Web-based Support for Population-based Medical Research

Presenting the QuON Survey System

F. A. Henskens

1,2

, D. J. Paul

1,2

, M. Wallis

, J. Bryant

, M. Carey

, E. Fradgley

, C. E. Koller

1,2

C. L. Paul

, R. W. Sanson-Fisher

and A. Zucca

Distributed Computing Research Group, University of Newcastle, NSW 2308, Callaghan, Australia

Health Behaviour Research Group, University of Newcastle, NSW 2308, Callaghan, Australia

Keywords: Survey Software, Web-based Surveys, Public Health, Health Research Support, eHealth.

Abstract: This paper discusses the needs of medical researchers working in the area of patient-centred medicine, in

particular their use of survey data in measuring patient opinions, needs, perceived quality of care received,

and priorities of health service interventions. Until quite recently, collection of survey data has been either

paper-based, or achieved using computer software that largely duplicated paper-based processes with

limited additional functionality. The authors investigate the use of web-based technology to support

collection of such data from patients, including experiences and observations on enhanced/additional

functionality made possible by its adoption. A novel software design termed QuON is presented, together

with examples of its capabilities and uses in current research projects.

1 INTRODUCTION

Public health research broadly focuses on measuring

health behaviours and evaluating the effectiveness of

innovations designed to improve the health of

groups of people. The scope of public health is

broad, covering for example: infectious diseases;

vaccination programs; lifestyle risk factors such as

tobacco smoking, diet, physical inactivity and

weight; and participation in cancer screening

programs. Detection and treatment of mental

illnesses such as depression and anxiety also falls

within this remit.

In 2001, the Institute of Medicine’s report

‘Crossing the Quality Chasm’ nominated patient

centredness as one of six domains of quality care

(Institute of Medicine, 2001). The inclusion of

patient-centred care in this seminal report signalled a

shift from traditionally physician-centred models of

care to patient-centred models that are structured

around patients’ physical, social and emotional

preferences, values, and experiences. This led to

changes, not only in the delivery of health care, but

also in the approach used in research. Patient centred

care emphasises the involvement of patients as

partners in their healthcare. Directly assessing

patient perspectives about whether healthcare meets

their needs using patient-reported surveys is

therefore considered best practice. Data collected via

survey can be used for a number of purposes,

including to: identify patient needs; measure

changes in patient outcomes; provide feedback to

healthcare systems about performance (e.g. as part

of quality assurance activities); and guide clinical

decision making.

Patient-centred health research often requires the

use of survey-based assessments to gather data on

health and health behaviours (McDowell, 2009). In

order to accurately capture a representative sample

of patients’ perspectives without expending

considerable resources, data collection must be

efficient with high levels of participant acceptability.

The production of high quality data requires that

sources of bias be limited by achieving high

response rates, minimising recall bias, obtaining

reliable and valid responses (which may involve

psychometric testing) and involving stakeholders in

development processes (von Elm et al., 2007). A key

element of both a patient-centred approach and

maximising research rigor is the minimisation of

participant burden in a manner that reduces

participant time (e.g. via flexibility and participant

tailoring), increases participant ease of use

(flexibility of format) and minimises the literacy

level required of participants.

To meet these requirements, a multidisciplinary

196

A. Henskens F., J. Paul D., Wallis M., Bryant J., Carey M., Fradgley E., E. Koller C., L. Paul C., W. Sanson-Fisher R. and Zucca A..

Web-based Support for Population-based Medical Research - Presenting the QuON Survey System.

DOI: 10.5220/0004738301960204

In Proceedings of the International Conference on Health Informatics (HEALTHINF-2014), pages 196-204

ISBN: 978-989-758-010-9

 2014 SCITEPRESS (Science and Technology Publications, Lda.)

collaboration between health behavioural scientists

and information technology experts was formed.

This collaboration has produced a web-based survey

package entitled QuON (Paul et al., 2013), which

includes specific features designed to achieve high

levels of participant acceptability.

2 SURVEYS AS A RESEARCH

TOOL

Survey-based measures need to be capable of

providing an accurate assessment of the behaviour or

health state of interest (McDowell, 2009). Measures

also need to be reliable, so that results are

reproducible. There are several factors that influence

measurement accuracy and reproducibility, which

need to be taken into account at the survey design

phase. These include the ease with which the

questions can be understood, and the acceptability

and relevance of the questions (Dillman et al., 2008).

