A constant in the history of modern medicine has been the struggle for successful diagnosis and treatment of disease.  Over time the recurrent efforts have proved both costly and essential, a notion the Commonwealth of Australia knows all too well. Each year Australia spends billions of dollars in healthcare costs, a large portion going directly towards treating a myriad of diseases. In 2004 alone the cost of treating Alzheimers was estimated at $3.6 billion and today there are currently 500,000 Australians living with dementia.  Cancer has also been a significant public health issue, with the number of new cases reaching 100,000 for the first time in 2005, and continuing to rise.  Along with rising health risks, the advent of improved medical records and technological advances has created an ever-increasing mountain of data that must be sifted through in order to analyze and track diseases.  Information such as lab research, clinical visits, and long-term patient data are all key components.  Australians have responded to the changes by turning to research that supplies new tools to the medical community.  

	 For over 40 years the Garvan Institute of Medical Research has been dedicated to relieving human suffering. The work done at the institute encompasses a wide range of diseases including cancer, diabetes, osteoporosis, arthritis, obesity, and neurodegenerative disorders such as Parkinsons and Alzheimers.  In all, the institute oversees over 40 research groups and is one of Australia's largest medical research institutions with more than five hundred scientists, students, and support staff.  

	A key driver of Garvans longevity and success has been its ability to compile and analyze research data.  When researchers can accurately obtain results, they are apt to produce more precise findings, thereby increasing the potential for medical breakthroughs.  As researchers share discoveries via articles for prestigious medical journals such as the New England Journal of Medicine, the forum for medical innovation widens.  In the long-term, the research leads to a better understanding of disease-management therapies and improved diagnosis; therefore the organization and communication of research findings is vital not only to the institute but to the healthcare industry at large.

	Maintaining the status of an internationally recognized leader in medical research is no simple task.  In fact, Garvans accolades have been attained solely through an enduring battle against disease.  The battle has required adaptations to evolutions in healthcare as well as medical and research technology, and the process is ongoing.  One particular adaptation is best evidenced by the institutes desire to generate detailed reports merging lab research with patients clinical information.  The choice made it clear that a reliable information architecture was necessary in order to deliver the many disparate systems into the hands of researchers.  Specifically, Garvan scientists needed the means to easily query vast collections of research and clinical data to form conclusions.  In the forum of peer-reviewed research journals, researchers can then publish findings that contribute to the quest for the cure.

	In solving the challenge, Garvan relied on an exceptional RDBMS platform to provide support for more than 30 separate databases.  The results have enabled researchers to easily extract data, draw conclusions, and author articles, bolstering efforts to improve the quality of life for the afflicted.  The novel ability to marry laboratory and patient data has also contributed to research in the international medical community, but Australians are the primary beneficiaries. Collaboration with national hospitals and institutions has strengthened the healthcare system, improved quality of life, and brought prestige to an entire continent. The battle against disease continues, but thanks to Garvans own ingenuity there is one more invaluable weapon to bring into the fray.

In 2008 Garvan Scientists produced 180 publications, namely, peer-reviewed articles in international journals, but also including book chapters, conference proceedings, letters, and educational articles.  Each respective journal has an impact factor which is used to measure its importance in a specific research field.  In 2008 Garvan achieved an average impact factor of slightly below eight for the top 80% of its publications, well above the international benchmark.
For seven years preceding 2008 its researchers averaged a remarkable 132 publications a year.  To say that the amount of research supporting such a world-class record--both in number and journal impact--is vast, would be an understatement.

	Presently, the increasing volume of information at scientists fingertips is extraordinary. From rates of clinical information gathering, to finer details collected against each participant, to giga-bytes of data produced by gene-chip experiments or microscope images, todays researchers require sophisticated analytical and information management tools. Acknowledging this fact, Garvans IT team searched for a platform on which to build a long-term solution.  As the institute sought ways to improve how its researchers compile and analyze data, it soon realized that the proper design of an information architecture was the linchpin in achieving system optimization.  
	
