Technology in Society: Sciencedirect

Technology in Society 59 (2019) 101186
Contents lists available at ScienceDirect
Technology in Society
journal homepage: www.elsevier.com/locate/techsoc
Evaluation and selection of mobile health (mHealth) applications using AHP T

and fuzzy TOPSIS
Manindra Rajak∗, Krishnendu Shaw
Department of Management Studies, Indian Institute of Technology (Indian School of Mines) Dhanbad, Jharkhand, India
ARTICLE INFO ABSTRACT
Keywords: mHealth is an innovative, mobile application based system, which has the ability to assist people to manage their
mHealth health better. With the increasing number of mHealth applications, it is very difficult to choose the ideal ap-
MCDM plication. As of now, very limited studies have been carried out on mHealth application selection. To fill the
AHP existing research gaps, this study endeavours to develop a model for mHealth application selection by adopting a
Fuzzy TOPSIS
combined approach of AHP and fuzzy TOPSIS. The hierarchical model has been developed using factors iden-
Fuzzy set theory
eHealth
tified from literature review and expert opinions. The ambiguity in comparing different mHealth applications
has been handled by applying fuzzy set theory. The AHP has been used to determine the weights of criteria and
sub-criteria, and the fuzzy-TOPSIS method has been used to obtain the final ranking of the applications. The
applicability of the proposed model has been discussed through a numerical case example. The sensitivity
analysis has been carried out by changing the weights of the criteria. In this study, user satisfaction, function-
ality, easy to learn and use, and information quality, have come out as important factors in mHealth application
selection. The proposed method will help users as well as medical practitioners to select the proper mHealth
application in this digital world.
1. Introduction Razmak and Belanger [3]; Li et al. [4]. Specifically, mHealth aims to
deliver healthcare services to patients through mobile or wireless de-
Over a period of time, the healthcare sector has witnessed sig- vices.
nificant transformation through the intervention of information tech- As compared to eHealth, mHealth is more applicable, due to the
nologies (IT) [1]. The penetration of information technology has helped portability and increased flexibility of the system. The system can be
the healthcare sector from various dimensions, ranging from online used in rural areas where conventional medical facilities are hard to
data collection, internet assisted surgery, managing medical data reach. mHealth can be considered a low-cost substitute for delivering
through messaging systems, and many more. In the literature, mobile health facilities to citizens of a country. A considerable number of
assisted healthcare delivery system has been conceptualized as mobile studies focusing on mHealth have been carried out in various parts of
health (mHealth). Unfortunately, no widely accepted definition of the world [5–11]. The efficacy of mHealth based systems as compared
mHealth has been established until now. The mHealth concept has to the traditional healthcare delivery system is indeed a debatable issue.
emerged from the earlier concept of electronic health (eHealth) that However, mHealth technology is expected to bring a revolution in the
intended to deliver healthcare facilities using fixed IT and computer healthcare sector in the upcoming time due to the flexibility involved in
peripherals. The Global Observatory for eHealth (GOe) defined it. Marcolino et al. [12] reviewed the efficacy of mHealth tools for
mHealth as “medical and public health practice supported by mobile managing diseases and found mixed results. mHealth may become an
devices, such as mobile phones, patient monitoring devices, personal important supplementary tool for managing health-related issues in the
digital assistants (PDAs), and other wireless devices” [2]. Recently, near future.
enhanced mobile technologies (for example: 3G, 4G, and 5G) and the As compared to traditional healthcare providers, mHealth has the
miniaturization of computing devices (e.g. Android, Windows-based potential to reduce the cost of healthcare services [13]. Further,
mobile phones, iOS) have successfully transformed the facility oriented mHealth can deliver all the health-related information to the individual
eHealth system into the portable mHealth system. There are various that in turn will lead toward a healthy lifestyle [14]. mHealth has been
studies conducted in the context of eHealth. A few notable studies are found to be a helpful tool in managing chronic diseases, reducing health
∗
Corresponding author.
E-mail addresses: [email protected] (M. Rajak), [email protected] (K. Shaw).
https://doi.org/10.1016/j.techsoc.2019.101186
Received 24 April 2019; Received in revised form 31 July 2019; Accepted 14 August 2019
Available online 23 August 2019
0160-791X/ © 2019 Elsevier Ltd. All rights reserved.
M. Rajak and K. Shaw Technology in Society 59 (2019) 101186
problems, and increasing an individual's knowledge of their health. keywords; however, no exactly similar study was found on the topic of
An mHealth based system can play a crucial role in delivering mHealth application selection. Thus, this section discusses similar stu-
healthcare facilities to human beings with the increasing number of dies related to mHealth application selection. Lohan et al. [19] sug-
smart phone users. The app based medical service delivery system is gested an MCDM model for assessing the performance of eHealth
expected to become a game changer, especially in rural areas where technology. The authors used five criteria to evaluate the performance
people lack basic medical facilities. As per a recent study, the number of of eHealth, namely cost, acceptability, ease of use, accuracy and false
smartphone users may reach approximately 3.8 billion across the world alarm. In the same line, Liao and Qiu [20] proposed a model to assess
by 2021 [15]. mHealth would be an innovative application for the the performance of cloud computing based healthcare service systems.
twenty-first century. Factors like data storage security, flexibility, software scalability, etc.
As of now, mHealth studies have been carried out into the following were identified by the researchers and a hierarchical structure was
areas: (a) technology adoption of mobile health applications; (b) de- developed using the aforesaid factors. An AHP was applied to select
veloping new software applications related to health and fitness, and (c) efficient computing services [21]. proposed a framework considering
clinical trials on the adoption of the mHealth system [3,12,16,17]. tangibles, responsiveness, reliability, information quality and empathy
Unfortunately, limited studies have been conducted on the develop- criteria and nineteen sub-criteria for selecting superior hospital web-
ment of frameworks for selecting mHealth applications. Due to the sites from the alternatives, which satisfy the needs and expectations of
higher penetration of Android and Windows based mobile phones in the customers. Dey et al. [22] developed a model integrating AHP and lo-
market across the world, the selection of mHealth application has be- gical framework (logframe) to manage the performance of healthcare
come an important agenda. Nowadays people are heavily using services. Sloane et al. [23] developed a model based on AHP for se-
smartphone-based applications in their day to day life [18]. However, a lecting newborn ventilators at a new hospital for women's healthcare.
proper framework to assess the efficacy of mobile applications in the Lo et al. [24] ranked the web services using TOPSIS under fuzzy
market is missing in the archived literature. environment. They have used five main criteria (runtime, transaction,
To fill the gaps in existing research, this study proposes an Analytic cost, configuration management, security) and seventeen sub-criteria to
Hierarchy Process (AHP) and fuzzy Technique for Order of Preference select the best web service according to their group performances.
by Similarity to Ideal Solution (fuzzy-TOPSIS) based framework for Zaidan et al. [25] examined open source electronic medical records
evaluating mHealth applications. The goal of this study is not to de- software packages, using combined AHP and TOPSIS methodologies.
velop a new complex multi-criteria decision-making framework invol- The model was developed after deliberation of 13 software packages, 7
ving sophisticated mathematical functions and operations, but rather to criteria and 30 sub-criteria. The identified 7 criteria were usability,
establish a widely accepted, generalised model for decision makers who functionality and features, security, user support, developer support,
can apply it easily in real life scenarios. To fulfil this objective, widely customizability, and ease of installation. Shafii et al. [26] developed a
accepted methodologies are chosen for the proposed model. The se- performance assessment model by using 5 criteria and 9 sub-criteria for
lection of mHealth applications deals with various factors, which are evaluating performances at the managers' level in running hospitals.
primarily qualitative in nature. To handle the qualitative factors, a Their considered criteria were functional, professional, organizational,
multi-criteria decision making (MCDM) approach is adopted. individual and human.
In this study, fuzzy set theory is adopted for handling the indis- Kim and Kim [27] proposed an AHP model for selection of Internet
tinctness associated with the decision maker's subjective judgment. The of Things (IoT) applications taking cognizance of three main criteria,
contributions of the proposed research are twofold. Firstly, it assists i.e. technological prospect, market potential and regulatory environ-
health service providers as well as users to identify the relevant factors ment, and 11 sub-criteria. The study carried out a comparison among
for choosing mHealth applications. Secondly, the study explains the various applications like IoT Healthcare, IoT logistics and IoT energy
procedure of mHealth application selection through a numerical case Management. Grigoroudis et al. [28] developed a performance eva-
example. The remainder of the paper is arranged as follows. Section 2 luation system considering the balance scorecard for public healthcare
discusses the relevant literature related to the application of MCDM services. This evaluation system included both financial and non-fi-
techniques for technology selection. Section 3 illustrates the proposed nancial performance indicators in order to maintain the quality of
approach. Section 4 explains the numerical example of the proposed service level, customer satisfaction, organizational system of self-im-
approach. Section 5 discusses sensitivity analysis. Finally, section 6 provement, the capability of adaptation and change in the organization.
concludes the study. Further, Striteska and Spickova [29] studied different performance
evaluation systems by analysing, comparing and summarizing the
2. Literature review strong and weak points through conducting a literature review. There
are various factors which are relevant for the selection of mHealth
mHealth has evolved from the earlier concept of eHealth, where application. However, this study has selected a few factors (given in
health related information is shared with the patients or the users using Table 1) for rating mHealth applications. The factors have been iden-
the fixed setup of information technologies. A few notable studies on tified taking opinions from academics as well as medical practitioners.
eHealth are as follows: Razmak and Belanger [3] explored eHealth
usage by medical practitioners and patients in the context of Canada. Li 3. Research methodology
et al. [4] found an increasing trend of adopting eHealth by the users in
USA. The current study applies combined methodologies, consisting of
As of now, a widely accepted definition of mHealth is still missing in AHP and fuzzy TOPSIS. The details of the methodologies are discussed
the literature. The present study intends to develop a framework for as follows. Firstly, a hierarchical structure is constructed using nine
mHealth application selection. Selecting mHealth applications is a criteria and thirty two sub-criteria identified through the literature
MCDM problem. Thus, the paper aims to shed light on the literature of review and taking opinions from the three experts. Table 1 depicts the
MCDM as well as the literature pertaining to mHealth. The relevant operational definitions of the various criteria.
studies in this regard are discussed below. Further, AHP is applied to determine the relative weights of criteria
and sub-criteria. Then these weights are further used in TOPSIS to rate
2.1. Related studies on the mHealth application selection different mHealth alternatives in a fuzzy environment. In this study,
fuzzy set theory is used to manage the uncertainties of the evaluation
The literature on mHealth was searched in popular scientific data- process. Finally, the proposed model is discussed through a case ex-
bases (Science Direct, Springer, Emerald, Google scholar) using various ample.
2
Table 1
mHealth application evaluation performance criteria, definition and its references.
Criteria Operational Definition References
User Satisfaction How the products and services meet the user's expectations. [30–33]
Compatibility A condition in which products or services can co-exist without conflict. [34–37]
Functionality Ability to perform defined operations in a specified manner. [38–40]
Security Protection of data/services against all kind of potential threats that may harm stored data [37,41–43]
Accessibility Authorization, and approachability to secured documents and retrieving information from that easily [44,45]
Easy to learn and Use Ease in accessing the product and service. [46–48]
Empathy Ability to perceive the expectation from that product keeping himself in position. [21,49–51]
Information Quality Usefulness of the information provided by given product or service based on user's expectations. [52–55]
Responsiveness Ability of product and service to react in a particular situation to provide prompt service. [56–58].
3.1. Analytic Hierarchy Process Table 2

