J Phys Fitness Sports Med, 4(2): 187-195 (2015)

DOI: 10.7600/jpfsm.4.187

JPFSM: Review Article

Physical activity and lifestyle intervention

Masato Nishiwaki1* and Naoyuki Matsumoto2
1
Faculty of Engineering, Osaka Institute of Technology, 5-16-1 Omiya, Asahi-ku, Osaka 535-8585, Japan
2
Faculty of Environmental Symbiotic Sciences, Prefectural University of Kumamoto, 3-1-100 Tsukide, Higashi-ku, Kumamoto
862-8502, Japan

Received: March 31, 2015 / Accepted: April 24, 2015

Abstract This review summarizes recent findings regarding the status of physical activity
and introduces effective methods of intervention. Data from our serial cross-sectional study
between 2003 and 2012 suggested that the decline in step counts over the last decade is mainly
related to a reduction in non-exercise activity associated with the increased use of cell phones
or computers and with playing video games. We then examined the effects of lifestyle inter-
ventions using an activity monitor with computerized game functions or an activity monitor
and Twitter on physical activity and body composition. These findings suggested that lifestyle
interventions using both of these strategies increases daily physical activity and reduces body
fat more effectively than using an activity monitor alone. In addition, changes in physical ac-
tivity and in body fat were significantly correlated. We also applied a randomized intervention
to examine an effective method of increasing physical activity levels among college physical
education students using a pedometer. We found that using a pedometer and inducing friendly
competition or encouragement from peers increased step counts more effectively during soccer
classes. These findings therefore have important implications for ensuring compliance with the
Physical Activity Reference for Health Promotion 2013 and Active Guide.
Keywords : activity monitor, health, cell phone, personal computer, physical education class,
weight loss

steps/day between 1997 and 20097). A final evaluation by

Introduction
Healthy Japan 21 also found a reduction in steps over the
Higher levels of physical activity (PA) and exercise are last decade8). If energy intake remains the same, a reduc-
associated with lower risk of non-communicable diseases tion of 1000 steps/day will induce a 1.0 kg increase in
(NCD), such as cardiovascular diseases and cancer1,2). body weight in a single year9). Because physical inactiv-
One study has found that insufficient PA and exercise is ity is the fourth leading cause of death worldwide10), the
the 3rd risk factor for mortality due to NCD in Japan, after decrease in PA is often considered a global pandemic11).
smoking and hypertension3). Thus, physical inactivity is Thus, it is essential to increase step counts or PA if the
a major risk factor for NCD4). Increasing PA and exercise risk for NCD is to be reduced.
is of primary importance to reduce the risk of mortality The cause of the decline in step counts during the last
due to NCD. This review summarizes recent findings re- decade has not been precisely defined. Physical activity
garding the status of PA in Japan and introduces effective can be regarded as consisting of exercise and non-exercise
interventions to increase levels of PA. activities12,13). Thus, we examined whether the decline in
step counts can be attributed to a reduction in the amount
of exercise or non-exercise activities in a serial cross-
Insufficient physical activity
sectional study of male college freshmen between 2003
A previous study reported that step counts is correlated and 2012 at the Osaka Institute of Technology9).
with the amount of PA performed at an intensity of ≥ 3 Although the characteristics of the participants and PA
metabolic equivalents (METs)5), and thus, step counts levels did not significantly differ, trends for average and
are widely applied as an objective index of moderate weekend step counts significantly decreased throughout
and vigorous PA1,6). The National Health and Nutrition the decade (Fig. 1A, B, and C). Furthermore, analyses of
Survey has assessed the annual number of step counts of daily behavior records indicated that weekday and week-
Japanese people since 1989. The results of the assessment end cell phone or computer usage significantly increased
showed that step counts among Japanese fell by 1000 and that the amount of weekday time spent playing video
games also significantly increased (Fig. 1D and E). No-
*Correspondence: masato.nishiwaki@oit.ac.jp tably, persons with low step counts used a cell phone or
188 JPFSM : Nishiwaki M and Matsumoto N

A Average
D

Weekday

Weekend

Playing video games (%)


Trend P < 0.05




Steps/day







 


 
 

 























 




 
*
















 




 








B Weekday

E

Cell phone or PC usage (%)


Trend P = 0.164







Steps/day

*

**
†










 


 





























 


















 



 


2003-2004
2007-2008
2011-2012

C Weekend


Trend P < 0.01



Steps/day





 
 
 
 

 



 
 


