Microsoft Word - Shephard_obesity and PA_part 2 June 2019.docx


Health & Fitness Journal 
of Canada 

Copyright © 2019 The Authors. Journal Compilation Copyright © 2019 Health & Fitness Society of BC 
Volume 12  June 30, 2019  Number 2 

	

NARRATIVE	REVIEW	
Can	Regular	Physical	Activity	Prevent	Obesity?	Part	2:	Empirical	cross-
sectional	evidence	of	efficacy.		
Roy	J.	Shephard1,*	
	
1	Faculty	of	Kinesiology	&	Physical	Education,	University	of	Toronto,	ON,	Canada.		
*Corresponding	Author:	royjshep@shaw.ca		
	

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	64 

	
	

	
	

Abstract	

Objective:	The	objective	of	this	narrative	review	is	to	examine	cross-sectional	empirical	data	suggesting	the	
efficacy	of	adequate	daily	physical	activity	in	the	prevention	of	obesity.	Methods:	Information	obtained	from	
Ovid/Medline	and	Google	Scholar	 through	 to	March	2019	was	supplemented	by	a	 search	of	 the	author's	
extensive	personal	files.	Results:	Average	skin-fold	readings	are	often	as	low	as	5	mm,	in	male	elite	long-
distance	 competitors	 who	 spend	 many	 hours	 per	 week	 in	 training,	 and	 values	 for	 female	 endurance	
competitors	are	also	often	only	a	half	of	those	found	in	the	general	female	population.	Physical	education	(PE)	
students	(particularly	men)	are	somewhat	thinner	than	their	peers	in	other	faculties,	but	this	advantage	is	
attenuated	 in	averaged	data	because	a	proportion	of	PE	students	have	a	 lesser	commitment	 to	an	active	
lifestyle,	and	also	some	students	 in	other	 faculties	engage	 in	physically	demanding	 inter-collegiate	sports.	
Early	 occupational	 studies	 showed	 low	 body	 fat	 levels	 in	 physically	 active	 occupations,	 and	 a	 similar	
difference	can	still	be	observed	in	elite	military	groups	where	high	daily	energy	expenditures	are	required.	
Amish	and	Old-order	Mennonite	families	who	eschew	modern	mechanical	devices	have	very	high	levels	of	
daily	physical	activity,	and	this	is	reflected	in	their	correspondingly	low	levels	of	body	fat.	Similar	findings	
have	been	reported	for	Inuit	who	still	follow	a	hunter-gatherer	lifestyle.	Studies	validating	questionnaires	and	
accelerometers	 in	 the	 general	 population	 have	 show	 correlations	 of	 up	 to	 -0.50	 between	 scores	 on	 such	
instruments	 and	 estimates	 of	 body	 fat	 content.	 Finally,	 cross-sectional	 comparisons	 within	 circumpolar	
communities	 have	 demonstrated	 higher	 levels	 of	 aerobic	 fitness	 in	 those	 populations	 with	 the	 least	
acculturation	to	a	sedentary	"white"	lifestyle,	and	this	sub-group	of	circumpolar	residents	also	has	a	low	body	
fat	 content.	Conclusions:	 Conclusions	about	 the	 impact	of	vigorous	daily	physical	 activity	upon	body	 fat	
content	are	limited	by	the	possible	self-selection	of	a	physically	active	lifestyle,	with	the	adoption	of	other	
favorable	 health	 behaviours	 by	 active	 individuals.	 Nevertheless,	 all	 of	 numerous	 population	 comparisons	
show	lower	levels	of	body	fat	in	those	groups	who	maintain	an	adequate	level	of	habitual	physical	activity.	In	
some	 cases,	 the	 levels	 of	 physical	 activity	 adopted	 have	 been	 very	 high,	 but	 there	 are	 also	 examples	 of	
populations	 where	 a	 favourable	 body	 composition	 has	 been	 associated	 with	 the	 regular	 performance	 of	
volumes	of	physical	activity	that	are	appropriate	for	average	middle-aged	adults.	Health	&	Fitness	Journal	of	
Canada	2019;12(2):64-96.		
https://doi.org/10.14288/hfjc.v12i2.273		
	
Keywords:	Acculturation,	Active	individuals,	Amish,	Heavy	workers,	Inuit,	Marathon	runners,	Occupational	
fitness,	Physical	activity,	Physical	education	students		



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	65	

Introduction	
Many	 nutritionists	 have	 argued	 over	

the	years	 that	regular	physical	activity	 is	
unlikely	 to	be	effective	 in	 the	prevention	
or	 the	 treatment	 of	 obesity	 (Epstein	 &	
Wing,	 1980;	 Jakicic,	 Clark,	 &	 Coleman,	
2001;	 Malhotra	 &	 Noakes,	 2015;	 Shaw,	
Gennat,	&	O'Rourke,	2006).	Such	authors	
have	pointed	to	the	large	total	volume	of	
energy	 expenditure	 needed	 to	 reduce	
body	 fat	 content	 significantly,	 the	 likely	
stimulation	 of	 appetite	 by	 vigorous	
exercise,	and	a	possible	reduction	of	both	
resting	 metabolic	 rate	 and	 spontaneous	
physical	 activity	 that	 counters	 the	
therapeutic	 response	 to	both	dieting	and	
vigorous	 exercise.	 A	 companion	 article	
(Shephard,	 2019)	 has	 recently	 examined	
(and	 largely	dismissed)	 these	arguments.	
The	 present	 narrative	 review	 offers	
empirical	 cross-sectional	 evidence	
suggesting	 that	 an	 adequate	 level	 of	
habitual	 physical	 activity	 is	 consistently	
associated	with	low	levels	of	body	fat	in	a	
wide	 variety	 of	 populations.	 Subsequent	
articles	 will	 substantiate	 this	 view	 with	
prospective	 data	 showing	 substantial	
favourable	changes	of	body	composition	in	
response	 to	 regular	 participation	 in	 an	
exercise	programme,	and	will	then	explore	
the	 role	 of	 exercise	 programmes	 in	 the	
more	 intractable	 problem	 of	 treating	
established	obesity.	
When	looking	at	reasons	behind	a	trim	

body	 figure,	 investigators	 inevitably	 face	
the	perennial	problem	of	disentangling	the	
closely	 intertwined	 influences	of	physical	
activity,	a	well-regulated	diet,	other	facets	
of	 a	 healthy	 lifestyle,	 and	 a	 genetic	
predisposition	 to	 a	 low	 body	 fat	 content	
among	 those	 who	 have	 achieved	 and	
maintain	 an	 appropriate	 body	 build	
(Shephard,	 2018).	 We	 will	 here	 look	
critically	at	evidence	derived	from	various	
cross-sectional	 studies	 that	 have	 related	

habitual	 physical	 activity	 to	 body	 fat	
content,	 including	 comparisons	 between	
the	 general	 population	 and	 supposedly	
active	groups	such	as	endurance	athletes,	
participants	 in	 other	 long-distance	 and	
ultra-long	 distance	 events	 (Table	 1),	
physical	 education	 students,	 workers	 in	
physically	 demanding	 jobs,	 traditional	
Amish	 farmers	 and	 unacculturated	 Inuit.	
We	 will	 also	 look	 at	 cross-sectional	
associations	 between	 self-reported	 or	
objectively	 measured	 habitual	 physical	
activity	 and	 obesity	 levels	 in	 the	 general	
population,	 and	 will	 compare	 the	 body	
composition	of	 indigenous	populations	at	
various	 stages	 in	 their	 adoption	 of	 a	
sedentary	western	lifestyle.	
	
Endurance	 athletes	 vs.	 the	 general	
population	 	
There	are	occasional	groups	of	athletes	

such	 as	 Sumo	 wrestlers,	 where	 the	
accumulation	of	fat	apparently	gives	them	
a	competitive	advantage,	but	most	athletes	
body	fat	quickly	accumulates	if	training	is	
interrupted	 by	 injury,	 and	 some	 of	 the	
thinnest	athletes	are	those	such	as	ultra-
marathoners	who	engage	in	many	hours	of	
endurance	training	each	day.		
An	 early	 analysis	 of	 6	 studies	 of	 elite	

male	 distance	 runners	 showed	 body	 fat	

Table	1:		Weighted	average	of	body	fat	content	
(%	fat)	as	reported	for	elite	endurance	athletes	
(Shephard	and	Åstrand,	2000).	

Sport	 Male	
competitors	

Female	
competitors	

Canoeing	 11.2	 	
Cycling	 8.6	 15.4	
Cross	country	
skiing	

7.9	 15.7	

Distance	
running	

4.7	 16.3	

Rowing	 8.5	 14.0	
Soccer	 10.3	 21.0	
Swimming	 11.6	 18.6	
Triathlon	 9.0	 14.1	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	66	

values	ranging	from	-0.5%	to	9.4%,	with	a	
weighted	average	of	only	4.7%	(Shephard		

&	 Âstrand,	 2000;	 Table	 2).	 This	 figure	
stands	 in	 considerable	 contrast	 with	

Table	2:		Estimated	body	fat	percentage	in	various	groups	of	distance	runners.	
	

Authors	 Sample	 Measure	of	body	fat	
Males	 Females	

Arrese	&	Ostáriz	
(2006)	

42	M,22	F	elite	runners,	Sum	of	6	
skin-folds,	3	km,	10	km,	and	
marathoners	

35.9,	38.4,	31.2	mm		 61.7,	50.4,	44.4	mm,		

Arazi	et	al.	(	2012)	 160	phys	ed.,	170	other	male	
students	

14.1%	(PE)	
21.1%	(other)	

	

Arazi	&	Hoseini	
(2011)	

140	student	athletes,	160	other	
female	students	

	 BMI	21.5	vs.	25.2	
kg/m2	

Bale	et	al.	(1986)	 20	elite	male	10	km	runners	 8.0%	fat	 	
Bartlett	et	al.	(1984)	 10	collegiate-level	female	distance	

runners	
	 14.8%	fat	

Coetzer	et	al.	(1993)		 9	white,	11	black	runners	(?	male)	
over	distances	>	3km.	

Average	skin-fold	6.3	
mm	vs.	5.0	mm	

	

Conley	&	
Krahenbuhl	(1980)	

12	male	10	km	runners	 Average	skin-fold	5.3	
mm	

	

Costill	et	al.	(1970)	 114	participants	in	1968	US	
marathon	trial	

7.5%	by	3	skin-fold	
equations	(12.5%	in	
other	students)	

	

Cureton	et	al.	(1978)	 4	men,	2	women,	distance	runners		 9.4%	 20.8%	
de	Garay	et	al.	
(1974)		

34	male	distance	runners	 -0.5%	 	

Eksterowicz	et	al.	
(2016)	

9	Kenyan	male	distance	&	marathon	
runners	

5.4%	 	

Flack	(1983)	 14	U.S.	marathon	runners	 6.4%	 	
Graves	et	al.	(1987)	 15	elite	and	12	good	female	

distance	runners	
		 14.3%	(elite)	

16.8%	(good)	
Hirata	et	al.	(1966)	 99	male	Tokyo	Olympic	

participants	
1.4%	 	

Knechtle	et	al.	
(2015)	

42	female	Recreational	half-
marathon	runners	

	 27.2%	

Kong	&	de	Heer	
(2008)	

6	male	Kenyan	distance	runners	 5.3%	 	

Pollock	et	al.	(1977)	 11	distance,	8	marathon	runners	
(male)	

5.0%	(distance)	4.3%	
(marathon)	

	

Rusko	et	al.	(1978)	 8	top	distance	malerunners		 8.4%	 	
Sprynarová	&	
Parizková	(1971)	

10	top	?	male	runners,	distance	not	
specified	

6.4%	 	

Upton	et	al.	(1983)	 38	female	middle-aged	
marathoners	vs.	35	sedentary	
controls	

	 15.6%	vs.	27.8%	

Vaccaro	et	al.	(	1981)	 10	female	distance	runners	
(Masters)	

	 18.3%	

Vernilio	et	al.	(2013)	 14	top	male	Kenyan	runners	 8.9%	 	
Wilmore	et	al.	
(1977)	

	42	female	world	class	distance	
runners	

	 16.9%	

Wilmore		&	Brown	
(1974)		

11	female	National	level	distance	
runners	

	 15.2%,	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	67	

typical	 values	 for	 healthy	 young	 North	
American	 adults	 (~15-22%	 in	 men,	 and	
~22-28%	in	women,	Lohman,	1992).	
	
Detailed	data	on	the	body	fat	content	of	

distance	runners	are	summarized	in	Table	
1.	In	addition	to	very	low	overall	values,	in	
general,	the	male/female	body	fat	ratio	is	
lower	 than	 in	 the	 general	 population,	
perhaps	 because	 a	 higher	 proportion	 of	
male	 competitors	 have	 a	 strong	
commitment	to	distance	running.	
Looking	 at	 details	 of	 some	 of	 the	

individual	 studies,	 Arrese	 &	 Ostáriz	
(2006)	 reported	 values	 for	 a	 group	 of	
highly	 trained	runners	 (130	men	and	54	
women)	 who	 were	 competing	 over	
various	 distances.	 Focusing	 on	 the	
distance	 competitors,	 data	 averaged	
across	6	skin-folds	showed	values	of	only	
5-6	mm	in	the	men,	and	7-10	mm	in	the	
women,	 as	 compared	 with	 averages	 of	 8	
and	 10	 mm	 for	 young	 men	 and	 women	
from	 the	 general	 population	 in	 Toronto	
(Shephard,	1979).	In	both	sexes,	the	skin-
fold	thickness	was	inversely	related	to	the	
athlete's	 competitive	 distance,	 and	 the	
lowest	 readings	 were	 seen	 in	 the	
marathoners.		
Bale	et	al.	(1986)	examined	60	male	10	

km	 runners	 at	 various	 levels	 of	
proficiency.	 In	 the	 20	 elite	 competitors,	
who	 presumably	 were	 the	 most	 heavily	
committed	 to	 training,	 the	 skin-fold	
predictions	 of	 body	 fat	 content	 averaged	
8.0%,	while	 the	 thickness	of	5	measured	
skin-folds	 averaging	 5	 mm.	 However,	 in	
second	and	 third	rank	competitors,	body	
fat	 content	 was	 usually	 substantially	
greater,	with	average	values	of	10.7%	and	
12.1%	respectively.		
Bartlett	 et	 al.	 (1984)	 compared	 10	

female	 collegiate-level	 distance	 runners	
with	10	university	gymnasts.	The	runners	
were	taller	than	the	gymnasts,	and	in	both	

groups	the	hydrostatic	estimates	of	body	
fat	 (averages	 of	 14.8%	 and	 16.8%,	
respectively)	 were	 much	 lower	 than	
among	 the	 general	 population	 of	
university	women.	
Coetzer	et	al.	(1993)	compared	body	fat	

in	11	"black"	and	9	“white"	South	Africans.	
Both	 groups	 were	 committed	 athletes,	
competing	 over	 distances	 >3	 km,	 and	
engaging	 in	 80-90	 km	 of	 training	 per	
week.	Averages	of	skin-fold	readings	taken	
at	7	measurement	sites	for	these	subjects	
were	 5.0	 and	 6.3	 mm,	 respectively.	 	 The	
"black"	 runners	 were	 not	 only	
systematically	thinner	(p=	0.012),	but	also	
had	 had	 consistently	 higher	 speeds	 than	
their	"white"	competitors,	mainly	because	
they	 trained	 at	 a	 higher	 percentage	 of	
their	 maximal	 oxygen	 intakes,	 and	 thus	
had	 a	 lesser	 lactate	 accumulation	 during	
an	event.	
Conley	&	Krahenbuhl	(1980)	studied	12	

of	 the	 top	 male	 finishers	 in	 a	 nationally	
known	 10	 km	 race.	 This	 highly	 trained	
group	 had	 an	 average	 maximal	 oxygen	
intake	 of	 71.7	 ml/[kg.min].	 Skin-folds	
measured	 at	 6	 sites	 had	 an	 average	
thickness	of	only	5.3	mm.	Inter-individual	
differences	 in	 performance	 among	 this	
group	seemed	due	mainly	to	differences	in	
running	economy.	
Costill	 et	 al.	 (1970)	 compared	 all	 114	

participants	 in	 the	 U.S.	 1968	 marathon	
trials	with	a	group	of	92	male	university	
students	 who	 were	 not	 participating	 in	
any	 organized	 physical	 training	
programme,	and	63	graduate	students	and	
faculty.	 Using	 the	 average	 of	 3	 different	
skin-fold	 predictions	 to	 determine	 body	
fat	content,	average	values	were	7.5%	for	
the	 distance	 runners,	 compared	 with	
12.5%	for	the	non-athletic	undergraduates	
and	16.3%	for	the	graduate	students	and	
faculty	 members.	 It	 was	 estimated	 that	
despite	 a	 relatively	 low	 total	 body	 mass,	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	68	

the	 runners	 had	 energy	 expenditures	 as	
high	as	25MJ/day.		
Cureton	et	al.	(1978)	reported	data	for	

