A Characterization of Potential Adopters and Rejecters in German
Ambulatory Care
Sebastian Dünnebeil, Ali Sunyaev, Ivo Blohm
Department of Informatics, Technische Universität München, Germany
Jan Marco Leimeister
Department of Economics,Universität Kassel, Germany
Helmut Krcmar
Department of Informatics, Technische Universität München, Germany
Keywords: Ambulatory care, Electronic health services, Health telematics infrastructure, Technology adoption,
Standardization, Data security, Practice equipment, Clustering.
Abstract: Germany is introducing a nation-wide health telematics infrastructure that enables various electronic health
services (EHS). Little is known about the burdens and drivers for potential adoption of these innovations.
Based on a quantitative study among German physicians participating in pilot test regions for health
telematics, this paper clusters potential adopters and rejecters of EHS, based on their usage intention as
determined with the UTAUT model. The study furthermore depicts opinions, attitudes, as well as equipment
of physicians in ambulatory care to find similarities in terms of IT diffusion, process and security
standardization, patient involvement, communication, documentation and general working patterns. The
clustering shows that “Supporters” and “Rejecters” of EHS differ significantly in many aspects investigated.
Based on these empirical findings, implications for design and introduction of e-health services can be
derived, ranging from a different way of approaching physicians in ambulatory care to incentive structures
for EHS usage.
As health information systems have the potential to
improve Healthcare quality (Shekelle et al., 2006),
German health authorities are currently building a
nationwide telematics infrastructure (TI) in order to
harmonize transactions and data storage of e-health
applications in the German public health system.
Universal accessibility of data without institutional
boundaries, via web services or common portable
data carriers, aims at reducing healthcare costs by
avoiding redundant examination of patients and
administration (Bundesministerium für Gesundheit,
2005). Unified telematics specifications can ensure
high data security standards, a standardized access
process and common data formats throughout the
national health system (Sunyaev et al., 2008).
Misuse can be avoided by mandatory encryption of
health data and role based access rules for health
care providers (gematik, 2008b, German Federal
Office for Information Security, 2008).
A major goal of the efforts is the enforcement of
patient centered treatment (Marschollek and
Demirbilek, 2006), standardization and transparency
of medical treatment. In Germany the telematics
infrastructure is used as the backbone for the
mandatory electronic health card (eHC) system. The
infrastructure is specified by a government
controlled institution, and connects existing
information systems of care providers via a common
Dünnebeil S., Sunyaev A., Blohm I., Marco Leimeister J. and Krcmar H. (2010).
DO GERMAN PHYSICIANS WANT ELECTRONIC HEALTH SERVICES? - A Characterization of Potential Adopters and Rejecters in German
Ambulatory Care.
In Proceedings of the Third International Conference on Health Informatics, pages 202-209
DOI: 10.5220/0002695602020209
network with shared data storage locations
(Fraunhofer Institut, 2005). This service oriented
architecture (SOA) provides services: the primary
systems (e.g., clinical information systems or
practice information system) of medical institutions
can invoke them to communicate with other care
providers and maintain, review or share medical data
objects. A local component, called “Connector,”
encapsulates all local services, as encryption or card
access and establishes a secure virtual private
network (VPN) connection to the central services if
needed (gematik, 2008a).
So far a number of different basic offline health
services, such as electronic prescription, emergency
data, and insurance basic data service, have been
tested in seven regions to verify the functions of the
technical infrastructure deployed. In the long term,
further services like electronic health records,
electronic physician letters, and electronic drug
documentation should be offered online via the
infrastructure. The infrastructure will be kept open
for the development of commercial value-added
services (Bernnat, 2006). Evaluation of the online
functionality of electronic health services will be
started with around 200 care providers in the test
region, involving about 100,000 citizens with health
Previous surveys have shown that a vast majority of
physicians rejects electronic services offered by the
National TI (Techniker Krankenkasse, 2009, Oliver
Kalthoff et al., 2008). Numerous campaigns have
been started by medical associations and politicians,
calling for a moratorium of the eHC project (Tuffs,
2008). The main reasons for rejection, as indicated
by the objectors, are safety concerns about central
storage of patients’ medical data as well as the
unsatisfactory performance during the offline testing
of services and infrastructure (Oliver Kalthoff et al.,
2008). Moreover, testing and introduction of the
eHC have been delayed many times: although the
initial introduction date was set for 2006, the testing
has still not been completed in 2009. The date of the
final introduction is not yet clear.
