Building the Free/Libre Open Source Health Care
(FLOSS-HC) Community: A Strategy for Pushing
Free/Libre Open Source in European Health Care
Thomas Karopka
IT Science Center Rügen gGmbH, Circus 14, 18581 Putbus, Germany
Abstract. Free/Libre Open Source Software (FLOSS) is both a process of
software development and a method of licensing. Although FLOSS plays a sig-
nificant role in several market areas, the impact in the health care arena is still
limited. This is mainly due to the special requirements and the different circum-
stances when compared to other market areas. However, FLOSS is thought to
be one of the most effective means to overcome the fragmentation in the health
care sector and provide a basis for more efficient, timely and cost effective
health care provision. In this position paper I will give a short overview of
FLOSS applications in the health care sector, I will briefly discuss the special
challenges and the identified barriers that need to be overcome to push FLOSS
in health care. I will then describe a possible strategy that may help in improv-
ing the situation for FLOSS in European health care.
1 Introduction
FLOSS is thought to be one of the driving forces in overcoming the fragmentation in
the health care market, which is one of the main efficiency barriers in the context of
ICT in health care. Other fields have already adopted the FLOSS business model to
some extend, e.g. Linux, MySQL, JBoss, Apache Tomcat or Mozilla Fire-
fox/Thunderbird are some popular examples of successful FLOSS applications. How-
ever, although quite successful in horizontal applications, i.e. that are useful in many
different industries, there is still limited success in vertical applications, i.e. applica-
tions that are specific to one single industry, such as the health care market. On the
other hand several governments are pushing open source in the public sector and the
European Commission’s initiative eEurope 2005 which builds on the Action Plan
(June 2000) sets the target “to promote the use of open source software in the public
sector and e-Government best practice through exchange of experiences across the
Union”. The European Union has set up the Open Source Observatory and Reposi-
tory for European public administrations (OSOR) (http://www.osor.eu) a site devoted
to the re-using, sharing and collaborative development of E-Government applications
for European public administrations. The site was official launched at the Open
Source World Conference in Malaga on October 20th 2008. So what about the health
care arena? In this paper I will give a non-comprehensive overview of FLOSS pro-
Karopka T. (2009).
Building the Free/Libre Open Source Health Care (FLOSS-HC) Community: A Strategy for Pushing Free/Libre Open Source in European Health Care.
In Proceedings of the 1st International Workshop on Open Source in European Health Care: The Time is Ripe, pages 3-7
DOI: 10.5220/0001828100030007
Copyright
c
SciTePress
jects in the health care sector, I will identify barriers for the adoption of FLOSS
strategies in health care and name challenges for the FLOSS community. The ideas
and theses in this paper where stimulated by discussions with various participants at
the EFMI special topic conference “Open Source in European Health Care” which
took place in September 2008 in London. My special thanks go to all participants of
this conference.
2 FLOSS Applications in Health Care
The “List of open source healthcare software” in Wikipedia [1] lists more than 120
different tools or projects in 19 categories. Some of the projects are apparently dead
or are not maintained for a long time. Other new initiatives like the “Open health
tools” project of the Eclipse foundation [2] or the newly founded Open eHealth
Foundation [3] are still missing. The largest category is the imaging/visualization
category with 39 applications followed by electronic health or medical records with
18 entries and medical practice management software with 12 entries.
If it comes to open source in healthcare often one of the most comprehensive and
most successful applications of FLOSS-HC, the VistA application [4], is mentioned.
This is an open source HIS application developed over several years by the federally
funded U.S. Veterans Administration. This software is further enhanced and adapted
for other healthcare systems under the WorldVista [5] project. As mentioned above,
hundreds of other tools have been developed with different levels of complexity and
maturity. To integrate this already huge amount of knowledge and person-years of
software development will be one of the top issues in the proposed strategy for Euro-
pean open source healthcare software.
