Challenges in Developing Software for the Swedish Healthcare Sector

Bilal Maqbool

and Sebastian Herold

Department of Mathematics and Computer Science, Faculty of Health, Science and Technology, Karlstad University,

Karlstad, Sweden

Keywords:

eHealth, Software Development, Exploratory Study, Empirical Study, Interview Study.

Abstract:

Context: High-quality software is essential to the progressing digitalisation of the Swedish healthcare sector.

Developing software with the desired high quality is far from trivial due to the sophisticated requirements of

the domain. Problem: Studies on healthcare digitalisation challenges in Sweden and other countries, however,

largely focus on the perceptions of healthcare professionals and patients and less on opinions of IT profession-

als. Method: In this exploratory study, we conducted semi-structured interviews with nine IT professionals

about observed challenges in developing software for the Swedish healthcare sector. A qualitative analysis

was performed to identify common themes. Results: We identiﬁed the prevalent challenges to be related to

data integrity, privacy and security, rules and regulations, engineering usability, and software testing. Con-

clusion: The results suggest that further research is required regarding agile methods, efﬁcient requirement

engineering, and testing in eHealth as well as in privacy and usability engineering.

1 INTRODUCTION

Since the late 1990s (Della Mea, 2001), modern in-

formation and communication technology has been

used around the world in healthcare (Bhavnani et al.,

2016). Electronic health records (EHR), patient por-

tals, and mobile applications are just some exam-

ples of tools applied to manage and monitor patients’

health (Ventola, 2014). Among the promises of the

progressing digitalisation of the healthcare sector are

increased efﬁciency, reduced administrative burden,

better insights for patients into their health, and over-

all reduced costs for the public healthcare (Ventola,

2014). Studies also suggest that eHealth is likely to

have a positive effect on the quality life both physi-

cally and psychologically (Wiklund Axelsson et al.,

2013).

eHealth solutions have become an integral part of

Sweden’s healthcare sector, too (Bjorkman, 2019).

For example, 95% of the documentation in the pri-

mary care is stored in EHR and 99% of pharmaceuti-

cal prescriptions are issued electronically (Bjorkman,

2019). Sweden spends approximately USD 1.2 bil-

lion annually for eHealth and has the strategic vision

to become a “Health Tech Leader” by 2025 (Bjork-

man, 2019).

The development of eHealth systems involves

great responsibility as the patient’s health and life

https://orcid.org/0000-0002-1309-2413

https://orcid.org/0000-0002-3180-9182

may depend on them (PwC, 2016). This particu-

larly applies to software, which is an essential part

of almost every eHealth system (PwC, 2016). Devel-

oping software for the healthcare sector is very de-

manding due to diverse requirements that software

solutions have to fulﬁll (Magrabi et al., 2019). The

need to meet high demands regarding safety (Holl-

mark et al., 2015; Rydenf

alt et al., 2019; Frennert

and Baudin, 2019), security (Jung and Loria, 2010;

Hollmark et al., 2015;

Ostlund, 2017; Barkman and

Weinehall, 2017), compliance with regulations (Holl-

mark et al., 2015;

Ostlund, 2017; Svanborg-Sj

ovall,

2014) and others require adequate methods and pro-

cesses to deliver high-quality eHealth solutions (Holl-

mark et al., 2015). Failing to maintain software qual-

ity can result in severe danger to patients’ health and

life (Galin, 2005) as well as to the credibility of the

software provider. For example, in a recent inci-

dent, recordings of patients seeking medical advice

via telephone were openly accessible because they

had been stored without proper encryption or authen-

tication mechanisms (BBC, 2019). This incident did

not only raise privacy and security concerns but also

damaged the credibility of the organizations involved.

While the digitalisation of the sector progresses,

investigating the challenges at hand as perceived by

the relevant stakeholders may provide valuable in-

sights in order to address open research gaps. Ex-

isting studies have mainly focused on the perceptions

of healthcare professionals and patients as the main

user groups of software in the domain. Studies that

take into account the complementing view of software

system and system providers are very scarce though.

In this study, we therefore describe the results of

analyzing semi-structured interviews with IT profes-

sionals about their perceptions regarding challenges

in developing software for the Swedish healthcare

sector. The objective is to identify potential short-

comings and research gaps in software development

methods applied for the healthcare domain.

The outline of the paper is as follows: In Sec. 2,

we will discuss related work. In Sec. 3, we explain

the chosen research methodology. Sec. 4 reports on

the ﬁndings of the study which are then discussed in

Sec. 5. The article ends with concluding remarks and

an outline of future work in Sec. 6.

2 RELATED WORK

For this section, published studies on the challenges

of using, developing and implementing Swedish

eHealth solutions/services were considered relevant.

In 2007, a qualitative study was conducted among

Swedish tele-nurses who were interviewed to explore

reasons for decision aid software not being used as in-

tended (Holmstrom, 2007). It was found that among

the most important reasons were usage for assessment

only due to insufﬁcient decision making support, in-

consistencies with clinical practices, and difﬁculties

in learning how to use the system. As a consequence,

participants reported to often discard recommenda-

tions provided by the system and using self-care ad-

vice to a large extent.

Jung and Loria conducted interviews to identify

barriers and intentions in the acceptance of Swedish

eHealth services by the elderly (Jung and Loria,

2010). They concluded that ease of use, trust in ser-

vice, and compatibility with citizen’s needs were the

main acceptance factors. It was added that better in-

formation about eHealth services are needed to in-

crease awareness for these services such that elderly

people can optimally utilize them.

In 2015, a study reviewed the literature to dis-

cuss challenges for implementing eHealth in Swe-

den and efforts done to overcome them (Hollmark

et al., 2015). General problems were identiﬁed simi-

lar to those that other researchers reported around the

world, such as interoperability, reimbursement, reg-

ulatory, and usability issues. The authors concluded

that interoperability efforts and standardization, better

communication between developers and stakeholder,

and developer expertise in safety, security and privacy

would be crucial for success.

In 2016, PwC conducted a survey involving doc-

tors and nurses from Sweden to understand their will-

ingness, needs, and perspectives regarding the trans-

formation of traditional healthcare to digital health-

care (PwC, 2016). According to this report, health-

care personnel fear the loss of face-to-face contact

with patients, and both care-givers and care-receivers

are worried about the potential negative impact on

the quality of care by using virtual care. Respon-

dents also have serious concerns about the failure of

such systems as the patient’s health can suffer badly.

Researchers suggest, at all levels, clear leadership is

needed, creating opportunities for better collaboration

between different healthcare provides and educating

and supporting staff to achieve a common goal, i.e.

Sweden’s 2025 vision.

In 2017,

Ostlund discussed challenges in digitiz-

ing healthcare in Sweden based on her experiences in

that domain (

Ostlund, 2017). She in particular noted

that requirements are often not gathered from the ac-

tual end-users, especially the elderly, and stressed that

active involvement of older users, and proactive and

cooperative efforts are crucial for the digitization of

healthcare. Moreover, she expressed that the health-

care sector needs to deal with the challenges of digital

arenas that would foster a better understanding the ef-

fect digitalisation in a controlled setting much closer

to the real environment.

