DEVELOPMENT OF A SOFTWARE QUALITY PLAN
FOR HOSPITALS
Vispi Shroff, Louise Reid, Sajid Hashmi, Gerard Mulligan, Daniel Sheehy
Keyang Xiang and Ita Richardson
Department of Computer Science and Information Systems, University of Limerick, Limerick, Ireland
Keywords: Hospital software quality plan, Health care software quality.
Abstract: This paper describes research into the development of a quality plan for the management of software in an
Irish Hospital. It studies relevant standards, models and legal acts. Synergies between the Irish Health
Service Executive’s Quality and Risk Management Standard and the Capability Maturity Model Integration
are utilised to build and study a quality plan. While exploring the possibility of utilising software
engineering quality standards to improve the quality standards within health care, this has also led to a
greater understanding of the interlinked issues within a hospital.
1 INTRODUCTION
The development of high-quality software is an issue
of great and growing importance throughout the
software industry (Gillies 1992). Medical software
in particular, which includes medical devices with
embedded software, can have a major impact on the
delivery of patient care. A good quality software
system accompanied by poor managerial practices
cannot provide the required quality of service.
Hospitals face challenges in terms of managing
software due to size, complexity of practices,
parallel management and resistance to change.
Clinicians are heavily dependent on information
to inform decision making on the management of
patients. Of concern in the Irish setting is the lack of
an integrated electronic patient record system.
Without quality software in place, there is little hope
of patients benefiting “both directly and indirectly
from improved data quality since accurate clinical
data are a prerequisite for high standards of care and
monitoring.” (Forster et al. 2008).
The accuracy of data, and consequently, of
information, is determined by the quality of the
software systems which produce that data. Within
Irish Hospitals there are a number of major and
minor systems in place. Patients, clinicians, nursing,
information technology, administration, data entry
personnel, researchers, governing bodies, and
external auditors, have different expectations,
understanding and requirements. The increasing use
of fourth generation databases such as Microsoft
Access is allowing staff with little expertise or
knowledge of data quality techniques to gather data
that is used in the crucial decision making processes
along the patient journey.
Current research yields few solutions to the
software quality issue. On reviewing the many
papers assessing the quality of clinical data there
appears to be a lack of recognition of correctness
and accuracy of data – which can be provided
through the provision of quality software systems.
Our aim is to address these issues by means of
some practical solutions. This paper proposes to
deliver an implementation of a software quality plan
for Irish hospitals through the use of recognised
healthcare and software quality models and
standards.
1.1 Research Methodology
A literature review of the major databases of
journals from 2000 was performed. This gave us the
opportunity to investigate quality issues (Brooks
1987) (Fitzgerald et al. 1979) within hospitals.
Interviews with hospital clinicians were also carried
out to understand the processes involved in the
provision of various inpatient services. Also,
existing standards for medical devices and health
record collection were investigated.
587
Shroff V., Reid L., Hashmi S., Mulligan G., Sheehy D., Xiang K. and Richardson I..
DEVELOPMENT OF A SOFTWARE QUALITY PLAN FOR HOSPITALS .
DOI: 10.5220/0003172305870591
In Proceedings of the International Conference on Health Informatics (HEALTHINF-2011), pages 587-591
ISBN: 978-989-8425-34-8
Copyright
c
2011 SCITEPRESS (Science and Technology Publications, Lda.)
Comparing this output with existing software
standards, we established which software
development processes and practices were relevant
to the hospital situation.
For the purposes of implementation, we utilised
the Quality and Risk Management Standard (HSE
2007) issued by the Irish Health Service Executive,
also known as the HSE. The HSE is an executive
body with a mandate to run all the public health
services in the Republic of Ireland. This standard
was mapped against the best practices in selected
processes areas in the Capability Maturity Model
Integration, also known as CMMI for Development
(SEI 2006). The Data Protection Act 1998 and 2003,
IEEE Std. 730-2002 - IEEE Standard for Software
Quality Assurance Plans, EUROSOCAP (The
European Standards on Confidentiality and Privacy
in Healthcare) and the Health Insurance Portability
and Accountability Act (HIPAA) of 1996 were also
taken into account.
1.2 Quality & Risk Management
Standard
The QRMS Standard includes a statement of
standard and supporting criteria supported by an
internal control model (See Figure 1). There are 3
levels to assess compliance with the standard.
Level 1- The service has approved
documentation which describes the process for
managing quality and risk.
Level 2- The service can demonstrate
implementation of the approved
documentation.
Level 3- The service can demonstrate that
there are processes in place to monitor the overall
effectiveness of the approved documentation.