Nine-Point scale suggested by Saaty [59].
AHP was conceptualized by Saaty [59]. AHP is perhaps the most Definition Intensity of importance
popular methodology that has been used in many MCDM problems
[60–62]. The advantages of AHP are manifold. The AHP method can be Equally important 1
successfully applied to qualitative as well as quantitative data. It can Moderately more important 3
Strongly more important 5
transform a complex, and multi-criteria problem into a hierarchical
Very strongly more important 7
structure [63]. The methodology requires minimal mathematical cal- Extremely more important 9
culations [64]. AHP is the only methodology which can take into Intermediate values 2, 4, 6, 8
consideration the consistency of the decision makers [65]. The hier-
archy levels are structured in such a way that, at the minimal level,
there is a set of alternatives, while, at the top level, a general objective (CI) is calculated. Equation (3) depicts the formula for the consistency
is placed. In between the minimal level and top level (i.e., at middle index.
level) criteria and sub-criteria are placed. The AHP computes the max n
weights of criteria and sub-criteria. The factors are pairwise compared CI =
n 1 (3)
using a predefined scale, suggested by Saaty. Further, the same scale is
used to assign weights for ranking the alternatives [57,58,61,66–68]. In Finally, the consistency ratio (CR) is calculated. The CR checks
this study, stand alone AHP is combined with fuzzy TOPSIS to rank the whether the evaluations are appropriately consistent or not. The CR can
mHealth applications. There are many studies where AHP has been be determined by taking the ratio of the CI and the random index (RI),
amalgamated with fuzzy logic to handle vagueness. However, in this suggested by Saaty [59].
study, we have not adopted the fuzzy AHP due to the limitations sug- CI
gested by Saaty and Tran [69]; inventors of the original AHP technique. CR =
RI (4)
As per Saaty and Tran [69]; incorporation of fuzzy logic into the tra-
ditional AHP will not make results more valid for real life. They also The acceptance limit of the CR is 0.1 or 10%. If the CR value exceeds
argued that incorporation of fuzzy technique into AHP will make the 0.1, then it is suggested to carry out the pairwise comparison again to
decision less accurate. The authors concluded that AHP is itself a fuzzy make the decision more consistent. According to a study, the value of
technique and there is no requirement to make it more fuzzy. consistency can be used as a parameter to measure the decision-maker's
The description of AHP is given below. In AHP, a set of criteria can ability to handle MCDM situations. It also measures the consistency of
be assumed as C = {Cj/j = 1, 2, 3 … n}. After pairwise comparison the overall hierarchy [42]. Table 2 shows the nine-point scale.
among “n” criteria, a (n × n) dimension matrix A is formed in which
each component, aij (i, j = 1, 2 … n), represents the weight of the cri- 3.2. Fuzzy TOPSIS method
terion given by the decision maker.
The number of pairwise comparisons can be determined based on The TOPSIS method was suggested by Hwang and Yoon in 1981
2
the formula n n , if “n” criteria are considered in the model. Equation [70]. It is a popular method for ranking the alternatives in MCDM
2
(1) represents the matrix of pairwise comparison. problems. This method has been used in various fields to solve decision-
making problems. The advantages of TOPSIS are as follows: TOPSIS has
a11 a1n ability to rank the best alternatives quickly [71]. The method works
1
A= , aii = 1, aji = , aij 0. upon the principles of compromise and has ability to handle conflicting
an1 ann aji (1)
situations [72]. In this method, the judgement data can be placed di-
After the pairwise comparison, mathematical computation is carried rectly without doing cumbersome calculations. The method has re-
out to establish the relative weights of criteria. Computation includes ceived significant attention from researchers due to its ease of use and
the calculation of normalized principle Eigenvector from the given low mathematical complexity. In addition, it has been also amalga-
matrix A . The relative weights are given by the Eigenvector (w) cor- mated with other methods to solve decision-making problems. Al-
responding to the largest eigenvalue ( max ), as though this method yields good results, it has some drawbacks. In a
decision-making scenario, judgments are characterized by crisp values.
Aw = max *w (2)
However, there are situations when decision-makers face difficulty in
If the rank of matrix A is 1 and max is n, then it can be conjectured assigning crisp numbers [58,73]. In this regard, the most challenging
that pairwise comparisons are completely consistent. The relative phase is how to assign a crisp value to a particular factor. Hence, in-
weights are obtained by normalizing any of the rows or columns of A stead of giving judgment in the form of a single numeric value, deci-
[57]. sion-makers can opt for an interval judgment, which is expected to be
Consistency is an important factor in AHP. At each stage, the con- more logical. For example, a decision maker's ambiguities, uncertainties
sistency has to be checked for each pairwise comparison matrix. For and vagueness cannot be resolved completely by using only crisp va-
checking the consistency of the comparison matrix, a consistency index lues. These problematic situations can be solved by applying fuzzy
3
Fuzzy Set U˜ = {(x,µũ (x ), x X)}
Definition 2. If ã is a triangular fuzzy number, then it is represented in

fuzzy theory by a triplet (u1, u2, u3) shown in Fig. 1. The membership
function µ ũ (x ) is defined as follows.
0, x u1
x u1
u2 u1
, u1 x u2
µu (x ) = x u3
u2 u1
, u2 x u3
0, x > u3 (5)
u˜ ( + ) v˜ = (u1, u2, u3) (+ ) (v1, v2, v3) = (u1 + v1, u2 + v2, u3 + v3),
(6)
Fig. 1. Triangular fuzzy number u. ̃ u˜ ( ) v˜ = (u1, u2, u3) ( ) (v1, v2, v3) = (u1 v3, u2 v2, u3 v1),
(7)
Table 3 u˜ ( × ) v˜ = (u1, u2, u3) ( × ) (v1, v2, v3) = (u1·v1, u2·v2, u3·v3), (8)
Membership functions of linguistic values.
u˜ (/) v˜ = (u1, u2, u3) (/) (v1, v2, v3) = (u1/v3, u2 /v2, u3/v1), (9)
Linguistic Value Fuzzy Numbers
kũ = (ku1, ku2, ku3). (10)
Very low (VL) (0, 0, 0.2)
Low (L) (0, 0.2, 0.4) In order to describe various operational laws of fuzzy theory, we
Medium (M) (0.2, 0.4, 0.6)
consider two triangular fuzzy numbers ũ and ṽ denoted by triplets (u1,
High (H) (0.4, 0.6, 0.8)
Very High (VH) (0.6, 0.8, 1) u2, u3) and (v1, v2, v3) respectively. The operative rules of these two
Excellent (E) (0.8, 1, 1) triangular fuzzy numbers are discussed below:
Definition 3. A linguistic variable is a variable whose values are in
linguistic terms. For example, ‘‘temperature” may be considered a
linguistic variable. It can be sensed as very cold, cold, normal, hot, and
Table 4
mHealth application evaluation performance criteria, sub-criteria and refer-
ences.
Criteria Sub-criteria References
User satisfaction (C1) Utility (C11) [99]

Efficiency (C12) [98,100–102]
Flexibility (C13) [103–105]
Customization (C14) [106,107]
Reliability (C15) [36,37]
Compatibility (C2) Quickness (C21) [21]
Easiness (C22) [65,108]
Functionality (C3) Accuracy (C31) [102,109–111]
Validity (C32) [34,36,112–114]
Robustly (C33) [115]
Fig. 2. Linguistic values and fuzzy numbers. Speed (C34) [21]
Availability (C35) [109,116–118]
Security (C4) User authentication (C41) [10,119–124]
theory. As of now, various researchers have incorporated fuzzy set Encrypted data Movement [10,117,119,120]
theory in TOPSIS to handle vagueness in decision-making problems (C42)
[74,75]. In fuzzy set theory, an uncertain situation can be modelled by Data Integrity (C43) [21,125–127]
Security breach deduction [48,128]
using different types of membership functions for example, triangular,
(C44)
trapezoidal etc. [76–82]. Physical Security (C45) [21]
It has been found that modelling using triangular fuzzy numbers for Audit trials (C46) [21]
solving decision-making problems with imprecise data is simple and Accessibility (C5) User Interface (C51) [5,21,112,129,130]
easy to use and, at the same time, it yields a better result [49,50,83]. Disability access and [21]
translation (C52)
Hence triangular fuzzy numbers are used in this study. As of now,
Content accessibility (C53) [21]
various vague situations have been modelled by using triangular Easy to learn and use Time to learn (C61) [21]
membership function by different authors [84–88]. Before applying (C6) Easiness to learn all functions [21]
fuzzy set theory in TOPSIS, it is imperative to discuss definitions related (C62)
Empathy (C7) Link (C71) [21,131]
to it. A few essential definitions of fuzzy set theory are discussed below
Customization (C72) [21,131]
[89–94]. Customer care (C73) [131,132]
Information Quality (C8) Information Richness (C81) [131–133]
Definition 1. A fuzzy set ũ in a universe of discourse X with elements x
Information Accuracy (C82) [21,132,133]
is a set of ordered pairs assigned a fuzzy index or a membership Information Up-date (C83) [21,131]
function µ ũ (x ) which associates with each element x in X usually lies in Responsiveness (C9) Customer Service (C91) [28,131,134]
[0,1]. The function value μÃ(X) is termed the grade of membership of x Technical performance (C92) [28,49,50,135,136]
in Ã. Interactivity (C93) [28,131,132]
4
Fig. 3. Hierarchy of mHealth application model.
very hot. These linguistic values can also be characterized by fuzzy

Table 5 numbers.
Pairwise comparison matrix for criteria.
Definition 4. If ũ and ṽ are two triangular fuzzy numbers defined as ũ
C1 C2 C3 C4 C5 C6 C7 C8 C9
= (u1, u2, u3) and ṽ = (v1, v2, v3), then the distance d (ũṽ ) between u
C1 1.00 6.00 0.50 3.00 6.00 0.50 5.00 0.50 8.00 and ṽ can be calculated as:
C2 1.00 0.143 0.25 3.00 0.125 0.25 0.125 0.50
C3 1.00 7.00 5.00 0.333 3.00 0.50 4.00 1
d (u˜, v˜) = [(u1 v1)2 + (u2 v2)2 + (u3 v3) 2]
C4 1.00 4.00 0.167 0.50 0.143 3.00 3 (11)
C5 1.00 0.200 0.167 0.20 0.50
C6 1.00 7.00 0.25 7.00 Definition 5. The weighted normalized fuzzy-decision matrix with
C7 1.00 0.25 4.00 different importance values of each criterion is constructed as -
C8 1.00 6.00
C9 1.00
K = [Kij] n × j i = 1,2, ..., n, j = 1,2, ..., J , (12)
5
Table 6
Summary of the evaluation criteria weight.
Criteria Local importance Sub-Criteria Local importance Global importance
User satisfaction 0.143 Utility 0.360 0.051

Efficiency 0.072 0.010
Flexibility 0.337 0.048
Customization 0.038 0.005
Reliability 0.193 0.027
Compatibility 0.024 Quickness 0.333 0.008
Easiness 0.667 0.016
Functionality 0.156 Accuracy 0.457 0.071
Validity 0.265 0.041
Robustly 0.058 0.009
Speed 0.151 0.023
Availability 0.068 0.011
Security 0.049 User authentication 0.176 0.008
Encrypted data movement 0.184 0.009
Data Integrity 0.472 0.023
Security breach deduction 0.033 0.016
Physical Security 0.087 0.004
Audit trials 0.048 0.002
Accessibility 0.022 User Interface 0.143 0.031
Disability access and translation 0.571 0.012
Content accessibility 0.286 0.006
Easy to learn and use 0.227 Time to learn 0.200 0.045
Easiness to learn all functions 0.800 0.181
Empathy 0.068 Link 0.309 0.021
Customization 0.582 0.039
Customer care 0.109 0.007
Information Quality 0.282 Information Richness 0.705 0.198
Information Accuracy 0.211 0.059
Information Up-date 0.084 0.023
Responsiveness 0.028 Customer Service 0.105 0.002
Technical performance 0.258 0.007
Interactivity 0.637 0.017
Where,
n
Dj = d (k˜ij, k˜i ) j = 1, 2, …….., J
A set of performance rating of Aj (j = 1, 2, … …. , J) with respect to j=1 (15)
criteria Ci (i = 1, 2, … , n) called
n
X̃ = {x ij , i= 1, 2, ….., n, j = 1, 2, ……,J }.
Dj = d (k˜ij, k˜i ) j = 1, 2, …….., J
A set of importance weights of each criteria wi (i = 1, 2, …, n). j=1 (16)
Based on the brief description of fuzzy theory above, fuzzy TOPSIS
steps can be explained as follows: Step 5: The similarity to an ideal solution is calculated using the
Step 1: Fuzzy TOPSIS starts with selection of the linguistic variables equation given in 17.
{x ij , i= 1, 2, ….., n, j = 1, 2, ……,J }. for alternatives with respect to Dj
criteria. In order to remove the normalization process, ranges of nor- CCj = = 1, 2, …….J .
Dj* + Dj (17)
malized triangular fuzzy numbers are kept between [0, 1] as in the
fuzzy linguistic rating (k˜ij ) . Step 6: At the end, the alternatives are ranked based on the CC j
Step 2: Now the weighted normalized fuzzy decision matrix is value. An alternative with maximum CC j can be ranked as best alter-
constructed. The weighted normalized value k̃ij calculated by Eq. (12).
native. The Example for calculating D1 , D1 and CC1 are given in the
Step 3: From the weighted normalized fuzzy decision matrix, posi-
appendix A10.
tive-ideal (A*) and negative ideal ( A ) solutions are calculated. The
fuzzy positive-ideal solution (FPIS, A*) and the fuzzy negative-ideal
solution (FNIS, A ) can be calculated by equations given below: 4. mHealth evaluation using the proposed model
The proposed model is illustrated with a numerical case example.

