Note: Where available, the PDF/Word icon below is provided to view the complete and fully formatted document
Senate Select Committee on Health
11/12/2015
Health policy, administration and expenditure

BROOKE, Ms Fiona Joy, Policy Adviser, National Rural Health Alliance

GREGORY, Mr Gordon, Chief Executive Officer, National Rural Health Alliance

PHILLIPS, Mr Andrew Robert, Policy Adviser, National Rural Health Alliance

PINSKIER, Dr Nathan Pinskier, Chair, Expert Committee—e-Health and Practice Systems, Royal Australian College of General Practitioners

STARLING, Dr Robert, Chief Information Officer, National Aboriginal Community Controlled Health Organisation

CHAIR: The committee will resume. I welcome everyone to our fourth roundtable for today. Thank you all for taking time to come and talk with us. We really appreciate you being here to share your perspectives with us. Even though you were going to appear separately, we are very pleased that we are able to have an extended period of time for discussion. I invite each of you to make an opening statement, perhaps starting with Dr Starling.

Dr Starling : Thank you for the invitation to attend. We note the committee's focus for this section is on improving the collection linkage and use of health data held by Commonwealth entities, and related issues.

Just briefly, NACCHO's member organisations now make up the most comprehensive, largest network providing comprehensive primary health care to Aboriginal and Torres Strait Islander people. There are a number of statistics that get bandied around. One is: 'Well, you only serve 33 or 37 per cent of the population.' That is true, but recently we have been doing studies on the catchment areas—that is, those areas in which Aboriginal people have access to an Aboriginal medical service—and the numbers of regular patients often exceed the estimated population within what we call a 60-minute drive time, which is the criterion that was used by AIHW. Obviously, data is what you make of it, but the role of the NACCHO member network is significant. We have a lot of expertise, then, within the organisation and, when you think about the fact that a lot of the services are over 40 years old, I think it does not get the right place at the table in terms of delivering its evidence and its expertise into a lot of the decision-making processes.

The Productivity Commission report on NIRA highlighted the need for consistent, timely information that is coordinated across the different programs, provider contracts, governments and organisations. I think consistency in data and timeliness of data are really crucial, and one of the issues we had that we raised with the committee in October was timely access to data so that it can be used in the planning process. In the recently released Public sector data management report, the one that came out at the beginning of the month, PM&C highlighted the lack of access to data to inform policy, budget and funding allocations and effectiveness within government. But I think that also applies to those who are closely related to service delivery—having access to that data in a timely fashion.

We are spending a lot of time trying to collate information across these different data sources. There is fragmentation of funding from Health to PM&C and to DSS now with its allocation also through PHNs. So the data from the reports from these projects is increasingly fragmented. There is a Productivity Commission report which said that there are over 917 current Indigenous health-specific programs. So there are 917 programs and 207 mainstream health programs with Aboriginal and Torres Strait Islander components. There is one current program for every 600 or so Indigenous people in the country. When we look at all of that, the fragmentation of the data and the reporting which is around that, we feel that that needs rationalisation and a clear data information strategy—and that will only be effective if the data can be used. We want to be one of the very heavy users of that data. Thank you.

CHAIR: Wonderful. Dr Pinskier, if I can jump you and continue with the rural theme and then come back to you. Ms Brooke.

Ms Brooke : The National Rural Health Alliance welcomes this opportunity to appear before the committee, and this time with a focus on issues relating to data and evidence. Evidence is critically important for the health sector, and the source of real evidence is good data appropriate for a specific purpose and properly analysed. Healthcare practitioners rely on accurate data to diagnose and treat. Health payment systems require data to enable them to pay for eligible encounters and to apportion payment. Health service managers and analysts use data to determine whether people are able to access the right care, in the right place, at the right time. But, too often, data is the last thing considered when new services are developed. This lack of planning results in datasets that may not adequately address the key evaluative issues or simply are not used. For example, data has been collected on the allied health workforce for a decade but is only now being examined.

Given Australia's egalitarian and universal service aspirations, one type of evidence that is important is that which compares health outcomes or access to services between different population groups. For the alliance, evidence of disparities between people in different parts of the nation is of particular importance. Usually such comparisons are between group averages, for people in one jurisdiction compared with people in another, for Aboriginal and Torres Strait Islander people compared with non-Indigenous people or for those in major cities compared with those in remote areas. If averages are used for comparisons between large groupings, the 'within group variation' is missed entirely. One example of particular interest to the alliance is the average prevalence of diabetes, say, for one of the primary health networks that includes the whole of a state's or territory's non-metropolitan area. It would tell us nothing about the communities or regions in which the need is greatest. A considerable amount of data is being collected by various organisations for various purposes but too often it is not being drawn together, cross-matched or triangulated. Often the rationale for collecting the data is poorly understood and its importance may not be recognised or the dissemination of the analyses may not have been included in the costings of the program or service.

We believe Australia needs an overarching health data plan managed by the AIHW. All elements of the plan would need to be appropriately funded, including data analysis and the dissemination of results. We believe that relevant evidence is needed to inform the development and targeting of services to meet the needs of particular need groups, including the almost seven million people who live outside the major urban centres. The latest available evidence shows that there is a healthcare deficit for people in rural and remote areas of $2.1 billion every year. That figure allows for the fact that rural and remote people use public hospitals at a higher rate than city people. Analysis of the data proves what is felt and experienced on the ground. Access to health professionals, including GPs, falls with increasing remoteness. A greater proportion of rural people delay seeing a health practitioner due to cost. Diagnosis, therefore, tends to be later, which means there is a need for more specialised care, which is very rarely available locally. Rates of survival after a diagnosis of cancer fall progressively with the patient's distance from the capital city.

Evidence shows that health risk factors are worse in rural areas. There are higher rates of smoking, a higher incidence of dangerous levels of drinking, and higher rates of sedentary lifestyles and obesity. These form the drivers of chronic disease and, unless they can be addressed, will result in poor health and higher costs into the medium and longer term.

The rural population is also older and has a higher proportion of people living with a disability. On average, incomes outside capital cities are 15 per cent lower than in the capital cities.

If only the right evidence were available, it would demonstrate that improved health for the people of rural and remote areas will continue to depend on five strategic approaches: improved access to integrated primary care, especially for infants and children; a holistic focus on the social determinants of health; sufficient resources and political leadership for the National Aboriginal and Torres Strait Islander Health Plan; practical integration of disability, aged and acute care; and investment in health-promoting infrastructure, including for broadband, education and training, job creation and research.

At the risk of presumption, let us end by comparing the alliance with the Australian government. The alliance depends on good data to make its case for special consideration of the situation in which the people of rural and remote areas find themselves. Further than that, we need evidence of the extent to which the special interventions provided in rural and remote areas are successful. For its part, a national government that is driven by return on investment, a worst-first approach and investments in productivity as the basis for hard decisions needs evidence that is unambiguous and transparent. The collection and analysis and publication of data for such purposes should be seen as an investment and not as a cost.

In preparing for today we also developed a very brief case study based on some data recently released in the Australian Atlas of Healthcare Variation. We are very happy to provide you with a copy of that rather than me reading it into Hansard. What we have tried to do in that case study is to give you examples of how we can have data that raises concerns but find, when we start deconstructing it, that there is still so much that we need to know before we can make accurate and valid judgements on what it is that needs to be done to address the issue that this originally raises. So we will be happy to provide you with a copy of that. Thank you.

