Applied Ergonomics 59 (2017) 132e142

Contents lists available at ScienceDirect

Applied Ergonomics
journal homepage: www.elsevier.com/locate/apergo

Review article

Prevalence and risk factors associated with musculoskeletal

complaints among users of mobile handheld devices: A systematic
review
Yanfei Xie*, Grace Szeto, Jie Dai
Department of Rehabilitation Sciences, The Hong Kong Polytechnic University, Hong Kong, China

a r t i c l e i n f o a b s t r a c t

Article history: This systematic review aimed at evaluating the prevalence and risk factors for musculoskeletal com-
Received 2 November 2015 plaints associated with mobile handheld device use. Pubmed, Medline, Web of Science, CINAHL and
Received in revised form Embase were searched. The methodological quality of included studies was assessed. Strength of evi-
11 August 2016
dence for risk factors was determined based on study designs, methodological quality and consistency of
Accepted 26 August 2016
results. Five high-quality, eight acceptable-quality and two low-quality peer-reviewed articles were
included. This review demonstrates that the prevalence of musculoskeletal complaints among mobile
device users ranges from 1.0% to 67.8% and neck complaints have the highest prevalence rates ranging
Keywords:
Mobile handheld devices
from 17.3% to 67.8%. This study also finds some evidence for neck flexion, frequency of phone calls,
Musculoskeletal complaints texting and gaming in relation to musculoskeletal complaints among mobile device users. Inconclusive
Systematic review evidence is shown for other risk factors such as duration of use and human-device interaction techniques
due to inconsistent results or a limited number of studies.
© 2016 Elsevier Ltd. All rights reserved.

Contents

1. Introduction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
2. Methods . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
2.1. Search strategies . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
2.2. Eligibility criteria . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 133
2.3. Methodological quality assessment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
2.4. Strength of evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
2.5. Data extraction . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 134
3. Results . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
3.1. Study selection . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
3.2. Quality of reviewed articles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 135
3.3. Study characteristics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
3.4. Results of prevalence rates . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
3.5. Risk factors and strength of evidence . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 136
4. Discussion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
4.1. Methodological considerations . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 137
4.2. Prevalence of musculoskeletal complaints . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 138
4.3. Evidence on risk factors . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
4.3.1. Posture . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
4.3.2. Total time spent on mobile devices . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140
4.3.3. Tasks frequently performed and human-device interaction techniques . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140

* Corresponding author. ST816, Department of Rehabilitation Sciences, The Hong

Kong Polytechnic University, Hung Hom, Kowloon, Hong Kong, China.
E-mail address: yanfei.xie@connect.polyu.hk (Y. Xie).

http://dx.doi.org/10.1016/j.apergo.2016.08.020
0003-6870/© 2016 Elsevier Ltd. All rights reserved.
 Y. Xie et al. / Applied Ergonomics 59 (2017) 132e142 133

4.4. Limitations of this review . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 140

5. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141
References . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 141

1. Introduction which can in turn facilitate prevention strategies. The purpose of

this systematic review is to (a) gain insight into the prevalence of
Mobile handheld devices, including personal digital assistants, musculoskeletal complaints among users of mobile devices; (b)
keypad phones, touchscreen smartphones and tablet computers, evaluate the existing evidence on risk factors for musculoskeletal
are information technology (IT) instruments commonly used for complaints associated with the use of these devices. For this pur-
communication and entertainment. Nowadays, nearly everyone, pose, mobile handheld devices studied include all items such as
from children to adults of all ages, owns at least one type of mobile different types of keypad mobile phones, smartphones and tablet
handheld device because of the low cost and convenience. Multiple computers with touchscreens as well as handheld game devices.
usability options and fascinating applications of mobile handheld
devices such as smartphones and tablet computers, together with
easy access to internet encourage users especially youngsters to 2. Methods
spend a large amount of time with their mobile devices. A survey in
Canada among 137 university students, staff and faculty showed 2.1. Search strategies
that the participants spent 4.65 hours daily on mobile handheld
devices (Berolo et al., 2011). The intensive use of mobile devices has Electronic database search and a reference search were the
been reported to be associated with non-specific symptoms of ill- major strategies to identify published studies that reported the
health and musculoskeletal complaints (Berolo et al., 2011; Chu prevalence and/or risk factors associated with musculoskeletal
et al., 2011). The association between electromagnetic field expo- complaints among users of mobile handheld devices. Electronic
sures from mobile phones and non-specific symptoms of ill-health databases of Pubmed, Medline (1946 þ via OvidSP), Web of Science,
such as headache, fatigue, depression, sleep disturbance and CINAHL (1982þ) and Embase (1980þ) were searched. Relevant
earache have been extensively investigated, according to several studies were identified based on a combination of three groups of
systematic reviews (Augner et al., 2012; Ro €o
€ sli et al., 2010; Valentini terms (MeSH and/or text word search terms). These three groups of
et al., 2007, 2010). However, there is a lack of systematic in- terms pertained to mobile handheld devices, musculoskeletal
vestigations of the prevalence and physical risk factors for complaints and prevalence/risk, respectively. The terms within
musculoskeletal complaints associated with the use of mobile each group were combined with “OR”, and three groups were
devices. linked with “AND”. Specific search terms used and full search
Musculoskeletal complaints affect a large number of people. syntax with truncation used in the Medline database is shown in
Prevalence is a useful indicator of the extent of musculoskeletal Table 1. Similar strategies were performed in other databases.
complaints in a population and risk factors for musculoskeletal Studies that were not related to the objective of this systematic
complaints serve as a basis for prevention and interventions review and did not meet the eligibility criteria were eliminated
(Walker, 2000). Several factors including excessive repetition, high through screening titles and abstracts of the articles. Regarding
physical and psychological demands, sustained awkward postures articles that could not be screened out through titles and abstracts,
and poor workstation designs have been identified as important a detailed review of full texts was performed to determine their
occupational factors associated with musculoskeletal complaints eligibility. Reference lists of papers which fulfilled the eligibility
(Co^ te
" et al., 2009; da Costa and Vieira, 2010; van Rijn et al., 2010). criteria were also systematically searched for additional relevant
Extensive research has been done in the past 2e3 decades on the studies. Furthermore, a forward search using Google Scholar was
relationship of computer use and musculoskeletal disorders, and conducted on May 4, 2016 to identify all related papers that made
these have also been summarized in a few systematic reviews reference to the eligible articles. The article screening and selection
(Mediouni et al., 2014; Thomsen et al., 2008; Waersted et al., 2010). were performed independently by two reviewers (YX and JD) and
These research studies were mainly conducted on the use of any disagreement was settled during a consensus discussion with
“desktop” computers, while more research has emerged about the the third reviewer (GS) (da Costa and Vieira, 2010; van der Windt
effects of using “laptop” computers in recent years (Asundi et al., et al., 2000).
2010; Gold et al., 2012; Malin " ska and Bugajska, 2010).
In the past 7e8 years, since the launch of the first generation of
“iPhone”, touchscreen handheld devices have dominated the IT
2.2. Eligibility criteria
market, and peoples' daily habits of using electronic devices have
been significantly affected by this technology. However, research on
The inclusion criteria for paper selection were studies that: (1)
risk factors and prevalence of musculoskeletal symptoms related to
had a primary purpose of examining the prevalence, or/and the risk
the use of these mobile touchscreen devices is only just emerging.
factors associated with musculoskeletal complaints among users of
Given the increasing reports on musculoskeletal complaints among
mobile handheld devices; (2) published in peer-reviewed English
users of mobile handheld devices (Ashurst et al., 2010; Fernandez-
language journals; (3) utilized cross-sectional, case-control, or
Guerrero, 2014; Williams and Kennedy, 2011), a comprehensive
prospective cohort study designs (4) reported results separately on
review is warranted to show a clear picture of the research about
prevalence of and/or risk factors for musculoskeletal complaints
the relationship of musculoskeletal complaints and mobile hand-
associated with the use of mobile devices. The exclusion criteria
held device use.
were as follows: (1) studies which investigated the use of mobile
Furthermore, a review would help lay a foundation for the
handheld devices such as phones as an intervention; (2) literature
development of clinical management and ergonomic guidelines,
reviews, conference papers and case reports.
134 Y. Xie et al. / Applied Ergonomics 59 (2017) 132e142

Table 1
The search strategy performed on April 12, 2016 in Medline (1946 þ via OvidSP) to identify potential articles for screening. Specific search terms, search fields including
Keyword Heading Word (kf), MeSH Subject Heading (sh) and Text Word (text), and combinations of terms are presented.

