 So, we might get started. So, I'd like to welcome everybody to today's information session on the Asada Infrastructure Development Workstream. To begin today's meeting, I'd like to acknowledge and celebrate the first Australians on whose traditional lands we meet, and I pay my respect to the Elders past and present, and I'm speaking to you all from the city of Canberra in Gunnawal Country. So, today's session is going to be a presentation followed by moderated Q&A. Please have questions to the chat as we go along to open the chat, there's a button in the bottom right. ARDC staff will be monitoring the chat channel and collecting the questions for the Q&A session after the presentation. As I said, chat window you should be able to open via a button on the bottom right of your screen, and also you might want to switch to speak of you. There's a layout button on the top right of your screen, that's personal preference, of course. So, the purpose of today's webinar is to discuss an infrastructure development opportunity being offered by ARDC, but as you'll see, the infrastructure will be developed to support clinical trials research data sharing in the first instance. So, a lot of discussion today will be pitched to infrastructure providers, but we welcome the researchers who are attending today and will be addressing the research context and opportunities along the way. So, to get us started and to give an introduction to who ARDC is, for those of you who don't know and what the Hissander Initiative is, I'm going to hand over to Adrian Burton. Hi everyone, I'm Adrian Burton, I work at the ARDC. Yeah, ARDC is part of the NCRIS network, NCRIS stands for the National Collaborative Research Infrastructure Strategy, the logo there on the screen, it's National Research Infrastructure for Australia, comes through the Department of Education and it's to build, as it says there on the screen, nationally significant assets, facilities and services to support leading edge research and think of it in terms of large scale facilities that wouldn't be possible by any one research institution or jurisdiction or something like that, that we, there's part of the Commonwealth Government's investment in research, is this National Research Infrastructure Program, there's a number of domain specific facilities that, you know, build or invest in telescopes or plant or mice breeding facilities to bring that scale up to the national scale. ARDC is slightly different, it's not necessarily a domain or discipline specific national facility, but we deal with the cross cutting challenges of data and data analytics, so high quality data collections and the analytic platforms that allow you to get value from that as well as underpinning infrastructure at the national scale for storage and compute. We are a cross cutting facility, as I said, and so this particular initiative, the Cassandra initiative, is a partnership between the ARDC and the Australian Health Research Community. And when we say a partnership, it's ARDC is putting significant resources from our own expertise and underpinning infrastructure and co-investment, you know, money, but the key thing is the strategic intent to build something at the national scale, so back to the the whole enquire introduced in Chris, is that our role here is to bring together a national partnership to underpin data sharing in the health research community. So, it is a partnership, you know, we are here to catalyse that and to move it along, but we do believe it's a community, it's an initiative that belongs to the whole research health research community. And why are we doing it through the consultation of probably 18 months before this stage? We have heard very strongly from this community that health data sharing and secondary use of the data that comes from health research projects such as clinical trials, cohort studies and registries, for example, that they can bring a lot of value, the reuse of that, the availability of that data for reuse. We bring value to the research community, we'd have broader increases in the impact of that research, make sense from the in health research investment point of view to build new research from our previous assets. And we've really focused this programme on saying, well, how could that secondary use have real translational health benefits. So, that's the objective. We're, you know, and the key things there are we're looking at something via a national infrastructure, ARVC is partnering with the whole community. And the idea is to unlock the value of the data outputs of health research for further reuse. I'll hand back to you, Kristen, just to take us through then the particular HASANDA programme and wherever. Okay, thank you, Adrian. So, as Adrian was saying, the priority for ARDC and for HASANDA is on enabling research impact and translation. And as he alluded to prior to the formal launch of HASANDA, ARDC held preliminary discussions with peak national health research organisations with a focus on these areas. These discussions identified that opportunity for ARDC to facilitate data sharing from clinical trials, clinical quality registries, cohort studies amongst other research areas to improve the impact and translation of their research. As part of the formal launch of the HASANDA initiative in late 2019, ARDC invited these organisations to form an advisory committee for HASANDA and provide feedback and advice on ARDC's strategy and direction for the programme. Based on those discussions with our advisory committee, ARDC selected investigator-initiated clinical trials as the first primary focus for HASANDA's infrastructure and as a proof of concept of what we're trying to do. This particular research area meets our criteria for supporting research translation. It's also one of the more mature health research data communities with respect to its data standards, its methodologies, its community coherence and organisation. So, in effect, what that means is there's a shorter path between their current data sharing capability and what HASANDA will deliver. So, they provide a good candidate for a proof of concept. So, ARDC identified three key priorities and work streams for the HASANDA programme. First work stream I'll mention is the infrastructure stream. It's the reason that we're here today and it's the stream under which we'll start building the national infrastructure to support the sharing and secondary use of clinical trials data. The key elements of this stream are the design, testing and deployment stages of the infrastructure. But as I'll explain shortly, it will be informed and supported by the other two work streams and ARDC will be coordinating the activity between the streams. The second stream I'll mention is data development. So, data development is a process for establishing the business requirements, the information needs and specifications of a national data asset. It's a process that's been developed by the Australian Institute for Health and Welfare for establishing a data asset. So, AIHW is a government agency with over three decades experience working with health data and statistics to inform and support health policy and service delivery. ARDC is going to be separating out this process into three stages. The first stage were consultations held last year and were an initial broad requirements gathering process. The second stage will be a focused design process and those of you who are invited into the infrastructure stream will be involved in the design process within this data development. It's essential for the success of Sander. So, we're currently initiating focused workshops with some of these groups and the feedback we receive will add to the requirements we already gathered from that initial broad community consultation conducted under the data development work stream. The culture and policies will also engage in work packages around ethics and consent and policies to enable the Sander's infrastructure and ongoing governance. So, the diagram on the screen is intended mainly as a communication tool to express our priorities, our resourcing and our high level approach to the Sander. It's not a detailed work plan and there is intersection across a bit between the streams as I've mentioned. I'll be presenting a work plan later but a point I want to make here is that Sander is a nationally coordinated initiative in which stakeholders will be working collaboratively and not competitively. I'd also note and health researchers will be aware that in other regions like the US with NIH and the UK's Medical Research Council there has been significant progress in their national approach to data