The authors’ own experiences and comprehensive

discussions revealed a raft of qualities required of a

‘good’ patient survey. It is valid, reliable,

psychometrically robust, acceptable, easy to score,

and easy to analyse (Clinton-McHarg et al., 2010).

To improve patient acceptability, surveys should

be designed to be visually pleasing, interactive, and

contain highly relevant item content. Electronic data

collection strategies have a number of advantages

over pen-and-paper modes of survey administration,

as described below.

2.1 Tailoring to Improve Relevance

Paper and pencil surveys require respondents to

follow (sometimes complex) instructions in order to

ensure that only questions relevant to their

circumstance are answered. For example,

respondents may be asked to report on the amount of

physical activity they did in the past week. Those

who indicate that they did not engage in physical

activity may be asked to skip a set of questions

related to activity type.

Instructions of this type introduce a risk that

respondents will misread or misinterpret

instructions, potentially causing confusion for the

respondent, and compromising data quality. In

contrast, electronic surveys can be customised for

each participant. For example, questions can be

alternately skipped or presented depending on a

respondent’s answers to any previous question, not

just the response immediately prior. This not only

minimises respondent burden in reading irrelevant

questions, but also results in better data quality

through reduction in respondent error.

2.2 Immediate Feedback

The survey customisation features mentioned above

can also be used to create immediate feedback for a

patient and/or service provider, based on their

answers. Tailoring of information to individual

needs and preferences improves relevance and recall

of health information (McPherson et al., 2001).

Answers can be inserted into a feedback

template and recommendations can be added, based

on the answers given. A number of strategies to

enhance recall and understanding of information can

be incorporated into feedback to patients or

clinicians. These include explicit categorisation of

information (Girgis and Sanson-Fisher, 1998,

NHMRC, 2004), repetition of important units of

information (Ley et al., 1973), and the use of plain

language (Fallowfield and Jenkins, 2004).

The ability to provide immediate and tailored

feedback from survey output opens up opportunities

for survey data to become part of health care

delivery. Surveys completed before the appointment

with a clinician allow for a more focused

consultation, with potential issues flagged and

automatically documented. This streamlining

enhances the overall provision of service. Data

provided to the participant can form part of a health

intervention. Large data sets aggregated for health

services provide the opportunity to monitor patient-

level perceptions and experiences, which are central

to the provision of quality care.

For optimal data analysis, survey results should

be available to the researcher in a timely manner.

Outputs appropriate for use in popular statistical

analysis packages such as STATA (Stata

Corporation, 2013) or SAS (SAS Institute, 2013)

give immediate access to group data.

2.3 Improved Comprehension

Electronic surveys may incorporate customised

modes of presentation to enhance survey

comprehension. For example, larger fonts or audio

recordings of questions can be used to assist vision-

impaired persons; the language of the survey text

can be customised (e.g. from English to

Vietnamese); and pop-up boxes can be used to

explain difficult terms without interrupting the flow

of the survey.

Graphics and interactive features may also be,

Web-basedSupportforPopulation-basedMedicalResearch-PresentingtheQuONSurveySystem

197

useful particularly for conveying complex

information or additional material for groups with

lower levels of literacy (Murphy et al., 2000).

2.4 Improved Data Quality

The quality of data collected using paper-and-pencil

surveys can be compromised due to missing data

and errors in responses. These quality issues can be

largely overcome through use of electronic surveys.

Respondent errors can be minimised by

programming features that notify respondents if they

have provided an answer outside an expected range,

and prompts them to re-enter their answer (e.g.

postcodes, duration of illness, age). Similar features

can be used to prompt for completion of missed

items, thus reducing the amount of missing data

(Boneveski et al., 1999).

Electronic surveys preclude the need for manual

data entry (either item by item, or by scanning pages

into a data reader), thus eliminating a time-

consuming process and minimising data errors.

Automatic summaries of survey data can be

produced quickly and easily, and the ability to

export data from the survey system to statistical

programs allows complex statistical analysis to be

undertaken.

Paper and pencil survey responses need to be

manually logged into a database or spreadsheet to

keep track of completion rates, and to determine

whether reminder letters need to be sent to non-

responders. Completion of electronic surveys,

however, can be monitored electronically so that

reminder emails, SMS, or prompts for researchers to

make reminder phone calls can be sent

automatically. This is a labour efficient feature,

which assists in maximising completeness of data.