In order to get a clear picture of the
characteristics of physicians in the ambulant sectors,
we analyzed the characteristics in depth of
physicians who, in contrast to care providers in
hospitals, decide independent of management on the
adoption of the TI and its services. Important fields
of medical work, intended to be improved by the
telematics, were the focus of the research. The goal
was to conclude whether rejection and support of the
telematics correlate with other aspects that are
important in daily work in ambulatory practices.
Therefore, the usage intention was measured directly
to get a clear picture of the general adoption
preferences. To scrutinize the publicly dispersed
criteria leading to rejection, we grouped physicians
according to their behavioral intention. The resulting
clusters were then analyzed in terms of the
remaining criteria to find out which characteristics
were common for the different adopter groups.
Consolidated findings, of the survey thus enabled
predictions about general attitude, working
characteristics, and equipment of physicians,
derivable from the usage intentions.
3.1 Survey Design
The survey was developed based on the Unified
Theory of Acceptance and Use of Technology
(UTAUT) to determine the usage intention for
electronic health services deployed via the national
TI for the support of ambulatory care (Venkatesh et
al., 2003). Based on medical goals documented by
leading institutions of the healthcare system and the
German government (Bundesrepublik Deutschland,
1988), we further asked for certain important
healthcare aspects, such as security concerns
(Sunyaev et al., 2009), treatment and administration
standards, patient involvement, and inter-
institutional communication (Hoppe and Richter-
Reichhelm, 2000, Haux, 2005), before posing the
questions about telematics and EHS. The question
sequence was designed to ensure minimum
influence of the public opinion on the answer
patterns not directly related to the telematics.
The questionnaire was named “Survey on IT
diffusion in ambulatory health care.” We
investigated the state of conventional Information
Technology in ambulatory practices and its usage by
physicians as well. The goal was to gain information
on the status quo of the testing region concerning the
aspects mentioned. Further, we were keen to find the
medical domain-related characteristics of medical
personnel and practices to be predictable when
measuring usage intention based on the UTAUT
and Rejecters in German Ambulatory Care
3.2 Sample Description
We distributed 500 questionnaires to all physicians
in ambulatory care of the Bavarian testing region
for healthcare telematics by mail, but also provided
the questionnaire on the Internet. The test region
has 452,000 inhabitants on 2,847 square kilometers.
The region is geographically well definable because
of its heterogeneous structure, which makes the
region suitable for field testing since it represents
the structure of Germany very well (ZTG Zentrum
für Telematik im Gesundheitswesen GmbH, 2009).
We received 117 responses, representing a response
rate of 23.2%.
Only 4 physicians used the online option to
submit their results, the remaining 113 returned it by
3.3 Clustering of the Respondents
The respondents were cluster analyzed according to
their usage intention. For this purpose a hierarchical
cluster analysis was performed, using ward’s
clustering algorithm and the squared Euclidean
distance as distance metric (Backhaus et al., 2006).
Usage intention was measured by the behavioral
intention scale as used in UTAUT by Venkatesh et
al. (Venkatesh et al., 2003). The rating scales were
Figure 1: Detailed Usage Intention.
slightly adapted to the given context (Table 1). They
ranged from 1 (intended adoption in less than 1 year)
to five (no intended adoption at all). As all variables
were measured on the same rating scales, there was
no need for normalization of the cluster variables.
Respondents that did not completely answer all three
items were excluded from the analysis. The total
mean reflects the combined opinion of all 117
4.1 Description of Clusters
The cluster analysis revealed two different clusters
that can unambiguously be interpreted as
“Supporters” (Cluster 1) (also referred to as
adopters/potential adopters) and “Rejecters” (Cluster
2) of telematics services. 49 physicians were
grouped into the supporters cluster and 29 into the
rejecters cluster. The remaining respondents were
excluded. This categorization is surprising; we
expected the group of supporters to be smaller, as
their opinion is merely perceived in the public
discussion of telematics services. The supporters
embrace the new technology and show a
significantly (p<0.001) higher intention, plan and
prediction of usage than the rejecters
(table 1).
Table 1: Usage Intention from electronic healthcare
services deployed by via the TI.
