Web-basedgeo-visualisationofspatialinformationtosupportevidence-basedhealthpolicy1.pdf
http://dx.doi.org/10.12826/18333575.2014.0004.Jardine http://dx.doi.org/10.12826/18333575.2014.0004.Jardine
Research
Web-based geo-visualisation of spatial information to support evidence-based health policy: a case study of the development process of HealthTracks
Andrew Jardine, Narelle Mullan, Ori Gudes, James Cosford, Simon Moncrieff, Geoff West, Jianguo Xiao, Grace Yun and Peter Someford
Introduction Providing analysts and decision-makers with timely access to reliable epidemiological, social demographic measures and environmental health measures at a high spatial resolution is a key priority to facilitate evidence-based decision-making (Brownson, Fielding & Maylahn 2009; Gudes et al. 2010; Jacobs et al. 2012). Australia is one of the few countries that have comprehensive, high quality datasets on many aspects of health measures, service utilisation and environmental health that contain spatial information (address or at least postcode) that can be geo-coded, such as deaths, hospitalisations, community mental health occasions of service, cancer incidence, commu- nicable disease notifications and air and water quality testing. However, the value of such datasets is only realised when they are used to inform evidence- based decision-making, policy development and research (Brownson, Fielding & Maylahn 2009). Visualisation of spatial health data offers a powerful tool for knowledge translation and decision support within health service agencies (Driedger et al. 2007; Schuurman, Leight & Berube 2008).
Mapping of health data has a long history in the form of static, pre-rendered maps and atlases (Cromley
2003; Evans & Sabel 2012). Traditionally, broad level government reports of individual datasets were routinely produced at specified time intervals (e.g. state level annual reports) and static maps generated by a limited number of analysts trained in the use of Geographical Information Systems (GIS) desktop software, thus wide-spread adoption of spatial infor- mation by decision-makers has generally been limited (Joyce 2009). However, with recent advances in web-mapping capabilities and the availability of online data, timely access to customised reports and maps of health information at a detailed spatial resolution can now be easily provided to health professionals and decision-makers (Croner 2003; Evans & Sabel 2012; Friedman, Parrish & Ross 2013; Gao et al. 2008; Studnicki & Fisher 2013). Online mapping and analyt- ical processing can provide a broad range of data and analytical tools for analysis, visualisation, modelling and planning without the need for expensive software and technical expertise (Kamadjeu & Tolentino 2006). Such interactive mapping tools have been demon- strated to be more useful and preferred by public health professionals over statics maps for comparing health outcomes and risk factors (Cinnamon et al. 2009).
Abstract Place is of critical importance to health as it can reveal patterns of disease spread and clustering, associations with risk factors, and areas with greatest need for, or least access to healthcare services and promotion activities. Furthermore, in order to get a good understanding of the health status and needs of a particular area a broad range of data are required which can often be difficult and time consuming to obtain and collate. This process has been expedited by bringing together multiple data sources and making them available in an online geo-visualisation, HealthTracks, which consists of a mapping and reporting component. The overall aim of the HealthTracks project is to make spatial health information more accessible to policymakers, analysts, planners and program managers to inform decision-making across the Department of Health Western Australia. Preliminary mapping and reporting applications that have been utilised to inform service planning, increased awareness of the utility of spatial information and improved efficiency in data access were developed. The future for HealthTracks involves expanding the range of data available and developing new analytical capabilities in order to work towards providing external agencies, researchers and eventually the general public access to rich local area spatial data.
(MeSH): Public Health Informatics; Evidence-Based Health Care; Access to Information; Geographic Mapping
HEALTH INFORMATION MANAGEMENT JOURNAL Vol 43 No 2 2014 ISSN 1833-3583 (PRINT) ISSN 1833-3575 (ONLINE) 7
http://dx.doi.org/10.12826/18333575.2014.0004.Jardine http://dx.doi.org/10.12826/18333575.2014.0004.Jardine
Research
Internationally, many online tools have been developed for specific disease outcomes or single healthcare services (Foley et al. 2010; MacEachren et al. 2008; Rinner et al. 2011; Shuai et al. 2006). However, the issues facing health and healthcare are complex and there is a growing need for tools that support the construction of knowledge from relevant, comprehensive, high quality and easy to interpret evidence to inform health policy (Bhowmick, Griffin et al. 2008; Dummer 2008; Gudes et al. 2010). The primary advantage of geospatial data is the ability to integrate, visualise and interrogate a wide range of data sources such as health outcomes, social demo- graphic measures, risk factor prevalence and health service utilisation. Furthermore, the role of small area, multidisciplinary local community health networks is increasing in Australia, as is the requirement for outcomes-based performance measurement at local public health unit level, thus intensifying the need for access to cross sectoral health, environment and demographic data at the local level (Baum et al. 2010; Gudes et al. 2010; Australian Governement, Medicare Locals 2013; National Health Performance Authority 2013; Studnicki et al. 2010).