3 THE QuON SURVEY SYSTEM

The QuON survey software system was jointly

developed by the Distributed Computing Research

Group (DCRG), and the Health Behaviour Research

Group (HBRG), at the University of Newcastle,

Australia, and initially funded by the Australian

Government Department of Innovation, Industry,

Science and Research. The HBRG had previously

used PC-based systems such as Digivey (CREOSO

Corporation, 2013), and web-based systems such as

Survey Monkey (Survey Monkey, 2013), and while

useful, had found these systems to be increasingly

restrictive. The joint development of QuON grew

out of a need to facilitate data discovery in ANDS

(Australian National Data Service, 2013), and a

realisation that the growing sophistication of HBRG

research required access to a richer set of survey

question types and software functionality.

QuON enables researchers to build complex and

personalised survey questions by providing a large

tool box of item types and branching patterns. It also

allows innovative question types, such as interactive

point allocation exercises, that would be too

burdensome to complete using a traditional pen and

paper survey. Examples of the powerful features of

QuON are described in Table 1. These are a

combination of successful design features from other

survey systems and powerful customised new

additions developed by the QuON team.

Table 1: Features of the QuON Survey Software System.

Feature Description Benefit

Tailoring to improve survey relevance to end users and to minimise respondent burden

Complex branching

capabilities

Allows questions to be included or exclude

based on either raw answers, or calculations (e.g.

ody mass index (BMI)) based on answers t

ANY previous questions, not just the answer t

the question immediately preceding the branch.

This is illustrated in Figure 1.

Allows creation of surveys that are high

relevant for participants, reducing participa

urden and survey completion time. This

especially important if data are being collected

healthcare settings where patients might be call

into their consultation at any time and need mo

than one session to complete the survey.

Question

composition features

Allows dynamic surveys to be modelled based

on participants’ responses on three levels:

o The number of questions received by an

individual based on previous answers;

o Building the question stem containing

content from previous answers,

o The possible response options available for

selection.

Increases sophistication of survey items and

improves comprehension of survey items for

patients

HEALTHINF2014-InternationalConferenceonHealthInformatics

198

Table 1: Features of the QuON Survey Software System (cont.).

Immediate and tailored feedback of data

Real-time feedback

Generation of on-the-spot printed or screen-

based feedback to patients based on survey

answers

Allows data to be used not only for research

purposes and to provide feedback to the

healthcare system about performance, but for

tailored feedback of health information to

patients with the potential to directly affect

patient health outcomes.

Prioritisation

Allocation of points to prioritise desired changes

(rather than simple ranking exercises)

Provides information and specificity about the

relative weight of desired changes, not just a

rank order.

Design factors that effect survey comprehension

Specification of

what is displayed

on each screen

The ability to have more than one question on a

screen

Reduces the time that it takes to load new pages

for each item (Couper et al., 2001), and makes

the survey more user friendly for participants.

This feature is not exclusive to QuON; (e.g. it is

also included in Survey Monkey), but QuON

allows greater levels of customisation).

Graphics

The capability to insert still or moving picture

content to the text on survey screens

Improves comprehension and the overall appeal

of the survey. Also allows creation of screens

that are used exclusively for provision of

content, without an accompanying question

Definition of

permissible

responses

Allows the researcher to specify limits on the

number of digits entered (e.g. limiting postcode

to four digits and numbers between 1,000 and

8,000), and prompts that allow participants to re-

enter their answer. Also allows specification of

answer type, e.g. number vs. text.

Enhances data quality assurance (Dillman et al.,

2008).

Calendar functions

and auto pop-up

number pads

For answers that require a numerical response a

number pad, or calendar, appears on the screen.

Reduces the amount of “free-style” typing

required by participants, thereby increasing user

friendliness and improving data quality

Multiple survey

completion.

Participation in identified and authenticated

surveys can be set to ‘once only’ or ‘multiple

permitted’,

The ability, for example, for a patient and their

family/carer to provide linked input or to track a

change over time.

Privacy and confidentiality

Flexible survey

design to facilitate

anonymity and/or

confidentiality

The confidentiality of surveys can be specified

so results can be anonymous; linked to

individual participants; or linked to groups of

related participants.

Ethical and analytical benefits.

Study specific

participant lists

The researcher has the option of adding

participant details into the secure QuON server.

Participant lists (identified and authenticated)

are study specific and only visible to individual

researchers from that study.

Ethical and pragmatic reasons.