Description Supporters Rejecters
From 1 (In less than
1 year) to 5 (Never)
Do see the
benefits the TI
and are willing to
use the services
in the short term
Do not see the
benefits of the TI
and are not
willing to use the
I intend to use the
system ***
1.94 4.83
I plan to use the
system. ***
2.17 4.90
I predict I would
use the system ***
*p < .05; **p < .01; ***p < .001
No significant distinction could be observed in terms
of age, experience, or size of the practice (number of
physicians, number of patients). However,
significantly (p<0.05) more medical specialists are
grouped into the supporters’ cluster than general
HEALTHINF 2010 - International Conference on Health Informatics
Table 2: Characteristics of physicians and practices.
49.7 48.9 50.9
(1=Female, 2=Male)
1.70 1.67 1.73
Specialization *
(1=Family Doctor,
1.59 1.62 1.44
Number of
physicians working
in the practice
2.1 1.7 2.3
consultations per
day ***
46.5 40.3 61.7
Average time spent
on patient contact
per day
7.4 6.3 7.0
Patients visiting the
practice per day
86.1 79.9 99.5
Number of
employees in the
4.5 4.8 4.0
*p < .05; **p < .01; ***p < .001
Moreover, physicians with a significantly lower
frequency of patient consultations are more often
(p<0.001) supporters of the new TI services, as they
are likely to have more time to treat every single
patient, thus using the services. When interpreting
this data, it has to be taken into account that the field
of medical specialists is very diverse in Germany,
and can only be compared with caution.
Dermatologists and orthopedists declared to
medicate regularly more than 80 patients per
working day, whereas specialists, such as
anesthetists or pediatricians, covered only a range of
20 to 40 patients.
4.2 Current use of Information
Technology in Ambulatory
The results seen in Figure 2 show widespread use of
conventional IT Technology in ambulatory care. For
more than 84% of all physicians, the Internet is
currently a source for medical information. While
Internet consultations for medical content is a
commodity for both groups, the practice of Internet
access is only a commodity for adopters,
significantly less (p<0.001) so for the cluster of the
rejecters, where only about half are connected to the
Internet. Internet access is apparently used
significantly more (p<0.05) by the supporters for
email communication within the health system with
colleagues, other institutions and patients.
Further, medical data are already distributed
among medical institutions, mainly by the adopter
cluster, even though a common and secure electronic
communication platform is not yet in place. Hence,
the inter-institutional communication differs
significantly, while very few differences can be seen
concerning the electronic documentation within the
practices. Physician-patient contacts are documented
electronically by 78% of the physicians, and for
73%, even with electronic medical records (EMR).
All respondents reported having an information
system in their practice. Most supporters already use
electronic utilities to represent their practice on the
Internet, to exchange patients’ medical data and to
communicate within the public health system. We
conclude thus that the supporters see the telematics
as complementary offers for extending the electronic
functionalities, which most of them are already
The rejecter cluster is using IT within their
practices but lags behind significantly in terms of
Internet based communication matters in every
criterion measured. The rejecter cluster cannot be
characterized as opponents of IT, as they are using it
in their practices and have similar usage patterns for
the Internet concerning their personal skill
enhancement. The question why the internal usage is
common but the joint usage a matter of strong
resistance will be discussed in the following
4.3 Process Standardization
As one of the goals of the TI is the overcoming of
institutional boundaries within the public health
system, we asked for the current status of inter-
institutional communication during the treatment
processes. A majority of 56% of all respondents
reported having regular contact during treatments
and 33% at least partly. Only 7% stated being
isolated in inter-institutional treatments of patients.
A significant difference between the adopter cluster
and the laggards cluster could not be observed. As
the potential adopters often transmit information
electronically, it is likely that the rejecters still use
traditional communication matters for the
coordination of their treatments.
Both clusters do not significantly differ
concerning process documentation either. Practices
seem to document their treatment processes
extensively , as the experienced level of process
and Rejecters in German Ambulatory Care
Figure 2: Adoption and Usage of conventional IT by Physicians in Ambulatory Care.
documentation standard is very high. 82% claim
that the processes are for the most part documented.
For the adopter cluster, significantly more processes
are derived from their IT systems, whereas the
laggards cluster has a lower level of IT orientation.
Whether these processes are based on best practice
or evidence-based-medicine cannot be determined in
this context. The potential adopters also show
significantly (p<0.001) higher interests in cost-
benefit analysis for their practices, while cluster 2
mostly does not regard this as being useful.
