www.nature.

com/scientificreports

OPEN Smartphones dependency risk

analysis using machine‑learning
predictive models
Claudia Fernanda Giraldo‑Jiménez 1,2, Javier Gaviria‑Chavarro 2
, Milton Sarria‑Paja 3*
,
Leonardo Antonio Bermeo Varón 3, John Jairo Villarejo‑Mayor 4,5
&
André Luiz Felix Rodacki 5
Recent technological advances have changed how people interact, run businesses, learn, and use
their free time. The advantages and facilities provided by electronic devices have played a major role.
On the other hand, extensive use of such technology also has adverse effects on several aspects of
human life (e.g., the development of societal sedentary lifestyles and new addictions). Smartphone
dependency is new addiction that primarily affects the young population. The consequences
may negatively impact mental and physical health (e.g., lack of attention or local pain). Health
professionals rely on self-reported subjective information to assess the dependency level, requiring
specialists’ opinions to diagnose such a dependency. This study proposes a data-driven prediction
model for smartphone dependency based on machine learning techniques using an analytical
retrospective case–control approach. Different classification methods were applied, including classical
and modern machine learning models. Students from a private university in Cali—Colombia (n = 1228)
were tested for (i) smartphone dependency, (ii) musculoskeletal symptoms, and (iii) the Risk Factors
Questionnaire. Random forest, logistic regression, and support vector machine-based classifiers
exhibited the highest prediction accuracy, 76–77%, for smartphone dependency, estimated through
the stratified-k-fold cross-validation technique. Results showed that self-reported information
provides insight into predicting smartphone dependency correctly. Such an approach opens doors
for future research aiming to include objective measures to increase accuracy and help to reduce the
negative consequences of this new addiction form.

Abbreviations
MD Mobile device
SDT Smartphone dependency test
OR Odds ratio
MSD Musculoskeletal disorder
ROC Receiver operating characteristic
SVM Support vector machine
PSU Problematic/Addictive smartphone use

The use of smartphones and mobile devices has experienced exponential growth in the last decade, and such
devices have become usual for work, education, daily tasks, and social l­ife1. An added value of smartphones is
based on personal use for everyday organization, communication, and entertainment, increasing the ubiquity
of digital tools during daily ­routines2. Despite many positive aspects, some adverse effects are derived from
extensive usage by young ­individuals3,4. The use of mobile devices is an occupational ­reality5. The impact of
smartphone usage on cognitive abilities for educational, occupational, and social functioning can be classified
as negative or positive from their socio-emotional ­components6. Moreover, the impact on children and teenag-
ers’ physical and mental health has been evidenced, modulated by exposure times and compulsive ­behaviors7.
Smartphones have a repertoire of tools that have altered consumption dynamics and how users interact within

1
Department of Health, Universidad Santiago de Cali, Cali, Colombia. 2Doctoral Program in Applied Sciences,
Universidad Santiago de Cali, Cali, Colombia. 3Department of Engineering, Universidad Santiago de Cali,
Cali, Colombia. 4Department of Electrical and Electronic Engineering, Federal University of Santa Catarina,
Florianopolis, Brazil. 5Department of Physical Education, Federal University of Paraná, Curitiba, Paraná,
Brazil. *email: [email protected]

Scientific Reports | (2022) 12:22649 | https://doi.org/10.1038/s41598-022-26336-2 1

Vol.:(0123456789)
 www.nature.com/scientificreports/

different ­environments8. There are many instances where other organizations (e.g., industrial, educational, com-
mercial, and advertising sectors) have created mobile applications for communication p ­ urposes9. These applica-
10
tions help to improve collaboration and facilitate information e­ xchange . They also provide the business sector
with information on the improvement and development of mobile applications to achieve business objectives,
cover new markets, and attract ­demand11. Available tools have three main features: accessibility, repetition, and
interactivity, which generate a high affinity towards these devices, whereas smartphones stand ­out12. Consumer
and information applications and social networks have high demand and influence individual communication
and lifestyle ­habits13.
Over the last decade, the use of mobile devices in different communities has become widespread, and its
lasting effects have multiplied. For instance, smartphones are effective learning tools in educational settings
to gain knowledge. There is a significant effect on the academic performance of undergraduate students when
mobile applications intermediate learning compared to traditional learning s­ chemes14. In this way, technologi-
cal improvements in the educational sector create the need to propose new strategies to offer students guidance
using efficient management of technical resources, to strengthen the learning ­process15. The excessive use of
smartphones is more prevalent among student populations than others. Easy access to the internet and big
screens for game interaction are factors significantly associated with blindness, deafness, and i­ nattentiveness23.
Mobile device dependency is a problem established in terms of frequency and excessive use. There is a preva-
lence of approximately 40% excessive use of mobile devices overall users, and about 42% of them belong to the
group of middle-low-class h ­ ouseholds16, with significant representation in the young p ­ opulation17. This habit is
negatively associated with inhibition, decision-making, memory performance, and sleep ­disorders18. Besides,
the simultaneous use of a cell phone during daily activities may represent an overload for some muscle groups
and constitute a risk factor for musculoskeletal onset p ­ roblems19.
Different studies on the problems derived from the use of mobile devices show preferences toward gaming.
However, users do not use these devices for gaming purposes but also multiple-purpose applications 20. These
applications are an integral part of modern life and, therefore, can create adverse dependency ­effects21. Conse-
quently, it is crucial to quantify the dependency using accurate scales and to incorporate ways of analyzing the
effects of excessive and harmful smartphone u ­ se22.
Implementing strategies to detect and monitor risk factors associated with smartphone dependency is impera-
tive. These strategies should promote participation in recreational activities and strengthen social relationships.
Reducing the adverse effects of smartphone dependency, postural problems, musculoskeletal symptoms, and
even deformities or chronic injuries may be prevented. The negative consequences on academic performance,
working, and social life can also be influenced.
In this study, the Smartphone Dependency Test (SDT) questionnaire was used to assess dependency among
university students. The SDT questionnaire was validated and linguistically adapted in 2016 for public and private
university students, with reliability for abstinence and tolerance (α = 0.901), for abuse and difficulty in controlling
the impulse (α = 0.853), and for problems caused by excessive use (α = 0.762)24.
Research involving predictive models to assess smartphone dependency is scarce. To the best of our knowl-
edge, there are no studies quantifying and using analytical techniques such as machine learning to model vari-
ables associated with smartphone dependency. It is worth mentioning that there is a significant advance in
using these tools to solve different research p ­ roblems25,26. However, they have not been widely used to generate
predictive models focused on smartphone dependence and have not been e­ stablished27,28.
This study proposed using self-reported information gathered through standardized questionnaires to train
predictive models using a machine-learning approach. It was hypothesized that the proposed questionnaires
could help to encode self-reported subjective information, which can be used to predict smartphone depend-
ency. The input variables consider factors related to personal data, family, environmental risks, physical loading,
device-specific risk factors, and musculoskeletal symptoms. Such an approach may reduce the bias during the
assessment process. This also may assist professionals in recommending actions to reduce the adverse effects of
overusing mobile devices. To our knowledge, no previous studies address this issue from a data-driven models’
standpoint. This study also provides insights that may entitle one to link subjective cues to objective measures
in future analyses.