Restricted access

Access to survey results is restricted so that only

the ‘owner’ of a survey is permitted to access

them.

For ethical reasons.

Improving usability for researchers

Survey item re-use

The ability to copy items from previously

published surveys into a new survey, and to edit

them for re-use.

Increases ease of use for research groups,

particularly for questions that are routinely used

e.g. gender, date of birth..

Auto-calculate

study ID numbers

The ability to generate study numbers based on

respondents’ answers.

Allows for test-retest or sub-group follow-up.

Timing statistics

QuON can be asked to record timing data such

as time taken to complete the survey, as well as

time spent on each question, and number of

pauses and restarts.

Provides valuable information on patients’

behaviour while taking the survey, for each

item, as well as overall.

Web-basedSupportforPopulation-basedMedicalResearch-PresentingtheQuONSurveySystem

199

Figure 1: Illustration showing complex branching

capabilities of the QuON survey software.

3.1 Design and Implementation

QuON is a Web application, built using the

CakePHP platform (Cake Software Foundation,

2012), that communicates with a MySQL database

(Oracle, 2012) used to store both survey

definitions and answers collected for each survey.

It uses a typical Model-View-Control (MVC)

approach (Krasner and Pope, 1988) and utilises

Web standards, such as HTML5 and Semantic

markup (W3C, 2013) so that the same survey can

be presented on different devices such as

computers, tablets, and smart phones. The QuON

application executes on a central server;

researchers and participants are clients of the

application, and interact with QuON using a local

web browser.

There are five kinds of QuON users:

 Administrators, who set up system-wide

properties, create new users and groups, and

assign users to groups;

 Researchers, who define and modify

surveys;

 Survey Owners, who are researchers with

the special assigned right to download

result data for a particular survey;

 Group Administrators, who are researchers

with the additional permission to perform

administrative tasks only for the group to

which they belong. For example, a group

administrator can add new users to their

group, change owners of surveys, etc.;

 Participants, who access and answer the

surveys that have been published and

allocated to them by the researchers.

Participants are specified on a per-survey basis,

and can be either:

 Anonymous: participants are not identified

before starting the survey and the results

are stored anonymously;

 Identified: participants need to be pre-

registered in the system by the researchers

responsible for the survey, and must

provide their username before they can take

the survey;

 Auto-identified: participants are asked to

provide a name (e.g. James0911 – given

name concatenated with day and month of

birth) to identify themselves before taking

the survey, but the username does not need

to be pre-registered by the researcher(s)

responsible for the survey;

 Authenticated: participants need to be pre-

registered in the system by the researchers

responsible for the survey, and must

provide their username and password before

they can take the survey.

Each QuON survey comprises an ordered set of

individually defined survey objects. A survey

object can be:

 A question, which displays some stimulus

and requests a response from the

participant;

 A calculation, which evaluates an

expression that is potentially based on

previous answers. The result is stored but

not displayed by the calculation object, and

can be used later in the survey as part of a

displayed message, or as an input into

branching logic. An example would be the

calculation of Body Mass Index from

answers on height and weight, which will

be used to generate appropriated feedback;

 A branch object, which customises the

order in which survey objects appear to

individual participants by ‘jumping’ to a

different location in the survey, based on a

conditional expression typically involving

previous participant responses;

 An information object, which displays text,

possibly augmented by still pictures or

video, and does not require a participant’s

answer.

Different question types supported by QuON

include: Informational; Text; Checkbox; Radio

Button; Button Option; Drop Down; Calendar;

Rank Order; Distribution of Points; and Likert

Scale. There is also support for dynamic questions

such as: Dynamic Checkbox; Dynamic Rank

Order; Dynamic Distribution of Points, in which

the presented options are based on a subset of

answers provided to previous questions; and so-

HEALTHINF2014-InternationalConferenceonHealthInformatics

200

called meta questions that allow multiple questions

to be conditionally displayed on the one screen.

Figure 2: Meta-question and Dynamic Checkbox.

Figure 2 shows a dynamic checkbox question

type combined with a meta question type. This

presents a dynamic set of checkbox answers based

on previous answers, as well as multiple questions

for the user to complete on a single screen.

Further, programmers can easily add new

question types to QuON by creating new CakePHP

Helpers (Cake Software Foundation, 2012). These

define the attributes survey designers (researchers)

are allowed to specify, how the question should be

displayed, how responses should be validated, and

how responses are stored.