 












 
































Fig. 1 Changes in daily step counts and daily behavior records between 2003 and 2012.
Average (A), weekday (B), and weekend (C) steps, playing video games per waking hours (D), and cell phone or computer us-
age during waking hours (E). PC, personal computer. Data are shown as means ± SEM. *P < 0.05 and **P < 0.01 vs. within the
same category in 2003-2004. †P < 0.05 vs. within the same category in 2007-2008.
Average and weekend step counts significantly decreased, and cell phone or computer usage significantly increased during
weekdays and weekends between 2011 and 2012 (Nishiwaki et al., 2014)9).

computer and played video games more frequently than findings indicate that declining step counts, especially
those with high step counts (Fig. 2A and B). Another among youth, are mainly related to a reduction in non-
recent study found that cell phone usage can disrupt PA exercise activity associated with increased cell phone or
and reduce cardiorespiratory fitness determined as peak computer usage and playing video games. Thus, increases
oxygen consumption14), and that reductions in step counts in daily PA habits17), improvements in PA modes18) and
are associated with increasing Internet addiction levels in better city planning that includes environmental improve-
a dose-response manner (Fig. 2C)15). Screen time is also ments 19,20) are needed to increase the amount of non-
inversely associated with isometric trunk muscle strength exercise activities.  



independently of cardiorespiratory fitness and other con-
founding factors in youth16). Thus, technological devel-
Monitoring physical activity and behavioral modification
opments such as the Internet and cell phones can reduce
daily PA since less energy is needed for non-exercise ac- Pedometers and activity monitors serve as surveillance
tivities. Although current evidence remains limited, these tools to assess step counts or the amounts of PA. How-
 JPFSM : Physical activity and intervention 189

A

High steps group
C

Trend P < 0.05

Playing video games (%)




Low steps group




P < 0.05

Steps/day































 

 













B






Cell phone or PC usage (%)


Low
High

Internet addiction



Fig. 2 Comparisons of daily behavior records.



P < 0.05
Daily amounts of time playing video games (A) and daily
cell phone or computer usage (B) during waking hours.

Step counts compared with Internet addiction levels among
groups I to V (C). PC, personal computer. Data are shown



as means ± SEM.






The group with more steps spends significantly less time



using cell phones and computers. The degree of Internet





addiction might be proportionally associated with reduced
Weekday
Weekend
step counts (Nishiwaki et al., 2014)9,15).

ever, a meta-analysis of eight randomized controlled trials reported that activity monitors and rewards such as being
and 18 observational studies found a significant increase able to watch television induce increases in PA among
of 2000 to 2500 steps/day among pedometer users com- children 27). Because participants become absorbed in
pared with control participants or baseline counts21). These playing interactive video games and perceive them as en-
findings suggested that the immediate feedback provided joyable, such games can serve as an additional interven-
by the activity monitor or pedometer is an important mo- tion to encourage compliance with exercise or rehabilita-
tivational feature that serves as a behavioral modification tion programs28-32). Therefore, wearing an activity monitor
tool. Clemes et al. have also found a significant mean in- as a motivational tool and incorporating behavior-based
crease of 1000 to 1800 steps/day when participants wear rewards or a computerized game element might have a
an unsealed pedometer and keep a personal activity log of collective effect on increasing daily PA. Thus, our pilot
their daily step counts, compared with wearing a covert crossover study analyzed the effects of a short-term life-
(uninformed device and measured values) or a sealed (un- style intervention using an activity monitor with comput-
informed measured values) pedometer22,23). That is, being erized game functions on PA33).
aware of wearing an activity monitor or pedometer and Healthy volunteers who participated in the 12-week
confirming the measured values can induce an increase crossover study were randomly assigned to either group
in daily PA or step counts. Therefore, in addition to being A (six-week game intervention followed by a six-week
a surveillance tool, pedometers and activity monitors can normal intervention) or group B (six-week normal inter-
serve as motivational devices that can induce behavioral vention followed by a six-week game intervention). The
modifications. Indeed, many studies have indicated that participants wore both a Lifecorder EX standard activity
lifestyle interventions using activity monitors or pedom-  