Methods
Participants and procedures. The study is an analytical observation using a retrospective case–control
approach involving 14,858 students from 19 undergraduate programs. The students were registered in four
schools of a private university in Cali, Colombia, in 2019. A 95% confidence level and a 5% margin of error were
used for the sample calculation, resulting in a sample of 1247 students. The sampling technique was randomly
stratified. The selection of the participants was performed by probability sampling using the epi-info™ s­ uite29.
Eighteen individuals were excluded after they met the exclusion criteria. These participants used the upper limbs
(arms and hands) in regular physical activities such as high-impact sports (basketball, volleyball, table tennis,
and weights in the gym) and repetitive movement in artistic activities (such as painting, embroidery crafts, and
playing musical instruments such as guitar and drum).
Consequently, the frequency and intensity of these activities could cause information bias, allowing the
control for selection bias. Therefore, the final sample was recalculated for 1228 students (95% CI; 5% error). The
Levene test confirmed data homogeneity, and the sample was comparable in age, sex, program, semester, and
marital status (0.157–0.740). The participants were then assigned according to their smartphone dependency. The
case group was composed of students with some smartphone dependency, while the control group was formed
of students with no smartphone dependency.

Scientific Reports | (2022) 12:22649 | https://doi.org/10.1038/s41598-022-26336-2 2

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 1.  Automatic system for predicting smartphone dependency.

The volunteers signed an informed consent form before participating in the study. Those individuals who
submitted an incomplete form or frequently played sports or artistic activities involving the upper limbs were
excluded.
The Smartphone Dependency Test is a free-to-use test created by Chóliz30, which was validated and linguisti-
cally adapted in 2016 for students receiving both public and private e­ ducation31. This test was used to measure
the level of independence of Mobile Devices (MD), which was assigned as the dependent variable. The test lasted
10 min and consisted of 22 items presented using a Likert-type scale. The scores range from 0 (zero) to 88 as
the maximum to determine whether the dependency was absent (0–29), low (30–38), medium (39–48), or high
(49–88). In addition, musculoskeletal disorders (MSD) were characterized via the Nordic Questionnaire, in its
Spanish version, whose application lasted 7 min. The questionnaire comprised two levels: (i) a general level that
sought to determine the occurrence of musculoskeletal discomfort by anatomical regions, and (ii) a specific
level that focused on delving into the chronology, frequency, duration, intensity, and impact of the discomfort
on their everyday activities.
The risk factors were the independent variables. The Risk Factors Questionnaire was designed and subjected
to internal validation by the researchers through the Delphi method by a group of 6 experts, obtaining a validity
of approximately 0.9, according to Chronbach’s alpha; its application lasted 7 min. This questionnaire included
the variables considered in the theoretical framework about sociodemographic, interpersonal, and contextual
factors related to the device and physical load. It was possible to identify the risk factors in the university student
­population32.
The study followed the principles of the Helsinki Declaration, guaranteeing confidentiality by coding and
signing the informed consent before participation. Regarding data collection, this study protocol was doubly
reviewed and endorsed by the Scientific Committee of Ethics and Bioethics of the Universidad Santiago de Cali
(act # 03 of 2019).

Data analysis. The data were recorded by a double entry in Excel. The information from the two databases
was compared, and unmatched data were cleaned, performing verification in the primary source.
To structure the model construction, the variables were transformed into categorical types for the processing
and analysis phase. The data allocation, which was 1%, was performed using the mode for qualitative variables
and the arithmetic mean for quantitative variables. Once the information was validated, a descriptive exploratory
analysis of the different variables was conducted to determine their behavior. Subsequently, a bivariate analysis
was performed to determine which were included in the model and selected for statistical significance with a
p-value < 0.05.
Figure 1 shows a schematic representation of the research approach. It indicates a general-purpose pattern-
recognition system adapted to address the overuse of smartphones. First, participants answered three question-
naires (i.e., the Smartphone Dependency Test, the Nordic Questionnaire—Spanish version, and the Risk Factors
Questionnaire) used by health professionals to assess the participant dependency level. Next, a selection strategy
and descriptive exploratory analyses of the different variables were performed to determine which predictors were
highly correlated to the target variables. As a result, 31 variables were selected and used to feed the data-driven
predictive model. Two groups of algorithms were applied—i.e., the classical approach and the deep learning
approach. The details of the algorithms are provided in the following section. Finally, based on these predictive
models, smartphone dependency and overusing were estimated.
Data processing, debugging, modeling, and validation were structured in six stages and are described in Fig. 2.