CHAIR: Thank you very much, Ms Brooke, for your clear articulation of a big problem and also, evidently, your passion for the work that you do and who you are representing. Mr Phillips, did you want to add anything?

Mr Phillips : I do not want to say anything; thank you.

CHAIR: Mr Gregory?

Mr Gregory : I am with her!

CHAIR: Over to you now, Dr Pinskier.

Dr Pinskier : Thank you for providing the Royal Australian College of General Practitioners with the opportunity to appear before this committee. I will begin by outlining the role of the RACGP. The RACGP is the specialty medical college for general practice in Australia. We are responsible for defining the nature of the discipline and scope for the profession. We set the standards and curriculum for education and training, maintain the standards for quality clinical practice and practices, and support general practitioners in their pursuit of excellence in patient care and community service. We have over 30,000 members who are working in or towards a career in general practice. As an aside, I might mention that we are running our next advertising campaign this weekend on national TV.

Since our last appearance before this committee in June, the college has launched its 'Vision for general practice and a sustainable healthcare system'. This is our model for how to best support general practice to improve patient health outcomes and have a sustainable healthcare system. We are keen to see a reorientation of the general practice service delivery model and its relationship to the patient or consumer in order to implement the patient-centric medical home model. This model of healthcare organisation and care shows much promise and has been delivering great results across a range of measures in the United States and other countries. For example, the patient-centred medical home model has led to reductions in emergency department presentations and hospitalisations, and has improved both patient and clinician satisfaction. It will also lead to improvements in healthcare delivery and, ultimately, healthcare outcomes. But the focus for today is on the area of the committee's interest: the collection, linkage and access to health data.

The role of healthcare professionals has changed from recording healthcare information related to a clinical condition and a consultation and managing medical records at a local level—which is the primary purpose—to contributing and transferring information across the healthcare sector to support a range of clinical and administrative processes and other secondary purposes. This has resulted in—perhaps forced—the move away from paper records to electronic healthcare records. General practice today relies on access to computers and associated digital technologies. The overwhelming majority of general practices use clinical information systems to store and manage their patient information. The days of pen and paper in our sector are, by and large, long gone. And, despite the adoption challenges, Australian GPs are in the main satisfied with their local electronic records systems. A published report as recently as this week indicates an 80 per cent GP satisfaction level.

These days, patient healthcare data is collected, stored and transferred by healthcare professionals in and across multiple databases, including local practice management and clinical information systems; hospital patient administration systems and clinical systems; pathology and diagnostic imaging systems; pharmacy; and government repositories such as Medicare, the My Health Record, the Australian Institute for Health and Welfare, the Australian Commission for Safety and Quality Atlas of Healthcare Variation, BEACH, and other places. There are multiple issues in relation to registration and ongoing technical and usage barriers for accessing national repositories. It is a challenge to access the Medicare system. It is a challenge to access, register and participate in the My Health Record system. Despite this, the potential benefits of electronic systems and electronic healthcare records are self-evident. Electronic records, in theory, improve access to legibility and usability of healthcare data. They also have the potential to be a rich source of data for secondary purposes.

However, this increased usage of and reliance on technology has not necessarily resulted in efficiency gains or significant improvements in healthcare delivery. Technology brings with it increased complexity and costs. In the world of e-health, the road to change and adoption has proved to be extremely challenging. This is further complicated by the fact that the methods for collection and storage vary, from direct data entry by the clinician at the point of care or point of delivery to a less useful, and frustrating, practice of scanning of letters and other clinical reports. As the volume of healthcare information and the number of data sources and databases grows, it is becoming increasingly difficult to manage the consolidation, the display, and the transfer of this data, which impacts upon the quality of the information and the visibility of the information, in both local clinical systems and national repositories.

Also impacting on data quality is the access to the recording, and the consistency of the underlying clinical terminology and data coding, which in turn impacts upon the ability to aggregate and reliably interpret the data. More resources need to be provided to general practice and other providers to ensure that their clinical information records systems contain high-quality data: data that is accurate, up-to-date, and appropriate to be transferred to other healthcare providers and other repositories. Healthcare data currently shared as part of a normal clinical referral, such as a referral and a specialist letter, or a hospital discharge summary, can still not be routinely sent between healthcare providers electronically by secure message delivery. The world of snail mail and faxing still dominates. The national building blocks to achieve this electronic interoperability have been funded, and they have been built, but they have not been consistently implemented.

Secure message delivery is a critical dependency towards achieving data interoperability. It was developed as an Australian standard to address the issue of multiple secure messaging products being available on the market, with none being able to communicate with each other. Today, those products can communicate with each other—but they don't. And, whilst the majority of general practices have the capability to send secure messages electronically, much of the broader healthcare sector cannot. This is best exemplified by using an analogy in the communications world, where a subscriber on the Telstra network is not able to communicate to a subscriber on the Optus or Vodafone network; or in banking, where a customer of one bank is not able to withdraw funds from any other bank's ATMs. Despite the fact that the technology standard and the underpinning infrastructure exists today, general practice and other healthcare providers are forced to use multiple messaging vendors in order to communicate. Imagine having to subscribe to three or more phone providers, having to carry three or more handsets and then having to search through a number of different and disparate directories or White Pages to locate the recipient and then having to make multiple attempts to place the call. It just would not work. It would all be too hard.

As a consequence and by way of example—this is one example only—only 18 per cent of Australian GPs report they were routinely notified when their patients were discharged from hospital, and the same low percentage report they were notified when one of their patients was seen in an emergency department. Letters and reports sent by snail mail or fax as opposed to electronically require double or triple handling. General practice databases are becoming clogged with scanned images of letters and reports. In my practice my clinical database is now over 45 gigabytes in size, of which over 44 gigabytes are scanned images. These images are not easily searchable, meaning valuable healthcare data cannot be retrieved or is lost. This is not an efficient way to run a 21st century healthcare system. It creates blockages for the transfer of data, which impacts upon clinical handover and continuity of care, ultimately affecting healthcare delivery.

Another example of where sharing of data is essential yet remains problematic is access to current medicines lists. For all medicines, current and accurate data about the drugs prescribed or dispensed to a patient is crucial in clinical decision making and assist GPs and other providers to prescribe safely, effectively and lawfully. There are multiple channels and databases available today to search for a patient's current medicines, yet there is no one source of true or reconciled list. This contributes to clinical safety risks and ultimately patient harms. This is especially critical in relation to the supply and dispensing of drugs of dependence.

In Tasmania the introduction of a real-time electronic system, the Drugs and Poisons Information System Online Remote Access, known as DORA, implemented by the Tasmanian Department of Health and Human Services, has successfully reduced inappropriate prescribing and dispensing of drugs of dependence. It is available to both prescribers and dispensers in real-time and is currently the only example of an electronic reporting and recording system in Australia that operates in real time. Despite its successes and calls by many health peak bodies, including the RACGP and consumers, and numerous state coroners, there is still no national system for real-time electronic reporting and recording of controlled drugs. DORA is a clear example where data is available that can improve patient safety and outcomes, that can save lives and that is shareable; yet it is not available as part of everyday, mainstream Australian general practice.

