Design of Blockchain-based Precision Health-Care Using Soft Systems Methodology
| DOI | https://doi.org/10.1108/IMDS-07-2019-0401 |
| Published date | 31 December 2019 |
| Date | 31 December 2019 |
| Pages | 608-632 |
| Author | Ravi Sharma,Charcy Zhang,Stephen C. Wingreen,Nir Kshetri,Arnob Zahid |
| Subject Matter | Information & knowledge management,Information systems,Data management systems,Knowledge management,Knowledge sharing,Management science & operations,Supply chain management,Supply chain information systems,Logistics,Quality management/systems |
Design of Blockchain-based
Precision Health-Care Using Soft
Systems Methodology
Ravi Sharma
Department of Accounting and Information Systems, College of Business and Law,
University of Canterbury, Christchurch, New Zealand
Charcy Zhang
Center for Inclusive Digital Enterprise (CeIDE), Christchurch, New Zealand
Stephen C. Wingreen
Department of Accounting and Information Systems,
College of Business and Law,
University of Canterbury, Christchurch, New Zealand
Nir Kshetri
University of North Carolina, Greensboro, North Carolina, USA, and
Arnob Zahid
Department of Accounting and Information Systems,
College of Business and Law,
University of Canterbury, Christchurch, New Zealand
Abstract
Purpose –The purpose of this paper is to describe the application of soft systems methodology (SSM) to
address the problematic situation of low opt-in rates for Precision Health-Care (PHC).
Design/methodology/approach –The design logic is that when trust is enhanced and compliance is better
assured, participants such as patients and their doctors would be more likely to share their medical data and
diagnosis for the purpose of precision modeling.
Findings –The authors present the findings of an empirical study that confronts the design challenge of
increasing participant opt-in to a PHC repository of Electronic Medical Records and genetic sequencing.
Guided by SSM, the authors formulate design rules for the establishment of a trust-less platform for PHC
which incorporates key principles of transparency, traceability and immutability.
Research limitations/implications –The SSM approach has been criticized for its lack of “rigour”and
“replicability”. This is a fallacy in understanding its purpose –theory exploration rather than theory
confirmation. Moreover, it is unlikely that quantitative modeling yields any clearer an understanding of
complex, socio-technical systems.
Practical implications –The application of Blockchain, a platform for distributed ledgers, and associated
technologies present a feasible approach for resolving the problematic situation of low opt-in rates.
Social implications –A consequence of low participation is the weak recall and precision of descriptive,
predictive and prescriptive analytic models. Factors such as cyber-crime, data violation and the potential for
misuse of genetic and medical records have led to a lack of trust from key stakeholders –accessors,
participants, miners and regulators –to varying degrees.
Originality/value –The applicationof Blockchain as a trust-enablingplatform in the domain of an emerging
eco-system such as precision health is novel and pioneering.
Keywords Design Science Research, Trust-less platform, Digital Healthcare
Paper type Research paper
Industrial Management & Data
Systems
Vol. 120 No. 3, 2020
pp. 608-632
© Emerald PublishingLimited
0263-5577
DOI 10.1108/IMDS-07-2019-0401
Received 7 August 2019
Revised 24 November 2019
Accepted 11 December 2019
The current issue and full text archive of this journal is available on Emerald Insight at:
https://www.emerald.com/insight/0263-5577.htm
The authors are grateful to the numerous participants in the SSM portion of the research who must
remain anonymous as per the requirements of the research ethics agreement. Many thanks are also due
to the reviewers and Editor of IMDS for their constructive comments and input which has led to a
much improved article. Paula Wingreen proof-read the first submission and Adil Bilal provided
research support for the Nvivo analytics.
608
IMDS
120,3
1. Challenges facing Precision Health-Care
An emerging trendin the practice of medicine known as PrecisionHealth-Care (PHC) has been
suggested as a promising service on digital health ecosystems or clouds. It is defined as the
development of a quantitative model which links the individual EHRs to the population and
derives the benefit of aggregating EHRs with consideration to social context (Colijn et al.,
2017). Although there are distinctions made between PHC and personalized medicine, a
simplifying assumption is that the former is a system-level perspective whereas the latter is
patient-centric. More specifically, PHC comprises health and medical records which are
networked to various front-end clients. It is an instance of an evidence-based approachwhich
provides personalized medicine, including clinical decisions, treatments and products to the
individualpatient (Lu et al., 2014; Zimmerman, 2019).This approach is driven by data analytic
models, whichpossess the ability of dealingwith large amounts of genome information which
combine genetic diagnoses with EHRs or EMRs (Mehta and Pandit, 2018).
With the support of robust data analytics and machine learning, PHC is capable of
descriptive, predictive or prescriptive diagnostics by benchmarking individuals with the
populationinorderto“discover”diseases, treatments or outcomes (Colijn et al.,2017).Typically,
patients of universal healthcare who opt-in, consent to share their medical data (including
genetic sequence data) into such a data base. They are the actual cases used for the construction
of regression or machine learning models, which formulate prescriptions and predictions
linking to diagnosis, treatments and outcomes with anonymised patient profiles as moderators.
When PHC is applied as a service, input will be the patient’s profile and diagnosis report(s); and
the output will be the treatment (e.g. therapy, medication, behavioural changes and surgical
procedure). In a best-case scenario, PHC learns from each health event of individuals who opt-in
and provides more precise predictions along with prescriptions. However, the reluctance of
individuals to opt-in and share genetic and medical data results in a weaker analytic model and
machine learning environment and hence a tragedy-of-the-commons scenario for PHC.
With Digital Healthcare having “crossed the chasm”, a large volume, velocity, variety
and veracity of health data is produced and shared among numerous players in the
eco-system such as Patients (Consumers), (Health-Care) Providers, Payers, Vendors,
Infomediaries and Regulators (Stephanie and Sharma, 2016). The twin issues of security and
privacy are hence critical. Security refers to protection against the unauthorized access or
modification of health data such as controls in place to limit who can access the information.
Privacy is harder to define, in part because user-specific details (Personally Identifiable
Information (PII)), preferences and contexts, but also because it refers to what many believe
to be a “human right”. The idea of what constitutes PII is an important aspect of security
and privacy in digital health and contribute significantly to user experience[1].
Cyber-security in health-care is a major problem with “hundreds”of reported violations
(Williams, 2019). In reality, most people are unwilling to share their health data, considering
that such disclosure might negatively impact on their privacy (Patil and Seshadri, 2014).
An intrusive aspect of PHC is the requirement for genetic sequencing information from
patients. Research by Ponemon (2016) suggests that 38 per cent are not willing to participate in
genetic testing because of deep-rooted distrust. In current PHC platforms, patients’privilege
and control over their medical data are limited by insufficient data tran sparency; patients often
do not know or control who accesses their health records and for what reason (Colijn et al.,
2017). Neither are players in the PHC eco-system explicitly accountable for data breaches
(Das et al., 2016). The Trusted Third Party (TTP) –a service provider –neither has the moral
right nor the technical ability to mediate. However, unlike conventional EHRs and EMRs,
genetic information is not common in established clinical practice (Nguyen et al., 2014). For
better treatment outcomes, sharing of personal health information (e.g. historic medical records
of profiles, symptoms, treatments and outcomes) with patients’consent is necessary.
Nevertheless, the contribution of sensitive genetic information for the d evelopment of
609
Design of
Blockchain-
based PHC
using SSM
To continue reading
Request your trial