monitor (Suzuken Co., Nagoya, Japan) and the Yuuhokei
eters can increase regular PA and help to prevent the onset activity monitor with computerized game functions (Ban-
of lifestyle-related diseases24-26). dai Co. Ltd., Tokyo, Japan) during the game intervention
and only the standard activity monitor during the normal
intervention.
Effects of an activity monitor with computerized game
Fig. 3A and B shows that significantly more daily steps
functions
were recorded and significantly more PA performed at an
We aimed to determine more effective methods of using intensity of ≥ 3 METs for the game, than for the normal
activity monitors or pedometers as follows. Others have intervention (both P < 0.01). The mean differences in
190 JPFSM : Nishiwaki M and Matsumoto N

steps and the amount of PA between the game and normal of daily PA levels might be limited. If the goal is 10000
interventions were 1673 ± 1738 steps (21.9 ± 24.0%) steps/day, then the target value for each game to proceed
and 0.7 ± 0.8 METs (33.6 ± 38.2%), respectively. The to the next scene should be set at 8000 steps, which ap-
Yuuhokei, which was used as the activity monitor with proximately corresponds to 10000 steps counted by the
computerized game functions in this study, is visually Lifecorder EX.
simple to understand and displays step counts until a goal
is achieved. Moreover, the game characters in Yuuhokei
Effects of an activity monitor and Twitter
usually encourage the users to achieve step-count mile-
stones or physical activity levels. Because the games in- One study has found that exercising with a virtual part-
cluded in Yuuhokei are based on stories and tales that are ner through a brief video chat via services such as Skype
very famous in Japan, such as Space Battleship Yamato, can improve aerobic exercise performance across multiple
the Rose of Versailles, and Section Chief Kousaku Shima, sessions38). These findings suggested that friendly compe-
participants easily became absorbed in increasing PA to tition or encouragement from online partners can increase
clear the game. Collectively, the game story, scenes, and motivation in both exercise and non-exercise activities,
characters might help to motivate individuals to walk or thereby increasing daily PA. Although recent reports have
be active and make them even more aware of daily PA described a web-based PA diary39-41), not everyone can
levels and step goals. Therefore, these findings suggest utilize large-scale systems. Thus, we considered social
that short-term interventions using an activity monitor networks such as Twitter, which enables users to send and
with computerized game functions increases PA more ef- read short 140-character messages (tweets) that an indefi-
fectively than those using a standard activity monitor. nite number of individuals can easily and immediately
However, despite a significant correlation between step share or exchange via tweets, retweets, and timelines. In
counts obtained from the Yuuhokei and the Lifecorder EX addition to being aware of wearing the monitor and con-
(r = 0.92), significantly fewer steps were counted by the firming measured values, adding Twitter might promote
former than the latter (7165 ± 4284 steps/day vs. 10132 daily PA or reduce sedentary behavior through friendly
± 4841 steps/day, P < 0.01)34). The accuracy of the step competition with, and/or encouragement from online part-
counts detected by Lifecorder EX is calibrated during the ners. Thus, we investigated the effects of a randomized
manufacturing process according to the Japanese Indus- lifestyle intervention using an activity monitor and Twit-
trial Standards (JIS), and the widely accepted error during ter on PA42).
walking on a treadmill is within 1%35-37). Therefore, these Healthy participants were randomly assigned to either
results infer that although relative changes in daily indi- a Normal intervention group that wore a Lifecorder EX
vidual step counts determined from the Yuuhokei might activity monitor or a Twitter intervention group that wore
help to evaluate daily PA, the application of absolute step the same activity monitor and tweeted about their daily
counts obtained from this device as a proxy assessment steps or PA for six weeks. An observer read the tweets
Physical activity (METs h/day)

A
B





Steps (Steps/day)


































Normal
Game
Normal
Game

Fig. 3 Comparisons of physical activity between Normal and Game interventions.

Comparisons of daily steps (A) and of amounts of daily physical activity performed at intensity of ≥ 3 METs (B)
for each individual value. Normal, normal intervention period; Game, intervention period using activity monitor
with computerized game functions; broken lines, Group A (six-week game intervention followed by six-week nor-
mal intervention); solid lines, Group B (six-week normal intervention followed by six-week game intervention).
Significantly more daily steps and physical activity were recorded for the game intervention than the normal inter-
vention (Nishiwaki et al., 2012)33).
 JPFSM : Physical activity and intervention 191