4	 male	 and	 2	 female	 distance	 runners;	
although	their	subjects	engaged	in	regular	
training,	 they	 were	 less	 fit	 than	 some	 of	
the	other	groups	discussed	in	this	section,	
with	 maximal	 oxygen	 intakes	 of	 ~58	
ml/[kg.min]	 for	 the	 men,	 and	 ~49	
ml/[kg.min]	for	the	women.		Body	fat	was	
determined	by	underwater	weighing,	and	
the	 respective	 estimates	 of	 9.4%	 and	
20.8%	were	somewhat	 lower	than	in	the	
general	 population.	 Cureton	 et	 al.	 noted	
that	 any	 experimental	 additions	 to	 body	
mass	 led	 to	 corresponding	 decreases	 in	
distance	running	performance.		
De	 Garay	 et	 al.	 (1974)	 based	 their	

estimates	 of	 body	 fat	 in	 Olympic	
champions	 on	 somatotyping	 and	 a	
measurement	 of	 bone	 diameters.	 They	
obtained	 very	 low	 (and	 often	 puzzlingly	
low	 negative)	 estimates	 of	 body	 fat	
content	in	44	male	distance	and	marathon	
runners.	 Hirata	 (1966)	 used	 a	 similar	
methodology	 on	 99	 male	 Olympic	
champions	in	Tokyo,	again	obtaining	very	
low	 values	 for	 the	 distance	 runners	 (an	
average	of	1.4%	fat).	
Eksterowicz	et	al.	(2016)	examined	the	

contribution	of	physical	characteristics	to	
the	outstanding	competitive	performances	
of	9	male	Kenyan	distance	and	marathon	
runners.	A	skin-fold	estimate	of	body	 fat	
content	yielded	an	average	of	5.4%,	with	
the	thickness	of	3	skin-folds	averaging	the	
very	low	figure	of	4.3	mm.	Kong	&	de	Heer	
(2008)	had	similar	findings	for	a	group	of	
6	 male	 Kenyan	 distance	 runners,	 with	 a	
skin-fold	 estimate	 of	 5.3%	 body	 fat.		
Vernilio	et	al.	(2013)	made	a	further	study	
of	 14	 male	 Kenyan	 runners,	 finding	 a	
somewhat	higher	(although	still	very	low)	
skin-fold	prediction	of	fat	(8.9%).	This	last	
study	is	the	most	recent,	and	it	is	possible	

that	 the	 nutrition	 of	 the	 top	 Kenyan	
runners	 may	 have	 improved	 over	 the	
course	of	these	3	studies.		 	
Flack	(1983)	used	both	hydrostatic	and	

anthropometric	 methods	 to	 estimate	 the	
body	fat	content	of	14	marathon	runners	
who	 were	 among	 528	 elite	 U.S.	 male	
athletes	 attending	 the	 U.S.	 Olympic	
Training	Center	 in	Colorado	Springs.	The	
average	 skin-fold	 prediction	 of	 body	 fat	
content	 for	 this	 sample	 was	 6.4%,	
although	 a	 higher	 value	 (12.5%)	 was	
obtained	 by	 hydrostatic	 weighing.	 Fleck	
noted	that	the	lowest	fat	values	were	seen	
in	 athletic	 disciplines	 where	 a	 high	
maximal	oxygen	intake	was	required.		
Graves	 et	 al.	 (1987)	 studied	 the	 body	

composition	 of	 15	 elite	 (international	 or	
national	 level)	 and	 12	 good	 (local	 level)	
female	 distance	 runners.	 The	 elite	 group	
had	 been	 running	 an	 average	 of	 104	
km/week	 during	 the	 previous	 year,	 and	
the	 good	 runners	 had	 covered	 63	
km/week	 during	 their	 training.	 The	
respective	 estimates	 of	 body	 fat	 by	
underwater	 weighing,	 at	 14.8%	 and	
16.3%,	 were	 low	 for	 women,	 and	 values	
for	the	"good"	group	were	not	significantly	
greater	 than	 those	 for	 the	 elite	 runners.	
Skin-fold	 and	 bio-impedance	 predictions	
both	 tended	 to	 underestimate	 the	
hydrostatic	predictions	in	this	sample.		
The	 data	 of	 Knechtle	 et	 al.	 (2015)	

illustrate	 the	 importance	 of	 commitment	
to	 distance	 running.	 In	 a	 group	 of	 42	
unselected	 female	 recreational	 half-
marathoners	 (participants	 in	 the	 Basel	
half-marathon),	 training	 distances	 were	
limited	to	an	average	of	32	km/week,	or	
3.6	hours	of	running	per	week,	and	a	skin-
fold	 prediction	 of	 body	 fat	 yielded	 the	
quite	 high	 average	 of	 27.2%.	 The	
percentage	 of	 body	 fat	 was	 quite	 closely	
correlated	with	race	times	(r	=	0.56),	and	
was	 also	 correlated	 negatively	 with	 the	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	69	

weekly	training	distance	(r	=	0.34	to	-0.69	
for	different	skin-folds).	
Pollock	et	al.	(1977)	made	skin-fold	and	

hydrostatic	predictions	of	body	 fat	 in	11	
elite	(international	or	national	class)	long-
distance	 and	 marathon	 runners,	 finding	
respective	 body	 fat	 levels	 of	 5.0%	 and	
4.3%,	with	figures	of	5.2	and	5.8	mm	for	
the	 average	 of	 7	 skin-folds.	 In	 "good"	
rather	 than	 elite	 runners,	 the	 average	
predicted	body	fat	increased	to	6.0%,	and	
in	 "average":	 young	 men	 they	 found	 a	
skin-fold	thickness	of	6.5	mm	
Rusko	 et	 al.	 (1978)	 studied	 athletes,	

about	a	half	of	whom	were	members	of	the	
Finnish	 national	 team.	 Skin-fold	
predictions	of	body	fat	in	8	male	distance	
runners	yielded	an	average	of	8.4%	body	
fat.	They	compared	this	figure	to	average	
values	of	11.0%	for	8	physical	education	
students,	and	14.4%	for	23	male	controls.	
The	relative	training	status	of	the	3	groups	
can	 be	 inferred	 from	 their	 respective	
levels	 of	 maximal	 oxygen	 intake	 (78.1,	
65.1	and	55,	ml/[kg.min]).	
Sprynarová	 &	 Parizková	 (1971)	

reported	 hydrostatic	 predictions	 of	 body	
fat	 in	 10	 top	 (presumably	 male)	
representative	 Czech	 runners	 (distance	
not	 specified;	 average	 maximal	 oxygen	
intake	 64	 ml/[kg.min]);	 values	 averaged	
6.3%	fat.	 	
Upton	 et	 al.	 (1983)	 compared	 38	

middle-aged	 female	 marathoners	 (aged	
31-50	years,	who	were	currently	training	
over	an	average	distance	of	74	km/week	
with	35	sedentary	women.	Differences	 in	
maximal	 oxygen	 intake	 indicated	 their	
respective	 levels	 of	 physical	 condition	
(55.5	 and	 31.4	 ml/[kg.min].	 Hydrometric	
estimates	 of	 body	 fat	 were	 15.6%	 and	
27.8%.	Vaccaro	et	al.	(1981)	also	studied	
older	women,	finding	an	average	body	fat	
of	 18.3%	 in	 10	 female	 Masters	 runners	
with	 an	 average	 age	 of	 44	 years;	 their	

body	fat	was	still	only	some	40%	of	values	
found	in	controls	of	similar	age.	
Wilmore	&	Brown	(1974)	and	Wilmore	

et	al.	(1977)	used	underwater	weighing	to	
determine	the	body	fat	in	two	samples	of	
top	(international	or	national	level)	female	
distance	runners,	finding	values	of	15.2%	
and	16.9%	respectively,	about	50%	of	the	
values	that	they	observed	in	their	control	
subjects.	 In	2	of	 the	best	 female	runners	
from	this	group,	the	body	fat	content	was	
less	than	6%.	The	maximal	oxygen	intake	
of	 the	 group	 averaged	 59.1	 ml/[kg.min],	
but	 in	 the	 thinnest	 individual	 a	 value	 of	
71.1	ml/[kg.min]	was	observed.	
In	 summary,	 various	 authors	 have	

consistently	demonstrated	very	low	skin-
fold	readings	and	predictions	of	body	 fat	
content	in	athletes,	particularly	in	distance	
runners.	Some	of	this	association	could	be	
explained	 by	 constitutional	 factors,	 with	
those	having	a	slim	body	build	gravitating	
to	 running	 as	 their	 preferred	 sport.	
However,	 long-distance	 runners	 also	
engage	in	very	large	volumes	of	endurance	
activity,	with	energy	expenditures	in	some	
cases	 as	 high	 as	 25	 MJ/day,	 and	 the	
contribution	of	this	physical	activity	to	low	
fat	levels	is	strongly	suggested	by	inverse	
relationships	to	several	measures	of	daily	
energy	 expenditure:	 the	 individual's	
preferred	 competitive	 distance,	 the	 level	
of	 competition,	 and	 the	 maximal	 oxygen	
intake.	
	
Participants	 in	 other	 distance	 and	
ultra-long-distance	running	events	
In	 support	 of	 the	 evidence	 garnered	

from	top	endurance	athletes,	we	may	look	
also	at	body	fat	data	obtained	on	distance	
runners	 of	 less	 than	 international	
standard,	 and	 other	 individuals	 who,	
although	not	highly	selected	athletes	have	
engaged	in	exceptional	bouts	of	sustained	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	70	

physical	activity,	such	as	running	coast-to-
coast	across	Canada.		
	
Other	distance	runners.			
Farrell	 et	 al.	 (1979)	 studied	 18	

experienced	 male	 distance	 runners	 aged	
28	years	who	had	been	competing	for	1	to	
10	years	and	had	a	maximal	oxygen	intake	
averaging	 61.8	 ml/[kg.min].	 Hydrostatic	
weighing	showed	an	average	of	9.3%	body	
fat	in	this	group.		
Lewis,	Haskell,	Klein,	Halpern,	&	Wood	

(1975)	 collected	 hydrostatic	 data	 on	 45	
physically	 active	 middle-aged	 men	
(average	 age	 47	 years),	 most	 of	 whom	
were	engaged	in	jogging	programmes	and	
were	 running	 distances	 >25	 km/week;	
many	 supplemented	 their	 running	 with	
swimming,	 cycling	 and	 active	 games.	
Treadmill	 times	 averaged	 16	 minutes	 on	
the	 Bruce	 protocol,	 demonstrating	 an	
excellent	 level	of	 fitness,	 and	hydrostatic	
data	showed	an	average	of	13.2%	body	fat	
in	this	group.	
Tudor-Locke	et	al.	(2010)	examined	the	

NHANES	 data	 for	 a	 large	 sample	 of	 the	
general	 U.S.	 population,	 demonstrating	
differences	 of	 actigraph-measured	 step-
counts	 collected	 for	 1	 waking	 day	 or	
longer	between	those	with	a	normal	BMI	
(8285	 steps/day	 in	 men	 and	 6486	
steps/day	 in	 women),	 those	 who	 were	
overweight	(7643	steps/day	in	men,	5482	
steps/day	in	women)	and	those	who	were	
obese	(BMI	>	30	kg/m2,	6664	steps/day	in	
men,	 5069	 steps/day	 in	 women).	 It	 is	
important	 to	 note	 that	 all	 individuals	 in	
this	 sample	 were	 relatively	 inactive,	 and	
that	 the	 step	 counts	 associated	 with	
maintenance	 of	 a	 normal	 BMI	 were	 well	
within	the	capacity	of	the	average	middle-
aged	individual.	
	
	
	

Ultra-long	distance	events.		
One	healthy	43-year-old	woman	ran	a	

total	of	7250	km	across	Canada	(Mertens	
et	 al.,	 1996),	 covering	 an	 average	 of	 65	
km/day	for	112	days	at	a	typical	pace	of	
~7.9	 km/h.	 During	 the	 course	 of	 this	
journey,	she	metabolized	16.7	kg	of	tissue,	
81.4%	 of	 which	 was	 fat.	 Clearly,	 in	 her	
case,	 the	 regular	 daily	 physical	 activity	
was	 very	 effective	 in	 reducing	 her	
immediate	body	fat	content,	although	we	
have	no	information	on	how	well	this	fat	
loss	was	sustained	in	subsequent	years.	A	
young	man	completed	a	similar	 run,	and	
although	 he	 made	 strenuous	 efforts	 to	
maintain	his	nutrition	over	the	journey,	he	
also	lost	about	8	kg	of	weight,	mainly	fat	
(Shephard	et	al.,	1975).		
Schutz	et	al.	(2012)	described	a	Trans-

European	foot	race,	where	44	participants	
who	had	previously	engaged	in	an	average	
of	18	years	of	distance	running	covered	a	
4,486	 km	 ultra-marathon.	 The	 bio-
impedance	 estimate	 of	 body	 fat	
percentage,	averaged	across	40	men	and	4	
women,	 was	 11.2%	 at	 the	 outset	 of	 this	
event.	
Several	 other	 authors	 have	 reported	

observations	on	ultra-distance	runners.	A	
60-year-old	 Quebec	 man	 ran	 6400	 km	
over	 a	 period	 of	 4	 months	 (Jobin,	
Tremblay,	 and	 Samson,	 1984),	 reducing	
his	 body	 mass	 by	 5.4	 kg	 (mostly	 fat).	
Bioelectrical	 impedance	 analysis	 also	
showed	 a	 decrease	 of	 body	 fat	 content	
after	a	woman	ran	1200	km	over	17	days,	
although	this	change	was	not	detectable	in	
skin-fold	 measurements	 (Knechtle	 et	 al.,	
2008).	Knechtle	&	Bircher	(2005)	had	an	
athlete	run	for	6	days,	finding	that	despite	
an	 increased	 intake	 of	 fat	 and	 protein	
during	 the	 run,	 there	 was	 a	 substantial	
loss	of	sub-cutaneous	fat.	Finally,	a	320	km	
run	across	Switzerland,	completed	over	54	
hours,	and	involving	a	7000	m	increase	of	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	71	

altitude,	led	to	a	1.02	kg	loss	of	lean	tissue	
and	 a	 0.30	 kg	 decrease	 of	 fat	 mass	
(Knechtle	&	Knechtle,	2007).		
	
Physical	 education	 students	 vs.	 other	
university	students	
Comparisons	of	body	fat	content	among	

university	 students	 have	 been	 based	 on	
either	 the	 individual's	 choice	 of	 faculty	
(physical	education	vs.	other	disciplines),	
or	 on	 questionnaire	 or	 objective	
measurements	 of	 actual	 activity	 patterns	
in	the	selected	groups	of	students.	
	
Choice	of	Faculty		
During	 term-time,	 physical	 education	

(PE)	students	usually	devote	a	substantial	
part	of	each	week	to	learning	the	physical	
skills	 that	 will	 be	 needed	 in	 their	
profession.	 A	 typical	 programme	 may	
involve	participation	in	8	weekly	one-hour	
practice	 in	disciplines	such	as	basketball,	
soccer,	 squash	 and	 tennis,	 and	 many	
students	 supplement	 such	 curricular	
requirements	by	vigorous	training	for	one	
or	 more	 inter-collegiate	 sports	 teams.	
Thus	 PE	 students	 might	 be	 thought	 to	
offer	 useful	 evidence	 on	 the	 impact	 of	
several	 years	 of	 sustained	 and	 vigorous	
physical	 activity	 upon	 the	 body	
composition	 of	 a	 young	 adult.	 There	 is	
evidence	 of	 some	 selective	 recruitment,	
with	more	obese	 individuals	avoiding	PE	
as	a	career	choice.	Nevertheless,	one	might	
anticipate	that	the	physical	characteristics	
of	 the	 average	 PE	 student	 would	 reflect	
the	 effects	 of	 regular	 vigorous	 physical	
activity	in	containing	the	accumulation	of	
body	 fat,	 despite	 the	 absence	 of	 any	
deliberate	 restrictions	 on	 the	 type	 and	
quantity	of	food	ingested.		
In	 general,	 the	 expectation	 of	 a	 low	

body	 fat	 content	 relative	 to	 students	 in	
other	faculties	is	met,	but	there	are	some	
exceptions.	 One	 factor	 complicating	 the	

analysis	 is	 a	 difference	 of	 entry	 grades	
between	 faculties.	 In	 prestigious	
universities,	 the	 minimum	 required	 high	
school	 academic	 grades	 tend	 to	 be	 less	
stringent	 for	 PE	 than	 for	 other	 faculties.	
Thus,	 individuals	 who	 are	 less	
intellectually	 gifted	 may	 opt	 to	 seek	 a	
placement	 in	 PE	 not	 because	 they	 have	
any	 strong	 personal	 commitment	 to	
physical	activity,	but	rather	because	they	
view	 such	 a	 choice	 as	 an	 easy	 means	 of	
gaining	 access	 to	 a	 major	 university	
despite	less	than	stellar	high	school	marks.	
Such	students	tend	to	select	physically	less	
demanding	 practica	 within	 a	 physical	
education	 programme	 (classes	 in	 golf,	
archery,	 ballroom	 dancing	 and	 yoga,	 for	
example),	and	even	in	these	activities	they	
may	 invest	 the	 minimum	 amount	 of	
interest	 and	 physical	 effort	 required	 to	
obtain	a	passing	grade.	
	
Intensity	 of	 programme.	 Stachoń	 &	
Pietraszewska	 (2013)	 obtained	 evidence	
on	 the	 importance	 of	 programme	
intensity.	 They	 classified	 252	 male	
physical	 education	 students	 in	 terms	 of	
their	apparent	commitment	to	a	high	daily	
level	of	physical	activity.	Those	members	
of	the	class	who	were	rated	as	vigorously	
active	 had	 a	 significantly	 lower	 bio-
impedance	estimated	percentage	of	body	
fat	 (19.6%)	 than	 those	 were	 classed	 as	
only	moderately	active	(21.8%).		
In	 addition	 to	 the	 student's	 personal	

commitment,	 a	 further	 potential	 issue	
affecting	the	intensity	of	daily	effort	is	that	
in	some	PE	programmes,	the	first	year	of	
instruction	 may	 focus	 on	 academic	
disciplines	 such	 as	 psychology	 and	
sociology.	 Thus,	 the	 initial	 physical	
demands	on	the	student	may	be	quite	low,	
and	even	insufficient	to	maintain	the	level	
of	physical	condition	observed	at	entry	to	
university.	 For	 instance,	 Meckel	 et	 al.	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	72	

(2011)	evaluated	85	male	and	89	 female	
Israeli	 physical	 education	 students	 on	
entry	to	the	Wingate	 Institute	and	at	 the	
end	 of	 their	 freshman	 year.	 Over	 this	
introductory	 year,	 both	 sexes	 showed	
some	evidence	of	a	loss	of	physical	fitness,	
with	small	 increases	 in	their	percentages	
of	body	fat,	and	a	parallel	slowing	in	their	
speeds	 over	 a	 2	 km	 run.	 Moreover,	 the	
deterioration	 of	 physical	 condition	 was	
most	marked	in	students	who	were	fittest	
on	admission	to	the	programme.		
	