With the increased use of technology across the health sector, there are increasing ways to collect and store data but insufficient avenues to make this data available and retrievable, especially for primary care researchers for secondary use of data. The ability to access and analyse healthcare data has the potential to foster ongoing research and future public healthcare planning and to identify populations and individuals who are at risk.

In summary, the data held in the electronic databases of Australian general practice is a valuable resource. It is a unique and underutilised national resource. GPs in the RACGP are keen to work with government, researchers and policy planners to enhance this valuable resource to improve the health and healthcare outcomes of all Australians. Thank you again for this opportunity to appear. We are happy to discuss.

CHAIR: Thank you very much. That was a very elegant description of the breakages in the system when we compare it to other technologies around communication, for example. Thank you for the lucidity of that.

Senator MOORE: I have been asking people about their knowledge of each other, not necessarily personally but in terms of organisations, to try to get a hold on the networking that goes on, particularly around this issue. Is it fair to say that each of your organisations is well aware of the other organisation and what they do?

Dr Starling : Yes. The network is strong. The main problem for us is finding time. There is so much going on. Traditionally there were conferences and so on, which we would all attend, and catch up afterwards.

We have stopped going to most because the registration fees are so high. By the time you add personnel costs, travel and everything else, it can be a $3,000 to $5,000 investment to go to a conference just for a couple of days. Those opportunities are not available now.

The main thing is trying to get a focus for those conversations as well. It is all very well just to have a yack but I think we are all getting very pragmatic now about why we are talking to each other and what we need from each other as well. With the issues that have been raised we are looking for shared solutions, but we are also looking for a shared focus for those solutions that is cumulative across our respective efforts, so that we are reducing duplication.

Senator MOORE: And it is fair to say that this issue of data draws people together, that it is a priority for discussion.

Dr Starling : It is getting consistent data so that we are each speaking after the same issues. That is so much of where discontinuities occur and discussions break down—because there are so many numbers from which to choose and there are no standards, there is no standard model. Harmonising the data that you are going to agree you are going to use is a preliminary exercise that we all have to go through now as well.

Dr Pinskier : The RACGP I think works widely across the profession through the professional colleges. Through the other peak general practice medical bodies, there are a number of alliances and also through the National Rural Alliance and the Indigenous sector as well. We have a national rural faculty, we have an Indigenous faculty and we have a huge number of representatives who attend a huge number of meetings. The challenge, as Dr Starling has indicated, is to synthesise that information, to bring it together and then to frame a policy which is acceptable at a peak level and also is acceptable to the membership. Quite often when you are dealing in the policy world you are exposed to a different understanding of how the world works because of interaction with governments and other peak agencies. That is not necessarily always easily translatable back to the grassroots operator, the GP who is sitting at a desk seeing 20, 30 or 40 patients a day. We work to achieve a policy framework that will work across the sector. It is an ongoing challenge.

Senator MOORE: Again, my point is: is data an issue where there is that—

Dr Pinskier : Data is an absolutely critical issue and I am seeing this increasingly. It is interesting that the Atlas of Healthcare Variation was addressed because I am the college's representative on the Atlas for Health Care variation electronic version, which is due to be implemented in June. We have had a good look at the data and trying to interpret the data is a real challenge. What does it really mean? Even if you could interpret it, having access to the ongoing resources to continually interpret it remains a challenge. Wearing another hat, I am also a board member of a Melbourne hospital network. Every meeting we receive hundreds of pages of reports of KPIs with longitudinal chronological tracking where people have interpreted it and given me the exact reason as to why the waiting time has gone from 4.1 minutes to 4.2 minutes. Hospitals have an enormous amount of resources and data. The tertiary sector is fairly well resourced. I would argue that the primary care and community sector is really poorly resourced.

Senator MOORE: And the rural sector, Mr Gregory?

Mr Gregory : I said to Nathan and Robert before we began formally that I hoped they understood that we worked for them because of course the organisations that they represent, being NACCHO and the RACGP, are two of our 36 member bodies. So the networking between us is happening in a sense on more than a daily basis, several times a day. In terms of data issues, yes it has become much more important to the Rural Health Alliance and its members over the period and I would like to say a couple of things in relation to what we are doing. The Rural Health Alliance runs a major biannual conference, as you know, and in the off year—meaning the other year—we run what is called the Rural and Remote Health Scientific Symposium, which is actually about data issues. One of the things that a networking body such as our usefully does is to bring together the national data agencies. We are all doing great jobs and are very busy but when we arrange meetings it is very gratifying, although perhaps surprising, to hear them say how nice it is for them to be meeting with each other because they are so busy doing their day-to-day work that they sometimes do not meet.

Another example of something that we have done, which relates to underpinnings of data, is the information session we had just last week about the modified Monash, which of course you are well aware of, which is one of the underpinning issues relating to data and data in the future. We expected a relatively small number, but we actually had 130 people from all over the Commonwealth come to Canberra for our session about what the Modified Monash Model is, what it looks like, how it can be used and misused. That is one example of the sort of interest there is in data and data issues.

Senator MOORE: That in itself is a test, isn't it—when you have put something out this close to Christmas, that that number of organisations and people would want to come to talk about the issue and listen to how it is going to work. One of the issues that we have been talking about—and there are so many, and to go into your individual contributions as well—is that the government put out an all-of-government statement on data in the first week of December. I am interested in your views about that policy statement; whether any of your organisations had been involved in discussions with the government before that document came out, in terms of consultation around the process; and, probably most importantly, where you think the process should go now. When you read the data statement, and I know you will have done so, it picks up a few of the themes that have been in your submissions and contributions—about access, priority, privacy, all those things. But we have now got the policy. From the perspectives of your organisations, all of whom have established interests and concerns about the area, how should that policy be turned into action? What should be the process now for organisations like yourselves to be engaged with the government and, in particular, Health, because Minister Ley has now associated herself with the same process? And what should the process be in terms of data that is not health? You have all indicated the fact that it is not just health databases, that you need to look at the wellbeing of people, you need to look at education, employment, poverty, all those things. I know they are three big questions, but I thought, to get discussion going, I would throw them all out and just see how you feel about where we go, now that the government has said that we have an all-of-government approach to make sure that we effectively look at data.

Ms Brooke : I will take the lead for us to begin, but I will throw over to my colleagues as well.

Senator MOORE: Thank you so much for starting.

Ms Brooke : It is a massive issue. When I read the statement, in preparing for today, one of the things that struck me was, again, some of the themes that we picked up in our statement, including the basic lack of planning. Government produces enormous reams of both administrative and program data, most of which does not get analysed, most of which just sits there and maybe someone is employed to set up a database. It might occasionally be looked at for answering KPIs but there is no real rationale and thought given to why they are collecting the data a lot of the time.