sharing and infrastructure to support their researchers and we would hope that the Sander program provides a way forward for the Australian health research sector. So, we're up to at the moment of what we've completed so far. So, prior to the launch of the infrastructure development activities later this year, ARDC will be completing a requirements gathering process. An initial set of requirements was identified via that development stream last year as I said which involved an open consultation with approximately 110 days from the clinical trials research community. The aim of that was to establish the research purpose uses and value proposition for the national data asset and the Sander infrastructure and also to test the community support and establish buy-in for Sander. AIHW were consultants for this process. They helped guide the design and help facilitate the workshops. The report itself was drafted by an external editorial team who contributed throughout the workshops and analyzed the feedback that we received. This report is published on our website and was referenced in the information for today's webinar. So, the consultations established a number of key things. First one being the research use for national data asset and you can see on the screen what the three primary research uses were, systematic reviews, secondary analysis and so on. It was identified that these in turn can facilitate things like policy development, new study design, health technology assessment, clinical guideline development. The consultations also identified the information needs for the data asset. Overwhelmingly, researchers want access to participant data or IPD and that's not a surprise for a lot of people here, but they also require study protocols to contextualize the participant data and data descriptions or dictionaries and quality statements so that the IPD can be used effectively and efficiently. Attendees didn't indicate a priority for the standardization of IPD itself, but they did integrate a need for a standard approach to cataloging the various data sets and documents produced by clinical trials. The reason for this is that cataloging these research outputs from trials allows them to be easily discovered by researchers who want to reuse the data as well as improving the planning of new research and the efficiency of managing access to data. Now, the number one feasibility concern that was identified was unsurprisingly around the ethics and consent requirements for data sharing and some attendees were also concerned about the additional resources that may be required to enable data sharing within their research groups. But attendees actually saw a nationally coordinated approach like ASANDA as an enabler of change and an opportunity to address these common issues that they face. Indeed, the overall community response to the ASANDA initiative was strikingly positive and researchers were incentivized by the prospect of increased research collaboration opportunities and impact that data sharing provides. The feedback we received also validated our vision that our approach must be collaborative and must coordinate between the key stakeholder groups in clinical trials research. So our wonderful editorial team distilled all this feedback into a set of principles and recommendations to guide ASANDA's infrastructure development and with their endorsement and with AIHW's support ARDC identified three key areas for investment and the design of the infrastructure. The first area is the development of coherent data practices. That's the name that we've given up. These will be a set of community defined expectations or best practice guidelines for data sharing, information standards for data sharing, data access practices, and ethics and consent practices. The consultations didn't identify any existing set of practices that researchers wanted to see mandated in Australia and there actually didn't appear to be any really dominant standards currently across the sector. The workshop attendees were keen for a set of national standards or approaches being developed that could give them guidance in these areas. However, based on our current information, we think it's unlikely that a standard one size fits all way of doing things will emerge, at least initially, but something like a framework to guide researchers and give them options to choose from seems practical and of value at this stage. So for example, the UK data service has a framework for data access which involves a three-tiered model. Incidentally, UK data service are the people that came up with FiveSafe, so I'm sure a lot of people have heard of that. So the three-tiered model defines the three levels of authorizing access to data either by making it open and freely available, which probably isn't going to be possible with clinical trials data. The second level of having mediated access in which the data custodian must grant permission for someone to access their data. Or a third most restricted level of controlled access in which people can access data but with the custodian's permission and only within a secure research environment approved or analytical environment approved by the data custodian. So the reason I bring this up is that there are practices out there which we could replicate or use as a basis for an Australian approach, but the key outcome of this part of the Cassandra infrastructure is to build consensus on a set of practices that we can adopt as a national standard. So the second major element of our infrastructure is what we call in coordinated data services and it's this element that is the focus of the current call and is what we're here to discuss with you today. The purpose of this part of the architecture is to establish a distributed network of infrastructure nodes embedded within institutions or potentially clusters of institutions who agree to work together to share a node. The nodes will supply the research data catalogues and outputs of clinical trials run at their respective locations into the national Cassandra data asset. By making their research catalogues visible to the network, institutional nodes can enable that data discovery research planning and so on that I mentioned before. We also expect nodes to adopt the coherent data practices from the last slide for providing access to their data, but nodes will actually be involved in the design of those practices and it's worth emphasizing for those out there who are concerned about this that the research data itself will stay with the custodian in the institution system or node and would only leave that node if access is granted by the node operator or the data custodian. So each node will hold its own data and it will do so using its own data infrastructure and systems. So with the standard ARDC is not looking to build a new single repository that we want to hold everyone's data for them. Instead what we want to build what we want to do is build capacity into the existing infrastructure at institutions so that they can meet a national standard that they themselves will help design. A simple example of how this might work is that each node makes use of its existing repositories and data catalogs that develops an API that can allow communication with the Hassander network. What Hassander will do is help the research community and institutions define what the information and functional requirements of that national network should be. The other key part of establishing the node network is the creation of a governance mechanism to oversee its ongoing management. ARDC sees itself as a catalyst and coordinator of this national infrastructure development. We expect that the nodes will play a key role in the ongoing governance of the wider network once it is established. So the final element of the design is Federation services. So Federation services are not part of the current open call that we're discussing today but as you can see they complete the design of the infrastructure and they explain the importance of the node network. So the Federation services will be the applications and interfaces that allow researchers to interact with the node network and that will enable research and data discovery access, secondary use so on. These services will rely on the foundation that that node network provides and a simple analogy, excuse me for this, is that the Federation services are to Hassander what Travago or booking.com are to hotels. So Travago is a simple interface that allows a person to see what kind of rooms are available at a wide range of hotels. You can look around at what hotels and rooms are available but if you want to access that room