Barkman and Weinehall conducted a comparative

study in Ethiopia, Ghana, and Sweden, comparing the

policymakers and mobile-based healthcare (Barkman

and Weinehall, 2017). According to the researchers,

despite the well-developed system, Sweden faces few

challenges: integration of health data, use of digital

decision support to develop personalized medicine,

future funding, and quality and efﬁciency of health-

care systems.

Oberg et al. conducted focus group interviews to

identify Swedish primary healthcare nurses’ concerns

and perceptions regarding the implementation and us-

age of digital healthcare systems (

Oberg et al., 2018).

Nurses were concerned because of the scope of their

operational responsibilities would potentially expand

and the need to adopt to new practices and policies as-

sociated with digital healthcare. The researchers con-

cluded that there is a dire need to educate and train

nurses so they can be involved in the eHealth imple-

mentation.

In 2019, a survey was conducted to study Swedish

home care nursing providers and found that respon-

dents have more expectations with remote monitor-

ing and automation (Rydenf

alt et al., 2019). It was

highlighted that a lack of suitable infrastructure, sys-

tems’ inability to adjust the context of use, and usabil-

ity are currently primary challenges in implementing

and adopting eHealth services. Furthermore, the au-

thors predicted that the digitalisation may massively

affects the workﬂow in healthcare organizations.

Frennert and Baudin performed a survey study

among representatives of different professions work-

ing within Swedish municipal eldercare (Frennert and

Baudin, 2019). It aimed at identifying challenges

and opportunities related to welfare technology in

Swedish municipal eldercare. The results show that

despite the fact that welfare technology is consid-

ered reliable and secure, it still faces many challenges

that slow down its implementation. Lack of funding

and infrastructure, high staff turn-over, uncertainties

about responsibility and laws, and difﬁculties with

procurement are key found challenges. Among other

mentioned challenges, resistance to change is the

same factor as identiﬁed by other researchers (Jung

and Loria, 2010). Lack of structured implementation

processes, availability and usage differences between

municipalities raise further issues.

This summary of related work shows that a sig-

niﬁcant amount of work has been carried out to ex-

plore challenges regarding digital healthcare in Swe-

den. This body of work mostly focuses on healthcare

professionals and their perceptions, such as literature

review about politicians, regulators, and policymak-

ers and their rules, regulations, and policies, inter-

viewing or surveying doctors, nurses, or other staff

members and comparative analysis between Sweden

and other countries’ eHealth sector. All work pre-

sented vital insights, but to the best of our knowl-

edge, no paper is concerned with challenges as per-

ceived by IT professionals producing systems for the

healthcare sector. This is a gap that we intend to

bridge to see whether and which software develop-

ment methodologies and techniques should be poten-

tially adapted/reﬁned. That’s why this paper is ﬁrst of

its kind; a comprehensive study conducted to explore

challenges in eHealth domain.

3 RESEARCH METHOD

The motivation for this study is to explore challenges

in developing software for the Swedish healthcare

sector as perceived by IT experts in order to identify

potential directions for research related to software

development methods. For this purpose, a qualitative

research design was chosen (Berg, 1989) as it allows

for a systematic exploration of the area based on peo-

ple’s opinions and experiences (Holmstrom, 2007).

Semi-structured interviews were chosen as mean of

data collection as they provide the ﬂexibility to fol-

low up on thoughts and ideas not envisaged upfront

(Corbin and Strauss, 2014). In the following subsec-

tions, we describe the applied research protocol in de-

tail.

3.1 Participants and Setting

The participants were selected through purposive,

snowball (Oates, 2005) and opportunistic sampling,

aiming at getting the preferred level of participants

having a wide range of experience related to the de-

velopment of Swedish digital healthcare solutions.

The respondents were informed that their participa-

tion was completely voluntary and anonymous. All

information from the interviews was be anonymised

before the publication of any results. Table 1 gives an

overview of the participants (detailed table

). Eight

interviewees were employed by in total six different

companies. Another participant worked as indepen-

dent developer. Participants P1-P6 and P8-P9 belong

to national and international organizations, interact-

ing with distributed stakeholders and software devel-

opment teams. Eight male (P1-P6, P8-P9) and one

female (P7) of variant backgrounds were interviewed

for approximately one hour on their opinions, expe-

riences, and knowledge of challenges faced during

the development of digital healthcare and welfare sys-

tems or services. Participants have a combined ex-

perience of approx. 34.75 years (mean= 3.86, me-

dian= 1), experiences ranging from 4 months (P5) to

15 years (P1), roles ranging from a software devel-

oper (P3-P6) to CEO (P1) and responsibilities rang-

ing from software development (P3-P6) to managing

teams (P8) in Sweden’s digital healthcare and welfare

sector.

All participants except P4 and P5 have played

more than one role or responsibility during working

in digital healthcare and welfare. Moreover, all par-

ticipants had the experience of working in different

roles with diverse responsibilities in additional do-

mains other than the digital healthcare and welfare

sector. We considered this diversity among the in-

terviewees essential to gather well-founded and rep-

resentative insights regarding the domain of interest.

3.2 Data Collection

Data collection was done through semi-structured in-

terviews for which we developed a protocol with top-

level themes. These themes touched upon perceptions

of challenges and their priorities as well as poten-

tial solutions while leaving enough space for more

speciﬁc follow-up questions. Each interview took

https://drive.google.com/ﬁle/d/

1CmzAIlsHYL0fv6DP5u0TX4BYB7DMZzkn/view.

Table 1: Participant Information.

P. ID Domains/Systems

(Other)

Role

(eHealth)

Responsibilities

(eHealth)

Experience

(eHealth/Total)

P1 Finance

Gaming

Business Developer

CEO

Domain Exploration

Market Survey

15/15 years

P2 Education

Telecom

Software Tester

Team Lead

Planning & Impl. Test

Testing Support

6/20 years

P3 Commercial Apps

Navigation

Software Developer Requirement Elicitation

Software Development

4/5 years

P4 Commerce

Education

Software Developer Software Development <1/5 years

P5 Real Estate

Finance

Software Developer Software Development <1/9 years

P6 Navigation

Finance

Software Developer Design Architecture

Software Testing

1/4 years

P7 Gaming Researcher

Software Developer

Research

Software Development

<1/10 years

P8 Commerce Business Analyst Analyze Business

Software Testing

1/2 years

P9 Business Intelligence

Word Processor

Software Tester

Test Lead

Test and Report Software

Quality

6/22 years

about one hour and were carried out via remote au-

dio or video calls between February-May 2019 and

January-March 2020. The interviewees gave their

consent to be recorded such that the participating re-

searcher could transcribe the interviews. All inter-

view transcripts were cross-reviewed by the second

author. During the interview, participants were al-

lowed to call for a stop of the recording if they felt

uncomfortable with being recorded.