In order to get a comprehensive understanding of the health status and needs of a particular area, a broad range of health, environmental, demographic and other information that can often be difficult and time consuming to obtain and collate, is required (Nutley & Reynolds 2013). Online geo-visualisation tools hold the promise of facilitating investigation of a broad range of large and complex datasets rapidly (Studnicki et al. 2010); for example, an investiga- tion into the varying contribution of community demographic composition within a defined region on multiple outcomes such as hospital separations, deaths, cancer incidence and communicable disease notifications. Other examples of the utility of spatial health data include disease surveillance and cluster detection; health service utilisation, needs, and resource allocation; investigation of health determi- nates and risk factor associations; and understanding inequalities in health outcomes and access to services (Boulos 2004; Cromley 2003; Gudes et al. 2010).
Insights from published evaluations of online health geospatial tools More than 35 state health departments in the United States have implemented a web-based data query system containing one or more health datasets (National Association for Public Health Statistics and Information Systems 2012). A review in 2005 showed that all systems allowed users to query data by geographic level (e.g. county) and about half
enabled users to produce static maps, but less than 20% included an interactive mapping component (Friedman & Parrish 2006). A more recent review of US-based query tools was not available at the time of development. A number of other interactive geospatial tools have been developed to enable interactive explo- ration of multivariate health data in map, chart or table format either as a stand-alone downloadable tool or online through a web browser. Examples described in the literature include: � GeoVISTA Studio • Exploratory Spatio-Temoral Analysis Toolkit
(ESTAT) (Robinson et al. 2005) • Visual Inquiry Toolkit (VIT) (Chen, Maceachren &
Guo 2008) • GeoVIZ Toolkit (Hardisty & Robinson 2011) • Visualisation System for Space-Time and
Multivariate Patters (VIS-STAMP) (Guo et al. 2006)
• Pennsylvania Cancer Atlas (Bhowmick, Griffin et al. 2008; Bhowmick, Robinson et al. 2008)
� Spatial Online Analytical Processing Visualization and Analysis Tool (SOVAT) (Scotch & Parmanto 2005, 2006)
� Adaptive Visualization for e-Science (ADVISES) (Sutcliffe et al. 2010)
� Epidemiologic Visual User Environment (EpiVue) (Yi et al. 2008). Important insights have been generated from
extensive evaluations of these tools. ADVISES was found to increase appreciation of, and facilitate creativity in, spatial data exploration among epidemi- ologists and health data analysts (Thew et al. 2011). SOVAT was demonstrated to be preferable and more efficient compared to using a combination of tradi- tional GIS and statistical software packages by public health professionals and students as it reduced the number of steps required during analysis and provided a more intuitive user interface (Scotch, Parmanto & Monaco 2008). Evaluation of ESTAT with end users determined that the most useful features were the highly interactive and flexible nature of the tool and the dynamic internal linkage between the interface components (bivariate map, scatterplot, parallel coor- dinate plot and time series animation slider tool) so that selection of an area on one highlights the corre- sponding areas on the others (Robinson 2007).
Bhowmick, Griffin et al. (2008) conducted in depth interviews with public health professionals working in cancer prevention and control and found large variation between the level of prior experience with spatial data and analysis tools and also the spatial scale and types of data required. Geo-visualisation software must therefore be simple and intuitive
8 HEALTH INFORMATION MANAGEMENT JOURNAL Vol 43 No 2 2014 ISSN 1833-3583 (PRINT) ISSN 1833-3575 (ONLINE)
http://dx.doi.org/10.12826/18333575.2014.0004.Jardine http://dx.doi.org/10.12826/18333575.2014.0004.Jardine
Research
enough for novice users, but also have extended func- tionality available to more advanced users who wish to explore and analyse the data in detail (Bhowmick, Griffin et al. 2008). They implemented a user centred design process to develop an interactive online cancer atlas and key design issues for the development of dynamic online health visualisation tools were iden- tified (Bhowmick, Robinson et al. 2008). Detailed tutorials demonstrating the application of the tool and context specific help documentation including both information of how to use the tool and defini- tions of key terms and concepts used were important. Even more important was the need for dynamically generated supplementary information in the form of explanatory text and tables to assist users to interpret the significance and key aspects of the data. Flexibility to query data at multiple geographies, across multiple variables and to export for later use in reports or pres- entations was also necessary (Bhowmick, Robinson et al. 2008).
This paper describes progress in the development of spatial visualisation and reporting tools in Western Australia and future plans to redevelop existing tools, building on current knowledge and best practice iden- tified in the published descriptions and evaluations of previous geo-visualisation tools discussed previously, to produce an integrated mapping and reporting tool that incorporates a broad range of health, environmental measures and demographic information.
Spatial visualisation and analysis tools in Western Australia Two applications with basic spatial functionality have been developed in collaboration between Curtin University of Technology and the Department of Health Western Australia (DoHWA) through the Cooperative Research Centre for Spatial Information (CRC-SI).
HealthTracks: Reporting is an epidemiological report generation tool and HealthTracks: Mapping is a web based mapping tool. The primary objective of these applications is to provide decision support tools to improve knowledge translation and utilisation for health decision-makers that leverages the power of spatial health information and provides instant access to a simple, yet valuable and innovative display of spatial health information.
HealthTracks: Reporting generates customised reports with informative epidemiological measures such as number of cases, age standardised rates, standardised rate ratios and associated 95% confi- dence intervals for a selected geographical area. An extensive range of datasets from a range of areas across the Health Department have been collated on a
central server and are updated at least annually (see Table 1). The data can be queried by health outcome, age, sex, Aboriginality and area from State to Health District level. All health outcome and population data tables are stored in a Microsoft SQL Server database and processed through an Active Server Page (ASP) web application based on the defined query param- eters. This provides users the ability to query the data according to their day-to-day needs and increase the efficiency by avoiding pre-processing pre-defined layers with multiple combinations.