1. “cell phone*”. kf, sh, tw.

2. “mobile phone*”. kf, tw
3. “mobile device*”. kf, tw.
4. “smartphone*”. kf, sh, tw.
5. “smart phone*”. kf, tw.
6. “touchscreen phone*”. kf, tw.
7. “hand-held device*”. kf, tw.
8. “tablet*”. kf, tw.
9. “tablet computer*”. kf, tw.
10. 1 or 2 or 3 or 4 or 5 or 6 or 7 or 8 or 9
11. “musculoskeletal disease*”. kf, sh, tw.
12. “musculoskeletal disorder*”. kf, tw.
13. “musculoskeletal symptom*”. kf, tw.
14. musculoskeletal discomfort. kf, tw.
15. “musculoskeletal complaint*”. kf, tw.
16. “cumulative trauma disorder*”. kf, sh, tw.
17. pain. kf, sh, tw.
18. 11 or 12 or 13 or 14 or 15 or 16 or 17
19. prevalence. kf, sh, tw.
20. risk. kf, sh, tw.
21. 19 or 20
22. 10 and 18 and 21
23. limit 22 to English language

Note: numbers 1e9, 11e17, 19e20 are the specific search terms; numbers 10, 18, 21 and 22 show how specific search terms were combined; number 23 is the limit for search
results; *represent truncation.

2.3. Methodological quality assessment 2.4. Strength of evidence

Three quality assessment tools were employed to evaluate the
methodological quality of cross-sectional, case-control and pro-
spective cohort studies, respectively. The methodological quality of
cross-sectional studies was assessed using a risk of bias tool which
was developed for evaluating the risk of bias in prevalence studies
(Hoy et al., 2012). The tool is comprised of four items (1e4) that
assess the external validity of a study, six items (5e10) that evaluate
the internal validity and an assessment of overall methodological
quality (Hoy et al., 2012). Inter-rater agreement was demonstrated
to be 93% with a Kappa value of 0.83 (95% CI, 0.78e0.88) for the 10
individual items on the risk of bias tool while 72% with a weighted
Kappa value of 0.48 (95% CI, 0.31e0.64) for the assessment of
overall methodological quality (Hoy et al., 2012). Two Scottish
Intercollegiate Guidelines Network (SIGN) checklists (Scottish
Intercollegiate Guideline Network, 2014) were adopted to eval-
uate the methodological quality of case-control and prospective
cohort studies, respectively. SIGN checklists cover four aspects of
bias, including selection and measurement bias as well as bias
related to confounding factors and statistical analysis. While there
are no standard checklists for assessing the methodological quality
of case-control and prospective cohort studies, SIGN checklists have
been reported to be the most appropriate, valid and useful among
currently available tools (Bai et al., 2012; Perestelo-Pe"rez, 2013).
The overall methodological quality of each included study was
rated as being either high quality (þþ) (low risk of bias), acceptable
(þ) (moderate risk of bias) or low quality (") (high risk of bias). The
overall methodological quality was rated based on raters' judgment
in the response that raters had given to each individual item in the
quality assessment tools (Hoy et al., 2012; Scottish Intercollegiate
Guideline Network, 2014). This method is consistent with Grades
of Recommendation, Assessment, Development and Evaluation
(GRADE) (Terracciano et al., 2010) and Cochrane approaches
(Higgins and Green, 2011). Methodological quality appraisal of each
included study was carried out by two reviewers (YX and JD)
independently; disagreement was identified and resolved during a
consensus meeting with the third reviewer (GS) (da Costa and
Vieira, 2010; van der Windt et al., 2000).
The method to establish the strength of evidence for potential
risk factors associated with musculoskeletal complaints among
users of mobile handheld devices was based on the methods pro-
posed by van der Windt et al. (2000) and Arie €ns et al. (2000). The
strength of evidence was classified as follows: (1) strong evidence:
consistent results in multiple prospective cohort or case-control
studies of high quality; (2) moderate evidence: consistent results
in multiple prospective cohort or case-control studies, among
which there was only one high-quality study; (3) some evidence:
results of one prospective cohort or consistent results in multiple
case-control studies regardless of quality, or consistent results in
multiple cross-sectional studies among which there was at least
one high-quality study; (4) inconclusive evidence: other condi-
tions, for example, inconsistent results in multiple studies or
consistent results in multiple cross-sectional studies of low quality
or findings being reported in only one case-control study regardless
of quality. Consistent results were defined by the fact that at least
75% of the studies measuring the effect of a certain risk factor re-
ported a strong association with musculoskeletal complaints, and
the definition of a strong association was a high odds ratio (OR)/
relative risk (RR) (>2.0), statistically significant (p < 0.05) or an
established cause-effect relationship (van der Windt et al., 2000).
2.5. Data extraction
Information from eligible studies was extracted as follows:
– Study characteristics: authors, the published year, study design,
the country, response rates, ages of participants and sampling
frames.
– Prevalence of musculoskeletal complaints: types of handheld
devices used, operational definitions of musculoskeletal com-
plaints to identify cases and results of prevalence rates.
– Risk factors: relevant results of risk factors assessed, including
the values of Pearson correlations (r), OR, 95% CI or p-value.
 Y. Xie et al. / Applied Ergonomics 59 (2017) 132e142 135

3. Results postures or human-device interaction techniques between symp-

tomatic and asymptomatic subjects when using a mobile phone. No
3.1. Study selection prospective cohort studies were identified.

The database searches and the application of the inclusion

criteria identified 15 full-text articles, reporting the results from 14
studies (Fig. 1). The articles from Gustafsson et al. (2010, 2011)
stemmed from the same study. Overall, there were 11 cross-
sectional survey studies and three case-control studies (4 full-text
articles) designed to compare the differences in muscle activity,
3.2. Quality of reviewed articles
Five papers were classified as being high quality (Berolo et al.,
2011; Hakala et al., 2006; Hegazy et al., 2016; Lui et al., 2011;
Shan et al., 2013), eight acceptable (Ali et al., 2014; Balakrishnan
et al., 2016; Eapen et al., 2010; Gustafsson et al., 2010, 2011; Kim,

Fig. 1. Flowchart of paper selection process.