3.3 Data Analysis

The interpretation and analysis of research data are

comprised of epoche and bracketing (Braun and

Clarke, 2006). We tried to minimize the involvement

of our biased knowledge, and synthesis was done

using thematic analysis (Braun and Clarke, 2006).

Micro-analysis (open and axial coding) was con-

ducted to identify in-vivo codes (MacLure, 2013),

and meaning coding categories emerged from the data

during the analysis. In the inductive analysis, a mixed

model of open coding and pre or initial coding list

was used for overall analysis. The pre-coding list acts

as the initial point for discovering deeper ideas from

the research data and was later transferred into the

axial coding. We used the NVivo12 tool to perform

the analysis (Walsh, 2003). All coding types, includ-

ing open and axial codes, were tracked and mapped,

highlighting the codes, their connections, and emer-

gent codes.

We followed several stages to conduct qualitative

analysis. In the ﬁrst stage, authors read the free-text

answers of each interview separately to familiarize

with data and to get a bigger picture of the interview

responses. In the next stage, the ﬁrst author generated

initial codes by analyzing text in-depth and assigned

applicable codes to them (Charmaz, 2014). Mean-

while, the second author kept reviewing codes and the

initial analysis done by the ﬁrst author. Then com-

ments and suggestions were made where required,

reﬁning the overall analysis. In the next stage, the

ﬁrst author performed the axial coding, and the sec-

ond author reviewed it in a similar routine. Interviews

and analyses were iteratively reviewed and discussed

during meetings between the authors. After agreeing

upon different comments and opinions, changes were

made. In the next stage, the authors analyzed for the

addition of new or deletion of old codes. In the ﬁ-

nal stage, authors searched for themes, reviewed them

and then classiﬁed and characterized identiﬁed codes

into different themes.

3.4 Ethical Considerations

We have followed the guidelines given by the Swedish

Research Council for research ethics (Gustafsson

et al., 2017). Information gathered or observed dur-

ing the interview is reported and analyzed accurately

and honestly.

4 FINDINGS

The analysis of the interview data revealed challenges

that were categorized into ﬁve main themes. In the

following subsections, we present the ﬁndings for

each of the ﬁve themes.

4.1 Challenges Related to Integrity,

Privacy, Protection and Security

Data is considered a key component and that can be

used to automate systems and services in the digi-

tal healthcare sector. More and more data is stored

and available for processing. Hence, ﬁve out of nine

participants (P2, P4, P5, P8, and P9) discussed chal-

lenges related to data, i.e. data integrity, privacy, data

protection, and security.

P2, P8, and P9 rate data integrity as one of the

most critical challenges faced during the development

of eHealth systems. P2 and P9 noted that patients, es-

pecially doctors, have concerns about the integrity of

the data because it is essential for making decisions in

many health-related processes. P2 stated “...really im-

portant, the personal integrity and security for the pa-

tient” because the health of individuals or even lives

may depend on critical data (as P9 stated). P8 added

that IT professionals need to consider various aspects

related to data integrity, such as how data is handled

in an organization because handling data will affect

the integrity of the data. P9 referred to a problem of

doctors having incomplete data about patients nowa-

days as an example of the challenge of data integrity:

“Today we don’t have all data in Sweden... the doc-

tors are not sure that they see everything about the

patient... that could be a problem”. Complete patient

information should be available to the doctor because

a system based on incomplete data will have bugs,

and it does not optimally support professionals and

patients as P9 explained “we can’t miss the big bugs

that could affect the decision of a doctor...”.

Three participants, P4, P5, and P6, talked about

challenges related to data protection but P6 did not

perceive this as a high priority challenge and hence

did not consider it for further discussion. P4 discussed

issues that need to be taken into consideration after re-

ceiving patient data. IT professionals need to ensure

that data is protected from any loss or unauthorized

access such that patients can trust the system. P5 also

stressed the potential impact on market reputation and

expressed: “if you lose that (patient) data, I think you

lose your reputation within the market, since no one

is able to trust you guys, not your app, not your com-

pany”.

Two participants P4 and P5, perceived data pri-

vacy as one primary challenge faced during the devel-

opment of eHealth systems. Patient data is very conﬁ-

dential, especially information regarding their disease

history. They expressed concerns about the conse-

quences of leaks of personal data that could like il-

legal sharing of patient data or offenses like black-

mailing. This challenge affects decisions taken re-

garding the process and approach taken for handling

patient data. While talking about PDA (Personal Dig-

ital Assistants), P4 gave an example: for “generating

an email... we should only use our mobile side, no

data should be kept at the server side due to the pri-

vacy”, to deal with such circumstances P4 mentioned

to use decentralized architectures.

According to participants P2 and P4, data security

is as important to the patient as the integrity of the

data. IT professionals have to ensure data security to

satisfy patients. Giving an example of data security,

a data-related challenge, P5 discussed the need to put

extra efforts during software development. If the sys-

tem lacks data security, it becomes a threat not only to

the organization but also to the patient. To ensure data

security, IT professionals need to add some additional

security layers.

Interviewees have a several recommendations on

how to address mentioned challenges. IT profession-

als need exhaustive software testing to ensure data

integrity, P9 stated “We need to do more and better

testing...” to make sure “. . . that the data isn’t lost

or transformed in some other way. It’s one typical,

very important testing”. The P8 recommends that

the software development team include people with

good knowledge of data integrity laws. This can help

in developing a product that complies with different

laws, regulations, or standards. The P8 spoke on data

handling and said that the workforce should be made

aware of the security rules and in compliance with

various regulations, especially the GDPR (EU’s Gen-

eral Data Protection Regulation). So, IT professionals

can ensure the security, accuracy and consistency of

data over its lifetime.

According to P5, decentralized architectures are

recommended to overcome data protection challenges

as there will be no single point of failure: “If one

server is compromised... 4 servers remain... if it’s

distributed. But you have to apply the security layer

on all, and you have to keep monitoring”. P6 brieﬂy

added a common known solution to address data pro-

tection challenges by implementing data encryption.

P4 mentioned that developers should be given

clear instructions on technical protocols to ensure the

privacy and security of the developing system. While

giving an example of how to send data from a system,

P4 explicitly mentioned that HTTP protocols should

be used: “... for the protocols such as HTTP should be

used to send (data)”. In addition, other business or ad-

ministrative policies that may affect privacy and secu-

rity decisions should be clearly communicated by the

hospital or the client. Close collaboration between so-

lution providers and stakeholders can help to address

such privacy and security challenges.

There are many security protocols available, P5

thinks that by following the security protocols and us-

ing them, IT professionals can ensure data security.

P5 suggested that centralized architectures should be

used to ensure data security: “Whenever you get the

data, it should be centralized somewhere”. Among

other solutions, as P6 mentioned about data protec-

tion, P5 also mentioned that data encryption and mul-

tiple security layers are needed to ensure data security

in a system: during the development cycle, data “has

to be secured with... password protection and there

should be multiple layers” of security.

