 Hello and welcome to today's webinar about data management in the revised national statement on ethical conduct in human research. Apologies for starting a few minutes late. We've just had a few technical issues. So I'm Caitlin May and I'm a senior research data specialist from the Australian Research Data Commons, which is an increased facility that has been built from the Australian National Data Service and NECTA and RDS. For more information on what the Australian Research Data Commons is doing now and into the future, please sign up for our newsletter. So I'm going to pass over to Jeremy Kenner, who's the expert advisor for ethics to the NHMRC's Research Quality and Priorities Branch. Jeremy, please take it away. Thank you Kate and welcome everybody. What I'm going to do here is to run you through a very very brief bit of background to how we got to the current revision of the national statement and what's in it and then drill down to the issues that are most pertinent for you all. As some of you may be aware, the national statement is Australia's singular principle guidance for conduct of ethical research and it is a particular type of document that is a bit different from some of the other documents of a similar type in other countries. One of the things that makes the difference different is that it's not as detailed as some of those documents and it applies to all human research but only to human research. It's based on a principles based approach to two research ethics and it's been around since about in one form or another since about 1992. The current version was based on a full review some years ago and published in 2007 and since then we have reviewed bits and pieces of it from a single word to a paragraph to a full chapter but we've never taken on a full section before one of the five sections which is what we did this time. So this national statement you should have now a slide that says national statement updated 2018. This was done over the last couple of years and published just last month with a fully revised section three with a new title and it was as I say released just last month. It made major changes to the structure and content as indicated based on an elements of research approach rather than a categories of research approach and the guidance in it applies to all research and then becomes more specific as it goes along and I'll explain that in the next series of slides. Chapter 3.1 which is the one that's relevant for this webinar is it relates to research in general and provides guidance related to specific types of research as necessary incorporates some chapters that were previously separate and one of the things that it adds is significantly greater guidance on ethical considerations related to collection use and management of data and information in research. There are a couple other specific chapters one addressing genomic research, one addressing human biospecimens and one addressing animal to human xenotransplantation. Now if you it also there were changes made to one of the other sections that were a consequence of the changes made to section three and those changes included the three items as noted. So this is just a for your later use just mapping out what the changes were to the document. Now in terms of implementation because of the the significant there's a substantial nature of the changes the current national statement which was last updated in 2015 remains in effect during the transition period from now through December. As I've said here users of the national statement are expected to gradually integrate the new and revised guidance into their proposal submissions and review during the transition period which will then end on the first of January at which point the 2018 version will become formally the national statement. So there's a period of I guess some ambiguity where people might be asking which one should we apply but we're hoping the common sense prevails on this point. So in section three which I'm going to run through not I'm not going to run through all section three. Section three relates to the ethical considerations then design development review and conduct of research. Right so there's an introduction to that section which you will read in due course I hope and then there is an introduction to chapter 3.1 which is the elements of research. It then goes through each of those seven elements. If we move ahead then you'll see guidelines and element one and then what we're concerned with is element four. Now the logic here is that when you use the national statement you should be using it as a whole. So section three is read in the context of sections one and two having been previously understood and refers back to concepts in those sections as well as specific paragraphs. When you're using section three everything is in the context of the section three sort of concept or philosophy and then in chapter 3.1 you're referring when you're using any of the elements such as element four it is understood that you should have absorbed what is in the other elements as well. So you're not seeing anything in a vacuum it's all contextualized and that's important to understand with respect to how to use this particularly when it comes to for our purposes issues related to the management of data and information that would be in other chapters which will then refer back to this chapter's discussion of collection use and management of data information. Having said that let's drill down now into some of the things that are maybe somewhat new or maybe differently worded versions of something that was already there. One of the main things to notice here is there is no specific chapter in the national statement now related to data banks which there was it was chapter 3.2 and a lot of the issues related to data were in that chapter but of course data banks is just a component of all of the issues and so it was felt that the issues related to collection use and management of data information related to a much greater set of expressions and and all research not