This presentation will share our experiences in Victoria, AU reducing stillbirths by working with hospitals, providers and consumers and how we adapted the UK bundles and made them practical and easy to use. We will describe an innovation project underway in the US to reduce stillbirths.
Dr. Jane Taylor works as an Improvement Advisor and Learning Designer. She enjoys supporting learning collaboratives, networks, and innovation projects by providing expertise in improvement approaches and measurement. Dr. Taylor has advised over 100 collaboratives for both adult and pediatric populations and have worked with many states on reducing infant mortality, maternal health, medical home, primary and specialty care, reducing readmissions; to name a few. She focuses on building quality improvement capability and improving healthcare around the world. In addition, Dr. Taylor has made a personal commitment to developing an approach to improvement and innovation that makes space for equity, inclusion and diversity. Her leadership experience as a hospital CEO and COO for over ten years and as an improvement consultant to healthcare for over 20 years has prepared her well for these efforts.
Dr. Taylor has disclosed that she does not have any real or perceived conflicts of interest in making this presentation.
Jillian Phillips: Dr. Jane Taylor works as an improvement advisor and learning designer. She enjoys supporting learning collaboratives, networks, and innovation projects by providing expertise in improvement approaches and measurement. Dr. Taylor has advised over 100 collaboratives for both adult and pediatric populations and has worked with many states on reducing infant mortality, maternal health, medical home, primary and specialty care, and reducing readmissions, just to name a few.
She focuses on building quality improvement capability and improving health care around the world. In addition, Dr. Taylor has made a personal commitment to developing an approach to improvement and innovation that makes space for equity, inclusion, and diversity. Her leadership experience as a hospital CEO and COO for over 10 years, and as an improvement consultant to healthcare for over 20 years has prepared her well for these efforts.
Dr. Jane Taylor: Well, thank you all for joining us today. I’m delighted to be able to present with my colleagues from Safer Care Victoria in Australia on a joint project that we did with the Institute for Healthcare Improvement to reduce avoidable stillbirths. We’re going to share approach we did, and then at the close of the presentation, I’ll talk about another exciting project in the US. I’m joined today with Felicity Czarnecki, who is the Collaborative Director in Victoria, Rebecca Reed, who is the Improvement Advisor, and I, Jane Taylor, who is also an Improvement Advisor on the project.
Our objectives is that at the end of this presentation, you’ll be able to state three or more interventions that can reduce avoidable stillbirths, appreciate the value of provider sharing lessons learned and collaborative learning and changing care practices. Also, you’ll be able to describe foundational elements of a Breakthrough Series Learning Collaborative model, and you’ll be able to describe a current innovation project underway in the US to reduce avoidable stillbirths.
Our agenda is we’re going to talk about our aim and why we did the work, how we use the Breakthrough Series Learning Collaborative model developed by IHI, how many teams we had and who was on the teams. Then we want to share with you our theory of change, what we worked on. The family of measures that we used, and our fantastic results. Then talk a little bit about the pandemic interruption and our restart, and then I’ll close talking about the Ariadne Labs project with Star Legacy. I’m going to turn this over to Felicity to talk about why we did the work.
Felicity Czarnecki: Thanks much, Jane. Back in 2018, there was a report from Australia Select Senate Committee inquiry into stillbirth research and education. This report recommended that the Australian government lead the process nationally to reduce the rate of stillbirth by 20% over the coming three years. At that time in Victoria, our stillbirth rate was 6.2 per thousand births after 20 weeks, and 2.2 per thousand births at 28 weeks or more. The bigger issue there was that this rate of stillbirth had not changed for two decades, and research was telling us that they believed many of these stillbirths could be prevented.
Obviously, we have a sector who care deeply about the harm experienced by women and their families when they experience stillbirth, and we had an opportunity in Victoria to do this significant piece of work. As we set off down the path of the Breakthrough Series Collaborative, we came up with this glorious aim that by July, 2020, we would reduce the rate of avoidable stillbirths at 28 weeks or more by 30% in participating maternity services. We did have an example set for us on the international stage in the UK with the Saving Babies’ Lives Care Bundle pointed to what could be possible in this space.