A = {k1 , k2 , ...ki } = maxkij i I , minkij ( i I ) ,i The detailed illustration is discussed below. An effective mHealth ap-
j j
plication can assist patients to improve their health. However, the se-
= 1,2, ...n, j = 1,2, ...J (13) lection of mHealth applications is a relatively difficult task. The pro-
blem can be solved by applying AHP and fuzzy TOPSIS techniques. The
combination of AHP and fuzzy TOPSIS has been found prudent in dif-
A = {k1 , k2 , ...ki } = minkij i I , minkij ( i I ) ,i ferent fields to solve MCDM problems. A few examples are discussed
below. Önüt and Soner [78] used the AHP and TOPSIS technique to
j j
= 1,2, ...n, j = 1,2, ...J (14) analyse the potential of solid waste transhipment sites, and to choose
the best alternative. Dağdeviren et al. [57] have developed a weapon
Where I' and I″ are related with benefit criteria and cost criteria, re- selection model by using AHP and fuzzy TOPSIS for the Turkish army.
spectively. Amiri [95] suggested the AHP and TOPSIS combination to select the
Step 4: Distances of alternatives from A* and A are calculated using best projects for Iranian Oil Company. Yu et al. [96] have proposed an
the following equations: evaluation model based on AHP and TOPSIS to rank the B2C e-
6
Fig. 4. Proposed model for mHealth application selection.
7
Table 7 commerce websites in China. The authors have used nine criteria for
Rating of alternatives based on sub-criteria and linguistic values in TOPSIS. developing the proposed model. Hayati et al. [97] used AHP and
Sub-criteria Ratings TOPSIS to rank the counties of Sistan and Baluchestan province in terms
C1 M M VH VH E E M L L VH of entitlement to health structural indicators in 2011, using 15 health
C2 L E L E L H E H VH M structural indicators for the research.
C3 L M E VL VH VH L L VL E In this paper, a combined approach of AHP and fuzzy TOPSIS
C4 M L E VL VH L H H VH E
C5 VH VH VL VL VH L L H VH H
method has been used to handle the problems. AHP has been used to
C6 M E VH M VL M L VL L H calculate the weights of different factors and fuzzy TOPSIS has been
C7 VL VL L VL E L L VH M VH used to rank the alternatives. Table 3 shows the linguistic and corre-
C8 L M E M L L VH H M E sponding fuzzy representation. Fig. 2 shows the membership function of
C9 H L H L VH VH H H H H
linguistic values.
C10 M M VL VH H L H H E L
C11 VL H H M L VL H L M M
C12 H M E L H H L VH H E
C13 VH M VL M VH M VH L M L 4.1. Identification of necessary criteria and sub-criteria for selection of
C14 VH VH VH VL L VH M L VH VH mHealth applications
C15 VH L H VL M VL VL H E E
C16 E L VL L H H VH VL E L
C17 E VL L L VH M VL L M VL
The selections of criteria and sub-criteria for mHealth applications
C18 VH H M L VL VH VH H VH H are very crucial. In the present study, a dual approach was adopted to
C19 E E VH VL E E H VH E E select the criteria and sub-criteria. In the first stage, various criteria
C20 E VL VH H M H M VH VL VL were identified by referring to the literature in the relevant field.
C21 M VL VL E H VH VL H VH VL
Subsequently, opinions from three experts from academia and industry
C22 VL VL H H VH E H E VL VL
C23 E H VL VL VL VL E L VH E were taken on the identified factors [21,98]. Based on the above, 9
C24 VL E H M E VL VL H E L criteria and 32 sub-criteria were selected. These criteria and sub-criteria
C25 E VL M E VH H H E VH VL are listed in Table 4.
C26 M L VH E E VL M E VL H After determining the criteria and sub-criteria, different mHealth
C27 L H VL M L M M VH H L
applications were explored [137]. The details of the mHealth applica-
C28 L VH VL VH E L L L E VH
C29 L VH M E M L H L H E tions are given in the appendix A11. In this study, the identified
C30 H VL H VH E E VL M E H mHealth applications were adopted as alternatives. Further, all alter-
C31 VL L M VL VH VL VH E E H natives have been evaluated based on the criteria and sub-criteria
C32 M M L E VH VL H H VH H
identified in Table 4. A conceptual model was then constructed, con-
sisting of objective functions, criteria and sub-criteria, as shown in
Fig. 3.
The model consists of four levels. In the first level, the goal is placed.
Table 8
Fuzzy evaluation matrix for the alternative mHealth Applications.
A/C A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 Weight
C11 0.2, 0.4, 0.6 0.2, 0.4, 0.6 0.6, 0.8, 1 0.6, 0.8, 1 0.8, 1, 1 0.8, 1, 1 0.2, 0.4, 0.6 0, 0.2, 0.4 0, 0.2, 0.4 0.6, 0.8, 1 0.051
C12 0, 0.2, 0.4 0.8, 1, 1 0, 0.2, 0.4 0.8, 1, 1 0, 0.2, 0.4 0.4.0.6, 0.8 0.8, 1, 1 0.4.0.6, 0.8 0.6, 0.8, 1 0.2, 0.4, 0.6 0.010
C13 0, 0.2, 0.4 0.2, 0.4.0.6 0.8, 1, 1 0, 0, 0.2 0.6, 0.8, 1 0.6, 0.8, 1 0, 0.2, 0.4 0, 0.2, 0.4 0, 0, 0.2 0.8, 1, 1 0.048
C14 0.2, 0.4, 0.6 0, 0.2, 0.4 0.8, 1, 1 0, 0, 0.2 0.6, 0.8, 1 0, 0.2, 0.4 0.4.0.6, 0.8 0.4.0.6, 0.8 0.6, 0.8, 1 0.8, 1, 1 0.005
C15 0.6, 0.8, 1 0.6, 0.8, 1 0, 0, 0.2 0, 0, 0.2 0.6, 0.8, 1 0, 0.2, 0.4 0, 0.2, 0.4 0.4.0.6, 0.8 0.6, 0.8, 1 0.4, 0.6, 0.8 0.027
C21 0.2, 0.4, 0.6 0.8, 1, 1 0.6, 0.8, 1 0.2, 0.4, 0.6 0, 0, 0.2 0.2, 0.4, 0.6 0, 0.2, 0.4 0, 0, 0.2 0, 0.2, 0.4 0.4, 0.6, 0.8 0.008
C22 0, 0, 0.2 0, 0, 0.2 0, 0.2, 0.4 0, 0, 0.2 0.8, 1, 1 0, 0.2, 0.4 0, 0.2, 0.4 0.6, 0.8, 1 0.2, 0.4, 0.6 0.6, 0.8, 1 0.016
C31 0, 0.2, 0.4 0.2, 0.4, 0.6 0.8, 1, 1 0.2, 0.4, 0.6 0, 0.2, 0.4 0, 0.2, 0.4 0.6, 0.8, 1 0.4.0.6, 0.8 0.2, 0.4, 0.6 0.8, 1, 1 0.071
C32 0.4.0.6, 0.8 0, 0.2, 0.4 0.4.0.6, 0.8 0, 0.2, 0.4 0.6, 0.8, 1 0.6, 0.8, 1 0.4.0.6, 0.8 0.4.0.6, 0.8 0.4, 0.6, 0.8 0.4, 0.6, 0.8 0.041
C33 0.2, 0.4, 0.6 0.2, 0.4, 0.6 0, 0, 0.2 0.6, 0.8, 1 0.4.0.6, 0.8 0, 0.2, 0.4 0.4.0.6, 0.8 0.4, 0.6, 0.8 0.8, 1, 1 0, 0.2, 0.4 0.009
C34 0, 0, 0.2 0.4.0.6, 0.8 0.4.0.6, 0.8 0.2, 0.4, 0.6 0, 0.2, 0.4 0, 0, 0.2 0.4.0.6, 0.8 0, 0.2, 0.4 0.2, 0.4, 0.6 0.2, 0.4, 0.6 0.023
C35 0.4.0.6, 0.8 0.2, 0.4, 0.6 0.8, 1, 1 0, 0.2, 0.4 0.4.0.6, 0.8 0.4.0.6, 0.8 0, 0.2, 0.4 0.6, 0.8, 1 0.4, 0.6, 0.8 0.8, 1, 1 0.011
C41 0.6, 0.8, 1 0.2, 0.4, 0.6 0, 0, 0.2 0.2, 0.4, 0.6 0.6, 0.8, 1 0.2, 0.4, 0.6 0.6, 0.8, 1 0, 0.2, 0.4 0.2, 0.4, 0.6 0, 0.2, 0.4 0.008
C42 0.6, 0.8, 1 0.6, 0.8, 1 0.6, 0.8, 1 0, 0, 0.2 0, 0.2, 0.4 0.6, 0.8, 1 0.2, 0.4, 0.6 0, 0.2, 0.4 0.6, 0.8, 1 0.6, 0.8, 1 0.009
C43 0.6, 0.8, 1 0, 0.2, 0.4 0.4.0.6, 0.8 0, 0, 0.2 0.2, 0.4, 0.6 0, 0, 0.2 0, 0, 0.2 0.4, 0.6, 0.8 0.8, 1, 1 0.8, 1, 1 0.023
C44 0.8, 1, 1 0, 0.2, 0.4 0, 0, 0.2 0, 0.2, 0.4 0.4.0.6, 0.8 0.4.0.6, 0.8 0.6, 0.8, 1 0, 0, 0.2 0.8, 1, 1 0, 0.2, 0.4 0.016
C45 0.8, 1, 1 0, 0, 0.2 0, 0.2, 0.4 0, 0.2, 0.4 0.6, 0.8, 1 0.2, 0.4, 0.6 0, 0, 0.2 0, 0.2, 0.4 0.2, 0.4, 0.6 0, 0, 0.2 0.004
C46 0.6, 0.8, 1 0.4.0.6, 0.8 0.2, 0.4, 0.6 0, 0.2, 0.4 0, 0, 0.2 0.6, 0.8, 1 0.6, 0.8, 1 0.4, 0.6, 0.8 0.6, 0.8, 1 0.4, 0.6, 0.8 0.002
C51 0.8, 1, 1 0.8, 1, 1 0.6, 0.8, 1 0, 0, 0.2 0.8, 1, 1 0.8, 1, 1 0.4.0.6, 0.8 0.6, 0.8, 1 0.8, 1, 1 0.8, 1, 1 0.031
C52 0.8, 1, 1 0, 0, 0.2 0.6, 0.8, 1 0.4.0.6, 0.8 0.2, 0.4, 0.6 0.4.0.6, 0.8 0.2, 0.4, 0.6 0.6, 0.8, 1 0, 0, 0.2 0, 0, 0.2 0.012
C53 0.2, 0.4, 0.6 0, 0, 0.2 0, 0, 0.2 0.8, 1, 1 0.4.0.6, 0.8 0.6, 0.8, 1 0, 0, 0.2 0.4, 0.6, 0.8 0.6, 0.8, 1 0, 0, 0.2 0.006
C61 0, 0, 0.2 0, 0, 0.2 0.4.0.6, 0.8 0.4.0.6, 0.8 0.6, 0.8, 1 0.8, 1, 1 0.4.0.6, 0.8 0.8, 1, 1 0, 0, 0.2 0, 0, 0.2 0.045
C62 0.8, 1, 1 0.4.0.6, 0.8 0, 0, 0.2 0, 0, 0.2 0, 0, 0.2 0, 0, 0.2 0.8, 1, 1 0, 0.2, 0.4 0.6, 0.8, 1 0.8, 1, 1 0.181
C71 0, 0, 0.2 0.8, 1, 1 0.4.0.6, 0.8 0.2, 0.4, 0.6 0.8, 1, 1 0, 0, 0.2 0, 0, 0.2 0.4, 0.6, 0.8 0.8, 1, 1 0, 0.2, 0.4 0.021
C72 0.8, 1, 1 0, 0, 0.2 0.2, 0.4, 0.6 0.8, 1, 1 0.6, 0.8, 1 0.4.0.6, 0.8 0.4.0.6, 0.8 0.8, 1, 1 0.6, 0.8, 1 0, 0, 0.2 0.039
C73 0.2, 0.4, 0.6 0, 0.2, 0.4 0.6, 0.8, 1 0.8, 1, 1 0.8, 1, 1 0, 0, 0.2 0.2, 0.4, 0.6 0.8, 1, 1 0, 0, 0.2 0.4, 0.6, 0.8 0.007
C81 0, 0.2, 0.4 0.4.0.6, 0.8 0, 0, 0.2 0.2, 0.4, 0.6 0, 0.2, 0.4 0.2, 0.4, 0.6 0.2, 0.4, 0.6 0.6, 0.8, 1 0.4, 0.6, 0.8 0, 0.2, 0.4 0.198
C82 0, 0.2, 0.4 0.6, 0.8, 1 0, 0, 0.2 0.6, 0.8, 1 0.8, 1, 1 0, 0.2, 0.4 0, 0.2, 0.4 0, 0.2, 0.4 0.8, 1, 1 0.6, 0.8, 1 0.059
C83 0, 0.2, 0.4 0.6, 0.8, 1 0.2, 0.4, 0.6 0.8, 1, 1 0.2, 0.4, 0.6 0, 0.2, 0.4 0.4, 0.6, 0.8 0, 0.2, 0.4 0.4, 0.6, 0.8 0.8, 1, 1 0.023
C91 0.4.0.6, 0.8 0, 0, 0.2 0.4.0.6, 0.8 0.6, 0.8, 1 0.8, 1, 1 0.8, 1, 1 0, 0, 0.2 0.2, 0.4, 0.6 0.8, 1, 1 0.4, 0.6, 0.8 0.002
C92 0, 0, 0.2 0, 0.2, 0.4 0.2, 0.4, 0.6 0, 0, 0.2 0.6, 0.8, 1 0, 0, 0.2 0.6, 0.8, 1 0.8, 1, 1 0.8, 1, 1 0.4, 0.6, 0.8 0.007
C93 0.2, 0.4, 0.6 0.2, 0.4, 0.6 0, 0.2, 0.4 0.8, 1, 1 0.6, 0.8, 1 0, 0, 0.2 0.4.0.6, 0.8 0.4.0.6, 0.8 0.6, 0.8, 1 0.4, 0.6, 0.8 0.017
8
Table 9
Weighted evaluation for alternative mHealth application.
A/C A1 A2 A3 A4 A5 A6
C11 0.012, 0.024, 0.036 0.012, 0.024, 0.036 0.036, 0.048, 0.06 0.036, 0.048, 0.06 0.048, 0.06, 0.06 0.048, 0.06, 0.06
C12 0, 0.0024, 0.0048 0.0096, 0.012, 0.012 0, 0.0024, 0.0048 0.0096, 0.012, 0.012 0, 0.0024, 0.0048 0.0048, 0.0072, 0.0096
C13 0, 0.0114, 0.0228 0.0114, 0.0228, 0.0342 0.0456, 0.057, 0.057 0,0, 0.0114 0.0342, 0.0456, 0.057 0.0342, 0.0456, 0.057
C14 0.0012, 0.0024, 0.0036 0, 0.0012, 0.0024 0.0048, 0.006, 0.006 0, 0, 0.0012 0.0036, 0.0048, 0.006 0, 0.0012, 0.0024
C15 0.0192, 0.0256, 0.032 0.0192, 0.0256, 0.032 0, 0, 0.0064 0, 0, 0.0064 0.0192, 0.0256, 0.032 0, 0.0064, 0.0128
C21 0.0016, 0.0032, 0.0048 0.0064, 0.008, 0.008 0.0048, 0.0064, 0.008 0.0016, 0.0032, 0.0048 0, 0, 0.0016 0.0016, 0.0032, 0.0048
C22 0, 0, 0.003 0, 0, 0.003 0, 0.003, 0.006 0, 0, 0.003 0.012, 0.015, 0.015 0, 0.003, 0.006
C31 0, 0.0168, 0.0336 0.0168, 0.0336, 0.0504 0.0672, 0.084, 0.084 0.0168, 0.0336, 0.0504 0, 0.0168, 0.0336 0, 0.0168, 0.0336
C32 0.0196, 0.0294, 0.0392 0, 0.0098, 0.0196 0.0196, 0.0294, 0.0392 0, 0.0098, 0.0196 0.0294, 0.0392, 0.049 0.0294, 0.0392, 0.049
C33 0.0022, 0.0044, 0.0066 0.0022, 0.0044, 0.0066 0, 0, 0.0022 0.0066, 0.0088, 0.011 0.0044, 0.0066, 0.0088 0, 0.0022, 0.0044,
C34 0, 0, 0.0056 0.0112, 0.0168, 0.0224 0.0112, 0.0168, 0.0224 0.0056, 0.0112, 0.0168 0, 0.0056, 0.0112 0, 0, 0.0056
C35 0.0052, 0.0078, 0.0104 0.0026, 0.0052, 0.0078 0.0104, 0.013, 0.013 0, 0.0026, 0.0052 0.0052, 0.0078, 0.0104 0.0052, 0.0078, 0.0104
C41 0.0054, 0.0072, 0.009 0.0018, 0.0036, 0.0054 0, 0, 0.0018 0.0018, 0.0036, 0.0054 0.0054, 0.0072, 0.009 0.0018, 0.0036, 0.0054
C42 0.0054, 0.0072, 0.009 0.0054, 0.0072, 0.009 0.0054, 0.0072, 0.009 0, 0, 0.0018 0, 0.0018, 0.0036 0.0054, 0.0072, 0.009
C43 0.0138, 0.0184, 0.023 0, 0.0046, 0.0092 0.0092, 0.0138, 0.0184 0, 0, 0.0046 0.0046, 0.0092, 0.0138 0, 0, 0.0046
C44 0.0016, 0.002, 0.002 0, 0.0004, 0.0008 0, 0, 0.0004 0, 0.0004, 0.0008 0.0008, 0.0012, 0.0016 0.0008, 0.0012, 0.0016
C45 0.0032, 0.004, 0.004 0, 0, 0.0008 0, 0.0008, 0.0016 0, 0.0008, 0.0016 0.0024, 0.0032, 0.004 0.0008, 0.0016, 0.0024
C46 0.0012, 0.0016, 0.002 0.0008, 0.0012, 0.0016 0.0004, 0.0008, 0.0012 0, 0.0004, 0.0008 0, 0, 0.0004 0.0012, 0.0016, 0.002
C51 0.0024, 0.003, 0.003 0.0024, 0.003, 0.003 0.0018, 0.0024, 0.003 0, 0, 0.0006 0.0024, 0.003, 0.003 0.0024, 0.003, 0.003
C52 0.0104, 0.013, 0.013 0, 0, 0.0026 0.0078, 0.0104, 0.013 0.0052, 0.0078, 0.0104 0.0026, 0.0052, 0.0078 0.0052, 0.0078, 0.0104
C53 0.0012, 0.0024, 0.0036 0, 0, 0.0012 0, 0, 0.0012 0.0048, 0.006, 0.006 0.0024, 0.0036, 0.0048 0.0036, 0.0048, 0.006
C61 0, 0, 0.0008 0, 0, 0.0008 0.0016, 0.0024, 0.0032 0.0016, 0.0024, 0.0032 0.0024, 0.0032, 0.004 0.0032, 0.004, 0.004
C62 0.0128, 0.016, 0.016 0.0064, 0.0096, 0.0128 0, 0, 0.0032 0, 0, 0.0032 0, 0, 0.0032 0, 0, 0.0032
C71 0, 0, 0.0044 0.0176, 0.022, 0.022 0.0088, 0.0132, 0.0176 0.0044, 0.0088, 0.0132 0.0176, 0.022, 0.022 0, 0, 0.0044
C72 0.0328, 0.041, 0.041 0, 0, 0.0082 0.0082, 0.0164, 0.0246 0.0328, 0.041, 0.041 0.0246, 0.0328, 0.041 0.0164, 0.0246, 0.0328
C73 0.0016, 0.0032, 0.0048 0, 0.0016, 0.0032 0.0048, 0.0064, 0.008 0.0064, 0.008, 0.008 0.0064, 0.008, 0.008 0, 0, 0.0016
C81 0, 0.0036, 0.0072 0.0072, 0.0108, 0.0144 0, 0, 0.0036 0.0036, 0.0072, 0.0108 0, 0.0036, 0.0072 0.0036, 0.0072, 0.0108