136 Y. Xie et al. / Applied Ergonomics 59 (2017) 132e142

2015; Stalin et al., 2016; Xie et al., 2016) and two low quality
(Goldfinch et al., 2011; Kim and Kim, 2015). Selection bias was
shown in most of the studies (Tables 2a and 2b). Additionally, there
was a lack of an acceptable case definition used and information on
the reliability and validity of the instrument used in a majority of
cross-sectional studies (Table 2a). Confounding factors were not
considered in some case-control studies (Table 2b).
3.3. Study characteristics
The included studies were from different countries with varied
response rates from 51% to 100% (Table 3). The population in these
studies was mainly students (Ali et al., 2014; Balakrishnan et al.,
2016; Eapen et al., 2010; Hakala et al., 2006; Hegazy et al., 2016;
Kim and Kim, 2015; Lui et al., 2011; Shan et al., 2013) while some
studies also recruited working populations (Berolo et al., 2011;
Gustafsson et al., 2010, 2011; Goldfinch et al., 2011; Kim, 2015;
Stalin et al., 2016; Xie et al., 2016). Types of mobile devices in the
selected studies included tablets (Shan et al., 2013), handheld
electronic game devices (Berolo et al., 2011; Lui et al., 2011) and
mobile phones which consisted of cell phones, keypad phones,
smartphones and touchscreen phones (Ali et al., 2014; Berolo et al.,
2011; Eapen et al., 2010; Goldfinch et al., 2011; Gustafsson et al.,
2010, 2011; Hakala et al., 2006; Kim, 2015; Kim and Kim, 2015;
Shan et al., 2013; Stalin et al., 2016; Xie et al., 2016). The duration
of daily use of mobile handheld devices ranged from less than one
hour to six hours among users (Balakrishnan et al., 2016; Berolo
et al., 2011; Hakala et al., 2006; Hegazy et al., 2016; Kim and Kim,
2015; Lui et al., 2011; Shan et al., 2013). Making phone calls, text-
ing messages, internet browsing and playing games were the
frequent activities among users of mobile handheld devices, as
reported in the included studies (Ali et al., 2014; Balakrishnan et al.,
2016; Berolo et al., 2011; Eapen et al., 2010; Kim and Kim, 2015; Lui
et al., 2011).
3.4. Results of prevalence rates
Ten studies examined the prevalence rates, but only seven (Ali
et al., 2014; Balakrishnan et al., 2016; Berolo et al., 2011; Eapen
et al., 2010; Kim and Kim, 2015; Shan et al., 2013; Stalin et al.,
2016) provided separate data on the prevalence of musculoskel-
etal complaints associated with the use of mobile handheld devices
(Table 4). In general, most studies reported on either lifetime or
period prevalence (Table 4). A variety of operational definitions of
musculoskeletal complaints were used to identify cases (Table 4),
which is the primary factor hampering the possibility of pooling
results using meta-analysis. Hence, the present review undertook a
narrative synthesis of information in terms of prevalence and risk
factors associated with musculoskeletal complaints among users of
mobile devices.
The reported body regions of musculoskeletal complaints
included the neck, upper limb regions, upper and low back. Four
studies reported neck complaints with six-month prevalence of
40.9%e44.1% (Shan et al., 2013), one-week prevalence of 67.8%
(Berolo et al., 2011), and lifetime prevalence of 55.8% (Kim and Kim,
2015). Stalin et al. (2016) reported a prevalence rate of 17.3% for
neck pain without specifying the sampling period. Berolo et al.
(2011) demonstrated that one-week prevalence of upper back
pain of any severity was 62.2%. Shan et al. (2013) presented six-
month prevalence of around 32% while Kim and Kim (2015)
showed lifetime prevalence of 29.8% for low back pain among
smartphone users and tablet computer users. Regarding complaints
in upper limb regions, Balakrishnan et al. (2016) reported a sum-
med prevalence rate of 72.5% for pain in all upper limb regions.
However, others reported prevalence rates of a specific region in
the upper limb, with a range of 1.0% (Eapen et al., 2010) to 56.9%
(Berolo et al., 2011), and complaints in shoulders, fingers and
thumbs were more commonly reported. Specifically, lifetime
prevalence of shoulder complaints ranged from 1.2% (Eapen et al.,
2010) to 54.8% (Kim and Kim, 2015). Complaints in the thumbs
were found to range from 9.8% (Eapen et al., 2010) to 56.9% (Berolo
et al., 2011), while in fingers were from 2.2% (Eapen et al., 2010) to
19.9% (Kim and Kim, 2015).
3.5. Risk factors and strength of evidence
Generally, the risk factors identified in the reviewed studies
could be grouped into four main categories: postures adopted
while using mobile handheld devices, time (or duration) spent,
tasks frequently performed and human-device interaction tech-
niques (Table 5). Some evidence was shown that neck flexion, as
well as frequency of phone calls, texting and gaming were risk
factors for developing musculoskeletal complaints among mobile
device users. Inconclusive evidence was found for the total time
spent on mobile devices in relation to musculoskeletal disorders
due to inconsistent results in the selected studies. As illustrated in

Table 2a
Methodological quality scores of 11 cross-sectional studies.

Included studies External validity criteria Internal validity criteria Overall quality

1 2 3 4 5 6 7 8 9 10 11

Ali et al. (2014) N N N Y Y Y N Y N Y þ

Balakrishnan et al. (2016) N N Y Y Y N N Y Y Y þ
Berolo et al. (2011) N N N Y Y Y N Y Y Y þþ
Eapen et al. (2010) N N N Y Y Y N Y N Y þ
Goldfinch et al. (2011) N N Y N Y N N Y N Y -
Hakala et al. (2006) Y Y Y Y Y N Y Y Y Y þþ
Hegazy et al. (2016) N Y Y Y Y N Y Y Y N þþ
Kim and Kim (2015) N N N Y Y N N Y N Y -
Lui et al. (2011) N N Y Y Y N Y Y Y Y þþ
Shan et al. (2013) N Y Y Y Y N Y Y Y Y þþ
Stalin et al. (2016) N Y N Y Y N N Y N Y þ

Note: N ¼ No; Y ¼ Yes; þþ ¼ high quality (low risk of bias); þ ¼ acceptable (moderate risk of bias); e ¼ low quality (high risk of bias); 1 e Was the study's target population a
close representation of the national population in relation to relevant variables, e.g. age, sex, occupation? 2 e Was the sampling frame a true or close representation of the
target population? 3 e Was some form of random selection used to select the sample, OR, was a census undertaken? 4 e Was the likelihood of non-response bias minimal? 5 e
Were data collected directly from the subjects (as opposed to a proxy)? 6 e Was an acceptable case definition used in the study? 7 e Was the study instrument that measured
the parameter of interest (e. g. prevalence of low back pain) shown to have reliability and validity (if necessary)? 8 e Was the same mode of data collection used for all
subjects? 9 e Was the length of the shortest prevalence period for the parameter of interest appropriate? 10 e Were the numerator(s) and denominator(s) for the parameter of
interest appropriate? 11 e Summary item on the overall risk of bias. (Hoy et al., 2012).
 Y. Xie et al. / Applied Ergonomics 59 (2017) 132e142 137

Table 2b
Methodological quality scores of four case-control studies.

Included studies Items Overall quality

1 2 3 4 5 6 7 8 9 10 11 12

Gustafsson et al. (2010) Y C Y C N Y Y Y Y Y N þ

Gustafsson et al. (2011) Y C Y C N Y Y Y Y C Y þ
Kim (2015) Y C Y C N Y N Y Y N N þ
Xie et al. (2016) Y C Y C N Y Y Y Y Y N þ

Note: N ¼ No; Y ¼ Yes; C ¼ Can't say; þ ¼ acceptable (moderate risk of bias); 1 e The study addresses an appropriate and clearly focused question; 2 e The cases and controls
are taken from comparable populations; 3 e The same exclusion criteria are used for both cases and controls; 4 e What percentage of each group (cases and controls)
participated in the study?; 5 e Comparison is made between participants and non-participants to establish their similarities or differences; 6 e Cases are clearly defined and
differentiated from controls; 7 e it is clearly established that controls are non-cases; 8 e Measures will have been taken to prevent knowledge of primary exposure influencing
case ascertainment; 9 e Exposure status is measured in a standard, valid and reliable way; 10 e The main potential confounders are identified and taken into account in the
design and analysis; 11 e Confidence intervals are provided; 12 e How well was the study done to minimize the risk of bias or confounding? (Scottish Intercollegiate Guideline
Network, 2014).

Table 3
Study characteristics of included studies.

First author (year) Study design Country Response rate (%) Age Sampling frame

Ali et al. (2014) Survey Pakistan 300/300 (100%) Unclear Undergraduate students
Balakrishnan et al. (2016) Survey Malaysia 200/200 (100%) 19e30 University students
Berolo et al. (2011) Survey Canada 137/140 (97.9%) Unclear Students, faculty and staff
Eapen et al. (2010) Survey India 1363/1500 (90.9%) 18e29 Students
Gustafsson et al. (2010, 2011) Case-control study Sweden 56/60 (93.3%) 19e25 Young adults from colleges and universities
Goldfinch et al. (2011) Survey New Zealand 240/471 (51.0%) 43.2 Staff in government agencies
Hakala et al. (2006) Survey Finland 6003/8810 (68.1%) 14e18 Adolescents
Hegazy et al. (2016) Survey Saudi Arabia 472/500 (94.4%) 19e25 Medical college students
Kim (2015) Case-control study South Korea 27/27 (100%) 20.6 ± 1.6 Young adults in a university
Kim and Kim (2015) Survey Korea 292/300 (97.3%) 21.4 ± 1.6 University students
Lui et al. (2011) Survey Hong Kong 476/600 (79.3%) 8e13 Primary 4 to 6 school children
Shan et al. (2013) Survey China 3016/3600 (83.9%) 15e19 High school students
Stalin et al. (2016) Survey India 2054/2121 (76.8%) >18 Adults
Xie et al. (2016) Case-control study Hong Kong 40/40 (100%) 23.9 ± 3.2 Young adults