The interviewees experienced most of the chal-

lenges mentioned above in other domains, too. As

compared to the banking sector and other domains,

P4 and P5 perceived the severity of data privacy chal-

lenges as higher in the eHealth sector. P5 perceived

severity of data protection and data security chal-

lenges high in the eHealth sector compared to the

banking sector. P8 perceived data integrity challenges

to be of higher severity than in other domains while

P9 had not faced such challenges in other domains so

far.

4.2 Challenges Related to Software

Testing

Four participants (P2, P3, P7, and P8) discussed

software testing challenges. According to P2, the

Swedish healthcare sector has experienced signiﬁcant

digital innovation, however, quality assurance and

testing were not taken seriously enough. This, accord-

ing to P2, resulted in poorly tested systems: “qual-

ity assurance and testing of these systems weren’t

prioritized” and “systems in this business (were) so

untested”.

Access to the real patient and other critical data

is often desirable for proper testing which is rarely

granted due to privacy and security concerns. As P7

stated “doctors were not willing to give me informa-

tion and of course I understand security reasons”.

Developers often use synthetic data for testing, which

carries the risk of not representing reality thoroughly.

This affects the efﬁciency of software development

as the creation of proper test data consumes signiﬁ-

cant time. Regarding test environments, P9 said, if

there are many different systems under the control of

different stakeholders (public and/or private) then it

will be a major challenge to access patient informa-

tion for creating test environments without spending

more time and money on this task.

P3 additionally highlighted the lack of testing in-

struments, especially medical hardware or equipment

that could be used for real environment testing. Due

to this challenge, solution providers are unable to

provide accurate results during software testing and

cannot predict the actual performance of the system.

Buying medical hardware for testing purposes will in-

crease production costs: “This was really costly for

me.. and I bought... just for testing purposes”.

Having multiple and different stakeholders creates

communication problems, P8 said. It becomes difﬁ-

cult to collaborate and communicate within the com-

pany or with other stakeholders: “a lot of the stake-

holders... want to (get) stuff (done) and it’s hard

for every sector... of the company to cooperate and

communicate...”. P8 added, it also becomes hard to

get and set functional requirements as companies set

higher demands. Due to unclear functional require-

ments and unrealistic deadline organizations often

sacriﬁce testing periods, so solution providers may

not meet the demands and software quality. There

will be bugs in the delivery system and it will not

work as intended. P8 stated “... their requirement in-

creases... (and) buggy system can... not save people’s

lives... (we) need more longer tests period”.

According to P2, lack of quality assurance and

software testing prioritization has created an oppor-

tunity for companies to enter the eHealth domain but

companies need to take quality assurance seriously

and quality needs to be prioritized to overcome such

challenges. If IT companies have close collaboration

and patient data-sharing agreements with stakehold-

ers then companies can overcome the lack of access

to patient information challenges. P7 stated, “coming

up with this idea to do agreement... can easily access

the information or data they (developer or researcher)

want about the patient and use for their research”.

For P3, with increased collaboration between de-

velopers and stakeholders, availability and access of

medical hardware for testing, IT professionals can

overcome challenges related to the lack of testing in-

struments: “bridge between developers and the sector

(is needed), (in) which developers get access to more

hardware and sensors..., like for testing of eHealth

systems or apps”. P9 assumes that this will take time,

but by working closely with different stakeholders

and developing or integrating pre-existing test envi-

ronments, IT professionals can overcome the prob-

lems related to the test environment. P8 added that

more time needs to be spend on exhaustive testing

to ensure the correctness of the system. Moreover,

IT professionals can use agile practices to bridge the

stakeholder communication gap.

According to P2 and P7, the lack of software

testing prioritization and lack of access to relevant

patient data are high in the eHealth sector as com-

pared to the telecommunication domain. Participant

P3 mentioned that the lack of testing instruments was

a less pronounced challenge when testing navigation

systems and/or applications other domains he had

worked in before. The lack of testing environments

are high in the eHealth sector than in other domains,

perceived by P9. According to P8, the gaps in com-

munication with stakeholders are equally common in

other domains.

4.3 Challenges Related to Usability and

User Experience

Three participants (P4, P5, and P7) discussed chal-

lenges of usability and user experience in detail. P3

and P6 also brieﬂy touched upon these topics. Ac-

cording to P4 and P5, IT professionals aim to make

applications easy to use but face some major chal-

lenges in the eHealth sector, especially for elder or

disabled users: “have to consider the old people as

well (as) disabled people... so I think the user experi-

ence matters a lot” as both groups may be affected by

poor eyesight, hearing problems, and other conditions

requiring special support for using software systems.

P5 went as far as to say that lacking usability de-

ﬁes the whole purpose of applications in this domain.

According to P7, the language barrier is an additional

usability and user experience challenge, especially for

newcomers to Sweden. Regarding usage problems P3

said, if IT professionals have less or missing infor-

mation about usage perspectives then it would also

decrease the usability of the software or system.

According to P4, P5, and P7, user interface de-

sign and other usability aspects need to be empha-

sized already early in the development process. P4

mentioned, these challenges will affect the develop-

ment process during prospective design. P5 said,

IT professionals need to research how they can im-

prove usability and user experience, and seek con-

stant feedback from customers, consumers, and user

experience (UX) teams. However, P5 also highlighted

that time constraints in development might restrict the

effort that could be spent on usability engineering.

Additionally, P5 elaborated on another challenge for

healthcare system providers, i.e. the need to guide and

train patients, doctors, and other professionals on how

to use novel applications. This would make systems

easier to use and affect their acceptance positively but

would require extra efforts from system providers.

P7 added, IT professionals need to involve end-

users in the design phase from the beginning and work

closely with them: “user involvement is very impor-

tant, extremely, especially in the ﬁrst phase of the de-

sign”. According to P5, IT professionals can keep

stakeholders on-board during the development lifecy-

cle. IT professionals need to collaborate with stake-

holders to better understand problems as it is impor-

tant to get feedback from customers on a regular basis.

P5 thinks, if IT professionals can provide better user

experience then they would have to put less effort into

training users.

P5 said, nowadays every company has specialized

UX designers and with their help developers can think

more about users and develop user-oriented solutions.

For P4, there should be clear design guidelines re-

garding making the application more useful and IT

professionals need to follow those guidelines to en-

hance user experience. P6 added, thinking about dif-

ferent user perspectives, user experience, and usage

scenarios, IT professionals can produce a product that

fulﬁlls user’s needs. Also IT professionals need to in-

volve special user interface (UI) designers to enhance

usability and user experience. P5 believes that beta

testing can also play an important role to understand

and mitigate ﬂaws in user experience.

P4 perceived challenges related to usability high

in the eHealth sector as compared to other domains

but P7 has no experience of usability challenges in

other domains. P5 perceives user experience chal-

lenges high in the eHealth sector as compared to the

banking sector but has no experience with other do-

mains. P5 perceived the same severity of user training

challenges in the eHealth sector as compared to soft-

ware for the construction, banking, or other domains.