from each participant and commented about PA. The pro- perimental period in the Twitter group (Fig. 4C). Friendly
tect mode was configured for each account to safeguard competition or encouragement from online partners such
the data of each participant. as other participants in the Twitter group in addition to be-
Daily PA (step counts and amount of PA) did not sig- ing aware of wearing an activity monitor and confirming
nificantly differ at week one between the groups and did the measured values might help to motivate participants
not significantly differ over the six-week experimental to become more active and cause them to become even
period in the Normal group. In contrast, step counts (Fig. more conscious of daily PA levels and step goals. There-
4A) and the amount of PA (Fig. 4B) gradually increased fore, these data indicated that daily PA increased more
from weeks one to six in the Twitter group from 8542 ± effectively when a lifestyle intervention comprised an ac-
3158 to 12700 ± 3935 steps/day and from 2.5 ± 1.2 to 4.6 tivity monitor and Twitter rather than an activity monitor
± 2.3 METs·h/day (both P < 0.01). Moreover, step counts alone.
significantly declined immediately after the end of the ex-

A
B
: Normal group

Physical activity (METs·h/day)





: Twitter group





Steps (steps/day)





 







** ** 

**



 ** ** **
 * *







 









 
1
2
3
4
5
6
(week)
1
2
3
4
5
6
(week)

Interaction F = 3.642, P < 0.01
Interaction F = 4.133, P < 0.01

C


: Normal group

: Twitter group

Steps (steps/day)

 **
**

**
**

**

**

**
: Twitter-stop group

**

**

















1
2
6
7
8
9
10
11
12
13
14
(week)

Fig. 4 Comparisons of time-course changes in physical activity during Twitter intervention.

Comparisons of time-course changes in step counts (A) and physical activity at an intensity of ≥ 3 METs (B). *P < 0.05 and **P
< 0.01 vs. week 1; †P < 0.05 and ††P < 0.01 vs. the Normal group. Data are shown as means ± SD in A and B.
Comparisons of time-course changes in steps in the Normal intervention group (n = 5), Twitter intervention group (n = 5), and
Twitter-stop group (n = 5) (C). Twitter-stop group, Twitter usage was stopped after the week 6 and the participants only wore the
activity monitor. **P < 0.01 vs. week 1; ††P < 0.01 vs. week 6. Data are shown as means ± SEM in C.
Daily physical activity in the Twitter group gradually increased from week 1 to week 6, but not in the Normal group. In addition,
step counts significantly declined immediately after the end of Twitter usage (Nishiwaki et al., 2013 and unpublished observa-
tion)42).
192 JPFSM : Nishiwaki M and Matsumoto N

pended 11020.6 ± 1364.2 kcal (amount of PA at intensity

Increases in physical activity and physiological param-
of ≥ 3 METs (METs·h) × body weight (kg) × 1.05 × pe-
eters
riod of six weeks). A 1.0 kg reduction in body fat gener-
Although epidemiological studies indicated the impor- ally requires the expenditure of 7000 to 8000 kcal32). Be-
tance of PA for health3,10), whether lifestyle intervention cause the Twitter group lost 1.1 ± 0.2 kg of body fat, the
can actually induce an improvement in physiological degree of fat reduction was approximately consistent with
parameters such as body composition or blood pressure the amount of PA in which the Twitter group performed.
(BP) remains unclear. A recent study found that in gen- Thus, our results indicated that more body fat can be lost
eral, an increase in PA is proportionally associated with a during lifestyle interventions associated with increases in
reduction in body fat within 16 weeks43). Therefore, if a PA levels.
lifestyle intervention using an activity monitor with com- Our data indicated that diastolic BP, mean BP, pulse
puterized game functions or an activity monitor and Twit- pressure, and pulse rate did not significantly change after
ter could increase the daily PA, then body weight and fat six weeks of interventions. However, systolic BP was
should become reduced and changes in body composition slightly, but significantly, reduced after the interventions
and PA should correlate. Thus, we examined the effects of (P < 0.05)44); it was reduced in not only interventions us-
increasing PA via a lifestyle intervention on body compo- ing Game and Twitter, but also in those using a normal
sition or BP44). activity monitor. Our results support previous findings
Six weeks of intervention using an activity monitor with that walking induces a reduction in BP45). Therefore, these
computerized game functions significantly reduced body findings suggest that short-term lifestyle intervention us-
weight, body mass index, and body fat determined using ing an activity monitor induced a reduction in systolic BP.
the impedance method (P < 0.01). Significantly more fat
was also lost in a Game, than in a Normal intervention
Application of a pedometer to physical education class
(P < 0.05)44) and body fat and waist circumference were
or sports activity
significantly reduced after an intervention using an activ-
ity monitor and Twitter. In fact, significantly more body Understanding that exercise is also important in in-
fat was lost after the Twitter, than the Normal intervention creasing the amount of PA, we performed a randomized
(-1.1 ± 0.2 kg vs. -0.1 ± 0.3 kg; P < 0.05) (Fig. 5A), and intervention study to determine a method of effectively
changes in PA and body fat significantly correlated (r = increasing PA levels during college physical education
-0.713, P < 0.05; Fig. 5B) (Nishiwaki et al. under review). (PE) classes using a pedometer46).
Although baseline PA and dietary intake were unclear in We randomly assigned 159 male college freshmen stu-
this population, the participants in the Twitter group ex- dents to a Control or an Intervention group based on af-