4.4 Processes and Infrastructure of the
Regarding the German telematics and processes,
infrastructures, applications and security standards
of the EHC, both clusters show a very low
knowledge level. Very few physicians have dealt
with the characteristics of the infrastructure as the
security standards and necessary hardware
infrastructure affecting their practices. Also, the
workflows which affect the practices when already
specified electronic health services such as
electronic prescriptions are introduced are hardly
known by both groups. Personal experience with the
particular EHC systems can therewith not be the
driver of the broad rejection among the physicians.
Just a small fraction has visited the sample
installation which is available in the testing region in
order to study a sample of the systems to be
deployed in all medical facilities in Germany. The
adopter cluster shows slightly more involvement in
the telematics; however, significant differences
cannot be seen, which excludes the system
knowledge as a criterion behind the adoption
4.5 Perceived Value of IT,
Documentation, Communication
and Patient Involvement
Ambulatory care needs IT in order to efficiently run
a practice today; this can be almost be regarded as
common sense within both groups. More than 80%
agree or strongly agree that they cannot productively
run their businesses without IT support. More than
70% do not see IT simply as a supportive
instrument, but believe that the technology provides
an added value to their practice. The adopter group
experiences a significantly higher value of IT
(p<0.01), even though the majority of both groups
see the necessity of using IT-Systems. As our results
have shown, electronic communication is already
very common among the group of potential
The belief that electronic communication will
be common in the public health domain within 5
years seems to follow from the daily experience of
this group. Documentation, communication, and
treatment standards all show significantly higher
support (p<0.001) among the potential adopters. A
comprehensive medical documentation maintained
by all involved care providers is seen as being
essential by the adopter cluster, while the opposing
cluster is undecided. Patient involvement in medical
documentation shows similar tendencies and
significant differences between the two groups
HEALTHINF 2010 - International Conference on Health Informatics
Figure 3: Process standardization and eHC related knowledge.
observed. The rejecters by trend doubt the benefits
of process standardization for their practices, and
fear an administrative overhead. They tend towards
the opinion that the daily work in practice is too
diverse to adopt standards for most workflows.
Standardized documentation for medical treatments
generally has lower support than does standardized
administrative documentation. While the adopter
cluster is rather supportive, the laggards are
undecided or dismissive. Expectedly, the potential
adopters regard e-health as a significant
development in medical treatment. Both groups
seem to have similar information providers
concerning e-health offers. As the total mean is
lower than the mean of both clusters, physicians
with missing contacts for information seeking have
not specified any preference concerning potential
adoption of electronic health services. Both groups
are rather undecided concerning the information
duty for patients about e-health risks and chances of
e-health services. As the adopter cluster has a higher
level of IT usage, the trend to see physicians as
information providers is well understandable since
the understanding of the services is likely to be
higher in this group.
4.6 Data Security
Ambulatory physicians generally show a very high
need for data security. National standards for the
handling of patients’ data are equally supported as
committing guidelines for the individual practices.
For most care providers, it is important to inform
their patients about the usage of medical data. The
adopters generally see a higher need for security
standards for their own practice and for the public
health system in general. Both groups are sceptical
about the central storage of medical data, and have a
tendency towards decentralized storage in patients’
hands to ensure their data sovereignty in the long
term. While the adopting cluster fears delays in
treatment processes when abandoning centralized
storage concepts, the laggards are significantly less
concerned (p<0.001). The group of potential
adopters certify themselves as having significantly
higher ability to inform their patients about the
usage of their medical data compared to the
opposing group.
The results show that the group of physicians ready
to adopt EHS offered via the TI significantly differs
from the more dismissive group in many aspects.
Tendencies of the data security topic, dominating the
public opinion concerning telematics adoption in
Germany, do not differ when comparing the
potential adopters and rejecters telematics services.
Both groups strongly support nationwide unified
security standards for their own practices also.
and Rejecters in German Ambulatory Care
Figure 4: Standards, patients’ involvement, communication standards, e-health, and IT-diffusion.
Adopters favor even higher security standards than
do the rejecters, independent of EHS offers. As the
eHC system tries to ensure exactly the demanded
security standards, other factors are more likely to be
responsible for the adoption decision. The two
identified clusters differ significantly in terms of
General IT adoption, process and documentation
standardization, patient involvement, cost-benefit
aspects, their medical specialization and their patient
throughput. The rejecters are not just lagging behind
in terms of IT adoption, but also in many aspects of
standardization, cost awareness and transparency. It
is likely that they regard the EHS as a mean to
enforce the aspects mentioned and therefore are less
supportive. The higher patient throughput of the
rejecters and their low interest in cost analysis for
their practices indicates that they are benefiting
above average from the status quo. EHS can be
further seen as a utility for patient involvement and
shared medical documentation, which show similar
answer patterns.