Table 1: Datasets contained in the current version of HealthTracks: Reporting
SOURCE DATASET
WA Hospital Morbidity Data System
Hospitalisations by International Classification of Disease (IDC) coding
Hospitalisations by Major Diagnostic Category (MDC) and Diagnostic Related Group (DRG) coding
Hospitalisations by external cause coding
Drug and alcohol related hospitalisations
Potentially preventable hospitalisations
WA Register of Births, Marriages and Deaths
Deaths by ICD coding
Drug and alcohol related deaths, avoidable deaths
WA Cancer Registry Cancer incidence
WA Notifiable Infectious Diseases Database
Communicable disease notifications
WA Mental Health Information System
Mental health occasions of service
WA Health and Wellbeing Surveillance System
Self-reported health service utilisation, risk factor and chronic disease prevalence
WA Emergency Department Data Collection
Emergency Department presentations
Australian Bureau of Statistics
Estimated resident population by age, sex, Aboriginality and area
Report output includes figures, charts, notes and importantly, expert knowledge and context specific interpretative text. The text has been written by experienced epidemiologists and incorporated into the reporting tool to facilitate interpretation by end users. The majority of reports result from users interrogating a single dataset in detail. However, a general overview report can also be created which collates data from multiple sources about a specific region in order to cater for a range of user needs (Figure 1).
HEALTH INFORMATION MANAGEMENT JOURNAL Vol 43 No 2 2014 ISSN 1833-3583 (PRINT) ISSN 1833-3575 (ONLINE) 9
http://dx.doi.org/10.12826/18333575.2014.0004.Jardine http://dx.doi.org/10.12826/18333575.2014.0004.Jardine
Research
HealthTracks: Mapping is a client-side GIS based application that allows users to spatially visualise health and demographic information together with data from external agencies or by uploading their own data. Particular attention was given to the thematic display of health, demographic and population datasets. HealthTracks: Mapping incorporates web- mapping services (WMS) to access spatial data from external organisations or custodians. The functionality meets the need of ‘Where Is?’ queries from agency staff with the geographical location of health services, multiple administrative boundaries (e.g. health regions, health districts, local government areas, statis-
tical local areas, suburbs and postcodes) and other contextual information (e.g. fast food outlets, public drinking water resources). To allow users to visualise differences between the various datasets, a tool was included to rapidly transition between two layers of interest. A selection of 16 commonly requested health conditions and diseases were included with considera- tion given to those that had reliable rates for Statistical Local Areas (SLAs) - the chosen geographic area for display. For example, Figure 2 depicts the display for the number of births in the Perth Metropolitan area in 2008 which is demographic layer list in HealthTracks: Mapping.
South Metro Health Region Metro State
Childhood immunisation (% fully immunised, age calculated 30th of September 2012)
[1]
12 - <15 months 88.5 90 90
24 - <27 months 90.6 90.2 90.8
60 - <63 months 89.2 89.4 89.8
Immunisation Provider (% of total vaccinations, Q3 2012) [1]
GPs 72.8 70.5 63.5
Community Health Centres 11.5 8.6 17.4
Maternity data (2011) [2]
Age-specific birth rate (per 1,000 women aged 15-44 years, not having had Hysterectomy) 66.2 63.4 65.2
Teenage births (%) 4.4 3.5 4.3
Birth in women aged 35 years and older (%) 19.8 22.4 21
Cervical cancer screening [3]
Participation rate (%, 2009 - 2010) 58.3 60.6 59.7
Normal smear result (%, 2006 - 2010) 90.6 90.6 90.6
Notifiable diseases (per 100,000 persons*, 2007 - 2011) [4]
Enteric disease 145.7 155 169
Vector borne diseases 61.3 55.2 63.2
STIs 424.7 410.5 490.3
Vaccine preventable diseases [5]
163 161.8 172.6
Hospitalisation for accidental falls (per 100,000 persons*, 2007 - 2011) [6]
Children (0-4 years) 652.6 712.2 684.5
Elderly (65 years and older) 3,740.20 3,930.70 3,895.10
Youth suicides (per 100,000 persons*, 2001 - 2010) [7]
Males (15-24 years) 17 16.7 20.7
Females (15-24 years) 5 4.4 4.9
Cancer Incidence (per 100,000 persons*, 2007 - 2011) [8]
Lung cancer 44.9 43.3 43.6
Breast cancer (females) 115.3 115.8 114.1
Cervical cancer (females) 6.9 7.3 7.7
Prostate cancer (males) 187.1 184.2 179.4
Melanoma 44.9 44.4 46.4 Notes on the sources of data: 1. Australian Childhood Immunisation Register. Immunisation provider data contains age groups of 12-<15, 24-<27 and 72-<75 months
2. WA Midwives Notification System. Age specific hysterectomy fractions from the report Cervical Screening in Australia 2009-2010 (AIHW, 2012) are applied to derive
number of women aged 15-44 years without hysterectomy. 3. WA Cervical Cytology Registry. Data for women aged 20-69 years only. 4. WA Notifiable Disease Registry.