4.4 Challenges Related to Rules and

Regulations

Four participants (P1, P2, P8, and P9) discussed chal-

lenges related to rules and regulations. P8 states, pro-

viding digital healthcare solutions has become more

challenging due to a variety of requirements caused

by administrative or legal constraints: “A lot of orga-

nizations (have) a lot of requirements regarding how

to develop solutions”. The digital healthcare system

involves many different stakeholders which makes

managing systems challenging.

This is especially true for Sweden’s decentralized

administration of the healthcare sector, which to a

large degree, is in the hands of the 21 counties. Ac-

cording to P9, this leads to integration issues: “...have

a lot of systems that are interconnected and... differ-

ent people would be responsible for different things...

the challenge will be to make sure that we have as

many real integrations as we can within reasonable

resources”. P1 added that the development of soft-

ware solutions is often delayed because the systems to

be developed have to be adapted to the regulations in

different counties. In the same way, P9 said that im-

plementing the technical means to ensure interoper-

ability between different systems is not the main chal-

lenge but the large number of stakeholders operating

the systems that adds to the complexity of designing

a system. IT professionals need to plan how com-

plex rules will affect the integration of different sys-

tems and the exchange of data between them. These

challenges are not difﬁcult to tackle, but they demand

extra time and resources, P9 stated “it takes time and

money to make the data available in the right format”.

P1 additionally highlighted that having complex

regulations and administrative decisions in each in-

dividual county multiply the effort required to adapt

to changes in the regulations. This is not only true

for existing companies in the healthcare sector but in

particular for new and small companies trying to en-

ter the healthcare domain in Sweden and improving it

with new innovations. They have to either face the is-

sue of having to comply with a large set of regulations

or limit their product to counties with similar regula-

tions which may not be feasible. P1 summarised:: “It

makes Sweden unattractive... because the market will

be so limited...”.

The lack of technical knowledge or consideration

of technical stakeholders during political decision-

making in healthcare was perceived as another chal-

lenge. P1 referred to the example of one county

that decided to force all healthcare providers to use

one speciﬁc EHR system, effectively putting them

into a vendor lock-in and shutting out other solution

providers from this market. The decision was moti-

vated by the believe that this would increase the data

security. However, according to P1, this would have

been solved in a more sustainable and less market-

restricting way instead by an administrative decision

on technical standards regarding secure storage and

processing of EHR.

With their experience with working in agile set-

tings, P1 and P8 discussed how issues are raised if

stakeholders or organizations are following different

software development practices. P8 said, the soft-

ware development industry is working in a modern

agile way, but the government organizations that for-

mulate and govern the regulations lack agility, mak-

ing it difﬁcult to cooperate and work effectively:

“They’re not that agile, so it’s hard for them to take

on changes. Two systems have different development

life-cycles...”.

According to the P2, due to different rules and reg-

ulations, developers also lack knowledge about laws.

Always there remains a gap between knowledge about

different laws applying to different types of users:

“Different laws are applicable to different cases... the

same information, but different laws”. Sometimes

laws are too rigid, perhaps unintentionally, preventing

innovation but IT professionals need to comply with

laws otherwise they will face legal issues.

According to P1, to overcome these challenges, IT

professionals need to understand the market to know

the boundaries that have been set by politicians for

the healthcare system. Hence rules and regulations

will become very clear. P8 added some regulations

are outdated and do not reﬂect the way new technol-

ogy is applied in healthcare and stressed that more

agility among those deciding on regulations would

be desirable. P2 expressed the opinion that in or-

der to overcome challenges regarding lack of knowl-

edge about laws, some regulations need to be relaxed

by the authorities. He added that software compa-

nies need to educate teams about rules and regula-

tions, follow regulations especially GDPR. In order to

overcome the interoperability challenges, P9 recom-

mended close collaboration with the different stake-

holders in the Swedish healthcare system. Further-

more, P1 proposed to intensify the efforts in estab-

lishing interoperability standards.

According to P1 and P8, following rules and regu-

lations is more challenging in the eHealth sector than

in ﬁnance and other domains. For P2, challenges re-

garding the lack of knowledge about laws are high in

the eHealth sector as compared to the telecommuni-

cation domain. P8 mentioned, due to different stake-

holders and systems, challenges of interoperability

are high in the eHealth sector as compared to ﬁnance

and other domains. P9 said that the involvement of

and dealing with different stakeholders in the eHealth

sector is more challenging than in other domains.

4.5 Challenges in Problem Domain

Analysis

Two participants (P3 and P6) discussed challenges in

problem domain analysis. Lack of documentation re-

lated to the problem domain and code pose further

challenges during the development phase. P3 reported

an example in which an existing system they were

asked to maintain was extremely difﬁcult to compre-

hend and hard to be run due to lacking documenta-

tion. P6 added, IT professionals need better insight

into the area of expertise or application to understand

and solve the actual problem. P6 highlighted that de-

velopers start developing an application with less in-

formation about the problem domain and ended up

with a faulty or incomplete system. Getting feed-

back from customers at the end of the development

phase consumes more time to rectify issues. More-

over, both participants stressed the importance of an

elaborative requirements elicitation, P3 stated “if the

requirements aren’t clear for each perspective, then

how can you (IT professionals) develop the entire ap-

plication or software”.

P3 discussed developer perspective and their lim-

ited knowledge of the medical domain, a major chal-

lenge related to developer’s perspective during the de-

velopment of the eHealth system. The mobile appli-

cation developer’s perspective focuses more on better

managing healthcare application platforms. IT pro-

fessionals have difﬁculty learning medical ideas and

their understanding of the medical domain becomes

limited. As P3 said, things are getting better day by

day, and over time, IT professionals became aware of

the ﬂow of the project. But we (authors) believe that

it has steep learning curve that uses critical time that

could be used to develop or test software or systems.

According to P3, collaboration between stake-

holders and stakeholders on-board can help to bet-

ter learn and understand the domain of the problem:

“to overcome this challenge. . . , on-board stakehold-

ers needed...”. P6 added that once the requirements

are elicited, instead of working directly on a project,

IT professionals need to do brainstorming and build

a prototype. Moreover, instead of getting feedback at

the end do it early to avoid time loss.

The eHealth sector lacks documentation related

to problem domains more than ﬁnance and other do-

mains. P6 added that the challenge is almost the

same, but IT professionals need to be more careful

in eHealth as they are dealing with important real-life

situations.

5 DISCUSSION

5.1 Discussion on Integrity, Privacy,

Protection and Security

In eHealth, data is being used to provide useful in-

formation not only to care providers but also to care

receivers (Frennert and Baudin, 2019). Challenges re-

garding incorrect or incomplete data were previously

reported when observing the use of the software by

tele-nurses (Holmstrom, 2007). Our ﬁndings conﬁrm

that this issue is considered a general challenge by

IT experts, too. Based on the replies of our partic-

ipants we assume that such issues are only partially

due to the complexity of the domain but are rather

caused to a large degree by incomplete requirements

elicitation, testing, and lack of knowledge about laws

related to data integrity. Further research is needed on

how these activities can be improved and proactive

solutions can be taken in eHealth.