Several features embedded in QuON enable the

survey designer to review the construct and check

for errors. Tracking of complex branching is

facilitated in survey preview, when a Branch

Preview will appear at the location where the

branch has been inserted. This displays the rule

and what has been entered for the positive and

negative destinations, as well as the outcome

determined from the previous question:

For example, Figure 3 shows that the primary

cancer site was not 2=breast (Result: false), thus

the program will progress to the negative

destination, in this case the branch which tests if

prostate was selected. Surveys can also be

validated so errors can be easily identified and

remedied.

QuON makes it easy to include customised

Cascading Style Sheets (CSS) to change the

appearance of any question or information survey

object. It is also possible to specify different styles

for mobile and non-mobile devices. Customised

Javascript (Arnold and Gosling, 2000) is also

possible, allowing researchers to implement

custom client-side logic on the participants’

devices while they are taking a survey.

Figure 3: Branch Preview.

3.2 Conducting a Survey

The survey sequence is presented in the order in

which the survey objects are inserted into the

survey. Questions are processed by displaying the

question, and waiting for the participant to provide

a valid response. If an invalid response is entered,

the respondent is presented with an error message,

and asked to enter a new response. Once a valid

response is provided, the system moves to the next

survey object. Calculation objects evaluate an

expression, after inclusion of values from the

respondent’s previous answers, before control

moves to the next survey object. Branches evaluate

their researcher-defined logical condition and, if it

is true, cause a jump to the survey object specified

in the branch’s positive condition; otherwise they

cause a jump to the branch’s negative destination.

A participant may choose to leave the survey early;

if the participant is an identified or authenticated

user, the survey can be re-joined later, at which

time the system presents the participant with the

question they were viewing at the time of leaving

the previous survey session. This functionality

caters for timeouts caused if the participant forgets

or is unable to complete a survey in one setting.

Once a participant has completed a survey,

he/she can be presented with feedback sheets based

on the participant’s responses. The content of

feedback sheets is defined using easily modifiable

templates.

The Owner of a survey is able to view results

and timing data online, or to download the data as

a Comma-Separated-Value (CSV) file at any time.

Web-basedSupportforPopulation-basedMedicalResearch-PresentingtheQuONSurveySystem

201

Thus aggregated survey results are available

immediately, and not delayed by the transcription

or scanning necessary for paper-based surveys.

QuON also allows researchers to provide survey

metadata, in a form suitable for ingestion by

ReDBox (Queensland Cyber Infrastructure

Foundation, 2012) systems, resulting in production

of RIF-CS (Global Registries, 2013) discoverable

entries in the Australian National Data Service

(ANDS, 2013). The adoption of open standards

such as CSV and the RIF-CS formats ensures that

researchers are not locked into any specific system

for result analysis.

Figure 4 presents a high-level overview of the

standard QuON deployment. Connectivity between

participants and the QuON server will typically be

by Internet, and may be either wired or wireless

(including WiFi and 3G/4G/GPRS). The

deployment options are flexible, with multiple

database servers and web servers being supported

under the CakePHP framework.

During the design of the QuON system a

strong emphasis was placed on open standards and

re-usable components. Open software development

platforms such as PHP, combined with the MIT

license structure ensures that any user of the

system can extend the platform to their specific

needs. This extension may, for example, add a new

Helper that introduces a custom question type, or a

custom ‘branding’ that tailors the survey output for

a specific device or group of users.

Figure 4: High-level overview of a QuON deployment.

3.3 User Documentation

The development of specialised QuON software

features was driven by the requirements of experts

in health behaviour, but implemented by experts in

software engineering and development. This

process required close collaboration between the

scientists and the software developers, and

effective transfer of information between their

respective disciplines.

To ensure that non-experts can put the

capabilities of the developed system to its best use,

it was necessary to prepare detailed documentation

that described the required steps (and rationales) in

a way that can be easily followed. This was

achieved by logical organisation of the user

document and detailed, step-by-step instructions,

which were illustrated by corresponding

screenshots.

Addition of new features such as extra question

types, and other enhancements, required occasion

updates of the user manual. This involved creation

of a new version with the relevant changes made

throughout the document, as well as a short

summary of the changes in the ‘Change History’

section of the last chapter of the user manual.

Features such as defining branch conditions

and the creation of the feedback sheet template

require the use of expressions similar to those used

in programming languages; these proved to be

difficult for the researchers to understand and use.