Empirical	 data.	 A	 large	 number	 of	
investigators	have	reported	empirical	data	
on	 the	 body	 fat	 content	 of	 physical	
education	 majors	 (Table	 3).	 Some	
investigations	 have	 shown	 relatively	 low	
levels	 of	 body	 fat,	 particularly	 in	 male	
students,	but	in	other	data	sets	the	values	
for	 PE	 students	 have	 differed	 little	 from	
those	 found	 in	 students	 from	 other	
faculties.	
A	 study	 of	 161	 male	 and	 74	 female	

students	 enrolled	 in	 a	 Polish	 physical	
education	department	 (Smolarczyk	et	al.,	
2012)	underlined	two	factors	limiting	the	
interpretation	 of	 such	 inter-faculty	
comparisons.	 There	 was	 a	 substantial	
element	 of	 programme	 self-selection,	 in	
that	students	who	entered	the	PE	classes	
were	 taller	 than	 their	 peers	 in	 other	
faculties.	 In	 general,	 the	 physical	
education	 students	 also	 had	 a	 greater	
mass	of	lean	tissue,	so	that	BMI	values	did	
not	provide	a	very	reliable	basis	for	rating	
obesity	 levels.	 However,	 the	 men	 in	 the	
Polish	PE	programme	also	had	a	very	low	
average	body	fat	content	as	determined	by	
bio-impedance	measurements	(an	average	
of	 12.5%).	 The	 average	 body	 fat	 in	 the	
women	(21.5%)	was	not	particularly	high,	
although	a	third	of	the	female	PE	students	
in	this	study	had	more	than	25%	body	fat.	

In	 one	 of	 the	 earliest	 studies	 of	 a	 PE	
programme,	 Shephard	 &	 Pimm	 (1975)	
evaluated	 a	 substantial	 sample	 of	 third-
year	 students	 (68	 men	 and	 74	 women)	
who	 were	 attending	 the	 University	 of	
Toronto.	 Some	 of	 the	 students	 in	 this	
group	 were	 enthusiastic	 athletes,	 but	
other	 individuals	seemed	 to	have	chosen	
the	 faculty	of	PE	not	because	of	a	strong	
personal	 interest	 in	 fitness,	 but	 rather	
because	they	thought	that	it	offered	them	
a	 means	 of	 entering	 the	 teaching	
profession	 despite	 modest	 high	 school	
academic	grades.	Perhaps	 for	 this	reason	
neither	average	BMI	values	(24.4	kg/m2	in	
the	 men,	 25.3	 kg/m2	 in	 women)	 nor	
average	 skin-fold	 estimates	 of	 body	 fat	
percentage	 (18.0%	 in	men,	26.5%	 in	 the	
women)	 were	 particularly	 remarkable	
relative	to	the	general	student	population.		
Wasiluk	et	al.	(2013)	also	found	few	inter-
faculty	 differences	 in	 body	 fat	 content	
among	 165	 first-year	 Polish	 university	
students	who	were	enrolled	in	courses	in	
physical	 education,	 physiotherapy,	
tourism	 and	 sport,	 and	 cosmetics.	
Likewise,	Bale	 (1980)	reported	relatively	
high	 skin-fold	 estimates	 of	 body	 fat	
(average	 24%)	 in	 53	 English	 female	
physical	education	students.		
Several	 reports	 made	 direct	

comparisons	of	body	fat	between	physical	
education	 students	 and	 their	 peers	 who	
had	enrolled	in	other	faculties.	Four	such	
investigations	 found	 lower	 fat	 levels	
among	the	physical	education	students	of	
both	 sexes	 (Arazi	 et	 al,	 2011,	 2012;	
Lutoslawska	 et	 al.,	 2014;	 Reuter	 et	 al.,	
2012),	and	another	found	low	fat	levels	in	
the	males	only	(Neculães	&	Lucaci,	2016).	
However,	one	report	that	considered	only	
female	students	found	no	lower	fat	levels	
in	 the	 physical	 education	 group	 than	 in	
students	 from	 other	 disciplines	 (Wasiluk	
et	al.,	2013).		 	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	73	

Table	3:	Estimates	of	percentage	body	fat	for	physical	education	students	and	their	peers	in	other	
faculties.	
	
Author	 Sample	 Body	fat	 Comments	

Male	 Female	
Arazi	et	al.	
(2011)	

140	female	"athletes"	
and	160	"non-athletes"	

	 BMI		20.5	vs.	
25.2	kg/m2	

Iran	

Arazi	et	al.	
(2012)	

160	male	POE	students,	
170	other	faculties	

BMI		21.9	vs.	
27.3kg/m2		

	 Iran	

Bale	(Bale,	
1980)	

53	Female	phys.	ed.	
students	

	 24.0%	 	

Kvintová	&	
Sigmund	(2016)		

114	active,	243	inactive	
female	university	
students	

	 20.9	(active)	
27.8	(inactive)	

"Active"	criterion	150	
min/wk	at	3-6	METS	or	
75	min/wk	at	6	METS	

Lutoslawska	et	
al.	(	2014)	

90	M,	87	F	phys.	ed.	
(pe)	students,	49	M,	46	
F	from	other	faculties	
(o)	

14.4	(pe)	
16.7	(o)	

25.2	(pe)	
30.9	(o)	

Poland;	PE	students	
physically	active	7-9	
h/week	

Meckel	et	al.	
(2011)		

85	M,	89	F	physical	
education	students	at	
entry	(e)	and	end	of	
freshman	year	(f),		

13.2	(e)	
14.0	(f)	

25.8	(e)	
26.6	(f)	

Israel.	Increase	of	fat	
paralleled	by	slowing	of	2	
km	run;	loss	of	fitness	
most	marked	in	those	
initially	fittest	

Morinaka	et	al.	
(2012)	

22	Japanese,	27	Thai	
university	students	

	 23.0	(Japanese)	
26.7	(Thai)	

	

Neculães	&	
Lucaci	(2016)	

PE,	Kinesiotherapy,	
Fitness,	Sports	(127	F,	
167	M	students)	

10.9	(PE)	
12.7	(K)	
13.8	(F)	
13.1	(S)	

21.5	(PE)	
20.5	(K)	
19.3	(F)	
20.7	(S)	

Romania	

Pribis	et	al.	
(2010)	

5101	US	university	
students,	1996	to	2008	

11.6	 22.4	 Body	fat	increased	
0.51%/yr	in	men	and	
0.65%/yr	in	women	

Reuter	et	al.	
(2012)	

Phys.	ed.	(15	M,	27	F)	
and	med.	(18	M,	25	F)	
students	

16.7	(pe)	
21.6	(med)	

28.3	(pe)	
31.1	(med)	

Brazil	

Shephard	and	
Pimm	(1975)	

Non-smoking	physical	
education	students	(68	
M,	74	F)	

18.0	
	
	
	
	

28.5	 Toronto	

Smolarczyk	et	al.	
(2012)	

161	men,	74	women	
physical	education	
students	

12.5	 21.5		 Poland;	body	fat	>	25%	
in	a	third	of	women	
students	

Stachoń	&	
Pietraszewska	
(2013)	

252	male	phys.	ed.	
students	

19.6	(va)	
21.9	(ma)	

	 Students	classed	as	
vigorously	(va)	or	
moderately	(ma)	active	

Tarnus	&	
Bourdon	(2006)	
	
	

14/23	male,	22/43	
female	"sportive"	
students		

22.1	(sportive)	
24.3	(non-
sportive)	

33.7	(sportive)	
35.9	(non-
sportive)	

	

Wasiluk	et	al.	
(2013)	

165	first-yr	students	in	
PE,	Physiotherapy	(pt),	
Tourism	(t),	Cosmetics	
(c).	

	
	

24.5	(pe)	
25.0	(pt)	
23.3	(t)	
23.8	(c)	

Poland	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	74	

Arazi	et	al.	(2012)	compared	160	male	
PE	 majors	 with	 170	 other	 male	 Iranian	
students,	 finding	 respective	 body	 fat	
estimates	of	14.1%	and	21.1%;	there		
were	 parallel	 differences	 in	 body	 mass	
index	 (BMI)	 and	 waist-hip	 ratios.	 	 140	
female	 "student	 athletes"	 attending	 the	
same	university	were	compared	with	160	
non-athletic	 females	 in	 terms	 of	 body	
mass	 indices	 (Arazi	 &	 Hoseini,	 2011);	
again	substantial	differences	favouring	the		
athletic	group	were	observed	(BMI	values	
of	 21.5	 vs.	 25.2	 kg/m2,	 with	 parallel	
differences	in	waist-hip	ratios).	
Lutoslawska	et	al.	(2014)	compared	90	

male	 and	 87	 female	 physical	 education	
students	 with	 49	 males	 and	 46	 females	
drawn	 from	 other	 faculties.	 They	 found	
somewhat	 lower	 skin-fold	 estimates	 of	
body	 fat	 in	 the	 PE	 students	 (many	 of	
whom	were	physically	active	for	as	much	
as	7-9	hours	per	week)	than	in	their	peers.	
Further	 data	 on	 Polish	 male	 physical	
education	 students	 (Saczuk,	 Wasiluk,	 &	
Wilczewski,	2016)	showed	a	small	secular	
trend	to	an	increase	of	BMI	over	25	years	
of	 observation:	 respective	 values	 were	
23.6	kg/m2	 in	1989,	23.8	kg/m2	 in	2004,	
and	24.1	kg/m2	in	2014.		
Morinaka	 et	 al.	 (2012)	 compared	

findings	 in	 small	 groups	 of	 female	
Japanese	 and	 Thai	 students.	 Body	 fat	
levels	 were	 unremarkable	 for	 students	
from	 faculties	 other	 than	 physical	
education.	 Nevertheless,	 when	
accelerometer	 measures	 of	 daily	 activity	
were	 compared	 with	 bio-impedance	
measures	 of	 body	 fat,	 a	 substantial	
negative	correlation	(r	=	-0.48)	was	seen	
in	 the	 Japanese	 students.	 However,	 this	
was	 not	 apparent	 in	 their	 Thai	 peers,	
perhaps	because	most	of	the	latter	group	
engaged	 in	 very	 little	 physical	 activity	
overall.	

Neculães	&	Lucaci	(2016)	evaluated	the	
body	fat	levels	of	217	female	and	167	male	
Romanian	students	who	were	enrolled	in	
4	 different	 health	 science	 and	 fitness-
oriented	 faculties.	 The	 body	 fat	
percentages	 were	 low,	 especially	 among	
the	 male	 students	 (10.9%-13.8%),	 with	
BMI	 values	 for	 the	 first	 3	 categories	 of	
student	 ranging	 from	22.9-24.4	kg/m2	 in	
the	 men,	 and	 20.7	 to	 20.9	 kg/m2	 in	 the	
women.		
Pribis	et	al.	(2010)	reported	quite	 low	

skin-fold	estimates	of	body	fat	(an	average	
of	11.6%)	in	a	sample	of	male	university	
students	 in	 Michigan	 who	 were	
participating	 in	 physical	 activity	 classes,	
but	the	average	values	for	female	students	
were	 higher	 (22.4%).	 Pribis	 et	 al.	
commented	 on	 a	 significant	 trend	 to	 an	
increase	of	body	fat	over	the	12	years	of	
observation,	between	1996	and	2008,	this	
had	averaged	0.51%/year	in	the	men	and	
0.65%/year	 in	 the	 women,	 and	 it	 was	
associated	 with	 a	 secular	 decrease	 of	
maximal	 oxygen	 intake	 in	 both	 sexes	
(Pribis	 et	 al.,	 2010).	 The	 suggested	
explanation	 was	 a	 decrease	 in	 habitual	
physical	 activity	 of	 the	 Michigan	 student	
population	 over	 the	 12-year	 period.	
However,	 such	 large	changes	 in	body	 fat	
content	could	also	reflect	secular	changes	
in	the	diet	and	ethnic	composition	of	the	
student	samples,	as	well	as	artifacts	from	
changes	in	the	personnel	making	the	skin-
fold	measurements.	 	
Reuter	 et	 al.	 (2012)	 compared	 small	

samples	of	physical	education	and	medical	
school	 students	 who	 were	 attending	 a	
Brazilian	university.	Data	on	body	 fat,	as	
obtained	 by	 dual-energy	 x-ray	
examination,	showed	a	small	advantage	to	
the	 physical	 education	 students	 of	 both	
sexes	 (16.7%	 and	 28.3%)	 relative	 to	 the	
medical	students	(21.6%	and	31.3%).	
	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	75	

Comparisons	 based	 on	 objective	
measurements	 of	 habitual	 physical	
activity.		 Additional	 proof	 on	 the	
efficacy	 of	 regular	 physical	 activity	 as	 a	
means	 of	 preventing	 obesity	 among	
university	 students	 can	 be	 found	 in	
studies	 where	 individuals	 have	 been	
classified	 based	 upon	 their	 reported	 or	
objectively-measured	 level	 of	 habitual	
physical	activity	rather	than	on	the	faculty	
in	which	they	were	enrolled.		
Thus,	 Kvintová	 &	 Sigmund	 (2016)	

categorized	 357	 female	 students	 on	 the	
basis	 of	 their	 reported	 physical	 activity;	
114	 of	 the	 group	 were	 classed	 as	
physically	 active	 (individuals	 spending	 >	
150	minutes/week	at	an	exercise	intensity	
of	 3-6	 METs,	 or	 75	 minutes/week	 at	 >6	
METS).	 This	 segment	 of	 the	 population	
had	a	substantially	lower	level	of	body	fat	
than	 their	 sedentary	 peers	 (20.8%	 vs.	
27.8%),	with	parallel	differences	of	BMI.		
Rutter	 (1994)	 studied	 the	 activity	

patterns	of	39	female	university	students	
using	a	Caltrac	personal	activity	computer.	
They	 reported	 correlations	 as	 high	 as	 -
0.38	 to	 -0.47	 between	 habitual	 physical	
activity	 and	 BMI,	 although	 perhaps	
because	of	an	initial	reactionary	response	
to	 wearing	 of	 the	 accelerometer,	
significant	 relationships	 were	 not	 seen	
until	the	accelerometer	had	been	fitted	for	
4-6	days.	
Tarnus	&	Bourdon	(2006)	examined	a	

small	sample	of	university	students	living	
on	 the	 island	 of	 La	 Réunion.	 All	 of	 this	
group	 were	 rather	 fat	 relative	 to	 their	
peers	in	other	parts	of	the	world,	but	there	
were	 nevertheless	 significant	 differences	
of	 obesity	 between	 students	 classed	 as	
"sportive"	(those	practicing	exercise	for	1	
hour	 or	 more	 per	 week)	 and	 the	 "non-
sportive;"	 respective	 levels	 of	 body	 fat	
were	 22.1%	 vs.	 24.3%	 in	 the	 men,	 and	
33.7	vs.	35.9%	in	the	women).		 	

Zanovec	 et	 al.	 (2009)	 used	 the	
International	 Physical	 Activity	
Questionnaire	 to	 classify	 142	 male	 and	
148	 female	 Baton	 Rouge	 university	
students	into	quartiles	of	habitual	physical	
activity,	 expressed	 in	 MET-hours	 of	
physical	 activity	 week.	 There	 was	 a	
substantial	 gradient	 of	 obesity	 between	
the	most	active	quartile	(23.0%	body	fat)	
and	the	least	active	(25.8%	body	fat),	and	
there	 was	 also	 a	 modest	 inverse	
correlation	 between	 the	 physical	 activity	
scores	 and	 an	 assessment	 of	 body	 fat	
content	 made	 by	 dual	 energy	 x-ray	
absorptiometry	(r	=	 -0.40).	However,	 the	
findings	 in	 this	study	were	not	co-varied	
for	 sex,	 and	 a	 spurious	 association	 may	
have	 been	 created	 because	 of	 a	
combination	of	 the	 inherently	greater	 fat	
content	 of	 the	 women	 and	 their	 lower	
average	levels	of	physical	activity.	Ruchan	
(2015)	 also	 demonstrated	 a	 relationship	
between	 body	 fat	 levels	 and	 habitual	
physical	activity	as	assessed	by	responses	
to	 the	 International	 Physical	 Activity	
Questionnaire.	 	 	
	
Summary		
Summarizing	the	reports	listed	in	Table	

3,	all	except	2	of	the	male	studies	showed	
fairly	low	levels	of	body	fat,	the	exceptions	
being	 the	 studies	 of	 Stachoń	 &	
Pietraszweska	 (2013)	 and	 Tarnus	 &	
Bourdon	(2006).	In	contrast,	fat	levels	for	
a	number	of	the	female	samples	have	not	
been	 much	 lower	 than	 would	 be	
anticipated	 in	 the	 general	 population	 of	
university	 age.	 Nevertheless,	 the	 data	
generally	 support	 the	 contention	 that	
university	students	who	are	involved	in	a	
physical	 education	 programme	 or	 some	
other	 form	 of	 regular	 vigorous	 physical	
activity	 maintain	 a	 lower	 percentage	 of	
body	fat	than	those	who	are	content	with	a	
sedentary	lifestyle.	Moreover,	this	benefit	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	76	

is	 apparently	 obtained	 without	 any	
deliberate	 restriction	 of	 dietary	 intake,	
and	in	some	instances	has	been	seen	with	
participation	in	quite	a	modest	volume	of	
weekly	activity.	
	
Physically	 demanding	 vs.	 sedentary	
occupations	
Early	 occupational	 epidemiology	 was	

based	 upon	 comparisons	 of	 people	 who	
were	engaged	 in	heavy	versus	 sedentary	
work.	 Automation	 and	 robotics	 has	 now	
eliminated	many	traditional	"heavy"	jobs,	
but	even	today	certain	occupations	such	as	
emergency	 workers	 and	 elite	 military	
units	 still	 find	 periodic	 demands	 for	
substantial	 occupational	 energy	
expenditure.	 Where	 information	 is	
available,	 it	 is	 instructive	to	compare	the	
body	composition	of	workers	in	such	jobs	
with	data	for	those	who	are	employed	in	
more	 sedentary	 professions.	 It	 is	 also	
useful	 to	 look	at	 the	 impact	of	employee	
fitness	 programmes	 upon	 the	 body	
composition	of	participants.		
	