Challenges related to data integrity, privacy, and

security were also reported by other researchers while

mentioning the importance of reliable infrastructure

and trust in healthcare, respectively (

Ostlund, 2017;

Jung and Loria, 2010). Additionally, we found that

private companies or organizations have to put extra

efforts to build such conﬁdence. We noticed different

perceptions of potential architectural solutions (cen-

tralised vs. decentralised) to privacy, data protection,

and security which might point at the need to clarify

the properties of different architectures and the qual-

ity attributes they support.

Additionally, several participants stressed the

need to use data encryption for data protection, which

seems surprising as this is a commonly known tech-

nology. Poor encryption are among the security con-

cerns and researchers also recommended the need to

encrypt data in eHealth, especially for transmitting

data over insecure public networks (Al-Issa et al.,

2019; Sahama et al., 2013). It seems worthwhile in-

vestigating to which degree sensitive data is actually

encrypted in the software landscape of the Swedish

healthcare sector.

We also recommend investigating standardized

methods, architectures, and additional security layers

(like HTTP over SSL) that are needed to be imple-

mented during the development phase to solve current

theme challenges. A survey paper also suggests sev-

eral possible solutions related to current theme chal-

lenges (Al-Issa et al., 2019). Furthermore, we sug-

gest exploring privacy by design approaches or any

conceptual framework to analyze privacy and security

threats and develop secure applications in the health-

care domain (Al-Issa et al., 2019).

5.2 Discussion on Software Testing

Our ﬁnding on low testing prioritization is in line with

previous research, which reported that testing budgets

are often limited and testing periods are short (Fren-

nert and Baudin, 2019). This was conﬁrmed in partic-

ular by participants responsible for testing. It seems

hence worthwhile, to ensure the quality of eHealth so-

lutions in the future, to investigate possibilities of test

automation. The feedback provided by our intervie-

wees also points to improving the efﬁciency and accu-

racy of other development activities, such as require-

ments engineering, in order to allow for more time

for testing. Many requirements engineering practices

have been ineffective in real world projects and yet

have been a problem in eHealth (Fricker et al., 2015).

It seems worthwhile to investigate which methods

in particular for requirements engineering are com-

monly used in the domain of interest and how their

efﬁciency can be increased.

We found that most IT professionals and re-

searchers do not have access to real data due to

privacy issues. We suggest investigating potential

challenges associated with patient data-sharing agree-

ments. What will happen if data owners belong to

different companies or countries that have different or

even conﬂicting rules regarding data sharing? We rec-

ommend exploring a better way that not only ensures

the privacy and protection of personal data, but also

gives access to patient and critical data to IT profes-

sionals and researchers working in the eHealth sector

as, for example, proposed by Frotoni et al. (Frontoni

et al., 2014).

Our ﬁnding shows that a lack of access to testing

instruments makes it challenging to test systems. In

addition to that, a researcher mentioned, the lack of

knowledge about social contexts also affects the test-

ing of eHealth systems (

Ostlund, 2017). We suggest

exploring different solutions for better access to test-

ing instruments like efﬁcient and safe testing of sys-

tems outside the controlled environments and simula-

tion of required instruments, for example, digitizing

new arenas (

Ostlund, 2017). We recommend explor-

ing better ways of test prioritisation while keeping

medical testing instrument availability challenges in

mind. Relevant techniques could be risk-based testing

(RBT) and cost-oriented prioritisation of test cases

(Huang et al., 2012).

Our ﬁnding shows that stakeholders’ communica-

tion issues are perceived as one cause of faulty sys-

tems. Considering the example of tele-nurses, soft-

ware programs were not fully up-to-date, inconsis-

tent with current practices, and much information

was missing (Holmstrom, 2007). Such issues can be

caused by unclear functional requirements, higher de-

mands, and unrealistic deadlines raised due to com-

munication gaps found during our ﬁndings. Our ﬁnd-

ing recommended agile approaches to ensure well-

structured teams and processes, better cooperation,

and work efﬁciency. Customer collaboration depends

on the application under development and organisa-

tional context of development (Robinson and Sharp,

2010). Therefore, we recommend further research,

whether agile practices help collaboration and com-

munication, or consume more resources than beneﬁts

in eHealth.

5.3 Discussion on Usability and User

Experience

We found that challenges on usability and user experi-

ence were also reported by (Svanborg-Sj

ovall, 2014;

Hollmark et al., 2015;

Ostlund, 2017;

Oberg et al.,

2018; Rydenf

alt et al., 2019; Frennert and Baudin,

2019) in their studies. It is difﬁcult to ensure the us-

ability and user experience for the elderly and dis-

abled. In addition, our research highlighted language

barriers and usage perspective challenges that can

lead to difﬁculties in using eHealth systems or soft-

ware. We suggest that more efforts be made to inves-

tigate the reasons for ignoring input from elder or dis-

abled people. Furthermore, we found that sometimes

usefulness is considered more important than the us-

ability of eHealth, also reported by (Jung and Loria,

2010;

Ostlund, 2017;

Oberg et al., 2018). Therefore,

we recommend exploring testing strategies to eval-

uate the usefulness of eHealth systems or services.

Eliciting design requirements is beyond the usabil-

ity knowledge of most developers, requirements en-

gineers, and users (Juristo et al., 2007). We therefore

recommend further investigation to determine who is

responsible for providing clear design guidelines and

whether this always guarantees the usefulness of the

application.

There is a belief that usability concerns can be

considered in the ﬁnal stages of the development pro-

cess (after testing), as it should not take time to rework

for this quality attribute (Juristo et al., 2007). But our

research and literature emphasizes the need to ensure

usability during the early stages of the development

process. We recommend inquiring whether IT com-

panies in eHealth focus entirely on user interface de-

sign and leave usability and user experience during

the design phase or if there are any other unexplored

issues.

Furthermore, we suggest investigating how to

keep elderly and disabled users on-board from the

beginning of development and get constant feedback

as multiple challenges are associated with it. In ad-

dition to guidelines (Smeltzer et al., 2017), imple-

mented practices in the IT sector can be explored for

better collaboration. Further research is needed to ex-

plore better methods for usability engineering in the

eHealth sector, as there is a belief that there should be

usability engineering throughout the lifecycle of an

eHealth application (Price et al., 2016).

Sometimes usability testing is confused with beta

testing and the ﬁrst is believed to be sufﬁciently cov-

ered if integrated in the latter stage as users are al-

ready involved in beta testing. However, usability

and beta testing are very different as developers and

testers can ask questions on the ﬂy during usability

testing but not so much in beta testing (Decker, 2014).

Moreover, beta testing is used to be performed to-

wards the end of application development while us-

ability should be evaluated continuously throughout

the design phase. That’s why we recommend explor-

ing how beta testing would be helpful in understand-

ing and mitigating ﬂaws in the user experience.