We have been trying to raise the issue that, before you go out to tender for any program, you should have already set up what your plans for evaluation are. You should have set up the key questions that you need to address through the program that is going out. You should have that data collection strategy in place for that program. So often, they are written after the contract has been signed, and often that it is too late to then go about collecting the data, because people are already setting up what they are going to do, and, if you come in after that point and say, 'We need this, this, this data,' they have set systems up that are not collecting it. It is too costly to then go back and rewrite those systems. The quality and the openness of government data really are of great interest to us. We would be very happy to be involved. I am not aware that we were consulted at all in development of the strategy, but we would certainly be very happy to sit and talk with Health and with any of the other organisations to talk through some of the ways that we use data. I am going to ask my colleague Andrew to talk further. One of the things that we found was that it was so hard to get all of the data together that we have actually set up our own website to pull that data together and make them available to people. Andrew, would you like to talk about this project?

Mr Phillips : About 15 years ago, there were very little health data around that we could use. But it really has exploded over the last decade and a half. Now, there is a huge amount of information from lots of different areas—a lot of it by remoteness areas, a lot of it by other geographic categories. When we are trying to provide governments or others with advice and our view on what is happening in rural and remote areas, and what may need to happen in those areas, it is a battle trying to get hold of up-to-date data from multiple different sources that cover both the social determinants, the prevalence of disease, service delivery and health outcomes. What we have done—in frustration, I guess—is start to put together a resource—

Senator MOORE: The little book?

Mr Phillips : The little book of rural health numbers. That is it. Exactly. Have you seen it?

Senator MOORE: No, I have not, but it has piqued my interest.

Mr Phillips : It was an in-house resource that we were using to shortcut our way to be able to provide good advice. We thought that it should be made publicly available, even though it is definitely not complete—it is patchy. We could spend the next year and one or two people full-time working on it to keep it up-to-date and to expand it, but we have one person two days a week—myself—to do that. So it as up-to-date as it can be. It is useful. We thought it would be very useful for people around the country to be able to utilise it.

Senator MOORE: As an organisation, are you encouraged that the federal government is now committing the government to specific action designed to optimise the use and reuse of public data? 'The government will remove existing barriers to sharing public data for policy analysis research and development. We acknowledge that the government holds extraordinary amount of unique data.' That is just selective quoting from the policy document. Does that give you encouragement?

Mr Phillips : Yes, it is very exciting.

Senator MOORE: Mr Gregory.

Mr Gregory : We have all been talking in hearings such as this and in many others for years about the desirability of a whole-of-government, or a joined-up government, approach to things. Of course, we have also talked endlessly—have we not—about the difficulties of such an approach. If we can get a start with data, meaning that if we can seriously obtain and use the data relating to a person who is unwell, with data pertaining not only to health professionals and the health sector per se but also to social and economic origins or circumstances, so to get a real picture, then that would be a great start. Of course, it still leaves the challenge of actually doing something about it in a whole-of-government approach. If you could see that the main thing with this person's ill health or mental illness is actually related to trauma or education, or whatever, still the challenge is actually to do something about it in a whole-of-government sense which is the most desirable. But at least having the data would be a start.

Senator MOORE: Dr Pinskier, from the Royal Australian College of GPs?

Dr Pinskier : From the Royal Australian College of GPs perspective, we strongly support an approach to open access and transparent data in a timely fashion. I can echo everyone else's comments here. In my experience, I am continually hearing from my GP research colleagues that it is extremely difficult to get access to timely information. I spent 10 years on the DHS Medicare's stakeholder advisory group. I was able to get access by going and knocking on people's doors, but other people spent hours, months and years knocking on doors and still did not get anywhere. By the time you get access to the data, the data are irrelevant, not contextual and time has moved on. We would like to see a model where data are readily available. We would like to see the agencies that provide support to the government, like AIHW, funded to be able to provide those conduits via the internet of things, which is available and is an easy channel to access. It is great to have a statement—it is wonderful to have a statement—but the proof is in the pudding and the delivery.

The AIHW is a really good example of an organisation that was targeted to be merged or shut down. It is chaired by my good colleague Dr Mukesh Haikerwal, former AMA president. It is fair to say that it was a struggle to keep the agency going as an independent authority. Fortunately, that appears to be the case now. But if you start to shut down these agencies—and I understand there are budgetary constraints; we get that—who is going to collect, maintain and harvest the data? What we would like to see is one agency responsible for coordinating the data for every other agency, so there is one channel or access point, so you do not have to register here or there and use a certificate for this or a credit card for that. There would be one point where I could go in and easily access and extract data and be able to interpret and analyse that data.

CHAIR: And you see that as the AIHW—

Dr Pinskier : One would argue that is a reasonably logical organisation to work through, because it has the capacity and track record and it understands health. That is a good starting point. The other issue is, once you have extracted and interpreted the data, we know where the evidence is, broadly. We know where the issues are around the social determinants of health and where health inequities are, but we as a society have to decide what we are going to do about that. It is all very nice to say we are going to build a brand new hospital, because that is easy to do: 'We'll just build it and put some people in it and magic will happen. We'll do a few liver and heart transplants and that'll win me another election.' That is terrific. But we know the real savings made in health, around the world, are in investments in preventative health and primary health. We are not seeing a willingness, from any side of politics, to make a commitment to invest in primary, preventative and population health.

There is overwhelming international evidence that supports that approach, that it keeps people healthier longer, keeps them more productive and keeps them out of hospital. Every dollar invested in primary care is a three-dollar-multiplier benefit, in terms of savings in secondary and tertiary care. Until we do that we are not going to see significant changes. To have the data and analyse the data we have to do something around the long-term policy planning, and we have to change the culture.

Senator MOORE: We had evidence this morning, in one of the roundtables, about the very point of having a central location, almost like a clearing house, where the various data sources would be available. They would not be analysing it. They would not be doing anything apart from saying, 'This is what is out there.' One of the people gave evidence that there should be a data dictionary so that you could say, 'This dataset is collected and this is what it is used for and this is what is defined.' It is an enormous job. Is that the kind of thing, Dr Pinskier, you are saying there should be?

Dr Pinskier : Absolutely. That is the conversation we are having with the online Atlas of Healthcare Variation.

Senator MOORE: Which is from the National Health Performance—

Dr Pinskier : The Australian Commission on Safety and Quality in Health Care. So now you have another organisation doing a different piece of work.

Senator MOORE: We had the National Health Performance Authority and they seemed to think they had some ownership as well.

Dr Pinskier : The NHPA. They have some ownership of that as well, and they are on the subcommittee as well. That is the problem. You have too many agencies playing in similar spaces. Somehow, we need to bring them together as one super-agency that manages the data, on behalf of the government, and makes it available to the consumers and to the researchers and users.

CHAIR: Is that a bid for the RACGP?

Dr Pinskier : We would be very happy to participate in that process, but we do not see ourselves as that superagency.

CHAIR: Fantastic. Mr Phillips, you wanted to add something.

Mr Phillips : I just wanted to say that the sorts of sentiments Dr Pinskier was expressing are the sorts of sentiments that we would strongly endorse, particularly in regard to MBS and PBS data. I have had personal experience of trying to get data, from the department, to work on it. It only took me about four months—

Unidentified speaker: That was quick!

Mr Phillips : sorry, 12 months—to get the data from them. That is while I was working for the Institute of Health and Welfare. It is also worthwhile to mention that even today the Institute of Health and Welfare does not have access to MBS or PBS data and that the institute has particular legislation that allows the accumulation, if you like, of data in a very safe and secure environment. They have a data dictionary, which has been operating for at least 15 years.

Senator MOORE: That is the AIHW?