A
Normal Twitter
B


r = -0.713, P = 0.031
y = -0.2059x -0.5727

Change in body fat (kg)



Change in body fat (kg)






-0.1 -1.1
± ± 




0.3
0.2



















P = 0.021

Change in amount of PA (METs h/day)

Fig. 5 Changes in body fat and amounts of physical activity.

Comparisons of changes in body fat (A). Relationship between changes in physical activity and body fat in the Twitter group (B).
Normal, normal intervention; Twitter, Twitter intervention; Amount of PA, amount of physical activity performed at intensity of
≥ 3 METs. Data are shown as means ± SEM.
Body fat decreased significantly more in the Twitter group than in the Normal group, and changes in physical activity and body
fat significantly correlate (Nishiwaki et al., under review).
 JPFSM : Physical activity and intervention 193

filiation courses. Both groups participated in four 90-min from baseline to fourth class for the Intervention group
PE classes and initially wore a sealed (uninformed mea- compared with the Control group and no significant dif-
sured values) pedometer to assess baseline step counts ference between the Control and the Half-intervention
at the first class. The Control group continued to wear a groups. Thus, in addition to being aware of wearing a
sealed pedometer for the next three classes, whereas the pedometer and confirming measured values, friendly
Intervention group wore an unsealed pedometer (informed competition or encouragement from other students tak-
measured values) and recorded their step counts in a team ing the same class might cause participants to become
activity log. Victory and defeat were decided by both the more active. Such PE classes using a pedometer might
step counts of team members and the soccer game scores. also positively affect game and point scores in college PE
Thus, the students in the Intervention group were encour- classes47). Collectively, our findings suggest that wear-
aged to increase their step counts. ing pedometers can increase step counts more effectively
Because we excluded all data missing values from sta- during PE classes when the players participate in friendly
tistical analyses, data for the remaining 43 (the Control competition or are encouraged by other students.
group) or 62 (the Intervention group) participants were
analyzed. Baseline step counts per class did not signifi-
Summary
cantly differ between the groups, and the step counts did
not significantly change over time in the Control group. A lack of physical activity is often considered as a glob-
However, step counts for the Intervention group gradu- al pandemic and the reductions in step counts or PA can-
ally increased between baseline and the fourth class (4502 not be ignored to reduce the risk for NCD. Some evidence
± 123 steps/class vs. 5539 ± 119 steps/class; P < 0.01; indicates that the decline in step counts over the last
Fig. 6A). Furthermore, we analyzed the effects of a half- decade is mainly related to a reduction in non-exercise
intervention on step counts. The participants in this half- activity associated with increased cell phone or computer
intervention (n = 51) wore an unsealed pedometer and re- usage and playing video games. Therefore, pedometers or
corded their step counts in an individual activity log, and activity monitors that can induce behavioral modifications
did not compete against other teams for step counts. Fig. should be used as motivational devices to increase step
6B shows a significantly greater change in step counts counts or PA. Our findings suggest that lifestyle interven-

A
B

Change in steps from baseline (%)

Steps/Physical education class


: Intervention group
P < 0.01

** 

: Normal group





** 

††

b


a


21.2 26.7


± ±
5.3
4.0


7.2
±



4.2

Baseline
1
2
3
Normal Half-
(class)
Intervention Intervention

Interaction F = 3.493, P < 0.05

Fig. 6 Comparisons of step counts during 90-min college physical education (soccer) class.
Comparisons of changes in step counts over time between the Intervention and Normal groups (A) and changes in step counts
from baseline to fourth class (B). Baseline step counts per class were determined from sealed (uninformed measured values)
pedometers during the first class. The Control group continued wearing the sealed pedometer. The Intervention group wore an
unsealed pedometer (informed measured values), recorded their individual step counts in a team activity log, and competed for
step counts against other teams. The Half-intervention group wore an unsealed pedometer, recorded their individual step counts
in a personal activity log, and did not compete for step counts against other teams. **P < 0.01 vs. baseline. ††P < 0.01, aP = 0.0569,
and bP = 0.0670 vs. the Normal group. Data are shown as means ± SEM.
The step counts gradually increased in the Intervention group from baseline to the fourth class.
(Nishiwaki et al. 2014 and unpublished observation)46).
194 JPFSM : Nishiwaki M and Matsumoto N