Body	 fat	 levels	 seen	 in	 classical	
occupational	epidemiology		
Morris	&	Raffle	(1956)	made	one	of	the	

earliest	occupational	comparisons	of	body	
build,	 fitness	and	 life	expectancy,	 looking	
at	 differences	 between	 the	 drivers	 and	
conductors	 of	 London	 Transport	 double-
deck	 buses.	 During	 the	 early	 part	 of	 the	
20th	 century,	 the	 conductors	 were	
required	to	run	up	to	the	top	deck	of	these	
vehicles	several	 times	every	kilometer	of	
the	bus	route	in	order	to	check	tickets	and	
collect	fares	from	the	passengers,	whereas	
the	drivers	spent	most	of	their	day	seated	
in	 the	 comfort	 of	 the	 driver's	 cab.	 The	
comparison	of	body	fat	content	seemed	a	
valid	 one,	 since	 both	 conductors	 and	
drivers	were	drawn	from	the	same	socio-
economic	milieu,	ate	in	the	same	canteens,	

and	 had	 few	 differences	 in	 working	
conditions	other	than	their	daily	physical	
activity.		
A	 substantial	 difference	 of	 waist	 girth	

was	 seen	 in	 company	 records	 of	 trouser	
sizes,	 the	 respective	 percentages	 of	
drivers	 and	 conductors	 having	 a	 waist	
measurement	 greater	 than	 36	 inches	
rising	from	7.9%	and	3.4%	when	recruited	
(at	ages	25-29	years)	to	39.7%	and	22.4%	
at	ages	45-49	years	and	63.2%	an	43.0%	
at	ages	55-59	years.	There	was	an	 initial	
difference	in	waist	circumference	of	about	
one	 inch	 when	 the	 men	 were	 first	
recruited	 to	 the	bus	system	(presumably	
reflecting	 some	 initial	 selection	 of	
sedentary	vs.	active	work),	but	by	the	time	
of	 retirement,	 this	 difference	 had	
increased	 to	 about	 2	 inches.	 Further	
studies	 of	 London	 busmen	 found	 a	
corresponding	1	mm	difference	of	average	
skinfold	 thicknesses	 between	 conductors	
and	 drivers	 (8	 vs.	 9	 mm)	 at	 recruitment	
(Oliver,	1967).	
	
Emergency	workers	
Emergency	 workers	 may	 on	 occasion	

be	 asked	 to	 perform	 very	 hard	 physical	
work,	and	entry	standards	often	require	a	
high	 level	 of	 physical	 fitness,	 but	
personnel	 do	 not	 necessarily	 show	 low	
levels	 of	 body	 fat	 even	 on	 entry	 to	 such	
occupations;	 indeed,	 one	 U.S.	 study	
(Tsismenakis	 et	 al.,	 2009)	 found	 that	 a	
third	 of	 recruits	 had	 an	 initial	 BMI	 >	 30	
kg/m2.	 After	 recruitment,	 body	
composition	may	worsen,	particularly	if	it	
is	 necessary	 for	 the	 worker	 to	 sit	 long	
periods	waiting	for	an	emergency	to	occur.	
This	problem	seems	particularly	prevalent	
in	firefighters.	In	a	sample	of	115	Missouri	
firefighters	(Poston	et	al.,	2011),	54	were	
rated	as	over-weight,	and	38	as	obese	(12	
with	class	II	or	class	III	obesity).	Studies	of	
firefighters	 have	 also	 found	 body	 mass	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	77	

increasing	by	2.6	kg	over	the	first	5	years	
of	 employment,	 with	 the	 prevalence	 of	
obesity	 increasing	 from	 33.7	 to	 40.4%	
(Soteriades	 et	 al.,	 2005),	 and	 body	 mass	
increasing	 by	 as	 much	 as	 1.4	 kg/year	
(Elliot	et	al.,	2007).	 In	addition	 to	sitting	
for	 long	 periods	 in	 the	 fire-hall,	
firefighters	face	problems	of	adaptation	to	
shift	work,	with	disturbances	of	their	diet	
and	sleep	patterns.	
Even	with	the	more	sustained	physical	

demands	 encountered	 in	 fighting	 forest	
fires,	 Budd	 (2001)	 found	 body	 fat	
percentages	 in	 Australian	 fire-fighters	
ranging	 from	 7%	 to	 27%.	 The	 focus	 in	
many	 of	 the	 fitness	 tests	 used	 when	
recruiting	wild-land	 firefighters	has	been	
on	aerobic	 fitness	and	muscular	strength	
rather	than	body	fat	content,	although	the	
latter	does	have	a	marked	influence	upon	
their	 risk	 of	 injury	 and	 thus	 of	
absenteeism.	
	
Employee	fitness	programmes		
A	 number	 of	 companies	 where	 most	

jobs	 involve	 little	 physical	 effort	 have	
introduced	worksite	 fitness	programmes;	
these	can	result	in	a	10-30%	reduction	of	
skin-fold	 readings	 among	 participants	
(Shephard,	1996).	
	
Military	recruits.	
Current	 evidence	 of	 an	 inverse	

relationship	 between	 the	 physical	
demands	of	occupational	activity	and	body	
fat	 content	 is	 most	 evident	 in	 studies	 of	
the	 armed	 forces,	 especially	 in	 reports	
concerning	elite	fighting	groups	(Table	4),	
where	 some	 extremely	 low	 values	 have	
been	observed.	
Campos	 et	 al.	 (2017)	 documented	 the	

effects	of	12	weeks	of	basic	training	upon	
the	 body	 composition	 of	 18-19-year-old	
male	 recruits	 to	 the	 regular	 Brazilian	
military.	 Over	 a	 period	 of	 preliminary	
conditioning,	 the	 average	 thickness	 of	 3	
skin-folds	decreased	from	14.1	mm	to	11.2	
mm,	and	the	corresponding	prediction	of	
body	 fat	 dropped	 from	 14.0%	 to	 11.6%,	
with	 a	 parallel	 dramatic	 increase	 in	
maximal	oxygen	intake	(from	35.2	to	49.8	
ml/[kg.min]).	
Farina	et	al.	 (2017)	examined	a	group	

of	 50	 male	 U.S.	 "special	 forces"	 who	
engaged	 in	 155	 min	 of	 aerobic	 training	
and	 180	 min	 of	 strength	 training	 every	
week.	A	skin-fold	based	determination	of	
body	 fat	showed	readings	of	10.4%	both	
initially	 and	 after	 3-6	 months	 of	
deployment.	
Malavolti	 et	 al.	 (2008)	 used	 3	

techniques	 (skin-folds,	 air	
plethysmography	 and	 dual	 x-ray	
absorptiometry	to	determine	the	body	fat	
content	 in	 27	 male	 naval	 recruits	 who	
successfully	 completed	 9	 months	 of	

Table	4:	Levels	of	body	fat	observed	in	studies	of	military	groups.	
	

Author	 Group	 Body	fat	 Methodology	
Campos	et	al.	(2017)	 130	Brasilian	military	

recruits,	12	wk	basic	training	
Initial	14.0%	
Final	11.6%	
Skin-folds	14.1,	11.2	mm	

Skin-fold	prediction	

Farina	et	al.	(2017)	 50	male	U.S.	Special	Forces	 10.4%	 Skin-fold	prediction	
Malavolti	et	al.	(2008)	 27	male	naval	recru9its	

undergoing	special	training	
Initial	13.0%,	
6	months	=	11.5%	
skin-folds	9.4	and	8.3	mm	

DXT	

Nindl	et	al.	(2007)		 50	male	participants,	U.S.	
Army	4-wk	Ranger	Course	

18.5%	initial	
8.8%	final		

DXT	

Paff	&	Kometa	(2000)	 39	Polish	paratroopers,	18	
months	training	

15.0%	initial	
13.1%	final	

Skin-fold	predictions	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	78	

intensive	training.	On	average,	dual	x-ray	
estimates	 of	 body	 fat	 content	 fell	 from	
13.0%	 to	 11.5%,	 while	 the	 average	 of	 6	
skin-fold	 thicknesses	 decreased	 from	 9.4	
mm	 to	 8.3	 mm	 during	 the	 9	 months	 of	
observation.	
Nindl	et	al.	(2007)	also	used	dual	x-ray	

absorptiometry	to	demonstrate	that	an	8-
week	 U.S.	 army	 ranger	 course	 with	 an	
energy	expenditure	of	32-40	MJ/day,	and	
an	energy	deficit	of	~4	MJ/day	decreased	
body	fat	from	an	initial18.5%	to	8.8%.	
Paff	&	Kometa	(2000)	noted	a	decrease	

of	 anthropometrically	 determined	 body	
fat	 in	 39	 presumably	 male	 Polish	
paratroopers	over	18	months	of	training,	
with	 values	 falling	 from	 15.0%	 at	
recruitment	to	13.1%	after	18	months	of	
military	 service.	 Anaerobic	 capacity	 and	
muscle	strength	also	increased,	but	there	
was	no	change	of	maximal	oxygen	intake.		
	
Traditional	 Amish	 and	 Mennonite	
groups	vs.	sedentary	North	Americans		
The	 Amish	 people	 have	 their	 historic	

roots	 in	 Swiss	 German	 Anabaptist	
churches.	 They	 are	 a	 community	 known	
particularly	 for	 their	 plain	 dress	 and	
simple	 lifestyle.	 Many	 of	 the	 old-order	
Amish	 believers	 migrated	 to	 North	
America	 in	 the	 early	 1700s;	 currently,	
there	 are	 some	 200,000	 living	 in	 the	
eastern	part	of	the	U.S.	and	around	1600	
in	 southern	 Ontario.	 Old-order	
Mennonites	are	a	related	sect,	originating	
in	 Switzerland	 and	 southern	 Germany;	
they	 also,	 have	 retained	 a	 plain	 lifestyle.	
The	 traditional	 Amish	 and	 a	 segment	 of	
Old-order	 Mennonites	 use	 no	 modern	
power	 tools	or	modern	vehicles	on	 their	
farms,	 and	 in	 consequence	 their	 daily	
routine	 requires	 very	 high	 daily	 energy	
expenditures	 (Table	 5).	 Their	 children,	
likewise,	 are	 allowed	 no	 mechanical	 or	
electronic	 devices,	 they	 are	 expected	 to	

help	 around	 the	 farm,	 and	 they	 usually	
walk	 a	 substantial	 distance	 to	 school,	 so	
that	they	also	incur	very	high	daily	energy	
expenditures.	
	
Adult	Amish	and	Old-order	Mennonites	
The	 impact	 of	 the	 simple	 and	

traditional,	non-mechanized	lifestyle	upon	
habitual	 physical	 activity	 can	 be	 seen	 in	
daily	step	counts	 for	 the	Amish	and	Old-
order	Mennonite	populations.		
Bassett	et	al.	 (2004)	examined	98	U.S.	

Amish	 adults	 aged	 18-75	 years	 who	 still	
engaged	 in	 labour-intensive	 and	 non-
mechanized	 farming,	 finding	 daily	
pedometer	 step	 counts	 averaging	 18,425	
for	 the	 men	 and	 14,196	 for	 the	 women.	
Moreover,	 use	 of	 the	 International	
Physical	Activity	Questionnaire	suggested	
that	 vigorous	 physical	 activity	 was	
undertaken	for	10	h/week	by	the	men	and	
3.4	 hours/week	 by	 the	 women.	 On	 the	
basis	of	BMI	(a	somewhat	fallacious	index	
in	 a	 group	 performing	 much	 heavy	
muscular	 work),	 only	 25%	 of	 men	 and	
27%	 of	 women	 were	 classed	 as	
overweight,	and	none	of	the	men	and	only	
9%	of	the	women	were	classed	as	obese.	
Bio-impedance	 data	 indicated	 body	 fat	
values	of	9.4%	 in	 the	men	and	25.3%	 in	
the	women.	
Glick	et	al.	(1998)	compared	traditional	

Old-order	 Mennonites	 (87	 men	 and	 72	
women)	with	 the	general	U.S.	population	
as	both	groups	were	examined	in	the	1987	
NHANES	survey.	The	Mennonite	men	and	
women	both	had	a	somewhat	 lower	self-
reported	BMI	than		
that	 seen	 in	 the	 general	 U.S.	 population,	
despite	 a	 larger	 food	 intake	 among	 the	
Mennonites	(20%	greater	in	the	men,	and	
39%	 greater	 in	 the	 women).	 Moreover,	
because	 the	 Mennonites	 used	 their	
muscles	extensively	to	undertake	physical	
tasks,	 the	 difference	 of	 body	 fat	 content	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	79	

from	sedentary	 individuals	was	probably	
substantially	 greater	 than	 that	 inferred	
from	the	BMI	data.	
In	 contrast	 to	 the	 low	 body	 fat	 levels	

seen	 in	 these	 two	 reports,	 Fuchs	 et	 al.	
(1990)	 examined	 400	 Amish,	 many	 of	
whom	 had	 moved	 from	 their	 farms	 to	
small	 towns	 in	Ohio.	Comparing	 the	self-
reported	 BMI	 of	 this	 group	 with	 that	 of	
773	 controls,	 they	 found	 no	 inter-group	
difference	 in	 the	 prevalence	 of	 obesity	
among	 the	 men	 and	 higher	 levels	 of	
obesity	 in	 the	Amish	women	 than	 in	 the	
controls.	 Precise	 measurements	 of	
physical	activity	were	not	obtained	for	this	

population,	 but	 the	 data	 suggested	 that	
protection	 against	 the	 accumulation	 of	
body	 fat	 was	 not	 an	 inherited	
characteristic;	the	advantage	of	the	Amish	
soon	 being	 lost	 when	 they	 ceased	 their	
involvement	in	the	heavy	physical	 labour	
of	 traditional	 agriculture.	 A	 third	 of	 the	
men	 studied	 by	 Fuchs	 and	 associates	
worked	in	small	local	industries,	and	many	
of	 the	remainder	of	 their	sample	worked	
as	 labourers	 on	 non-Amish	 farms,	
although	about	 two	 thirds	of	 the	women	
were	 still	 full-time	 home-makers.	 Among	
factors	 contributing	 to	 the	 shift	 in	 body	
composition,	 Fuchs	 et	 al.	 (1990)	 noted	

Table	5:	Habitual	physical	activity	and	measures	of	obesity	in	traditional	Amish	and	Mennonite	
populations.	
	
Author	 Sample	 Habitual	physical	

activity	
Measure	of	obesity	 Comments	

Bassett	
et	al.	
(2004)	

98	Amish	adults,	aged	
18-75	yr	

Daily	step	count	18,425	
(M),	14,196	(F)	

Overweight	25%	(M),	
27%	(F),	obese	0%	
(M),	9%	(F)	
Impedance	fat	9.4%	
(M),	25.3%	(F)	

Vigorous	PA		
10	h/wk	(M),		
3.4	h/wk	(F)		

Bassett	
et	al.	
(\2007)	

139	Amish	youth	aged	
6-18	yr		

Weekday	step	count	
17,525	

Only	7.2%	
overweight,		
1.4%	obese	

	

Esliger	et	
al.	
(2010)	

68	Amish,	120	
Mennonite,	132	rural	
&	93	urban	
contemporary	
children	aged	8-13	yr	

Actigraph	data	show	less	
moderate	&	vigorous	
activity	in	contemporary	
children	than	in	Amish	&	
Mennonite	

BMI	17.2	vs.	19.8	vs.	
19.0,	18.8	kg/m2;		
Triceps	skin-fold	10.6	
vs.	16.5,	vs.	17.7,	18.0	
mm.		

	

Fuchs	et	
al.	
(1990)	

400	Amish	adults,	773	
controls	

	 No	difference	of	
obesity	in	men	
(28%),	Amish	women	
more	obese	than	
controls	(40%	vs.	
31%)	

Amish	no	longer	
pursuing	
traditional	
agriculture	

Glick	et	
al.	
(1998)	

80	M,	69	F	Old-order	
Mennonite,	vs.12,275	
M,	14,815	F	US	
NHANES	population	

	 BMI	24.8,	24.7	in	
Mennonites,	vs.	26	
kg/m2	in	M	&	F	
NHANES	sample		

Old-order	
Mennonites	vs.	
1987	NHANES	
survey	

Tremblay	
et	al.	
(2008)	

127	Mennonite,	275	
contemporary	
children	aged	8-13	yr	

Moderate	&	vigorous	
physical	activity:	
Mennonite	12	(urban)	or	
18	(rural)	more	min/day	
than	peers	

Skin-fold	thicknesses	
greater	in	both	urban	
and	rural	
contemporary	
children	than	in	
Mennonites	

Mennonite	
children	also	had	
greater	aerobic	
fitness	and	
strength	than	their	
peers	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	80	

that	food	was	usually	still	very	plentiful	in	
the	 urban	 Amish	 households,	 and	
everything	tended	to	be	eaten	from	over-
filled	 plates	 to	 avoid	 any	 sinful	 waste	 of	
food.	
Most	Amish	still	complete	only	a	limited	

amount	of	 formal	education	(nearly	80%	
of	 the	 Fuchs	 sample	 had	 not	 even	
completed	 Grade	 8),	 so	 there	 is	 still	 a	
dearth	 of	 Amish	 professionals	 such	 as	
physicians,	 and	 the	 Amish	 are	 often	
reluctant	to	seek	health	advice	from	those	
living	 outside	 of	 their	 religious	
community.	In	the	U.S.,	reluctance	to	visit	
a	 physician	 is	 reinforced	 by	 a	 low	
monetary	 income	 and	 a	 lack	 of	 health	
insurance.	 Further,	 the	 Amish	 seem	 less	
likely	 than	 their	 peers	 to	 engage	 in	
voluntary	leisure	activity	if	they	find	that	
they	 are	 gaining	 weight	 (Levinson	 et	 al.,	
1989).	Some	17%	of	the	Amish	studied	by	
Fuchs	 et	 al.	 were	 following	 exercise	
programmes,	 as	 compared	 with	 37%	 of	
their	 peers.	 The	 Amish	 women	 also	
experienced	 a	 higher	 average	 number	 of	
pregnancies	 than	 their	 peers	 in	 the	
general	 Ohio	 population.	 Finally,	 since	
obesity	 was	 less	 stigmatized	 among	 the	
Amish	than	in	the	rest	of	U.S.	society,	more	
of	 them	may	have	been	willing	 to	report	
high	body	weights	than	their	peers.	
	