Table 5, 76.2% (16/21) of risk factors involving postures, tasks per-
formed, and human-device interaction techniques were examined
by only one study. As a result, evidence on associations between
these risk factors and musculoskeletal complaints was largely
inconclusive.
4. Discussion
This review, to our knowledge, is the first systematic review to
identify the prevalence of and risk factors for musculoskeletal
complaints among users of mobile handheld devices. This review
shows that the prevalence of musculoskeletal complaints in
different body regions ranges from 1.0% to 67.8%. Consistently, the
neck has the highest prevalence rate, which is from 17.3% to 67.8%.
Some evidence is found for a neck flexion posture, as well as fre-
quency of phone calls, texting and gaming as important risk factors.
4.1. Methodological considerations
Several major methodological limitations have been identified
in the included studies. Firstly, most of the included studies showed
a selection bias. For one thing, a random selection of samples was
not performed in most of the studies (see Tables 2a and 2b). For
another, the inclusion and exclusion criteria were unclear in eight
included studies (Ali et al., 2014; Goldfinch et al., 2011; Hakala et al.,
2006; Hegazy et al., 2016; Kim and Kim, 2015; Lui et al., 2011; Shan
et al., 2013; Stalin et al., 2016). Eligibility criteria are important
information for determining possible risk of bias in observational
studies and it should be clearly documented (Guyatt et al., 2011).
Without this information, it is not clear whether the selected
studies recruited those who had medical conditions or traumatic
history affecting the musculoskeletal system, probably resulting in
higher prevalence rates of musculoskeletal complaints. Another
methodological limitation is the case definition (see item 6 in
Table 2a). Firstly, it is not clear in most of the included studies
whether a threshold of pain level was required for case identifi-
cation. Two studies (Balakrishnan et al., 2016; Berolo et al., 2011)
identified those who reported a pain score of 1/10 or higher as
cases. However, low grade pain such as a pain score of 1/10 may not
be clinically relevant. Furthermore, most of the studies did not
provide detailed descriptions or body diagrams of specific
anatomical locations. For example, some studies have only named
the “neck” region without specifically describing the extent of
anatomical boundaries (e.g. Kim and Kim, 2015; Shan et al., 2013;
Stalin et al., 2016). Berolo et al. (2011) was the only cross-
sectional study that used a body diagram to illustrate the exact
region referred to as “neck”. Some studies referred to the problem
of “neck pain” while others used the term “neck-shoulder pain”
(see Table 4). This issue has also been discussed in previous studies
(Guzman et al., 2009; Hoy et al., 2010). Without the use of pain
thresholds and body diagrams, results on prevalence rates are very
likely to be biased. Future studies should aim to overcome meth-
odological limitations of sample selections and case definitions.
Reliability and validity of instruments adopted to measure the
prevalence or exposures were not assessed in most of the studies.
Self-reports, observational methods and direct measurements are
commonly employed to examine exposures to risk factors for
musculoskeletal complaints (David, 2005). All included cross-
sectional studies (Table 2a) adopted the self-reported question-
naires to collect data of prevalence rates and exposures to mobile
handheld devices. Among these, only one study (Ali et al., 2014)
confirmed cases by physical examination and four studies (Hakala
138 Y. Xie et al. / Applied Ergonomics 59 (2017) 132e142

Table 4
Prevalence rates of musculoskeletal complaints (MSC) among users of mobile devices.

First author (year) Types of mobile Definitions of MSC to Prevalence Results of prevalence rates
handheld devices identify cases

Ali et al. (2014) Cell phones, De Quervain's Lifetime prevalence 42% were experiencing pain in the thumb/wrist and 49.67% showed
touchscreen phones tenosynovitis positive results in Finkelstein test.
and keypad phones
Balakrishnan et al. Handheld devices such Musculoskeletal One-week 72.5% had mild to severe pain in the upper limb; 50% had mild to
(2016) as cell phones disorders of upper prevalence extreme pain, 48% had mild to extreme tingling sensation, 42% had mild
extremity to severe weakness and 44% had mild to extreme stiffness in the arm,
shoulder and hand while performing any specific activity.
Berolo et al. (2011) Mobile handheld Pain One-week 84% of participants reported pain of any severity in at least one body
devices including prevalence part.
mobile phones, Pain of any severity: 37.1% in the left thumb, 56.9% in the right thumb;
personal digital 16.4% in right fingers, 10.7% in left fingers; 32.1% in the right elbow,
assistant, handheld 27.2% in the left elbow; 45.7% in the left shoulder, 52.1% in the right
video game devices and shoulder; 67.8% in the neck and 62.2% in the upper back.
game controllers
Eapen et al. (2010) Cell phones Cumulative trauma Lifetime prevalence Symptoms: 9.8% in the thumb, 2.7% in the elbow, 2.4% in the wrist, 2.2%
disorders including in the fingers, 1.9% in the hand, 1.3% in the forearm, 1.2% in the shoulder
pain, fatigue, and 1.0% in the arm. Symptoms reported in this group included pain
discomfort, stiffness (61.7%), fatigue (44.3%), stiffness (16.6%) and weakness (15.8%).
and weakness Duration of symptoms: 10.1% had symptoms lasting less than 5 min and
did not affect daily activities; 10.6% needed to shift the phone to the
other hand because of symptoms and 4.3% felt discomfort affected their
activities of writing or holding fine/small objects.
Goldfinch et al. Cell phones Pain in the back and Unclear These two studies provided prevalence of MSC associated with
(2011) upper limbs electronic devices, but no separate data on prevalence of MSC associated
Hakala et al. (2006) Mobile phones Neck-shoulder or low Six-month with handheld device use.
back pain prevalence
Kim and Kim Smartphones Pain Lifetime prevalence Pain in body regions: 55.8% in the neck, 54.8% in the shoulder, 19.2% in
(2015) arms, 19.2% in hands, 27.1% in wrists, 19.9% in fingers, 29.8% in the low
back.
Lui et al. (2011) Small-screen handheld Discomfort One-month It provided prevalence of MSC associated with electronic devices, but no
devices including prevalence separate data on prevalence of MSC associated with handheld device
PlayStation Portable, use.
Game Boy and mobile
phones
Shan et al. (2013) Mobile phones and Neck/shoulder or low Six-month Mobile phone users: 40.9% of neck/shoulder pain and 32.2% of low back
tablets back pain prevalence pain.
Tablet users: 44.1% of neck/shoulder pain and 32.9% of low back pain.
Stalin et al. (2016) Mobile phones Neck pain and painful Unclear 17.3% of neck pain and 4.0% of painful fingers among mobile phone
fingers users.