just the ones that use data banks specifically. Now what I've done here is I've highlighted a few provisions in red I don't know if that's visible from the screen that you're using I hope it is but I'm just going to instead of going through the whole thing which is several quite a number of pages we're going to highlight some some some things that have been emphasized that that could be emphasized. The first is of course that that all of this information is contextualized by the ethical principles discussed in section one of the national statement and a second is that the purpose of the national statements ethical guidelines is served by a positive attitude towards the use of population-wide data sets and that and the idea that this promotes the core principle ethical principle of justice which has to do with the sharing fair sharing of benefits and inclusion in in research participation in research. Some other some other things that need to be understood in order to understand what we've done here. Anyone working in this field knows that the term data itself is a bit ambiguous it means different things to different people how it relates to the term information there is no consensus view on that however we've attempted to and and I would credit ARDC here formally A&DS with having a lot of input into this discussion. We have decided to define these terms in a particular way for the purposes of this document which may not be defensible more broadly applied and so we we've we've operating from the premise that data information are often used interchangeably and then we define those terms for the purposes of this document again this is in red for the purposes of national statement data is intended to refer to and then there's a bit of a of a list there of things that that are included and this in understanding the rest of the statements it needs to be understood that those these are the definitions that we're using. Now identifiability of information is a huge issue maybe even 90 percent of the issue that or whatever that we are addressing here and we had a bit of a long debate about whether to use the terms that were in the national statement the terms were personally identified not reidentifiable and nonidentifiable versus the terms used in privacy legislation which were not the same generally they are essentially identified and de-identified. The national statement terms have been used in 2007 intentionally to not reflect the legislative terms which were felt to be inaccurate in the context of research or not as easily applied so the debate ensued about whether which which set of terms to use the ones previously in national statement are the ones that were that were in the privacy legislation and we eventually ended up with a decision which surprised all of us which was to use none of them and we wanted to discuss the issues related to identifiability without getting caught on specific labels and terms and so there's a footnote here that says what what we've decided to do and essentially why and I hope that that's visible to you or that you can read it at another time but we took the position that the identifiability of information was a characteristic that existed on a continuum which was affected by a large number of context of contextual factors and then we start discussing that more and importantly we we tried to point out that identifiability of information changes during the life cycle of a research project and therefore researchers and reviewers have to focus on the risk of harm that arises out of the the identifiability of information issues at various stages of the research project not just at the beginning for example and then we listed some factors that needed to be taken into consideration when determining a degree of identifiability of information and when evaluating the risks that are associated with them we then encourage people to go beyond the guidance included in the national statement to consult the office of information Australian Information Commissioner and its state and territory equivalents the Bureau of Statistics the ANDS now the ARDC in addition to the national statement as there are of course very valuable resources available from those sources that that can be used we of course consulted all of these entities in the development of this of this chapter now we we did talk about the methods that might be used to reduce the identifiability and consequent risks and they're listed here there are several of them in relation to data management we're pushing the idea of development of a data management plan for every research project now this is not a requirement you'll notice the use of the word should rather than must so it's not something that can be imposed as a requirement of the national statement but it's obviously strongly encouraged the word should and must are important in reading the national statement and I won't elaborate on that I think it's sort of obvious what the difference is now one of the things you might notice is that people tend to use different terms when they're talking about collection use management of information and we've tried to use all of them so we've got here saying data management plan should address intentions related to generation collection access use analysis disclosure storage retention disposal sharing and reuse of data and information all of these are in our view related to the management of data and the plan should be developed as importantly as early as possible in the process and then there are some a list of eight things that were considered to be necessary to include in any data management plan now all of these paragraphs I would say of course are important to read but I'm we couldn't go through all of them so I'm trying to highlight a few of them for you for your consideration we have one here it's number 3.1.50 although it's not listed that way on your on the slide and it's the last one related to data management and this one is important for for this discussion that in the absence of justifiable ethical reasons and it lists them