Supervised machine learning techniques. Machine learning has been successfully used in several
research areas with applications in medical signal processing, computer-assisted systems, language processing,
and healthcare, among others. From the classical point of view to more recent deep learning techniques, data-
driven models try to capture the inner structure of data derived from external systems. These models help make
predictions on new unseen d ­ ata26,33,34. There is a wide range of applications that vary from healthcare, transpor-
tation, social networks, banking, security, and education. Internet of Things (IoT) Networks is widespread in
many industrial applications. Machine Learning models help identify and avoid malicious traffic attacks, which
can affect network security and essential s­ ervices35–38. These techniques have been used to improve the user’s
experience and decision-making process, which are more subjective scenarios and more dependent on the user’s
psychological ­characteristics39,40. It is important to note that in such scenarios, it is necessary to analyze people’s
opinions, sentiments, perceptions, etc., to help develop tools in multiple situations to allow users’ interaction

Scientific Reports | (2022) 12:22649 | https://doi.org/10.1038/s41598-022-26336-2 3

Vol.:(0123456789)
 www.nature.com/scientificreports/

Figure 2.  Information processing flowchart to find out the model.

with applications, products, and ­services40–42. This is the possibility explored in this study, in which users are
required to respond to a self-report standardized questionnaire that can be linked to smartphone dependency.
To have a precise notation, x(i) denotes the input variables arranged as an n-dimension vector, also known as
features, while y(i) indicates the output or target variable (i.e., the predicting variable). The pair (x(i),y(i)) is a train-
ing example. The dataset containing the information from m training examples {(x(i),y(i))}; i = 1… m, is known as
the training set. Typically, X and Y are used to denote the space representations of the input and output variables,
respectively. When a classification problem is approached, the variables in the Y space take discrete values cor-
responding to the classes or categories defined in the learning problem. For the specific problem addressed in
this work, y ∈ {0, 1}, where a value y = 0 has been defined to indicate a person with a negative diagnosis, whereas
y = 1 indicates a person with a positive diagnosis of smartphone dependency.
A supervised learning problem estimates a function h ­ ɵ(x): X → Y, such that given an input x, ­hɵ(x) predicts
the y value. The function h ­ ɵ(x) is also known as the hypothesis function.
Several approaches have been applied to define the h ­ ɵ (x) function. From classical approaches such as logis-
tic ­regression43, Support vector machines (SVM) with polynomial and Radial Basis Functions (RBF) kernels,
which is considered a discriminative a­ pproach44, Decision t­ ree45, and Random ­forest46, to modern approaches
based on deep learning (DL) such as multilayer perceptron (MLP)33, and tabular data such as ­TabNet47, as is
the particular case of the present study. A detailed description of previously mentioned techniques is out of the
scope of this paper.
Deep learning techniques are well known for their performance when solving problems related to images,
audio, and ­text25,26. One of the shortcomings of training a deep learning model is having sufficient data for a
proper parameter ­estimation26. Some approaches include transiently modifying the output to fit the requirements
and then fine-tuning learning, where a previously trained model can be ­applied25. However, in this work, the
amount of data was relatively limited to infer that a deep neural network would be adequately trained; neither
three are pre-trained models of adjacent problems so that transfer learning can be used. Hence, classical machine
learning techniques are expected.

System validation. The assisted diagnosis process using automated systems is imperfect. The result
obtained from a classification system represents a probability rather than a correct answer with irrefutable cer-
tainty. Different diagnostic measures are thus employed to verify and assure that the results are repeatable and to
validate the ability of a system to identify the presence or absence of disease.
In particular, random cross-validation (tenfold) was used in these experiments. The available data were
used for data training (70%), and the remaining data (30%) to test the proposed ­model33. It is important to note
that the folds were randomly assembled using a shuffle-split methodology in its stratified version to guarantee
a proportional distribution in each s­ et34. Each classification approach was evaluated using logistic regression,
support vector machine, decision tree, random forest, multilayer perceptron, and TabNet. For assessing the
performance of each model, diagnostic measures such as sensitivity, specificity, accuracy, and precision are used.

Scientific Reports | (2022) 12:22649 | https://doi.org/10.1038/s41598-022-26336-2 4

Vol:.(1234567890)
www.nature.com/scientificreports/

Additionally, the area under the curve (AUC) of the receiver operating characteristics (ROC) was determined
­ odel48,49.
for each m
TP = true positive
TN = true negative
FP = false positive
FN = false negative
TP + TN
Accuracy = (1)
TP + TN + FP + FN

TN
Specificity = (2)
TN + FP

TP
Sensitivity = (3)
TN + FN

TP
Precision = (4)
TP + FP

Results
The data analyses indicated that 70% of the participants presented smartphone dependence. Initially, a prelimi-
nary analysis was conducted to identify variables with a more prominent relationship with the response variable.
Hence, the chi-square test for categorical variables and the odds ratio (OR) for dichotomous qualitative vari-
ables were applied. According to this analysis, the following variables were identified as related to smartphone
dependency in students: (i) Academic program; (ii) school; (iii) marital status; (iv) socioeconomic status; (v) Is
it possible to express oneself in the family? (vi) May the student be identified as not having a smartphone? (vii)
Arguments about spending much time with a smartphone; (viii) residence area; (ix) the type of access to the
network; (x) most used space; (xi) time of acquisition; (xii) average use time per day; (xiii) The posture you use
when interacting with the phone: sitting on the floor, lying on the side, lying on the back; (xiv) the amount of
time with body discomfort; and (xv) duration of each episode of wrist discomfort.
Table 1 shows the discriminated results for each variable. The risk factors are presented, and the variables and
their corresponding sub-categories are indicated. The frequency and percentage of students classified as having
dependency (cases) are also shown.
The responses associated with the identification of musculoskeletal discomforts indicated the wrist as the body
area with the highest risk factor (OR = 1.93, CI 95% = 1.47–2.54)). The neck, shoulder, back, and elbow regions
showed similar risk factors (OR = 1.42, 1.62, 1.88, and 1.89, respectively). The results are summarized in Table 2.
Table 3 shows the results for the discomfort in the previous 12 months according to smartphone dependency.
The results found the elbow (OR = 1.45) and shoulder (OR = 1.69) with the highest risk for discomfort, while the
back area with the lowest.