	A platform was needed that would merge all key components including patient information such as symptoms, demographics, surgery results, and after-surgery results, with collected lab research.  Sets of information had to reside in a single database and provide the ability to easily write stored procedures accessing combined information for researcher queries.  The database had to allow for the simple creation of query systems researchers could access on their own and it was critical that the systems, in particular the ability to extract key information to elucidate novel findings, be delivered directly to them. The process would prove most efficient if researchers were able to pose questions without queuing for time with an IT specialist.  Being able to acquire data, pose questions, and receive immediate results would encourage a new level of independent productivity.

	To aid in its endeavor Garvan turned to Sybase Adaptive Server Enterprise (ASE). Since deploying the new project Garvan has developed 25 separate production and 12 test databases all on one server. By combining specific patient information with lab research data, the databases have made it possible to easily create stored procedures that merge relevant information. Using these query responses, researchers can draw more reliable conclusions such as the relationship between gene product activity and patient survival.  ASEs ease of use is also of significant importance.  Users interact with databases via an in-house application with which they are able to build queries and once built, they can be e-mailed to other researchers in order to share and discuss results. The system lets researchers easily browse information and provides for a more thorough examination of the data.

	Accessing data under the protection of a rich security model protects private patient information while allowing access to more nondescript records.  The query builder, written in Realbasic programming language, easily accesses information when connected to the RDBMS and researchers simply use a point-and-click tool to query tables. Depending on the role defined for the user, patient identity is withheld and only research-oriented information such as gender and date of birth is presented. The results of queries can then be pasted into customized statistical tools. The fundamental integration of meta-data allows for the same query tool to operate against all databases including cancer, osteoporosis, or pituitary and neurodegenerative disease.  Simply put, the technology allows Garvan researchers to be more autonomous in their work, a measure of great importance when managing the explosion of research.

Yes

Yes

If not for the creation of an exceptional information architecture to support the work of Garvan researchers, a great disservice would have befallen the field of medicine.  With all the valuable and private data the Institute oversees, a lack of organization is unacceptable.  Instead of hampering researchers efforts, Garvan maintained its standing as a leader in medical research by deploying a vital project.  The technology provided has contributed directly to the indispensable duty of gathering data, analyzing it, and producing results.  With each publication, scientists benefit the cause of improving the quality of life for the national and international community.

	The implementation of a new information architecture has opened up many bonus opportunities for researchers as well.  Because of how well the project has performed, user-demand for query services has grown steadily over the years. With queries becoming more sophisticated as researchers realize the flexibility of the system, the IT team has recently upgraded the current server and is considering adding a second.  As the architecture allows room for growth, the databases offer support to combat the volatility of research collection.

	The fluctuating nature of medical research promotes the movement of staff between various institutes. Researchers average 1-2 years at Garvan to conduct their targeted studies so it is imperative that information be retained and made available to subsequent researchers. Since the longevity of research collection frequently outlasts the tenure of the researchers, well documented and well modeled databases are essential.  Since the implementation of Sybase ASE and Garvans new information architecture, the data server has been upgraded four times.  With each upgrade users have gained performance without sacrificing the information architectures key features and the subsequent data that is so precious to their work.

	The efficiency of Garvans new information architecture has largely contributed to the maintenance of a lean IT staff.  By enabling researchers to better manipulate data, the institute now uses just four full-time equivalents to manage the server and database applications. The flexibility provided by the new architecture has meant increased agility for researchers, as they are more equipped to handle complex comparisons. Having the advantage of retrieving gene data regarding a patient with breast cancer, along with her history of symptoms, treatment, and outcome, is a task that can now be fulfilled by the researcher without requesting help from IT.  The expedition of this process provides opportunities for scientists to gather data more quickly, analyze and discuss treatment outcomes with colleagues, and publish articles at an increased rate.

Beyond the projects initial focus, the success of the RDBMS and information architecture has spread to unexpected areas.  A prime example is Garvans prostate cancer information system. A database developed 7 years ago has since become a useful tool in analyzing how a particular gene product impacts prostate tissue.  The database has made for easy comparisons between how patients respond to chemotherapy versus radiation with the hope that results will lead to a more accurate treatment plan with the least amount of unfortunate side effects.