then you need to follow a specific process to get access to that room. Similarly a Federation service in Hassander might let you see what research documents or data are being held at different nodes or institutions but if you want to access those data or documents then there will be a clearly defined process you need to follow. The other thing to consider with Federation services is they don't just need to be data catalogs and request systems, they can be as advanced as your imagination allows but they will rely on the foundation of the node network. Okay so that was an overview of the progress we've made so far and the directions we're going with this infrastructure design. So now we're going to focus in on how we will go about establishing the node network and the purpose of this webinar is to open the call for research institutions and organizations to establish themselves as a node in the Hassander network and to be part of those nationally coordinated data services. So starting in March we'll be asking institutions to register their interests to become a node in the Hassander network. To be a node you will need to be able to find three key things and I referred to these before but they're worth repeating as they're the key characteristics. So firstly nodes will need to be the owners or the custodians of clinical trials data or they will need to be able to provide the node service to data custodians and they will also need to share metadata about their trials via the network. So in this context because metadata means different things to different people but in this context it means the descriptions of the data and documents produced by the trials. So in first instance you won't need to provide the actual files themselves initially that description. The formats that the documents and data need to be shared in will be determined during the initial design phase for the node network and again nodes will provide input into those designs. Nodes must also be willing to respond to data requests to access their data but there is no unconditional requirement that all requests must be improved must be approved. Instead we would expect nodes to commit to a framework for determining whether data requests should be approved and again the nodes will contribute the design of that data access framework. The second thing that nodes will contribute nodes will provide their own data infrastructure so Hassander won't be mandating specific IT systems or software that nodes must use but we will facilitate the establishment of that national standard for the information and functional requirements of an institution's systems and what those systems must meet so that that will enable them to join the node network. Nodes will then be responsible for implementing that national standard within their own operational environments and the third element finally is once the network is established we expect nodes to have ongoing input into the governance and oversight of the network but as I said before ARDC will support this by working with the nodes to facilitate the design and roll out of the governance model so those were the business functions that nodes will provide. Now if establishing a node at your institution and being part of this national network is something you'd like to do or interested in then as a node operator we'd be expecting you to contribute to Sander to the program in these three ways firstly node operators are expected to contribute to most aspects of Hassander's design this is mutually advantageous as nodes will be key operators in the infrastructure so your input in the design will ensure that Hassander is feasible but node operators are also going to want their priorities and requirements factored into the designs Secondly once those designs are complete then nodes are expected to implement and embed them within their operational environments. ARDC is facilitating the requirements gathering and design stages but it will be up to node operators how to adapt their infrastructure to meet those designs and requirements and the third element final key element is that we want to see these infrastructure designs embedded within the nodes that they will become a normal part of business's usual and can be sustained by the node operators so our approach to establishing the nodes will essentially follow a design development deployment approach but before I take you through the indicative work plan for that I want to just pause for a moment to emphasize the ethos that ARDC is bringing to Hassander so ARDC sees Hassander as a co-creation project amongst the key stakeholder groups in health and clinical trials research so the trialists the researchers the organizations the participants health consumers and key funders and policymakers like NHMRC ARDC's role is to coordinate these stakeholder groups and facilitate the consensus building and design processes as well as investing in the development and deployment process once the initial set of nodes has been identified they will provide key input into the design of the infrastructure as I said but then assume primary responsibility for implementing those design requirements at their respective locations so this is a nationally coordinated and collaborative approach that involves more groups than just the nodes but node operators will have agency throughout this process so our work plan reflects that ethos and will be carried out via four main stages as discussed requirements gathering started last year and will continue to the middle of this year and are being conducted coordinated by ARDC and its contracted partners for this stage so we've already had the requirements of the broader research community as I discussed earlier and over the next few months we'll be completing focus consultations with clinical trials networks and research participants to add to and refine to the principles and requirements that we've already obtained I don't want to preempt the outcomes of these consultations but I'd expect that they will refine some of the details around government governance issues like ethics and consent data access and issues like research acknowledgement and data custodians rights and responsibilities ultimately the outputs of this requirements gathering phase will inform the next stage which is the node design phase which will run through the second half of this year ARDC will again coordinate this phase but the node design will be decided with the input of the node operators and other groups who can provide expert guidance on issues like data access ethics informatics and so on and the output of this process will be the information functional and data access specifications for the nodes so following that we'll move into the node development phase which will run for approximately 12 months at this point node operators will take those design specifications from the previous phase and build them into their local operational environments ARDC again will provide support by coordinating these activities across the nodes and by providing investment into the nodes the work packages here will be the fairly typical dev test deploy scenario and ARDC will confirm the successful deployment of the node by assessing that it meets the specifications for federation we'll also require node operators to document the nodes design and operation specifications for posterity and ongoing sustainability the final phase in this work plan is a six month test and deployment of at least one federation service and the goal of that is to ensure that the nodes operate successfully as a network ARDC will run the federation service development as a separate activity to this node establishment but will coordinate with the node development activities also during this final phase we plan to have the initial governance policies and procedures for the node network put in place and that will come from the establishment of a governance structure that the nodes will contribute to as I mentioned so that's the overview but before I hand back to Adrian I just want to emphasize again that the nodes are key contributors to design aspects and have agency throughout the development process but you won't be going into line and ARDC will provide coordination and support and there will be additional expertise and input provided by other key stakeholders and groups so that's that in a whirlwind I'll hand over to Adrian who's going to take us through the ROI process copy napping Christian I'm back um thanks for that Christian uh so Christian was talking about the timeline there that you know we're um continuing the consultation process and then we want to start the node design more