Children	 of	 traditional	 Amish	 and	
Mennonite	communities	
Traditional	 Amish	 and	 Mennonite	

children	have	a	much	higher	level	of	daily	
physical	 activity	 than	 their	 North	
American	peers,	in	part	because	they	help	
with	chores	around	the	farm,	and	in	part	
because	 they	do	not	occupy	 their	 leisure	
time	 by	 playing	 sedentary	 electronic	
games.		
Tremblay	et	al.	(2005;	2008)	compared	

58	 Amish	 and	 106	 Mennonite	 children	
(the	latter	drawn	from	a	group	that	used	

some	electrical	equipment	on	their	farms)	
with	267	contemporary	Canadian	children	
aged	 8-13	 years.	 The	 level	 of	 moderate	
and	 vigorous	 physical	 activity	 was	 much	
higher	in	the	Amish	and	Mennonite	youth	
than	 in	 the	 other	 Canadians.	 Thus,	 the	
Mennonite	 boys	 engaged	 in	 3	 h/day	 of	
vigorous	 physical	 activity,	 and	 the	
Mennonite	girls	were	vigorously	active	2	
h/day	 (Tremblay	 et	 al.,	 2005).	 The	
Mennonite	 boys	 helped	 extensively	 with	
farm	 chores,	 while	 the	 Mennonite	 girls	
spent	much	of	their	evenings	sewing	and	
cooking.	 The	 Mennonite	 youth	 showed	
correspondingly	 lower	 skin-fold	 readings	
and	 higher	 levels	 of	 aerobic	 fitness	 and	
muscular	 strength	 than	 their	
contemporary	 peers.	 Triceps	 skin-fold	
values,	 as	 estimated	 from	 the	 published	
graphs,	were	around	11.5	and	12.5	mm	for	
the	Amish	boys	and	girls,	respectively,	15	
and	 17	 mm	 for	 the	 Mennonite	 boys	 and	
girls	 and	 16	 and	 19.5	 mm	 for	 male	 and	
female	controls.	
Bassett	et	al.	(2007)	measured	the	step	

count	in	139	Amish	youth	aged	6-18	years,	
finding	a	very	high	average	weekday	count	
(17,525	steps/day).	Values	 for	weekends	
were	 with	 somewhat	 lower,	 in	 part	
because	of	time	spent	sitting	in	church	on	
Sundays;	 the	 7-day	 average	 was	 thus	
15,563	 steps/day.	 Skin-folds	 were	 not	
measured	 in	 this	 sample,	 but	 measured	
height	and	weight	data	indicated	that	only	
7.2%	 of	 this	 group	 would	 be	 classed	 as	
overweight	and	1.4%	as	obese.	
Esliger	 et	 al.	 (2010)	 compared	 Amish,	

Mennonite	 (not	 strict	 Old-order)	 and	
contemporary	 Canadian	 children,	 the	
latter	 being	 drawn	 from	 both	 urban	 and	
rural	settings.	Accelerometer	data	showed	
that	 the	 Amish	 children	 (and	 to	 a	 lesser	
extent	the	Mennonite	children)	engaged	in	
more	 moderate	 and	 vigorous	 physical	
activity	 than	 their	 contemporary	 peers.	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	81	

There	 were	 parallel	 differences	 in	 the	
average	BMI	for	the	4	groups	(Amish	7.2	
kg/m2,	Mennonites	19.8	kg/m2,	rural		
controls	 19.0	 kg/m2,	 and	 urban	 controls	
18.8	kg/m2,	with	much	larger	inter-group	
differences	in	triceps	skin-fold	thicknesses	
(respective	values	of	10.6,	16.5,	17.7	and	
18.0	mm).			

	
Unacculturated	Inuit	
Another	 group	 who	 had	 very	 high	

energy	expenditures	when	they	were	first	
studied	 (around	 the	 year	 1970)	 were	
unacculturated	 Inuit	 living	 on	 Canada's	
Arctic	coast.	Many	of	this	population	then	
followed	 a	 traditional	 hunter-gatherer	
lifestyle.	 Measurements	 made	 by	 a	
combination	 of	 Kofrany-Michaelis	
respirometer	 and	 time-and-motion	
analyses	 suggested	 an	 average	 daily	
energy	 expenditure	 of	 15.4	 MJ	 during	
hunting	 trips	 (Shephard	 &	 Rode,	 1996).	
This	 total	 was	 accumulated	 mainly	
through	 very	 prolonged	 bouts	 of	
moderately	heavy	endurance	exercise.	In		
the	summer	months,	if	the	weather	was	

favourable	 for	 hunting,	 physical	 activity	
would	 continue	 over	 much	 of	 a	 24-hour	
period.	
Those	 following	 this	 lifestyle	 showed	

correspondingly	 low	 indices	 of	 body	 fat	
content	(Table	6).	The	body	mass	indices	
were	not	remarkable,	in	part	because	the	
population	 was	 muscular,	 and	 also	

because	 of	 a	 short	 stature.	 However,	 the	
average	thickness	of	the	triceps	skin-folds	
was	 remarkably	 low,	 especially	 in	 the	
male	 members	 of	 the	 community.	 There	
also	seemed	evidence	for	the	development	
of	obesity	 in	 the	oldest	group	of	women,	
most	 of	 whom	 had	 undergone	 multiple	
pregnancies.	 Comparisons	 of	 body	 fat	
content	 with	 the	 urban	 Canadian	
population	 must	 however	 be	 made	 with	
care.	Not	only	were	the	Inuit	much	more	
active	 physically,	 but	 they	 were	 also	
obtaining	a	large	proportion	of	their	daily	
food	from	their	hunting	activities.		
Percentages	 of	 body	 fat	 in	 the	 Inuit	

sample	were	determined	not	only	by	skin-
fold	 measurements,	 but	 also	 by	 the	
dilution	 of	 deuterated	 water	 (Table	 7).	
The	 latter	 methodology	 requires	 making	
assumptions	 about	 tissue	 hydration,	
which	 may	 differ	 between	 the	 Inuit	 and	
urban	Canadians.	Certainly,	 the	estimates	
of	body	fat	content	made	by	this	method	

(which	 reflects	 total,	 rather	 than	
superficial	 fat)	 were	 somewhat	 higher	
than	 those	 obtained	 from	 skin-fold	
predictions.	 Possibly,	 the	 Inuit	 store	 a	
greater	 than	 normal	 proportion	 of	 their	
body	 fat	 internally.	 Nevertheless,	 a	 low	
total	body	 fat	 content	was	confirmed	 for	
the	males	at	all	ages.	Values	for	the	women	
were	also	lower	than	would	be	expected	in	
sedentary	 urban	 city-dwellers,	 although	

Table	6:	Average	BMI	and	triceps	skin-fold	
thickness	of	Inuit	residents	of	Igloolik	in	
1970	(Shephard	and	Rode,	1996).	
	
Age	
(yr)	

BMI	(kg/m2)	 Triceps	skin-
fold(mm)	

Men	 Wome
n	

Men	 Women	

20-29	 24.4	 23.2	 5t.5	 8.5	
30-39	 24.9	 23.9	 6.3	 9.2	
40-49	 25.3	 23.7	 5.4	 7.0	
50-59	 25.8	 27.5	 7.9	 19.0	

Table	7:		Percentage	of	body	fat	in	Igloolik	
Inuit	in	the	year	1970,	as	determined	by	the	
dilution	of	deuterated	water	(Mertens	et	al.,	
1996).	
	
Age	group	
(yr)	

Body	fat	(%)	
Men	 Women	

15-19	 12.8	 18.4	
20-29	 15.3	 20.8	
30-39	 11.9	 25.4	
40-49	 -	 26.1	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	82	

there	was	apparently	some	accumulation	
of	fat	as	the	Inuit	women	became	older.		
Low	 levels	 of	 skin-fold	 thickness	 are	

also	 apparent	 in	 data	 for	 indigenous	
circumpolar	 children	 from	 this	 era,	
whether	 measured	 at	 Igloolik	 in	 the	
Canadian	arctic,	or	on	nGnassan	children	
living	 in	 the	 Russian	 arctic	 (Shephard	 &	
Rode,	1996)	(Table	8).		
	
Physically	 active	 vs.	 sedentary	 urban	
populations	
Coefficients	 of	 correlation	 between	

estimates	of	body	fat	content	and	habitual	
physical	 activity	 have	 been	 examined	
frequently	 in	 modern	 urban	 populations,	
often	 in	 the	 context	 of	 evaluating	 the	
reliability	and	validity	of	physical	activity	
questionnaires.	 The	 findings	 are	 of	
particular	interest	in	the	context	of	public	
health,	 as	 (unlike	 groups	 such	 as	
marathoners	 and	 hunter-gatherers),	 the	
activities	involved	are	at	an	intensity	and	
volume	that	are	likely	to	prove	acceptable	
to	many	current	urban	populations.	Only	a	
few	of	 the	many	available	studies	can	be	
discussed	 here.	 Many	 of	 these	
investigations	are	based	on	questionnaire	
estimates	of	habitual	physical	activity,	but	
some	 have	 also	 focused	 on	 objective	
measurements	obtained	by	pedometer	or	
accelerometer.	
	
Subjective	data	
In	 general,	 self-reported	 habitual	

physical	activity	and	body	fat	have	shown	
modest	negative	inter-correlations	(Table	

9),	although	because	of	errors	inherent	in	
the	 questionnaire	 approach	 to	 the	
assessment	 of	 physical	 activity,	
associations	 have	 necessarily	 been	
attenuated	 relative	 to	 their	 true	 values.	
For	the	best	designs	of	questionnaire,	and	
in	samples	with	substantial	differences	of	
physical	activity	between	those	classed	as	
active	 and	 sedentary,	 correlations	 have	
been	as	high	as	-0.40,	but	in	other	studies	
where	few	of	the	sample	have	engaged	in	
any	 significant	 physical	 activity,	 much	
lower	 correlation	 coefficients	 have	 been	
reported.	Correlations	have	generally	been	
larger	for	reports	of	vigorous	activity	than	
for	 moderate	 or	 light	 physical	 activity,	
probably	because	engagement	in	vigorous	
activity	 is	 clearly	 perceived	 by	 the	
individual,	and	is	reported	more	precisely.	
Differences	 of	 body	 fat	 content	 between	
the	most	active	and	the	least	active	groups	
of	 individuals	 have	 been	 in	 the	 range	 3-
4%.	
Ball	 et	 al.	 (2001)	 studied	 1302	

Australians	 aged	 18-78	 years,	 looking	 at	
relationships	between	leisure	activity,	BMI	
and	an	estimate	of	body	fat	based	on	the	
average	 of	 6	 skin-folds.	 They	 noted	 that	
the	 odds	 of	 finding	 a	 BMI	 within	 the	
normal	range	was	significantly	related	to	a	
high	reported	level	of	leisure	activity	(less	
so	 in	men	than	in	women,	the	respective	
odds	 ratios	 being	 1.8	 in	 men,	 and	 2.6	 in	
women,	p	=	0.04,	<0.01).	After	controlling	
for	 age	 and	 differences	 of	 educational	
level,	 in	 the	 women,	 moderate	 or	 high	
intensity	 leisure	 activity	 was	 also	

Table	 8:	 Average	 skin-fold	 thicknesses	 (mm)	 as	 measured	 in	 Igloolik	 children	 (1989/90)	 and	
nGnassan	children	(1991)(Shephard	and	Rode,	1996).	
	
Age	(yr)	 Boys	 Girls	

Igloolik	 nGnassan	 Igloolik	 nGnassan	
11.0-12.9	 7.9	 6.7	 11.7	 8.0	
13.0-14.9	 9.0	 3.1	 11.6	 10.2	
15.0-16.9	 7.9	 6.1	 18.2	 11.2	
17.0-18.9	 9.6	 6.4	 16.8	 12.7	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	83	

associated	with	a	low	body	fat	content.		
However,	 the	 BMI	 and	 skin-fold	

readings	 were	 unrelated	 to	 estimates	 of	

occupational	or	domestic	physical	activity	
in	either	sex.	

Table	9:	Relationship	between	self-reported	habitual	physical	activity	and	body	fat	content.	
Author	 Sample	 Fat	measure	 Correlation	with	body	

fat	or	other	measure	
Comments	

Ball	et	al.	
(2001)		

1302	M	&	F	
aged18-78	yr	

BMI	and	6	skin-
folds	

Odds	of	normal	BMI	if	
high	leisure	activity	1.76	
(M),	2.59	(F)	(p	=	0.04,	
<0.01)	

Adjusted	for	age	&	
education;	no	effect	of	
occupation	or	
domestic	physical	
activity	

Bouchard	et	
al.	(1983)	

150	children,	
150	adults	

Skin-fold	estimate	
of	body	fat	

-0.08	to	-	0.13	 	

Bradbury	et	
al.	(2017)	

119,230	M,	
140,578	F	
	

BMI,	Bio-
impedance	
estimate	of	body	fat	

Body	fat	23.4%	vs.	
26.3%	M,	33.9%	vs.	
37.9%	F).	

Most	active	(>100	
MET-h/wk	vs.	least	
active	(<5	MET-h/wk)	

Ekelund	et	al.	
(2005)	

17-yr	old	
students	(190	
M,	255	F)	

Air-displacement	
plethysmography	

-0.18	(M)	
-0.03	(F)	

All	activity	levels	low	
in	F	students	

Folsom	et	al.	
(1985)	

3448	Minnesota	
adults	aged	25-
74	years	

BMI	 -0.10	(F)	to		
-	0.11	(M)	with	Log	of	
reported	heavy	
activities	per	day	

Activity	levels	
generally	low	in	this	
population	

Godin	&	
Shephard	
(1985)	

163	men	and	
143	women	

Skin-fold	estimate	
of	body	fat	

-	0.21	 Significant	r	for	
strenuous	but	not	for	
moderate	or	light	
physical	activity	

Jacobs	et	al.	
(1993)	

78	men	and	
women	aged	20-
59	yr	

Hydrostatic	
measures	of	body	
fat	

-0.37	to	-	0.44	for	best	
questionnaires,	0.52	for	
accelerometer	

Comparison	of	10	
questionnaires	

Martinez	et	al.	
(1999)	

15,239	M	&	F	
aged	>	15	yr	

Self-reported	BMI	 Low	exercise	
participation	
(p<0.0001),	lack	of	
interest	in	PA,	sitting	at	
work	(p	=	0.0004)	all	
predictors	of	obesity	

	

Richardson	et	
al.	(1995	)	

78	men	and	
women	aged	20-
59	years	

Hydrostatic	
estimate	of	body	fat	
%	

-0.37	in	men,		
-0.44	in	women	

Correlation	seen	for	
sport	and	deliberate	
exercise,	less	apparent	
for	non-sport	light	
physical	activity	

Wanner	et	al.	
(2016)	

1440	M,	1602	F	 Bio-impedance	
estimate	of	body	
fat,	BMI,	waist	
measurements	

OR	total	physical	
activity	0.60		
Moderate	phys.	activity	
0.67	Vigorous	phys.	
activity	0.59	
Sitting	1.59	

Odds	ratio	for	>32%	
fat,	high	vs.	low	level	of	
activity	

Washburn	et	
al.	(1991)	

306	M,	417	F,	
average	age	
39.5	years	
Boston	
population	

Log	BMI	 -0.13	(-0.11	M;	-	0.17	F);	
overall	value	of	-0.41	
after	co-variate	
adjustment	

BMI	for	most	active	
tertile	25.1,	least	
active	26.7	kg/m2	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	84	

