Informed Consent as Patient Driven Policy for Clinical Diagnosis and
Treatment: A Smart Contract Based Approach
Md Al Amin
a
, Amani Altarawneh
b
and Indrajit Ray
c
Computer Science Department, Colorado State University, Fort Collins, Colorado, U.S.A.
Keywords:
Informed Consent, Patient, Provider, Clinical Treatment, Disease Diagnosis, Blockchain, Smart Contract.
Abstract:
Digitized healthcare systems improve services, make it easier for healthcare providers to work together, im-
prove the accuracy of diagnoses, and get the most out of each treatment. They provide healthcare services
that are better, faster, more reliable, and less expensive. With the help of information technology, comput-
ing resources, and digitized health records, medical researchers are trying to solve critical health problems
like COVID-19. However, electronic healthcare systems significantly risk patients’ data privacy and security.
Anyone with credentials can access patients’ healthcare data. Patients grant consent to share or access data.
But they need a way to ensure informed consent is done right and on time. Due to the centralized authority
in present healthcare systems, healthcare-covered entities perform all operations. As a result, many unwanted
events and security incidents happen in healthcare systems. Patients must know how their data is accessed, by
whom, and when. Therefore, a blockchain and smart contract-based patient-informed consent management
system is proposed. Where patients provide informed consent to share or access their health records, as well as
methods to ensure that informed consent is properly completed. The immutability and auto-triggering proper-
ties of blockchain and smart contracts ensure the integrity and accountability of the given informed consent.
1 INTRODUCTION
Compared to paper-based systems, electronic health
record (EHR) systems make it easier for doctors to
work together, make diagnoses more accurate, speed
up treatment, and give doctors ready access to patient
medical records. As healthcare data become more
digitized, distributed, and interactive, the healthcare
ecosystem is increasingly becoming more concerned
about the security of EHR information and systems.
Several factors contribute to the increased vulnera-
bility of EHR systems. Health workers are often
under-trained and under-experienced in handling and
securely maintaining information systems. Software
bugs, security flaws, and human errors allow unau-
thorized users to enter these databases. Insider ad-
versaries can also get their hands on protected med-
ical information, which can cause sensitive patient
information to be lost, misused, or shared. Conse-
quently, healthcare providers’ responsibility to pro-
tect patient privacy and health records confidentiality
a
https://orcid.org/0000-0003-1700-7201
b
https://orcid.org/0000-0002-4885-350X
c
https://orcid.org/0000-0002-3612-7738
has increased significantly due to these factors in elec-
tronic health data processing (Sulmasy et al., 2017).
While better security and privacy technology are
needed to better protect patient data from these types
of breaches, there is ample evidence that shows that
improper policy adoption, implementation, and en-
forcement cause a significant amount of unauthorized
access without a “need to know” to EHR data
(Staff, 2016; Seh et al., 2020; Marchand-Melsom
and Nguyen Mai, 2020). Intentionally or unintention-
ally, access privileges are assigned to users when they
should not be. Policies are not followed correctly, and
access control rules are not checked or implemented
promptly. In some cases, it has been observed that
the same roles and privileges are assigned to all em-
ployees. Often, individual patient-level policies are
not enforced to the word. Gaps in auditing and moni-
toring have also been found; they are not done unless
there are serious complaints or a legal requirement to
do so. Informed consent policies, in particular, are
one group of policies that suffer considerably because
of such gaps in policy specification and enforcement.
Informed consent (Parvin, 2022; Lorenzini et al.,
2022) is a legal and ethical concept in healthcare that
refers to the process in which a patient voluntarily
Al Amin, M., Altarawneh, A. and Ray, I.
Informed Consent as Patient Driven Policy for Clinical Diagnosis and Treatment: A Smart Contract Based Approach.
DOI: 10.5220/0012090600003555
In Proceedings of the 20th International Conference on Security and Cryptography (SECRYPT 2023), pages 159-170
ISBN: 978-989-758-666-8; ISSN: 2184-7711
Copyright
c
2023 by SCITEPRESS Science and Technology Publications, Lda. Under CC license (CC BY-NC-ND 4.0)
159
agrees to a medical intervention, procedure, or treat-
ment after being fully informed of the risks, benefits,
and alternatives or agree to share personal health data
to participate in clinical trials or research experiments
(Falagas et al., 2009). Informed consent empowers
patients to make decisions about their health and well-
being that are in their best interests. The purpose of
informed consent is to protect the autonomy and dig-
nity of the patient by giving them control over their
health and treatment decisions. It helps to build trust
between the patient and the healthcare provider. In-
formed consent is typically governed by laws and reg-
ulations such as the General Data Protection Regula-
tion (GDPR) in Europe or the California Consumer
Privacy Act (CCPA) in the United States to avoid fu-
ture legal issues and conflicts (Gopal et al., 2023).
While informed consent plays a critical role in the
patient’s healthcare, it is quite challenging to man-
age informed consent properly, preserving the privacy
and security of patient health data. Patients often
consent to users like treatment team members, physi-
cians, nurses, support staff, lab technicians, insurance
companies, family member(s), and other providers.
When patients need more advanced care or counsel-
ing from doctors, like specialists, they must go to a
different hospital. Due to work transfers and family
movements, patients must move to other regions, like
new states or countries. It is sometimes unavoidable
to change health plans or insurance company cover-
age. Patients must revoke access rights granted in the
past but no longer needed or valuable. Patients also
give emergency access permissions where there might
be an emergency, like an accident or life-and-death
situation, and patients cannot approve access permis-
sion instantly. In this case, assigned emergency treat-
ment team members must access admitted patients’
data without their consent (de Oliveira et al., 2023).
We believe the following problems must be ad-
dressed to protect healthcare data from unauthorized
access and preserve patients’ autonomy over their
consent and healthcare resources.
- Individual patient-level policies are not enforced
properly, otherwise known as informed consent.
- Centralized hospital system acts as a single point
failure and source of truth of access audit trails.
- Lack of consent provenance information or unal-
tered source of given consent execution.
- Patients are not assured that the intended users ex-
ecute given consents. Other requests are denied.