C82 0, 0.001, 0.002 0.003, 0.004, 0.005 0, 0, 0.001 0.003, 0.004, 0.005 0.004, 0.005, 0.005 0, 0.001, 0.002
C83 0, 0.0004, 0.0008 0.0012, 0.0016, 0.002 0.0004, 0.0008, 0.0012 0.0016, 0.002, 0.002 0.0004, 0.0008, 0.0012 0, 0.0004, 0.0008
C91 0.0012, 0.0018, 0.0024 0, 0, 0.0006 0.0012, 0.0018, 0.0024 0.0018, 0.0024, 0.003 0.0024, 0.003, 0.003 0.0024, 0.003, 0.003
C92 0, 0, 0.0014 0, 0.0014, 0.0028 0.0014, 0.0028, 0.0042 0, 0, 0.0014 0.0042, 0.0056, 0.007 0, 0, 0.0014
A/C A7 A8 A9 A10 A* A-
C11 0.012, 0.024, 0.036 0, 0.012, 0.024 0, 0.012, 0.024 0.036, 0.048, 0.06 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C12 0.0096, 0.012, 0.012 0.0048, 0.0072, 0.0096 0.0072, 0.0096, 0.012 0.0024, 0.0048, 0.0072 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C13 0, 0.0114, 0.0228 0, 0.0114, 0.0228 0, 0, 0.0114 0.0456, 0.057, 0.057 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C14 0.0024, 0.0036, 0.0048 0.0024, 0.0036, 0.0048 0.0036, 0.0048, 0.006 0.0048, 0.006, 0.006 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C15 0, 0.0064, 0.0128 0.0128, 0.0192, 0.0256 0.0192, 0.0256, 0.032 0.0128, 0.0192, 0.0256 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C21 0, 0.0016, 0.0032 0, 0, 0.0016 0, 0.0016, 0.0032 0.0032, 0.0048, 0.0064 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C22 0, 0.003, 0.006 0.009, 0.012, 0.015 0.003, 0.006, 0.009 0.009, 0.012, 0.015 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C31 0.0504, 0.0672, 0.084 0.0336, 0.0504, 0.0672 0.0168, 0.0336, 0.0504 0.0672, 0.084, 0.084 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C32 0.0196, 0.0294, 0.0392 0.0196, 0.0294, 0.0392 0.0196, 0.0294, 0.0392 0.0196, 0.0294, 0.0392 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C33 0.0044, 0.0066, 0.0088 0.0044, 0.0066, 0.0088 0.0088, 0.011, 0.011 0, 0.0022, 0.0044 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C34 0.0112, 0.0168, 0.0224 0, 0.0056, 0.0112, 0.0056, 0.0112, 0.0168 0.0056, 0.0112, 0.0168 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C35 0, 0.0026, 0.0052 0.0078, 0.0104, 0.013 0.0052, 0.0078,0.0104 0.0104, 0.013, 0.013 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C41 0.0054, 0.0072, 0.009 0, 0.0018, 0.0036 0.0018, 0.0036, 0.0054 0, 0.0018, 0.0036 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C42 0.0018, 0.0036, 0.0054 0, 0.0018, 0.0036 0.0054, 0.0072, 0.009 0.0054, 0.0072, 0.009 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C43 0, 0, 0.0046 0.0092, 0.0138, 0.0184 0.0184, 0.023, 0.023 0.0184, 0.023, 0.023 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C44 0.0012, 0.0016, 0.002 0, 0, 0.0004 0.0016, 0.002, 0.002 0, 0.0004, 0.0008 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C45 0, 0, 0.0008 0, 0.0008, 0.0016 0.0008, 0.0016, 0.0024 0, 0, 0.0008 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C46 0.0012, 0.0016, 0.002 0.0008, 0.0012, 0.0016 0.0012, 0.0016, 0.002 0.0008, 0.0012, 0.0016 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C51 0.0012, 0.0018, 0.0024 0.0018, 0.0024, 0.003 0.0024, 0.003, 0.003 0.0024, 0.003, 0.003 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C52 0.0026, 0.0052, 0.0078 0.0078, 0.0104, 0.013 0, 0, 0.0026 0, 0, 0.0026 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C53 0, 0, 0.0012 0.0024, 0.0036, 0.0048 0.0036, 0.0048, 0.006 0, 0, 0.0012 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C61 0.0016, 0.0024, 0.0032 0.0032, 0.004, 0.004 0, 0, 0.0008 0, 0, 0.0008 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C62 0.0128, 0.016, 0.016 0, 0.0032, 0.0064 0.0096, 0.0128, 0.016 0.0128, 0.016, 0.016 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C71 0, 0, 0.0044 0.0088, 0.0132, 0.0176 0.0176, 0.022, 0.022 0, 0.0044, 0.0088 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C72 0.0164, 0.0246, 0.0328 0.0328, 0.041, 0.041 0.0246, 0.0328, 0.041 0, 0, 0.0082 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C73 0.0016, 0.0032, 0.0048 0.0064, 0.008, 0.008 0, 0, 0.0016 0.0032, 0.0048, 0.0064 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C81 0.0036, 0.0072, 0.0108 0.0108, 0.0144, 0.018 0.0072, 0.0108, 0.0144 0, 0.0036, 0.0072 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C82 0, 0.001, 0.002 0, 0.001, 0.002 0.004, 0.005, 0.005 0.003, 0.004, 0.005 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C83 0.0008, 0.0012, 0.0016 0, 0.0004, 0.0008 0.0008, 0.0012, 0.0016 0.0016, 0.002, 0.002 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C91 0, 0, 0.0006 0.0006, 0.0012, 0.0018 0.0024, 0.003, 0.003 0.0012, 0.0018, 0.0024 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
C92 0.0042, 0.0056, 0.007 0.0056, 0.007, 0.007 0.0056, 0.007, 0.007 0.0028, 0.0042, 0.0056 ṽ*1= (1,1,1) ṽ-1= (0,0,0)
Sets of criteria and different sub-criteria are placed in the second and 4.2. Computation the weights of criteria
third level, respectively. Finally, on the last level, different alternatives
are placed. The primary objective of this model is to choose best In the first phase, three decision makers from academia and
mHealth applications. In this model, AHP has been applied to calculate healthcare industry with more than fifteen years of experience were
the weights of criteria and sub-criteria. Furthermore, TOPSIS has been asked to carry out pairwise comparisons among the identified factors
applied to rank the alternatives. (user satisfaction, compatibility, functionality, security, accessibility,
easy to learn and use, empathy, information quality, responsiveness)
9
using the scale suggested by Saaty [59]. The results obtained from the Table 10
pairwise comparison matrix are shown in Table 5. After carrying out a Weighted Fuzzy TOPSIS results.
pairwise comparison among the criteria with respect to the goal, the Alternatives Dj Dj CCj
three decision makers were again asked to carry out pairwise compar-
isons among the sub-criteria with respect to criteria. Table A1 in the A1 (Cody) 31.74047905 0.28061731 0.008763513
Appendix shows the evaluation of sub-criteria with respect to user sa- A2 (Hot5 Fitness) 31.75635457 0.264106109 0.008248042
tisfaction. The consistency ratio (CR) was found to be 0.04. Table A2 in A3 (Pact) 31.65567102 0.359066569 0.011215665
A4 (Carrot fit) 31.75377846 0.266238647 0.008314757
the Appendix shows the evaluation of sub-criteria with respect to A5 (Human) 31.64487194 0.371652412 0.011608144
compatibility, whereas Table A3 in Appendix shows the evaluation of A6 (Moves) 31.73325929 0.287305377 0.008972527
sub-criteria with respect to functionality. Table A4 in Appendix shows A7 (LoseIt) 31.72130402 0.298220293 0.009313702
the evaluation of sub-criteria with respect to security. Table A5 in the A8 (Noom Weight Loss Coach) 31.59382059 0.462005687 0.014412534
A9 (Healthy Out) 31.58966027 0.464350392 0.014486499
Appendix shows the evaluation of sub-criteria with respect to accessi-
A10 (Zipongo) 31.63006332 0.383529413 0.011980205
bility. Similarly, Tables A6–A9 in Appendix show the pairwise com-
parisons with respect to criteria such as easy to learn and use, empathy,
information quality, responsiveness, respectively. Table 6 shows the
final results of AHP. In this study, the respondents have given higher Table 11
importance to the factors easy to learn and use (C6), information Unweighted Fuzzy TOPSIS results.
quality (C8), functionality (C3), and user satisfaction (C1) for the se- Alternatives Dj Dj CCj
lection of mHealth applications. The global weights of sub-criteria have
been calculated by multiplying local importance of criteria with sub- A1 (Cody) 17.22213505 16.72734689 0.49271288
A2 (Hot5 Fitness) 19.04121173 14.88744939 0.43878682
criteria. In the AHP, information richness and easiness to learn all
A3 (Pact) 17.86132245 16.06701722 0.47355743
functions have been emerged as important factors with weights of A4 (Carrot fit) 18.84589754 15.06065699 0.44418129
0.198 and 0.181, respectively. The values of the CR for different pair- A5 (Human) 13.88068497 20.15240170 0.59214146
wise comparisons were observed at less than or equal to 0.10. The CR A6 (Moves) 18.68248077 15.28782199 0.45003491
confirms the consistency of the decisions. The global weights of various A7 (LoseIt) 18.10890057 15.84186715 0.46661293
A8 (Noom Weight Loss Coach) 16.47346522 17.51894414 0.51537812
factors have been further used in TOPSIS to rate the mHealth applica-
A9 (Healthy Out) 13.47020008 20.47896042 0.60322435
tions. Fig. 4 shows the diagram of the proposed model for mHealth A10 (Zipongo) 15.55108680 18.33496834 0.54107710
application selection.
Table 12
4.3. Assessment of alternatives and determine the final rank Weighted and un-weighted ranking.
After computing the weights of the sub-criteria, the next step is to Rank Weighted CCj Weighted Ranking Un-weighted Un-weighted
CCj Ranking
select the best mHealth application from the alternatives. Using TOPSIS
for selecting the alternatives, the same experts were asked to assign 1 0.014486499 A9 (Healthy Out) 0.60322435 A9 (Healthy Out)
weights to different alternatives based on the importance of the sub- 2 0.014412534 A8 (Noom Weight 0.59214146 A5 (Human)
criteria. The weights were assigned in a linguistic term given in Table 7. Loss Coach)
3 0.011980205 A10 (Zipongo) 0.54107710 A10 (Zipongo)
Further, the linguistic values were converted to the fuzzy numbers
4 0.011608144 A5 (Human) 0.51537812 A8 (Noom Weight
based on the scale given in Table 3. The fuzzy evaluation matrix is Loss Coach)
shown in Table 8. The weighted evaluation matrix was constructed by 5 0.011215665 A3 (Pact) 0.49271288 A1 (Cody)
multiplying sub-criteria weights calculated by AHP with the corre- 6 0.009313702 A7 (LoseIt) 0.47355743 A3 (Pact)
sponding fuzzy numbers. The fuzzy weighted decision matrix is shown 7 0.008972527 A6 (Moves) 0.46661293 A7 (LoseIt)
8 0.008763513 A1 (Cody) 0.45003491 A6 (Moves)
in Table 9. After determining the weighted decision matrix, it was ne-
9 0.008314757 A4 (Carrot fit) 0.44418129 A4 (Carrot fit)
cessary to judge the nature of sub-criteria for estimating the final rank 10 0.008248042 A2 (Hot5 Fitness) 0.43878682 A2 (Hot5 Fitness)
of mHealth applications. The ideal positive and negative solutions for
all the sub-criteria were decided based on the nature of the factors. In
this study, the positive and negative distances are represented as D* and 5. Sensitivity analysis
D‾, respectively. In our research, sub-criteria ranging from C11 to C93
are considered as benefit criteria. For benefit criteria, the values of The sensitivity analysis was carried out by varying the weight of one
fuzzy positive-ideal solution (FPIS, A*) and the negative-ideal solution criterion at a time, keeping other factors unchanged. For example, the
(FNIS, A ) can be considered as Vi = (1, 1, 1) and Vi = (0, 0, 0) , weight of criterion C1 (user satisfaction) was exchanged with the
respectively. Now using Equation (15) and Equation (16), the distances weight of criterion C2 (compatibility), and subsequently, the weight of
from the positive and negative ideal solutions were calculated for al- C1 (user satisfaction) was exchanged with the weight of C3 (function-
ternatives. Table 10 and Table 11 show the distances for the alter- ality). A total of 36 experiments were conducted and results are tabu-
natives. Finally, CCj values were computed using Equation (17) for all lated in Table 13. Fig. 5 represents variations in the final ranking of
alternatives using the D* and D‾ values [138]. different mHealth applications with the change of criteria. In sensitivity
The weighted and unweighted CCj values for all the mHealth ap- analysis, A9 (Healthy Out) has emerged as the best mHealth application
plications are shown in Tables 10 and 11, respectively. Furthermore, followed by A8 (Noom Weight Loss Coach) and A10 (Zipongo). The