Bouchard	 et	 al.	 (1983)	 used	 a	 3-day	
diary	record	to	estimate	habitual	physical	
activity	in	a	group	of	150	children	and	150	
adults	 who	 were	 participating	 in	 the	
Quebec	family	study.	They	noted	quite	low	
negative	 correlations	 between	 skin-fold	
assessments	 of	 body	 fat	 content	 and	 the	
diary	records	of	physical	activity	(-0.08	to	
-0.13).	
Bradbury	et	al.	(2017)	stratified	a	very	

large	 sample	 of	 English	 adults	 (119,230	
men	and	140,578	women	participating	in	
the	U.K.	Biobank	2006-2010)	 in	 terms	of	
their	bio-impedance	estimates	of	body	fat	
content.	 The	 results	 showed	 that	
individuals	 reporting	 the	 most	 physical	
activity	on	the	International	Physical		
Activity	 Questionnaire	 (100	 MET-

h/week)	had	consistently	 lower	body	 fat	
content	than	the	 least	active	members	of	
the	 same	 sample	 (<5	 MET-h/week)	
(23.4%	vs.	26.3%	in	men,	33.9%	vs.	37.9%	
in	 women).	 The	 measured	 BMI	 values	
were	also	lower	for	the	active	group	(27.1	
vs.	 28.0	 kg/m2	 in	 the	 men,	 25.7	 vs.	 27.2	
kg/m2	in	the	women).		
Folsom	 et	 al.	 (1985)	 examined	 a	

representative	 sample	 of	 Minnesota	
residents	 aged	 25-74	 years.	 Levels	 of	
physical	 activity	 in	 this	 sample	 were	
generally	 low,	 but	 nevertheless	 small	
negative	 correlations	 were	 shown	
between	 the	 log	 of	 reported	 leisure-time	
physical	 activity	 (kCal/day)	 and	 body	
mass	 index,	 both	 in	 men	 (-0.11)	 and	 in	
women	 (-0.10).	 Leisure	 activity	 was	
associated	 with	 a	 good	 education	 and	 a	
type	A	personality,	but	the	authors	did	not	
adjust	 for	 these	 factors	by	a	multivariate	
analysis.	
Godin	&	Shephard	(1985)	sought	data	

to	 validate	 their	 simple	 physical	 activity	
questionnaire	tin	a	sample	of	163	men	and	
143	women	recruited	from	the	University	
of	Toronto	community.	Correlations	with	a	

four	site	skin-fold	assessment	of	body	fat	
were	substantial	and	highly	significant	for	
reports	 of	 strenuous	 leisure	 physical	
activity	(activity	at	an	estimated	intensity	
of	9	METs,	 r	=	 -	0.21),	but	were	smaller	
and	not	statistically	significant	for	reports	
of	 moderate	 (6	 METs)	 or	 light	 (3	 METs)	
intensity	physical	activity.	
Jacobs	 et	 al.	 (1993)	 involved	 78	 men	

and	 women	 aged	 20-59	 years	 in	 a	
simultaneous	 assessment	 of	 10	 different	
physical	 activity	 questionnaires,	 using	
accelerometer	 measurements	 as	 their	
criterion	 data.	 Hydrostatic	 estimates	 of	
body	fat	showed	correlations	ranging	from	
-0.37	 to	 -0.44	with	what	were	 judged	as	
the	best	of	the	10	questionnaires,	and	0.52	
with	the	accelerometer	data.	
Martinez	 et	 al.	 (1999)	 examined	 the	

self-reported	 BMI	 of	 15,239	 Europeans	
aged	>15	years.	About	10%	of	 the	group	
was	 obese	 on	 this	 criterion,	 and	 the	
prevalence	of	overweight	was	35.6%	and	
25.6%	in	men	and	women	respectively.	A	
low	 participation	 in	 various	 leisure-time	
physical	activities,	a	lack	of	interest	(a	lack	
of	pre-contemplation)	in	being	involved	in	
exercise/physical	 activity	 and	 an	
increasing	number	of	hours	sitting	down	
at	work	were	all	predictors	of	obesity.	
Richardson	 et	 al.	 (1995)	 used	 the	

Baecke	 questionnaire	 and	 an	
accelerometer	 to	 assess	 physical	 activity	
patterns	in	78	adults	aged	20-59	years.	A	
hydrostatic	estimate	of	 the	percentage	of	
body	 fat	 showed	 a	 relatively	 strong	
negative	relationship	(r	=	-0.37	in	men,	 -
0.44	in	women)	with	participation	in	sport	
and	 deliberate	 exercise,	 but	 less	
substantial	 relationships	 were	 found	 for	
other	forms	of	light	activity	(-0.09	in	men,	
-0.51	 in	 women).	 Unfortunately,	 the	
associations	 between	 body	 fat	 and	
accelerometer	data	were	not	 reported	 in	
this	study.		



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	85	

Wanner	et	al.	(2016)	estimated	body	fat	
levels	 in	 a	 sample	 of	 Swiss	 adults	 (1440	
males	 and	 1602	 females),	 using	 the	 bio-
impedance	 technique.	 Relationships	 with	
habitual	 physical	 activity	 as	 assessed	 by	
the	 International	 Questionnaire	 of	
Physical	Activity	were	evaluated	in	terms	
of	the	odds	of	developing	>32%	body	fat.	
On	 comparing	 the	 highest	 versus	 the	
lowest	 tertiles,	 the	 odds	 of	 accumulating	
>32%	fat	were	reduced	by	 total	physical	
activity	 (0.60),	 moderate	 intensity	
physical	 activity	 (0.67)	 and	 vigorous	
physical	 activity	 (0.59),	 and	 were	
increased	with	sitting	time	(1.59).	Similar	
relationships	 were	 shown	 for	 BMI,	 waist	
circumference,	 and	 the	 weight-to-height	
ratio.	
Washburn	 et	 al.	 (1991)	 sought	 to	

validate	 general	 use	 of	 the	 Harvard	
Alumni	 Physical	 Activity	 Questionnaire.	
They	examined	306	men	and	417	women	
from	the	Boston	area	with	an	average	age	
of	 39.5	 years.	 Activity	 levels	 were	
separated	 into	 3	 categories,	 as	 proposed	
by	 Paffenbarger	 (<2MJ/week,	 2-8	
MJ/week	 and	 >8	 MJ/week).	 In	 the	 men,	
the	 sample	 was	 divided	 rather	 equally	
between	the	3	activity	categories,	but	39%	
of	 the	 women	 expended	 <2	 MJ/week	 on	
physical	 activity.	 The	 study	 showed	 a	
modest	 unadjusted	 correlation	 between	
the	 questionnaire	 assessment	 of	 habitual	
physical	activity	and	the	log	of	BMI	(-0.11	
in	 men,	 -0.17	 in	 women),	 but	 the	
correlation	 coefficient	 for	 the	 entire	
sample	was	increased	to	-0.41	when	data	
were	 adjusted	 for	 significant	 co-variates	
that	 included	 ethnic	 group,	 cigarette	
smoking	 and	 the	 mean	 monthly	
temperature	at	the	time	of	the	interview.		
Ekelund	 et	 al.	 (2005)	 examined	

correlations	 in	 a	 somewhat	 younger	 age	
group,	looking	at	the	relationship	between	
body	 composition	 as	 assessed	 by	 an	 air-

displacement	 plethysmograph	 and	 a	
questionnaire	estimate	of	physical	activity	
in	 Swedish	 children	 aged	 ~17	 years.	 A	
significant	 negative	 relationship	 between	
the	2	variables	was	found	in	the	boys	(r	=	-
0.18),	but	not	in	the	girls	(r	=	-0.03),	who	
in	most	cases	had	very	low	average	levels	
of	 habitual	 physical	 activity.	 The	
difference	 in	 exercise	 behaviour	 was	
reflected	in	the	average	body	fat	levels	for	
boys	and	girls	(16.3%	vs.	29.3%).	
  
Objectively	monitored	physical	activity	
Estimates	 of	 habitual	 physical	 activity	

as	 determined	 by	 pedometer,	
accelerometer	 or	 Sense-Wear	 arm	 band	
are	 generally	 more	 accurate	 than	 those	
inferred	from	responses	to	questionnaires,	
although	 there	 remain	 some	 forms	 of	
physical	 activity	 such	 as	 climbing	 a	 hill,	
cycling	 and	 swimming	 that	 are	 still	 not	
assessed	 accurately	 by	 a	 pedometer	 or	
accelerometer.	 Probably	 because	 of	 the	
greater	 precision	 of	 the	 objective	
measurements,	correlations	with	the	body	
mass	index	or	the	estimated	percentage	of	
body	 fat	 are	 usually	 greater	 for	 the	
objective	than	for	the	subjective	data,	with	
correlation	 coefficients	 sometimes	
reaching	 0.50	 or	 higher	 (Table	 10).	
However,	 the	 apparent	 magnitude	 of	
associations	 has	 unfortunately	 been	
exaggerated	in	some	studies,	because	the	
investigators	have	failed	to	separate	data	
for	male	and	female	subjects;	the	men		
have	 tended	 to	 be	 both	 more	 active	 and	
thinner	than	the	women.	
In	a	study	of	78	men	and	women	aged	

20-59	years,	Jacobs	et	al.	(1993)	found	an	
age	 and	 sex	 adjusted	 correlation	 of	 0.52	
between	 body	 fat	 as	 determined	 by	
hydrostatic	 weighing	 and	 Caltrac	
accelerometer	data	expressed	in	kCal/day,	
although	the	correlation	decreased	to	-	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	86	

0.08	if	the	Caltrac	output	was	expressed	
as	MET-min/day).	
McClung	 et	 al.	 (2000)	 examined	 201	

patients	 who	 were	 undergoing	 surgical	
treatment	for	orthopedic	disorders.	In	this	
sample,	 there	 was	 only	 a	 very	 weak	
correlation	 between	 BMI	 and	 pedometer	
estimates	 of	 daily	 step	 count	 (r	 =	 0.05).	
Nevertheless,	the	patients	with	the	highest	
activity	 levels	 in	 this	 sample	 (>8237	
steps/day)	 had	 an	 average	 BMI	 of	 25.4	
kg/m2,	compared	with	27.4	kg/m2	for		
those	with	very	low	levels	of	daily	physical	
activity	(<3704	steps/day).		
Scheers	et	al.	(2013)	used	a	Sense-Wear	

arm-band	 to	 define	 physical	 activity	
patterns	 in	 370	 middle-aged	 Flemish	
adults.	Abdominal	obesity,	as	indicated	by	
waist	 circumferences,	 showed	 a	 strong	
negative	 partial	 correlation	 (0.61)	 with	

habitual	 physical	 activity	 as	 indicated	 by	
the	 arm-band	 output.	 The	 odds	 ratio	 of	
showing	 abdominal	 obesity	 in	 the	 most	
active	group	(active	>	60	min/day)	was	a	
dramatic	0.15,	relative	to	those	in	the	least	
active	group	(active	<30	min/day).	
Tryon	et	al.	 (1992)	related	body	mass	

index	 to	 a	 pedometer	 measurement	
(distance	 traveled	 per	 hour)	 in	 127	
women	ranging	from	14%	underweight	to	
99%	 overweight;	 they	 found	 a	 small	
negative	 correlation	 between	 the	 2	 data	
sets	(r	=	-0.22).		
Tudor-Locke	et	al.	(2001)	assessed	the	

relationship	 between	 pedometer	 step	
counts	and	body	fat	content	in	a	sample	of	
109	older	men	and	women	(average	age	
45	 years).	 The	 percentage	 body	 fat	
increased	from	26.2%	in	the	most	active		

Table	10:		Relationships	seen	between	objectively	measured	habitual	physical	activity	and	body	
mass	index	or	body	fat	content.	
	
Author	 Sample	 Fat	

measureme
nt	

Correlation	 Comments	

Jacobs	et	al.	
(1993)	

78	men	and	
women	aged	20-
59	yr	

Body	fat	by	
hydrostatic	
weighing	

-0.52	(age	and	sex	
adjusted	kCal/d),	-
0.08	MET-min/d)	

	

Lohman	et	
al.	(2008)	

1440	eighth-
grade	
schoolgirls	

Skin-fold	
measures	of	
body	fat	

-0.12	(mv)	
-0.15	(v)	

Correlations	for	moderately	
vigorous	(mv)	and	vigorous	(v)	
physical	activity	

McClung	et	
al.	(2000)		

209	orthopedic	
patients	

BMI	 -	0.05	 BMI	for	high	phys.	activity	(8237	
steps/d)	25.4	kg/m2,	low	phys.	
activity	(<3704	steps/d)	27.4	
kg/m2	

Rowlands	et	
al.	(1999)	

34	children	aged	
8-10	yr	

Skin-fold	
measures	of	
body	fat	

-0.42	 	

Scheers	et	al.	
(2013)	

370	middle-
aged	adults	

Waist	
circumferenc
e	

-0.61	partial	
correlation	
coefficient	

Sense-wear	arm	band	data;	odds	
ratio	for	abdominal	obesity	in	
active	individuals	0.15	

Tryon	et	al.	
(1992)	

127	women	
aged	19	to	55	
years	

BMI	 -0.22	 	

Tudor-Locke	
et	al.	(2001)	

109	men	&	
women,	average	
age	45	yr	

Impedance	
estimate	of	
body	fat	

-0.27;	similar	
correlation	with	
BMI	(-0.21)	

Body	fat	26.2%	most	active	tertile,	
32.3%	least	active	tertile	(but	not	
co-varied	for	sex)	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	87	

tertile	 (>9357	 steps/day)	 to	 32.3%	 in	
the	least	active	tertile	(<	5267	step/day),	
although	this	comparison	was	somewhat		
confounded	 because	 there	 was	 a	

greater	percentage	of	women	in	the	least	
active	tertile.			 	
Associations	between	habitual	physical	

activity	 and	 fatness	 have	 also	 been	
demonstrated	in	 children.	 Lohman	 et	 al.	
(2008)	related	skin-fold	estimates	of	body	
fatness	 to	 accelerometer	 measures	 of	
habitual	physical	activity	 in	1440	eighth-
grade	 U.S.	 schoolgirls,	 finding	 slightly	
stronger	 negative	 correlations	 for	
vigorous	 (-0.15)	 than	 for	 moderately	
vigorous	(-0.12)	physical	activity.	
Rowlands	 et	 al.	 (1999)	 also	 found	

negative	 correlations	 between	 skin-fold	
assessments	 of	 body	 fat	 and	
accelerometer	 data	 in	 8-12-year	 old	
children;	however,	correlations	as	high	as		
r	 =	 -0.42	 must	 unfortunately	 be	
discounted,	 because	 data	 were	 not	
separated	 by	 sex,	 despite	 substantial	
differences	 in	 physical	 activity	 levels	
between	the	boys	and	the	girls.		
	
Unacculturated	 vs.	 sedentary/inactive	
circumpolar	residents	
Various	 cross-sectional	 comparisons	

can	 be	 drawn	 between	 circumpolar	
residents	 who	 have	 conserved	 a	 very	
active	hunter-gatherer	lifestyle,	and	those	
who	 have	 adopted	 a	 more	 sedentary,	
inactive,	and	westernized	pattern	of	living.	

Although	higher	levels	of	body	fat	content	
can	be	linked	to	the	adoption	of	sedentary	
living,	 it	 is	 important	 to	 note	 that	
westernization	 has	 also	 brought	 changes	
in	 diet	 and	 other	 aspects	 of	 personal	
lifestyle	 to	 the	 more	 acculturated	
segments	of	the	circumpolar	populations.	
	
Comparisons	 of	 body	 fat	 within	 the	
Igloolik	population	
During	 the	 1970s,	 Inuit	 living	 in	 the	
Canadian	 arctic	 moved	 to	 small	 coastal	
settlements	 constructed	 by	 the	 Canadian	
government	 around	 schools	 and	 nursing	
stations.	One	such	recently	built	but	 still	
isolated	hamlet	was	Igloolik,	near	the	tip	
of	the	Melville	Peninsula.	 	At	this	time,	 it	
was	 possible	 to	 distinguish	 3	 groups	 of	
men	 within	 the	 community	 -	 those	 who	
continued	 living	 off	 the	 land,	 pursuing	
their	 traditional	 hunter-gatherer	
economy,	 those	 who	 had	 become	
acculturated	 to	 a	 sedentary	 and	 inactive	
western	 lifestyle,	 and	 an	 intermediate	
group,	in	the	course	of	making	a	transition	
between	the	two	ways	of	 life	(Table	11).	
Differences	 in	 habitual	 physical	 activity	
between	the	three	groups	can	be	inferred	
from	 the	 gradation	 of	 directly	 measured	
maximal	 oxygen	 intakes	 between	 the	
three	 groups.	 Although	 the	 change	 in	
lifestyle	 was	 of	 fairly	 recent	 date,	 there	
were	already	differences	 in	the	thickness	
of	skin-folds,	both	during	the	summer	and	
the	winter	months.	During	the	darkest		

Table	11:	Average	aerobic	fitness	and	skin-fold	thickness	found	in	Igloolik	men	in	1970.	Comparison	
of	traditional	hunters,	their	acculturated	peers,	and	an	intermediate	group	in	the	course	of	making	
transition	to	a	sedentary	lifestyle.	Based	on	the	data	of	Rode	and	Shephard	(1973).	
	
Current	Lifestyle	 Maximal	oxygen	intake	

[mL/(kg.min)]	
Average	skin-fold	thickness	(mm)	

Summer	 Winter	 Summer	 Winter	
Traditional	hunters	(n	=	
20)	

56.6	 56.2	 5.8	 6.4	

Transitional	(n	=	22)	 54.9	 54.9	 6.1	 6.7	
Acculturated	(n	=	18)	 51.1	 50.1	 6.7	 7.9	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	88	

winter	 months,	 extreme	 cold	 restricted	
the	 physical	 activity	 of	 all	 three	 groups,	
with	 associated	 decreases	 of	 maximal	
aerobic	power	and	significant	increases	in	
body	fat	content.	
	

Cross-sectional	 circumpolar	
comparisons	
Igloolik	 first	 became	 a	 settled	

community	 around	 1970,	 and	 it	 is	 thus	
possible	 to	 make	 cross-sectional	
comparisons	 between	 the	 Inuit	 living	 in	
this	hamlet	and	other	circumpolar	groups	
where	 there	 has	 been	 a	 longer	 and	
stronger	 contact	 with	 western	 ways	 of	
living	and	an	urban	 lifestyle.	The	Human	
Adaptability	 Project	 of	 the	 International	
Biological	Programme	undertook	such	an	
analysis,	 comparing	 skin-fold	 thicknesses	
between	 the	 Igloolik	 Inuit	 and	 more	
acculturated	 circumpolar	 populations,	
including	the	Inuit	of	Wainwright,	Alaska,	
and	nomadic	Lapps	(Shephard,	1980).	The	
men	 of	 Igloolik	 were	 found	 to	 be	
substantially	thinner	than	the	Wainwright	
Inuit	 or	 the	 nomadic	 Lapps,	 but	
differences	were	less	apparent	among	the	
women,	 probably	 because	 they	 did	 not	
usually	participate	in	the	long	and	arduous	

expeditions	 of	 their	 traditional	 hunter-
gatherer	partners.	
More	recently,	Young	has	demonstrated	

similar	cross-sectional	gradients	of	fatness	
between	 different	 groups	 of	 North	
American	Indians	and	Meti,	related	to	the	
extent	 of	 their	 contacts	 with	 urban	
communities	(Young,	1994).	
	