1.3 Capability Maturity Model
Integrated (CMMI)
CMMI is a model based on a collection of best
practices that assist organisations in the
improvement of their software processes. This
model was chosen as it is open and freely available
for use. It is widely adopted and validated by
numerous companies and across various different
industries. While it is heavily used in the software
industry, there are also case studies available on the
implementation of CMMI in the health care services
industry (Forrester 2008).
CMMI was found to be the Model that mapped
most successfully to the Irish Health Service’s
Quality and Risk Management Standard.
Levels 1 and 2 in the QRMS map to the
Managed & Defined Levels in CMMI respectively.
Level 3 in the QRMS looks at the control and
monitoring of the effectiveness of the processes
involved. The Quantitatively Managed and
Optimising levels in CMMI facilitate that control
and drive towards continuous improvement and
optimisation.
This intuitive mapping demonstrates the
complimentary nature of standards and models and
creates a compelling argument for implementation
together.
2 QUALITY MANAGEMENT
PLAN
Based on the literature review and interviews of
clinicians within the hospitals, the following areas of
concern were identified. These areas of concern are
then addressed with the QRMS and CMMI models
guiding the activities to be carried out.
2.1 Project Planning
Within CMMI, the Project Planning and Project
Monitoring and Control process areas bring a degree
of structure to the hospital software quality plan.
(Cooke-Davies 2002) outlined the importance of
project and operations management working
together to as a critical factor to achieving project
success. When developing the hospital quality plan,
following a review of CMMI processes, we
developed goals, standards and practices. We also
developed a system to constantly review and action
the findings of the plan.
The HSE provided the following statement of
standard which was adopted for this quality plan
“Healthcare quality and risk are effectively managed
through the implementation of an integrated quality
and risk management system that ensures continuous
quality improvement” (HSE 2007, p.5).
Within a hospital quality plan, each software
system, its type, interfaces, stakeholders and
requirements must be defined. A communication
strategy for stakeholders must be developed. The
potential risks associated with each s system must be
identified and rated according to severity and
likelihood. A plan generated and implemented to test
each system for Integrity and Availability, the
use/under use and fit for use of each system must
continually be assessed and improved.
HEALTHINF 2011 - International Conference on Health Informatics
588
2.2 Risk Management
The use of medical devices and health information
systems in a hospital are an inherent risk to the
patient. Storing patient information on these systems
further exacerbates the level of this risk factor
(Sicotte et al. 2006). Failure of software in these
systems/devices can have potentially catastrophic
effects, leading to injury of patients or even death
(Burton et al. 2006).
The complexity of software has long been
considered a critical IT project risk factor (Sicotte et
al. 2006). Risk Management must be an integral part
of software project management processes and
include proactive risk assessment and reactive
incident management to avoid incidents recurring.
(Kavaler & Speigel 2003) define risk management
as “an organized effort to identify, assess and reduce
where appropriate, risk to patients, visitors, staff and
organizational assets”. These risks can be wide
ranging from scheduling and timing risks to
personnel management risks. The hospital’s
software development team can cope with these
classes of software risks by applying appropriate
systematic risk management activities to the
software development process (Galin 2004).
The CMMI divides Risk Management into three
main activities - defining a risk management
strategy, identify and analysing potential risks and
managing and mitigating risks which do take place.
Its purpose is to identify potential problems before
they occur so that risk-handling activities can be
planned and invoked as needed across the life of the
project to mitigate adverse impacts on achieving
objectives (Dhlamini et al. 2009).
Risk management for software and systems is
not just part of a software quality plan, but should be
an integral part of the overall risk management plan
for the services which hospitals provide.
2.3 Patient Data Security
Storing patient data in electronic form raises
concerns about Patient Privacy and Data Security
(Haak et al. 2003). To comply with regulations,
software systems and medical devices must
guarantee adequate protection of the confidentiality
integrity and availability of patient information. The
framework outlined in this paper assists a hospital to
adhere to the HIPAA Standard, EUROSOCAP and
the Data Protection Acts of 1998 and 2003.
However, having complete protection of patient data
in practice may not be feasible or plausible as it may
inhibit the doctor’s work (Anderson 1996).
There have been many published breaches of
patient information in the Irish press that highlight
some major security breaches with regard to medical
information (O'Hora 2010). These breaches include
unauthorized secondary use of patient records,
disclosure of patient records by hackers and
commercial vendors, and use of patient records by
employers (Baumer et al. 2000). Complying with
data security law and regulations is a difficult
challenge for hospital managers. Hospital workers
demand more and easier access to patient
information in order to provide the best care to their
patients. Also, vendors of healthcare software use
words like flexible, easy-to-use, accessible,
streamlined, and multidisciplinary to promote their
products. This is at odds with principles of data
security which talk about privacy and confidentiality
(Waldo 1999).