the ranking of alternatives was carried out based on CCj values. sensitivity analyses show interesting scenarios with the variations of
Table 12 shows the comparison between weighted and unweighted weights of the criteria. In this study, A2 (Hot5 Fitness) has been ranked
ranking. Interestingly, a significant variation is observed between the as the worst alternative among the applications.
weighted and unweighted ranking processes. MCDM is often influenced After carrying out the sensitivity analyses, it is vital to carry out a
by the weights of the factors and sub-factors assigned by the decision comparative analysis of our proposed model with the existing similar
maker. Hence, it will be interesting to observe the behaviours of the literature. Table 14 shows the Comparative analysis of present study in
model by changing the weights of the factors. terms of model parameters. Although our model adopted the tested
AHP and fuzzy TOPSIS combination to select mHealth applications, it is
10
M. Rajak and K. Shaw
Table 13
Sensitivity analysis result.
Ex no Criteria Definition (condition) A1 A2 A3 A4 A5 A6 A7 A8 A9 A10 Highest rank Name of the app
EX 1 C1–C2 (User Satisfaction- Compatibility) 0.0077 0.0073 0.0098 0.0072 0.0104 0.0069 0.0086 0.0106 0.0099 0.0112 0.0112 Zipongo (A10)
EX 2 C1–C3 (User Satisfaction- Functionality) 0.0081 0.0075 0.0100 0.0076 0.0107 0.0082 0.0083 0.0092 0.0093 0.0107 0.0107 Human (A5) & Zipongo (A10)
EX 3 C1–C4 (User Satisfaction-Security) 0.0096 0.0074 0.0102 0.0073 0.0100 0.0074 0.0089 0.0100 0.0110 0.0111 0.0111 Zipongo (A10)
EX 4 C1–C5 (User Satisfaction- Accessibility) 0.0098 0.0067 0.0104 0.0088 0.0100 0.0084 0.0090 0.0113 0.0101 0.0093 0.0113 Noom Weight Loss Coach (A8)
EX 5 C1–C6 (User Satisfaction- Easy to learn and use) 0.0126 0.0113 0.0133 0.0099 0.0122 0.0095 0.0125 0.0118 0.0129 0.0148 0.0148 ZIpongo (A10)
EX 6 C1–C7 (User Satisfaction-Empathy) 0.0088 0.0075 0.0103 0.0088 0.0110 0.0077 0.0091 0.0108 0.0107 0.0100 0.0110 Human (A5)
EX 7 C1–C8 (User Satisfaction-Information quality) 0.0113 0.0127 0.0142 0.0119 0.0161 0.0129 0.0122 0.0138 0.0141 0.0164 0.0164 Zipongo (A10)
EX 8 C1–C9 (User Satisfaction- Responsiveness) 0.0082 0.0072 0.0095 0.0090 0.0111 0.0066 0.0099 0.0110 0.0116 0.0108 0.0116 Healthy Out (A9)
EX 9 C2–C3 (Compatibility- Functionality) 0.0078 0.0078 0.0092 0.0072 0.0121 0.0082 0.0069 0.0139 0.0134 0.0114 0.0139 Noom Weight Loss Coach (A8)
EX 10 C2–C4 (Compatibility- Security) 0.0083 0.0082 0.0112 0.0083 0.0118 0.0089 0.0092 0.0146 0.0142 0.0121 0.0146 Noom Weight Loss Coach (A8)
EX 11 C2–C5 (Compatibility- Accessibility) 0.0088 0.0082 0.0112 0.0083 0.0116 0.0090 0.0093 0.0144 0.0145 0.0120 0.0145 Healthy Out (A9)
EX 12 C2–C6 (Compatibility- Easy to learn and use) 0.0103 0.0108 0.0141 0.0097 0.0159 0.0110 0.0111 0.0182 0.0169 0.0169 0.0182 Noom Weight Loss Coach (A8)
EX 13 C2–C7 (Compatibility- Empathy) 0.0082 0.0082 0.0110 0.0075 0.0113 0.0088 0.0090 0.0140 0.0139 0.0127 0.0140 Noom Weight Loss Coach (A8)
EX 14 C2–C8 (Compatibility- Information quality) 0.0105 0.0114 0.0148 0.0101 0.0170 0.0115 0.0114 0.0191 0.0175 0.0181 0.0191 Noom Weight Loss Coach (A8)
EX 15 C2–C9 (Compatibility- Responsiveness) 0.0088 0.0083 0.0112 0.0083 0.0116 0.0090 0.0093 0.0278 0.0145 0.0120 0.0278 Noom Weight Loss Coach (A8)
EX 16 C3–C4 (Functionality- Security) 0.0100 0.0078 0.0097 0.0072 0.0113 0.0085 0.0076 0.0134 0.0148 0.0111 0.0148 Healthy Out (A9)
11
EX 17 C3–C5 (Functionality- Accessibility) 0.0102 0.0071 0.0099 0.0090 0.0116 0.0099 0.0074 0.0146 0.0136 0.0092 0.0146 Noom Weight Loss Coach (A8)
EX 18 C3–C6 (Functionality- Easy to learn and use) 0.0125 0.0110 0.0131 0.0097 0.0133 0.0107 0.0139 0.0168 0.0180 0.0162 0.0180 Healthy Out (A9)
EX 19 C3–C7 (Functionality-Empathy) 0.0091 0.0078 0.0099 0.0089 0.0123 0.0080 0.0075 0.0145 0.0146 0.0099 0.0146 Healthy Out (A9)
EX 20 C3–C8 (Functionality-Information quality) 0.0109 0.0123 0.0141 0.0118 0.0147 0.0118 0.0131 0.0188 0.0189 0.0160 0.0189 Healthy Out (A9)
EX 21 C3–C9 (Functionality- Responsiveness) 0.0084 0.0076 0.0089 0.0092 0.0127 0.0078 0.0085 0.0143 0.0154 0.0109 0.0154 Healthy Out (A9)
EX 22 C4–C5 (Security- Accessibility) 0.0087 0.0081 0.0113 0.0087 0.0117 0.0093 0.0093 0.0147 0.0142 0.0116 0.0147 Noom Weight Loss Coach (A8)
EX 23 C4–C6 (Security- Easy to learn and use) 0.0141 0.0109 0.0143 0.0096 0.0146 0.0113 0.0121 0.0172 0.0194 0.0165 0.0194 Healthy Out (A9)
EX 24 C4–C7 (Security- Empathy) 0.0088 0.0082 0.0112 0.0079 0.0113 0.0089 0.0092 0.0141 0.0144 0.0122 0.0144 Healthy Out (A9)
EX 25 C4–C8 (Security- Information quality) 0.0150 0.0116 0.0150 0.0101 0.0155 0.0120 0.0122 0.0181 0.0206 0.0174 0.0206 Healthy Out (A9)
EX 26 C4–C9 (Security- Responsiveness) 0.0085 0.0082 0.0111 0.0087 0.0118 0.0089 0.0095 0.0146 0.0146 0.0120 0.0146 Healthy Out (A9) &Noom Weight Loss Coach (A8)
EX 27 C5–C6 (Accessibility- Easy to learn and use) 0.0139 0.0098 0.0152 0.0124 0.0152 0.0136 0.0117 0.0193 0.0173 0.0135 0.0193 Noom Weight Loss Coach (A8)
EX 28 C5–C7 (Accessibility- Empathy) 0.0089 0.0080 0.0113 0.0080 0.0111 0.0094 0.0092 0.0142 0.0139 0.0120 0.0142 Noom Weight Loss Coach (A8)
EX 29 C5–C8 (Accessibility- Information quality) 0.0152 0.0101 0.0162 0.0135 0.0161 0.0148 0.0123 0.0205 0.0179 0.0138 0.0205 Noom Weight Loss Coach (A8)
EX 30 C5–C9 (Accessibility- Responsiveness) 0.0088 0.0082 0.0113 0.0083 0.0116 0.0091 0.0093 0.0144 0.0144 0.0310 0.0310 Zipongo (A10)
EX 31 C6–C7 (Easy to learn and use- Empathy) 0.0130 0.0109 0.0141 0.0124 0.0162 0.0114 0.0128 0.0191 0.0193 0.0142 0.0193 Healthy Out (A9)
EX 32 C6–C8 (Easy to learn and use- Information quality) 0.0167 0.0168 0.0139 0.0136 0.0163 0.0137 0.0191 0.0217 0.0243 0.0210 0.0243 Healthy Out (A9)
EX 33 C6–C9 (Easy to learn and use- Responsiveness) 0.0113 0.0106 0.0134 0.0128 0.0168 0.0103 0.0134 0.0188 0.0200 0.0161 0.0200 Healthy Out (A9)
EX 34 C7–C8 (Empathy-Information quality) 0.0133 0.0117 0.0149 0.0142 0.0179 0.0122 0.0128 0.0208 0.0205 0.0140 0.0208 Noom Weight Loss Coach (A8)
EX 35 C7–C9 (Empathy- Responsiveness) 0.0084 0.0082 0.0109 0.0082 0.0115 0.0087 0.0095 0.0142 0.0145 0.0125 0.0145 Healthy Out (A9)
EX 36 C8–C9 (Information quality- Responsiveness) 0.0117 0.0111 0.0139 0.0140 0.0183 0.0106 0.0143 0.0199 0.0214 0.0170 0.0214 Healthy Out (A9)
Main 0.0088 0.0082 0.0112 0.0083 0.0116 0.0090 0.0093 0.0144 0.0145 0.0120
Technology in Society 59 (2019) 101186
Fig. 5. Sensitivity Analysis by varying the weights of the criteria. Series 1: Cody (A1). Series 2: Hot5 Fitness (A2). Series 3: Pact (A3). Series 4: Carrot Fit (A4). Series
5: Human (A5). Series 6: Moves (A6). Series 7: LoseIt (A7). Series 8: Noom Weight Loss Coach (A8). Series 9: Healthy Out (A9). Series 10: Zipongo (A10).
worthy to note that no exact similar model is published in the referred factors for mHealth technology selection. Healthy Out has emerged as
journals considering the same set of criteria and sub-criteria. the best mHealth application among all the alternatives followed by
Noom Weight Loss Coach. Interestingly, a significant variation of the
ranking has been observed for weighted and unweighted methods. The
6. Discussion and conclusion sub-criteria play an important role in mHealth technology selection.
However, a decision maker can only use the fuzzy TOPSIS to rank the
Mobile health has become necessary recently due to the digitization alternatives.
of society. This system can manage diseases more promptly if it can be This study makes a theoretical contribution to the body of literature.
amalgamated with the traditional healthcare system. In this digital As per our knowledge, there are hardly any frameworks reported which
world, an effective mHealth application can help people manage their consider all the mentioned factors for evaluating the performances of
health and become more informed regarding health. mHealth applications. This study can be one of the stepping stones for
However, with the increasing number of health related applications, evaluating the performances of mHealth applications. The sensitivity of
it is very difficult to decide which applications are suitable for the users the study is quite unique and such sensitivity analysis has not been
and which are not. This research has tried to shed light on this direc- reported in any kind of eHealth or mHealth studies.
tion. The aim of this research is to form a model based on AHP and The study has several managerial implications also. The framework
fuzzy TOPSIS to compare and evaluate the performances of different can be adopted by mHealth application developers for ranking the
mHealth applications. For evaluating performances of different applications. The functionality of the application can be improved using
mHealth applications, nine criteria and thirty-two sub-criteria have the result from the framework. By using this framework, doctors can
been considered. Different mHealth applications are compared and suggest that patients use the application suitable for their healthcare.
prioritized based on the above thirty-two factors. To capture the un- Although AHP and TOPSIS methods have been widely used in var-
certainty pertaining to the human decision-making process, fuzzy ious sectors to solve different problems, these methods also have a few
concepts have been applied. User satisfaction, functionality, easy to limitations. The AHP and TOPSIS methods have limitations pertaining
learn and use, and information quality were observed as important
Table 14
Comparative analysis of present study in terms of model parameters.
Present study by the GülçinBüyüközkan and Shafii et al. [26] Redenovic and Veselinovic [141] Liao and Qiu [20] Liu et al. [142]
author (2019) GizemÇifçi [21].
User satisfaction - Tangibles - Functional - Number of users of information systems for - Health care cloud - Usefulness
Compatibility - Reliability - Professional support to electronic health software - Ease of use
Functionality Assurance - Human - Data redundancy on an annual basis in the - Cloud service -Quality of
Security Empathy - Organizational preparation and encoding electronic delivery technological service
Accessibility Information quality invoices - Cloud management - Compatibility
Easy to learn and use Responsiveness - Monthly increase of interoperability rates at issues - Trust
Empathy the beginning of the implementation period - Perceived value
Information - Rate of return on investment in - Social factors
Quality communication infrastructure on a monthly - Product image
Responsiveness basis in the year of implementation of
electronic health
- Monthly increase of the utilization of data
(big data) based on cloud approach and web
orientation in percentage
- Assessment of compliance with HIPPA
principles on a scale of 0–10
12
to the number of alternatives to be ranked. The AHP method can be In addition, the model can be tested in real world settings in different
successfully applied for maximum of nine alternatives [139]; whereas parts of the world.
TOPSIS can rank a maximum of ten alternatives [140]. The model can
suffer with the bias of the respondent. The current study tries to es- Acknowledgement
tablish a framework for evaluating mHealth applications. In this study,
a significant stress has been given on developing a conceptual frame- The authors are thankful to the editor-in-chief Charla Griffy-Brown,
work. Future research can be carried out applying different multi-cri- and the reviewers whose insightful comments and suggestions have
teria decision making techniques, like fuzzy-ANP, Promethee, Vikor etc. significantly helped to improve the quality of the paper.
Appendix A
Table A1
Evaluation of sub-criteria with respect to User satisfaction (CR = 0.04).
C11 C12 C13 C14 C15
Utility (C11) 1.00 5.00 1.00 6.00 3.00

