideas to flow, Lloyd said: “A lot of the time we get caught up in our current processes. We need to break out of that and think differently.”
12:00 Each workshop, all attending teams are asked to present on their project progress so far. Team presentations are an opportunity for teams to update the teams and faculty, share learning and face challenges with others. Following each presentation, the room are asked to feedback their reflections and apply that to their own projects.
The first team to present is DreaMS, Christies, Cancer Vanguard team who aim: for 95% of patients (in the pilot group) will have a documented Goals of Care (GOC) conversation that addresses what is important to them when deciding on treatment.
Carol shares how their aim has changed multiple times throughout their project so far. However, they are at PDSA 5, which involves wider testing on 46 people (30 patients) in 3 oncology team groups, having built up from just 4 patients in PDSA 1.
They are very pleased with the improvement with regards the depth of discussions that are being documented between specialist and patient as part of the GOC discussions and Carol shared one particularly good documented conversation that occurred. This documentation is then sent to the patient’s GP where they can continue the conversation. Despite these discussions, clinics have not appeared to run over time.
Lloyd advised that sharing the initial data when it is available is important, to see how it is received by the teams involved.
12:40pm Break for lunch.
13:25pm Following lunch, Haelo’s Chief Executive and IS4 expert faculty, Professor Maxine Power is next to take the stage.
In this session Maxine delivers, Engaging the Attention Economy, in the hope of building the will to change.
“What were you doing in 1990?” Tim Berners-Lee had invented the internet one year previously; Maxine describes how meanwhile she was using traditional mail to communicate with colleagues in her role as a speech and language therapist. The internet ended up changing the world and brought the ability to spread ideas quickly. The teams were asked to think about what platforms they were currently using to promote their projects.
Maxine explains the importance of using digital technologies to create excitement about work being done. “You need to spend as much time thinking about what you do in the digital space, as you do in real life. When people show interest in a project [in the digital space], this inspires those involved in the work.”
To do this, Maxine explores the power of engaging people through film and media, “When you want people interested in to your work, it is not a technical act, it is an emotional act. You have to tap into how they feel. This is about ‘innovating with messages’.”
The room are shown a short film produced by the PolyPals team from a previous cohort, led by Dr Dai Green who the teams heard from this morning. The task is the same for our teams as it was for PolyPals – produce a short film within this leanring session.
13.25pm Inspired by the power of video, the exciting next session, Lights Camera Improvement! is led by Haelo’s expert film team. Stephen Miller, Multimedia Content Producer and Thomas Harvey, Multimedia Project Assistant, support Haelo’s mission to be digital by default; leading the way in healthcare films.
The practical film workshop shares simple methods and techniques to create engaging films. The duo inspires the room to project films into their project! Teams are tasked to develop a short film related to their project, which will be premiered on the final day of the workshop.
15:50pm We return to team presentations, with the final presentations of the day from:
Gateway C, Greater Manchester Cancer Vanguard team, who aim to: Increase the proportion of referrals for suspected cancer made by 8 pilot GP practices through the 2 week wait pathway by May 2017 by XX%. This ‘XX%’ has just been added to the aim and is yet to be determined but baseline data have just been collected and it will be based upon its analysis.
Cathy and Ewan kick us off with their project. Straight away they noted one positive aspect of their work – they have produced a network of individuals across the different organisations involved.
They showcased their online portal with case studies, pathways and online learning; to change practice and behaviour of GPs, in early detection of cancer. This is especially important with conditions like colorectal cancer, where initial symptoms can be unclear.
There’s exciting work going forward to collect more data on an ongoing basis, rolling out existing learning modules to more GPs, and designing more modules. The aspiration is to have this rolled out to 5000 GPs!
Next up is I-Said, NHS Quest team from Royal United Hospitals Bath aim: To reduce adverse incidents involving insulin for adult inpatients with diabetes by 75% by May 2017.
Mark explores the team’s progress so far. He noted their driver diagram has gone through many iterations as the work has proceeded. This is followed by the process map, and he comments this was more complex than first anticipated.
An impressive number of PDSA cycles have been conceived. “We’re not lacking in ideas, just the time to try them”. One innovative test is allowing patients (who are familiar with self-management) to control their own insulin while an inpatient. With PDSAs in general, he emphasises the value the team have found in keeping them small before building up, a theme touched on already today.
Being a small team, challenges remain about undertaking time-intensive work, especially regarding measurement. Maxine added that Charles Vincent’s Measuring and Monitoring of Safety Framework encourages use of real-time data, and that in fact having ‘just enough’ of this type of data is perfectly fine to tell an improvement story.