et al., 2006; Hegazy et al., 2016; Lui et al., 2011; Shan et al., 2013) studies are warranted before solid conclusions can be drawn.
reported test-retest reliability of the questionnaires used. Other
studies may not furnish a strong validity in the assessment of
4.2. Prevalence of musculoskeletal complaints
prevalence rates and exposures without testing the reliability or
validity of the questionnaire (Heinrich et al., 2004; IJmker et al.,
Although complaints in the neck, shoulders, thumbs and fingers
2008), which could be an important source of bias. Previous
were more commonly reported among users of mobile handheld
studies (Homan and Armstrong, 2003; Gold et al., 2015; Berolo
devices compared with other body regions, wide ranges of preva-
et al., 2015) found that there is low agreement between
lence rates were noted. Previous systematic reviews also found
measuring exposures such as time spent on mobile devices by self-
similar wide ranges of period or lifetime prevalence for neck and
report questionnaires and by direct and objective measurements
upper limb complaints in the general population (Fejer et al., 2006;
such as a phone bill, phone activity measure applications and ac-
Hogg-Johnson et al., 2008; Huisstede et al., 2006; Luime et al.,
tivity monitors. Regarding the case-control studies (Gustafsson
2004). For example, Hogg-Johnson et al. (2008) reported
et al., 2010, 2011; Kim, 2015; Xie et al., 2016) included in this re-
12emonth prevalence of 30%e50% while Fejer et al. (2006) illus-
view, they employed direct measurements such as using surface
trated that period prevalence was between 6.0% and 75.1% and
electromyography and motion tracking systems to evaluate muscle
lifetime prevalence ranged from 0.2% to 71% for neck pain in a
activity and neck flexion angles. Exposures such as human-device
worldwide general population. Furthermore, studies reviewing the
interaction techniques were observed by two experienced ergon-
prevalence of musculoskeletal disorders among specific working
omists in Gustafsson et al. (2011). Future studies should confirm
populations such as hospital physicians (Hengel et al., 2011) and
musculoskeletal complaints by physical examination and measure
pianists (Bragge et al., 2006) also reported a broad range of prev-
exposures to risk factors among users of mobile devices through
alence rates. It was proposed that the large variation in prevalence
direct measurements in order to provide more accurate data.
rates was likely due to the broad definition used to identify cases,
Finally, to date, no available prospective studies have investi-
including definitions of the disorder itself and anatomical areas
gated the risk factors contributing to musculoskeletal complaints
(Bragge et al., 2006; Fejer et al., 2006; Hengel et al., 2011; Luime
among users of handheld devices. Cross-sectional study designs
et al., 2004). This may also account for the wide range of preva-
could not determine a causal relationship between exposures and
lence rates found in the current systematic review. Alternatively,
musculoskeletal complaints. Further high-quality and prospective
different types of prevalence assessed in the evaluated studies
 Y. Xie et al. / Applied Ergonomics 59 (2017) 132e142 139

Table 5
Summary of risk factors assessed in the included studies for musculoskeletal complaints (MSC) associated with the use of mobile handheld devices.

Group Exposures/Risk Number Studies assessed and relevant results Strength

factors of studies of evidence

Postures Neck flexion 2 Gustafsson et al. (2011): Proportions of subjects in neck flexion $ 40% between symptomatic and Some
asymptomatic subjects –42% vs 7%, p < 0.05;
Kim (2015): Upper and lower cervical flexion angles during smartphone texting in the neck pain group
were significantly higher compared to pain-free group.
Thumb 1 Gustafsson et al. (2010): No significant differences in the thumb median angle, median velocity, median Inconclusive
postures power frequency and the pause percentage between symptomatic and asymptomatic subjects.
Standing 1 Shan et al. (2013): Standing and neck pain–AOR ¼ 1.08, 95%CI ¼ 0.92e1.27; and low back pain– Inconclusive
AOR ¼ 1.31, 95%CI ¼ 1.11e1.55.
Sitting 1 Shan et al. (2013): Sitting and neck pain–AOR ¼ 0.67, 95%CI ¼ 0.58e0.79; and low back pain–AOR ¼ 0.86, Inconclusive
95%CI ¼ 0.73e1.01.
Lying 1 Shan et al. (2013): Lying while using phones and neck/shoulder pain–AOR ¼ 1.42, 95%CI ¼ 1.20e1.68; Inconclusive
and low back pain–AOR ¼ 1.27, 95%CI ¼ 1.06e1.52.
Lying while using tablets and neck/shoulder pain–AOR ¼ 1.13, 95%CI ¼ 0.85e1.53; and low back pain–
AOR ¼ 1.46, 95%CI ¼ 1.08e1.98.
Semi-reclining 1 Shan et al. (2013): Semi-reclining while using phones and neck/shoulder pain–AOR ¼ 1.07, 95%CI ¼ 0.90 Inconclusive
e1.28; and low back pain–AOR ¼ 0.85, 95%CI ¼ 0.82e1.18.
Semi-reclining while using tablets and neck/shoulder pain–AOR ¼ 1.02, 95%CI ¼ 0.78e1.33; and low
back pain–AOR ¼ 0.93, 95%CI ¼ 0.70e1.23.
Time Total time 9 Berolo et al. (2011): Association between total time $2.4 h/day and pain in the left shoulder–AOR ¼ 2.06, Inconclusive
95%CI ¼ 1.00e4.24; right shoulder–AOR ¼ 2.55, 95%CI ¼ 1.25e5.21; neck–AOR ¼ 2.27, 95%CI ¼ 1.24
e5.96.
Eapen et al. (2010): 72.7% of subjects reported the cause for MSC was excessive use of mobile devices.
Goldfinch et al. (2011): Time was not a significant predictor of pain in the upper back, lower back,
shoulder and arm.
Hakala et al. (2006): No significant association with neck/shoulder/low back pain when seven
confounding factors* adjusted.
Hegazy et al. (2016): Association between total time on texting/gaming and back pain–AOR ¼ 8.63, 95%
CI ¼ 2.94e25.36.
Kim and Kim (2015): Use hours of smartphones were not significantly correlated with MSC.
Lui et al. (2011): Association between total time $2 h/day and discomfort in any regions–AOR ¼ 4.75,
95%CI ¼ 1.06e21.36.
Shan et al. (2013): Association between total time $2 h/day and neck/shoulder pain–AOR ¼ 1.49, 95%
CI ¼ 1.20e1.86; low back pain–AOR ¼ 1.84, 95%CI ¼ 1.46e2.32.
Stalin et al. (2016): Association between mobile phone use and neck pain–AOR ¼ 2.05, 95%CI ¼ 1.50
e2.80; painful fingers–AOR ¼ 3.15, 95%CI ¼ 1.48e6.69.
Tasks Frequency of 3 Berolo et al. (2011): Association with moderate/severe pain in the left shoulder–AOR ¼ 3.00, 95% Some
frequently phone calls CI ¼ 1.19e7.55; right shoulder–AOR ¼ 2.89, 95%CI ¼ 1.24e6.75; neck–AOR ¼ 2.48, 95%CI ¼ 1.03e4.34.
performed Eapen et al. (2010): Proportions of subjects who make phone calls frequently in symptomatic vs
asymptomatic groups – 96.5% vs 63.8%, p < 0.001.
Hegazy et al. (2016): Association with neck pain–AOR ¼ 4.01, 95%CI ¼ 1.72e9.37.
Frequency of 3 Ali et al. (2014): Proportions of subjects who frequently use the phone for texting showing positive vs Some
texting negative Finkelstein test–64% vs 37%, p < 0.001.
Berolo et al. (2011): Texting and moderate/severe pain in the right shoulder–AOR ¼ 2.34; 95%CI ¼ 1.02
e5.78; left shoulder–AOR ¼ 2.97; 95%CI ¼ 1.17e7.53.
Eapen et al. (2010): Proportions of subjects in symptomatic vs asymptomatic groups–96.5% vs 63.1%,
p < 0.001.
Frequency of 1 Berolo et al. (2011): Web browsing and pain in the middle of the right thumb–AOR ¼ 2.61; 95%CI ¼ 1.01 Inconclusive
web browsing e6.76.
Frequency of 2 Berolo et al. (2011): Gaming and pain of any severity in the right shoulder–AOR ¼ 4.09; 95%CI ¼ 1.77 Some
gaming e9.42.
Eapen et al. (2010): Proportion of subjects in symptomatic vs asymptomatic groups–65.2% vs 57.8%,
p ¼ 0.029.
Frequency of 1 Berolo et al. (2011): Scheduling and pain of any severity in the left hand–AOR ¼ 2.06, 95%CI ¼ 1.00e4.24. Inconclusive
scheduling
Entertainment 1 Berolo et al. (2011): Entertainment and pain of any severity in neck–AOR ¼ 2.23, 95%CI ¼ 1.03e4.82. Inconclusive
Human-device Without back 1 Gustafsson et al. (2011): Proportions of subjects in symptomatic vs asymptomatic groups–56% vs 23%, Inconclusive
interaction support p < 0.005.
techniques Without arm 1 Gustafsson et al. (2011): Proportions of subjects in symptomatic vs asymptomatic groups–34% vs 7%, Inconclusive
support p < 0.005.
One-handed 1 Gustafsson et al. (2011): Proportions of subjects in symptomatic vs asymptomatic groups–34% vs 47%, Inconclusive
grip p > 0.005.
Key press with 1 Gustafsson et al. (2011): Proportions of subjects in symptomatic vs asymptomatic groups–68% vs 47%, Inconclusive
one thumb p > 0.005.
Key press with 1 Gustafsson et al. (2011): Proportions of subjects in symptomatic vs asymptomatic groups–68% vs 60%, Inconclusive
the medial side p > 0.005.
of the thumb
Eye-to-screen 1 Shan et al. (2013): No correlations with neck/shoulder and low back pain. Inconclusive
distances
Others Screen sizes 1 Kim and Kim (2015): Correlations with low back pain–r ¼ 0.129, p < 0.05 Inconclusive
Increased 1 Xie et al. (2016): Significantly increased activity in upper trapezius in young adults with chronic neck- Inconclusive
muscle activity shoulder pain compared with those without pain, p < 0.05.