tions using an activity monitor with computerized game global action for public health. Lancet 380: 294-305.
functions or an activity monitor and Twitter increase daily 12) The Office for Lifestyle-Related Diseases Control, General
PA and reduce body fat more effectively than an activity Affairs Division, Health Service Bureau, Ministry of Health,
monitor alone. In addition, using pedometers along with Labour and Welfare of Japan. 2006. Exercise and Physical
Activity Guide for Health Promotion 2006 - To Prevent Life-
friendly competition or encouragement from peers during
style-related Diseases - (in Japanese).
PE activities might increase step counts among college
13) The Office for Lifestyle-Related Diseases Control, General
students more effectively than participating in such activi- Affairs Division, Health Service Bureau, Ministry of Health,
ties without these stimuli. These findings therefore have Labour and Welfare of Japan. 2006. Exercise and Physical
important implications for ensuring compliance with the Activity Reference for Health Promotion 2006: Physical Ac-
Physical Activity Reference for Health Promotion 2013 tivity, Exercise, and Physical Fitness (in Japanese).
and Active Guide48,49). 14) Lepp A, Barkley JE, Sanders GJ, Rebold M and Gates P.
2013. The relationship between cell phone use, physical and
sedentary activity, and cardiorespiratory fitness in a sample
Conflict of Interests of U.S. college students. Int J Behav Nutr Phys Act 10: 79.
The authors declare that there is no conflict of interests 15) Nishiwaki M, Kiuchi A and Nakamura T. 2014. The rela-
tionship between Internet addiction and steps -A cross-sec-
regarding the publication of this article.
tional study in a sample of male college freshmen in Japan-.
Tairyoku Kagaku (Jpn J Phys Fitness Sports Med) 63: 445-
References 453 (in Japanese).
16) Grøntved A, Ried-Larsen M, Froberg K, Wedderkopp N,
  1) Miyachi M. 2012. Measures of physical activity and exercise Brage S, Kristensen PL, Andersen LB and Møller NC. 2013.
for health promotion by the Ministry of Health, Labour and Screen time viewing behaviors and isometric trunk muscle
Welfare. J Phys Fitness Sports Med 1: 467-472. strength in youth. Med Sci Sports Exerc 45: 1975-1980.
  2) Sawada SS. 2014. Physical fitness for health. J Phys Fitness 17) Oka K, Shibata A and Ishii K. 2014. Association of dog own-
Sports Med 3: 377-384. ership and dog walking with human physical activity. J Phys
  3) Ikeda N, Inoue M, Iso H, Ikeda S, Satoh T, Noda M, Mizoue Fitness Sports Med 3: 291-295.
T, Imano H, Saito E, Katanoda K, Sobue T, Tsugane S, 18) Miyaji A, Nishiwaki M, Hamada Y and Matsumoto N. 2013.
Naghavi M, Ezzati M and Shibuya K. 2012. Adult mortality An examination of a new walking method to fulfill the Physi-
attributable to preventable risk factors for non-communicable cal Activity Reference in a shorter time. Tairyoku Kagaku
diseases and injuries in Japan: a comparative risk assessment. (Jpn J Phys Fitness Sports Med) 62: 383-390 (in Japanese).
PLoS Med 9: e1001160. 19) Iwasa T, Takamiya T, Ohya Y, Odagiri Y, Kikuchi H, Fuku-