5. WA Hospital Morbidity Data System. 6. WA Hospital Morbidity Data System. 7. WA Death Registrations and the Australian Bureau of Statistics Mortality Database. 8.
WA Cancer Registry. * These rates have been age-standardised to the Australian 2001 population. SLA level, geocoded records have been used to estimate mortality
and cancer incidence rates.
Figure 1: Extract of HealthTracks: Reporting output showing summary statistics and interpretive text
generated from a query containing multiple data sources for South Metropolitan Health Region, Perth Metropolitan Area and Western Australia as a whole
10 HEALTH INFORMATION MANAGEMENT JOURNAL Vol 43 No 2 2014 ISSN 1833-3583 (PRINT) ISSN 1833-3575 (ONLINE)
http://dx.doi.org/10.12826/18333575.2014.0004.Jardine http://dx.doi.org/10.12826/18333575.2014.0004.Jardine
Research
Development and release The development of HealthTracks: Mapping involved a series of discussions, consultation meetings and testing by three stakeholder groups from the DoHWA. The first stakeholder group included experienced GIS staff, the second was executives from the Public Health Division and the third was end-users with a range of roles and levels of responsibility within the health system. GIS staff were able to examine a database of requests to their branch to inform the choice of datasets to include and executive staff gave insight into their expectations and health measures required. Users were consulted to prioritise the list of required functionalities and inform the interface design. Information collected from these consultation meetings was incorporated into a detailed specifications document to guide development. The process with users was then followed by three qualita- tive evaluation sessions in which users were observed interacting with early versions of the system and asked to undertake a series of tasks to gauge usability. Users were then asked to provide feedback that was used to improve the final version before release.
Notification of the availability of HealthTracks: Mapping was broadcast to DoHWA staff in August 2010. Word of mouth and smaller presenta-
tions provided more detailed demonstrations of HealthTracks: Mapping capabilities to user groups within DoHWA. An average of 2.7 unique users accessed the application per working day, primarily amongst those users who were previously familiar with the existing Geographic Information System services provided by DoHWA.
In contrast, HealthTracks: Reporting was designed to provide more efficient access to data that was commonly requested and which previously had to be manually extracted and analysed by Epidemiology Branch staff. It was initially developed by an expe- rienced epidemiologist and was refined over several years based on user feedback and suggestions. However, it was decommissioned in 2005 after the initial developer was no longer able to provide technical support. As part of the current project, the system architecture was updated in 2011 and user awareness and engagement was conducted over a six month period. This involved high-level discus- sions to obtain support at executive level, followed by presentations at meetings during which development priorities and requirements were gathered.
A passive dissemination process was used to facili- tate a staged re-release of HealthTracks: Reporting in
Figure 2: Snapshot of HealthTracks: Mapping output showing number of births in 2008 per postcode in the Perth Metropolitan Areas, Western Australia
HEALTH INFORMATION MANAGEMENT JOURNAL Vol 43 No 2 2014 ISSN 1833-3583 (PRINT) ISSN 1833-3575 (ONLINE) 11
http://dx.doi.org/10.12826/18333575.2014.0004.Jardine http://dx.doi.org/10.12826/18333575.2014.0004.Jardine
Research
June 2011. Firstly, managers were informed and asked to disseminate to relevant staff. This was followed by an email to all internal staff that had previously requested data. Finally, a number of groups within DoHWA requested further details, so information sessions were delivered to increase user engagement.
Uptake and impact Uptake of the reporting tool has been encouraging. From release at the end of June 2011 until the end of March 2013, 309 unique users within the Department registered to access the system and used it at least once. At total of over 18,000 reports were generated during this period (Figure 3).
An adapted version of the Missouri Information for Community Assessment (Intervention MICA) participant feedback survey published in Freidman and Parrish (2008) was used to evaluate end users experience with HealthTracks in April 2012. Users who had produced at least 10 reports (n=69) were invited to participate in the online survey, of whom 24 completed it (35% participation rate). Users were asked for feedback regarding a range of aspects including design, navigi- ation, tehnical elements and content using a five point Likert scale. Feedback was generally positive with 91% of respondents indicating they either agreed or strongly agreed that HealthTracks was straightforward to use. Furthermore, 96% indicated that the data presented was useful and 88% were satisfied with the tool overall. Some areas for improvement were also
identified. For example, only 70% of users agreed or strongly agreed that the tool was flexible enough to easily extract the required data and only 64% were able to easily extract data and import into an external spreadsheet, database or report. Qualitative practical examples of the impact the tools have had on accessing data to inform decision making are discussed in the case studies below.
Case Study 1: Primary health care needs analysis A rural General Practice Network requested a primary healthcare needs analysis to be compiled within a short timeframe to form part of their tender for Medicare Local funding and identify gaps and key areas of need to better inform and target primary health care services under the new reforms. This needs analysis required a broad range of data including population demographics, hospitalisations, mortality, cancer incidence, immunisation coverage, health service utilisation and health behaviour and risk factor prevalence. Previously, such a request would have taken many weeks to complete due to the complexity and broad range of data required. However, a large majority of the required data was obtained through the HealthTracks: Reporting application, and in conjunction with some further manually extracted data, the needs analysis was completed within the required timeframe.