3 Challenges and Barriers for FLOSS-HC
The health care sector provides special challenges for FLOSS applications. While a
large part of traditional FLOSS applications is either in the professional IT sector
(server software, operating system software i.e. Linux) or in the traditional office
sector (OpenOffice, Firefox, Thunderbird) where the end users are used to IT applica-
tions, the situation in health care is different. Hospital Information Systems are very
complex systems often comprising hundreds of different small applications. Each
sector within the hospital often has its own standards and tools. This has led to a
fragmentation in the IT sector which hinders an effective information exchange be-
tween sectors or other health professionals that are part in the care process. The situa-
tion in the practice management sector is comparable. For example in Germany there
are more then 100 different providers of practice management systems. This leads to
a similar situation like in the HIS sector. Additionally the end users, the health pro-
fessionals, are often not trained sufficiently in the use of IT applications. Often they
feel uncomfortable with IT applications and perceive them as barriers instead of sup-
port for their work. These are challenges for IT in health care in general. For FLOSS
applications there are several additional barriers. Some of them are:
4
Lack of professional support
No road map (Sustainability?)
IT administrators need someone to sue in case of failure
Functional gaps
Publish early – publish often not the right strategy in health care
Health care needs mission critical applications
At the top of the agenda is the lack of professional support. It is absolutely necessary
for hospitals to get professional support if something is not working as expected.
Often changes have to be made to adapt the system to local specialties of the hospital.
Another point is the lack of a roadmap. Hospitals need to have sustainable solutions
and the possibility to plan the future extension of the system. FLOSS projects are
often driven by single persons or groups. If these persons loose interest in the project
or move to other fields, the product may not be supported further. This is absolutely
inacceptable in the health care market. Functional gaps need to be closed. If this can
not be done by the hospital IT group, the vendor should be able to provide this sup-
port. In the case of FLOSS it can not be expected that the community fulfils the re-
quests of the hospital because the developers usually work on features that they are
interested in or on a pre-specified work package. The usual strategy in FLOSS
projects is “publish early – publish often”. The community is part of the development
process. The test phase is done by the clients in the field. This is also not acceptable
in health care. Last but not least the security and reliability demands in health care are
extremely high. It is often not clear to the management in health care whether they
can trust the software in respect to these demands.
4 Success Factors for FLOSS
In this section I will look at the success factors for FLOSS applications. Probably the
most successful FLOSS application is the operating system LINUX. One of the rea-
sons for this success is the inventor of LINUX, Linus Torvald. He still coordinates
and integrates the Linux kernel together with input from the community. Other suc-
cessful FLOSS projects have different organizational structures. In the case of the
Apache project and the Eclipse Project a Foundation provides the professional frame
for managing and coordinating the development of the different applications and
products. Also from the sister field of bioinformatics we have learnt that successful
projects need either a professional structure or it depends on the leading role and the
support of an individual or a group, e.g. Ensembl [6] or BioPAX [7].
5 A Strategy for FLOSS in European Health Care
In the US FLOSS applications have already gained some success in the health care
market. This is mainly due to the professionalization of the sector. Companies like
ClearHealth Inc. [8], MedSphere [9], Akaza Research [10], WebReach Inc. [11] or
Tolven [12] have adopted the FLOSS business model. One advantage of the US mar-
ket is the size and the number of potential customers. It is far easier to reach the criti-
5
cal number of customers. In Europe this is more difficult due to the smaller countries,
the different health care systems and the language barrier. Several systems have been
developed in a European context (e.g. Care2x [13], O3 [14]) but the use is still li-
mited due to the aforementioned barriers to the market. Despite the differences in the
healthcare system and the language barrier, it is possible to work on the basis of
common standards and modules as several existing projects are trying to implement.
To push FLOSS HC application development in a European context I propose the
adoption of the following strategy:
1. FLOSS-HC inventory
(What is already there?)
In this first step an inventory should be carried out to provide an overview of
what is already there. To date several projects start inventing the wheel again.