  4) Pratt M, Sarmiento OL, Montes F, Ogilvie D, Marcus BH, shima N, Oka K, Kitabatake Y, Shimomitsu T and Inoue S.
Perez LG, Brownson RC and Lancet Physical Activity Series 2015. Differences in the physical activity patterns among
Working G. 2012. The implications of megatrends in infor- young old adults by three residential locations in Japan.
mation and communication technology and transportation for Tairyoku Kagaku (Jpn J Phys Fitness Sports Med) 64: 145-
changes in global physical activity. Lancet 380: 282-293. 154 (in Japanese).
 5) Oshima Y, Hikihara Y, Ohkawara K, Takata K, Miyake 20) Matsumoto N, Kurobe K, Nishiwaki M and Hayashi N. 2013.
R, Ebine N, Tabata I and Tanaka S. 2012. Daily steps cor- Making physical activity maps for promoting a non-exercise
responding to the reference quantity of physical activity of activity and its utilization for regional development. SSF
Exercise and Physical Activity Reference for Health Promo- Journal of Sport for Everyone 2: 72-78 (in Japanese).
tion 2006 (EPAR2006) assessed by accelerometer. Tairyoku 21) Bravata DM, Smith-Spangler C, Sundaram V, Gienger AL,
Kagaku (Jpn J Phys Fitness Sports Med) 61: 193-199 (in Lin N, Lewis R, Stave CD, Olkin I and Sirard JR. 2007. Us-
Japanese). ing pedometers to increase physical activity and improve
  6) Tanaka S. 2012. Status of physical activity in the Japanese health: a systematic review. JAMA 298: 2296-2304.
population. J Phys Fitness Sports Med 1: 491-497. 22) Clemes SA, Matchett N and Wane SL. 2008. Reactivity: an
  7) The Japan Ministry of Health, Labour and Welfare. 2014. The issue for short-term pedometer studies? Br J Sports Med 42:
National Health and Nutrition Survey Japan in 2012 (in Japa- 68-70.
nese). 23) Clemes SA and Parker RA. 2009. Increasing our understand-
 8) The Japan Ministry of Health, Labour and Welfare. 2011. ing of reactivity to pedometers in adults. Med Sci Sports Ex-
“Healthy Japan 21” final evaluation (in Japanese). erc 41: 674-680.
  9) Nishiwaki M, Kiuchi A and Nakamura T. 2014. The exami- 24) Iwane M, Arita M, Tomimoto S, Satani O, Matsumoto M,
nation of a decline in steps in the last decade and its reason Miyashita K and Nishio I. 2000. Walking 10,000 steps/day or
-A serial cross-sectional study in male college freshmen-. more reduces blood pressure and sympathetic nerve activity
Tairyoku Kagaku (Jpn J Phys Fitness Sports Med) 63: 231- in mild essential hypertension. Hypertens Res 23: 573-580.
242 (in Japanese). 25) Swartz AM, Strath SJ, Bassett DR, Moore JB, Redwine BA,
10) WHO. 2010. Global Recommendations on Physical Activity Groer M and Thompson DL. 2003. Increasing daily walking
for Health. improves glucose tolerance in overweight women. Prev Med
11) Kohl HW 3rd, Craig CL, Lambert EV, Inoue S, Alkandari JR, 37: 356-362.
Leetongin G, Kahlmeier S and Lancet Physical Activity Se- 26) Tudor-Locke C, Bell RC, Myers AM, Harris SB, Ecclestone
ries Working G. 2012. The pandemic of physical inactivity: NA, Lauzon N and Rodger NW. 2004. Controlled outcome
 JPFSM : Physical activity and intervention 195