I just wanted to flag this little cutout, this newspaper article on the screen. This came out in Victoria in one of our state newspapers on the day we commenced this work in June, 2019, and it was such an energizing moment for our teams to know that they were part of a national effort and that their work was being recognized. That really set us off on a great path. I’m going to hand over to Rebecca now to talk a bit more about that Breakthrough Series Collaborative structure.
Rebecca Reed: Thanks, Felicity. A Breakthrough Series Collaborative, as Jane highlighted as part of our introduction, is an execution strategy that was designed by the Institute for Healthcare Improvement, and designed as a specific methodology to support spread and scale of known interventions that will make a difference to the experience and outcome that you’re trying to shift.
Some really important, I suppose, components to call out as a part of a Breakthrough Series Collaborative, that is the essential nature of bringing subject matter experts together to create a shared understanding of what the system is saying. A shared understanding of what consumers who are experiencing the care of the area that you’re focusing on, how they understand and see the system in order to create a shared understanding and a vision of a way forward.
You’ll see that there’s some black arrows on the left-hand side of the screen. A big portion of this work is spending a great deal of time, pulling that information together and constructing a theory of how you’re going to make an impact in the system that you’re focusing on. Jane will look through that in a little bit more with you shortly.
As we move through to the right, you’ll see there’s some circles and boxes and squares. This is where the action takes place as a part of our time together. Most collaboratives operate between 12 to 18 months. Ours has been a little bit disrupted, obviously, because of the pandemic that’s occurred and Felicity will speak to us a little bit more in detail, but I guess some of the important things to understand about a Breakthrough Series Collaborative is, where we see these boxes with Ls1, LS2, LS3, they’re key moments in times where teams come together, where that word collaboration becomes really important, to share what they’re learning, to share their results, and to shamelessly steal from each other the efforts that they’re doing to create the improvement that they’re wanting to see in relation to the piece of work.
During those boxes, you’ll see there’s AP1, AP2, AP3, these are action periods where teams are actually out doing Plan-Do-Study-Act cycles to actually introduce the changes and contextualize those, so they’re actually creating sustainable improvements in their systems in relation to the processes and outcomes we’re trying to shift.
For us here in Victoria, whilst there’s the standard collaborative model, we were starting at a base and working with clinicians that were really new to doing improvement science. Our action period calls, whilst in the standard model there is coaching through emails, visits, phone conferences, and teams are doing reporting, et cetera, one of the key things that we did differently as a part of ours and was absolutely significant because of the low capability and starting place for our teams in relation to improvement science was actually the relationship building that Felicity and I did with going out and actually visiting every single one of our health services to create that trust and understanding that we’re going to work through this process together.
Absolutely, what we’ve seen throughout the course of this collaborative is an increase in confidence. That sounds like it might not mean much, but we’ve had individual clinicians telling us that, “This felt really confusing and really hard, and I didn’t quite see the benefit of the hard work that doing Plan-Do-Study-Act cycles with rigor would actually mean,” but by the end and halfway through action period 2, we’ve had teams saying, “It’s given me a new language, I’m influencing leaders. I’m able to bring champions on board to actually continue to support our work.”
I think that’s perhaps sometimes something we overlook as a part of the benefits of doing a Breakthrough Series Collaborative, but the capability it builds in your people, the confidence it builds in your people to actually then pick up this information and methodology to use it beyond this piece of work is absolutely phenomenal. It’s well worth a read. Go to the IHI website and look for the white paper on Breakthrough Series Collaborative if you’re not very familiar with it. Thanks, Jane. I’ll go to the next slide and hand over to you.
Dr. Taylor: Thanks. The model that we used with the teams that were involved in this project was the model for improvement. The model for improvement has three questions. What are we trying to accomplish? Well, we’re going to reduce avoidable stillbirths by 30%. How will we know a change is an improvement? Rebecca’s going to talk about our family of measures in a moment. What changes can we make that will actually result in improvement? Taking those changes, customizing them for the unique settings of each health service and have them do that by using this quasi scientific method called Plan-Do-Study-Act. Where did these teams come from? Tell us about that, Felicity.