Efficiency (C12) 1.00 0.20 3.00 0.25
Flexibility (C13) 1.00 8.00 2.00
Customization (C14) 1.00 0.16
Reliability (C15) 1.00
Table A2
Evaluation of sub-criteria with respect to Compatibility (CR = 0.00).
C21 C22
Quickness (C21) 1.00 0.50

Easiness (C22) 1.00
Table A3
Evaluation of sub-criteria with respect to Functionality (CR = 0.04).
C31 C32 C33 C34 C35
Accuracy (C31) 1.00 2.00 5.00 4.00 7.00

Validity (C32) 1.00 4.00 2.00 5.00
Robustly (C33) 1.00 0.33 0.50
Speed (C34) 1.00 3.00
Availability (C35) 1.00
Table A4
Evaluation of sub-criteria with respect to Security (CR = 0.05).
C41 C42 C43 C44 C45 C46
User authentication (C41) 1.00 1.00 0.33 4.00 2.00 6.00

Encrypted data Movement (C42) 1.00 0.25 6.00 4.00 3.00
Data Integrity (C43) 1.00 8.00 7.00 8.00
Security breach deduction (C44) 1.00 0.20 0.50
Physical Security (C45) 1.00 2.00
Audit trials (C46) 1.00
Table A5
Evaluation of sub-criteria with respect to Accessibility (CR = 0.00).
C51 C52 C53
User Interface (C51) 1.00 0.25 0.50

Disability access and translation (C52) 1.00 2.00
Content accessibility (C53) 1.00
13
Table A6
Evaluation of sub-criteria with respect to Easy to learn and use (CR = 0.00).
C61 C62
Time to learn (C61) 1.00 0.25

Easiness to learn all functions (C62) 1.00
Table A7
Evaluation of sub-criteria with respect to Empathy (CR = 0.00).
C71 C72 C73
Link (C71) 1.00 0.50 3.00

Customization (C72) 1.00 5.00
Customer care (C73) 1.00
Table A8
Evaluation of sub-criteria with respect to Information Quality (CR = 0.03).
C81 C82 C83
Information Richness (C81) 1.00 4.00 7.00

Information Accuracy (C82) 1.00 3.00
Information Up-date (C83) 1.00
Table A9
Evaluation of sub-criteria with respect to Responsiveness (CR = 0.03).
C91 C92 C93
Customer Service (C91) 1.00 0.25 0.14

Technical performance (C92) 1.00 0.33
Interactivity (C93) 1.00
A 10: Example for calculating D1 , D1 and CC1
(1 0.0016) 2
D1 = 31.7405
(0 0.0016) 2
D1 = 0.28062
D1 0.28062
CC1 = = = 0.00876351
D1 + D1 31.7405 + 0.28062
Appendix A11: Details of various mHealth applications
Cody
Cody is a new virtual learning app that facilitates a friendly exercise social platform where the users can share and perform workouts by following
new as well as old friends. This app enables users to monitor their own progress through a time bound manner and, at the same time, their daily
workout activities can be followed, liked and commented on by other users within the community. It has various interesting features, like an easy
interface and intuitive navigation.
Hot5 Fitness
Hot5 Fitness App provides step-by-step high-quality workouts by the industry's leading trainers. The application provides a range of instructors
and courses such as cardio, yoga, CrossFit, and traditional training. Hot5 provides video workouts of experts of different difficulty levels. The app is
very simple, attractive and it can be integrated very easily with mobile phones. The best part of this app is that it can be operated anywhere even
without the availability of internet.
Pact
Pact (previously recognized as GymPact) is a healthcare app. It is an incentive based mHealth application where the users receive money for
completing a required exercise level.
14
Carrot Fit
Carrot Fit is the newly developed app from the makers of Carrot To-Do and Carrot Alarm. Carrot Fit is a simple weight tracker which helps users
get necessary information regarding diet and exercise to maintain weight.
Human
Human is a newly launched mHealth app, which keeps a vigilant watch on the activities of users. The app recommends how many minutes you
moved today and how many minutes you need to move. It is very easy to use and very user-friendly compared to other competitive fitness
technology.
Moves
Moves is a day-to-day activity monitoring app that helps users to stay in better shape. This pedometer tracks the activities of users every minute in
a day and provides them with statistics of their activities like steps, movements, pedals and even stairs. It calculates how many calories are burned.
LoseIt
LoseIt is a calorie counting application that helps users in achieving their target of reducing weight. This app minutely takes care of everything
relating to diet and fitness. Based on users' general health goals and weight loss plans, the app suggests a customized weight loss plan for users. It
works as a personal trainer and nutritionist at their end. The app tracks the calorie requirements of the users.
Noom Weight Loss Coach

Noom operates as a dietary coach. It suggests proper meals and workouts for users through counting their steps all day long. It sends users dietary
recipes at regular intervals to keep them on track and away from boredom.
Healthy Out
Healthy Out is a dietary app that enables users to search meals in local restaurants that fulfil particular dietary demands quickly and efficiently.
By providing their post code, users can choose their nutritional needs like heart-healthy, gluten-free, fat-free, or any combination of one or more.
Zipongo
This app is freely available to iOS and Android users. Zipongo app facilitates shoppers to plan an ideal meal in day-to-day life. The app helps users
choose healthy foods with sufficient nutrients and vitamins in a cost effective manner.
Appendix B. Supplementary data
Supplementary data to this article can be found online at https://doi.org/10.1016/j.techsoc.2019.101186.
References Statista Research Department, 2019 Extracted on 29/07/2019. Available at:

https://www.statista.com/statistics/330695/number-of-smartphone-users-
worldwide/.
[1] T.U. Daim, C. Griffy-Brown, Infrastructure and service integration as a key process [16] L. Khansa, Z. Davis, H. Davis, A. Chin, H. Irvine, L. Nichols, N. MacMichael, Health
in technology management, Technol. Soc. 43 (2015) 95–96, https://doi.org/10. information technologies for patients with diabetes, Technol. Soc. 44 (2016) 1–9.
1016/j.techsoc.2015 .11.003. [17] D. Spatar, O. Kok, N. Basoglu, T. Daim, Adoption factors of electronic health re-
[2] WHO, mHealth New Horizons for Health through Mobile Technologies, World cord systems, Technol. Soc. 58 (2019) 101144.
Health Organization, 2011 Extracted on 2/7/219. Available at: https://www.who. [18] C. Chen, C. Watanabe, C. Griffy-Brown, The co-evolution process of technological
int/goe/publications/goe_mhealth_web.pdf. innovation—an empirical study of mobile phone vendors and telecommunication
[3] J. Razmak, C.H. Belanger, Comparing Canadian physicians and patients on their service operators in Japan, Technol. Soc. 29 (1) (2007) 1–22.
use of e-health tools, Technol. Soc. 51 (2017) 102–112. [19] E.S. Lohan, P.F. e Silva, A. Basiri, P. Peltola, Location Based Services Analysis
[4] Y.B. Li, L. James, J. McKibben, Trust between physicians and patients in the e- Through Analytical Hierarchical Processes: an e-Health-Based Case Study, Multi-
health era, Technol. Soc. 46 (2016) 28–34. Technology Positioning, Springer International Publishing, 2017, pp. 283–301.
[5] C. Liu, Q. Zhu, K.A. Holroyd, E.K. Seng, Status and trends of mobile-health ap- [20] W.H. Liao, W.L. Qiu, Applying analytic hierarchy process to assess healthcare-
plications for iOS devices: a developer's perspective, J. Syst. Softw. 84 (11) (2011) oriented cloud computing service systems, SpringerPlus 5 (1) (2016) 1030.
2022–2033. [21] G. Büyüközkan, G. Çifçi, A combined fuzzy AHP and fuzzy TOPSIS based strategic
[6] H. Li, T. Zhang, H. Chi, Y. Chen, Y. Li, J. Wang, Mobile health in China: current analysis of electronic service quality in healthcare industry, Expert Syst. Appl. 39
status and future development, Asian J. Psychiatr. 10 (2014) 101–104. (3) (2012) 2341–2354.
[7] D.M. West, Improving Health Care through Mobile Medical Devices and Sensors [22] P.K. Dey, S. Hariharan, O. Despic, Managing healthcare performance in analytical
vol. 10, Brookings Institution Policy Report, 2013, pp. 1–13. framework, Benchmarking Int. J. 15 (4) (2008) 444–468.
[8] J.G. Kahn, J.S. Yang, J.S. Kahn, ‘Mobile’health needs and opportunities in devel- [23] E.B. Sloane, M.J. Liberatore, R.L. Nydick, W. Luo, Q.B. Chung, Using the analytic
oping countries, Health Aff. 29 (2) (2010) 252–258. hierarchy process as a clinical engineering tool to facilitate an iterative, multi-
[9] B.M. Silva, J.J. Rodrigues, I. de la Torre Díez, M. López-Coronado, K. Saleem, disciplinary, microeconomic health technology assessment, Comput. Oper. Res. 30
Mobile-health: a review of current state in 2015, J. Biomed. Inform. 56 (2015) (10) (2003) 1447–1465.
265–272. [24] C.C. Lo, D.Y. Chen, C.F. Tsai, K.M. Chao, April. Service selection based on fuzzy
[10] L.H. Iwaya, M.A. Gomes, M.A. Simplício, T.C.M.B. Carvalho, C.K. Dominicini, TOPSIS method, Advanced Information Networking and Applications Workshops
R.R. Sakuragui, M.S. Rebelo, M.A. Gutierrez, M. Näslund, P. Håkansson, Mobile (WAINA), 2010 IEEE 24th International Conference on, IEEE, 2010, pp. 367–372.
health in emerging countries: a survey of research initiatives in Brazil, Int. J. Med. [25] A.A. Zaidan, B.B. Zaidan, A. Al-Haiqi, M.L.M. Kiah, M. Hussain, M. Abdulnabi,
Inform. 82 (5) (2013) 283–298. Evaluation and selection of open-source EMR software packages based on in-
[11] K. Ganapathy, A. Ravindra, mHealth: A Potential Tool for Health Care Delivery in tegrated AHP and TOPSIS, J. Biomed. Inform. 53 (2015) 390–404.
India, Rockefeller foundation, 2008. [26] M. Shafii, S.M. Hosseini, M. Arab, E. Asgharizadeh, F. Farzianpour, Performance
[12] M.S. Marcolino, J.A.Q. Oliveira, M. D'Agostino, A.L. Ribeiro, M.B.M. Alkmim, analysis of hospital managers using fuzzy AHP and fuzzy TOPSIS: Iranian
D. Novillo-Ortiz, The impact of mHealth interventions: systematic review of sys- Experience, Glob. J. Health Sci. 8 (2) (2016) 137.
tematic reviews, JMIR mHealth uHealth 6 (1) (2018) e23. [27] S. Kim, S. Kim, A multi-criteria approach toward discovering killer IoT application
[13] S. Kumar, W.J. Nilsen, A. Abernethy, A. Atienza, K. Patrick, M. Pavel, D. Hedeker, in Korea, Technol. Forecast. Soc. Chang. 102 (2016) 143–155.
Mobile health technology evaluation: the mHealth evidence workshop, Am. J. [28] E. Grigoroudis, C. Litos, V.A. Moustakis, Y. Politis, L. Tsironis, The assessment of
Prev. Med. 45 (2) (2013) 228–236. user-perceived web quality: application of a satisfaction benchmarking approach,
[14] D. West, How mobile devices are transforming healthcare, Issues Technol. Innovat. Eur. J. Oper. Res. 187 (3) (2008) 1346–1357.
18 (1) (2012) 1–11. [29] M. Striteska, M. Spickova, Review and comparison of performance measurement
[15] A. Holst, Number of Smartphone Users Worldwide from 2016 to 2021 (In Billions), systems, J. Organ. Manag. Stud. 2012 (2012) 1.
15
[30] D.Y. Lee, M.R. Lehto, User acceptance of YouTube for procedural learning: an multiattribute decision-making methods, Int. J. Approx. Reason. 14 (4) (1996)
extension of the technology acceptance model, Comput. Educ. 61 (2013) 193–208. 281–310.
[31] P. Legris, J. Ingham, P. Collerette, Why do people use information technology? A [64] M.B. Javanbarg, C. Scawthorn, J. Kiyono, B. Shahbodaghkhan, Fuzzy AHP-based
critical review of the technology acceptance model, Inf. Manag. 40 (3) (2003) multicriteria decision making systems using particle swarm optimization, Expert
191–204. Syst. Appl. 39 (1) (2012) 960–966.
[32] B.H. Wixom, P.A. Todd, A theoretical integration of user satisfaction and tech- [65] O. Duran, J. Aguilo, Computer-aided machine-tool selection based on a Fuzzy-AHP
nology acceptance, Inf. Syst. Res. 16 (1) (2005) 85–102. approach, Expert Syst. Appl. 34 (3) (2008) 1787–1794.
[33] S. Akter, J. D'Ambra, P. Ray, Service quality of mHealth platforms: development [66] O. Kulak, C. Kahraman, Fuzzy multi-attribute selection among transportation
and validation of a hierarchical model using PLS, Electron. Mark. 20 (3–4) (2010) companies using axiomatic design and analytic hierarchy process, Inf. Sci. 170 (2)
209–227. (2005) 191–210.
[34] F.C. Tung, S.C. Chang, C.M. Chou, An extension of trust and TAM model with IDT [67] M. Dağdeviren, İ. Yüksel, Developing a fuzzy analytic hierarchy process (AHP)
in the adoption of the electronic logistics information system in HIS in the medical model for behavior-based safety management, Inf. Sci. 178 (6) (2008) 1717–1733.
industry, Int. J. Med. Inform. 77 (5) (2008) 324–335. [68] C. Kahraman, D. Ruan, I. Doǧan, Fuzzy group decision-making for facility location
[35] E. Sezgin, S.Ö. Yıldırım, A literature review on attitudes of health professionals selection, Inf. Sci. 157 (2003) 135–153.
towards health information systems: from e-Health to m-Health, Procedia Technol. [69] T.L. Saaty, L.T. Tran, On the invalidity of fuzzifying numerical judgments in the
16 (2014) 1317–1326. Analytic Hierarchy Process, Math. Comput. Model. 46 (7–8) (2007) 962–975.
[36] J.H. Wu, S.C. Wang, What drives mobile commerce?: an empirical evaluation of [70] G.H. Tzeng, J.J. Huang, Multiple Attribute Decision Making: Methods and
the revised technology acceptance model, Inf. Manag. 42 (5) (2005) 719–729. Applications, CRC press, 2011.
[37] J.H. Wu, S.C. Wang, L.M. Lin, Mobile computing acceptance factors in the [71] D.L. Olson, Comparison of weights in TOPSIS models, Math. Comput. Model. 40
healthcare industry: a structural equation model, Int. J. Med. Inform. 76 (1) (7–8) (2004) 721–727.
(2007) 66–77. [72] I. Chamodrakas, D. Martakos, A utility-based fuzzy TOPSIS method for energy
[38] D. Shao, D. Shao, A Proposal of a Mobile Health Data Collection and Reporting efficient network selection in heterogeneous wireless networks, Appl. Soft
System for the Developing World, Master thesis project. Malm hgskola School of Comput. 12 (7) (2012) 1929–1938.
technology. Department of computer science, 2012, pp. 13–74. [73] M. Dağdeviren, Decision making in equipment selection: an integrated approach
[39] H. Kiljander, Evolution and Usability of Mobile Phone Interaction Styles, Helsinki with AHP and PROMETHEE, J. Intell. Manuf. 19 (4) (2008) 397–406.
University of Technology, 2004. [74] M.G. Kharat, S. Murthy, S.J. Kamble, R.D. Raut, S.S. Kamble, M.G. Kharat, Fuzzy
[40] M. Mohageg, E. Bergman, Information appliance design at sun microsystems, multi-criteria decision analysis for environmentally conscious solid waste treat-
CHI'00 Extended Abstracts on Human Factors in Computing Systems, ACM, 2000, ment and disposal technology selection, Technol. Soc. 57 (2019) 20–29.
April, pp. 63–64. [75] M.V. Khatir, Z. Akbarzadeh, Elucidation of structural relationships of SWOT: a
[41] H.E. Rippen, E.C. Pan, C. Russell, C.M. Byrne, E.K. Swift, Organizational frame- mixed method approach based on FMADM for formulating science and technology
work for health information technology, Int. J. Med. Inform. 82 (4) (2013) e1–e13. strategies, Technol. Soc. 56 (2019) 44–56.
[42] B.M. Silva, J.J. Rodrigues, F. Canelo, I.C. Lopes, L. Zhou, A data encryption so- [76] G. Büyüközkan, O. Feyzioğlu, E. Nebol, Selection of the strategic alliance partner
lution for mobile health apps in cooperation environments, J. Med. Internet Res. in logistics value chain, Int. J. Prod. Econ. 113 (1) (2008) 148–158.
15 (4) (2013). [77] T.Y. Chen, C.Y. Tsao, The interval-valued fuzzy TOPSIS method and experimental
[43] R.M. Brian, D. Ben-Zeev, Mobile health (mHealth) for mental health in Asia: ob- analysis, Fuzzy Sets Syst. 159 (11) (2008) 1410–1428.
jectives, strategies, and limitations, Asian J. Psychiatr. 10 (2014) 96–100. [78] S. Önüt, S. Soner, Transshipment site selection using the AHP and TOPSIS ap-
[44] S. Dünnebeil, A. Sunyaev, I. Blohm, J.M. Leimeister, H. Krcmar, Determinants of proaches under fuzzy environment, Waste Manag. 28 (9) (2008) 1552–1559.
physicians' technology acceptance for e-health in ambulatory care, Int. J. Med. [79] Y.M. Wang, T.M. Elhag, Fuzzy TOPSIS method based on alpha level sets with an
Inform. 81 (11) (2012) 746–760. application to bridge risk assessment, Expert Syst. Appl. 31 (2) (2006) 309–319.
[45] A. Mehrpouya, R. Samiolo, Performance measurement in global governance: [80] T. Yang, C.C. Hung, Multiple-attribute decision making methods for plant layout
ranking and the politics of variability, Account. Org. Soc. 55 (2016) 12–31. design problem, Robot. Comput. Integr. Manuf. 23 (1) (2007) 126–137.
[46] S. Salsbury Lyons, T. Tripp-Reimer, B.A. Sorofman, J.E. DeWitt, B.J. Boots Miller, [81] D. Yong, Plant location selection based on fuzzy TOPSIS, Int. J. Adv. Manuf.
T.E. Vaughn, Information technology for clinical guideline implementation: per- Technol. 28 (7) (2006) 839–844.
ceptions of multidisciplinary stakeholders, J. Am. Med. Inform. Assoc. 12 (2005) [82] C. Kahraman, G. Büyüközkan, N.Y. Ateş, A two phase multi-attribute decision-
64–71. making approach for new product introduction, Inf. Sci. 177 (7) (2007)
[47] J.K. Ayeh, Travellers' acceptance of consumer-generated media: an integrated 1567–1582.
model of technology acceptance and source credibility theories, Comput. Hum. [83] Y.H. Chang, C.H. Yeh, A survey analysis of service quality for domestic airlines,
Behav. 48 (2015) 173–180. Eur. J. Oper. Res. 139 (1) (2002) 166–177.
[48] J.M.O. Egea, M.V.R. González, Explaining physicians' acceptance of EHCR sys- [84] C. Kahraman, A. Beskese, D. Ruan, Measuring flexibility of computer integrated
tems: an extension of TAM with trust and risk factors, Comput. Hum. Behav. 27 (1) manufacturing systems using fuzzy cash flow analysis, Inf. Sci. 168 (1) (2004)
(2011) 319–332. 77–94.
[49] I.C. Chang, H.G. Hwang, W.F. Hung, Y.C. Li, Physicians' acceptance of pharma- [85] H.J. Zimmermann, Fuzzy set theory, Wiley Interdisciplinary Rev.: Comput. Stat. 2
cokinetics-based clinical decision support systems, Expert Syst. Appl. 33 (2) (3) (2010) 317–332.
(2007) 296–303. [86] J.F. Ding, G.S. Liang, Using fuzzy MCDM to select partners of strategic alliances for
[50] Y.H. Chang, C.H. Yeh, S.Y. Wang, A survey and optimization-based evaluation of liner shipping, Inf. Sci. 173 (1) (2005) 197–225.
development strategies for the air cargo industry, Int. J. Prod. Econ. 106 (2) [87] Z.S. Xu, J. Chen, An interactive method for fuzzy multiple attribute group decision
(2007) 550–562. making, Inf. Sci. 177 (1) (2007) 248–263.
[51] C.C. Sun, A performance evaluation model by integrating fuzzy AHP and fuzzy [88] E.E. Karsak, E. Tolga, Fuzzy multi-criteria decision-making procedure for evalu-
TOPSIS methods, Expert Syst. Appl. 37 (12) (2010) 7745–7754. ating advanced manufacturing system investments, Int. J. Prod. Econ. 69 (1)
[52] A. Bhattacherjee, G. Premkumar, Understanding changes in belief and attitude (2001) 49–64.
toward information technology usage: a theoretical model and longitudinal test, [89] S.M. Chen, Evaluating weapon systems using fuzzy arithmetic operations, Fuzzy
MIS Q. (2004) 229–254. Sets Syst. 77 (3) (1996) 265–276.
[53] E.A. Wahdain, M.N. Ahmad, User acceptance of information technology: factors, [90] C.T. Chen, C.T. Lin, S.F. Huang, A fuzzy approach for supplier evaluation and
theories and applications, J. Inf. Syst. Res. Innovat. 6 (2014) 17–25. selection in supply chain management, Int. J. Prod. Econ. 102 (2) (2006) 289–301.
[54] F.C. Tung, S.C. Chang, Nursing students' behavioral intention to use online [91] C.H. Cheng, Y. Lin, Evaluating the best main battle tank using fuzzy decision
courses: a questionnaire survey, Int. J. Nurs. Stud. 45 (9) (2008) 1299–1309. theory with linguistic criteria evaluation, Eur. J. Oper. Res. 142 (1) (2002)
[55] M.M. Yusof, J. Kuljis, A. Papazafeiropoulou, L.K. Stergioulas, An evaluation fra- 174–186.
mework for Health Information Systems: human, organization and technology-fit [92] S. Alharbi, S. Drew, Using the technology acceptance model in understanding
factors (HOT-fit), Int. J. Med. Inform. 77 (6) (2008) 386–398. academics' behavioural intention to use learning management systems, Int. J. Adv.
[56] S.S. Nudurupati, U.S. Bititci, V. Kumar, F.T. Chan, State of the art literature review Comput. Sci. Appl. 5 (1) (2014) 143–155.
on performance measurement, Comput. Ind. Eng. 60 (2) (2011) 279–290. [93] P.A. Raj, D.N. Kumar, Ranking alternatives with fuzzy weights using maximizing
[57] M. Dağdeviren, S. Yavuz, N. Kılınç, Weapon selection using the AHP and TOPSIS set and minimizing set, Fuzzy Sets Syst. 105 (3) (1999) 365–375.
methods under fuzzy environment, Expert Syst. Appl. 36 (4) (2009) 8143–8151. [94] T.C. Wang, T.H. Chang, Application of TOPSIS in evaluating initial training air-
[58] F.T. Chan, N. Kumar, M.K. Tiwari, H.C. Lau, K.L. Choy, Global supplier selection: a craft under a fuzzy environment, Expert Syst. Appl. 33 (4) (2007) 870–880.
fuzzy-AHP approach, Int. J. Prod. Res. 46 (14) (2008) 3825–3857. [95] M.P. Amiri, Project selection for oil-fields development by using the AHP and
[59] T.L. Saaty, Decision making with the analytic hierarchy process, Int. J. Serv. Sci. 1 fuzzy TOPSIS methods, Expert Syst. Appl. 37 (9) (2010) 6218–6224.
(1) (2008) 83–98. [96] X. Yu, S. Guo, J. Guo, X. Huang, Rank B2C e-commerce websites in e-alliance
[60] Y.C. Chou, Y.Y. Hsu, H.Y. Yen, Human resources for science and technology: based on AHP and fuzzy TOPSIS, Expert Syst. Appl. 38 (4) (2011) 3550–3557.
analyzing competitiveness using the analytic hierarchy process, Technol. Soc. 30 [97] H. Hayati, S. Karimi, J. Sadeghifar, J. Ebrahimzadeh, S. Afshari, B. Khosravi,
(2) (2008) 141–153. E. Ashrafi, Determining the entitlement to structural indicators of health by means
[61] Z.D.U. Durmuşoğlu, Assessment of techno-entrepreneurship projects by using of fuzzy AHP and TOPSIS: a case study in Sistan and Baluchestan, Iran, J.
analytical hierarchy process (AHP), Technol. Soc. 54 (2018) 41–46. Pharmacoecon. Pharm. Manag. 1 (3/4) (2015) 61–64.
[62] T. Daim, D. Yates, Y. Peng, B. Jimenez, Technology assessment for clean energy [98] A. Panayides, Z. Antoniou, M.S. Pattichis, C.S. Pattichis, A.G. Constantinides,
technologies: the case of the Pacific Northwest, Technol. Soc. 31 (3) (2009) August. High efficiency video coding for ultrasound video communication in m-
232–243. health systems, Engineering in Medicine and Biology Society (EMBC), 2012
[63] E. Triantaphyllou, C.T. Lin, Development and evaluation of five fuzzy Annual International Conference of the IEEE, IEEE, 2012, pp. 2170–2173.
16
[99] M. Serpoush, M.H. Askariazad, F. Yaghmaie, An AHP framework for hospital in- [122] D.D. Luxton, R.A. Kayl, M.C. Mishkind, mHealth data security: the need for
formation systems' evaluation: the case Of tehran private hospitals, Int. J. Bus. HIPAA-compliant standardization, Telemed. e-Health 18 (4) (2012) 284–288.
Manag. Stud. 5 (2) (2013). [123] S. Fong, Y. Zhuang, Using medical history embedded in biometrics medical card
[100] K. Vincová, Using DEA models to measure efficiency, Biatec 13 (8) (2005) 24–28. for user identity authentication: data representation by AVT hierarchical data tree,
[101] S. Marshall, K. Haywood, R. Fitzpatrick, Impact of patient‐reported outcome BioMed Res. Int. 2012 (2012).
measures on routine practice: a structured review, J. Eval. Clin. Pract. 12 (5) [124] R. Yudkin, Popularity of mobile devices brings risk, Health Manag. Technol. 33 (4)
(2006) 559–568. (2012) 32-32.
[102] E. van der Geer, H.F. van Tuijl, C.G. Rutte, Performance management in health- [125] K. Malhotra, S. Gardner, W. Mepham, A novel implementation of signature, en-
care: performance indicator development, task uncertainty, and types of perfor- cryption and authentication (SEA) protocol on mobile patient monitoring devices,
mance indicators, Soc. Sci. Med. 69 (10) (2009) 1523–1530. Technol. Health Care 16 (4) (2008) 261–272.
[103] T.A. Sanner, L.K. Roland, K. Braa, From pilot to scale: towards an mHealth ty- [126] T. Escobar-Rodríguez, E. Carvajal-Trujillo, Online purchasing tickets for low cost
pology for low-resource contexts, Health Pol. Technol. 1 (3) (2012) 155–164. carriers: an application of the unified theory of acceptance and use of technology
[104] C. López-Nicolás, F.J. Molina-Castillo, H. Bouwman, An assessment of advanced (UTAUT) model, Tour. Manag. 43 (2014) 70–88.
mobile services acceptance: contributions from TAM and diffusion theory models, [127] M.S. Househ, E.M. Borycki, W.M. Rohrer, A.W. Kushniruk, Developing a frame-
Inf. Manag. 45 (6) (2008) 359–364. work for meaningful use of personal health records (PHRs), Health Pol. Technol. 3
[105] N.C. Goodwin, Functionality and usability, Commun. ACM 30 (3) (1987) 229–233. (4) (2014) 272–280.
[106] C. De Blok, B. Meijboom, K. Luijkx, J. Schols, The human dimension of modular [128] B. Suh, I. Han, The impact of customer trust and perception of security control on
care provision: opportunities for personalization and customization, Int. J. Prod. the acceptance of electronic commerce, Int. J. Electron. Commer. 7 (3) (2003)
Econ. 142 (1) (2013) 16–26. 135–161.
[107] R.E. Glasgow, S. Askew, P. Purcell, E. Levine, E.T. Warner, K.C. Stange, [129] T.A. Al-Saadi, T.M. Aljarrah, A.M. Alhashemi, A. Hussain, A systematic review of
G.A. Colditz, G.G. Bennett, Use of RE-AIM to address health inequities: application usability challenges and testing in mobile health, Int. J. Account. Financ. Report. 5
in a low-income community health center-based weight loss and hypertension self- (2) (2015) 1–14.
management program, Translat. Behav. Med. 3 (2) (2013) 200–210. [130] A. Nischelwitzer, K. Pintoffl, C. Loss, A. Holzinger, November. Design and devel-
[108] S.Y. Park, An analysis of the technology acceptance model in understanding uni- opment of a mobile medical application for the management of chronic diseases:
versity students' behavioral intention to use e-learning, J. Educat. Technol. Soc. 12 methods of improved data input for older people, Symposium of the Austrian HCI
(3) (2009) 150. and Usability Engineering Group, Springer, Berlin, Heidelberg, 2007, pp. 119–132.
[109] M. Hirst, P.F. Luckett, K.T. Trotman, Effects of feedback and task predictability on [131] R.U. Bilsel, G. Büyüközkan, D. Ruan, A fuzzy preference‐ranking model for a
task learning and judgment accuracy, Abacus 35 (3) (1999) 286–301. quality evaluation of hospital web sites, Int. J. Intell. Syst. 21 (11) (2006)
[110] G. Eysenbach, T.L. Diepgen, The role of e-health and consumer health informatics 1181–1197.
for evidence-based patient choice in the 21st century, Clin. Dermatol. 19 (1) [132] F. Patsioura, S. Kitsiou, A. Markos, July. Evaluation of Greek public hospital
(2001) 11–17. websites, ICE-B, 2009, pp. 223–229.
[111] A.D. Patil, A.V. Shekdar, Health-care waste management in India, J. Environ. [133] E. Grigoroudis, E. Orfanoudaki, C. Zopounidis, Strategic performance measure-
Manag. 63 (2) (2001) 211–220. ment in a healthcare organisation: a multiple criteria approach based on balanced
[112] L. Wu, J.Y. Li, C.Y. Fu, The adoption of mobile healthcare by hospital profes- scorecard, Omega 40 (1) (2012) 104–119.
sionals: an integrative perspective, Decis. Support Syst. 51 (3) (2011) 587–596. [134] E. Cristobal, C. Flavián, M. Guinaliu, Perceived e-service quality (PeSQ)
[113] R.H. Shroff, C.C. Deneen, E.M. Ng, Analysis of the technology acceptance model in Measurement validation and effects on consumer satisfaction and web site loyalty,
examining students' behavioural intention to use an e-portfolio system, Australas. Manag. Serv. Qual.: Int. J. 17 (3) (2007) 317–340.
J. Educ. Technol. 27 (4) (2011). [135] B. Yoo, N. Donthu, Developing a scale to measure the perceived quality of an
[114] K. Park, J. Park, Y.D. Kwon, Y. Kang, J.W. Noh, Public satisfaction with the Internet shopping site (SITEQUAL), Q. J. Electron. Commer. QJEC 2 (1) (2001)
healthcare system performance in South Korea: universal healthcare system, 31–45.
Health Policy 120 (6) (2016) 621–629. [136] Q. Ma, L. Liu, The role of Internet self-efficacy in the acceptance of web-based
[115] F. Lucivero, B. Prainsack, The life stylisation of healthcare? ‘Consumer genomics’ electronic medical records, Contemp. Issues End User Comput. 17 (1) (2005)
and mobile health as technologies for healthy lifestyle, Appl. Transl. Genom. 4 38–57.
(2015) 44–49. [137] J. Cohen, 11 health and fitness apps that achieve top results, Retrieved February
[116] B. Kamsu-Foguem, C. Foguem, Telemedicine and mobile health with integrative 14, 2017, from, 2015. http://www.forbes.com/sites/jennifercohen/2015/01/07/
medicine in developing countries, Health Pol. Technol. 3 (4) (2014) 264–271. the-11-top-health-fitness-apps-that-achieve-the-best-results/#73d69bac1aca.
[117] A.M. Mosadeghrad, Factors influencing healthcare service quality, Int. J. Health [138] C. Kahraman, S. Çevik, N.Y. Ates, M. Gülbay, Fuzzy multi-criteria evaluation of
Policy Manag. 3 (2) (2014) 77. industrial robotic systems, Comput. Ind. Eng. 52 (4) (2007) 414–433.
[118] W.L. Currie, J.J. Seddon, A cross-national analysis of eHealth in the European [139] T.L. Saaty, M.S. Ozdemir, Why the magic number seven plus or minus two, Math.
Union: some policy and research directions, Inf. Manag. 51 (6) (2014) 783–797. Comput. Model. 38 (3–4) (2003) 233–244.
[119] B.S. Correa, B. Gonçalves, I.M. Teixeira, A.T. Gomes, A. Ziviani, AToMS: a ubi- [140] X. Li, K. Wang, L. Liu, J. Xin, H. Yang, C. Gao, Application of the entropy weight
quitous teleconsultation system for supporting AMI patients with prehospital and TOPSIS method in safety evaluation of coal mines, Procedia Eng. 26 (2011)
thrombolysis, Int. J. Telemed. Applicat. 2011 (2011) 2. 2085–2091.
[120] A.D.F. dos Santos, C. de Souza, N.R. de Queiroz, G.C. e Penna, H.J. Alves, [141] Z. Redenovic, I. Veselinovic, Integrated AHP-TOPSIS method for the assessment of
Incorporation of telehealth resources in Belo Horizonte's SAMU: qualifying and health management information systems efficiency, Econ. Themes 55 (1) (2017)
improving care, eHealth, Telemedicine, and Social Medicine, 2009. eTELEMED'09. 121–142.
International Conference on, IEEE, 2009, February, pp. 72–76. [142] Y. Liu, Y. Yang, Y. Liu, G.H. Tzeng, Improving sustainable mobile health care
[121] J.E. van Gemert-Pijnen, H.C. Ossebaard, Proceedings 5th International Conference promotion: a novel hybrid MCDM method, Sustainability 11 (3) (2019) 752.
on eHealth, Telemedicine, and Social Medicine eTELEMED 2013, 2013.
17