- No guarantee that consents are executed only
when included conditions are satisfied; otherwise,
denied.
- Patients lack control over their consent to control
their health record access.
In this paper, we propose a decentralized dis-
tributed ledger (DLT) (Altarawneh et al., 2020) pow-
ered with smart contracts (Buterin et al., 2014) to
address the above-mentioned challenges and require-
ments. We propose a blockchain and smart contract-
based informed consent management and enforce-
ment framework, which runs on top of a public
blockchain such as Ethereum (Buterin et al., 2014).
This smart contract-based approach provides an au-
tomated system and guarantees the integrity and ac-
countability of the given informed consent. However,
this proposed approach does not address patients’ pro-
tected healthcare data. For this reason, we do not fo-
cus on healthcare data security and privacy issues.
DLTs such as blockchain provide proofs that sub-
mitted acts are immutable; it keeps the audit trail’s in-
tegrity and can detect any changes that shouldn’t have
been made. Blockchain security properties, such as
nonrepudiation, are also guaranteed, where no partic-
ipants can deny submitting changes. Smart contracts
ensure the proper execution of informed consent and
prevent random access. Also, smart contracts emit
event information if there are any operations. To the
best of our knowledge, this work is the first to cap-
ture patients’ informed consent for disease diagnosis
and clinical treatment. Our contributions include the
following:
- A novel way of implementing Informed Consent
for medical diagnosis and treatment. Also, storing
in decentralized and distributed networks to over-
come a single point of truth sources and failure.
- Enforcing mechanism to authorization module
ensuring informed consent is evaluated and en-
forced while making authorization to access pa-
tient health records.
- Some consent services are proposed with sample
graphical representations to provide patients with
concise and informative reflections on informed
consent.
- Certain fundamental consent management consid-
erations and operations to support patients’ treat-
ment flow and avoid unwanted health data access.
The remainder of the paper is organized as fol-
lows. We discuss some related works in Section 2.
Section 3 explains the proposed system with required
components. Sections 4 and 5 discuss consent ser-
vices and management. Section 6 contains the exper-
imental evaluation of the proposed model. In Section
7, we wrap up the paper with what to do next.
SECRYPT 2023 - 20th International Conference on Security and Cryptography
160
2 RELATED WORK
Several suggestions exist for adopting blockchain
technology in healthcare and e-health systems. So far,
this research has been on how blockchain can protect
medical information and store and share medical data,
analytics, and informed consent systems for clinical
or research experiments. Research on informed con-
sent for clinical diagnosis and treatment is focused to
some extent (Ploug and Holm, 2012). To our knowl-
edge, ours is the first work on employing blockchain
and smart contracts for clinical treatment informed
consent management and enforcement.
Authors in (Cunningham et al., 2022) propose
Non-Fungible Tokens (NFTs) as the mechanism for
recording and transmitting records of patients’ con-
sent for medical data use. The proposed model en-
ables subjects to record signed documents of consent
that permit Data Consumers to request medical data
from Data Providers in line with the consent given
by the subjects to whom that data pertains. However,
how NFTs can be used for provenance with regulatory
policies such as HIPAA/GDPR remains to be seen.
In (Azaria et al., 2016), authors propose MedRec,
a blockchain-based healthcare data access and per-
mission management system to handle electronic
medical records. The model addressed four signif-
icant issues: fragmented, slow access to medical
data; system interoperability; patient agency; and im-
proved data quality and quantity for medical research.
MedRec gives patients a complete, unchangeable log
of their medical information and makes it easy to get
it from their providers and treatment sites. References
to different kinds of medical data are put together and
encoded on a blockchain ledger to create an accessi-
ble trail for medical history.
Xia et al. (Xia et al., 2017) propose a blockchain-
based data-sharing framework that addresses the ac-
cess control challenges associated with sensitive data
stored in the cloud using the blockchain’s immutabil-
ity and built-in autonomy properties. Yue et al. (Yue
et al., 2016) propose a blockchain-based app, Health-
care Data Gateway, where patients securely own, con-
trol, and share their data. Untrusted entities can han-
dle healthcare data using secure multi-factor comput-
ing to protect patient privacy. In (Fan et al., 2018),
the authors propose a blockchain-based information
management system, MedBlock, that allows efficient
EMR access and retrieval. It protects users’ pri-
vacy with custom access control protocols and en-
cryption technology while sharing data. Zyskind et
al., (Zyskind et al., 2015) propose blockchain for ac-
cess control management and secure data storage. En-
crypted data is kept on servers that can be trusted.
A blockchain-based dynamic consent manage-
ment architecture, ConsentChain, is offered by (Al-
balwy et al., 2021) in facilitating the exchange of
clinical genomic data. ConsentChain is built on the
Ethereum platform, and smart contracts are used to
model the actions of patients (who can give or take
back permission to share their data), data creators
(who collect and store patient data), and data re-
questers (who need to query and access the patient
data). However, this work mainly focuses on pa-
tient genome data sharing with clinicians, researchers,
and bioinformaticians. Clinical treatment imposes
different requirements for consent management than
genome data sharing. Many users perform various op-
erations in treatment processes, such as reading, writ-
ing, modifying, and others, and access privileges are
assigned based on the role of the users. We propose
a consent management framework to address com-
plex permission assignment requirements for treat-
ment team members, insurance agents, external doc-
tors, pharmacists, etc.
The researchers,(Tith et al., 2020), propose an
e-consent management model that uses Hyperledger
Fabric blockchain and a purpose-based access control
scheme. All patient records, consents, and metadata
about data access are written on the blockchain and
are shared among the organizations taking part. A
Chaincode performs business logic for managing pa-
tient consent. Patients can create, update, and with-
draw their consent in the blockchain. The proposed
model can be used for data donation for biobank re-
search purposes besides sharing patient data. How-
ever, the Hyperledger blockchain is a permissioned
blockchain network where participants are limited
to the organizations. It doesn’t provide the public
eye to provide trust. We adopt the public Ethereum
blockchain network where participants with stakes
can join and maintain the ledger to provide immutable
information to untrusted network participants. Most
important, Ethereum’s public consensus mechanism
adds more transparency than a permissioned network.