formally by the middle of the year by June July so between now and then we're looking for to establish you know who those nodes will be so we've got a little process over the next three months that will allow people to register in interest for us to then be able to facilitate between some of those expressions of interest and then finally to invite proposals proposals to become nodes we should we hope to have that all confirmed by the middle of year and that will allow us to kick off that node design phase that Kristen uh walked us through so that's the uh sort of general idea there a rigid registration of interest is exactly what it says it should be fairly lightweight and but enough for us to say uh oh well perhaps there's clusters developing here that we could facilitate conversations between uh also ideas of how the coverage might go it's an uh it's a point where you can say well here's what we understood and you can discuss again with the ARDC to say um you know to clarify your understanding of what um being a node means um so that's the then so that's the registration of interest and then facilitation at that point yes we asked people to specifically submit a proposal Kristen alluded to the fact that it's not meant to be a competitive process um so facilitation and uh is an important part of this um we've already been through and continuing this uh community consultation about what people want and there's a design there of how you know that can be delivered through this node model so this is not the kind of I don't know like an ARC thing where you're competing on a new or shiny idea or a different thing this is more about the capacity to respond to this model and and to be part of that collaborative design process to be able to contribute to that and to be one of a collaborative set of nodes so that's more the kind of thing we're looking here for rather than a something you know a competition on some new ideas or something like that um I think we'll go on to the next slide some of the things that we'd be looking at and these are indicative and we'll clarify these with you with some of the feedback during that facilitation process and certainly before the official submissions but if you're looking at an individual node criteria um you know what kind of research data would be likely to come through this node what are the established relationships with particular research programs or institutions that would mean that this node really has a good um a good chance of being a really rich source of clinical trials data that could be used for guidelines and secondary analysis etc we do if you look through the background information that's on our website which I encourage you to do there is this idea of nominated trials each of the nodes will be asking them to say well apart from the generic relationship that you might have with I know anything from these three institutions you know that general idea of what data what research might be covered we will be asking you to nominate some specific trials that you know are underway and that could that by the end of this development process we'll be using them as the test of the infrastructure that yet we can deliver these particular trials through the through the new node and we'll be yes they'll be part of that testing of that final phase of testing to say yep we have an infrastructure and we're actually sharing a number of nominated trials through those so that's that will be another component of the the coverage there and then of course yeah the future research activity that would come through there we'd be looking at you know being satisfied that you know that the data is not only of high value to for research and and research translation but also that it it's feasible you know that actually the it's data that is can be fair you know findable accessible is interoperable and reusable and that there is a kind of a track record of having data that can is shareable another track record thing we'd be looking from from the individual knows would be have they ever operated any infrastructure before or do they have relationships with institutions that have you know enterprise IT that that can partner with them to really deliver a research infrastructure rather than a research project you know IT you know this is really stepping up to a new level of delivery and of course we're looking for this that collaborative spirit and the ability and people who can contribute to the node design phase that Kristen was talking about and then of course we'll be looking you know are the institutions and the alliances involved are they likely to be long-term players in the health research system and therefore have a compelling confidence that after the establishing project that we have ways of embedding this back into business as usual so that's you know how you know the kinds of things we'd be looking for from a good note and then there's a portfolio that we'll also have to from the ARDC point of view look at say okay if we have and we are at the moment running with a rule of thumb of around 10 nodes so from the number of initial nodes that we establish is there a story that says that this is a you know a good significant first step we don't we know that you know this is you know a 10-year plan as far as we're concerned this is the first two or three years of the establishment but we do want to establish a network of nodes that has you know the best possible coverage of the different kinds of research the different kinds of disease the different types of institutions a good coverage of the major institutions that are involved in health research that there is coverage of the jurisdictions other you know and other important things for the program like different communities being involved and that idea that we brought up that the whole purpose of this is for translational research so do we have all the right partners there that are heading towards new guidelines and meta-analysis in different areas so there's that that is the portfolio criteria that we have to look at across the different nodes as well but again hopefully you'll you'll see that these are the criteria that just reflect the objectives of the whole initiative and that's what we're using so then just to come back to what the next steps for you are if you've listened to this and you're interested the key next step is this registration of interest between March the 1st and the 31st we will then be taking those registration of interest through to the facilitation phase so you need to get yourself into that into the pipeline during that registration of interest phase and then as I said we'll be working through any of the connections and coverage questions during the facilitation phase and then getting some proposal submissions in May I think we'll pause there and that's all we've got to say as far as the information sharing part of it is where I've been for questions you did ask for some questions during the as you all registered so we've got a few of those to go through we go through those first and then we also have some that have come through on chat if you have questions feel free to put those in chat and if we have time that's we cover them now and if we don't have time we will answer these questions in a frequently asked questions kind of page and as I mentioned at the top we had some colleagues I wish there's been monitoring the chat and it's going to be moderating this session so yeah raise over to you so we've got a few questions first beforehand in the chat around the scope perhaps we can deal with those first so for example there's been talk about clinical trials is there going to be support for other kinds of studies and in particular sienna sienna goals asked about things like observational studies clinical registries other kinds of things so perhaps some comment about the scope of his sander sure so we're starting with clinical trials as a proof of concept because obviously anything in in health there's a line in the sand we ARDC has longer term plans to scale up the sander model to incorporate those other research areas and maybe Adrienne you have better sense of the timeline for that so yes the all the stuff we talked to you about today is a network that is really focused on that first phase of getting coverage of clinical trial outputs health studies is definitely the full scope for her sander in fact the hd at the beginning of her sander's health studies so cohort studies and clinical quality registries and related things are in scope for us and we want to keep that in mind for example when we're thinking about how the nodes the portfolio of different nodes were established