Machine learning based prediction system. All the significant variables from the different models per-
formed were included. A total of 31 variables related to smartphone dependence were identified. Table 4 shows
the results for all classifiers in which the accuracy, specificity, sensitivity, precision, and area of the ROC curve
of five diagnostic measures are presented. For the random forest, n_e is the number of estimators or trees in the
forest. For SVM C is the regularization parameter, γ is the kernel coefficient for both polynomial and radial basis
functions, and d is the degree of the polynomial kernel. In the case of the multilayer perceptron, we use a DNN
with six hidden layers with 50, 50, 50, 20, 20, and 10 neurons using relu activation functions connected to an
output layer with one single neuron using a sigmoidal activation function.
Differences were observed among the methods under study, considering the metrics to assess their perfor-
mance. For example, the TabNet model and the decision tree have the lowest overall rates; however, the decision
tree presented the highest specificity value, above 50%. In contrast, for logistic regression, random forest, and
both support vector machine approaches, better sensitivity rates were achieved (above 91%), but specificity
was significantly reduced (below 41%). As expected, neither the TabNet model nor the multilayer perceptron
performed better than the classical approaches.
To perform a global evaluation for each classifier, the AUC of the ROC curve was determined (Fig. 3). It was
observed that the classifier with the lowest performance was the TabNet model, followed by the decision tree.
On the other hand, the similar AUC of the five models (AUC ~ 0.72) makes it challenging to determine which
approach offers the best performance. Overall, considering the model’s simplicity, the number of parameters,
and the performance achieved by the logistic regression classification approach, such an approach is a suitable
predictive model for the task at hand. However, the SVM or random forest classifiers constitute attractive alter-
natives, given that these approaches have comparable high performances.
It is worth mentioning that a highly sensitive system can correctly identify participants where smartphone
dependency is suspected. Hence, self-reported information gathered through standardized questionnaires con-
tains discriminative features to train predictive models. However, the perceptual and subjective nature of the
information can also hamper the potential of predictive models. This may be the reason for achieving low
specificity. In the early stages of a diagnosis, it is helpful to include the assessment of multiple professionals to

Scientific Reports | (2022) 12:22649 | https://doi.org/10.1038/s41598-022-26336-2 5

Vol.:(0123456789)
 www.nature.com/scientificreports/

Risk factors Variables Cases Control p-value

Academic program
Administration 42 (66.7) 21(33.3) 0.000
Bioengineer 36(60.0) 24(40.0)
Accountancy 27(42.9) 36(57.1)
Law 201(68.1) 94(31.9)
Economics 29(46.0) 34(54.0)
Nursing 41(59.4) 28(40.6)
Finance 21(33.3) 42(66.7)
Physiotherapy 44(89.8) 5(10.2)
Speech Therapy 29(100.0) 0(0.0)
Industrial Engineering 60(100.0) 0(0.0)
Electronic Engineering 37(61.7) 23(38.3)
Engineering Energy 60(100.0) 0(0.0)
System Engineering 60(100.0) 0(0.0)
Surgical Instrumentation 21(60.0) 14(40.0)
Medicine 47(83.9) 9(16.1)
Marketing 31(64.6) 17(35.4)
Dentistry 37(90.2) 4(9.8)
Psychology 20(74.1) 7(25.9)
Sociodemographic Respiratory Therapy 20(74.1) 7(25.9)
Faculty
Health 259(77.8) 74(22.2) 0.000
Economic Sciences 150(50.0) 150(50.0)
Engineering 253(84.3) 47(15.7)
Law 201(68.1) 94(31.9)
Civil status
Single 737(72.0) 287(28.0) 0.021
Married 48(66.7) 24(33.3)
Separated 8(53.3) 7(46.7)
Divorced 3(100.0) 0(0.0)
Widowed 1(100.0) 0(0.0)
Cohabitation 66(58.4) 47(41.6)
Socioeconomic stratification (Lower class: 1—Upper class: 6)
1 60(63.2) 35(36.8) 0.000
2 167(55.5) 134(44.5)
3 367(74.1) 128(25.9)
4 159(78.3) 44(21.7)
5 96(82.8) 20(17.2)
6 14(77.8) 4(22.2)
Who do you live with?
Both parents 255(67.3) 124(32.7) 0.001
Mother/father 153(72.2) 59(27.8)
Another family member 244(78.5) 67(21.5)
Friend 75(72.1) 29(27.9)
Partner 75(61.0) 48(39.0)
Alone 61(61.6) 38(38.4)
Can you be discriminated for not owning a smartphone?
Interpersonal No 537((66.5) 270(33.5) 0.000
A little 139(74.7) 47(25.3)
Some 112(81.2) 26(18.8)
A lot 61(82.4) 13(17.6)
Extremely 14(60.9) 9(39.1)
Arguments for spending too much time on the smartphone
No 496(65.0) 267(35.0) 0.000
A little 158(78.6) 43(21.4)
Some 110(76.9) 33(23.1)
Continued

Scientific Reports | (2022) 12:22649 | https://doi.org/10.1038/s41598-022-26336-2 6

Vol:.(1234567890)
www.nature.com/scientificreports/

Risk factors Variables Cases Control p-value

A lot 74(80.4) 18(19.6)
Extreme 25(86.2) 4(13.8)
Are there any rules for the smartphone use at home?
No 243(58.1) 175(41.9) 0.000
A little 414(74.3) 143(25.7)
Some 102(79.1) 27(20.9)
A lot 74(82.2) 16(17.8)
Extreme 30(88.2) 4(11.8)
Type of internet connection
Mobile data 432(65.3) 230(34.7) 0.000
Wifi 422(76.0) 133(24.0)
Both 9(81.8) 2(18.2)
Context sensitive
Places of more smartphone use
Home 430(66.3) 219(33.7) 0.004
University 379(74.3) 131(25.7)
Shopping malls 54(78.3) 15(21.7)
Time since you acquired your first cell phone
Less than 6 months 42(80.8) 10(19.2) 0.000
From 6 months to 1 year 250(75.1) 83(24.9)
1–3 years 87(60.4) 57(39.6)
3–6 years 140(59.6) 95(40.4)
Related to the mobile device More than 6 years 343(74.1) 120(25.9)
Average time of use per day
Less than an hour 35(61.4) 22(38.6) 0.000
1–3 h 173(58.4) 123(41.6)
3–6 h 205(65.9) 106(34.1)
More than 6 h 449(79.8) 114(20.2)
Do you use your smartphone sitting on the floor?
Less than an hour 496(66.3) 252(33.7) 0.001
1–3 h 339(77.2) 100(22.8)
3–6 h 24(72.7) 9(27.3)
More than 6 h 4(50.0) 4(50.0)
Do you use your smartphone lying on one side?
Less than an hour 495(64.2) 276(35.8) 0.000
1–3 h 311(80.2) 77(19.8)
3–6 h 46(86.8) 7(13.2)
More than 6 h 11(68.8) 5(31.3)
Do you use your smartphone lying on your back?
Less than an hour 394(63.4) 227(36.6) 0.000
1–3 h 371(77.8) 106(22.2)
Physical load 3–6 h 75(78.1) 21(21.9)
More than 6 h 23(67.6) 11(32.4)
How long have you had any discomfort?
Less than a month 532(65.8) 277(34.2) 0.000
Between 2 and 3 months 133(79.6) 34(20.4)
Between 4 and 6 months 81(78.6) 22(21.4)
Between 7 and 9 months 30(83.3) 6(16.7)
Between 10 and 12 months 87(77.0) 26(23.0)
Duration of each wrist episode
Less than an hour 730(68.8) 331(31.2) 0.014
Between 1 and 24 h 80(82.5) 17(17.5)
Between 1 and 7 days 30(85.7) 5(14.3)
Between 1 and 4 weeks 9(69.2) 4(30.8)
More than a month 14(63.6) 8(36.4)