In addition, Ethereum smart contracts are the most
used, and there are many projects of development and
refinements.
3 PROPOSED APPROACH
The main idea is to integrate components of informed
consent into the patient-provider agreement. Then
create and deploy smart contracts for informed con-
sent components in the blockchain network. The
authorization module calls the corresponding smart
contract for an access request to enforce informed
Informed Consent as Patient Driven Policy for Clinical Diagnosis and Treatment: A Smart Contract Based Approach
161
consent. The request specifies which subject wants
to perform which operation on what objects under
what constraints or conditions. Once the smart con-
tract is called and the authorization module decides,
the corresponding event information is recorded as
logs in the blockchain network. Smart contracts
auto triggering feature automatically emits event or
activity data without missing any specified and re-
quired components. Event or activity data is stored
on the blockchain network. The blockchain net-
work provides a distributed, immutable, and decen-
tralized storage platform. This ensures that deployed
smart contracts provide original consent made with
the agreement of the patient and provider. It also
ensures that event logs are the same as created and
stored without any modification. Figure 1 shows the
proposed approach. With the necessary parameters,
the following discusses the patient-provider agree-
ment, informed consent components, consent smart
contract generation, consent enforcement, and patient
interaction with the blockchain network services.
3.1 Patient-Provider Agreement (PPA)
The patient-provider agreement aims to determine
who is responsible for what in treatment. The goal
is to improve outcomes, lower risks, and educate pa-
tients better. A multi-center study (Pergolizzi et al.,
2017) evaluated the utility of the PPA, how readily pa-
tients understood it, its ability to educate patients in an
unbiased way about treatment, and the feasibility of
incorporating a PPA in clinical practice. Both patients
and doctors believe this PPA helped them decide on
a course of treatment and was fair in laying out the
treatment’s risks and benefits. Most patients reported
the PPA to be ”somewhat helpful” or ”very helpful”
in deciding on a course of treatment and ”easy to un-
derstand. A PPA, also known as a contract, differs
from organization to organization. Healthcare organi-
zations adjust what they need from patients and what
they expect from them to match those needs, treat-
ments, and responsibilities. This is done based on the
nature and needs of treatment and services. Also, the
components and representation of the PPA depend on
the hospital or clinic. Examples include general hos-
pitals, emergency rooms, urgent care or walk-in clin-
ics, dental care, cancer treatment, physiotherapy, etc.
The patient-provider agreement is depicted in Fig-
ure 1 with necessary components. The Patient-
Provider Agreement formally is composed of four tu-
ples:
PPA = (PC, PrC, ROC, ICC)
satisfying the following requirements:
(i) PC is a finite set of patient components contain-
ing patient’s personal information, contact in-
formation, mailing information, pharmacy infor-
mation, billing and insurance information, emer-
gency contact, and others. The patient is responsi-
ble for providing and maintaining valid informa-
tion for these components.
(i) PrC is a finite set of provider components, in-
cluding the treatment team, anonymous data shar-
ing for research, prescription, and others. Treat-
ment team members for a patient include doctors,
nurses, support staff, lab technicians, and billing
officers. As the treatment period for a patient, ev-
erything from treatment to insurance coverage and
billing is considered.
(iii) ROC is a finite set of regulatory and other com-
ponents. It has applicable security and privacy
policy to comply with the local government, reg-
ulatory agencies (HIPAA, GDPR), federal gov-
ernment, and foreign government requirements if
necessary.
(iv) ICC is a finite set of informed consent compo-
nents. It indicates the permission given by the pa-
tient to access healthcare data.
This work mainly focuses on ICC and does not
consider and discuss PC, PrC, and ROC. They are
the future scope of this paper. Algorithm 1 shows the
step-by-step instructions for creating a PPA with PC,
PrC, ROC, and ICC. A patient-provider agreement is
formed when a patient visits a hospital. The terms and
conditions of the contract make it invalid after a cer-
tain period. There may be several contracts for a sin-
gle patient. Several patient-provider agreements must
be created and properly documented to deliver health-
care services. Managing many contracts involves var-
ious things, such as contract creation, development,
testing, updating, etc. If the requests contain con-
tracts, the authorization module must consider those
with other required policies when making access de-
cisions. From Figure 1, it is seen that the proposed
model stores the integrity of a PPA to the blockchain
network to ensure the detection of any modification,
intentionally or unintentionally.
3.2 Informed Consent Components
Before giving consent, patients need to know every-
thing about the particular consent. Figure 2 shows
the informed consent conceptual framework structure.
The Informed Consent formally is composed of four
tuples:
IC = (U, O, OP, CON)
satisfying the following requirements:
SECRYPT 2023 - 20th International Conference on Security and Cryptography
162
Figure 1: Proposed Smart Contract-Based Informed Consent Management Framework.
Algorithm 1: Patient-Provider Agreement (PPA).
Input : Patient Components (PC), Provider Components (PrC),
Regulatory and Other Components (ROC), and
Informed Consent Components (ICC).
Output: An formal agreement between patient and provider
(PPA).
1 Input Parameters Initialization Patient-Provider Agreement
PPA
i
:= {PC
i
, PrC
i
, ROC
i
, ICC
i
} where i indicates patient
identity; PC := {PC
1
, PC
2
, PC
3
, ........PC
M
}
2 PrC := {PrC
1
, PrC
2
, PrC
3
, ........PrC
N
}
3 ROC := {ROC
1
, ROC
2
, ROC
3
, ........ROC
P
}
4 ICC := {ICC
1
, ICC
2
, ICC
3
, ........ICC
R
}
5 Agreement Parameters Hash Calculation /* Hash(input) is
a function to calculate hash of input */
6 Hash
PC
Hash(PC
1
, PC
2
, PC
3
, ........PC
M
)
7 Hash
PrC
Hash(PrC
1
, PrC
2
, PrC
3
, ........PrC
N
)
8 Hash
ROC
Hash(ROC
1
, ROC
2
, ROC
3
, ........ROC
P
)
9 Hash
ICC
Hash(ICC
1
, ICC
2
, ICC
3
, ........ICC
R
)
10 Hash
PPA
i
Hash(Hash
PC
, Hash
PrC
, Hash
ROC
, Hash
ICC
)
11 Patient-Provider Agreement Finalization
12 if PPA
i
is complete then
13 /* complete means presence of PC, PrC, ROC,
and ICC */
14 add PPA
i
to patient-provider agreement repository and
return ID
PPA
i
;
15 add ID
PPA
i
and Hash
PPA
i
to blockchain network;
16 else
17 Error: PPA
i
can not be created
18 /* incomplete patient-provider agreement */
19 end if
(i) U is a finite set of authorized users denoted as
{u
1
, u
2
, u
3
, .....}. The user can perform certain
operations on healthcare resources when certain
conditions are satisfied.