Discussion	and	conclusions	
Almost	all	of	the	various	cross-sectional	

comparisons	 considered	 in	 this	 article	
have	 pointed	 towards	 a	 strong	 and	
consistent	 association	 between	 regular	
physical	 activity	 and	 avoidance	 of	 the	
accumulation	 of	 an	 excessive	 amount	 of	
body	fat.		However,	one	must	be	cautious	
in	assuming	that	the	relationship	is	causal.	
Physically	active	individuals	could	have	a	
specific	 genotype	 that	 limits	 their	 fat	
accumulation,	 and	 the	 active	 individuals	
may	 also	 have	 been	 more	 careful	 with	
their	 diet	 and	 other	 aspects	 of	 their	
personal	 lifestyle.	 Moreover,	 in	 some	 of	
the	 cross-sectional	 comparisons,	 an	
excessive	 body	 fat	 content	 could	 have	
been	 the	 cause	 rather	 than	 the	
consequence	 of	 a	 low	 level	 of	 habitual	
physical	 activity.	 However,	 the	 rapid	

Table	12:	Comparison	of	average	skin-fold	thicknesses	between	the	Igloolik	Inuit	of	1970,	the	more	
acculturated	Inuit	populations	of	Wainwright,	Alaska,	and	nomadic	Lapps	(Shephard,	1980).		
	
Group	 Age	(yr)	 Triceps	(mm)	 Subscapular	

(mm)	
Suprailiac	
(mm)	

Average	(mm)	

Igloolik	men	 20-29	 4.4	 6.6	 7.2	 6.0	
	 30-39	 4.7	 7.3	 7.5	 6.5	
	 40-49	 4.1	 5.7	 5.9	 5.2	
	 50-59	 4.4	 6.8	 6.0	 5.7	
Lapps	(men)	 15-55	 7.0	 8.2	 6.1	 7.1	
Wainwright	(men)	 25-65	 9.6	 12.2	 	 10.9	
Igloolik	women		 20-29		 10.2	 9.0	 8.7	 9.3	
	 30-39	 14.0	 12.3	 10.6	 12.3	
	 40-49	 11.7	 10.0	 9.0	 10.2	
	 50-59	 28.0	 23.8	 22.6	 24.8	
Lapps	(women)	 15-21	 25.3	 15.5	 11.4	 17.4	
Wainwright	(women)	 25-65		 12.2	 14.9	 	 13.6	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	89	

accumulation	 of	 fat	 in	 previously	 thin	
populations	with	transition	to	a	sedentary	
lifestyle	 argues	 against	 a	 genetic	
explanation	 of	 the	 associations	 noted	 in	
this	 article,	 and	 in	 some	 of	 the	
comparisons	 (such	 as	 those	 between	
university	 faculties),	 there	 is	 also	 no	
strong	 reason	 to	 suppose	 dietary	
differences	 between	 thin	 and	 fatter	
individuals.	
The	 association	 between	 high	 energy	

expenditures	and	low	levels	of	body	fat	is	
particularly	obvious	in	endurance	athletes,	
but	 nevertheless	 there	 are	 examples	 of	
other	 less	 highly	 selected	 and	 only	
moderately	 fit	 individuals	 who	 have	
undertaken	 100	 or	 more	 days	 of	
marathon-style	exercise,	thereby	reducing	
their	body	fat	content	by	6-8	kg	or	more.			
In	 the	 case	 of	 physical	 education	

students,	 the	 impact	 of	 their	 educational	
choice	 upon	 body	 fat	 content	 is	 more	
evident	in	male	than	in	female	students.	In	
some	 schools,	 complications	 of	
interpretation	 have	 arisen	 from	 terms	
assigned	to	academic	rather	than	physical	
education	 instruction,	 and	 not	 all	 PE	
students	have	had	a	strong	commitment	to	
their	 physical	 development.	 Inferences	
would	be	clarified	by	further	studies	using	
objective	methods	to	measure	the	amount	
of	 weekly	 physical	 activity	 actually	
undertaken	 by	 individual	 students,	 and	
also	 by	 exploring	 the	 extent	 of	
involvement	of	supposed	control	students	
in	 physically	 demanding	 inter-collegiate	
sports	programmes.	
Early	 occupational	 studies	 often	

focused	on	cardiac	risk	factors	other	than	
fat	 accumulation,	 but	 comparisons	 of	
waist	circumferences	between	conductors	
and	 bus	 drivers	 showed	 a	 significant	
advantage	 to	 the	 conductors,	 despite	
similarities	 in	 socio-economic	 status	 and	
lifestyle	 between	 the	 two	 classes	 of	

employee.	 Today,	 there	 is	 only	 a	 limited	
possibility	of	making	further	occupational	
comparisons;	 bus	 drivers	 rather	 than	
conductors	 now	 collect	 fares	 from	
passengers,	 and	many	 jobs	 that	 formerly	
required	 hard	 physical	 work	 have	 been	
eliminated	 by	 automation	 and	 robotics.	
However,	heavy	physical	demands	are	still	
made	 on	 elite	 military	 groups,	 and	
available	 reports	 document	 that	 such	
troops	consistently	have	very	low	levels	of	
body	fat.	
The	 high	 levels	 of	 physical	 activity	

found	in	traditional	Amish	and	Old-order	
Mennonite	 families	 are	 reflected	 in	
correspondingly	 low	 levels	 of	 body	 fat,	
and	 the	 changes	 of	 body	 composition	 in	
Amish	 who	 have	 abandoned	 their	
traditional	 rural	 life	 to	 work	 at	 more	
sedentary	 jobs	 within	 Ohio	 towns	
provides	 evidence	 that	 their	 traditional	
thinness	is	a	consequence	of	a	heavy	daily	
energy	expenditure	rather	than	a	genetic	
trait	or	some	other	facet	of	their	personal	
lifestyle.	 Nevertheless,	 communal	 meals	
appear	 to	 play	 an	 important	 role	 in	
traditional	Amish	families,	and	it	would	be	
useful	to	obtain	confirmatory	data	on	the	
nature	 and	 extent	 of	 their	 daily	 food	
intake.			
Cross-sectional	 studies	 of	 the	 general	

population	 seeking	 to	 validate	 both	
physical	 activity	 questionnaires	 and	
accelerometers	 have	 shown	 correlation	
coefficients	as	high	as	-0.50	with	estimates	
of	 body	 fat	 content.	 Correlations	 have	
been	 largest	 for	 the	 best	 of	 the	
questionnaires,	and	have	also	been	greater	
for	reports	of	vigorous	than	for	moderate	
or	light	activity,		
Inuit	 who	 have	 maintained	 a	 hunter-

gatherer	 lifestyle	are	another	group	with	
very	 low	 levels	 of	 body	 fat.	 	 Here,	 there	
remains	 scope	 for	 further	 study	 of	 fat	
distribution	between	superficial	and	deep	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	90	

body	depots,	and	to	examine	whether	the	
tendency	 to	 internal	 fat	 storage	 is	
important	in	maintaining	thermal	balance	
when	undertaking	heavy	physical	work.	It	
also	remains	interesting	to	clarify	why	the	
oldest	 Inuit	 females	 carry	 substantial	
amounts	 of	 fat;	 is	 this	 a	 consequence	 of	
multiple	 pregnancies,	 or	 are	 there	
substantial	 sex-based	 differences	 in	
activity	patterns	within	the	household?	
		Although	 the	 various	 comparisons	

discussed	in	this	article	point	strongly	to	
the	efficacy	of	vigorous	physical	activity	in	
containing	the	accumulation	of	body	fat,	it	
may	 be	 asked	 whether	 the	 amounts	 of	
activity	 needed	 are	 realistic	 for	 the	
general	 middle-aged	 population.	 The	
average	 person	 is	 certainly	 not	 going	 to	
engage	in	a	cross-Canada	run,	or	even	to	
run	in	a	marathon	event,	and	in	many	of	
the	 comparisons	 we	 have	 cited	 the	
physically	 active	 group	 is	 undertaking	
much	 more	 than	 the	 minimum	 public	
health	recommendation	for	daily	physical	
activity.	 However,	 the	 minimum	 volume	
needed	 to	prevent	obesity	 is	probably	 in	
the	range	of	150-250	minutes	of	moderate	
to	vigorous	aerobic	activity	per	week,	and	
this	target	seems	well	within	the	capacity	
of	 the	 average	 middle-aged	 adult.	 The	
challenge	to	the	public	health	professional	
is	 to	 maintain	 enthusiasm	 for	 this	
attainable	goal.	
		
Acknowledgements	
The	 author	 acknowledges	 no	 funding	

relationships	or	other	conflicts	of	interest.		
	
Author’s	Qualifications	
The	 author's	 qualifications	 are	 as	

follows:	 Roy	 J.	 Shephard,	 C.M.,	 Ph.D.,	
M.B.B.S.,	 M.D.	 [Lond.],	 D.P.E.,	 LL.D.,	 D.Sc.,	
FACSM,	FCSEP,	FFIMS,	FAAPE.	
	
	

References	
Arazi,	 H.,	 Hoseini,	 R.,	 &	 Behrozi,	 A.	 (2012).	 A	

comparison	of	body	fat	and	blood	pressure	
between	 physical	 education	 and	 non-
physical	 education	 	 major	 male	 students.	
Phys	 Ed	 Sport	 (Iran),	 10(2),	 127-134.	
http://facta.junis.ni.ac.rs/pe/pe201202/p
e201202-06.pdf		

Arazi,	H.,	&	Hoseini,	R.	 (2011).	Body	composition	
and	 blood	 pressure	 between	 athlete	 and	
non-athlete.	 Biology	 Exerc,	 7(2)	 5-14.	
https://www.biologyofexercise.com/imag
es/issues/721.pdf		

Arrese,	 A.L.,	 &	 Ostériz	 E.S.	 (2006).	 Skinfold	
thicknesses	 associated	 with	 distance	
running	 	 performance	 in	 highly	 trained	
runners.	 J	 Sports	 Sci,	 24(1),	 69-76.	
https://doi.org/10.1080/0264041050012
7751	

Bale,	P.	 (1980).	The	relationship	of	physique	and	
body	composition	to	strength	in	a	group	of	
physical	 education	 students.	 Br	 J	 Sports	
Med,	 14(4),193-198.	
https://doi.org/10.1136/bjsm.14.4.193	

Bale,	 P.,	 Bradbury,	 D.,	 &	 Colley,	 E.	 (1986).	
Anthropometric	 and	 training	 variables	
related	to	10	km	running	performance.	Br	J	
Sports	 Med,	 20(4),	 170-173.	
https://doi.org/10.1136/bjsm.20.4.170	

Ball,	K.,	Owen,	N.,	Salmon,	 J.,	Bauman,	A.,	&	Gore,	
C.J.	 (2001).	 Associations	 of	 physical	
activity	with	body	weight	and	 fat	 in	men	
and	women.	Int	J	Obes	Relat	Metab	Disord,		
25,	 914-919.	
https://doi.org/10.1038/sj.ijo.0801622	

Bartlett,	H.L.,	Mance,	M.J.,	&	Buskirk,	E.R.	(1984	).	
Body	composition	and	expiratory	reserve	
volume	 in	 female	 gymnasts	 and	 runners.	
Med	Sci	Sports	Exerc,	16(3),	311-315		

Bassett,	 D.R.,	 Schenider,	 P.L.,	 &	 Huntington,	 G.E.	
(2004).	 Physical	 activity	 in	 an	 Old	 Order	
Amish	 community.	 Med	 Sci	 Sports	 Exerc,	
36(1),	 79-85	
.https://doi.org/10.1249/01.MSS.000010
6184.71258.32	

Bassett,	 D.R.,	 Tremblay,	 M.S.,	 Esliger,	 D.W.,	
Copeland,	 J.L.,	 Barnes,	 J.D.,	 &	 Huntington	
G.E	 (2007	 ).	 Physical	 activity	 and	 body	
mass	 index	 of	 children	 in	 an	 old	 order	
Amish	 community.	 Med	 Sci	 Sports	 Exerc,	
39(3),	 410-415.	
https://doi.org/10.1249/mss.0b013e3180
2d3aa7	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	91	

Bouchard,	 C.,	 Tremblay,	 A.,	 Leblanc,	 C.,	 Lortie,	 G.,	
Savard,	 R.],	 &	 Thériault,	 G	 .	 (1983).	 A	
method	 to	 assess	 energy	 expenditure	 in	
children	 and	 in	 adults.	 Am	 J	 Clin	 Nutr,	
17(3),	 461-466.	
https://doi.org/10.1093/ajcn/37.3.461	

Bradbury,	 K.E.,	 Guo,	 W.,	 &	 Cairns,	 B.J.	 (2017).	
Association	between	physical	activity	and	
body	 fat	 percentage,	 with	 adjustment	 for	
BMI:	a	large	cross-sectional	analysis	of	UK	
Biobank.	 	 BMJ	 Open,	 7:e011843.	
http://dx.doi.org/10.1136/bmjopen-
2016-011843		

Budd,	 G.M.	 (2001).	 How	 do	 wildland	 firefighters	
cope?	 Physiological	 and	 behavioural	
temperature	 regulation	 in	 men	
suppressing	Australian	summer	bushfires	
with	hand	tools.	J	Therm	Biol,	26(4-5),	381-
386.	 https://doi.org/10.1016/S0306-
4565(01)00048-1	

Campos,	L.C.B.,	Campos,	F.A.D.,	Bezerra,	T.A.R	 ,	&	
Pellegrinotti,	 I.L..	 (2017)	 Effects	 of	 12	
weeks	 of	 physical	 training	 on	 body	
composition	 and	 physical	 fitness	 in	
military	recruits.	Int	J	Exerc	Sci,	10(4,)	560-
567.	
https://www.ncbi.nlm.nih.gov/pubmed/2
8674600		

Coetzer,	P.,	Noakes,	T.D.,	Sanders,	B.,	Lambert,	M.I.,	
Bosch,	 A.N..,	 Wiggins,	 T.,	 &	 Dennis,	 S.C.	
(1993).	Superior	 fatigue	resitance	of	elite	
black	 distance	 runners.	 J	 Appl	 Physiol,	
75(4),1822-1827.	
https://doi.org/10.1152/jappl.1993.75.4.1
822.		

Conley,	D.L.,	&	Krahenbuhl,	G.S.	 (1980).	Economy	
and	 distance	 running	 performance	 of	
highly	 trained	 athletes.	 Med	 Sci	 Sports	
Exerc,	 12(5),	 357-360.	
https://doi.org/10.1249/00005768-
198025000-00010	

Costill,	 D.L,	 Bowers,	 R.,	 &	 Kammer,	 W.F.	 (1970).	
Skinfold	 estimates	 of	 body	 fat	 among	
marathon	 runners.	 Med	 Sci	 Sports,	 2(2),	
93-95.	
https://doi.org/10.1249/00005768-
197000220-00005	

Cureton,	 K.J.,	 Sparling,	 P.B.,	 Evans,	 B.W.,	 Johnson,	
S.M.,	 Kong,	 U.D.,	 &	 Purvis,	 J.W.	 (1978).	
Effect	of	experimental	alterations	in	excess	
weight	 on	 aerobic	 capacity	 and	 distance	
running	 performance.	 Med	 Sci	 Sports,	
10(3),	 194-199.		

https://www.ncbi.nlm.nih.gov/pubmed/7
23510		

de	 Garay,	 A.L.,	 Levine,	 L.,	 &	 Carter,	 J.E.L.	 (1974).	
Genetic	 and	 anthropometric	 studies	 of	
Olympic	 champions.	 New	 York,	 NY,	
Academic	Press,	pp.1-296.			

Ekelund,	 U.,	 Neovius,	 M.,	 Linné,	 Y.,	 Brage,	 S.,	
Wareham,	 M.J.,	 &	 Rӧssner,	 S.	 (2005).	
Associations	between	physical	activity	and	
fat	 mass	 in	 adolescents:	 the	 Stockholm	
weight	development	study.	Am	J	Clin	Nutr,	
81(2),	 355-360.	
	https://doi.org/10.1093/ajcn.81.2.355.	