and to promote access to the benefits of research researchers should collect and store data information generated by research projects in such a way that they can be used in future research projects where researchers believe they're valid reasons for not making this successful it must be justified so the default is that it should be available for future research and and if not needs to be justified now those justifications may be there may be many justifications that people can employ so we didn't take an absolute position but we obviously took a strong position on this with respect to secondary use of data information which is a section of this chapter I'm highlighting here the idea that it there are other ways of demonstrating respect for individuals other than obtaining their consent so if you can't obtain your consent that isn't the end of the story you need to then consider other ways of showing respect amongst them community consultation ensuring that the research results are translated into improvements and services and practices acknowledging the source of the data or information in publications and or publishing research results in a location and language suitable for the general community data information then we then we try and wrestle with the the whole problem of inter use of information obtained from the internet which as we all know is very challenging and we have quite a bit to say about this there's some highlighted language here that I won't read to you but I'll emphasize that the guiding principle for researchers is that although the data or information may be publicly available this does not automatically mean automatically mean the individuals with whom the data information associated have necessarily granted permission for its use and therefore you need to consider these these things here however we're still saying the consent is the gold standard but if it's not available then a waiver of requirement for consent may be obtainable when again about social media platforms researchers should take account of any terms and conditions applicable to these platforms when using data information from these sources or platforms now understanding of course that most people don't read the small print nevertheless if if there are requirements or conditions in terms in about the use of those particular social media platform those need to be understood and adhered to in conducting research using that that information in terms of sharing of data or information the some requirements here is emphasis on the importance of identifying a custodian and the variety of people that or entities that that could be that the data punishment management plan needs to be shared with that custodian so that they're aware of what is being promised to the participants or advise to them it's important to distinguish in proposals to share a closed research disclose research data or information the distinction between disclosure to specific third parties sharing with other researchers and disclosure to the public and to clarify which you're talking about in consent documents in thinking about all this there are expectations that people will have and this is independent even of the consent that's obtained but it needs to be be under be considered by researchers that there are expectations and policies regarding the sharing or reuse of information that should consider the value of the data and information for future research and that these expectations should be clear to participants the last statement I've highlighted here is that shared or banked data or information that's stored in a form that can identify individuals can sometimes be used in research that qualifies as negligible or low risk research however it cannot be used in research that is exempt from ethics review and that has to do with other clauses in the national statement at five point one point two two and and two three I believe so that's that's the the some of the issues that I wanted to highlight from the the new text the new content of chapter three point one of section three of the national statement there's some take home messages that I've written down for your consideration the first is that we we want it should be clear that the national statement is not obstructive it intends to support data linkage research secondary use of data and data sharing as long as it's in accordance with the basic principles of the national statement and the guidance provided in section three that if that there's no one size fits all approach for that researchers can rely on that would be artificial and that it's all about context and proportionality that is that care needs to be taken in to the extent possible but also in proportion to the the risks associated with what would happen if inappropriate care were taken and that's this is important for hrc's as well to consider not to make overly restrictive requirements for research that bear that that is accompanied by uh minimal risk that appropriate strategies to minimize the risk of unauthorized disclosure by identifying information need to be used that planning ahead makes a lot of this a lot easier if you develop and communicate data management plans after discussions amongst yourselves as researchers and then share those that information with with data custodians and hrc's then it's the importance of drafting contracts and other agreements with with any of the parties involved as as early as possible um to ensure that the use of data is consistent with the expectations of those who provided it the consent obtained if any and or any waiver granted and to consult experts like a rdc as well as relevant review bodies so i'll end it there um and if uh we have some questions which i hope we will i'm happy to address those thank you very much jeremy um so there's a couple of questions that have come through in the question pod the first one is previously the term reidentifiable was used um in the um old statement um where data was coded to researchers but they could still re-identify the patients um the person is