Felicity: Thanks, Jane. In the top left-hand corner of the screen, you can see the map of Australia and that little red portion shows you where Victoria is, which is where we’re calling in from today. Then in the middle of the screen is a much bigger picture of Victoria, our lovely state and all these red dots that you can see indicate roughly where our teams were located. We had the great privilege of working with 19 of Victoria’s maternity services, a mix of public and private across 23 hospital sites.
Now, these represented at the time about 50% of the births that were happening in Victoria on any given year. You can see clustered around the center of the map of Victoria, that’s where Melbourne is. That’s where our metropolitan services are. These tend to be larger services with higher birth rates, greater complexity of women coming for their pregnancy and birth, and usually different models of care operating within any particular services.
As you get further out away from the city, you get into our regional and rural health services where the birth numbers tend to be a little lower, the complexity is a little lower, and the number of models of care at any given hospital are fewer. That certainly doesn’t mean there are fewer challenges. Our teams, big and small, metropolitan and regional and rural, certainly, face different challenges during the time of this work.
The teams were multidisciplinary, predominantly midwives and predominantly led by midwives, but working closely with their obstetric colleagues, whether they were staff specialists or visiting obstetricians or GP obstetricians. Quite a few of our teams also worked with quality improvement people from within their service as well.
As Rebecca noted, we had the good fortune to be able to go and visit all our teams as well. The furthest one for Melbourne where we are is probably about four and a half, five hours away. We did a lot of driving around the state to meet with our teams and build those relationships and to build our own understanding of the context in which they were working in as well so we could learn as we went along how best to support them to grow their improvement in clinical capabilities.
The other thing I would note here is that we were fortunate to work with some key partners and agencies during the work. One of those being Quit Victoria, who’s a lead smoking cessation agency. With the Stillbirth Center of Research Excellence, a national research body aimed at reducing stillbirth, they are delivering work in this space at a national level, and with the Stillbirth and Neonatal Death Society, Sands, who have now combined with Red Nose. They work with families and services in the stillbirth space.
Dr. Taylor: Thank you, Felicity.
Felicity: Welcome. I will hand back to Jane to talk us through that theory of change.
Dr. Taylor: Thanks. Any improvement project is grounded in what it’s going to take to bring about improvement, and we call this a theory of change. This is actually a living document. This went through three major revisions over the life of our project. Starts with the aim on the left-hand side, and then it has primary drivers, which are the structures or the norms or the key processes that have to change in order to bring about improvement. We had partnering with women and we really ended up with the application of evidence-based clinical care and partnering with women.
The secondary drivers are discrete moments in time. When did we need to partner with women? At booking, at pregnancy care appointments, when planning for the timing of birth. Or it could also be places where this happens during the care appointment, or finally, it could be key steps in the process that really need to be highly reliable and functioning well to get results.
A driver diagram is what is necessary and sufficient to get results. One of our original driver diagrams had leadership on it, but it turned out that it wasn’t so much a driver as we worked to distribute the leadership and thread it throughout the work so that the empowerment was felt at all levels of the organization. Then I’ll point you to the change ideas, which are the specific evidence-based changes that need to happen, say, at booking or at the pregnancy appointments.
Basically, it comes down to five key focus areas. Smoking cessation, recognizing and responding to changes in fetal movement, and rapid response, I would add. Recognition and management of fetal growth restriction, maternal sleep position, and then finally, we added this, promoting appropriate timing of birth, and we added engaging women in birth timing decision-making. We’ll be talking about that a little bit more in a moment, but this was our key theory of change, and you can see that in the change ideas, it’s mapped back to the UK bundles for preventing avoidable stillbirth. Rebecca, tell us how we would know if a change was an improvement. How would we learn if we were improving in this work?
Rebecca: Excellent. Thanks, Jane. With any improvement effort, we create what’s known as a family of measures. Really what that’s designed to do is to create a system view of the area that you’re working in in order to understand what actions you’re taking actually leading to the improvement that you want to see. Within a family of measures, there are three key components. Firstly, outcome measures, which relate to what we’re trying to make a difference in. Then process measures, which are the key steps or actions that we know that we need to make reliable and consistent in order to shift that outcome in the direction that we’re wanting it to move.