Day one has drawn to a close. Thanks you to all the teams and we look forward to seeing you tomorrow! News from day two, coming soon.
— Haelo (@_Haelo) February 7, 2017
Very Excited to welcome our #IS4L teams back for day 2 of 3 in our final workshop. Today we explore planned experimentation
— Kurt Bramfitt (@KurtBramfitt) February 8, 2017
09:10am Lloyd Provost is next to address the room, introducing Planned Experimentation Design Principles. This session explores the theory and methods involved in designing a multi factor test.
First we do our tests in a very small scale, then we scale up. Planned experimentation comes in with wider-scale testing, involving more structured testing to accelerate learning beyond that available from individual PDSAs. Lloyd said: “An experiment is a study designed to provide the basis for action. Planned experimentation is a collection of approaches and methods to help increase the rate of learning about improvements to systems, processes or products.”
— Haelo (@_Haelo) February 8, 2017
10:40am Back from a short break, Lloyd presents Running a Factorial Design.
The bubble game illustrates factorial design, a type of planned experimentation. Each team has two ‘workers’ to blow bubbles, one ‘quality inspector’ to assess the size of the bubbles, one ‘data analyst’ to record the data on a spreadsheet and one ‘manager’ to co-ordinate the team.
Each team initially identified their ‘best’ worker after producing I-charts to visualise their data. However the problem appeared more complex than initially thought. What if there were differences in the bubble solution, or the blowing wand?
The teams use the principles of factorial design by studying different factors together. This provides further learning on which interventions, or combination of interventions, are most effective in producing larger bubbles.
Teams can relate this to their own projects in complex health and social care systems, where there can be multiple factors that have influence on a problem.
12:30pm Break for lunch
13:15pm In Lloyd’s final session of the day, we look at Applying Planned Experimentation. This is an opportunity for teams to apply learning to their own project, creating a factorial design and planning methods for analysis. He explores the difference between methods for research and methods for improvement. Research methods such as confidence interval and p-values have their place (and rely upon random sampling), but in improvement we need ways to look at things. Lloyd said: “In improvement traditional research methods can keep us constrained and stop us from learning.”
Lloyd then gives examples of factorial design projects he has been involved in to illustrate the main points covered and encourages the teams to apply today’s learning on factorial design to their own projects. They then break out to work on this.
14:55pm After a short refreshment break, we are delighted to welcome our guest speaker, Graham Martin, Professor of Health Organisation and Policy, University of Leicester. In this session we explore qualitative methods and their use in quality improvement.
— Kurt Bramfitt (@KurtBramfitt) February 8, 2017
Graham starts by exploring the role and usefulness of qualitative data. He states there is value in “…adding some qualitative texture to quantitative data.”
Using a qualitative analysis of one of the most famous QI initiatives ever performed, the Keystone Michigan project, as an example: “There was an awful lot behind the scenes that the quantitative data didn’t pick up…Why did it work better in some sites and not others?…If you don’t do this kind of work, it’s harder to then reproduce things in other settings.”
— Graham Martin (@Graham_P_Martin) February 8, 2017
Graham said: “Studies that make the best of qualitative data are those that look at the interactions between things… Quantitative methods are good at exploring ‘what’ and ‘whether’… What qualitative methods are best for is answering ‘why?’ and ‘how?’.”
We then explore interviews and ethnography, discussing the benefits and limitations of each method. In particular, there is a human-interaction element of qualitative inquiry that illuminates aspects of a situation that would be hard to obtain from purely quantitative data.
He has a couple of ideas for how the teams could use qualitative data to enhance their projects:
15:50pm Before we close day two, the room are set to hear two team presentations, both NHS Quest teams.
The first from Silver Strength, from Wigan, Wrightington and Leigh NHS whose aim is : To undertake the Comprehensive Geriatric Assessment (CGA) on all patients over the age of 75 within 24 hours of being admitted to the Age Well unit.
Micky kicks off by describing the intervention, which is a ‘lengthy’ MDT assessment, and talks about efforts to streamline the patient pathway. One focus area has been to improve identification of eligible patients early on in A&E and arrange an appropriate Age Well unit bed, where the CGAs would then take place.
Josie then takes the mic and further describes their work including, in a similar vein to the qualitative data presentation from Prof Graham Martin earlier, the team “…have kept a log that looks at positive and negative issues”. This is something other teams may find useful. They both explain recent difficulties with the project considering the high volume of patients over the winter period but feel they have regrouped now and are ready to move forward!
And finally Hackney in IIT from Homerton University Hospital who aim to: Increase the numbers of service users receiving multidisciplinary care within seven days of the start of the team’s intervention by 50% by March 2017.