Technology in Society: Sciencedirect

Uploaded by

Copyright:

Available Formats

Technology in Society: Sciencedirect

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Technology in Society: Sciencedirect

Uploaded by

Copyright:

Available Formats

Technology in Society 59 (2019) 101186

Contents lists available at ScienceDirect

Evaluation and selection of mobile health (mHealth) applications using AHP T

ARTICLE INFO ABSTRACT

3.1. Analytic Hierarchy Process Table 2

Fuzzy Set U˜ = {(x,µũ (x ), x X)}

Definition 2. If ã is a triangular fuzzy number, then it is represented in

User satisfaction (C1) Utility (C11) [99]

Fig. 3. Hierarchy of mHealth application model.

very hot. These linguistic values can also be characterized by fuzzy

User satisfaction 0.143 Utility 0.360 0.051

The proposed model is illustrated with a numerical case example.

Fig. 4. Proposed model for mHealth application selection.

C11 C12 C13 C14 C15

Utility (C11) 1.00 5.00 1.00 6.00 3.00

Quickness (C21) 1.00 0.50

C31 C32 C33 C34 C35

Accuracy (C31) 1.00 2.00 5.00 4.00 7.00

C41 C42 C43 C44 C45 C46

User authentication (C41) 1.00 1.00 0.33 4.00 2.00 6.00

C51 C52 C53

User Interface (C51) 1.00 0.25 0.50

Time to learn (C61) 1.00 0.25

C71 C72 C73

Link (C71) 1.00 0.50 3.00

C81 C82 C83

Information Richness (C81) 1.00 4.00 7.00

C91 C92 C93

Customer Service (C91) 1.00 0.25 0.14

A 10: Example for calculating D1 , D1 and CC1

Appendix A11: Details of various mHealth applications

Noom Weight Loss Coach

Appendix B. Supplementary data

Supplementary data to this article can be found online at https://doi.org/10.1016/j.techsoc.2019.101186.

References Statista Research Department, 2019 Extracted on 29/07/2019. Available at:

You might also like