Case Study 2: Regional Aboriginal health profiles to inform the Closing the Gap initiative A series of regional Aboriginal Health Profiles were produced to inform Aboriginal Regional Health Forums in prioritising and allocating funding under the Closing the Gap initiative. The profiles include popu- lation data, morbidity data and mortality data and were initially created before HealthTracks: Reporting was available, which was a fairly labour intensive exercise that required a working group of five analysts from who worked solely on the profiles for a number of weeks. The profiles were updated recently and HealthTracks: Reporting was used to source the majority of the epidemiological data required, reducing the amount of time required to prepare the profiles by approximately two thirds. Staff within the unit noted the value of having up to date epidemiolog- ical data readily available in a user friendly platform, describing HealthTracks: Reporting as ‘extremely bene- ficial’ and a ‘leap forward’ in efficient access to data to provide the evidence base for a variety of planning and decision making purposes and meeting tight reporting deadlines.
Case Study 3: Hospital service planning Providing planners and clinicians in Princess Margaret Hospital for Children, the principal paediatric hospital
100
150
200
250
300
350
6,000
8,000
10,000
12,000
14,000
16,000
18,000
20,000
N u
m b
er o
f u
se rs
N u
m b
er o
f re
p o
rt s
ge n
er at
ed
0
50
100
0
2,000
4,000
6,000
30 J
u n
1 1
31 A
u g
11
31 O
ct 1
1
31 D
ec 1
1
29 F
eb 1
2
30 A
p r
12
30 J
u n
1 2
31 A
u g
12
31 O
ct 1
2
31 D
ec 1
2
28 F
eb 1
3
Date
Reports Users
Figure 3: Number of HealthTracks: Reporting users registered and reports generated between 30 June 2011
and 31 March 2013
12 HEALTH INFORMATION MANAGEMENT JOURNAL Vol 43 No 2 2014 ISSN 1833-3583 (PRINT) ISSN 1833-3575 (ONLINE)
http://dx.doi.org/10.12826/18333575.2014.0004.Jardine http://dx.doi.org/10.12826/18333575.2014.0004.Jardine
Research
in Western Australia, with access to spatial informa- tion and the ability to map their own data has been valuable in trauma service planning. HealthTracks: Mapping was used to upload and display a subset of cases by postcode, which was then used as a decision support document in subsequent face-to-face stake- holder meetings by providing visual insight into where trauma patients were coming to the hospital from. Based on the preliminary investigation using HealthTracks: Mapping, further detailed spatial infor- mation was requested and used for trauma service planning and identifying areas of research interest that could be pursued further.
Case Study 4: Plan for tackling drug and alcohol problems in the population The Western Australia Drug and Alcohol Office used HealthTracks: Mapping to upload and visualise the geographical locations of drug and alcohol related service-centre calls from residents in the Perth metro- politan area. A series of maps were also created including socioeconomic index for area and projected population as well as rates of drug-related hospitalisa- tions, mortality, ambulance callouts, police reported drug-related offences, treatment episodes recorded by drug and alcohol services. These maps were visually compared and contrasted to identify the at-risk popu- lations and areas where services were most needed, to evaluate the association between socioeconomic status and drug-related health outcomes, and to estimate the future service need for different geographical areas. The service planning would not be achieved with ease without the use of HealthTracks: Mapping. The secure single-session data storage was an advantage for those users working with confidential data.
Conclusion The primary aim of the HealthTracks project is to provide a decision support tool to assist health policy- makers and service planners with evidence based planning and policy development. While researchers may also benefit through rapid access to data and analytical tools, it is important that all end users are aware that causal inferences cannot be drawn from ecological associations between spatial health outcome and risk factor data (Robinson 1950). Therefore, such applications can be used as a research tool for hypoth- esis generation but not hypothesis testing, and studies using individual level data and utilising high resolu- tion analysis are still required to investigate causal links (Subramanian et al. 2009; Wakefield & Shaddick 2006).
Privacy and reliability are the two most important considerations in the provision of spatial health data
that can be queried across multiple dimensions arising from the resultant small numbers in each cell or polygon (Boulos 2004; Rudolph, Shah & Love 2006; Scandol & Moore 2012). This is particularly significant in Western Australia where the population distribution is highly uneven, with almost 80% of the population living in the greater Perth metropolitan area (0.3% of the land area), while the population density in the rest of the state averages only 0.2 persons per square kilometre (Australian Bureau of Statistics 2011). Consistent with most current public health reporting standards, counts of less than five in a specific area are not reported due to privacy considerations and rates based on counts of less than 20 are deemed unreliable and suppressed (Rudolph, Shah & Love 2006).
Advances in information technology and software development have made it cost-effective to provide powerful and flexible mapping, reporting and analytic capabilities and the field continues to rapidly evolve (Highfield et al. 2011). Globally, a number of geo- visualisation tools have been developed for health data, and user testing has revealed important insights in published evaluations (Robinson 2007; Scotch, Parmanto & Monaco 2008; Thew et al. 2011). The release of the online mapping and reporting appli- cations has increased the uptake and use of spatial information in the DoHWA. However, they were always intended to be proof of concept applications and there is a large scope to further enhance usability, functionality and the range of data available.