Similar to Silver Strength, a more streamlined process is a part of the project and Ruth talks us through that new process. She then moves on to some of the PDSAs undertaken, including an example of unsuccessful PDSAs such as aspects of MDT meetings with geriatrician involvement. These unsuccessful tests are often more useful with regards learning than successful ones! However there have been some great successes for the team too, such as time for input from psychotherapy.
Plus the overall outcome measure is also looking like it’s nearly met! The team have identified that having tried many PDSAs at once, it’s hard to tell which ones have led to the improvement and Ruth acknowledges the material on factorial design today may be relevant. Keep up the great work!
Another packed day of learning closes with the opportunity to meet with faculty members for coaching. Coaching is available to teams first thing and at the end of each day.
— Haelo (@_Haelo) February 8, 2017
Welcome to the final day of IS4L cohort two, workshop three. We begin the day with a film showcase! Led by Ste and Tom from Haelo’s video production team, the teams present their films created on day one in the Lights Camera Improvement session.
— Haelo (@_Haelo) February 9, 2017
Broughton Believers have agreed to use their video
Considering some teams were pushed outside their comfort zone, it was great to see what was produced in just 1 hour of work! There was a variety of different styles, conveying different messages. The use of visual media can be very powerful in conveying a message, and building ‘grassroots’ support for new ideas or ways of working. We hope the teams consider the value this could bring to their projects. Resources have been sent out via email that could be useful in any future video work the teams might like to undertake
— Mark Taylor (@MTaylorOxford) February 9, 2017
Onto the first team presentations for day three. We kick off with Team Imp4act from CMFT who aim to reduce self-reported non-adherence to Methotrexate, as the exemplar drug, in patients with rheumatoid arthritis attending the Kellgren Centre for Rheumatology, CMFT, by 50% by Dec 2017.
Rachel takes us through the overall rationale behind the project – that adherence to methotrexate for rheumatoid arthritis is poor based upon its perceptions as ‘a cancer drug’. After a brief exploration of the driver diagram and measurement strategy, she then presents the tests of change employed – motivational interviewing for the clinic staff, an ‘agenda setting’ tool, a virtual referral system and standardised advice to patients about methotrexate and alcohol.
The data for the outcome measure looks great, with a clear improvement since the start of the project. We’re looking forward to seeing how things look in three months’ time.
Next up is Arun from A-Team, whose aim is to ensure 85% of outpatients under the care of SRFT intestinal failure rate their outpatient experience as ‘positive’ by February 2017.
He explains their area of work is a relatively new field, so it has been an interesting area to undertake improvement. There’s some great data on display tied in well to the measurement strategy, which gets us at Haelo very excited! Perhaps the most striking chart is one showing the amount of cumulative travel miles patients have been saved by the use of telemedicine, and the resultant improvement in the carbon footprint.
Members of other teams in the room offer helpful suggestions, such as getting more feedback from patients on their experience using the telemedicine approach. Lloyd adds a quick reminder to distinguish the baseline from the action period on the charts. Overall some very useful feedback from the room. That’s one benefit of these three days, to share the learning and get different perspectives on the great work being undertaken.
11:00am Lloyd Provost is now as the front, as we go into our final set of skills sessions. ‘Holding the Gains’ develops people’s understanding of what is needed to sustain and spread improvement. Lloyd guides the room to embed measurement and processes, starting with the difference between testing and implementation:
— Tazeem Shah (@ShahTazeem) February 9, 2017
Testing is trying and adapting existing knowledge on small scale, learning what works in the system. Implementation is making the change part of the routine day-to-day operation of the system in your pilot population.
Do you have high confidence in your change idea(s)? Are your staff ready for change? If so, it’s time to “shift gears” from testing to implementation!
— Meghna Jani (@MeghnaJani) February 9, 2017
Lloyd explains, “The model is the same. Continue to use the PDSA for implementation.” He then goes on to explore further aspects of implementation that differ from testing, for example, greater consideration of the social aspects of the change or ways to publicise it. This segment finishes with a team exercise to reflect on implementation of their change ideas.
12:00 noon In our penultimate presentation session, we welcome two more teams into the spotlight. The IMPS team, from CMFT & USHM, aim to improve experience of antenatal care and outcomes of pregnancy for parents who have experienced a stillbirth in a previous pregnancy by May 2017.
Alex explains this aim has been revised following feedback and the driver diagram is on about iteration number 11! This often happens as a project progresses, and is completely fine.
One helpful aspect of the work is automated data collection. This can be a huge benefit to projects if this option is available. The team have also added in a qualitative data source and believe the talk from Prof Graham Martin yesterday will be useful in helping them work through that. One result is that the service has managed to increase their capacity (number of appointments) by 25%, with no negative effects on the balancing measure of non-attendance.