as far as doing that we hope that during the timeline that you just saw we would start to do some of the background consultation with the cohort studies community and the clinical registries communities in Australia and by the time we get to the infrastructure deployment stage here that we'd be ready to do a second phase with the broader of studies data in scope thanks Adrienne and I think a sort of a related question is somebody's asked about international programs so what happens when for example the work that's going on with the trial data here is part of an international setting how does that affect the sander's work and the design of the nodes and the design of the project it's a good question the international landscape is also emerging as you know as we develop here in Australia the very first thing to say is that by having a coordinated national infrastructure approach in Australia we should be able to come to the table to some of these larger international initiatives and we will particularly what's the word explicitly reach out to the European and US models that are emerging there as far as shared data outputs from international trials that is usually part of the setup of the trial itself but we again I would hope that by having good quality capability and capacity in Australia we would actually be a more attractive partner to some of those international initiatives okay and I just once just come in from from Tim also is it we're talking about things that have finished completed trials which obviously takes a while to finish all we're talking about having the the sander initiative to include things are in progress is there any sense of that from the consultations we focused in on the shareable the objective there is to an infrastructure that supports sharing and secondary use of health data health research data so that would depend if there's outputs that are shareable as interim parts of the project then yes they should be able to fit into this program okay I think there I think there'll be a balance there it do come up during the consultations the research community said that going back and if modifications needed to be made to data and files and documents from completed trials to make them available in her sander then there's obviously a gap there that they'd need to close and that is much easier to fit into a sander if they know from the outset of a trial that that is going to be one of the outcomes of the project so so that was a comment in the consultations but as Adrian said part of our evaluation criteria for nodes will be whether they're able to nominate trials that could be included into sander from around the time of launch and we understand that there is a mix of status there between trials the completion say the readiness for publishing and so on and so I think part of the facilitation phase is to have those discussions with potential nodes to understand the situations and and how feasible it would be for them to to make the data available okay thanks um there may be some around scope in a minute I might come back to those are finding more but perhaps a little bit about the the nodes themselves so a few people asked about this one is around that's come into the forehand is the costs so what kind of node costs would be part of AIDC's investment you know is it around equipment and licenses staff costs other kind of costs what's the sort of current thinking well as far as these projects are concerned you know that they are they are the node establishment projects and so the co-investment from AIDC is around establishing the functionality of the node um so the kind of normal things that you'd expect there are the the staff um the obviously the staff and development costs there there are some you know licensing and other um independent infrastructure things that are required as part of the development of the node um we'd be happy to get more feedback from people during the facilitation phase as to the kinds of activities that kind of costs that are related to the establishment of these nodes yep thank you um I've got here if someone is else is hosting a world standard secure data hub would it seem to be willing to deploy its data holdings into another hub and it was a similar question which was around how will data be distributed to those who are for example not nodes I think it's a related question but it's basically um I guess that's about our data holdings becoming available to others well uh it's really no question if someone else is hosting a world standard secure data hub if you're the person who did this you feel free to clarify in the chat whilst we're answering but I'll answer what I think the question is asking um if someone is if there's a world standard hub for doing this redeployed that through sander uh as Kristen pointed out the ethos here is uh not for our DC to mandate any software or even governance approaches at each of the nodes um but for you to for for the nodes to be able to dovetail this back into the systems and processes that you have uh at those nodes so in in principle we are just looking for the outcome and the functions and if you have a world standard um way of delivering this we're super happy to have that um so Jadrin you see Alison's question I thought I naively thought that uh that saying the sort of things that they would invest in would answer Alison's question that she's asking can what's the incentive to be a node um I presume she means financial incentive but there could be others uh and now I'm a bit lost as to which question we're answering so Alison's question was what's what's what is the incentive is what is there any incentive for an organisation to become a known um well you've seen what the uh the kind of mission and objectives of the sander initiative is so the very first thing was we'd be looking at uh institutions who perhaps for different reasons from the ARDC but for the same you know societal good reasons that they are looking to what's the word make sure that there is a sorry my phone just keeps ringing and I can't stop it for room we'd be looking for institutions that share the same objective of unlocking the value from research data of the outputs of these health studies the research data in that area um so yes we would be looking at people who are institutions that have an aligned mission with us that would be the number one incentive um to be part of the national leadership in how data sharing happens in Australia would be you know perhaps another incentive to be able to be on the forefront of getting access you know if you're part of this node network then you will be right there and the members of your institution of your cluster will be right there getting access to the outputs of other research um so yes we're we're looking for organisations who actually have an ethos and would therefore contribute to the initiative if there are other incentives we are providing incentives to establish the nodes as far as um establishment money is concerned and we will be also looking for uh longer term operational um support for this within the research uh system as well so I've been a related question to that it's after the project period so after 2023 is the institutional node expected to be financially and operationally sustainable um after that that was one of the questions that came in beforehand yeah so look continued uh first thing ARDC will be continued to be part to participate in this as I said we're from our point of view the first two three years are the establishment phase and we fully from our point of view to be committed for you know a 10-year period to realise the benefits of this so from our point of view yes we will be uh continue to be involved after the projects um we uh I said that that is one of the criteria that we'll be looking at is is there a compelling you know um opportunity that the node will continue after the uh project period as far as sustainability and the models are concerned um we will that will be again part of the working group amongst the node operators to look at the different sustainability models um and we are you know the spirit we would bring to this is this idea that it's the health if this is the health research community and we're building a cooperative infrastructure where everyone's getting a benefit and that it is part of national infrastructure so we'd be wanting to have the um the easiest possible access for the end users that's our spirit to begin with and we will be convening a sustainability working group across the uh nodes to look at the different sustainability models that could be applied um after the project but it will be something we'll