HealthTracks is currently in the process of being rebuilt in order to integrate the mapping and reporting functions into a single analytical tool and make use of the latest software tools. Open-source software will be used given the wider variety of tools available compared with proprietary software, especially in the domain of lightweight client and server-sided solutions (Anderson & Moreno-Sanchez 2003; Moreno-Sanchez et al. 2007). User feedback and lessons learnt from the demonstration applications, along with the recom- mendations made in the published evaluations of other spatial tools reviewed previously will be incor- porated. The challenge of balancing privacy with the demand for more detailed spatial queries, in order to work towards facilitating access by external agencies, researchers and the general public, will be addressed by performing server-side analysis of individual-level data, with only aggregated information deployed to the client, provided the requested combination of dimensional cross-sections meets specified privacy and reliability criteria. Data will be stored in an Online Analytical Processing (OLAP) based data warehouse to minimise processing time and maximise data storage efficiency (Studnicki et al. 2010; Tremblay et al.
HEALTH INFORMATION MANAGEMENT JOURNAL Vol 43 No 2 2014 ISSN 1833-3583 (PRINT) ISSN 1833-3575 (ONLINE) 13
http://dx.doi.org/10.12826/18333575.2014.0004.Jardine http://dx.doi.org/10.12826/18333575.2014.0004.Jardine
Research
2007). Further details and technical specifications are described in Moncrieff et al. (2013). The rebuild will also address the issues identified in the user feedback survey and expand the range of data available to include socioeconomic and remoteness measures, burden of disease and injury, and environmental moni- toring in order to assist in identifying and addressing health inequalities and accessibility, emerging health issues and the ever-increasing demand for spatially based information for effective and timely health service planning and policy development (Granell, Fernández & Díaz 2013).
Acknowledgments The work has been supported by the Cooperative Research Centre for Spatial Information, whose activi- ties are funded by the Australian Commonwealth’s Cooperative Research Centres Programme. We also wish to acknowledge the Department of Health Western Australia for providing support and additional funding for this project.
References Anderson, G. and Moreno-Sanchez, R. (2003). Building
web-based spatial information solutions around Open Specifications and Open Source Software. Transactions in GIS 7(4): 447-466.
Australian Bureau of Statistics (2011). Regional population growth, 2009-10. Cat. no. 3218.0. Canberra, ABS. . Available at: http://www.abs.gov.au/AUSSTATS/abs@. nsf/allprimarymainfeatures/DBE284E4CA0D40BFCA2 579D0000BFEDE?opendocument (accessed 13 March 2014).
Australian Government, Medicare Locals (2013). Medicare Locals. Available at: http://www.medicarelocals.gov.au (accessed 24 September 2013).
Baum, S., Kendall, E., Muenchberger, H., Gudes, O. and Yigitcanlar, T. (2010). Geographical Information Systems: an effective planning and decision-making platform for community health coalitions in Australia. Health Information Management Journal 39(3): 28-33.
Bhowmick, T., Griffin, A.L., MacEachren, A.M., Kluhsman, B.C. and Lengerich, E.J. (2008). Informing geospatial toolset design: understanding the process of cancer data exploration and analysis. Health & Place 14(3): 576-607.
Bhowmick, T., Robinson, A.C., Gruver, A., MacEachren, A.M. and Lengerich, E.J. (2008). Distributed usability eval- uation of the Pennsylvania Cancer Atlas. International Journal of Health Geographics 7: 36.
Boulos, M.N. (2004). Towards evidence-based, GIS-driven national spatial health information infrastructure and surveillance services in the United Kingdom. International Journal of Health Geographics 3(1): 1.
Brownson, R.C., Fielding, J.E. and Maylahn, C.M. (2009). Evidence-based public health: a fundamental concept for public health practice. Annual Review of Public Health 30: 175-201.
Chen, J., Maceachren, A.M. and Guo, D. (2008). Supporting the process of exploring and interpreting space-time multivariate patterns: the Visual Inquiry Toolkit. Cartography and Geographic Information Science 35(1): 33-50.
Cinnamon, J., Rinner, C., Cusimano, M.D., Marshall, S., Bakele, T., Hernandez, T., Glazier, R.H. and Chipman, M.L. (2009). Evaluating web-based static, animated and interactive maps for injury prevention. Geospatial Health 4(1): 3-16.
Cromley, E.K. (2003). GIS and disease. Annual Review of Public Health 24: 7-24.
Croner, C.M. (2003). Public health, GIS, and the Internet. Annual Review of Public Health 24: 57-82.
Driedger, S.M., Kothari, A., Morrison, J., Sawada, M., Crighton, E.J. and Graham, I.D. (2007). Using participa- tory design to develop (public) health decision support systems through GIS. International Journal of Health Geographics 6: 53.
Dummer, T.J.B. (2008). Health geography: supporting public health policy and planning. Canadian Medical Association Journal 178(9): 1177-1180.
Evans, B. and Sabel, C. (2012). Open-Source Web-based Geographical Information System for health exposure assessment. International Journal of Health Geographics 11(1): 2.
Foley, D.H., Wilkerson, R.C., Birney, I., Harrison, S., Christensen, J. and Rueda, L.M. (2010). MosquitoMap and the Mal-area calculator: new web tools to relate mosquito species distribution with vector borne disease. International Journal of Health Geographics 9: 11.