Mr Phillips : That is right. We thought you could have whichever organisation you wanted to be the central repository or whatever—there could be several of them, I guess; but the institute seems to be well placed. It has massive experience in this area. It has good infrastructure in this area and good legislation in this area to collect the data.

Dr Pinskier : And good governance.

Mr Phillips : Yes, and good governance.

Senator MOORE: One of the department's submissions—we have several from them—says that the department is 'providing the AIHW with five years of MBS and PBS data to support the construction of an enduring dataset that can be accessed in a secure manner by responsible researchers looking to investigate any health related issue'. That would be a relatively recent movement. One of the questions that people have raised is: why five years? We will be asking the institute about this. There is another round table coming up with government agencies. I am interested in the five-year figure and whether that is a resources issue or whether it is a scientific decision. What they have said—and this is picking up on your point—is that there has been this demand for MBS and PBS data as a bottom line in doing anything in health. So they have committed to doing that as part of their plan.

Dr Pinskier : It cannot be a resourcing issue, because if you establish a personally controlled electronic record or a My Health Record, your data from MBS and PBS persists. Once it is in there, it persists. So the data is available. Part of the challenge that I have seen is that government agencies typically—and I am sure you are more familiar with this than I am—operate in a silo fashion. They lock things down. They are risk averse. They only want to release as much as they need to release. The challenge is to be able to move the response and the obligation for the data management, the data curation, the data custodianship into another agency. It should not be a five-year question. It should be a case of: 'There's a link. Make it work. It's a secure link. Make it work. Let's move on.'

Senator MOORE: I know that Mr Phillips wants to contribute, but, Dr Starling, I have not got to you with my specific question about the government's statement. Does it encourage you? How do you feel about the fact that the whole of government has made a statement about the importance of data and their openness in this space?

Dr Starling : Just as a digression: my background for many years has been in international standards for geographic information systems. Other work has been with the United Nations environmental program in Nairobi, which is about how you look at linkages across national and regional datasets in relation to the millennium goals and so on. There is a thing called the Open Geospatial Consortium, which is the international standards for geographic information. A few years ago, we tried to get a chapter of that going here in Australia. It was refreshing to read the report on the public sector data management. It would have been nice 15 years ago, because what it is saying is no different from what was being said at that time, especially when the opportunities of the internet were being explored across a wide range of agencies. Now, instead of having to hold data in central repositories, you store directories, you store road maps. So the data could be kept at source, within its respective custodial arrangements or authoritative sources and then be accessed on an as needed basis. That is the architecture which we are putting together for the NACCHO health information strategic plan. We are working through the Queensland Aboriginal and Islander Health Council as well. So, for groups like EFRI and John Glover's group and so on and so forth, we are saying: 'No, you give us access to your data. You have a data dictionary. We will access and acquire information on an as needed basis.' We do not have the resources. We do not want to store it, because as soon as you store it you have to update it.

The model we are hoping will come out of this would say, 'No, you're the authoritative source for this. You may abrogate that to a custodial role for AIHW to act as a custodian, but you're still responsible for that data.' Identifying the authoritative sources then also forces you to decide about who is responsible for which data items, because at the moment so many people claim responsibility or claim authority over data. A key question is: what is the authoritative source of the data? That comes out in that report as well—who is responsible for what, where and when? How is it going to be licensed for use? I do not like the idea of having to pay for it. One of the problems we have with AIHW is: if we want to send back the data that the sector has actually contributed to AIHW, we have to pay to have a report generated.

Senator MOORE: How much do you have to pay?

Dr Starling : It is in the tens of thousands of dollars—it is not 50 bucks here or there.

Senator MOORE: I will ask AIHW whether there is a pricing scale.

Dr Starling : I think it is on a level of effort—an anticipated cost basis. I think what is being highlighted by us all is that data per se is just a grab bag—fill it up with data and get a fishing line. The thing we are really trying to focus on are the questions we are trying to answer. I was most interested in your point, Fiona, about major projects being funded and then six or nine months later you get the reporting criteria. We have classic situations with the Aboriginal and Torres Strait Islander CQI framework. We are coming up to the first period of reporting on that but the framework itself is yet to be finalised, let alone the reporting requirements, which are to be met by 29 January.

There is a lack of planning, vision and foresight. Try to map the National Aboriginal and Torres Strait Islander Health Plan and the implementation plan to the AHMAC Indigenous framework's two-yearly reports or to the CQI or to the NIRA report. They are inconsistent and they are not cumulative. A frustration that we are feeling is that we have such limited resources between us but we see outsourced research groups getting all this money to look at questions. We say, 'We have that data; we know what that is. Why aren't you engaging with the sector, with the data originators and data owners?' It is like the GP practices—use that expertise to inform the process.

The bit that is so important, and the statement will help us, is build capacity with our own respective sectors. At the moment stuff keeps getting outsourced—building expertise in people who have no long-term commitment to the sector or the industry or the population with whom they are working. We really want to say, by asking relevant questions and by taking a multidisciplinary and multiagency approach to it: 'Right, if we can get that question answered, we can set up data models and put them in place. We can set up the request and response mechanisms to get the information.' That then becomes a sustainable entity as well, and then we can say, 'Knowing this, we can move on.' There will always be leaders and there will always be laggards in the process. Until we start to get specific questions on the table and specific processes in place with specific outcomes in mind, we are just going to wander off and not see the real opportunity that the strategy does present. The proof will be in the pudding, but opening a door does not do anything. Why are you opening it and what do you want it open for? How then can you use quality improvement within the data, now that you are actually using that data, to improve that over time and reduce the duplication of measures?

So it is a bit ominous. We have a very clear pattern as to where we wish to go and which data cookie jars we want to be able to access but it is longer than my lifetime is going to be at the current rate.

Senator MOORE: A number of you have mentioned the atlas and we have had information this morning about that as well. My understanding is that there is no attempt in the atlas to link data; it is just a straight snapshot of known data and it is there and is very valuable as a point but in terms of what it actually gives you it is a list of data that is already available in different ways looking across the country. It is very valuable in its own way but it makes no attempt to get data to talk to each other and to come up with analysis. That is my understanding.

Ms Brooke : That is correct. It is very high level.

Senator MOORE: You have to point to her. She is in charge. People are putting their hands up.

CHAIR: I can see Mr Phillips is waiting there too.

Senator MOORE: Sorry, Ms Brooke. We have had a number of groups this morning and people have been so well-behaved they have put their hands up—it took me about 30 years. I have thrown you right off. I do apologise.

Ms Brooke : I have lost my train of thought.

Senator MOORE: I hope it comes back.

CHAIR: Mr Phillips, you be the trigger.

Mr Phillips : Two things: one was a comment previously about whole of government and from our perspective this is particularly important. A lot of the work we engage in, because of the way the data is organised and some of the political issues, et cetera, are issues around workforce, service delivery and so on, but so many of the problems for people who live in rural and remote areas and for Indigenous people, most of whom live in rural and remote parts of Australia, are issues related to access to work, career choices, access to education, the relevance of education, how they are going to get to university, law and order in small communities and those sorts of things. So the advantage of a whole-of-government approach is that you might get a joined-up approach to health that goes back to deal with the determinance of what comes afterwards that the medical profession have to deal with. The other thing I would like to say relates to data linkage, which you mentioned, and MBS data and PBS data and the Institute of Health and Welfare. The institute, for quite a few years now, has been doing quite a lot of data linkage work. You cannot link data unless you have access to the raw data itself and you have the opportunity to link. So an advantage of having data sets like MBS and PBS available in its raw form within the institute would be that there is an opportunity to link other data sets—for example, mortality or hospital data—to that data and get a much richer understanding of what is going on with people.