(i) O is a finite set of protected objects otherwise
known as protected healthcare resources. A
finite set of protected objects (O) denoted as
{o
1
, o
2
, o
3
, .....}.
(iii) OP is a finite set of operations denoted by
{op
1
, op
2
, op
3
, ...}. Operations represent the sys-
tem actions that authorized users can perform on
the objects. Examples of operations are read,
write, and update.
(iv) CON is a finite set of conditions. It indicates the
conditions that must be satisfied by the user to
perform operations on the protected objects. A
finite set of conditions, CON, can be denoted as
{con
1
, con
2
, con
3
, ...}.
There are many users in the healthcare system.
Each user plays a different role and responsibility in
performing their job. Treatment team members for a
patient include doctors, nurses, support staff, lab tech-
nicians, billing officers, the patient’s emergency con-
tact person, and other hospital employees assigned by
the authority. Some outsider members are insurance
agents, pharmacists or pharmacy technicians, doctors
or lab technicians from another hospital. As the treat-
ment period for a patient, everything from treatment
to insurance coverage and billing is considered. In-
formed consent users can be anyone from five groups
of people: (i) treatment team member, (ii) emergency
contract, (iii) external users, (iv) insurance company
agent, and (v) pharmacy. External users are from dif-
ferent hospitals when a patient is transferred for better
treatment if the situation demands it. Usually, exter-
nal users have temporary access to admitted patients’
health records.
The term object refers to an electronic version of a
patient’s medical history kept on file by the healthcare
provider over time. It may include all the administra-
Informed Consent as Patient Driven Policy for Clinical Diagnosis and Treatment: A Smart Contract Based Approach
163
Figure 2: Informed Consent Components.
tive and clinical information pertinent to the patient’s
care under a specific provider, such as demographics,
progress notes, issues, medications, vital signs, previ-
ous medical history, immunizations, laboratory infor-
mation, and radiology reports. These objects must be
protected from unauthorized users. The main purpose
of informed consent permitting by patients to users to
perform certain operations.
Many operations are executed by the authorized
in the healthcare industry to perform required op-
erations. Some common operations are view/read,
add/write, update/modify, delete, etc. In the view
operation, users can only view or read healthcare
records or resources if the request is valid and com-
plies with all applicable policies. The state of the
data is not changed in this operation, ensuring data
integrity. But it can compromise confidentiality and
privacy if the access requested is granted without ap-
propriate credentials. On the other hand, the write op-
eration changes the state of the records or healthcare
data. If proper policy enforcement is not ensured, it
breaks the integrity of the data.
There might be various constraints or conditions
under which certain consent can be enforced, rejected,
revoked, and others. The conditions can be but are not
limited to:
(i) Time Constraints. In time constraints, any user
can access a patient’s healthcare data within a cer-
tain time. For example, the time condition for
consent is regular office hours: 8 am-5 pm. In this
case, the request is rejected if any subject wants to
access the patient’s record beyond this time. The
attempt is recorded as an audit trail event.
(ii) Date Constraints. The date constraints limit the
calendar date. No access request is granted be-
yond the intended date.
(iii) Day Constraints. Day conditions can in-
clude work days (Monday-Friday), weekends
(Saturday-Sunday, holidays, etc. Based on the
day, the subject can access data. Suppose a reg-
ular doctor has a duty on workdays. On weekends
no access is given to that doctor.
(iv) Location-Based Constraints. The location-based
condition allows users to access information from
a certain location, like a hospital building, inside
an emergency room for treating emergency pa-
tients, and others.
(v) IP-Based Constraints. IP-based condition lim-
its healthcare users from accessing resources from
certain IPs. Devices IPs must be from the known
list; otherwise, no access is granted.
(vi) Access Frequency Dependent. A user can op-
erate for a certain number in access frequency-
dependent conditions. Suppose an external doctor
is given five times view permission. Once the doc-
tor reads the patient’s specified records five times,
the given consent is expired, and access is denied.
There is no access without getting new consent.
The above mentions list is not fixed for the con-
ditions, but we consider them for this study. There
might be other conditions depending on the treatment
SECRYPT 2023 - 20th International Conference on Security and Cryptography
164
nature, patient characteristics, provider business pol-
icy, nature, etc. With sophisticated technology, ma-
licious attackers can spoof the conditions to fool the
system into accessing healthcare data and other com-
promised credentials. Proper layered defense mecha-
nisms must be deployed to ensure that conditions’ cre-
dentials are accurate, not fabricated or manipulated.
3.3 Consent Smart Contract Generation
Once a patient-provider agreement or PPA is cre-
ated and stored in the repository, all informed consent
components are deployed as smart contracts. The pa-
tient owns all the deployed contracts. The authoriza-
tion module needs to access these smart contracts to
make decisions with other components such as sub-
ject attributes, object attributes, operations attributes,
environmental attributes, organizational policies, reg-
ulatory and other policies, and others required. Algo-
rithm 2 shows the steps to develop and deploy smart
contracts for informed consent components. The
smart contract deployment unit, SCDU, collects all
consent components from PPA and checks integrity
to confirm that collected consents are not modified de-
liberately or inadvertently. In step 3 in Figure 1, PPA
integrity as the hash from Algorithm 1 (Hash
PPA
i
) is
stored in the blockchain network along with PPA id.