	In 2004 the Australian national research funding body offered funds to prostate cancer researchers across Australia to encourage pooling for prostate cancer cases. The resulting body, the Australian Prostate Cancer Collaboration (APCC) chose to implement Garvans prostate cancer information solution (CanSto). The implementation of a single solution, with a common data model, allowed for the pooling of cases from all nodes in the collaboration to add statistical power to the research effort.  Associate Professor David Horsfall, national project manager for the APPC BioResource relates his story:

	Five years ago, when I was appointed as Project Manager to the Australian Prostate Cancer BioResource, I was tasked with identifying and deploying a suitable database to underpin this national tissue bank. I originally thought this was going to be quite difficult, but to my surprise the IT group at the Garvan Institute had developed an in-house system (CanSto) that was functioning very well for their various institutional tissue banks, including the prostate bank. To the credit of the Garvan IT group, they rolled this system out to the other prostate cancer banking nodes in Adelaide, Brisbane and Melbourne, and performed whatever operational software configuration work was necessary, including site visits and training seminars. This system has been of immense value to the Australian Prostate Cancer BioResource as it has engendered a tangible commonality between the nodes of our virtual tissue bank. The Garvan Query Application (GQA) database interrogation tool and the monthly accrual reports have been particularly useful for banking administration. Over the last 12-18 months, our banking network has progressed from solely tissue accrual to a phase of accrual and distribution. The integration of patient clinicopathological data and the associated sample type and storage coordinates makes the identification and retrieval of tissue samples with predetermined histological features, or from patients from a selected disease outcome grouping, a particularly straightforward procedure. The Australian Prostate Cancer BioResource is now continuing to provide Australia's prostate cancer researchers with quality tissues to support the national translation of laboratory results into clinical practice. The quality support received by the Australian Prostate Cancer BioResource from the Garvan Institute IT group has been critical to us achieving our goals.

	Another field that has profited from Garvans solution concerns the bone disease osteoporosis.  Professor John Eisman, director of the Bone Program at Garvan explains: 

The FunnyBone architecture and the Garvan Query Application (GQA) have provided an essential infrastructure for the millions of data points that are critical for the Dubbo Osteoporosis Epidemiology Study. The structure allows us to extract specific data sets from the more than 3000 people collected over more than 20 years. 

	 Regardless of the large amount of patient data over such a long period of time, the pairing of GQA with Funny Bone proved a worthy application:  

GQA allows precise selections of unique data sets for specific questions. These extractions underpin the analyses that have allowed the high productivity of the Dubbo Osteoporosis Epidemiology Study and its international recognition. FunnyBone and GQA have been essential for the translation of these findings into publications in highly ranked international peer-reviewed scientific journals. Says Eisman.

Garvans research includes the collection of osteoporosis information from a dedicated clinic in country New South Wales. The information systems deployed into the clinic run against the same database that scientists in Sydney analyze. When a patient visits the clinic, their details are entered into the database and made immediately available to researchers in Sydney.  This tight interaction is achieved through the implementation of the clinical and research interfaces against a single, well-modeled database. Questions such as the cost of the osteoporosis disease to a community, the relationship between bone density to muscle strength, and even the propensity of a person to fracture a bone during a fall can all be assessed and the results disseminated through publication in research journals. 

	By collecting blood from each participant every two years, Garvan is also able to conduct genetic research to understand the relationship of genes to osteoporosis, which can lead to possible therapies and interventions that mitigate the disease or reduce its impact. Garvan has researched genes, gene products and the onset of disease across more than 4,000 participants.  During each visit, the institute collects approximately 600 data items that are loaded into the database.  With the mountain of research garnered from the osteoporosis study, an online software program was successfully created for public use that acts as a predictor of the likelihood to incur a fracture due to the disease.  It is a rare case indeed when scientists are able to supply valuable findings about a widespread disease directly to those who suffer from it.  No less remarkable is the fact that the information is provided free of charge in a format easily understandable to the general public.  