Friedman, D.J. and Parrish, R.G. (2008). Issues in evalu- ating health department web-based data query systems: working papers. The Robert Wood Johnson Foundation. Available at: http://research.policyarchive.org/21435.pdf (accessed 25 January 2012).
Friedman, D.J. and Parrish, R.G.I. (2006). Characteristics, desired functionalities, and datasets of state web-based data query systems. Journal of Public Health Management and Practice 12(2): 119-129.
Friedman, D.J., Parrish, R.G.I. and Ross D.A. (2013). Electronic health records and US public health: current realities and future promise. American Journal of Public Health. 103(9): 1560-1567.
Gao, S., Mioc, D., Anton, F., Yi, X. and Coleman, D.J. (2008). Online GIS services for mapping and sharing disease information. International Journal of Health Geographics 7: 8.
Granell, C., Fernández, Ó.B. and Díaz, L. (2013). Geospatial information infrastructures to address spatial needs in health: collaboration, challenges and opportunities. Future Generation Computer Systems In press. DOI: 10.1016/j.future.2013.04.002.
14 HEALTH INFORMATION MANAGEMENT JOURNAL Vol 43 No 2 2014 ISSN 1833-3583 (PRINT) ISSN 1833-3575 (ONLINE)
http://dx.doi.org/10.12826/18333575.2014.0004.Jardine http://dx.doi.org/10.12826/18333575.2014.0004.Jardine
Research
Gudes, O., Kendall, E., Yigitcanlar, T., Pathak, V. and Baum, S. (2010). Rethinking health planning: a framework for organising information to underpin collaborative health planning. Health Information Management Journal 39(2): 18-29.
Guo, D., Chen, J., MacEachren, A.M. and Liao, K. (2006). A visualization system for space-time and multivariate patterns (VIS-STAMP). IEEE Transactions on Visualization and Computer Graphics 12(6): 1461-1474.
Hardisty, F. and Robinson, A.C. (2011). The GeoViz Toolkit: Using component-oriented coordination methods for geographic visualization and analysis. International Journal of Geographical Information Science: IJGIS 25(2): 191-210.
Highfield, L., Arthasarnprasit, J., Ottenweller, C. and Dasprez, A. (2011). Interactive web-based mapping: bridging technology and data for health. International Journal of Health Geographics 10(1): 69.
Jacobs, J.A., Jones, E., Gabella, B.A., Spring, B. and Brownson, R.C. (2012). Tools for implementing an evidence-based approach in public health practice. Preventing Chronic Disease 9: 110324.
Joyce, K. (2009). “To me it’s just another tool to help understand the evidence”: public health decision-makers’ perceptions of the value of geographical information systems (GIS). Health & Place 15(3): 801-810.
Kamadjeu, R. and Tolentino, H. (2006). Web-based public health geographic information systems for resources-con- strained environment using scalable vector graphics technology: a proof of concept applied to the expanded program on immunization data. International Journal of Health Geographics 5(1): 24.
MacEachren, A., Crawford, S., Akella, M. and Lengerich, G. (2008). Design and implementation of a model, web-based, GIS-enabled cancer atlas. The Cartographic Journal 45(4): 246-260.
Moncrieff, S., West, G., Cosford, J., Mullan, N. and Jardine, A. (2013). An open source, server-side framework for analytical web mapping and its application to health. International Journal of Digital Earth. In press. DOI: 10.1080/17538947.2013.786143
Moreno-Sanchez, R., Anderson, G., Cruz, J. and Hayden, M. (2007). The potential for the use of Open Source Software and Open Specifications in creating Web-based cross-border health spatial information systems. International Journal of Geographical Information Science 21(10): 1135-1163.
National Association for Public Health Statistics and Information Systems (2012). Web-based Data Query Systems (WDQS). National Association for Public Health Statistics and Information Systems, Silver Spring, MD. Available at: https://naphsis-web.sharepoint.com/Pages/ WebbasedDataQuerySystemsWDQS.aspx (accessed 25 January 2012).
National Health Performance Authority. (2013). Our work. Available at: http://www.nhpa.gov.au/internet/ nhpa/publishing.nsf/Content/Our-work (accessed 24 September 2013).
Nutley, T. and Reynolds, H.W. (2013). Improving the use of health data for health system strengthening. Global Health Action 6: 20001.
Rinner, C., Moldofsky, B., Cusimano, M.D., Marshall, S. and Hernandez, T. (2011). Exploring the boundaries of web map services: the example of the Online Injury Atlas for Ontario. Transactions in GIS 15(2): 129-145.
Robinson, A.C. (2007). A design framework for exploratory geovisualization in epidemiology. Information visualization 6(3): 197-214.
Robinson, A.C., Chen, J., Lengerich, E.J., Meyer, H.G. and Maceachren, A.M. (2005). Combining usability techniques to design geovisualization tools for epidemiology. Cartography and Geographic Information Science 32(4): 243-255.
Robinson, W.S. (1950). Ecological correlations and the behavior of individuals. American Sociological Review 15(3): 351-357.
Rudolph, B.A., Shah, G.H. and Love, D. (2006). Small numbers, disclosure risk, security, and reliability issues in Web-based data query systems. Journal of Public Health Management and Practice 12(2): 176-183.
Scandol, J.P. and Moore, H.A. (2012). Health Statistics NSW: getting the right balance between privacy and small numbers in a web-based reporting system. NSW Public Health Bulletin 23(1-2): 1-4.