Some research studies have highlighted the chal-

lenges of educating consumers to use healthcare soft-

ware programs (Holmstrom, 2007) and (

Oberg et al.,

2018). Our ﬁnding also agreed to this challenge and

discussed the additional efforts that IT professionals

put into training users and considering their usage per-

spectives. We have found that usability is indirectly

proportional to efforts to train consumers, but still,

further investigation is needed to conclude.

5.4 Discussion on Rules and Regulations

We found challenges related to the fragmented and

distributed administration which was also referred to

in previous studies (Wadmann et al., 2009; Hollmark

et al., 2015; Barkman and Weinehall, 2017;

Ostlund,

2017; Frennert and Baudin, 2019). We have found

that fragmented administration makes it difﬁcult for

all counties to develop the same system. We recom-

mend looking for software product lines that can help

IT professionals to understand the general needs of

different counties (Strobl et al., 2010).

We found that there is a lack of technological un-

derstanding in political decision making. Sometimes

complex and rigid laws prevent innovation in the

eHealth domain. A study added that such challenges

could result in time, cost inefﬁciency, and untested

systems (Hollmark et al., 2015).

Furthermore, we have found that unclear and dif-

ferent regulations in different counties make it difﬁ-

cult for new companies to develop eHealth solutions.

In general, administrative rules and regulations don’t

reﬂect technological advancements. Such issues raise

interoperability challenges, loss of market, and cus-

tomer issues. We suggest exploring standardisation

efforts like HL7 and we have to ﬁnd out what pieces

are missing in current efforts for interoperability in

the Swedish eHealth sector. Furthermore, we suggest

investigating how understanding the market will help

to raise awareness of the limitations that politicians

have set on the healthcare system and how it will in-

crease awareness and acceptance level among people.

Regarding the lack of agility in administration, we

suggest that researchers need to explore how to han-

dle the use of traditional software methods in govern-

ment organizations. Furthermore, we believe that the

transition from the traditional process to agility will

present itself with a variety of challenges. If there

is no other way than transforming, then how can this

transformation be made in order to work effectively

and efﬁciently with other organizations while keeping

discovered challenges in mind.

Our ﬁndings show that knowledge of laws or stan-

dards among developers could be extended. Develop-

ers need to put extra effort and time to acquire such

knowledge. We believe that higher staff turnover will

create more problems, as the participant points out

that the gaming industry is attracting more develop-

ers. Agile practices could help in better communi-

cation and early detection of issues. Continuous in-

volvement of stakeholders would help to get a clear

understanding of rules and regulations. Furthermore,

we suggest investigating techniques that could efﬁ-

ciently solve problems of knowledge about different

laws applicable to different users and regulatory com-

pliance in the eHealth sector.

5.5 Discussion on Problem Domain

Analysis

Our ﬁnding shows that if IT professionals don’t have

proper documentation, then it becomes challenging

to understand and analyze the problem domain. This

can lead to problems with implementation and usabil-

ity, especially during the maintenance of a project.

Our ﬁndings suggest requirement engineers need to

ensure that developers must have complete informa-

tion regarding the problem domain so that they can

effectively meet customer needs. We recommend

exploring efﬁcient requirement engineering methods

and domain-driven design and development concepts.

Our ﬁnding on lack of medical domain knowl-

edge among developers is in line with previous re-

search, which reported that tele-nurses believe that IT

experts lack insight into the problem domain analy-

sis and medical domain knowledge (Hollmark et al.,

2015). We think teaching medical domain knowledge

to every developer has a steep learning curve and it

consumes essential resources that could be used for

other important activities. Considering the case of P1

from our ﬁndings, software companies can dedicate

a few key people responsible for having medical do-

main knowledge, and they will share it with IT pro-

fessionals. Therefore, not every IT professional will

need to have this knowledge explicitly.

On-board stakeholders, brainstorming, prototyp-

ing, and close collaboration may be solutions to the

lack of medical domain analysis and knowledge chal-

lenges. Since these are key features of agile practices

that most IT professionals apply, it seems worthwhile

to investigate whether they actually work in the do-

main of interest of this study.

It can be seen, many of the recommendations

found during our study were the most common in ag-

ile practices, but still, the eHealth sector faces vari-

ous challenges. We suggest investigating this issue

further, whether the technology or requirements are

evolving or changing rapidly compared to the recom-

mended solutions, or there is something else that is

still unexplored.

5.6 Threats to Validity

One aspect inﬂuencing the external validity of a study

is the number of observed cases (Yin, 1994). Eight

cases are often considered a reasonable minimum

threshold to draw generalizable conclusions in qual-

itative studies (McCracken, 1988). We hence believe

that with the given number of participants (n=9), we

managed to obtain generalizable ﬁndings about exist-

ing challenges in the domain of interest. However, a

larger number of cases would be desirable to get a bet-

ter picture of the prevalence of challenges. We deem

the variety among interviewees regarding their level

of experience and area of expertise to reﬂect the tar-

get population fairly and hence to support the external

validity of the study.

The performed data analysis is heavily based on

human interpretation of collected data. Therefore, it

might be compromised by biases, inconsistencies, and

missed aspects in the interpretation, reducing the reli-

ability of the study. To counteract this threat, contin-

uous mutual reviewing was performed by the authors

to ensure the developed code and themes quality.

All interviews were conducted before the Covid-

19 pandemic seriously affected the Swedish health-

care system. We are aware of potential shifts in the

perceived challenges based on experiences with de-

veloping and operating eHealth systems during the

pandemic. A replication of the study today might

hence lead to slightly different conclusions. We be-

lieve though that the identiﬁed potential research di-

rections are relevant in the long run.

6 CONCLUSIONS

It has been noticed that a lot of work has been done

to explore challenges in the Swedish eHealth sector,

but it was not done from the software developers’ per-

spective. The results suggest that further research is

required regarding agile methods, efﬁcient require-

ment engineering, and testing in eHealth as well as

in privacy and usability engineering. There are also

some problems in realizing the vision, but better regu-

lations, user-centric, pro-active, and collaborative ef-

forts are needed to fulﬁll Sweden’s 2025 vision. Other

domains primarily, the banking sector is working hard

to increase the effectiveness and quality of the ser-

vice to increase customer satisfaction. We think there

is hope, potential, and expectation that soon eHealth

will have the same. Further research is required to

look into how other domains in Sweden or digital

healthcare solution providers in other countries are

dealing with such challenges.

ACKNOWLEDGEMENTS

This work was funded by Region V

armland via the

DigitalWell Arena project (Dnr RV2018-678).

REFERENCES

Al-Issa, Y., Ottom, M. A., and Tamrawi, A. (2019). ehealth

cloud security challenges: A survey. Journal of

Healthcare Engineering, 2019.

Barkman, C. and Weinehall, L. (2017). Policymakers and

mhealth: Roles and expectations, with observations

from Ethiopia, Ghana and Sweden. Global health ac-

tion, 10(sup3):1337356.

BBC (2019). Millions of medical calls exposed online.

https://www.bbc.com/news/technology-47292887.

Accessed: 23.08.2020.