2.4 Integrity & Availability
Quality data can be defined as data which is fit for
use or purpose (Bertoni et al. 2009), (de Lusignan
2006). In addition, (Welzer et al. 2002) state that
quality data must be accurate, available, have
integrity, consistency, timeliness and completeness.
In some cases no data is better than inaccurate data.
(Welzer et al. 2002) found that data needs to be
modeled correctly first and then be correct, adequate
and available. It is then necessary to be quality
validated again.
(Bertoni et al. 2009) and (Berndt et al. 2001)
recognized the importance of placing emphasis on
data analysis when designing a database. This is also
backed up in the research by (Treweek & Flottorp
2001) pointing to the fact that it is natural that
stakeholders would like to make use of available
information but that “a major problem, however, is
simply getting at the data
2.5 Verification and Validation
Verification and Validation are the processes which
determine “whether or not products of a given
development or maintenance activity conform to the
requirement of that activity and whether or not the
final software product fulfils its intended purpose
and meets user requirements” (Abran et al. 2004).
Our research demonstrates that a software quality
plan for a hospital requires that each system is
audited regularly to ensure accurate and complete
data. It is inevitable that errors or misunderstandings
between data providers and database managers will
occur in a highly specialized area. Actions must be
DEVELOPMENT OF A SOFTWARE QUALITY PLAN FOR HOSPITALS
589
put in place to continually improve both the
timeliness and accuracy of the reports. The data
managers must receive feedback to ensure
continuous quality improvement.
Verification and validation are two distinct
process areas within the CMMI-Dev model. On a
reactive level, validation through CMMI processes
ensures that software and systems currently in use
actually do perform the activities they were
originally intended for. On a proactive level,
validation for new systems and software can take
place at all levels of creation of the software and
systems, be it at the time of design, implementation
or verification.
Verification is the other side of the coin. Having
ensured that the software or system being built is the
right system, verification ensures that what is being
built actually meets the requirements originally set
out for the software or system.
3 DISCUSSION
During the course of our research we found that
there are many problem areas within hospitals which
must be addressed to enhance the quality of service.
The root causes of many problems lie in the absence
or lack of enforcement of standards, legislation and
management processes. Additionally, software
systems are often not seen as being governed by the
existing health standards. For such use, health
standards need to be modified to take software
standards into account. Existing research on quality
in health care has focused on single issues - little has
been done to consider quality of service as a whole.
In this research, we have tried to address the overall
quality issues in hospitals by focusing on software
systems and the definition of relevant quality
processes. These ultimately will support guidelines
that hospitals may follow to improve their quality of
service and avoid many hazards. For future research,
we recognise that the validation of this approach is
very important and we aim to investigate the
advantages and disadvantages of using our proposed
plan. We are interested in understanding how quality
of service is improved through the implementation
and improvement of existing software processes.
4 CONCLUSIONS
This study investigated the problems in relation to
software quality, to uncover the root causes of
information and data which lacks quality, and to
propose the suitable strategies to address the issues.
Lack of processes and management of systems and
software within the hospital environment results in
poor quality of service and makes the goal of patient
safety more difficult. Different standards, models,
and published literature exist at the moment.
However, it is impossible to use any single one to
solve the existing diverse and complicated problem.
Through literature review and interviews with
clinicians, we have identified software process
requirements for a hospital quality plan and we have
proposed some guidelines based on existing health
care standards, quality standards and published
resources. We could not address all problem areas in
detail; instead we emphasized those issues which
were reported by existing literature to be the more
critical ones.
Quality assurance is an ongoing process which
must be monitored and controlled. Quality assurance
and improvement is the responsibility of all
stakeholders and these must be involved in all
iterations of the quality cycle.
ACKNOWLEDGEMENTS
This work is partially supported by Science
Foundation Ireland grant 3/CE2/I303_1 to Lero.
This work is also partially supported by
TRANSFoRm. TRANSFoRm is partially funded by
the European Commission – DG INSFO (FP7
247787).
REFERENCES
Abran, A. et al. eds., 2004. SWEBOK - Guide to the
Software Engineering Body of Knowledge, IEEE
Computer Society. Available at: http://www.computer.
org/portal/web/swebok/html/contents [Accessed July
23, 2010].
Anderson, R. J., 1996. Security in clinical information
systems. Available at: http://citeseerx.ist.psu.edu/
viewdoc/download?doi=10.1.1.31.4380&rep=rep1&t
ype=pdf [Accessed July 22, 2010].
Baumer, D., Earp, J. B. & Payton, F. C., 2000. Privacy of
medical records: IT implications of HIPAA. SIGCAS
Comput. Soc., 30(4), pp.40–47.
Berndt, D. J. et al., 2001. Healthcare Data Warehousing
and Quality Assurance. IEEE Xplore, pp.56-65.