Finally, the last component is balancing measures, and they refer to the outer mechanisms of your system. They’re components that perhaps you want to keep an eye on that because we know we’re working in one portion to address the outcome that we’re trying to see, we just want to make sure that we’re not causing disruption anywhere else. Sometimes some common language might be an unintended consequence. Sometimes we think about those as being a negative unintended consequence, but sometimes you can have balancing measures also that are looking out for positive unintended consequences as a result of your work.
Creating a family of measures starts at the very start of that Breakthrough Series Collaborative when we bring those subject matter experts together. In doing this work, knowing that there’s a high degree of evidence around those care bundles across the world really in terms of where we should be targeting, actually distilling that down to the type of measures that you want to be able to capture was a really challenging process for teams. Particularly for our teams, this was the first time that they had started conceptualizing measures from a learning perspective, as opposed to learning for judgment, learning for performance.
Creating a suite of measures that felt representative of the work that they needed to do, captured everything that they wanted to capture and then place the lens over it that I want to look at it continuously so I can learn and respond and alter the actions that we’re doing on a day-to-day basis as a part of our improvement effort takes a lot of coaching with teams, particularly when they’re new in this space.
Just to highlight what our outcome measures were, we had an outcome of rate of stillbirths at 28 or more weeks gestation. We also included the ceasing smoking between conception and birth as an outcome measure, knowing that the reduction in smoking was probably one of the biggest interventions in terms of reducing preventable stillbirths.
Our key process measures included the intervention of ask, advice, help, which is a conversation that aids smoking cessation in women. Process measures that focused around fetal growth restriction and monitoring. Provision of education, key education for women around decreased fetal movements and maternal sleep position, as well as capturing women being involved as much as they would like in terms of decision-making about timing of birth. Our balancing measures focused on the percentage of women who were receiving inductions of labor or caesareans prior to 39 weeks, and the number of babies admitted to special care nursery.
I’ll invite Felicity and Jane to pop in and add any bits of information, but I think for teams endeavoring to do this work, this measurement process took a lot of refining. When we started, it was a really big challenge for us to get teams to condense the suite of measures that they want to be continuously capturing information and reporting on and learning from, from about 20 measures down to 12. That was a slow process for us to influence the teams, but it wasn’t until they actually realized the benefit of having specific information as opposed to too much information to be able to make sense of. I think that was a real learning curve for our teams to select measures that you could actually learn from.
Where did that take us and what did that look like and how do we continue to build that learning? We used these charts, and I’ll start here, but our process measures in terms of that day-to-day learning is where that occurred, and I’ll dive into that in a little bit. Excitingly, our teams made a difference. This process makes a difference.
The chart on the left there is our overall aggregate for our entire collaborative. Both those regional rural centers and metropolitan centers. Whilst the chart demonstrates that we haven’t created a new level of normal, when we compare our baseline period to our active collaborative period, we would have anticipated having 79 stillbirths during the same phase of time where the collaborative’s been running. Instead we’ve had 54 stillbirths, which is just absolutely amazing and accounts for a 32% reduction in stillbirths.
Excitingly, we can see that we’re on the verge of making and creating that new level of performance in our system. Excitingly, for our regional teams, they have actually already created a new level of normal. Moving their rate of stillbirth from 5.4 to 1.4. It’s just our largest services that are going to tip us over the edge and make that overall next great big jump, which is just fabulous. Thanks, Jane.
I’ll just take you through some of our process measures. This is the outcome measure for anyone who might be new to improvement. Outcome measures are the last measure that you will start to see moving. It is really important that you have process measures that not only capture the key things that you want to make reliable and consistent within your system, but this is where your teams will get momentum. This is where they’ll get engagement, because they’ll start to see these measures move more quickly in relation to the effort that they’re putting in as a part of their Plan-Do-Study-Act cycles of change.
These two measures focusing around ceasing smoking and the ask, advise, help intervention. You can see when you look at those charts that our teams have made some significant improvement really early on and then have predominantly held that improvement. If you look to the far right, you can see that, particularly for the ask part of intervention, which is the bottom chart, you can start to see we’re on the verge of a new rise as well.