Dr Pinskier : It is really interesting because the Atlas of Healthcare Variation really just provides the statistics and makes no attempt to interpret the statistics. One of the issues the college has is that, if you start to make crude interpretations, you might draw erroneous conclusions as to why there are significant variations for all sorts of reasons.

CHAIR: Yes. This goes to Ms Brooke's opening statement about analysis capacity.

Dr Pinskier : Absolutely. We have had a look at them. When I sit around with our quality groups, it is quite challenging. People give you a million reasons as to why things happen in certain ways. When you sit with all healthcare professionals you get lots of different interpretations. We clearly do not know a lot of things about why health care is delivered in certain ways. The other point I want to make about the Medicare data is really interesting. When we developed the personally controlled electronic health record, there was a view that the MBS and PBS data was purely administrative data and had no clinical or healthcare related value. It was purely a set of information attributed to a particular service delivery.

I think there is a really interesting debate to be had about what the role of the Medicare data is—the MBS, the PBS, the ACIR, the organ donor and other datasets that they hold—and how we can actually extract value out of them, because at the moment I do not think there is a uniform view around this.

CHAIR: Dr Starling?

Dr Starling : There are a couple of things that come to mind as well. When we did get the MBS and PBS data, we used to get it at the division of general practice level, so that gave a geographic granularity to it, for the 130-odd—

Dr Pinskier : One hundred and twenty-one.

Dr Starling : Then we got the Medicare Locals and we got 61 of those. Now we have 31 PHNs.

Senator MOORE: It is getting lower and lower!

Dr Starling : Look at remote and rural WA. I used to work for the Kimberley-Pilbara Medicare Local, and we had a million square kilometres. Country Western Australia is now 2.5 million square kilometres. There is quite a bit of variation within it.

CHAIR: Yes, I know. On the Central Coast there was a discrete unit; there was the Hunter; then there was New England. Now we have 1.1 million in a primary health—supposedly local—network, that goes from the Hawkesbury River to the Queensland border. I think there are six extraordinary ones like that across the country.

Dr Starling : I was in the Mount Isa this week and that goes from the gulf to the New South Wales border as well. I think we really do need to think carefully about the geographic granularity of this data. We have been rerunning analyses—I think the AIHW ran them—using SA 1 level data is supposed to SA 2 or 3. What you are looking for there are patterns. You are not looking for absolutes. So: 'Where are there gaps in service delivery?' That is why we have been doing the heat-mapping and all those sorts of things, so that we can at least just start to visualise things. As you probably know, we are very much into geo-referencing everything that we possibly can, and using maps and dashboards as a way of visualising that, to use within our own groups but also in our community consultations.

A focus I would like to suggest around this data is also something that is concerning us, and that is: the PHNs got another $900 million or something to distribute with the mental health announcement in the last few weeks. We are concerned about how that commissioning process is going to work. How then are the metrics which you are describing going to be made consistent across those 31 PHNs in their contractual requirements for their commissioned services? At the Medicare Local level, we were able to basically set our own performance measures for the providers who were delivering those services. They were probably quite different across in the Territory or in Barwon—even though there was no guidance from the department and from the government as to: 'What are the measures that we really want to collect uniformly across the country when we are distributing this amount of money?'

CHAIR: And the capacity to actually articulate useful datasets will improve if a cycle of research, investigation and evidence-based decision making starts to be more clearly established. At the moment it is, 'Make it up as you go,' really, isn't it? You choose your own adventure, in terms of what you measure and what you match up.

Before we shift off this—and I will come back to you in a minute, Dr Pinskier—I will go to your submission, Mr Gregory, because you made some comments in some detail about the geocoding advances by the ABS, but you also note substantial variation between jurisdictions around datasets for geographically identifiable data—Mr Phillips.

Mr Phillips : This came up particularly when we had our Modified Monash workshop the other day. That statement is based on comments from the Institute of Health and Welfare and the people who do geography within ABS, and there was some concern that Modified Monash may not be able to be reported for all datasets. For some, definitely, it would, but for quite a lot it would not be able to be reported because of the granularity of the Modified Monash Model and because of the fact that quite a lot of the data that comes together nationally comes from the various jurisdictions, from states and territories, and the way in which that data is geocoded—the address information—differs quite a lot. The institute and the ABS are undertaking a process of trying to get those various jurisdictions to provide better, tighter geographic identifying information. How long that process will take is difficult to say. But it is certainly in the forefront of a lot of people's minds.

CHAIR: Dr Pinskier?

Dr Pinskier : The previous comments by Dr Starling, I think, are really relevant. We have gone from an era when general practice, you could argue, was reasonably well supported, possibly oversupported, with 121 divisions, to probably a reasonable balance of 60, 61 Medicare Locals to an era now of Primary Health Networks, where, if they have any visibility on the ground, it is almost non-existent. The question that we as a college are trying to determine in our vision for general practice and vision for a patient-centred medical home is: how are we going to support that transition if we do not have access to the data and the ability to analyse the data and then the ability to make the changes in private and primary practice? You can have all the heat maps you want being produced everywhere, but nothing will change.

We have to have evidence that will drive a case for change in the way that we deliver health care on the ground. We know that there is a reform process happening, but I am going to, reasonably, argue that, if we do not invest resources at the local level to make those changes, in 20 years time general practice will not have changed that much and we will still be saying that we are not providing optimal care because we have not actually defined what we need to around change and adoption. We need to be able to translate the evidence into action on the ground. At the moment, we do not have the resources to be able to do that.

CHAIR: Is it a funding gap? Is it a gap in will?

Dr Pinskier : I think it is probably the model. There is an issue around the model and there is probably an issue around the funding. One of things that we have advocated for is to develop a capacity within general practice for data analytics. We have invested a lot in practice nurses, through the practice nurse incentive payment. We do not have a strategy to necessarily educate those nurses to support data capture. And if you look at practices around the country—in my practice we might run data analytics once a month; we might do it once a year. There is no system. There is no process to do it sustainably.

We have invested in the Improvement Foundation, based out of South Australia, which runs some interesting collaborative work. But, again, as soon as practices drop out of that collaborative and they look at their data and they work out why their waiting times are so long, why their coronary heart disease risk factors are higher or why their HbA1c tests for diabetes are higher than at the practice up the road—that works for three, six, nine months and then it drops off the radar.

Unless you get ongoing resources and support, people who are invested in the sector—not consultants, I agree; it is not consultants. It is the people who live and breathe the sector who understand. So we do not want the McKinseys and the PricewaterhouseCoopers coming and knocking on our door for three months and disappearing. We want to teach people to fish, and we need people in the regions to come and do the work. So it is more than just having a PHN; it is how you then commission ongoing, continuous support to make those sustainable changes.

CHAIR: Dr Starling?