When Garvan began work over 10 years ago on the early aspects of a new information architecture, it did not wholly envisage where it would eventually lead. Initially, only a meta-data driven query tool was deployed. Funding was and remains the key constraint and at the early stages, not even one FTE was available to start work. Garvan was able, however, to foster solutions over time by demonstrating that each year more value was being delivered.  Slowly, the applications team grew. A sound architecture was crucial to this growth as was an industrial-strength RDBMS and server solution that could continue to evolve. Organic, evolutionary growth was key. At no point could significant funds and resources be brought to bear a redesign phase or significant changes in direction of the deployment platform.

	Further challenge came in the form of the heterogeneous nature of researchers computers. The majority of Garvan staffers use Apple Macintosh computers, yet all information systems delivered as part of the information architecture are deployed on both Mac and PC clients. The selection of the REALbasic interactive development environment allowed the link to occur easily. REALbasic builds for both Mac and PC platforms from a single project file; therefore a widespread technical problem was avoided.

The approach Garvan took, indeed the approach it was forced to take, meant it could not seek significant funds from the outset. The hardware deployed was initially a second-hand server donated to Garvan. The deployment of Sybase ASE as the RDBMS was only possible because Sybase also donated the product as part of its commitment to health research and delivery.  Only after some years of demonstrating a significant value to science did Garvan seek funds to purchase current server hardware and staff permanent analyst and programmer positions.

	To overcome any potential resistance to the projects direction, Garvan has continually sought to add value beyond the simple management of information.  Rather than extend all efforts towards one inflexible solution, it has adopted a malleable approach.  For example, the development of a general-purpose labeling solution has been applicable across many applications. The technique allows the programmer to simply plug in a component to print labels for blood and tissue samples for storage in -20C, -80C, and liquid nitrogen. The attraction of the added value ensures that Garvans solutions remain in demand and funding remains as secure as any research finding.

Exceeded

Garvans project was designed to address needs stemming from researchers. As such, the new information architecture already supports research objectives at Garvan as well as nine external research centers (QUT, Hanson Institute, Monash, POWMRI, RPA, RNA, Concord Hospital, St George Hospital and Charles Gardner Hospital). The architecture is also fundamental to a number of solutions that support the administration of science at Garvan. For example, global collaborations are set up and documented through a web-based material transfer agreement system; computer assets and their authorized access to the network are managed through a web-based solution and internal cost recovery is handled through solutions deployed on the same platform.  The project has reached and exceeded its primary goals no doubt because of the agility provided by the proper RDBMS tool. 

	The multi-disciplinary nature of Garvan's research has made it a cornerstone in Australias medical community and major collaborative programs with national and international institutions have placed an emphasis on establishing links with hospitals and industry. The constant interaction between scientists of differing expertise has led to exciting collaborations and publications that share the goal of exploring the interplay of different body systems and organs in the disease process, with a view to developing cures. Since the deployment of a researcher-based initiative, the global medical research community has felt the effects as a new method for patient and research data collection and analysis continues to positively affect millions.  The struggle towards successful diagnosis and treatment of disease is sure to outlast the researchers themselves, but their costly and essential body of work is effectively safeguarded for future innovations.

The first application Garvan deployed was a new solution for tracking patients details and blood and tissue samples for the Clinical Facility. The application was immediately embraced by all users. Garvan then cut over its Osteoporosis research Clinic and research group from an old Clinical Reporting System (CRS) solution to a new in-house solution built on the early information architecture. Again, the solution was immediately heralded by both Clinic and research staff. Other more recent applications boast over 300 users, both internal and external, that log up to 1000 distinct sessions each week.  This information provides evidence of how the benefits of Garvans drive for systemic organization trickle down from researchers, to external users, to those who suffer from disease directly.  As quickly as Garvan staff members have embraced new research innovations, so will patients embrace evolving healthcare techniques that keep them healthy.