Felicity we’ll sort of talk through some of the impacts that occurred as a result of COVID, but you’ll see, as we start to move through some of these chats, that for some of them, interventions have held even throughout the pandemic period, which is just amazing a tribute to really the rigor teams have placed in doing their Plan-Do-Study-Act cycles. The other areas perhaps the change wasn’t able to be held maybe for a number of reasons, some of which were that teams may have not got through the amount of rigor that they could have with the Plan-Do-Study-Act cycles, but also because of the complete turnaround teams needed to do with their service provision. Felicity, we’ll talk a little bit more about that.
These next two measures focus on fetal growth, either measuring and plotting and/or screening. Again, you can see that our teams made some really early gains in relation to these measures and have predominantly held that improvement, which is really exciting. You can see when moving from sometimes really low numbers to quite high percentages of women who previously didn’t receive these interventions, but were certainly now more reliably and consistently receiving these interventions. Thanks, Jane.
These next two measures focus on key education that’s important for women in terms of being able to have some self-determination in their care and advocating for in relation to how they feel their pregnancy is progressing, particularly with decreased fetal movements. This felt like a really easy place to start. The provision of decreased fetal movement material and education.
We also had included really early on as a part of our phase one activities, the time it was taking from the time a woman would contact a service to report decreased fetal movements to the time she was arriving and receiving appropriate scans in relation to that. The benefit of collecting data over time and teams using it as a learning rather than a judgment was that our teams were able to really quickly say, “Hey, we’re actually far exceeding what the state expectation is in relation to this timeframe.”
That’s actually not our issue. The issue isn’t about women receiving the scan once they’ve recognized the decreased fetal movement. It’s actually the woman contacting the service because she’s not responding to the decreased fetal movement or change. That actually sparked a dearth, I suppose, in a real depth of learning across our teams in terms of engaging with women to understand what that was, what’s the culture and norm that was sitting around and driving that behavior, not only in terms of a woman picking up the phone, but the response she got when she called the service in relation to that.
We saw an amazing amount of work in relation to education for clinicians around their response when a woman calls as opposed to automatically assuming there’s just anxiety there and nothing in playing it down to actually actively responding and actively encouraging women, “Even if you go home in your home for 15 minutes, even though we’ve just checked, please ring us back if you’re still feeling nervous because it’s our role to care for you.” That’s been amazing and a number of our larger services have certainly seen some changes in relation to the reduction of time between women first feeling or noticing some decreased fetal movements to actually activating it.
I think before we flip, sorry, the last thing that I think is really important in this slide is just thinking about, I suppose, the subgroups that sit within this. A number of our health services have real rich multicultural demographics, and actually what diving deeply into this, particularly into the decreased fetal movement space, it actually has helped teams recognize there’s particular cultural norms that they need to address and particular cultural barriers that exist that they need to address in order to support that population of women. Thinking about what are the subgroups in relation to each of these measures has been a really important part of this work. Thanks. We’ll go to the next one.
The final process measure, just to share with you, was the one around shared decision-making. Felicity we’ll probably dive into this one a little bit more when she talks about our pandemic impact, but you can certainly see that this chart is quite different in comparison to our others. Again, we had that initial early rise, but this measure became about measurement as opposed to the intervention we were trying to impact. This was a really key learning for us as a collaborative, around creating real clarity of the operational definition and the why behind capturing this measure.
I think really early on this measure became about collecting the measure and whether or not a woman was asked if she was involved in decision-making, as opposed to actually, what we’re really wanting to know is the answer of whether or not she was involved in the impact, whether or not she was involved in that shared decision-making head.
I think we saw that initial rise because people were just ticking, “Yes, we’ve asked the question.” During a pandemic period, services were disrupted, and you can see that that dropped, but I think that’s probably more relating to documentation, and you can see in the latter half of this chart that it started to pick up. I’m going to leave that with Felicity to talk to because actually that became the focus of really the second portion of our collaborative. Thanks.
Felicity: Thank you. Unsurprisingly, for the past year, work was a little bit disrupted due to the COVID-19 pandemic. In Victoria, though we had quite low case numbers compared to other places in the world, we had some of the tightest restrictions both on individual movement, but also some of, I suppose, the most significant disruptions in terms of what was happening in our health services, we could start to see this in February, March of 2020.