Although this might be desirable to some extend, it is more effective to try to
work together and reuse modules as often as possible. A very valuable re-
sources to achieve this is the Wikipedia list of open source software in health-
care [1].
2. FLOSS HC Communication platform and software repository
(Where can I find FLOSS HC applications and information about FLOSS-
HC?)
The effort from the first package should result in a communication platform
backed up by a database which contains links to the resources assembled in
the first package. The platform should provide means for communication like
a wiki and/or a forum for discussion about the projects and the road map. The
forum should be moderated. Otherwise it is not possible to work towards a
common aim.
3. FLOSS HC use case database
(What is really needed? What is the impact of the proposed package? How
realistic are the chances of success?)
The database of already existing software and projects should be comple-
mented by a database of use cases where the necessities from the field are
stored in form of use cases. These use cases and may be workflows should
describe the use cases from the health professional point of view. This is
probably one of the most important but also most difficult parts to achieve.
4. FLOSS HC knowledge base
(Description of modules and their capabilities)
A sheer database of FLOSS software is not sufficient. With technologies like
the Web Services Description Language (WSDL) or an ontology about
FLOSS-HC applications it is possible to describe the capabilities of the soft-
ware and issue complex queries for searches in the knowledge base. A good
example from the bioinformatics realm is the TAVERNA tool [15] which
may be used to build workflows based on the orchestration of services from
over 1.000 registered service applications.
To successfully apply the strategy outlined above it is necessary to have some kind of
professional corporate body. Probably the most appropriate would be the foundation
of a European Center for FLOSS in European Health Care. This non-for-profit public
corporate body should provide monetary support and also the manpower in form of a
steering team that is responsible for guiding the community process and if possible
6
also organizational work. The concentration on the European level does not exclude
projects and initiatives from non-European states but mainly provides a means for
European funding and a thematic focus on issues that are relevant in a European con-
text. Having said this it is clear, that the initiative is open for all interested people and
that it should not be an effort to reinvent the wheel. There should be an exchange
with already existing similar efforts. Finally the communication platform should use
Web 2.0 technologies to provide the best possible means for communication and
interaction for all stakeholders.
6 Conclusions
FLOSS has become a viable alternative to proprietary software development in sever-
al business areas. However, in the health care IT sector open source is still rare. To
improve the situation and chances for FOLSS in health care, a coordinated action of
all stakeholders in FLOSS-HC is needed. To make FLOSS a viable alternative to
proprietary software in the health care arena, several barriers need to be overcome.
The proposed strategy may be one way to accomplish this aim.
Acknowledgements
The author likes to thank the organizers of the EFMI special topic conference “Open
source in European health care” as well as all participants for an inspiring conference
which provided the basic inspiration for this position paper.
References
1. Wikipedia list of open source healthcare software. http://en.wikipedia.org/wiki/
List_of_open_source_healthcare_software. 2008
2. Open Health Tools. http://www.openhealthtools.org/. 2008
3. Open eHealth Foundation. http://www.openehealth.wikispaces.net/. 2008
4. VistA CPRS Demo Site. http://www1.va.gov/cprsdemo/. 2008
5. WorldVista. http://worldvista.org/. 2008
6. Ensembl. http://www.ensembl.org. 2008
7. BioPAX. http://www.biopax.org/. 2008
8. ClearHealth Inc. http://www.clear-health.com. 2008
9. MedSphere. http://www.medsphere.com/. 2008
10. Akaza Research. http://www.akazaresearch.com. 2008
11. WebReach Inc. http://www.webreachinc.com/. 2008
12. Tolven. http://www.tolven.org/. 2008
13. care2x. http://www.care2x.org/. 2008
14. Open Three (O3) Consortium. http://gnbts9.units.it. 2008
15. Oinn T, Greenwood M, Addis M et al. Taverna: lessons in creating a workflow environ-
ment for the life sciences: Research Articles. Concurr Comput: Pract Exper 2006;
18(10):1067-100.
7