evaluation of the First Step Program: a daily physical activity DL. 2012. Aerobic exercise is promoted when individual per-
intervention for individuals with type II diabetes. Int J Obes formance affects the group: A test of the Kohler motivation
Relat Metab Disord 28: 113-119. gain effect. Ann Behav Med 44: 151-159.
27) Roemmich JN, Gurgol CM and Epstein LH. 2004. Open-loop 39) Barwais FA, Cuddihy TF and Tomson LM. 2013. Physi-
feedback increases physical activity of youth. Med Sci Sports cal activity, sedentary behavior and total wellness changes
Exerc 36: 668-673. among sedentary adults: a 4-week randomized controlled
28) Agmon M, Perry CK, Phelan E, Demiris G and Nguyen HQ. trial. Health Qual Life Outcomes 11: 183.
2011. A pilot study of Wii Fit exergames to improve balance 40) Gotsis M, Wang H, Spruijt-Metz D, Jordan-Marsh M and Va-
in older adults. J Geriatr Phys Ther 34: 161-167. lente TW. 2013. Wellness partners: design and evaluation of
29) Biddiss E and Irwin J. 2010. Active video games to promote a web-based physical activity diary with social gaming fea-
physical activity in children and youth: a systematic review. tures for adults. JMIR Res Protoc 2: e10.
Arch Pediatr Adolesc Med 164: 664-672. 41) Nose H, Morikawa M, Yamazaki T, Nemoto K, Okazaki K,
30) LeBlanc AG, Chaput JP, McFarlane A, Colley RC, Thivel D, Masuki S, Kamijo Y and Gen-No H. 2009. Beyond epide-
Biddle SJ, Maddison R, Leatherdale ST and Tremblay MS. miology: field studies and the physiology laboratory as the
2013. Active video games and health indicators in children whole world. J Physiol 587: 5569-5575.
and youth: a systematic review. PLoS One 8: e65351. 42) Nishiwaki M, Nakashima N, Ikegami Y, Kawakami R, Ku-
31) Miller CA, Hayes DM, Dye K, Johnson C and Meyers J. robe K and Matsumoto N. 2013. Effects of a lifestyle in-
2012. Using the Nintendo Wii Fit and body weight support tervention by the concurrent use of an activity monitor and
to improve aerobic capacity, balance, gait ability, and fear of Twitter on physical activity -A randomized intervention
falling: two case reports. J Geriatr Phys Ther 35: 95-104. study-. Tairyoku Kagaku (Jpn J Phys Fitness Sports Med) 62:
32) Tripette J, Murakami H, Gando Y, Kawakami R, Sasaki A, 293-302 (in Japanese).
Hanawa S, Hirosako A and Miyachi M. 2014. Home-based 43) Ross R and Janssen I. 2001. Physical activity, total and re-
active video games to promote weight loss during the post- gional obesity: dose-response considerations. Med Sci Sports
partum period. Med Sci Sports Exerc 46: 472-478. Exerc 33: S521-S527; discussion S528-S529.
33) Nishiwaki M, Kuriyama A, Ikegami Y, Nakashima N and 44) Nishiwaki M, Kuriyama A, Ikegami Y, Nakashima N and
Matsumoto N. 2012. Effects of a lifestyle intervention using Matsumoto N. 2014. A pilot crossover study: effects of an
an activity monitor with game functions on physical activ- intervention using an activity monitor with computerized
ity -A randomized crossover study-. Tairyoku Kagaku (Jpn J game functions on physical activity and body composition. J
Phys Fitness Sports Med) 61: 335-341 (in Japanese). Physiol Anthropol 33: 35.
34) Nishiwaki M, Kuriyama A, Ikegami Y, Nakashima N and 45) Nemoto K, Gen-no H, Masuki S, Okazaki K and Nose H.
Matsumoto N. 2012. An examination of steps and physical 2007. Effects of high-intensity interval walking training on
activity amount obtained from a new activity monitor with physical fitness and blood pressure in middle-aged and older
game functions in daily life. J Train Sci Exerc Sport 24: 193- people. Mayo Clin Proc 82: 803-811.
201 (in Japanese). 46) Nishiwaki M, Kiuchi A and Nakamura T. 2014. An examina-
35) Kumahara H, Schutz Y, Ayabe M, Yoshioka M, Yoshitake Y, tion of a method for effectively increasing physical activity
Shindo M, Ishii K and Tanaka H. 2004. The use of uniaxial amount in college physical education class: A randomized
accelerometry for the assessment of physical-activity-related intervention study. Japan Journal for Health, Physical Edu-
energy expenditure: a validation study against whole-body cation, Recreation, and Dance in Universities 11: 21-29 (in
indirect calorimetry. Br J Nutr 91: 235-243. Japanese).
36) Park J, Ishikawa-Takata K, Tanaka S, Mekata Y and Tabata 47) Kamuro R and Nishiwaki M. 2014. Effects of physical ed-
I. 2011. Effects of walking speed and step frequency on es- ucation basketball class using a pedometer on steps, game
timation of physical activity using accelerometers. J Physiol score, and winning point in college students. Japan Journal
Anthropol 30: 119-127. for Health, Physical Education, Recreation, and Dance in
37) Tanaka C and Tanaka S. 2009. Daily physical activity in japa- Universities 11: 31-38 (in Japanese).
nese preschool children evaluated by triaxial accelerometry: 48) The Japan Ministry of Health, Labour and Welfare. 2013.
the relationship between period of engagement in moderate- Physical Activity Reference for Health Promotion 2013 (in
to-vigorous physical activity and daily step counts. J Physiol Japanese).
Anthropol 28: 283-288. 49) The Japan Ministry of Health, Labour and Welfare. 2013. Active
38) Irwin BC, Scorniaenchi J, Kerr NL, Eisenmann JC and Feltz Guide (in Japanese).