Syntax information was usually provided in the

form of templates, with placeholders used to

represent the data required by the template. The

correct format for provision of such data was

described, allowing relatively easy transfer of the

syntax used in sample expressions into the specific

context required for each survey situation. Detailed

examples of desired outcome scenarios were

provided which included the syntax used to

achieve that outcome; an explanation of the syntax;

and the output produced. However, despite the

details provided, this aspect of survey design

presents the biggest challenge to the researcher.

4 EVALUATION

A functional and comparative evaluation of the

QuON system conducted by the Health Behaviour

Research Group (HBRG) at the University of

Newcastle. QuON was compared with the two

incumbent survey systems, Survey Monkey and

Digivey. The features presented in Table 1 were

used to produce the comparisons shown in Table 2.

Some evaluation results benefit from extra

explanation, which is provided in the list of caveats

that immediately follows Table 2.

HEALTHINF2014-InternationalConferenceonHealthInformatics

202

Table 2: Comparison of Features.

Feature QuON Di

ive

Surve

Monke

Complex branching

capabilities

Yes Partial (1) Yes

Question composition features Yes

Real-time feedbac

Yes

oYes

Prioritisation Yes

Specification of what is

displayed on each screen

Yes No Partial (2)

Graphics Yes Yes Yes

Definition of permissible

responses

Yes Yes Yes

Calendar functions and auto

op-up number pads

Yes No Yes

Multiple surve

completion. Yes Partial (3) Yes

Flexible survey design to

facilitate anonymity and/or

confidentialit

Yes Yes Yes

Stud

specific participant lists Yes

oYes

Restricted access Yes Partial (4) Yes

Surve

item re-use Yes

o Partial (5)

Auto-calculate study ID

numbers

Yes No No

Timin

statistics Yes

oYes

The following caveats apply to the above

comparison:

1. Digivey does support complex branching via

its “skip” and “branch” features, though

researchers found the two separate concepts

confusing and hard to work with in

comparison to QuON’s single “branch

object” approach.

2. Survey Monkey does permit definition of

multiple questions to be displayed on a screen

but the selection is static. There is no ability

to dynamically choose the displayed

questions on the basis of previous answers

3. Digivey supports multiple runs of a survey

but support for resuming a half-complete

survey is limited.

4. Digivey stores all survey answers on the local

PC, and unless the researcher chooses the

encryption option these are accessible by any

user of that PC.

5. Survey Monkey permits the export and

import of survey questions but does not

provide the survey or department level re-use

that QuON implements.

5 CONCLUSIONS

Researchers in patient-centred medicine often

gather data on health and health behaviours using

survey-based assessments. Accurate representation

of patients’ perspectives is more readily achieved

when data collection is efficient, and when

participants feel comfortable with the tools used to

elicit their opinions. Collected data is most useful

when it is available in a timely manner, and in a

form suitable for statistical analysis.

This paper presents QuON, a software system

that supports the definition and conduct of web-

based surveys. QuON provides a rich set of

question types, together with the ability to define

surveys that are tailored to the circumstance of

each individual participant. The system design

supports a high level of flexibility in survey and

participant administration, while ensuring

appropriate confidentiality of participant

responses. QuON has been specifically developed

to fill gaps in the existing incumbent offerings to

ensure surveys are as effective as possible a

research tool in the public health domain.

The QuON software architecture allows

programmers to easily add new question types.

Additionally, appropriately skilled researchers can

change the way question or information screens are

displayed, including different configurations for

mobile and non-mobile devices.

QuON continues to evolve to meet its users’

needs. The initial version may be downloaded

Web-basedSupportforPopulation-basedMedicalResearch-PresentingtheQuONSurveySystem

203

from http://code.google.com/p/quon/.

ACKNOWLEDGEMENTS

The work presented in this paper was funded by

grant DC17 from the Australian Government

Department of Innovation, Industry, Science and

Research under the ANDS scheme. Dr. Jamie

Bryant is supported by an Australian Research

Council Post-Doctoral Industry Fellowship.

REFERENCES

ANDS. 2013. Australian National Data Service

[Online]. Available: http://www.ands.org.au

[Accessed September 2013].

Arnold, K. & Gosling, J. 2000. The Java Programming

Language, 3rd ed., Addison Wesley.

Australian National Data Service. 2013. Better Data:

Better Research [Online]. Available:

http://www.ands.org.au/betterdata/index.html

[Accessed September 2013].