Schuurman, N., Leight, M. and Berube, M. (2008). A Web-based graphical user interface for evidence-based decision making for health care allocations in rural areas. International Journal of Health Geographics 7: 49.
Scotch, M. and Parmanto, B. (2005). SOVAT: Spatial OLAP Visualization and Analysis Tool. Proceedings of the 38th Hawaii International Conference on System Sciences: 1-7.
Scotch, M. and Parmanto, B. (2006). Development of SOVAT: a numerical-spatial decision support system for community health assessment research. International Journal of Medical Informatics 75(10-11): 771-784.
Scotch, M., Parmanto, B. and Monaco, V. (2008). Evaluation of SOVAT: an OLAP-GIS decision support system for community health assessment data analysis. BMC Medical Informatics and Decision Making 8: 22.
Shuai, J., Buck, P., Sockett, P., Aramini, J. and Pollari, F. (2006). A GIS-driven integrated real-time surveillance pilot system for national West Nile virus dead bird surveillance in Canada. International Journal of Health Geographics 5: 17.
Studnicki, J., Fisher, J.W., Eichelberger, C., Bridger, C., Angelon-Gaetz, K. and Nelson, D. (2010). NC CATCH: Advancing public health analytics. Online Journal of Public Health Informatics 2(3). Available at: http:// ojphi.org/ojs/index.php/ojphi/article/view/3348/2754 (accessed 13 March 2014).
Studnicki, J. and Fisher J.W. (2013). Determining community health status priorities in a online analytic processing (OLAP) environment. Online Journal of Public Health Informatics 5(2): 210. Available at: http://ojphi.org/ojs/ index.php/ojphi/article/view/4634/3720 (accessed 13 March 2014).
HEALTH INFORMATION MANAGEMENT JOURNAL Vol 43 No 2 2014 ISSN 1833-3583 (PRINT) ISSN 1833-3575 (ONLINE) 15
http://dx.doi.org/10.12826/18333575.2014.0004.Jardine http://dx.doi.org/10.12826/18333575.2014.0004.Jardine
Research
Subramanian, S.V., Jones, K., Kaddour, A. and Krieger, N. (2009). Revisiting Robinson: the perils of individualistic and ecologic fallacy. International Journal of Epidemiology 38(2): 342-360; author reply 370-343.
Sutcliffe, A., Thew, S., De Bruijn, O., Buchan, I., Jarvis, P., McNaught, J. and Procter, R. (2010). User engagement by user-centred design in e-Health. Philosophical Transactions of the Royal Society A: Mathematical, Physical and Engineering Sciences 368(1926): 4209-4224.
Thew, S.L., Sutcliffe, A., de Bruijn, O., McNaught, J., Procter, R., Jarvis, P. and Buchan, I. (2011). Supporting creativity and appreciation of uncertainty in exploring geo-coded public health data. Methods of Information in Medicine 50(2): 158-165.
Tremblay, M.C., Fuller, R., Berndt, D. and Studnicki, J. (2007). Doing more with more information: changing healthcare planning with OLAP tools. Decision Support Systems 43(4): 1305-1320.
Wakefield, J. and Shaddick, G. (2006). Health-exposure modeling and the ecological fallacy. Biostatistics 7(3): 438-455.
Yi, Q., Hoskins, R., Hillringhouse, E., Sorensen, S., Oberle, M., Fuller, S. and Wallace, J. (2008). Integrating open- source technologies to build low-cost information systems for improved access to public health data. International Journal of Health Geographics 7(1): 29.
Corresponding author: Andrew Jardine, PhD, MAE Research Fellow CRC for Spatial Information Department of Spatial Sciences Curtin University GPO Box U1987 Perth WA 6845 email: [email protected]
Narelle Mullan, MSc Program Manager CRC for Spatial Information Department of Spatial Sciences Curtin University Perth WA 6845 email: [email protected]
Ori Gudes, PhD, MA Research Fellow CRC for Spatial Information Department of Spatial Sciences Curtin University Perth WA 6845 email: [email protected]
James Cosford, BSc Software Developer CRC for Spatial Information Department of Spatial Sciences Curtin University Perth WA 6845 email: [email protected]
Simon Moncrieff, PhD Senior Research Fellow CRC for Spatial Information Department of Spatial Sciences Curtin University Perth WA 6845 email: [email protected]
Geoff West, PhD Professor CRC for Spatial Information Department of Spatial Sciences Curtin University Perth Western Australia, 6845 email: [email protected]
Jianguo Xiao, PhD Epidemiologist, Epidemiology Branch Department of Health Western Australia East Perth WA 6004 email: [email protected]
Grace Yun, PhD Senior Research Analyst, Epidemiology Branch Department of Health Western Australia East Perth WA 6004 email: [email protected]
Peter Somerford, MSc Principal Epidemiologist, Epidemiology Branch Department of Health Western Australia East Perth WA 6004 email: [email protected]
16 HEALTH INFORMATION MANAGEMENT JOURNAL Vol 43 No 2 2014 ISSN 1833-3583 (PRINT) ISSN 1833-3575 (ONLINE)
Copyright of Health Information Management Journal is the property of Health Information Management Association of Australia Ltd. and its content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's express written permission. However, users may print, download, or email articles for individual use.