To verify PPA integrity, SCDU calls the corre-
sponding smart contract function to retrieve the PPA
hash value stored in the network. Any modification of
consent components voids the consent. If there is no
modification, then SCDU creates and deploys smart
contract(s) to the blockchain network. Once the con-
tracts are deployed, the contract addresses are added
to the patient’s profile and hospital systems. The con-
tract address is an identifier for a smart contract in the
blockchain network.
3.4 Consent Enforcement
Enforcing informed consent is essential to protect pa-
tient’s rights and autonomy and ensure that medical
treatments are performed with the patient’s complete
understanding and agreement. Healthcare organiza-
tions and providers must take the informed consent
process seriously and provide patients with all the
information they need to make informed decisions
about their medical care. Consent enforcement en-
sures that related consents are executed while mak-
ing access decisions for the requests. In the proposed
model, all consents are stored on the blockchain net-
work as smart contracts and can not be enforced un-
til they are called. The authorization module (AM)
considers a patient’s consent while making an autho-
Algorithm 2: Informed Consent Smart Contract.
Input : Informed Consent Component ICC.
Output: Smart contract contains informed consents elements
ICC from patient-provider agreement PPA.
1 Initialization Informed Consent IC
i
:= {Sub, Op, Ob j, Cond}
where Sub, Op, Ob j, Cond represent one or more individual
attribute(s) and i represents patient identity; Subject Attributes
Sub := {SubAttr
1
, SubAttr
2
, ........SubAttr
M
} Operation
Attributes Op := {OpAttr
1
, OpAttr
2
, ........OpAttr
N
} Object
Attributes Ob j := {Ob jAttr
1
, Ob jAttr
2
, ........Ob jAttr
P
}
Conditions Cond := {CondAttr
1
, CondAttr
2
, ........CondAttr
R
}
2 Smart Contract Generation and Deployment if IC
i
is complete
then
3 /* complete means presence of Sub, Op, Ob j,
and Cond with patient consent */
4 IC
i
is added to smart contract
5 else
6 Denied: smart contract for IC
i
can not be created and
deployed
7 /* incomplete informed consent component */
8 end if
rization decision. The user submits requests to AM
for authorization to operate on an object. The user
also must provide the required credentials for identi-
fication purposes. The AM needs to consider differ-
ent attributes with required policies for making deci-
sions. It also requires considering informed consent
from the patient for that requested subject. The other
points are subject, object, operation, and environmen-
tal attributes.
Figure 3: Informed Consent Enforcement Process.
Figure 3 shows the sequence diagram of AM and
informed consent interaction. When a subject submits
a request to AM. Then the AM queries the blockchain
Informed Consent as Patient Driven Policy for Clinical Diagnosis and Treatment: A Smart Contract Based Approach
165
network through the corresponding smart contract to
get the informed consent information for the sub-
ject. If the subject has consent from the patient for
the requested resources and other conditions, then the
AM evaluates other parameters and informed consent
to make final authorization. Finally, the AM gen-
erates the granted decision otherwise rejected for a
legitimate request. After deciding, AM sends audit
trail information to the smart contract to write in the
blockchain network. For this study, it is considered
that the authorization module is not compromised or
tampered with. Also, the communication channel be-
tween AU and the smart contract access points or apps
is secured from malicious users.
3.5 Patient Interaction
Patients must interact with the proposed system using
a GUI, apps, or abstraction. Different wallets, such as
Coinbase and MetaMask, interact with the blockchain
network to sign the transactions and manage cryp-
tocurrencies or tokens (He et al., 2020). Wallets store
private keys and other credentials for the users. Var-
ious types of users need to use the system. Some
users need specially designed software interfaces or
apps to interact with the systems (Mtshali and Khu-
bisa, 2019). These users include older adults or se-
nior citizens with limited knowledge of information
technology, physically disabled people, minors, and
other notable people. However, this work focuses on
designing mechanisms to record patients’ informed
consent for disease diagnosis and clinical treatment.
Hospitals or healthcare providers can serve the unique
needs of particular users for their patients. Patients
do not need to understand the underlying technolo-
gies, such as blockchain, smart contracts, distributed
systems, consensus mechanisms, etc. We assume that
patients’ devices, as well as apps, are protected from
intruders. Also, communication between patients’ de-
vices/apps and blockchain network nodes is secured.
3.6 Consent Operational Costs
Some blockchain-based frameworks need transaction
fees, like Gas in Ethereum. The gas consumption or
transaction fee is considered for research and techni-
cal aspects, not from the patients’ or users’ perspec-
tive. Healthcare providers can spend on infrastructure
expenses such as blockchain network nodes, apps for
mobile devices to interact with hospitals, blockchain
systems, and others. There are direct costs regarding
storing informed consent on public blockchain net-
works like Ethereum. The patients, insurance com-
panies, and others can cover these costs, like doctors’
fees, medications, pathology lab tests, radiology lab
tests, and other direct/indirect costs related to treat-
ment. In blockchain networks, state change opera-
tions require spending money, while reading from the
network does not need monetary expenditure. Once
informed consents are deployed to the blockchain net-
works, relevant users can access them without spend-
ing charges.
3.7 Consent Indexing and Query
We know the importance of informed consent index-
ing and query mechanism to provide efficient queries
for consent enforcement and patient interaction. For
this work, we assume that the search operations are
done by the corresponding smart contracts efficiently
from the blockchain network. However, detailed in-
dexing and search mechanisms are our future com-
munication.
4 CONSENT SERVICES
Patients need to know they can get information about
their consent given to whom, for what purpose, on
what resources, and under what conditions. They
can also learn how their consent is carried out, such
as who does what operation and when. This section
discusses user and resource-oriented consent services
available for patients through the proposed framework
with the required credentials. There might be other
services for given and executed consents, such as
operation-oriented, date-oriented, and so on (Albalwy
et al., 2021). Additional services are not discussed
in this paper because of space limitations. However,
there should be some assurances regarding the given
and executed consents. Guarantees can be served
through different consent services. The main objec-
tive is to provide patients with concise and informa-
tive reflections on informed consent.