Note: AOR, adjusted odd ratio; h/day ¼ hours per day; *Seven confounding factors including age, gender, parents' level of education, school success, time of puberty, efficiency
of physical activity and stress symptoms.
140 Y. Xie et al. / Applied Ergonomics 59 (2017) 132e142
could also be an explanation for the wide range of prevalence rates
reported.
4.3. Evidence on risk factors
4.3.1. Posture
Some evidence is found for neck flexion in relation to muscu-
loskeletal disorders among users of mobile devices. This result is
consistent with previous systematic reviews supporting neck
flexion as an important risk factor associated with work-related
musculoskeletal disorders (Arie €ns et al., 2000; da Costa and
Vieira, 2010; Erick and Smith, 2011). The mechanical demand on
neck muscles posed by a flexed cervical spine during sitting has
been demonstrated to be 3e5 times that of a neutral neck posture
(Vasavada et al., 2015). A sustained flexed neck posture adopted by
mobile device users is probably one of the key factors to explain the
high prevalence rate of neck complaints in the current review. The
result implies that correcting awkward neck postures while using
mobile devices is an important strategy to reduce or prevent neck
pain among users of mobile devices. Yet, to provide a more com-
plete picture of healthy use of mobile devices, well-designed in-
vestigations are needed to identify what degree of neck flexion is
critical for increasing the compressive loading on the cervical spine
and what duration of holding such a flexion posture is a “safe” or
critical threshold for not over-loading the neck structures in using
mobile devices.
4.3.2. Total time spent on mobile devices
Interestingly, inconclusive evidence is found on total time spent
on mobile devices in relation to musculoskeletal complaints. Of
nine studies that evaluated time as a risk factor, six studies showed
significant association between total durations of device use and
musculoskeletal complaints such as the neck, shoulder and low
back pain while three studies found no association. Inconsistent
results of time as a risk factor in the eligible studies may be due to a
wide variation of adjustment for confounding variables in the
included studies. Among the nine studies measuring time as a risk
factor, five cross-sectional studies (Berolo et al., 2011; Hegazy et al.,
2016; Lui et al., 2011; Shan et al., 2013; Stalin et al., 2016) showed
moderate to strong association, with adjusted OR ranging from 1.49
to 8.63, between a long duration of mobile handheld device use and
musculoskeletal complaints. While some studies (Eapen et al.,
2010; Goldfinch et al., 2011; Kim and Kim, 2015; Shan et al.,
2013; Stalin et al., 2016) did not adjust or adjusted only limited
confounding factors such as gender and age, other studies adjusted
many confounding factors simultaneously (Berolo et al., 2011;
Hakala et al., 2006; Hegazy et al., 2016; Lui et al., 2011). Hakala
et al. (2006) found that using mobile phones more than five
hours daily was significantly associated with neck-shoulder pain
(OR ¼ 2.2, 95%CI ¼ 1.1e3.7) when age and gender were adjusted.
However, the association became weaker and even disappeared
when more confounding factors including parents' level of educa-
tion, school success, time of puberty, efficiency of physical activity
and stress symptoms in addition to age and gender were adjusted
(Hakala et al., 2006). Although confounder adjustment is an
important method to minimize bias, unnecessary or over adjust-
ment could result in negative rather than positive effects on the
precision of results (Schisterman et al., 2009). More attention
should be paid to identify potential confounders and make proper
adjustment in future studies.
Differences in defining the threshold of time spent on mobile
handheld devices as an exposure may also contribute to inconsis-
tent results in the included studies. Furthermore, the included
studies only measured total daily duration on mobile handheld
devices as an exposure. Details on continuous time spent on mobile
handheld devices and human-device interaction-rest patterns
were not provided. However, human-device interaction-rest pat-
terns may be an important factor in the development of musculo-
skeletal complaints. The phenomenon of superficial muscles being
continuously activated has been proposed to be a key contributing
factor for chronic musculoskeletal disorders in computer users
(Ha€gg, 1991; Szeto et al., 2005). Regular rest breaks during work is
demonstrated to be beneficial to relieve work-related musculo-
skeletal complaints among office workers (Galinsky et al., 2007;
van den Heuvel et al., 2003). Therefore, work-rest patterns while
interacting with mobile devices in addition to total daily duration of
use should be included as a more comprehensive variable while
evaluating the risk factors for musculoskeletal complaints associ-
ated with mobile handheld devices.
4.3.3. Tasks frequently performed and human-device interaction
techniques
Some evidence is found for the frequency of phone calls, texting
and gaming associated with musculoskeletal complaints among
mobile device users. According to Berolo et al. (2011) and Hegazy
et al. (2016), frequency of phones calls, texting and gaming are
strongly associated with pain in the neck or shoulder, with OR
ranging from 2.48 to 4.01. Eapen et al. (2010) and Ali et al. (2014)
reported that compared with asymptomatic persons, symptom-
atic individuals made phone calls, texted message and played
games on mobile phones significantly more frequently. Holding a
phone near the ear during phone calls for a long time is associated
with a prolonged static posture while texting and gaming are more
correlated with repetitive movements, both of which involve sus-
tained contraction of neck and shoulder muscles in order to
maintain the posture or control the movement (Gustafsson et al.,
2010; Kietrys et al., 2015; Ning et al., 2015; Xie et al., 2016). Sus-
tained static postures and repetitive movements of the finger and
the upper limb have also been demonstrated in previous studies as
risk for the development of neck, shoulder, forearm or thumb dis-
orders (Barr et al., 2004; Larsson et al., 2007; Gupta and
Mahalanabis, 2006; Thomsen et al., 2008). Yet, to gain further
insight into the relationship of frequency of phone calls, texting and
gaming with musculoskeletal complaints, further research is
required to provide more concrete evidence on how these different
tasks affect the muscle activity and postures.
Many risk factors including different human-device interaction
techniques were examined by only one study (see Table 5).
Therefore, the results of inconclusive evidence for these risk factors
need to be interpreted with caution. Different conclusions might be
drawn when new studies emerge in the future. Additionally, other
possible risk factors, for example, speeds of text entry or browsing,
different input methods like using index fingers to interact with
mobile devices and using a mobile device in different positions
such as an extreme neck extension in prone or during walking have
not been investigated so far. The smartphone technology is also
rapidly evolving and new products with different screen sizes and
different input methods are being produced for the consumers all
the time. Hence the research needs to keep pace with the new
products in order to produce knowledge about the current state of