Boneveski, B., Sanson-Fisher, R., W., Campbell, E.,

Carruthers, A., Reid, A. L. A. & Ireland, M. 1999.

Randomized controlled trial of a computer strategy

to increase general practitioner preventive care. .

Preventive Medicine, 29(6), 478-486.

Cake Software Foundation 2012. CakePHP Cookbook,

Cake Software Foundation.

Clinton-Mcharg, T., Carey, M., Sanson-Fisher, R.,

Shakeshaft, A. & Rainbird, K. 2010. Measuring the

psychosocial health of adolescent and young adult

(AYA) cancer survivors: a critical review. Health &

Quality of Life Outcomes, 8(25).

Couper, M. P., Traugott, M. W. & Lamias, M. J. 2001.

Web Survey Design and Administration. Public

Opinion Quarterly, 65(2), 230-253.

Creoso Corporation. 2013. Available:

http://www.digivey.com [Accessed September

2013].

Dillman, D. A., Christian, L. M. & Smyth, J. D. 2008.

Internet, Mail, and Mixed-Mode Surveys: The

Tailored Design Method, John Wiley & Sons, Inc.

Fallowfield, L. & Jenkins, V. 2004. Communicating sad,

bad, and difficult news in medicine. . Lancet,

363(9405), 312-19.

Girgis, A. & Sanson-Fisher, R. W. 1998. Breaking bad

news. 1: Current best advice for clinicians.

Behavoural Medicine, 24(2), 53-59.

Global Registries. 2013. The Registry Interchange

Format - Collections and Services (RIF-CS)

[Online]. Available: http://www.globalregistries.org

[Accessed September 2013].

Institute of Medicine 2001. Crossing the quality chasm :

a new health system for the 21st century [Committee

on Quality of Health Care in America], National

Academies Press.

Krasner, G. E. & Pope, S. T. 1988. A cookbook for

using the model-view controller user interface

paradigm in Smalltalk-80. Journal of Object-

Oriented Programming, 1(3), 26-49.

Ley, P., Bradshaw, P. W., Eaves, D. & Walker, C. M.

1973. A method for increasing patients’ recall of

information presented to them. . Psychological

Medicine, 3, 217-20.

Mcdowell, I. 2009. Measuring Health. A guide to rating

scales and questionnaires.

, Oxford University Press.

Mcpherson, C. J., Higginson, I. J. & Hearn, J. 2001.

Effective methods of giving information in cancer: a

systematic literature review of randomized

controlled trials. Journal of Public Health Medicine,

23(3), 227-34.

Murphy, P. W., Chesson, A. L., Walker, L., Arnold, C.

L. & Chesson, L. M. 2000. Comparing the

effectiveness of video and written material for

improving knowledge among sleep disorders clinic

patients with limited literacy skills. Southern

Medical Journal, 93(3), 297-304.

NHMRC 2004. General Guidelines for Medical

Practitioners on Providing Information to Patients.

Oracle 2012. MySQL Reference Manual, Oracle.

Paul, D. J., Wallis, M., Henskens, F. A. & Nolan, K.

QuON - A Generic Platform for the Collation and

Sharing of Web Survey Data. Web Information

Systems and Technologies (WEBIST2013), 2013

Aachen, Germany. 111-116.

Queensland Cyber Infrastructure FOUNDATION. 2012.

ReDBox - Mint [Online]. Available: http://

www.redboxresearchdata.com.au [Accessed 11

January 2013].

SAS Institute. 2013. SAS: Business Analytics and

Business Intelligence Software [Online]. Available:

http://www.sas.com [Accessed September 2013].

Stata Corporation. 2013. Stata: Data Analysis and

Statistical Software [Online]. Available:

http://www.stata.com [Accessed September 2013].

Survey Monkey. 2013. Available: http://www

.surveymonkey.com [Accessed September 2013].

Von Elm, E., Altman, D. G., Egger, M., Pocock, S. J.,

Gøtzsche, P. C. & Vandenbroucke, J. P. 2007. The

Strengthening the Reporting of Observational

Studies in Epidemiology (STROBE) statement:

guidelines for reporting observational studies.

Bulletin of the World Health Organisation, 867-872.

W3C. 2013. A vocabulary and associated APIs for

HTML and XHTML - Editor's Draft [Online].

Available: http://www.w3.org/html/wg/drafts/

html/master/ [Accessed November 2013].

HEALTHINF2014-InternationalConferenceonHealthInformatics

204