asking what term should be used in future to reflect a similar status of the way that data will be handled well it's not so much that reidentifiable um is a is a term that's verboten it's that we we wanted to to to emphasize the idea that identifiability was a bit of a movable feast and that re-identification was more which should be best understood as a process that changes the character of the information rather than an absolute characteristic of the information itself um information can be re-identifiable at one stage of the project but then let's call it non-identifiable at another stage of the project by virtue of it having been de-identified which again is a process we would argue that information isn't de-identified a process of de-identification has been applied to it making it difficult or impossible to um to uh identify um and so we just didn't want to get trapped in the labels the word using the word re-identifiable um data is not a problem uh it's just not something that we wanted to use in the national statement guidance obviously if people understand what it means but we I would give you one example and that is that um there's a lot of argument in the genomics research community for example that there is no such thing as non-identifiable or fully de-identified or permanently de-identified information that if people work hard enough and triangulate enough databases they'll ultimately be able to re-identify information which everyone assumed was fully de-identified and so that becomes problematic do you call it non-identifiable for all practical purposes do you call it re-identifiable because you're being purist in your understanding of what it is um so we run into more problems it sort of creates more problems than itself that's a long-winded answer to a simple question I apologize thanks Jeremy uh so another person has asked is this retrospective for existing studies or prospective from December 2018? Prospective. Fabulous um another person has asked will the HREA be revised by January 2019 to reflect the new 2018 national statement? It will we're working on it now it'll be ready by the time that by by January 1st possibly earlier. Fabulous um we'll keep an eye out for that um so another person has asked um about section 55 to 54 um in the section 3.1 is this guidance only relevant to internet data or information or all publicly available data? Okay let me find what the reference is here so this is um in the so this for some reason the way this printed out it's supposed to say 3.1 point but um the um he's saying is this which which what which paragraphs are we uh is he asking about or is she? 51 to 54 I believe it's the part that was about the social media and internet um data on the internet. Yeah I think the idea is that it should apply um just see what's directly before it um that's interesting I can't seem to go up so give me a second there we go um I don't think it's it's meant to be limited to um internet related data and we actually had a problem with we even had a section originally on internet data but we wanted to um consider it more broadly so I I I'd have to um to look at it again but my instinctive response is that it's intended to apply to all secondary use of data information not just that obtained from the internet if we don't say in the paragraph with respect to data obtained from the internet then we we mean it more broadly. Excellent. There are there are some questions we've been sorry there are some questions we've been asked about what constitutes publicly available data for example um if you've got if you've got information that's in a newspaper article that wasn't collected um in the same way as we consider data collected but if you want to use information from a newspaper article or a book published in 1856 do you consider that publicly available data or not and so we there are there is a little bit of ambiguity here and again um we're hoping that mostly people are using a common sense approach. Excellent. Jeremy I'll just ask you one more quick question there's a couple more that have come in but we'll address them afterwards in a question and answer document that will be made available to everyone um this last question here is the option for granting a waiver for consent currently exists and leads to different and inconsistent decisions by ethics committees will the revised guidelines help to reduce this inconsistency? Well inconsistency of application unfortunately isn't really within the scope of the NHMRC's authority to resolve the fact that things are that provisions of the national statement are applied inconsistently can have to do with um you know poorly written provisions or it can have to do with different um philosophies and attitudes that that that reviewing bodies have um we can't arbitrate that and say this one's right and this one's wrong we're not authorized to do that um all we can do is reiterate our interpretation of our own documents which we often do in the form of responses to queries that people give us. I'd have to know more about the nature of the inconsistency in order to comment more intelligently on that question. Okay thank you so much Jeremy for your time today. You're happy to write in and ask um and the uh the last slide here is uh some contact information so um again you're welcome we'll respond. Excellent so thank you very much Jeremy for your time today and I'd just like to remind all of the people who have joined our webinar that ARDC has a lot of resources around data management and some of the topics that Jeremy touched on such as de-identification and data management plans. Our resources will be updated to include the new requirements from the national statement on ethical conduct in human research. Currently they are all still sitting on the ANNs website so if you go to ANNs that's a and d s dot org dot au forward slash medical that's a great page to start on to um find all of the resources that are related to these topics that Jeremy has been touching on today so thank you Jeremy for speaking to us. Thank you everyone.