In terms of where we were at in our collaborative timeline, we were deep into our second action period and in the thick of planning for our third learning session. The work was due to end in June. We were feeling like we were running towards the finish line really strongly. It became clear though at the start of March that with everything that was happening in Victoria, as we moved into a very long period of lockdown and our health services had to introduce significant changes to the way they delivered care that we actually had to take our foot off the accelerator and pause the work to take the pressure off our teams.
What that looks like from our end was that we reached out to them all and said, “We’re going to hit pause. We’re going to remove any expectation of you reporting to us, submitting data to us, or doing any active work that you don’t feel you have capacity to do at the moment. We’re still here,” we said to them, “and we’re very much happy to connect with you if you want to, if you want to talk through your work, if you want to share anything with us, we will still be here.” We know that some work continued in some services, and in others, they really had to completely down tools as their service was disrupted.
In a period of lockdown, I think for just over a hundred days as a community, and certainly the interruption, the health services rippled out on either side of that quite dramatically.
As we emerged though from that lockdown period towards the end of 2020, it looked like we had a nice opportunity to pick up and to actually finish off the work definitively and not just let it fizzle out and bid everyone farewell.
We designed a six-month end to the program that will be conducted virtually. Teams would have the opportunity to connect as a whole cohort through milestone virtual forums, and they would have the opportunity to connect frequently with us, either for one-to-one coaching or for small group coaching, where we would bring services together who had either similar challenges or similar locations or sizes to talk through their work and have access to coaching from our team.
Of our 19 health services, 16 of them returned and were able to have the resources to commit to that six-month period, which was terrific. We continued work towards that shared aim. We’re really delighted to say that teams were keen to take advantage to connect with one another again both to debrief a bit of that pandemic experience, but also to re-energize their work and really learn from each other again. It was pretty lovely to see when everyone started to come back together.
The other thing that was really striking and I wouldn’t recommend having a pandemic in the midst of collaborative, but one opportunity that it offered was that teams could reflect as they started to backfill some of their data, they had a really clear view of what changes stuck through that period of disruption and what had kind of fizzled away. That was a great learning for them around the rigor of their improvement methodology. It really sparked some great maturity in their improvement, knowledge and skill, and they started to work during this second virtual phase more strongly in some of those more challenging areas like shared decision-making. It was a great period of learning for them and for us, and we are wrapping up that work this week with them. Jane, I think you’re going to tell us now a bit more about Ariadne Labs.
Dr. Taylor: Yes, and thank you for that and congratulations on our work. It’s just fantastic. I wanted to share very briefly of an emerging project that is being done with Ariadne Labs and the Star Legacy Foundation. The Ariadne Labs is a joint center for health systems innovation at Brigham and Women’s Hospital in Boston and the Harvard T.H. Chan School of Public Health.
Ariadne Labs intends to develop simple, scalable solutions that dramatically improve the delivery of healthcare at critical moments to save lives and reduce suffering. They’ve turned their attention now to avoidable stillbirths. Their vision is that every health system delivers the best possible care for every patient everywhere every time. How do they do that? They work anywhere in the space or the continuum between prototyping to wide-scale dissemination around the world, and they call it an arc where they design, test, and implement solutions.
So far, what we’ve done is extensive literature rereview and evidence exploration, in-depth interviews with those women who have a lived experience of a stillbirth, in-depth interviews with clinicians involved for caring for women who’ve experienced a stillbirth, and then others who have been involved in previous improvement efforts like Felicity and Rebecca around reducing avoidable stillbirths. Then that is synthesized, and recommendations on scale up and a broader implementation plan will be created. Standby, we’re right in the early days of this work.
Thank you all very much. Have a great rest of your conference. And thank you, Felicity and Rebecca.
This presentation was part of the Stillbirth Summit 2021. This individual lecture will be awarded .5 hours of continuing education credit to include viewing the lecture and completing evaluation and post-test. Once received a certificate will be emailed to the address you provide in the post-test. If you did not register for the Summit WITH Continuing Education, you can purchase the continuing education by clicking here. This purchase will provide you access to all Stillbirth Summit 2021 lectures including continuing education credit. There is no charge for viewing the presentation.
Please feel free to ask questions of the presenter. We will obtain their answers/comments and provide them here as received.