Dr Starling : What we have done with our work is focus not just on the patient home idea but on the patient journey. The AMSs have a comprehensive holistic model anyway. So it is more of a team arrangement in terms of diagnosing and managing people's and families' conditions. The connectivity between the different providers is the area on which we are concentrating at the moment, because, as Dr Pinskier mentioned, unless you have a clean handover of data to a person of responsibility—also liability!—everything is left in limbo, if you like. The referral information does not get to the right person in the right time. The report does not get back to the right person in the right time. It is the communications between the providers which add complexity to the journey for the patient. So what we are focusing on in our own discussion and around the CQI model is how we can use that continuous quality improvement to get referrals working to local GPs, allied health, mental health, hospitals, but then also monitor how well they are working. What is the waiting time? Do people attend? Could that best be serviced by teleconference rather than face-to-face? Do we have to fly people to Brisbane or to Perth or somewhere in order to get to an intervention, if we are not diligent in helping them manage their care plan and something goes wrong? The thing is, there is the higher-level stuff we have talked about, there is the general population health, there is the feedback to policy. But it is about getting it back to me as the patient. I am a patient as well. I have my patient journey. I know where the complexities lie. I know the frustrations in having to repeat filling out forms and doing consents—as we all do.

We want to start looking at the data linkages and data access, to identify areas of need and populations at risk—but then also: how can we improve the patient journey for those people in those situations? And that is—like you mentioned earlier, Senator—about collaboration between different groups. And we are talking to some strange partners at the moment! Just because there was no real context and no outcome that we could see, in forming relationships with some groups. Relate it back to the patient journey; it could be an environmental health issue, or a broader education issue, or something like this. We have now got a context, because we have got that patient journey model. And I think that is something which we are working through.

In terms of where we go with this new policy, and the idea of nominating certain projects coming through the Digital Transformation Office now, which we would love to participate in—and be funded for work in that area, because we don't have any money!—it is about the connectivity of the different service providers. And that does feed back into the commissioning process, because that is then showing how effectively you are using that money for the patient—it is a means to an end for the patient, not a means to an end for the business.

CHAIR: And are you saying that we will not get that unless we get the big data picture right, because that will not drive the innovation that you are talking about?

Dr Starling : It will promote the silos, and the continuation of the silos—because I can meet my compliance criteria for my contract—tick, tick, tick, tick, tick—but it has actually not done anything for my patient's journey.

CHAIR: Okay. Dr Pinskier, do you want to add anything?

Dr Pinskier : No; it is what I emphasised in the opening remarks: that the critical issue from our perspective is the lack of interoperability. The inability to be easily able to transfer data captured at source, and send it on to other parties, other providers—to be able to share that data, we say either point-to-point, which is between providers, or point-to-share, which is up to national repositories for use by unknown providers at a future point in time. The technology all exists. There is a lack of wherewithal to just close a few gaps and make it happen.

If we were going to be critical, one of the major criticisms we had around the National E-Health Strategy is that it was heavily focused on point-to-share. What we said is that you need to get the point-to-point environment right. The business-as-usual stuff that a clinician does every day in conjunction with their patients—that is, I see a patient, I need to refer them, I want to send something electronically; I will support them to send things out and receive back electronically. Once you are doing that, it is a very small step then to send point-to-share. But if you focus on the end of the journey—point-to-share—and you say, 'well, every time you see a patient, do a care plan; push something up to a national repository', then the GPs and the clinicians will say: 'Well, that is nice. I am not interested. I do not have the time.' It has got to be leveraged, not the primary purpose. And because that part of the journey has not been consolidated, and because there has been no investment in getting the interoperability between all the secure messaging brokers, we have a whole lot of data that just does not go anywhere or gets lost in the system.

CHAIR: But the data that needs to move with patients across the system is not moving.

Dr Pinskier : It is not moving in a way that you would like. Again, 10 years ago one could argue that a national, big-box, big-data, Big Brother solution was the way to go. Today, there are so many more agile, local exchange models that have been implemented rapidly by all sorts of providers. We would argue that a better model would be to allow the sector to innovate and create the conduits.

There are solutions in the states such as virtual handshakes. If a patient goes to a hospital, the hospital opens up their repository and says to the patient: 'Here is your data. Now you can share that with any healthcare provider or any other person you want. Just let them log in and see the data.' The patient journey is then immediately made available to anyone.

CHAIR: Are there problems there with different jurisdictions, though? If they are all created in such a variety of different ways, there are points at which the different systems cannot communicate easily with one another. We have people who move around the country.

Dr Pinskier : That is a challenge at the moment, so what we want to look at is national standards and national interoperability so that they can be sent. The example that I gave is that it does not matter if you are on Vodafone or Optus. If I make a call, they can place the call. I do not need to think about which server the call has to go through. It just works. If we build the standards, the compliance points and the conformance points, we can make all this work.

The health information exchanges are starting to work in the states. They are working internationally. We need to build the same sort of standard and not rely on the government continually building bigger and bigger boxes and having to invest lots and lots of taxpayer dollars to continue to support them. If the policy changes then the box may disappear, and what are we left with? It is a single point of failure.

CHAIR: In the Australian context, are you aware of any jurisdictions or even pilot programs within jurisdictions where any of what you are articulating is being attempted? Is there any research going on? Are there any points that would give us an indication of what Australia might be capable of?

Dr Pinskier : There are lots of examples of projects that have commenced, run for a period of time and been shut down. We have not continually resourced projects on a sustainable basis, because in the last seven or eight years the Commonwealth, with the National E-Health Transition Authority, has basically gazumped a lot of those projects. But there are local hospitals that provide access. It has happened in New South Wales, in Barwon—

CHAIR: This was some of the innovation of the Medicare Locals. They found that this was a problem, and so we did get spots of innovation.

Dr Pinskier : Yes, absolutely.

CHAIR: This is what I am looking for. I know it has been going on, but where has it happened? Is it continuing anywhere? Are there any bright spots?

Dr Pinskier : There are some great spots. Barwon Health is a really good example. It worked outside the Victorian government's HealthSMART strategy and built its own solutions around interconnectivity and interoperability. It is a really good example of an e-health environment that is working. But every time you get a change—divisions to Medicare Locals to primary healthcare networks—a lot of the knowledge gets lost, transferred or relocated. I am seeing that even in Peninsula Health, where I am a director. They are now having trouble engaging with a PHN which runs from the Yarra all the way to Portsea and then around to Cranbourne and back. It is too big, because the hospital network does not even cover half of that. So the requirements of the PHN do not align with the requirements of the health network. That is a problem in itself.

Where there have been projects, they have not been either maintained or sustained, or they have had to rely—

CHAIR: Or they have had deliberate political interruptions.

Dr Pinskier : Or there have been interruptions. The Tasmanian one, with real-time reporting, was a great one. The Commonwealth signed on to that four or five years ago, and then they said they funded the states. The states said, 'We haven't got the money,' and then it does not happen and you end up with people having adverse events and dying from an overdose of S8s because the poor clinician at the end of the day did not know they should not have prescribed it and the pharmacist was not aware they should not have dispensed it. We can stop a lot of this, because the national standards are actually there. It is now about closing the gaps and creating the conforming repositories that we have talked about. I think that is the challenge. Let people leverage the repositories and create the portal views rather than trying to build the repositories.