look at that the nodes are up from the beginning so if it just looks like a fly-by-night you know one person uh application to become a node then that will not compare um favourably to uh something that comes in that says look we are a cluster of research institutions that have been working for 100 years in health research and we uh want to be part of the uh ongoing uh solution here actually Adrienne I'm going to jump in on that because I saw a question come through chat around how do you define research institutions um and I don't have chat open in front of me because I'm on screen share I can't remember full extent of that but I think it's probably worth addressing the question if um or the issue around clustering I'm not sure if we um if we spent much time on that during the presentation. The question was uh you know what do we mean by institution could it be partnerships between several institutions universities emporis for example that's the question. Yes absolutely and remember I said we were going off a rule of thumb of around 10 initial nodes um we think that that then drives us towards a model of uh clusters either around um subject matter area or regional or jurisdictional or you know um clusters of organizations that already work together in some way uh we would be looking that that there is the enterprise there is a solidity behind this that there there is that there are long lived institutions behind the cluster and behind the node but we do think that um there are possibly at least 100 you know just research organizations doing trials in Australia uh and we don't think that a model that that starts off with you know 100 different solutions is a good role we're looking for something a simplification for the end user and so we are encouraging our institutions to work together in uh clusters. A question which my answer is sort of relates to that is somebody asked if if you're not a member of a node um the federation the federated process later is that restricted to those who are only nodes or will it be available to a wider group? I think I know the answer to that is yes it will be even the spirit of this is we are building national research infrastructure to support research um you know to actually remove the barriers to access to data and increase research across Australia um without preempting the legitimate requirement you know that the legitimate requirements of the researchers and others that we'll be taking into account during those stakeholder engagement things that um Kristen talked about earlier so uh that you know access is definitely something that that you know needs to be a nice balance and you know the node operators obviously will be there leading the initiative but the in principle answer to your question is yes this is open infrastructure for to increase research all the way across Australia there will be access uh restraints um you know based on the legitimate needs of the patient groups and the researchers themselves. And Adrienne to follow on which I know is not the order of questions received Tori's asked once established is the exchange of the data from a node to another to other parties going to be commercialized or is the idea here that it's a purely of you know a free exchange of information which I think I'm putting that one up now because I think it follows on from what you just did. Yes and again remember I started the whole thing with the establishment you know the interest was part of national research infrastructure so the spirit of this is to you know build an infrastructure that will increase the amount of data re-use in Australia and so we'd be trying to at all at all options to reduce the friction on reuse. Okay um we have a little bit longer so a couple of questions about the design of the time frame some of the technical aspects of the process so we've talked about there being a time frame we're just going to find the one obviously the consultation process to this day just decided on the time frame the question was how did they ever how did we arrive at a time frame that was sort of three to four years rather than a longer time frame because there seems to be a lot to do it was that did the time frame that you've decided for the current step come out of the consultation process or the person asking the question was concerned of course but about readiness do we know if this if that's realistic is that feasible or is the time frame being how we arrived at the time frame for the project? It's a very good question um it's in discussion just it has come out of the consultation and it's a balance between the urgency of people to actually establish an infrastructure we're looking at it as a feasible time to remember what we're saying these are node establishment projects so we think it's a feasible time to establish a set of nodes and the infrastructure to deliver some data I do agree with the premise of the question that potentially the readiness of data and the culture and and potentially going back over all data sets if necessary that could be a much longer term initiative but yes this is both on the the feasible bit about actually establishing the underpinning infrastructure for us. Yes and there have also been a couple of questions which sort of impinge on workflow about preparedness of data quality to be accepted so will they you know how do we make sure the information coming from the different nodes is homogeneous or at least consistent meaning a minimal data quality standard and therefore will be acceptable if you like how do how will we do that? Sure it's just about to cross to Christian today but he's been handed an important piece of paper there um yeah sorry yep uh timing was a little awkward but question was around the information standards was it? Well it's it's there've been a couple of questions paraphrasing them in a bit of ask the question about the difficulties of data quality in the first case making sure that there's high data quality and you know garbage in garbage out of course and I think that there is a second one which I'm going to ask now which is about automation obviously there's a lot of movement of data is that going to have to be manual is it going to be automated how will we deal with that with the quality issue and the automation of the movement of the data? So I think the issues around data quality and standards and as well as metadata quality and standards will be part of that design no design process that we're talking about and that hopefully we'll soon confirm that AHW will be leading that process for us um so yes there will be standards introduced there that the goal there of course number one objective has to be to meet the research reuse requirements but it also has to be feasible and practical for it to to proceed the second question was on automation I wasn't exactly sure what aspects we're talking about automation for hang on let me see why we saw that question so I'm happy to jump in there oh thanks yeah now we're talking about the federation model and you know delivering descriptive descriptions of the of the data sets to a federated sort of discovery portal if you like um and could that be metadata could that be automated automated yes we fully support that and that would be one of the functional requirements of a node to be able to you know supply for the technical point of view the equivalent of an API that could be created to pull that information which would have lots of advantages it cements we're not manually packaging stuff up and sending it through email and we always want this system to be to have a currency so that you know as soon as a node does ingest new data to be shared then the description of that data isn't immediately available in any discovery systems that are working over the time and we've done this before this is a kind of thing that ARDC does all the time and we have there a number of automation models there we would be hoping to uh what's the word leverage off some of the international standards potentially the ones that are in in clay in the health and medical area but there are a number of IT standards for that kind of metadata exchange okay now there are some questions that are coming in that are quite complicated in terms of trying to answer it i don't know christian if you're watching these as well what um the question Steve McEachins asks is the federated model here want a federated data or a federated metadata or both i think the answer that is both we're starting we're definitely starting with federated metadata yes and part of the reason for these nodes is that there are different custodian requirements related to each of the nodes and