Table 1.  Qualitative variables: university students with and without smartphone dependency.

Scientific Reports | (2022) 12:22649 | https://doi.org/10.1038/s41598-022-26336-2 7

Vol.:(0123456789)
 www.nature.com/scientificreports/

Significative variables Cases n Controls n OR (CR95%)

Do you experience any discomfort in
Neck
Yes 502 180 1.429 (1.118–1.827)
No 361 185
Shoulder
Yes 305 92 1.622 (1.233–2.134)
No 558 273
Back
Yes 347 96 1.884 (1.439–2.468
No 516 269
Elbow
Yes 163 40 1.892 (1.307–2.739)
No 700 325
Wrist
Yes 341 92 1.938 (1.476–2.547)
No 522 273

Table 2.  Bivariate analysis. Discomfort in undergraduate students with and without smartphone dependency.

Significant variables Cases n Controls n OR (CR95%)

Discomfort during the last 12 months
Neck
Yes 399 127 1.611 (1.251–2.077)
No 464 238
Shoulder
Yes 238 67 1.694 (1.250–2.296)
No 625 298
Back
Yes 309 101 1.458 (1.115–1.906)
No 554 264
Elbow
Yes 149 40 1.696 (1.168–2.462)
No 714 325
Wrist
Yes 580 280 1.607 (1.213–2.129)
No 283 85

Table 3.  Bivariate analysis. University students experiencing discomfort in the last 12 months with and
without dependency on the smartphone.

Classifier Accuracy Sensitivity Specificity Precision AUC​

Decision tree 70.5 ± 2.0 78.9 ± 2.9 50.1 ± 2.7 79.2 ± 0.9 0.639 ± 0.01
Logistic regression 76.4 ± 1.5 91.1 ± 1.4 39.7 ± 3.4 78.4 ± 1.0 0.721 ± 0.02
Random forest
n_e = 20 76.6 ± 1.3 91.4 ± 0.7 40.6 ± 4.7 78.8 ± 1.2 0.731 ± 0.03
SVM poly
C = 1, γ = 5, d = 2 76.4 ± 1.2 93.1 ± 0.5 38.8 ± 3.9 77.9 ± 1.0 0.729 ± 0.01
SVM rbf
C = 10, γ = 3 77.2 ± 1.3 92.2 ± 0.9 39.7 ± 3.6 78.7 ± 1.0 0.729 ± 0.01
Multilayer perceptron 73.6 ± 2.3 94.7 ± 5.2 22.7 ± 8.0 75.04 ± 3.1 0.720 ± 0.02
TabNet 67.4 ± 1.3 86.4 ± 5.3 21.3 ± 8.0 72.8 ± 2.2 0.611 ± 0.02

Table 4.  Predictive Performance of the models.

Scientific Reports | (2022) 12:22649 | https://doi.org/10.1038/s41598-022-26336-2 8

Vol:.(1234567890)
www.nature.com/scientificreports/

Figure 3.  Receiver Operating Characteristic Curve (ROC curve) for all classification systems.

reject or confirm dependency. It would be necessary to include objective measurements to improve the system’s
prediction capabilities in future works.

Discussion
The classification models yielded satisfactory smartphone dependency predictions. Likewise, a relationship
between university students with and without smartphone dependency and multiple risk factors was found,
which should motivate establishing high-priority preventive actions. The results indicate that student enrollment
was significantly correlated with smartphone dependency, and an important prevalence was identified, especially
among engineering (84.3%), health (77.8%), law (68.1%), and economic sciences students (50.0%). Similar results
have been reported, although the highest dependency rate was identified in the medical academic p ­ rogram50.
Marital status (72%) was related to smartphone dependency, which is in line with previous ­studies51–54. How-
ever, being single cannot be included as a risk factor. It can be hypothesized that being involved in a romantic
relationship may reduce smartphone users’ time. Nevertheless, this is a factor that requires additional analysis.
The high-income socioeconomic stratification was also meaningful for smartphone users, as it facilitates
access to new technology, gadgets, pay-per-use applications, etc.52,53,55. Our data corroborate previous reports
that high family income is more likely to develop smartphone d ­ ependency17. In addition, young students may
feel discriminated against for not having a cell phone and not satisfying a communication prerequisite to belong
to a particular social group. Cellphone ownership is highly relevant in today’s society, where social networks are
at the core of personal and social relationships. It might have also accelerated the first cell phone acquisition, as
dependency is more pronounced (74.1%) in those who used it for the first time more than six years ago. Others
have also reported a similar dependence (77.5%)56. Further investigations are necessary to explore the causes of
its acquisition and excessive use.
Adverse domestic situations can also be a predictor related to smartphone ­dependency57. It has been shown
that students who reported domestic conflict or adversities (e.g., parent alcohol and drug use, mental health,
incarceration, suicide, intimate partner violence, separation/divorce, and homelessness) are also more likely to
have problematic/addictive smartphone use. A strong association between household dysfunction and psycho-
logical and behavioral health issues was reported. However, this association requires further research to explain
this association further.
A significant difference was found between those who access the internet by paying for data packages and
illimited access. Having internet access with no limitations facilitates surfing the internet, making video calls,
gaming, sending text messages anytime, etc. The result showed that having a data plan increases the probability
of developing smartphone dependency by 50%, as the number of hours is also likely to be greater than others
with more limited access.
The amount of time spent using cell phones is also a strong indicator of dependence. In this study, the par-
ticipants with smartphone addiction reported periods of usage longer than 6 h. It has been reported that the
likelihood of developing smartphone addiction is proportional to the number of hours of use (3–4 h: OR = 5.79;