User-Oriented Services. In this service mode, a
patient is aware of a specific user’s consent. It dis-
plays a list of resources for which the user has ob-
tained the patient’s permission to perform certain op-
erations. The application conditions, such as access
frequency, start and end dates, and other specific re-
quirements, are also included. The audit trails are
available to the patient if a user performs any opera-
tion on any resource. So that the patient can check all
of the user’s consents. Figure 4 depicts all given con-
sents for a doctor, David, on patient Jordan
s health
records: Visit Notes, Prescription, Radiology Lab Re-
port, Pathology Lab Report, and Immunization His-
tory with operations and conditions. While all exe-
SECRYPT 2023 - 20th International Conference on Security and Cryptography
166
cuted consents are shown in Figure 5 with operations,
access frequency, and access time.
Figure 4: User-Oriented Given Consents.
Figure 5: User-Oriented Executed Consents.
Resource-Oriented Services. Patients may need to
know the given and executed consent for a particu-
lar resource or object. In an object-oriented given
consent service, all permissions are listed, like who
has which operation consent under what conditions.
Figure 6 shows a sample presentation of an object-
oriented given consent. Each user has some opera-
tions and requirements for the object. While Figure
7 depicts the executed authorizations for an object. It
shows several events where each event contains which
subject performs what operation when and other in-
formation.
Figure 6: Object-Oriented Given Consents.
Patient consent service gives an overview of the is-
sued and executed consents. A patient needs to know
Figure 7: Object-Oriented Executed Consents.
the moment consent status. This service must be un-
ambiguous, uninterrupted, unaltered, real-time, and
easily understandable for the patients. In this work,
we present some possible applications of consent ser-
vices. In future directions, we want to implement
consent services using graph databases. Consent-
related information is collected from the blockchain
network and processed as graph databases to reflect
the consent status more presentable, understandable,
and meaningful for the patients.
5 CONSENT MANAGEMENT
This section briefly explains consent management
requirements and various operations: consent gen-
eration, modification, withdrawal, expiration, and
archiving. It is essential to ensure that operations do
not introduce privilege conflicts, leakages, or incom-
plete treatment teams. The most important point is
ensuring that consent modification, withdrawal, or ex-
piration does not interrupt the treatment process. For
example, if consent for an insurance agent is revoked,
that agent cannot access or process patient coverage
billed by the hospital. They can refuse the treatment if
the hospital is not paid promptly. Consent can include
information about the people using it, the resources
it will use, the activities it will be used for, certain
conditions, and other relevant information. Due to
page constraints, technical mechanisms or specifica-
tions for operations are not included, and they are the
future scope of this work.
Management Considerations. Consent manage-
ment is integral to providing healthcare services be-
cause it involves collecting, storing, and managing
information about patient consent. Effective consent
management requires paying close attention to sev-
eral important factors, such as: (i) legal and ethical
requirements; (ii) patient autonomy; (iii) data pri-
vacy and security; (iv) interoperability; (v) user ex-
perience; (vi) continuous improvement; and (vii) oth-
ers. Consent management is a complicated, multi-
step process that needs to be carefully thought out
to work well, be efficient, and meet all relevant stan-
dards and requirements.
Informed Consent as Patient Driven Policy for Clinical Diagnosis and Treatment: A Smart Contract Based Approach
167
Consent Generation. In this process, new consent is
created or generated. The detailed functionalities are
given in section 3 with the necessary components and
their interactions. Further consent can cause conflict
with the existing consent. So, it needs to be checked
before adding permission to the repository to ensure
no conflict of privileges. Once the verification is done,
the approval is activated for the users.
Consent Modification. Sometimes, it is neces-
sary to update a consent for various reasons: (i) hav-
ing error(s), (ii) modifying current user(s), object(s),
or condition(s), (iii) adding new user(s), entity(s), or
condition(s), (iv) dropping user(s), object(s), or con-
dition(s), and others. If any modification occurs, the
old consent is added to the consent archive reposi-
tory and referred to as a pointer to the new permis-
sion. When giving and taking consent, the wrong peo-
ple, resources, operations, or conditions can be added
unintentionally. This could lead to some unwanted
events, including security incidents. Once a mistake
is realized, it must be fixed immediately to avoid un-
wanted incidents.
Consent Withdrawn. Approval withdrawal happens
if patients do not want to share their data anytime.
In another case, if consent is given to the wrong per-
son or includes some extreme conditions, the given
authorization can be revoked by the patient or hos-
pital authority. When consent is withdrawn, related
users or entities must be notified. It is added to the
consent archive database to solve legal and regulatory
problems if consent is ever taken away. If given con-
sents are essential to the patient’s treatment and are
taken out before the treatment is over, there may be
some consequences, like an interruption in treatment
or drug supply.
Consent Expiration. A given consent can be voided
if unmet conditions exist. Conditions could be a spe-
cific date, access frequency, and so on. For example,
a doctor consents to view or read healthcare data for a
particular patient. The approval includes a condition
that access can be granted five times. The consent ex-
pires when the doctor reads or views the patient’s data
five times. If the doctor tries to access it for the sixth
time, the system fails to authorize it since there is no
consent for this access attempt. There can be multiple
conditions for approval that keep the support active or
valid. All conditions should be checked automatically
by systems instead of manually. Manual checking can
introduce delays or be overlooked by the correspond-
ing users.
Consent Archiving. Modified, withdrawn, and ex-
pired consents are added to this read-only repository.
No consent is active in this database. The main goal
of this repository is to store consent metadata that can
be used to answer any questions that come up because
of legal or regulatory requirements. Patients can also
see all the previously given consents that have been
modified, withdrawn, or expired.
Implementation. We implement a smart contract,
”ConsentManagement, in Ethereum using Solidity
language to perform consent management operations.
Consent generation is done by the proposed smart
contract deployment unit in Section 3.3 using Algo-
rithm 2. In addition, the ConsentManagement con-
tract would perform consent modification and with-
drawal invoked by the consent owner or patient. Con-
sent expiration and archiving operations must be done
automatically as default functions when the condi-
tions are present.