sometimes they're different state legislations or just different you know requirements from partnerships with health services etc and that's part of the idea of starting up with the the node model is to be able to at the data level the data you know access etc but that is handled separately by each of the nodes and i think i'm happy to cross to christian on this but certainly the message that we got at the consultation phase is that there was a bit that absolute harmonization of the actual data content and the data structures is a phase two let's call that you know that the you know that the and that was in recognition of the fact that clinical trials can be about a lot of different things and a lot of different data types can possibly be represented as the output of a clinical trial we want to be as inclusive as we can as far as the data reuse opportunities are concerned and so we wouldn't i wouldn't start with a an absolute data sort of homogenized harmonization program that's not where we'll start but we do hope that by building a community of researchers of research infrastructure that are working together on this over 10 years that we would also improve that kind of data standards across the whole christian anything yeah uh yeah definitely so i think yeah on that last point around ipd standardization no that wasn't definitely not identified in the consultations as a requirement of the research community and those of you who work in health research know as adrian mentioned it is just such a far reach for where we currently are and you know potentially could be argued in some cases that it's not it's not really the approach of exploratory science and research the other points that i'd make in it because i guess people are asking they're trying to get out you know like a grand design here and a goal that is um uh you know that could see is highly desirable everything is automated everything is standardized um everyone does everything it's everything the exact same way um but how many steps it's going to take to get there and i think aides is being very realistic about this and the response that we had from the consultations was very realistic about this so well we can't you know do 100 steps in one step but we can take the first key steps and the key things that we heard back the business requirements or priorities for researchers was that for people wanting to wanting to see what research was out there and what data was out there i could get access to that data that they're having having significant problems and significant inefficiencies uh in in doing that because they'd have to go around effectively they were sending emails to every single research group that they could you know find on on some pretty rudimentary searching and having to get information that way and that's clearly inefficient and it was actually um preventing a lot of uh meta-analysis and translational stuff being done um so so researchers were quite keen on uh the idea that if there was just a single place that they could go to to at least see what trials were out there and what kind of data they were collecting that would be uh one massive step forward from where we currently are so that's one key goal from the uh data custodian or the trialess point of view um and what might be a motivation um to engage in something like Cassandra uh i just make the point and i think everyone's aware that data sharing um and requirements around that have really ramped up over uh the last few years and there was lots of policy signals um in Australia that look like in the future they might translate into policy requirements um that data sharing will be mandated in um uh research that's funded by by those key health and medical funders um even if if the trials are only getting part of its funding from there there will be some level of requirement um and so uh with that and with the with the other data sharing policies that already out there um say by publishers um i think most health researchers would have experienced that the the data sharing options we have available uh to meet those requirements aren't great um and something where they could have input into how data sharing works in Australia uh and more control uh which they're required to have often in in ethics approval and other compliance more control over who has access to their data rather than just having to hand it over to some third party of repository is also a highly desirable thing yeah um there's a couple of questions come in Adrian about the relationship between this and other things that are similar and are often interested by ARDC for example there's Cardray there's Erica um even the imaging um project so the perhaps a comment about how this relates to other ARDC investments and platform projects for example sure for those new idiots that don't know the ARDC uh this is uh a specific a different kind of program the ARDC we're choosing you know with the community this community uh a real challenge and we're just working along with the community to to build something up we also have other programs previously where we've run you know an article say what's an interesting idea to do with the platform and we've and through that process we actually have uh some partnerships in developing some of these secure e-research platforms like uh Sirlep and Erica and the Cardray project which is there around the policies for this kind of access we fully intend now to pull those you know the outputs of those projects into this um sort of strategic push uh to say well how could some of those uh developments that be made in in for example secure e-research uh platforms uh how could some of the nodes you know where uh applicable and appropriate how could some of the learning from those projects be implemented into this node network um so we will do that um what's the word proactively okay as sort of a related question is um asking how this meta data collection relates to the Australian clinical trials registry which i know is not an ARDC fund the thing but it is a if you like a related data set or related project so that was the clinical trials registry clinical trials registry yes yes um is uh highly related to the ANZ CTR the clinical trials registry it's meant to actually complement the registry um the registry is set up for a particular purpose around integrity and the fact that you know that a trial is happening and prospective registration and there's whole sets of policies for that this is meant to just complement that with the um access to the data that comes from a clinical trial we've been working very closely with the ANZ CTR they're on that advisory committee that uh Kristen um uh outlined at the beginning so they've been part of the driving force and the design for this um on some more specific things uh during actually the consultation phase people said i'd you be a pity if someone had to write in all the description of the the trial you know into the registry and then they have to go and type something in in another place around similar information about the data set so we've taken that on as a kind of functional requirement that we should be able to pull information from master data sources like the clinical trials registry to inform the descriptive data that will support the data sets and standard um so yes it's highly highly related and relevant to the ANZ CTR I think also that what the registry does and you know we've selected clinical trials is our starting point of proof concept um something like a trials registry which not a lot of other areas in health search have uh is a one uh a single point for those that aren't familiar with it that actually you know is is a massive data set of what research projects are being conducted and some key information about that so um I think and that information is very useful uh an essential for secondary use of of trials data so I think they can actually act as a model for how um of the kinds of information information model that Cassandra might require in the future for other research types on the clinical trials yep um now I mean there are a few other questions but I think some of them might actually be answered along the way and some I'm taking the liberty of saying a really some detailed stuff that will be answered in the really the process of the consultation some of them are quite detailed Kristen is there something in there that you think we've missed substantial that we've missed these obviously the questions that are answered now as we approach the end will be answered in a in a separate document um there was one other area perhaps um was about how the there's been a couple of people ask how the