the art in this technology era.
4.4. Limitations of this review
There are some limitations in this systematic review. A lack of
effort to identify unpublished peer-reviewed studies and non-
English written articles as well as the exclusion of laboratory
studies may introduce bias in this review. In addition, some studies
have probably been missed out although an extensive literature
search was performed.
 Y. Xie et al. / Applied Ergonomics 59 (2017) 132e142
5. Conclusion
Five articles of high quality, eight of acceptable quality and two
of low quality were evaluated in this systematic review. A wide
range of prevalence rates of musculoskeletal complaints in
different body regions has been reported, with the highest preva-
lence commonly found in neck complaints. This review has only
found some evidence showing that a neck flexion posture as well as
the frequency of phone calls, texting and gaming are associated
with musculoskeletal disorders among users of mobile handheld
devices. Interpretation of the results of risk factors in this review
should be treated with caution since many risk factors have been
examined by only one study. More high-quality and prospective
studies are needed to establish the relationship between the use of
mobile handheld devices and musculoskeletal complaints.
References
Ali, M., Asim, M., Danish, S.H., Ahmad, F., Iqbal, A., Hasan, S.D., 2014. Frequency of De
Quervain's tenosynovitis and its association with SMS texting. Muscles Liga-
ments Tendons J. 4 (1), 74e78.
Arie€ns, G.A.M., van Mechelen, W., Bongers, P.M., Bouter, L.M., van der Wal, G., 2000.
Physical risk factors for neck pain. Scand. J. Work Environ. Health 26 (1), 7e19.
Ashurst, J.V., Turco, D.A., Lieb, B.E., 2010. Tenosynovitis caused by texting: an
emerging disease. J. Am. Osteopath. Assoc. 110 (5), 294e296.
Asundi, K., Odell, D., Luce, A., Dennerlein, J.T., 2010. Notebook computer use on a
desk, lap and lap support: effects on posture, performance and comfort. Ergo-
nomics 53 (1), 74e82.
Augner, C., Gnambs, T., Winker, R., Barth, A., 2012. Acute effects of electromagnetic
fields emitted by GSM mobile phones on subjective well-being and physio-
logical reactions: a meta-analysis. Sci. Total Environ. 424, 11e15.
Bai, A., Shukla, V.K., Bak, G., Wells, G., 2012. Quality Assessment Tools Project
Report. Canadian Agency for Drugs and Technologies in Health, Ottawa.
Balakrishnan, R., Chinnavan, E., Feii, T., 2016. An extensive usage of hand held de-
vices will lead to musculoskeletal disorder of upper extremity among student in
AMU: a survey method. Int. J. Phys. Educ. Sports Health 3 (2), 368e372.
Barr, A.E., Barbe, M.F., Clark, B.D., 2004. Work-related musculoskeletal disorders of
the hand and wrist: epidemiology, pathophysiology, and sensorimotor changes.
J. Orthop. Sports Phys. Ther. 34 (10), 610e627.
Berolo, S., Wells, R.P., Amick, B.C., 2011. Musculoskeletal symptoms among mobile
hand-held device users and their relationship to device use: a preliminary
study in a Canadian university population. Appl. Ergon. 42 (2), 371e378.
Berolo, S., Steenstra, I., Amick, B.C., Wells, R.P., 2015. A comparison of two methods
to assess the usage of mobile hand-held communication devices. J. Occup.
Environ. Hyg. 12 (4), 276e285.
Bragge, P., Bialocerkowski, A., McMeeken, J., 2006. A systematic review of preva-
lence and risk factors associated with playing-related musculoskeletal disor-
ders in pianists. Occup. Med. Lond. 56 (1), 28e38.
Chu, M.K., Song, H.G., Kim, C., Lee, B.C., 2011. Clinical features of headache associ-
ated with mobile phone use: a cross-sectional study in university students.
BMC Neurol. 11, 115.
Co
^ te
", P., van der Velde, G., Cassidy, J.D., Carroll, L.J., Hogg-Johnson, S., Holm, L.W.,
Carragee, E.J., Haldeman, S., Nordin, M., Hurwitz, E.L., 2009. The burden and
determinants of neck pain in workers: results of the bone and joint decade
2000e2010 task force on neck pain and its associated disorders. J. Manip.
Physiol. Ther. 32 (2), S70eS86.
da Costa, B.R., Vieira, E.R., 2010. Risk factors for work-related musculoskeletal dis-
orders: a systematic review of recent longitudinal studies. Am. J. Ind. Med. 53
(3), 285e323.
David, G.C., 2005. Ergonomic methods for assessing exposure to risk factors for
work-related musculoskeletal disorders. Occup. Med. Lond. 55 (3), 190e199.
Eapen, C., Kumar, B., Bhat, A.K., 2010. Prevalence of cumulative trauma disorders in
cell phone users. J. Musculoskelet. Res. 13 (3), 137e145.
Erick, P.N., Smith, D.R., 2011. A systematic review of musculoskeletal disorders
among school teachers. BMC Musculoskelet. Disord. 12, 260.
Fejer, R., Kyvik, K.O., Hartvigsen, J., 2006. The prevalence of neck pain in the world
population: a systematic critical review of the literature. Eur. Spine J. 15 (6),
834e848.
Fernandez-Guerrero, I.M., 2014. WhatsAppitis. Lancet 383 (9922), 1040e6736(14)
60519-5.
Galinsky, T., Swanson, N., Sauter, S., Dunkin, R., Hurrell, J., Schleifer, L., 2007. Sup-
plementary breaks and stretching exercises for data entry operators: a follow-
up field study. Am. J. Ind. Med. 50 (7), 519e527.
Gold, J., Driban, J., Yingling, V.R., Komaroff, E., 2012. Characterization of posture and
comfort in laptop users in non-desk settings. Appl. Ergon. 43 (2), 392e399.
Gold, J.E., Rauscher, K.J., Zhu, M., 2015. A validity study of self-reported daily texting
frequency, cell phone characteristics, and texting styles among young adults.
BMC Res. Notes 8 (1), 120.
Goldfinch, S., Gauld, R., Baldwin, N., 2011. Information and communications
141
technology use, e-government, pain and stress amongst public servants. New
Tech. Work Employ. 26 (1), 39e53.
Gupta, A.D., Mahalanabis, D., 2006. Study of hand function in a group of shoe fac-
tory workers engaged in repetitive work. J. Occup. Rehabil. 16 (4), 657e684.
Gustafsson, E., Johnson, P.W., Hagberg, M., 2010. Thumb postures and physical loads
during mobile phone use - a comparison of young adults with and without
musculoskeletal symptoms. J. Electromyogr. Kinesiol. 20 (1), 127e135.
Gustafsson, E., Johnson, P.W., Lindegard, A., Hagberg, M., 2011. Technique, muscle
activity and kinematic differences in young adults texting on mobile phones.
Ergonomics 54 (5), 477e487.
Guyatt, G.H., Oxman, A.D., Vist, G., Kunz, R., Brozek, J., Alonso-Coello, P., Montori, V.,
Akl, E.A., Djulbegovic, B., Falck-Ytter, Y., 2011. GRADE guidelines: 4. Rating the
quality of evidencedstudy limitations (risk of bias). J. Clin. Epidemiol. 64 (4),
407e415.
Guzman, J., Hurwitz, E.L., Carroll, L.J., Haldeman, S., Co ^ te
", P., Carragee, E.J.,
Peloso, P.M., van der Velde, G., Holm, L.W., Hogg-Johnson, S., Nordin, M.,
Cassidy, J.D., 2009. A new conceptual model of neck pain: linking onset, course,
and care: the bone and joint decade 2000-2010 task force on neck pain and its
associated disorders. J. Manip. Physiol. Ther. 32 (2 Suppl.), S17eS28.
Ha€gg, G., 1991. Static work loads and occupational myalgia-a new explanation
model. Electromyogr. Kinesiol. 9, 141e144.
Hakala, P.T., Rimpela, A.H., Saarni, L.A., Salminen, J.J., 2006. Frequent computer-
related activities increase the risk of neck-shoulder and low back pain in ado-
lescents. Eur. J. Public Health 16 (5), 536e541.
Hegazy, A.A., Alkhail, B.A., Awadalla, N.J., Qadi, M., AI-Ahmadi, J., 2016. Mobile
phone use and risk of adverse health impacts among medical students in
Jeddah, Saudi Arabia. Br. J. Med. Med. Res. 15 (1), 1e11.