6 EXPERIMENTAL EVALUATION
We implement smart contracts to interact with the
Ethereum test network, Goerli, to store and retrieve
given informed consent. We use the Python web3
framework Brownie and Ethereum solidity. We set
up a multi-signature wallet to perform consent man-
agement operations. It is two members and two out
of two (2 out of 2) control. Every member must sign
in to store the consent in the blockchain network. The
Healthcare Provider System deploys smart contracts
in the Goerli network and transfers ownership to the
corresponding patients. We use Metamask wallet to
sign the transaction to interact with the Goerli net-
work. Goerli faucet ethers are used as gas to send the
transaction. We leverage Infura API to interact with
the blockchain networks.
The Ethereum blockchain platform is selected for
the experimental purpose. It offers Turing complete
smart contract language, Solidity, and Vyper, to im-
plement the logic for the proposed model. Since de-
ployed codes stay in the blockchain network and can
not be changed. Deploying smart contract to the main
network with bugs or errors cost money and reputa-
tion. So it is essential to test smart contracts before
deploying them to the Ethereum main network. Go-
erli and Sepolia Ethereum test networks (Sivaselvan
et al., 2023) are currently available for testing smart
contracts. In the following, we discuss the main re-
sources: (i) time, (ii) computational, and (iii) gas re-
quired to deploy and evaluate the functionalities of
the proposed approach. Other resources are not con-
sidered for this work.
Time. Smart contract deployment and execution
stages are the basis of the time cost associated with
on-chain activities. Since the time it takes to double-
check and sign a transaction is entirely up to the user,
SECRYPT 2023 - 20th International Conference on Security and Cryptography
168
it is disregarded. The duration of a smart contract’s
deployment and execution is determined mainly by
how long it takes for the appropriate transaction to
be included in a block by miners. So, on average,
the duration of either process is about the same as
the time it takes for a transaction to be confirmed on
the blockchain technology of ones choosing. A new
block is added to the Ethereum blockchain every 12
seconds, which is ideal for our purposes. It was 15
seconds in 2019 (Pierro and Rocha, 2019). So long
as there is sufficient place in new blocks, a new trans-
action would take, on average, no more than 12 sec-
onds. The time it takes for a transaction to propagate
is more difficult in practice. If block congestion oc-
curs, the time it takes for a transaction to be included
in a block might increase. However, blockchain users
can influence this time by paying more gas for faster
confirmation. Given that users may artificially extend
the confirmation time of their transactions.
Computational Resource. To deploy smart contracts
on the Ethereum test network, Goerli, the proposed
method uses Infura blockchain API services. Infura
(Panda and Satapathy, 2021) is a suite for building
blockchain applications. It has tools for developers
and application programming interfaces (APIs). In-
fura also gives developers quick and reliable access
to the Ethereum network, which lets them build so-
phisticated next-generation software and Web3 apps
that can grow to meet user demand. So, there is no
need to keep a blockchain node running to use smart
contracts. We assume that other resources like the
CPU, HDD, and communication bandwidth on the lo-
cal machine are negligible.
Gas Consumption. All the read calls of smart con-
tracts are gas-free. Gas is needed for any activity
on Ethereum that involves writing or changing data
(Wood et al., 2014). Some functions are sending ether
(or any other ERC20 token), minting and sending
NFTs, deploying smart contracts, changing the state
of the blockchain, and so on. For this work, we only
need to consider smart contract deployment and func-
tion calling costs to write data on the blockchain net-
work. The cost of smart contract deployment is pro-
portional to the size of the code. This is a one-time
cost for a single-contract deployment. How much it
costs to call a function depends on how many times
it is called and how much data needs to be stored or
changed on the blockchain network. Code optimiza-
tion is crucial to reduce gas costs. But instead of re-
ducing gas use, we focus on the model’s functions.
7 CONCLUSIONS
Consent to treatment is a crucial aspect of ethical and
practical healthcare delivery. It is necessary to ensure
that patients are fully informed, involved, and in con-
trol of their health and treatment decisions. Therefore,
healthcare providers must take the necessary mea-
sures to ensure patients understand their treatment
options and make educated decisions regarding their
treatment processes. In addition, it helps build trust,
empower patients, and improve the quality of care
provided. Informed consent is an ongoing process;
patients can change their minds and withdraw their
consent anytime. Also, it is not a one-time event but a
continuous process that starts before the intervention
and continues throughout the patient’s care.
Smart contract-based patient consent management
frameworks are emerging as a promising solution to
the challenges of managing health data in a secure and
privacy-preserving manner. Such frameworks ensure
that patients have control over their protected health
information (PHI) and can provide informed consent
for its use by healthcare providers, researchers, and
other stakeholders. In addition, blockchain technolo-
gies can ensure that consent is managed securely and
efficiently. At the same time, they can provide decen-
tralization, transparency, and immutability that can be
leveraged to improve healthcare auditability and ac-
countability for all stakeholders involved. Such in-
herent characteristics make it a potential solution for
healthcare data systems concerning sharing and pa-
tient privacy.
This work identifies some important consent man-
agement operations for the uninterrupted treatment
process. However, there is a need for those oper-
ations to be specified and executed technically. In
the future direction, the detailed technical approach
will be worked on. Future research could also exam-
ine the ethical considerations of using blockchain and
smart contract technology in healthcare, implement-
ing and adopting smart contract-based systems in ac-
tual healthcare settings, and integrating smart con-
tracts with other emerging technologies.
ACKNOWLEDGEMENTS
This work was partially supported by the U.S. Na-
tional Science Foundation under Grant No. 1822118
and 2226232. Any opinions, findings, and conclu-
sions or recommendations expressed in this material
are those of the authors and do not necessarily re-
flect the views of the National Science Foundation,
or other federal agencies.
Informed Consent as Patient Driven Policy for Clinical Diagnosis and Treatment: A Smart Contract Based Approach
169
REFERENCES
Albalwy, F., Brass, A., Davies, A., et al. (2021). A
blockchain-based dynamic consent architecture to
support clinical genomic data sharing (consentchain):
Proof-of-concept study. JMIR medical informatics,
9(11):e27816.