Lloyd suggests that data in different method of delivery could be presented in p-charts.
Next up is Query Cancer, from Greater Manchester Cancer Vanguard Innovation, who aim to reduce the time to diagnosis for suspected cancer patients to 14 days. This involves development of an MDT clinic and an evaluation of whether this clinic can reduce times to diagnosis.
Maryna sets the scene around the difficulties with cancer referral pathways and the focus areas for improvement, such as preventing the system being saturated with the current 97% of suspected cancer referrals who (luckily) end up being disease-free. Another area is with diagnostic testing traditionally being done sequentially, involving multiple visits. The team are attempting a shift to more rapid diagnostic process in the ‘one-stop’ clinic.
There looks to have been lots of activity. Remember to record these in the PDSA framework to enable useful reflections and drive further improvement ideas!
She also talks about how the team are considering the use of factorial design going forward off the back of the session yesterday. The faculty are here to help if any of the teams would like more guidance on this!
12:30pm Break for lunch
13:15pm: We move on to the next team presentations starting with Team Screenology, another Cancer Vanguard team who aim: To increase the uptake of the 3 national cancer screening programmes by 10% by 2021, by focusing on improving the access to screening services across GM. Graham whizzes through the project. However at the moment there is a separate RCT looking at breast screening being undertaken and the project has been paused for the moment, so as not to affect that RCT.
Overall there are plans to affect 480 GP surgeries and their breast screening practices and he identifies work to be done when things start up again after completion of the RCT, including promoting the project and working up further interventions using the methodology covered at IS4. An example is that the team that wants to make use of factorial design and describes the session yesterday as ‘an epiphany’ as to one way to pursue this work in the future.
And last but certainly not least are from Salford CCG, the Broughton Believers, who aim to improve flu vaccination uptake across the groups defined in the driver diagram by 3035 vaccinations from 2015-16 baseline by end March 2017.
Fraser starts by discussing the mission to engage patients and GP practices in Broughton, and they believe they can make a big difference when it comes to flu vaccine uptake! He explores the driver diagram (version 7), including PDSA cycles hanging off the drivers. A nice tidy driver diagram always goes down well.
There are some fantastic data in P-charts showing an increase in flu uptake in some GP practices. However Fraser emphasises the team’s insight that it’s about more than the outcome data itself, but also that the learning gained from undergoing this improvement journey is valuable for future flu seasons.
Thanks to all the teams for their presentations over the past three days. Great to see how things are going and we look forward to hearing more at the graduation in May!
14:00pm: Now for the final time, we hear from Lloyd in Scaling Up and Spreading Improvement. This session will again equip attendees for their final action period.
Teams develop an understanding of the difference between implementation and scale up. They also explore models and frameworks to support scale up and spread of their change ideas.
Using the example of James Lancaster in 1601, who discovered that 3tsp of lemon juice prevented scurvy in sailors, he explores the concept of ‘spread of ideas’ and the fact that Lancaster’s idea took 146 years to spread to the Navy. There are techniques to make sure our good ideas spread faster than Lancaster’s!
One first step is to recognise the important distinction between ’scale-up’ and ‘spread’. ‘Scale-up’ describes movement of ideas around institutional restraints (within the original site) and may include ‘winning people over to the idea’. ‘Spread’ is replicating the work in new institutions or sites where the issue is one of resources, rather than will.
— Helen Cottrill (@_HelenBarrett_) February 9, 2017
There are numerous ways to spread ideas that avoid the 17 year wait Helen has tweeted about, such as Breakthrough Series Collaboratives, campaigns, diffusion agents etc. See the Prezi slides at the end of this blog for more info on methods for both scale-up and spread.
One key concept is Roger’s Diffusion Curve. When looking at this concept, and applying it to our projects, the question we want to answer is “How do we engage our likely early adopters?” This is because their engagement sets off the more rapid adoption by the majority.
Close Finally we hear once again from Kurt Bramfitt as he brings the event to a close. We look ahead to the requirements for the Graduation event and consider next steps to ensure teams are prepared for action period three.
The teams are then invited to get refreshments and reflect on the last three days. It’s been a packed three days for the collaborative; sharing learning, celebrating progress and overcoming challenges together.
See you at the summit!
Charles Vincent, Measuring and Monitoring of Safety
The Improvement Guide: A Practical Approach to Enhancing Organizational Performance. Gerald J. Langley (Author), Ronald D. Moen (Author), Kevin M. Nolan (Author), Thomas W. Nolan (Author), Clifford L. Norman (Author), Lloyd P. Provost (Author).
Prof Graham Martin, posted via twitter