Bertoni, M. et al., 2009. A Case Study on the Analysis of
the Data Quality of a Large Medical Database. In
DEXA '09: Proceedings of the 2009 20th International
Workshop on Database and Expert Systems
HEALTHINF 2011 - International Conference on Health Informatics
590
Application. Washington, DC, USA: IEEE Computer
Society, pp. 308–312.
Brooks, J., 1987. No Silver Bullet Essence and Accidents
of Software Engineering. Computer, 20(4), pp.10–19.
Burton, J., McCaffery, F. & Richardson, I., 2006. A risk
management capability model for use in medical
device companies. In WoSQ '06: Proceedings of the
2006 international workshop on Software quality.
New York, NY, USA: ACM, pp. 3–8.
Cooke-Davies, T., 2002. The "real" success factors on
projects. International Journal of Project
Management, 20(3), pp.185 - 190.
Dhlamini, J., Nhamu, I. & Kaihepa, A., 2009. Intelligent
risk management tools for software development. In
SACLA '09: Proceedings of the 2009 Annual
Conference of the Southern African Computer
Lecturers' Association. New York, NY, USA: ACM, pp.
33–40.
Fitzgerald, M., Hanna, F. & Taylor, D., 1979. The use of a
microprocessor in routine cardiac assessment. Journal
of Medical Engineering Technology, 3(4), pp.175-80.
Forrester, E., 2008. CMMI for Services - Overview
Presentation. Available at: https://bscw.sei.cmu.edu/
pub/bscw.cgi/d683162/CMMI-SVC%20Health%20Ca
re%20Example%20v3.ppt [Accessed July 22, 2010].
Forster, M. et al., 2008. Electronic medical record
systems, data quality and loss to follow-up: survey of
antiretroviral therapy programmes in resource-limited
settings. Bulletin of the World Health Organization,
86(12), pp.939-947.
Galin, D., 2004. Software quality assurance: From Theory
to Implementation 1st ed., Essex, UK: Pearson
Education Limited.
Gillies, A. C., 1992. Software Quality: Theory and
Management, London, UK, UK: Chapman & Hall,
Ltd.
Haak, M. V. D. et al., 2003. Data security and protection
in cross-institutional electronic patient records.
International Journal of Medical Informatics, 70(2-3),
pp.117 - 130.
HSE, 2007. Quality & Risk Management Standard.
Available at: http://www.hse.ie/eng/About/Who/
OQR009_20080221_v3_Quality_and_Risk_Managem
ent_Standard.pdf [Accessed July 22, 2010].
Kavaler, F. & Speigel, A. D., 2003. Risk Management in
Health Care Institutions: A Strategic Approach 2nd
ed., London: Jones and Bartlett Publishers, Inc.
Available at: http://books.google.com/books?hl=en&
lr=&id=5gwsBRhEd70C&oi=fnd&pg=PR19&dq=Ris
k+management+in+health+care+institutions:+a+str
ategic+approach&ots=_ypn09BqTP&sig=2lC0a7dJ2
UC_xbb-9y_ynzAIabk#v=onepage&q&f=false
[Accessed July 22, 2010].
de Lusignan, S., 2006. The optimum granularity for
coding diagnostic data in primary care: Report of a
workshop of the EFMI Primary Care Informatics
Working Group at MIE 2005. Informatics in Primary
Care, 14(2), pp.133-137.
O'Hora, A., 2010. Watchdog warns HSE for repeated
security breaches of patient data. Irish Independant.
Available at: http://www.independent.ie/health/watch
dog-warns-hse-for-repeated-security-breaches-of-
patient-data-2131144.html [Accessed July 22, 2010].
SEI, 2006. CMMI for Development : Version 1.2.
Available at: http://www.sei.cmu.edu/downloads/
cmmi/CMMI-DEV-v1.2.doc [Accessed July 22, 2010].
Sicotte, C. et al., 2006. A Risk Assessment of Two
Interorganizational Clinical Information Systems.
Journal of the American Medical Informatics
Association, 13(5), pp.557- 566.
Treweek, S. & Flottorp, S., 2001. Using electronic medical
records to evaluate healthcare interventions. Health
Informatics Journal, 7(2), pp.96 -102.
Waldo, B. H., 1999. Managing Data Security: Developing
a Plan to Protect Patient Data. Nursing Economic$,
17(1), pp.49 - 52.
Welzer, T. et al., 2002. Medical Diagnostic and Data
Quality. In Proceedings of the 15th IEEE Symposium
on Computer-Based Medical Systems. 15th IEEE
Symposium on Computer-Based Medical Systems.
IEEE Xplore.
DEVELOPMENT OF A SOFTWARE QUALITY PLAN FOR HOSPITALS
591