Scientific Reports | (2022) 12:22649 | https://doi.org/10.1038/s41598-022-26336-2 9

Vol.:(0123456789)
 www.nature.com/scientificreports/

5–6 h: OR = 10.78)17. Indeed, the risk almost doubled for those using the device for 5–6 h compared to those
with fewer hours (i.e., 3–4 h per day)58.
Sitting was the most predominant posture while using a smartphone (66.3%), despite the short period it was
sustained (i.e., less than an hour). It may explain why the wrist and the neck areas showed the largest prevalence
(OR. 1.93 and 1.42, respectively). It has been reported that office workers with excessive smartphone use are
approximately six times more likely to have neck p ­ ain59. It reinforces that smartphone dependency is highly asso-
ciated with neck pain. Nonetheless, the prevalence was lower than reported by Derakhshanrad and ­colleagues59.
There can be multiple reasons for this difference, including the location, target population, and instrument
applied. In this study, university students with smartphone dependency reported discomfort or musculoskel-
etal symptoms for less than one month (n = 532, 65.8%). Hence, the presence and duration of musculoskeletal
discomfort in the last 12 months contribute to the prediction of smartphone dependency.
The variables used in the model show that sociodemographic characteristics determine a level of smartphone
dependency. However, the age and gender variables must be ruled out. For instance, Nikhita and collaboratives
reported that female users had a higher p ­ revalence60, while Matoza-Báez and c­ olleagues61 showed a higher
prevalence of male users. The age of more than 90% of our participants ranged between 18 and 32, and a more
comprehensive range is required to discard age as an explanatory factor.
This is a cross-sectional analysis, and longitudinal studies are required before establishing a cause-effect
relationship. The inclusion and analysis of variables related to academic performance, mental health, and sleep
disorders may be considered for future studies. Although the number of participants included in the present
study is not trivial, the amount of data affects the training process of the models, and it remains an open problem
to address in future studies, including deep learning techniques. Once risk factors and variables related to smart-
phone dependency are identified, it is essential to mention that strategies to reduce these risks and adverse effects
are paramount for society. It should involve a multidisciplinary approach. Campaigns to raise awareness about the
negative consequences of physical and mental health and how to address these problems or where people can find
professional advice may constitute a relevant strategy to counteract the adverse impacts of overusing technology.

Conclusions
Smartphones are ubiquitous and part of our daily life. The adverse effects of excessive use of smartphones are con-
cerning, as dependency is becoming a public health problem requiring special attention due to its consequences
on physical and mental health. Machine learning helped identify several dependency factors while using a large
number of independent variables. The support vector machine and random forest presented the highest predic-
tion precision for smartphone dependency, obtained through the stratified-k-fold cross-validation technique.
The variable selection is more critical than the choice of a specific model itself.
This study shows that self-reported information obtained using standardized questionnaires contains dis-
criminative information to predict smartphone dependency using data-driven models. These results open doors
for future studies aiming to reduce the adverse effects of overusing mobile devices. In many cases, a correct
assessment of dependency levels and the corrective actions to be taken require the intervention of experienced
health professionals. This is not always possible in the early stages, while late interventions can be costly and
may bring adverse effects. Further research in this area is still required, as the perceptual and subjective nature
of the information may hamper the potential of predictive models. For future work, it is necessary to introduce
objective measures. Using electronics to measure physiological activity can add important information instead
of subjective self-reported variables.

Data availability
Datasets analyzed during the current research are available to the corresponding author upon reasonable request.

Received: 11 August 2022; Accepted: 13 December 2022

References
1. Demir, K. & Akpinat, E. The effect of mobile learning applications on students’ academic achievement and attitudes toward mobile
learning. Malays. Online J. Educ. Technol. 6, 48–59 (2018).
2. Abadiyan, F., Hadadnezhad, M., Khosrokiani, Z., Letafatkar, A. & Akhshik, H. Adding a smartphone app to global postural
re-education to improve neck pain, posture, quality of life, and endurance in people with nonspecific neck pain: A randomized
controlled trial. Trials 22, 274 (2021).
3. Osorio-Molina, C. et al. Smartphone addiction, risk factors and its adverse effects in nursing students: A systematic review and
meta-analysis. Nurse Educ. Today 98, 104741 (2021).
4. Osailan, A. The relationship between smartphone usage duration (using smartphone’s ability to monitor screen time) with hand-
grip and pinch-grip strength among young people: An observational study. BMC Musculoskelet. Disord. 22, 186 (2021).
5. Hitti, E., Hadid, D., Melki, J., Kaddoura, R. & Alameddine, M. Mobile device use among emergency department healthcare profes-
sionals: prevalence, utilization and attitudes. Sci. Rep. 11, 1917 (2021).
6. Shaygan, M. & Jaberi, A. The effect of a smartphone-based pain management application on pain intensity and quality of life in
adolescents with chronic pain. Sci. Rep. 11, 6588 (2021).
7. Sohn, S. Y., Krasnoff, L., Rees, P., Kalk, N. J. & Carter, B. The association between smartphone addiction and sleep: A UK cross-
sectional study of young adults. Front. Psych. 12, 629407 (2021).
8. Wilkerson, G. B. et al. Wellness survey responses and smartphone app response efficiency: Associations with remote history of
sport-related concussion. Percept. Mot. Skills 128, 714–730 (2021).
9. Thornton, L. et al. A multiple health behavior change, self-monitoring mobile app for adolescents: Development and usability
study of the Health4Life App. JMIR Format. Res. 5, e25513 (2021).
10. Joo, E., Kononova, A., Kanthawala, S., Peng, W. & Cotten, S. Smartphone users’ persuasion knowledge in the context of consumer
mHealth apps: Qualitative study. JMIR Mhealth Uhealth 9, e16518 (2021).