nodes what's the success factors I mean how will you know if it's successful perhaps a some comment if there's been some given to some thought to um going forward what will we see as a successful node looking like would that be uh yeah that's I feel like that's very much an Adrian I think with apologies you've answered some of that but are there any other comments you wanted to make about how how we would evaluate the success of a node um again our overall objective here is to get clinical trial clinical trial data reused in further research applied in um guidelines health you know emerging health guidelines so that's the the overall principle that we'll be looking at um how how will this node help us to get to to that point um and yes I think I went over some of those towards the end of the presentation anyways some of the actual stuff there so we don't have anything further to add there on the details are there's a couple of things there around standards and yes in OMOP and um and I think Kristen has has indicated that in the first instance we are looking for absolute in a shared set of descriptive metadata for all these data assets that are around the place so that they can be discovered findable and then accessible um we will probably look at the interoperable and reusable uh you know as a second phase uh and obviously uh identifying key common standards and I note that Dougie Boyle's on the call here today we are working with Dougie in a number of other projects in promoting the OMOP model um so yes we fully intend to um leverage that work here as well I'm not sure if you want to add anything Dougie but I'd like to just say that it's a healthy relationship when we intend to leverage it yeah I noticed there were a few things about standards I mean the broad answer to that is of course we're not trying to reinvent something and we're not going to ignore that the vast quality of good quality work that's been going on um and Steve's asked perhaps he's asked who knows as the knows the discussion but can you advise Brithy on the likely process for the standards development work program no Steve you're not allowed to take us off track you can say that to Steve I should point out to people that I know Steve so that's right um I guess the broader issue there is there is a substantial body of work to do in the standard program overall and we are trying to get this infrastructure piece underway which is what today is really about and it's been a helpful to sort of focus on that we're not ignoring for a minute there's other several other work streams which and someone's talked about the human infrastructure aspect and that sort of cultural aspect is absolutely important we do appreciate that we'll have that going as if you like a parallel stream as Christian Scherger in that in that diagram there are several things going at once but in the first instance we want to get this one underway get the work packages underway yeah thanks Rhys that's a that's a really good response I did see that question come through I don't want to dismiss Steve's question um but Adrian did you want to adjust that now or or separately I'm just going through some of the other questions there before we get there will more directly will AADC be co-funding the co-development of node infrastructure yes so yeah providing cash cash for that and you know and as well as the facilitation for that co-design to happen and of course you know we're looking to to work with organizations that are also committed to that goal is there any concept of an affiliate membership to a node look yes that was the idea of a node actually being a cluster of you know like-minded new institutions that maybe come together on these or region or jurisdiction or some other coverage point of view so one would be actually be encouraging and that's what I partly what I mentioned in the facilitation phase that we'll be looking to say look these people are want to be involved they don't necessarily want to host it let's put them in touch with another node emerging node there that they can participate in yeah I'm actually jumping because there were there've been a few questions that we've kind of adjusted but the theme is around the readiness of clinical trials data the ability to share data whether it's past data whether it's future data there was a question that came in prior about you know if the data is co-owned say by a commercial entity or and I'm not sure if we've addressed that fully or directly but I think we understand the knowledge that there are existing constraints and arrangements in place for a lot of data and around the intellectual property and the custodianship of that data and where it's not feasible there won't be a requirement to bring in the lawyers and break the contracts around that data for example but what we want to do is as part of facilitation phase at least is to understand what data you might potentially be able to nominate for Cassandra and it might be around then you might have some ready to go there might be others that it looks like you know you could give the name of the trial and that's it but we want to understand what that is and understand the pathways and hopefully use those can also identify potential pathways for that data to be made available via this infrastructure and I mean this is a quicker side I have over a decades experience doing this prior to this role in ARDC and getting data out of cohort studies into a shareable space and I think for people that haven't tried it before there's a lot of unknowns it feels like a lot of work and maybe there's an initial step in understanding what the approach is and engaging with ethics and understanding what the appropriate approvals or waivers sometimes from ethics committees are but there often are pathways and so that's the discussion that we can have with people interested in this. There was a related question which I think was about fair principles and is there a problem between fair principles competing with the principles or a responsible research conduct and the answer is no there's not fair is a set of aspirations which as Adrian has explained you work towards as an incorrect if I'm wrong got your advice but if some institutions are not at the point where they consider to be as fair as others that doesn't matter we it's as open as possible but as as controlled as it needs to be and I think that's come across a couple of times so a few people have expressed concern around that and in practice as Kristen's alluding to if you work through that with the custodians in a particular situation of that trial you can work through it but as the principle ARDC is working towards fair as a if you like the aspirational set of principles for all of the data sets we have in the end of our time we are come let me 10 seconds as somebody there's a few questions about that that scope of clinical trials and other really valuable health data remember how should you use health studies and we are looking you know the bigger picture here is to again build up into these nodes the capacity to share a lot of different types of data it's just from an IP project you don't start with a number you know too many scenarios you know we build something up but with the idea of that it can and will support other types of health studies data and there's nothing stopping even in this first phase the individual nodes experimenting with you know if they've got a fairly generic secure environment for facilitating data sharing there is nothing at all stopping them and we would encourage it trialling it with the other other data plan with really odd on 230 I am just hope I think I might just close the meeting by saying thank you very much for your participation the ARDC is lives off its community relationships with the research communities we've had some really very very good input here and the chat we will keep that very carefully and make sure that every issue that has been raised here is identified and and at least you know put into the further consultations that are happening again just as a parting action if you are interested contact the ARDC at this address here to get further information about it and then make sure that you get that deadline of the 31st of March to register your interest yeah and I'll just add into that that once the ROI is open and there's a form there to complete that we will be contacting everyone who's registered for this event just to notify them thanks everyone very much it's been a very lively you know question and answer session thank you very much for your commitment and we look forward to working with you all into the future yep thank you everyone