CHAIR: Thank you. Dr Starling?

Dr Starling : There were two issues there. In the early e-health days the idea was to set up local repositories which would then be federated. They would be joined together by a smart switchboard system at the top. It would be similar to a smart index, as I was referring to earlier, which would tell you where someone's data resided rather than having it all sitting in one pot. I was disappointed when the national e-health strategy changed from that federated model where, as I was intimating earlier, you could get areas of innovation and excellence and movement like Tasmania which could then move out and be adopted by others because you would have a comms framework within which that connectivity could occur. So that is one aspect.

The other one, though—again within the AMS sector—is that they have shared full clinical electronic health records throughout the Kimberley.

Unidentified speaker: MMEx.

CHAIR: Yes.

Dr Starling : Yes, through a software approach—

CHAIR: The committee has undertaken over 40 hearings since we came into being earlier in this parliament. We did an amazing investigative tour across the northern part of Australia, and we got unbelievably positive responses from the Northern Territory and from just across the border in the eastern Kimberley about the way in which the use of electronic records was impacting very significantly and positively, particularly on Indigenous health, as it is such a moving community. But their concerns were about being overridden by a Commonwealth system that was not as fleet of foot and certainly not familiar to them. These interjurisdictional issues are currently pressing in a place where it looks like the Northern Territory have got their e-health record going along quite happily. What are you hearing about this?

Dr Pinskier : I can tell you, because I attended the early meetings when the Commonwealth decided to morph the Northern Territory shared record into the national system. The people who developed the Northern Territory record, notwithstanding a lot of it comes from Commonwealth funding—but that is the nature of the Northern Territory—developed it for a specific purpose. The shared record was developed essentially to support local Indigenous community medical centres where their patients were wandering and seeing another centre 100, 200 or 300 kilometres up the road. They were saying, 'We're part of that same service, but we don't have access to the clinical information in the other service, so let's build an outreach.' Essentially, the share was an extension of the local record system, and from there it was built into a statewide system to support Indigenous healthcare delivery, and it worked really well for that specific requirement. It was bolted together, cobbled together. The technology may not have been state of the art, but it worked.

The moment the Commonwealth said, 'Okay, we're going to take those 60,000 Indigenous Australians and now migrate them into the national system,' the clinicians were not happy about it, because they had customised it. They were getting pathology reports and lots of useful information that was not available on the national system. The way the reports are displayed in the national system is not consistent with the way they appear in the local record systems of the share. There was a lot of frustration. So now you are running parallel systems, and I think it will take a long time before the clinicians are comfortable with the change.

There has been a lot of retrofitting to try to make it work. But by doing all that retrofitting you change the primary purpose of the person that controlled the electronic health record, which is that it is a consumer-held record which is made available to clinicians at the choice of the consumer, whereas the share was a clinical record, with no consumer view, to support clinical healthcare delivery. So they were two quite different things, and you cannot easily just morph one into the other, because you end up creating a problem. You end up with two different products and, I would argue, a bastard outcome, quite frankly. It is neither a clinical record nor a consumer record, and that is where we have ended up.

CHAIR: That sounds terrible, Dr Pinskier.

Dr Pinskier : It is unfortunate.

CHAIR: But it does seem to me, given everything that has been achieved in the history of the world, that we could come up with some system that would meet those requirements plus the big data requirements that have been articulated today if there were enough forethought into what it is that we are trying to get from this data. There are multiple users of it. Is it impossible to conceive of one place in which those multiple users could get their information collected?

Dr Pinskier : I think we can have one aggregation methodology and then multiple viewing points and allow different people to view—

CHAIR: To see different things of that aggregated data.

Dr Pinskier : Yes. Then, if you standardise the method of collection, you standardise the underlying clinical terminologies and coding, so the data means the same thing to you and to me and to a machine, and then you can start to really interrogate it; it becomes quite robust data. But because we collect it differently and we code it differently, and one machine can read it and the other machine cannot read it, how do you interpret it?

CHAIR: This is why I worry about all these separate little innovation projects happening, rather than something much more.

Dr Pinskier : We have a national coding system, SNOMED. It has not been broadly implemented, for all sorts of reasons. I think we are probably getting closer today. We have a national coding system for Australian Medicines Terminology. Again, it has not been broadly implemented. If we focused on linking the national classification and coding systems to the data entry, then we can start to leverage the data that is collected, otherwise we are continually running these research projects all the time to try to figure out what we really have collected in another way.

CHAIR: What carrots and sticks need to be used to advance those critical changes?

Dr Pinskier : The college's view is a collaborative arrangement with the people who play in the sector, with the vendors who play in the sector. It is critical to get all those legacy products into the room as part of the conversation. We are all going to benefit from being able to access and interpret this data, but for vendors it is a challenge because they have already built a whole lot of systems and databases and software. For them to retrofit costs money. The Commonwealth comes along and says, 'Upgrade this; change that; change the MBS requirements; add on a PCHR; do a CDA-conforming document'—there is so much terminology it drives you nuts—and the vendors say, 'Why? My customers aren't asking, and it will cost a lot of money and it doesn't fit into my product development life cycle.' We need to work with industry as partners. Ultimately if we want access to that data and we want it done simply and to do it once and do it right then we need to support industry in tooling up to be able to meet those requirements.

CHAIR: We heard about the need for 10-year funding cycles for this sort of structural and cultural change earlier on. Would you support those calls?

Dr Pinskier : Absolutely. I think it is about giving certainty to the key participants. We want to see innovation. We do not want to stifle innovation. But we also want to support industry to move towards standardised models. The analogy I give is: when you get into a car, the steering wheel looks like a steering wheel; the brake looks like a brake, the dashboard looks like a dashboard. When you use software and the underlying data models that underpin it, there are so many different models it is complex and confusing. That is why we cannot extract the data in a form where we can interrogate it easily.

CHAIR: We are going to have to draw to a close very shortly, but can I just give you the opportunity to make one key recommendation? If we are going to implement a recommendation, what would your No. 1 be? I will go to Dr Starling first because he looks the most relaxed with that question!

Dr Starling : I think it would be to use the recommendations of the Productivity Commission of the public sector data management and focus on the patient and the patient's journey through the health system as a way of bringing the highly varied parties together with a common interest which is actually the patient.

CHAIR: Thank you. Dr Pinskier?

Dr Pinskier : I would say it is: listen to the needs of the sector so that infrastructure can be developed that allows interoperability across the sector. Once you achieve interoperability, it means that you can start to move data around and you have all sorts of possibilities.

CHAIR: Get Optus to talk to Vodafone.

Dr Pinskier : Absolutely.

CHAIR: Okay. Ms Brooke?

Ms Brooke : I think Andrew will—

Mr Phillips : I will proxy.

Ms Brooke : We have just caucused and—

Mr Phillips : I draw your attention to page 2 of our submission. In that we say:

What is missing is a single, overarching long term plan clearly identifying the range of health data needs and the rationale that should underpin what data is gathered and by whom, how it is brought together and stored and how it will be analysed and disseminated with appropriate funding attached.

CHAIR: Thank you for coming to spend your time with us this afternoon and share your insights. It has been very helpful.

Proceedings suspended from 15:48 to 16:03