Scientific Reports | (2022) 12:22649 | https://doi.org/10.1038/s41598-022-26336-2 10

Vol:.(1234567890)
www.nature.com/scientificreports/

11. Mergany, N. N., Dafalla, A. E. & Awooda, E. Effect of mobile learning on academic achievement and attitude of Sudanese dental
students: A preliminary study. BMC Med. Educ. 21, 121 (2021).
12. Maharjan, S. M. et al. Passive sensing on mobile devices to improve mental health services with adolescent and young mothers in
low-resource settings: the role of families in feasibility and acceptability. BMC Med. Inform. Decis. Mak. 21, 117 (2021).
13. Kwon, S. E., Kim, Y. T., Suh, H. Won & Lee, H. Identifying the mobile application repertoire based on weighted formal concept
analysis. Expert Syst. Appl. 173, 114678 (2021).
14. Konok, V. et al. Mobile use induces local attentional precedence and is associated with limited socio-cognitive skills in preschool-
ers. Comput. Hum. Behav. 120, 106758 (2021).
15. Ellahi, A., Zaka, B. & Sultan, F. A study of supplementing conventional business education with digital games. J. Educ. Technol.
Soc. 20, 195–206 (2017).
16. Giraldo-Jiménez, C. F. et al. Dependence on mobile devices among health science university students: A cross-sectional analytical
study. World Trans. Eng. Technol. Educ. 20, 45–51 (2022).
17. de Freitas, B. H. B. M., Gaíva, M. A. M., Bernardino, F. B. S. & Diogo, P. M. J. Smartphone addiction in adolescents, part 2: Scoping
review-prevalence and associated factors. Trends Psychol. 29, 12–30 (2021).
18. Warsaw, R. E. et al. Mobile technology use and its association with executive functioning in healthy young adults: A systematic
review. Front. Psychol. 12, 643542 (2021).
19. Choi, S., Kim, M., Kim, E. & Shin, G. Changes in low back muscle activity and spine kinematics in response to smartphone use
during walking. Spine 46, E426–E432 (2021).
20. Yue, H. et al. The relationships between negative emotions and latent classes of smartphone addiction. PLoS ONE 16, e0248555
(2021).
21. Annoni, A. M., Petrocchi, S., Camerini, A.-L. & Marciano, L. The Relationship between social anxiety, smartphone use, dispositional
trust, and problematic smartphone use: A moderated mediation model. Int. J. Environ. Res. Public Health 18, 2452 (2021).
22. Conlin, M.-C. & Sillence, E. Exploring British adolescents’ views and experiences of problematic smartphone use and smartphone
etiquette. J. Gambl. Issues 46, 279–301 (2021).
23. Chen, P. L. & Pai, C. W. Pedestrian smartphone overuse and inattentional blindness: An observational study in Taipei, Taiwan.
BMC Public Health 18, 1342 (2018).
24. Gamero, K. et al. Estandarización del Test de Dependencia al Celular para Estudiantes Universitarios de Arequipa. Persona 19,
179–200 (2016).
25. Goodfellow, I., Bengio, Y. & Courville, A. Deep Learning (The MIT Press, 2016).
26. Benois-Pineau, J. & Zemmari, A. Multi-faceted Deep Learning. Models and Data. (Springer Nature, Switzerland, 2021).
27. Lu, Y., Lu, G., Li, J., Zhang, Z. & Xu, Y. Fully shared convolutional neural networks. Neural Comput. Appl. 33, 8635–8648 (2021).
28. Sarker, I. H., Kayes, A. S. M. & Watters, P. Effectiveness analysis of machine learning classification models for predicting personal-
ized context-aware smartphone usage. J. Big Data 6, 57 (2019).
29. U.S. Department of Health & Human Services. Centers for Disease Control and Prevention. Epi InfoTM (2021). Disponible en:
https://​www.​cdc.​gov/​epiin​fo/​index.​html.
30. Chóliz, M. Mobile-phone addiction in adolescence: The test of mobile phone dependence (TMD). Progr. Health Sci. 2, 33–44
(2012).
31. Chóliz, M. et al. Development of a brief multicultural version of the test of mobile phone dependence (TMDbrief) questionnaire.
Front. Psychol. 7, 650 (2016).
32. Dimate, A. E. & Rocha, A. I. Percepción de desórdenes musculoesqueleticos y aplicación del método RULA en diferentes sectores
productivos. Revista de la Universidad Industrial de Santander. 49, 57–74 (2017).
33. Bishop, C. M. Pattern recognition and machine learning (Springer, 2006).
34. Duda, R. O., Hart, P. E. & Stork, D. G. Pattern classification (Jhon Wiley & Sons Inc, 2001).
35. Shafiq, M., Tian, Z., Bashir, A. K., Du, X. & Guizani, M. CorrAUC: A malicious Bot-IoT traffic detection method in IoT network
using machine-learning techniques. IEEE Internet Things J. 8, 3242–3254 (2021).
36. Shafiq, M., Tian, Z., Sun, Y., Du, X. & Guizani, M. Selection of effective machine learning algorithm and Bot-IoT attacks traffic
identification for internet of things in smart city. Futur. Gener. Comput. Syst. 107, 433–442 (2020).
37. Shafiq, M., Tian, Z., Bashir, A. K., Du, X. & Guizani, M. IoT malicious traffic identification using wrapper-based feature selection
mechanisms. Comput. Secur. 94, 101863 (2020).
38. Stergiou, C. L., Plageras, A. P., Psannis, K. E. & Gupta, B. B. Secure Machine Learning Scenario from Big Data in Cloud Computing
via Internet of Things Network. en Handbook of Computer Networks and Cyber Security: Principles and Paradigms (eds. Gupta,
B. B., Perez, G. M., Agrawal, D. P. & Gupta, D.) 525–554 (Springer Nature Switzerland AG, 2020). https://​doi.​org/​10.​1007/​978-3-​
030-​22277-2_​21
39. Cui, Z. et al. Personalized Recommendation System Based on Collaborative Filtering for IoT Scenarios. IEEE Trans. Serv. Comput.
13, 685–695 (2020).
40. Chen, T. Y., Chen, Y. M. & Tsai, M. C. A status property classifier of social media user’s personality for customer-oriented intelligent
marketing systems: Intelligent-based marketing activities. Int. J. Semant. Web Inf. Syst. 16, 25–46 (2020).
41. Birjali, M., Kasri, M. & Beni-Hssane, A. A comprehensive survey on sentiment analysis: Approaches, challenges and trends.
Knowl.-Based Syst. 226, 107134 (2021).
42. Singh, S. K. & Sachan, M. K. Classification of code-mixed bilingual phonetic text using sentiment analysis. Int. J. Semant. Web Inf.
Syst. 17, 59–78 (2021).
43. Tolles, J. & Meurer, W. J. Logistic regression relating patient characteristics to outcomes. JAMA Guide Stat. Methods 316, 533–534
(2016).
44. Vapnik, V. Statistical learning theory (John Wiley and Sons Inc., 1998).
45. Priyam, A., Gupta, R., Rathee, A. & Srivastava, S. Comparative analysis of decision tree classification algorithms. Int. J. Curr. Eng.
Technol. 3, 334–337 (2013).
46. Breiman, L. Using iterated bagging to debias regressions. Mach. Learn. 45, 261–277 (2001).
47. Arik, S. Ö. & Pfister, T. Tabnet: Attentive interpretable tabular learning. Proc. AAAI Conf. Artif. Intell. 35, 6679–6687 (2021).
48. Fletcher, R. H. & Fletcher, S. W. Clinical epidemiology: The essentials (Lippincott Williams & Wilkins, 2019).
49. Kumar, R. & Indrayan, A. Receiver operating characteristic (ROC) curve for medical researchers. Indian Pediatr. 48, 277–287
(2011).
50. Laurence, P. G., Busin, Y., da Cunha Lima, H. S. & Macedo, E. C. Predictors of problematic smartphone use among university
students. Psicologia: Reflexão e Crítica 33, 1–13 (2020).
51. Ivanova, A. et al. Mobile phone addiction, phubbing, and depression among men and women: A moderated mediation analysis.
Psychiatr. Q. 91, 655–668 (2020).
52. Zhao, S. Z. et al. Association of mobile instant messaging chat group participation with family functioning and well-being:
Population-based cross-sectional study. J. Med. Internet Res. 23, e18876 (2021).
53. Xie, Y. J. et al. Relationships between the usage of televisions, computers, and mobile phones and the quality of sleep in a Chinese
population: Community-based cross-sectional study. J. Med. Internet Res. 22, e18095 (2020).
54. Jeong, Y.-W., Han, Y.-R., Kim, S.-K. & Jeong, H.-S. The frequency of impairments in everyday activities due to the overuse of the
internet, gaming, or smartphone, and its relationship to health-related quality of life in Korea. BMC Public Health 20, 954 (2020).