Berg, B. L. (1989). Qualitative Research Methods for the

Social Sciences, volume 1. Pearson Boston, MA.

Bhavnani, S. P., Narula, J., and Sengupta, P. P. (2016). Mo-

bile technology and the digitization of healthcare. Eu-

ropean heart journal, 37(18):1428–1438.

Bjorkman, J. (2019). Healthcare resource guide:

Sweden. https://2016.export.gov/industry/health/

healthcareresourceguide/eg main 116247.asp. Ac-

cessed: 04.07.2020.

Braun, V. and Clarke, V. (2006). Using thematic analysis

in psychology. Qualitative research in psychology,

3(2):77–101.

Charmaz, K. (2014). Constructing Grounded Theory. Sage.

Corbin, J. and Strauss, A. (2014). Basics of Qualitative

Research: Techniques and Procedures for Developing

Grounded Theory. Sage publications, 4th edition.

Decker, R. (2014). Usability vs beta testing (the

what & when). http://www.uxsisters.com/2014/08/

usability-vs-beta-testing. Accessed: 24.11.2020.

Della Mea, V. (2001). What is e-health (2): The death of

telemedicine? Journal of medical Internet research,

3(2):e22.

Frennert, S. and Baudin, K. (2019). The concept of welfare

technology in swedish municipal eldercare. Disability

and Rehabilitation, pages 1–8.

Fricker, S. A., Grau, R., and Zwingli, A. (2015). Require-

ments engineering: best practice. In Requirements En-

gineering for Digital Health, pages 25–46. Springer.

Frontoni, E., Baldi, M., Zingaretti, P., Landro, V., and Mis-

ericordia, P. (2014). Security issues for data sharing

and service interoperability in ehealth systems: the

nu. sa. test bed. In 2014 International Carnahan Con-

ference on Security Technology (ICCST), pages 1–6.

IEEE.

Galin, D. (2005). Software Quality Assurance: From The-

ory to Implementation. Pearson Education India.

Gustafsson, B., Hermeren, G., and Petersson, B. (2017).

Good Research Practice. Stockholm: Veten-

skapsr

adet, Sweden.

Hollmark, M., Skjoldebrand, A. L., Andersson, C., and

Lindblad, R. (2015). Technologyrready to be

launched, but is there a payer? challenges for im-

plementing ehealth in sweden. In PHealth 2015:

Proceedings of the 12th International Conference on

Wearable Micro and Nano Technologies for Personal-

ized Health, Sweden, volume 211, page 57. IOS Press.

Holmstrom, I. (2007). Decision aid software programs

in telenursing: not used as intended? experiences

of swedish telenurses. Nursing and health sciences,

9(1):23–28.

Huang, Y.-C., Peng, K.-L., and Huang, C.-Y. (2012). A

history-based cost-cognizant test case prioritization

technique in regression testing. Journal of Systems

and Software, 85(3):626–637.

Jung, M. L. and Loria, K. (2010). Acceptance of swedish

e-health services. Journal of multidisciplinary health-

care, 3:55–63.

Juristo, N., Moreno, A., and Sanchez-Segura, M.-I. (2007).

Guidelines for eliciting usability functionalities. IEEE

Transactions on Software Engineering, 33(11):744–

758.

MacLure, M. (2013). Classiﬁcation or wonder? coding as

an analytic practice in qualitative rresearch. Deleuze

and research methodologies, pages 164–183.

Magrabi, F., Habli, I., Sujan, M., Wong, D., Thimbleby, H.,

Baker, M., and Coiera, E. (2019). Why is it so difﬁcult

to govern mobile apps in healthcare? BMJ Health and

Care Informatics.

McCracken, G. (1988). The Long Interview (Qualitative

Research Methods). Sage, 1st edition.

Oates, B. J. (2005). Researching Information Systems and

Computing. Sage.

Oberg, U., Orre, C. J., Isaksson, U., Schimmer, R., Lars-

son, H., and H

ornsten,

A. (2018). Swedish primary

healthcare nurses’ perceptions of using digital ehealth

services in support of patient self-management. Scan-

dinavian journal of caring sciences, 32(2):961–970.

Ostlund, B. (2017). Digitizing health care: Welfare technol-

ogy as a way to meet digital and demographic chal-

lenges in sweden. In 2017 4th International Confer-

ence on Systems and Informatics (ICSAI), pages 78–

83. IEEE.

Price, M., Weber, J., Bellwood, P., Diemert, S., and Habibi,

R. (2016). Evaluation of ehealth system usability and

safety. Handbook of eHealth Evaluation.

PwC (2016). The digital patient is here — but is healthcare

ready?

Robinson, H. and Sharp, H. (2010). Collaboration, commu-

nication and co-ordination in agile software develop-

ment practice. In Collaborative software engineering,

pages 93–108. Springer.

Rydenf

alt, C., Persson, J., Erlingsdottir, G., and Johansson,

G. (2019). ehealth services in the near and distant fu-

ture in swedish home care nursing. CIN: Computers,

Informatics, Nursing, 37(7):366–372.

Sahama, T., Simpson, L., and Lane, B. (2013). Security

and privacy in ehealth: Is it possible? In 2013 IEEE

15th International Conference on e-Health Network-

ing, Applications and Services (Healthcom 2013),

pages 249–253. IEEE.

Smeltzer, S., Mariani, B., and Meakim, C. (2017). Com-

municating with people with disabilities. http:

//www.nln.org/professional-development-programs/

teaching-resources/ace-d/additional-resources/

communicating-with-people-with-disabilities. Ac-

cessed: 24.11.2020.

Strobl, S., Bernhart, M., and Grechenig, T. (2010). An ex-

perience report on the incremental adoption and evo-

lution of an spl in ehealth. In Proceedings of the 2010

ICSE Workshop on Product Line Approaches in Soft-

ware Engineering, pages 16–23.

Svanborg-Sj

ovall, K. (2014). Swedish healthcare is the

best in the world, but there are still lessons to learn.

http://www.theguardian.com/public-leaders-network/

2014/jan/03/sweden-healthcare-coordinate-oecd.

Accessed: 06.07.2020.

Ventola, C. L. (2014). Mobile devices and apps for health

care professionals: Uses and beneﬁts. Pharmacy and

Therapeutics, 39(5):356.

Wadmann, S., Strandberg-Larsen, M., and Vrangbæk, K.

(2009). Coordination between primary and secondary

healthcare in denmark and sweden. International

journal of integrated care, 9.

Walsh, M. (2003). Teaching qualitative analysis using QSR

NVivo. The Qualitative Report, 8(2):251–256.

Wiklund Axelsson, S., Nyberg, L., N

aslund, A., and Me-

lander Wikman, A. (2013). The anticipated positive

psychosocial impact of present web-based e-health

services and future mobile health applications: An in-

vestigation among older swedes. International Jour-

nal of Telemedicine and Applications, 2013.

Yin, R. K. (1994). Case Study Research: Design and Meth-

ods. Sage, Thousand Oaks, CA, 1st edition.