The German healthcare system has a group of
very progressive care providers. Conventional IT
can be considered to be a commodity for this group
of physicians. They are open for standards and
patient involvement, and therefore are keen to utilize
telematics services for this purpose. As both clusters
claim to maintain contact during treatments with
other medical institutions, but only the supporters
utilize IT for this purpose, the difference in IT usage
should not result in treatment quality, but in
efficiency increase. To investigate this, further
studies should be done to measure the impact of IT
in ambulatory care on both efficiency and medical
HEALTHINF 2010 - International Conference on Health Informatics
Figure 5: Security Standards for Medical Data.
Backhaus, K., Erichson, B., Plinke, W. & Weiber, R.
(2006) Multivariate Analysemethoden: Eine
Anwendungsorientierte Einführung, Berlin, Springer.
Bernnat, R. (2006) Kosten-Nutzen-Analyse der
Einrichtung einer Telematik-Infrastruktur im
deutschen Gesundheitswesen. Booz Allen Hamilton
Bundesministerium Für Gesundheit (2005) The German
eHealth Strategy (Target and strategy, concept, legal
framework, activities/roll-out plan, costs and return of
investment, European perspective). Berlin/Bonn.
Bundesrepublik Deutschland (1988) Sozialgesetzbuch
(SGB) Fünftes Buch, Gesetzliche
Fraunhofer Institut (2005) Spezifikation der
Lösungsarchitektur zur Umsetzung der Awendungen
der elektronischen Gesundheitskarte. Fraunhofer,
Projektgruppe FuE-Projekt.
Gematik (2008a) Einführung der Gesundheitskarte -
Gesamtarchitektur. gematik GmbH.
Gematik (2008b) Übergreifendes Sicherheitskonzept der
Telematikinfrastruktur. Specification.
German Federal Office for Information Security (2008)
Common Criteria Protection Profile - electronic Health
Card (eHC)
Haux, R. (2005) Health information systems – past,
present, future. International Journal of Medical
Informatics, 75, 268-281.
Hoppe, J.-D. & Richter-Reichhelm, M. (2000)
Medizinische Über- , Unter- und Fehlversorgung in
Deutschland. Ärztliche Zentralstelle
Qualitätssicherung im Auftrag von
Bundesärztekammer und Kassenärztlicher
Marschollek, M. & Demirbilek, E. (2006) Providing
longitudinal health care information with the new
German Health Card - a pilot system to track patient
pathways. Computer Methods and Programs in
Biomedicine, 81, 266-271.
Oliver Kalthoff, O., Marsden, N., Kalthoff, S. & Drescher,
F. (2008) Abschlussbericht Evaluation der Einführung
der elektronischen Gesundheitskarte in der Testregion
Shekelle, P., Morton, S. C. & Keeler, E. B. (2006) Costs
and Benefits of Health Information Technology.
Evidence Report - Technology Assessment.
Sunyaev, A., Kaletsch, A., Mauro, C. & Krcmar, H.
(2009) Security Analysis of the German electronic
Health Card’s Peripheral Parts. Proceedings of the
11th International Conference on Enterprise
Information Systems.
Sunyaev, A., Leimeister, J. M., Schweiger, A. & Krcmar,
H. (2008) IT-Standards and Standardization
Approaches in Healthcare. IN WICKRAMASINGHE,
N. & GEISLER, E. (Eds.) Encyclopedia of Healthcare
Information Systems. Idea Group.
Techniker Krankenkasse (2009) Branchenbarometer E-
Health. F.A.Z. - Institut fuer Management-, Markt-
und Medieninformation, 1.
Tuffs, A. (2008) Germany plans to introduce electronic
health card Medical publication of the year.
Venkatesh, V., Morris, M., Davis, G. & Davis, F. (2003)
User Acceptance of Information Technology: Toward
a Unified View. MIS Quarterly, 27, 425-478.
Ztg Zentrum Für Telematik Im Gesundheitswesen Gmbh
(2009) Testregionen in Deutschland.
and Rejecters in German Ambulatory Care