Scientific Reports | (2022) 12:22649 | https://doi.org/10.1038/s41598-022-26336-2 11

Vol.:(0123456789)
 www.nature.com/scientificreports/

55. Romero-Rodríguez, J.-M., Aznar-Díaz, I., Marín-Marín, J.-A., Soler-Costa, R. & Rodríguez-Jiménez, C. Impact of problematic
smartphone use and Instagram use intensity on self-esteem with university students from physical education. Int. J. Environ. Res.
Public Health 17, 4336 (2020).
56. Mustafaoglu, R., Yasaci, Z., Zirek, E., Griffiths, M. D. & Ozdincler, A. R. The relationship between smartphone addiction and
musculoskeletal pain prevalence among young population: A cross-sectional study. The Korean J. Pain 34, 72–81 (2021).
57. Forster, M. et al. Adverse childhood experiences and problematic smartphone use among college students: Findings from a pilot
study. Addict. Behav. 117, 106869 (2021).
58. Haug, S. et al. Smartphone use and smartphone addiction among young people in Switzerland. J. Behav. Addict. 4, 299–307 (2015).
59. Derakhshanrad, N., Yekaninejad, M. S., Mehrdad, R. & Saberi, H. Neck pain associated with smartphone overuse: cross-sectional
report of a cohort study among office workers. Eur. Spine J. 30, 461–467 (2021).
60. Nikhita, C. S., Jadhav, P. R. & Ajinkya, S. Prevalence of mobile phone dependence in secondary school adolescents. J. Clin. Diagn.
Res. 9, 6–9 (2015).
61. Matoza-Báez, C. M. & Carballo-Ramírez, M. S. Nivel de Nomofobia en Estudiantes de Medicina de Paraguay, Año 2015. Cimel
21, 28–30 (2016).

Author contributions
All authors contributed to the writing and design of the study to the best of their skills and knowledge. C.F.G.-
J. contributed to study design, data collection, and manuscript writing. J.G.-C. contributed to data processing
tasks such as data normalization, statistical analysis, and variable selection. M.O.S.-P. and J.J.V.-M. contributed
to training and validating machine learning models, consolidating results from the statistical analysis, and writ-
ing several sections of the manuscript. L.A.B.V. and A.L.F.R. contributed to the analysis and interpretation of
results coming from different stages of the study. They added central insights to enrich the final discussion of the
manuscript and made significant contributions to writing several sections of the manuscript.

Funding
This research has been funded by Universidad Santiago de Cali, Dirección General de Investigaciones under
call No. 01-2022.

Competing interests
The authors declare no competing interests.

Additional information
Correspondence and requests for materials should be addressed to M.S.-P.
Reprints and permissions information is available at www.nature.com/reprints.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and
institutional affiliations.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International
License, which permits use, sharing, adaptation, distribution and reproduction in any medium or
format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the
Creative Commons licence, and indicate if changes were made. The images or other third party material in this
article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the
material. If material is not included in the article’s Creative Commons licence and your intended use is not
permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from
the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/.

© The Author(s) 2022

Scientific Reports | (2022) 12:22649 | https://doi.org/10.1038/s41598-022-26336-2 12

Vol:.(1234567890)