Heinrich, J., Blatter, B.M., Bongers, P.M., 2004. A comparison of methods for the
assessment of postural load and duration of computer use. Occup. Environ.
Med. 61 (12), 1027e1031.
Hengel, K.M.O., Visser, B., Sluiter, J.K., 2011. The prevalence and incidence of
musculoskeletal symptoms among hospital physicians: a systematic review. Int.
Arch. Occup. Environ. Health 84 (2), 115e119.
Higgins, J., Green, S., 2011. Cochrane Handbook for Systematic Reviews of In-
terventions Version 5.1.0. The Cochrane Collaboration. Available at: http://
handbook.cochrane.org/ (accessed 22.04.16.).
Hogg-Johnson, S., van der Velde, G., Carroll, L.J., Holm, L.W., Cassidy, J.D.,
Guzman, J.D., Co " te
", P., Haldeman, S., Ammendolia, C., Carragee, E., Hurwitz, E.,
Nordin, M., Peloso, P., 2008. The burden and determinants of neck pain in the
general population: results of the bone and joint decade 2000-2010 task force
on neck pain and its associated disorders. Spine (Phila Pa 1976) 33 (4 Suppl.),
S39eS51.
Homan, M.M., Armstrong, T.J., 2003. Evaluation of three methodologies for
assessing work activity during computer use. AIHA J. (Fairfax, Va) 64 (1), 48e55.
Hoy, D., Protani, M., De, R., Buchbinder, R., 2010. The epidemiology of neck pain.
Best. Pract. Res. Clin. Rheumatol. 24 (6), 783e792.
Hoy, D., Brooks, P., Woolf, A., Blyth, F., March, L., Bain, C., Baker, P., Smith, E.,
Buchbinder, R., 2012. Assessing risk of bias in prevalence studies: modification
of an existing tool and evidence of interrater agreement. J. Clin. Epidemiol. 65
(9), 934e939.
Huisstede, B.M., Bierma-Zeinstra, S.M., Koes, B.W., Verhaar, J.A., 2006. Incidence and
prevalence of upper-extremity musculoskeletal disorders. A systematic
appraisal of the literature. BMC Musculoskelet. Disord. 31, 7.
IJmker, S., Leijssen, J.N., Blatten, B.M., van der Beek, A.J., van Mechelen, W.,
Bongers, P.M., 2008. Test-retest reliability and validity of self-reported duration
of computer use at work. Scand. J. Work Environ. Health 34 (2), 113e119.
Kietrys, D.M., Gerg, M.J., Dropkin, J., Gold, J.E., 2015. Mobile input device type,
texting style and screen size influence upper extremity and trapezius muscle
activity, and cervical posture while texting. Appl. Ergon. 50, 98e104.
Kim, M.S., 2015. Influence of neck pain on cervical movement in the sagittal plane
during smartphone use. J. Phys. Ther. Sci. 27 (1), 15e17.
Kim, H.J., Kim, J.S., 2015. The relationship between smartphone use and subjective
musculoskeletal symptoms and university students. J. Phys. Ther. Sci. 27 (3),
575e579.
Larsson, B., Søgaard, K., Rosendal, L., 2007. Work related neck-shoulder pain: a re-
view on magnitude, risk factors, biochemical characteristics, clinical picture and
preventive interventions. Best. Pract. Res. Clin. Rheumatol. 21 (3), 447e463.
Lui, D.P., Szeto, G.P., Jones, A.Y., 2011. The pattern of electronic game use and related
bodily discomfort in Hong Kong primary school children. Comput. Educ. 57 (2),
1665e1674.
Luime, J.J., Koes, B.W., Hendriksen, I.J., Burdorf, A., Verhagen, A.P., Miedema, H.S.,
Berhaar, J., 2004. Prevalence and incidence of shoulder pain in the general
population; a systematic review. Scand. J. Rheumatol. 33 (2), 73e81.
Malin " ska, M., Bugajska, J., 2010. The influence of occupational and non-occupational
factors on the prevalence of musculoskeletal complaints in users of portable
computers. Int. J. Occup. Saf. Ergon. 16 (3), 337e343.
Mediouni, Z., de Roquemaurel, A., Dumontier, C., Becour, B., Garrabe, H.,
Roquelaure, Y., Descatha, A., 2014. Is carpal tunnel syndrome related to com-
puter exposure at work? a review and meta-analysis. J. Occup. Environ. Med. 56
(2), 204e208.
Ning, X., Huang, Y., Hu, B., Nimbarte, A.D., 2015. Neck kinematics and muscle ac-
tivity during mobile device operations. Int. J. Ind. Ergon. 48, 10e15.
Perestelo-Pe "rez, L., 2013. Standards on how to develop and report systematic re-
views in psychology and health. Int. J. Clin. Health Psychol. 13 (1), 49e57.
Ro€o
€sli, M., Frei, P., Mohler, E., Hug, K., 2010. Systematic review on the health effects
142 Y. Xie et al. / Applied Ergonomics 59 (2017) 132e142
of exposure to radiofrequency electromagnetic fields from mobile phone base
stations. Bull. World Health Organ 88 (12), 887e896.
Schisterman, E.F., Cole, S.R., Platt, R.W., 2009. Overadjustment bias and unnecessary
adjustment in epidemiological studies. Epidemiology 20 (4), 488e495.
Scottish Intercollegiate Guidelines Network, 2014. Available at: http://www.sign.ac.
uk/guidelines/fulltext/50/ (accessed 22.04.16).
Shan, Z., Deng, G., Li, J., Li, Y., Zhang, Y., Zhao, Q., 2013. Correlational analysis of neck/
shoulder pain and low back pain with the use of digital products, physical ac-
tivity and psychological status among adolescents in Shanghai. PLoS One 8 (10),
e78109.
Stalin, P., Abraham, S.B., Kanimozhy, K., Prasad, R.V., Singh, Z., Purty, A.J., 2016.
Mobile phone usage and its health effects among adults in a semi-urban area of
southern India. J. Clin. Diagn. Res. 10 (1), LC14eLC16.
Szeto, G.Y.P., Straker, L.M., O'Sullivan, P.B., 2005. A comparison of symptomatic and
asymptomatic office workers performing monotonous keyboard workd1: neck
and shoulder muscle recruitment patterns. Man. Ther. 10 (4), 270e280.
Terracciano, L., Brozek, J., Compalati, E., Schunemann, H., 2010. GRADE system: new
paradigm. Curr. Opin. Allergy Clin. Immunol. 10 (4), 377e383.
Thomsen, J.F., Gerr, F., Atroshi, I., 2008. Carpal tunnel syndrome and the use of
computer mouse and keyboard: a systematic review. BMC Musculoskelet.
Disord. 9, 134.
Valentini, E., Curcio, G., Moroni, F., Ferrara, M., De Gennaro, L., Bertini, M., 2007.
Neurophysiological effects of mobile phone electromagnetic fields on humans:
a comprehensive review. Bioelectromagnetics 28 (6), 415e432.
Valentini, E., Ferrara, M., Presaghi, F., De Gennaro, L., Curcio, G., 2010. Systematic
review and meta-analysis of psychomotor effects of mobile phone electro-
magnetic fields. Occup. Environ. Med. 67 (10), 708e716.
van den Heuvel, S.G., de Looze, M.P., Hildebrandt, V.H., The ", K.H., 2003. Effects of
software programs stimulating regular breaks and exercises on work-related
neck and upper-limb disorders. Scand. J. Work Environ. Health 29 (2), 106e116.
van der Windt, D.A., Thomas, E., Pope, D.P., de Winter, A.F., Macfarlane, G.J.,
Bouter, L.M., Silman, A.J., 2000. Occupational risk factors for shoulder pain: a
systematic review. Occup. Environ. Med. 57 (7), 433e442.
van Rijn, R.M., Huisstede, B.M., Koes, B.W., Burdorf, A., 2010. Associations between
work-related factors and specific disorders of the shoulderda systematic re-
view of the literature. Scand. J. Work Environ. Health 36 (3), 189e201.
Vasavada, A.N., Nevins, D.D., Monda, S.M., Hughes, E., Lin, D.C., 2015. Gravitational
demand on the neck musculature during tablet computer use. Ergonomics 58
(6), 990e1004.
Waersted, M., Hanvold, T.N., Veiersted, K.B., 2010. Computer work and musculo-
skeletal disorders of the neck and upper extremity: a systematic review. BMC
Musculoskelet. Disord. 11, 79.
Walker, B.F., 2000. The prevalence of low back pain: a systematic review of the
literature from 1966 to 1998. J. Spinal Disord. 13 (3), 205e217.
Williams, I.W., Kennedy, B.S., 2011. Texting tendinitis in a teenager. J. Fam. Pract. 60
(2), 66e67.
Xie, Y., Szeto, G.P., Dai, J., Madeleine, P., 2016. A comparison of muscle activity in
using touchscreen smartphone among young people with and without chronic
neck-shoulder pain. Ergonomics 59 (1), 61e72.