Altarawneh, A., Herschberg, T., Medury, S., Kandah,
F., and Skjellum, A. (2020). Buterin’s scalability
trilemma viewed through a state-change-based classi-
fication for common consensus algorithms. In 2020
10th Annual Computing and Communication Work-
shop and Conference (CCWC), pages 0727–0736.
Azaria, A., Ekblaw, A., Vieira, T., and Lippman, A. (2016).
Medrec: Using blockchain for medical data access
and permission management. In 2016 2nd Inter-
national Conference on Open and Big Data (OBD),
pages 25–30. IEEE.
Buterin, V. et al. (2014). A next-generation smart contract
and decentralized application platform. white paper,
3(37):2–1.
Cunningham, J., Davies, N., Devaney, S., Holm, S., Hard-
ing, M., Neumann, V., and Ainsworth, J. (2022). Non-
fungible tokens as a mechanism for representing pa-
tient consent. Studies in Health Technology and Infor-
matics, 294:382–386.
de Oliveira, M. T., Verginadis, Y., Reis, L. H., Psarra, E., Pa-
tiniotakis, I., and Olabarriaga, S. D. (2023). Ac-abac:
Attribute-based access control for electronic medical
records during acute care. Expert Systems with Appli-
cations, 213:119271.
Falagas, M. E., Korbila, I. P., Giannopoulou, K. P., Kondilis,
B. K., and Peppas, G. (2009). Informed consent: how
much and what do patients understand? The American
Journal of Surgery, 198(3):420–435.
Fan, K., Wang, S., Ren, Y., Li, H., and Yang, Y. (2018).
Medblock: Efficient and secure medical data shar-
ing via blockchain. Journal of medical systems,
42(8):136.
Gopal, R. D., Hidaji, H., Kutlu, S. N., Patterson, R. A., and
Yaraghi, N. (2023). Law, economics, and privacy:
Implications of government policies on website and
third-party information sharing. Information Systems
Research.
He, D., Li, S., Li, C., Zhu, S., Chan, S., Min, W., and
Guizani, N. (2020). Security analysis of cryptocur-
rency wallets in android-based applications. IEEE
Network, 34(6):114–119.
Lorenzini, G., Shaw, D. M., Arbelaez Ossa, L., and El-
ger, B. S. (2022). Machine learning applications in
healthcare and the role of informed consent: Ethical
and practical considerations. Clinical Ethics, page
14777509221094476.
Marchand-Melsom, A. and Nguyen Mai, D. B. (2020). Au-
tomatic repair of owasp top 10 security vulnerabil-
ities: A survey. In Proceedings of the IEEE/ACM
42nd International Conference on Software Engineer-
ing Workshops, pages 23–30.
Mtshali, P. and Khubisa, F. (2019). A smart home appliance
control system for physically disabled people. In 2019
Conference on Information Communications Technol-
ogy and Society (ICTAS), pages 1–5. IEEE.
Panda, S. K. and Satapathy, S. C. (2021). An investigation
into smart contract deployment on ethereum platform
using web3. js and solidity using blockchain. In Data
Engineering and Intelligent Computing: Proceedings
of ICICC 2020, pages 549–561. Springer.
Parvin, A. (2022). Jurisprudential justification of informed
consent in medical practice: A critical approach. Issue
5 Int’l JL Mgmt. & Human., 5:956.
Pergolizzi, J. V., Curro, F. A., Col, N., Ghods, M. P.,
Vena, D., Taylor, R., Naftolin, F., and LeQuang, J. A.
(2017). A multicentre evaluation of an opioid patient–
provider agreement. Postgraduate medical journal,
93(1104):613–617.
Pierro, G. A. and Rocha, H. (2019). The influence factors on
ethereum transaction fees. In 2019 IEEE/ACM 2nd In-
ternational Workshop on Emerging Trends in Software
Engineering for Blockchain (WETSEB), pages 24–31.
IEEE.
Ploug, T. and Holm, S. (2012). Pharmaceutical infor-
mation systems and possible implementations of in-
formed consent-developing an heuristic. BMC Medi-
cal Ethics, 13(1):1–12.
Seh, A. H., Zarour, M., Alenezi, M., Sarkar, A. K., Agrawal,
A., Kumar, R., and Ahmad Khan, R. (2020). Health-
care data breaches: insights and implications. In
Healthcare, volume 8, page 133. Multidisciplinary
Digital Publishing Institute.
Sivaselvan, N., Bhat, V., Rajarajan, M., and Das, A. K.
(2023). A new scalable and secure access control
scheme using blockchain technology for iot. IEEE
Transactions on Network and Service Management.
Staff, C. P. S. (2016). Regulatory compliance/lessons
learned from hipaa enforcement. Journal of the Cali-
fornia Dental Association, 44(11):703–704.
Sulmasy, L. S., L
´
opez, A. M., and Horwitch, C. A. (2017).
Ethical implications of the electronic health record: in
the service of the patient. Journal of general internal
medicine, 32(8):935–939.
Tith, D., Lee, J.-S., Suzuki, H., Wijesundara, W., Taira, N.,
Obi, T., and Ohyama, N. (2020). Patient consent man-
agement by a purpose-based consent model for elec-
tronic health record based on blockchain technology.
Healthcare Informatics Research, 26(4):265–273.
Wood, G. et al. (2014). Ethereum: A secure decentralised
generalised transaction ledger. Ethereum project yel-
low paper, 151(2014):1–32.
Xia, Q., Sifah, E. B., Smahi, A., Amofa, S., and Zhang,
X. (2017). Bbds: Blockchain-based data sharing for
electronic medical records in cloud environments. In-
formation, 8(2):44.
Yue, X., Wang, H., Jin, D., Li, M., and Jiang, W. (2016).
Healthcare data gateways: found healthcare intelli-
gence on blockchain with novel privacy risk control.
Journal of medical systems, 40(10):218.
Zyskind, G., Nathan, O., et al. (2015). Decentralizing pri-
vacy: Using blockchain to protect personal data. In
2015 IEEE Security and Privacy Workshops, pages
180–184. IEEE.
SECRYPT 2023 - 20th International Conference on Security and Cryptography
170