Eksterowicz,	J.,	Napierala,	M.,	&	Zukow,	W.	(2016).	
How	the	Kenyan	runner's	body	structure	
affects	 sports	 results.	 Hum	 Movement,	
17(1),	 8-14.	
https://doi.org/10.1515/humo-2016-
0002		

Elliot,	 D.L.,	 Goldberg,	 L.,	 Kuehl,	 K.S.,	 Moe,	 F.L.,	
Berger,	 R.K.R.,	 &	 Pickering,	 M.A.	 (2007).	
The	PHLAME	(Promoting	healthy	lifestyles	
alternative	 models'	 effects)	 firefighter	
study:	 Outcomes	 of	 two	 models	 of	
behavior	 change.	 J	 Occup	 Environ	 Med,	
49(2),	 204-213.	
https://doi.org/10.1097/JOM.0b013e318
0329a8d	

Epstein,	L.H.,	&	Wing,	R.R.	(1980).	Aerobic	exercise	
and	 weight.	 Addict	 Behav,	 5(4),	 371-388.	
https://doi.org/10.1016/0306-
4603(80)90011-8	

Esliger,	D.W.,	Tremblay,	M.S.,	Copeland,	J.L.,	Barnes,	
J.D.,	 Huntington,	 G.E.,	 &	 Bassett,	 D.R.	
(2010).	 Physical	 activity	 profile	 of	 old	
order	 Amish,	 Mennonite	 and	
	contemporary	 children.	 Med	 Sci	 Sports	
Exerc,	 42(2),	 296-303.	
https://doi.org/10.1249/MSS.0b013e318
1b3afd2	

Farina,	E.K.,	Taylor,	J.C.,	Means,	G.E.,	Williams,	K.W.,	
Murphy,	N.E.,	Margolis,	L.M.,	 .	 .	 .	McClung,	
J.P.	 (2017).	 Effect	 of	 combat	 deployment	
on	 anthropmetrics	 and	 physiological	
status	 of	 U.S.	 army	 special	 operations	
forces	soldiers.	Mil	Med,	182,	e1659-e1668.	
https://doi.org/10.7205/MILMED-D-16-
00022	

Farrell,	P.A.,	Wilmore,	J.H.,	Coyle,	E.F.,	Billing,	J.E.,	&	
Costill,	 D.L.	 (1979).	 Plasma	 lactate	
accumulation	 and	 distance	 running	
performance.	 Med	 Sci	 Sports,	 11(4),	 338-
344.	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	92	

https://www.ncbi.nlm.nih.gov/pubmed/5
30025		

Fleck,	 S.	 J.	 (1983).	 Body	 composition	 of	 elite	
American	athletes.	Am	J	Sports	Med,	11(6,)	
398-403.	
https://doi.org/10.1177/0363546583011
00604	

Folsom,	A.R.,	Caspersen,	C.J.,	&	Tasylor,	H.L.	(1985).	
Leisure	 time	 physical	 activity	 and	 its	
relationship	 to	 coronary	 risk	 factors	 in	 a	
population-based	 sample.	 Am		
	Epidemiol,	 121(4),	 570-579.	
https://doi.org/10.1093/oxfordjournals.aj
e.a114035	

Fuchs,	 J.A.	 Levinson,	 R.M.,	 Stoddard,	 R.R.,	 Mullet,	
M.E.,	 &	 Jones,	 D.H.(1990).	 Health	 risk	
factors	 among	 the	 Amish:	 results	 of	 a	
survey.	 Health	 Educ	 Q,	 17(2),	 197-211.	
https://doi.org/10.1177/1090198190017
00206	

Glick,	 M.,	 Michel,	 A.C.,	 Dorn,	 J.,	 Horowitz,	 M.,	
Rosenthal,	 T.,	 &	 Trevisan,	 M.	 (1998).	
Dietary	 cardiovascular	 risk	 factors	 and	
serum	 cholestrol	 in	 an	 old	 order	
Mennonite	 community.	 Am	 J	 Publ	 Health,	
88(8),	 1202-1205.	
https://doi.org/10.2105/AJPH.88.8.1202	

Godin,	G.,	&	Shephard,	R.J.	(1985).	A	simple	method	
to	 assess	 exercise	 behavior	 in	 the	
community.	Can	J	Appl	Spt	Sci,	10(3),	141-
146.	
https://www.ncbi.nlm.nih.gov/pubmed/4
053261		

Graves,	 J.E.,	Pollock,	M.L.,	&	Sparling,	P.B.	 (1987).	
Body	composition	of	elite	female	distance	
runners.	Br	 J	 Sports	Med,	8(Suppl.	2),	 96-
102.	
https://www.ncbi.nlm.nih.gov/pubmed/3
692659		

Hirata,	 K.	 (1966).	 Physique	 and	 age	 of	 Tokyo	
Olympic	 champions.	 J	 Sports	 Med	 Phys	
Fitness,	 6(4),	 207-222.	
https://www.ncbi.nlm.nih.gov/pubmed/5
980984		

Jacobs,	D.R.,	Ainsworth,	R.E.,	Hartman,	T.J.,	&	Leon,	
A.S.	 (1993).	A	simultaneous	evaluation	of	
10	 commonly	 used	 physical	 activity	
questionnaires.	Med	Sci	Sports	Exerc,	2(1)5,	
81-91.	
https://doi.org/10.1249/00005768-
199301000-00012	

Jakicic,	 J.M.,	 Clark,	 K.,	 Coleman,	 E.,	 Donnelly,	 J.E.,	
Foreyt,	 J.,	 Melanson,	 E.,	 	 .	 .	 .	 Volpe,	 S.	
(2001).	 Appropriate	 intervention	

strategies	 for	weight	 loss	and	prevention	
of	weight	regain	for	adults.	Med	Sci	Sports	
Exerc,	 33(12),	 2145-2156.		
https://doi.org/10.1097/00005768-
200112000-00026	

Jobin,	J.,	Tremblay,	A.,	&	Samson,	P.	(1984).	Effect	
of	a	6400-km	run	on	a	60-year	old	man.	
Phys	 Sportsmed	 12(9),	 53-61.	
https://doi.org/10.1080/00913847.1984.
11701943	

Knechtle,	B.,	&	Bircher,	S.	(2005).	Veränderung	der	
Körperzusammensetzung	 bei	 einem	
Ultralauf.	 [Changes	 in	 body	 composition	
during	 an	 extreme	 endurance	 run].	
	Praxis,	 94(10),	 371-377.	
https://doi.org/10.1024/0369-
8394.94.10.371	

Knechtle,	 B.,	 Duff,	 B.,	 Schulze,	 I.,	 &	 	 Kohler,	 G.	
(2008).	 The	 effects	 of	 running	 1,200	 km	
within	17	days	on	body	composition	 in	a	
female	 ultrarunner-Deutschlandlauf	
	2007.	 Res	 Sports	 Med,	 16(3),	 167-188.	
	https://doi.org/10.1080/1543862080210
4120	

Knechtle,	B.,	&	Knechtle,	P.	(2007).	In	einem	Lauf	
vom	Bodensee	an	den	Genfersee	–	Effekt	
auf	 Fett-	 und	 Muskelmasse	 (Run	 across	
Switzerland-	 effect	 on	 body	 fat	 and	
	muscle	 mass).	 Praxis,	 96(8),	 281-286l	
https://doi.org/10.1024/1661-
8157.96.8.281	

Knechtle,	 B.,	 Knechtle,	 P.,	 Barandun,	 U.,	 &	
Rosemann,	T.	(2015).	Anthropometric	and	
training	variables	related	to	half-marathon	
performance	 in	 recreational	 female	
runners.	 Phys	 Sportsmed,	 39(2),	 158-166.	
https://doi.org/10.3810/psm.2011.05.19
07	

Kong,	P.W.,	&	De	Heer,	H.	(2008).	Anthropometric,	
gait	and	strength	characteristics	of	Kenyan	
distance	 athletes.	 J	 Sports	 Sci	 Med,	 7(4),	
499-504.	
https://www.ncbi.nlm.nih.gov/pubmed/2
4149957		

Kvintová,	 J.,	 &	 Sigmund,	 M.	 (2016).	 Physical	
activity,	 body	 composition	 and	 health	
assessment	 in	 current	 female	 university	
students	with	active	and	inactive	lifestyles.	
J	Phys	Ed	Sport,	16(Suppl.	1),	627-632.		

Levinson,	 R.M.,	 Fuchs,	 J.A.,	 Stoddard,	 R.R.,	 Jones,	
D.H.,	&	Mullet,	M.	 (1989).	Behavioral	risk	
factors	in	an	Amish	community.	Am	J	Prev	
Med,	 5(3),	 150-156.	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	93	

https://doi.org/10.1016/S0749-
3797(18)31096-1	

Lewis,	 S.,	 Haskell,	 W.L.,	 Klein,	 H.,	 Halpern,	 J.,	 &	
Wood,	 P.D.	 (1975).	 Prediction	 of	 body	
composition	 in	 habitually	 active	 middle-
aged	 men.	 J	 Appl	 Physiol,	 39(2),	 221-225.	
https://doi.org/10.1152/jappl.1975.39.2.2
21	

Lohman,	T.G.	(1992).	Advances	in	body	composition	
assessment	Champaign,	IL,	Human	Kinetics,	
pp.	1-150.		

Lohman,	 T.,	 Ring	 ,	 K.,	 Pfeiffer,	 K.,	 Camhi,	 S.,	
Arredondo,	 E.,	 Pratt,	 C.,	 .	 .	 .	 Webber,	 L.S.	
(2008).	Relationships	among	fitness,	body	
composition	and	physical	activity.	Med	Sci	
Sports	 Exerc,	 40(6),	 1163-1179.	
https://doi.org/10.1249/MSS.0b013e318
165c86b	

Lutoslawska,	 G.,	 Malara,	 M.,	 Tomaszwski,	 P.,	
Mazurek,	 K.,	 Czajkowska,	 A.,	 Kȩska,	 A.,	 &	
Tkaczyk,	 J.	 	 (2014).	Relationship	between	
the	percentage	of	body	 fat	and	surrogate	
indices	 of	 body	 fat	 in	 male	 and	 female	
Polish	 active	 and	 sedentary	 students.	 J	
Phys	 Anthropol,	 33:10.	
https://doi.org/10.1186/1880-6805-33-
10	

Malavolti,	M.,	Battistini,	N.C.,	Dugoni,	M.,	Bagni,	B.,	
Bagni,	I.,	&	Pietrobelli,	A.	(2008).	Effect	of	
intense	 military	 training	 on	 body	
composition.	 J	 Strength	 Cond	 Res,	 22(2),	
503-508.	
https://doi.org/10.1519/JSC.0b013e3181
63441f	

Malhotra,	A.,	Noakes,	T.,	&	Phinney,	S.	(2015).	It	is	
time	to	bust	the	myth	of	physical	inactivity	
and	obesity:	you	cannot	outrun	a	bad	diet.	
Br	 J	 Sports	 Med,	 49(15),	 967-968.	
https://doi.org/10.1136/bjsports-2015-
094911	

Martinez,	J.A.,	Kearney,	J.M.,	Kafatos,	A.,	Paquet,	S.,	
&	 Martinez-Gonzalez,	 M.A.	 (1999).	
Variables	 independently	 associated	 with	
self-reported	 obesity	 in	 the	 European	
Union.	Publ	Health	Nutr,	2	 (1a),	125-133.	
https://www.ncbi.nlm.nih.gov/pubmed/1
0933632		

McClung,	C.D.,	Zahiri,	C.A.,	Higa,	J.K.,	Amstutz,	H.C.,	
&	 Schalzried,	 T.S.	 (2000).	 Relationship	
between	body	mass	 index	and	activity	 in	
hip	or	knee	arhtroplasty	patients.	J	Orthop	
Res,	 18(1),	 35-39.	
https://doi.org/10.1002/jor.1100180106	

Meckel,	 Y.,	 Galily,	 Y.,	 Nemet,	 D.,	 &	 Eliakim,	 A.	
(2011).	 Changes	 in	 weight	 indexes	 and	
aerobic	 fitness	 of	 physical	 education	
students	over	three	years	of	college.	J	Hum	
Sports	 Exerc,	 8(1),112-121.	
https://doi.org/10.4100/jhse.2011.61.13	

Mertens,	D.J.,	Rhind,	S.,	Berkhoff,	 S.,	Dugmore,	D.,	
Shek,	 P.N.,	 &	 Shephard,	 R.J.	 (1996).	
Nutirtional,	 immunologic	 and	
psychological	responses	to	a	7250	m	run.	J	
Sports	 Med	 Phys	 Fitness,	 36(2),	 132-138.	
http://europepmc.org/abstract/MED/889
8521		

Morinaka,	T.,	Limtrakul,	P.,	Makonkawkeyoon,	L,	&	
Sone,	Y.	(2012).	Comparison	of	variations	
between	 percentage	 of	 body	 fat,	 body	
mass,	 and	 daily	 physical	 activity	 among	
young	Japanese	and	Thai	female	students.	J	
Physiol	 SAnthropol	 31:21.	
https://doi.org/10.1186/1880-6805-31-
21	

Morris,	 J.N.,	 &	 Raffle,	 P.A.B.	 (1956).	 Physique	 of	
London	 busmen.	 The	 epidemiology	 of	
uniforms.	 	 Lancet,	 271(6942),	 569-570.	
https://doi.org/10.1016/S0140-
	6736(56)92049-9	

Neculães,	M.,	&	Lucaci,	P.	(2016).	Statistical	study	
on	 body	 weight	 and	 body	 fat	 among	
students	from	the	physical	education	and	
sport	 faculty	 in	 Iaşi.	 Timişoara	 Phys	 Ed	
Rehab	 J,	 2016-0005.	
https://doi.org/10.1515/tperj-2016-0005		

Nindl,	 B.C.,	 Barnes,	 B.R.,	 Alemany,	 J.A.,	 Frykman,	
P.N.,	 Shippee,	 R.L.,	 &	 Friedl,	 K.L.	 (2007).	
Physiological	 consequences	 of	 U.S.	 army	
ranger	 training.	 Med	 Sci	 Sports	 Exerc,	
39(8),1380-1387.	
https://doi.org/10.1249/MSS.0b013e318
067e2f7	

Oliver,	R.M.	(1967).	Physique	and	serum	lipids	of	
young	 London	 busmen	 in	 relation	 to	
ischaemic	heart	disease.	Br	J	 Industr	Med,	
24,	 181-188.	
http://dx.doi.org/10.1136/oem.24.3.181	

Paff,	 J.,	 &	 Korneta,	 K.	 (2000).	 Changes	 in	 aerobic	
and	 anaerobic	 fitness	 in	 the	 Polish	 army	
paratroopers	during	their	military	service.	
Aviat	Space	Environ	Med,	71(9),	 920-924.	
http://europepmc.org/abstract/med/110
01345		

Pollock,	M.L.,	Gettman,	L.R.,	 Jackson,	A.,	Ayres,	 J.	 ,	
Ward,	 A.,	 &	 Linnerud,	 A.C.	 (1977).	 Body	
composition	of	elite	class	distance	runners.	
Ann	 NY	 Acad	 Sci,	 301(1),	 361-370.	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	94	

https://doi.org/10.1111/j.1749-
6632.1977.tb38213.x	

Poston,	 W.S.C.,	 Jitnarin,	 N.,	 Haddock,	 K.C.,	 Jahnke,	
S.A.,	 &	 Tuley,	 B.C.	 (2011).	 Obesity	 and	
injury-related	 absenteeism	 in	 a	
population-based	 firefighter	 cohort.	
Obesity,	 19(10),	 2076-2081.	
https://doi.org/10.1038/oby.2011.147	

Pribis,	P.,	Burtnack,	C.A.,	McKenzie,	S	.O.,	&	Thayer,	
J.	 (2010).	 Trends	 in	 body	 fat,	 body	 mass	
index	and	physical	fitness	among	male	and	
female	 college	 students.	 Nutrients,	 2(10),	
1075-1085.	
https://doi.org/10.3390/nu2101075	

Reuter,	C.,	Stein,	C.E.,	&	Vargas,	D.M.	(2012).	Bone	
mass	 and	 body	 composition	 in	 college	
students.	Rev	Assoc	Med	Bras,	58(3),	328-
334.	
https://www.ncbi.nlm.nih.gov/pubmed/2
2735225	

Richardson,	M.T.,	Ainsworth,	B.E.,	Wu,	H.C.,	Jacobs,	
D.R.,	 &	 Leon,	 A.S.	 (1995).	 Ability	 of	 the	
atherosclerosis	 risk	 in	 communities	
(ARIC)/Baecke	 questionnaire	 to	 assess	
leisure-time	 physical	 activity.	 Int	 J	
Epidemiol,	 24(4),	 685-693.	
https://doi.org/10.1093/ije/24.4.685	

Rode,	 A.,	 &	 Shephard,	 R.J.	 (1973).	 Fitness	 of	 the	
Canadian	Eskimo:	The	influence	of	season.	
Med	 Sci	 Sports	 Exerc,	 5(3),	 170-173.	
https://doi.org/10.1249/00005768-
197323000-00014	

Rowlands,	 A.V.,	 Eston,	 R.G.,	 &	 Ingledew,	 D.K.	
(1999).	 Relationship	 between	 activity	
levels,	aerobic	fitness	and	body	fat	in	8-	to	
10-yr	 old	 children.	 J	 Appl	 Physiol,	 86(4),	
1428-1435	
https://doi.org/10.1152/jappl.1999.86.4.1
428	

Ruchan,	 I.	 	 (2015).	 An	 evaluation	 of	 the	 physical	
activity	 levels	 and	 body	 compositions	 of	
university	students.	Anthropologist,	20(3),	
430-436.		 	
https://doi.org/10.1080/09720073.2015.
11891746	

Rusko,	H.,	Havu,	M.,	&	Karvinen,	E.	(1978).	Aerobic	
performance	capacity	in	athletes.		Eur	 J	
Appl	 Physiol,	 38(2),	 151-159.	
https://doi.org/10.1007/BF00421531	

Rutter,	 S.	 (1994).	 Comparison	 of	 energy	
expenditures	 in	 normal	 weight	 and	
overweight	 women	 using	 the	 Calrac	
personal	 activity	 computer.	 Int	 J	 Eating	
Disord,	 15(1),	 37-42.	

https://doi.org/10.1002/1098-
108X(199401)15:1<37::AID-
EAT2260150105>3.0.CO;2-8	

Saczuk,	J.,	Wasiluk,	A.,	Wilczewski,	R.	&	Wilczewski,	
A.	 (2016).	 Differences	 in	 body	 build	 and	
physical	 fitness	 of	 PE	 students	 from	 the	
Faculty	of	Physical	Education	and	Sport	in	
Biala	Podlaska	in	the	years	1989,	2004	and	
2014.	 Pol	 J	 Sport	 Tourism,	 23,	 212-215.	
https://doi.org/10.1515/pjst-2016-0028	

Scheers,	 T.,	 Philippaerts,	 R.,	 &	 LeFevre,	 J.	 (2013).	
Sense-wear	 determined	 physical	 activity	
and	 sedentary	 behavior	 and	 metabolic	
syndrome.	Med	Sci	Sports	Exerc	45(3),	481-
489.	
https://www.ncbi.nlm.nih.gov/pubmed/2
3034646		

Schutz,	 U.W.,	 Schmidt-Trucksüss,	 A.,	 Knechtle,	 B.,	
Machann,	J.,	Weidelbach,	H.,	Erhardt,	M.,	.	.	.	
Bilich,	 C.	 (2012).	 The	 Trans-Europe	
footrace	 project:	 longitudinal	 data	
acquisition	 in	 a	 randomized	 mobile	 MRI	
observational	 cohort	 study	 on	 44	
endurance	runners	at	a	64	stage	4,486	km	
transcontinental	ultramarathon.	BMC	Med,	
10:78.	 https://doi.org/10.1186/1741-
7015-10-78		

Shaw,	K.A.,	Gennat,	H.C.,	O'Rourke,	P.,	&	Del	Mar,	C.,	
(2006).	Exercise	for	overweight	or	obesity	
(Review).	 Cochrane	 Database	 Syst	 Rev,	
2006	 (4)	 CD003817	
https://doi.org/10.1002/14651858.CD00
3817.pub3	

Shephard,	 R.J.,	 &	 Rode,	 A.	 (1996).	 The	 health	
consequences	 of	 'modernization.'	 Evidence	
from	circumpolar	peoples	Cambridge,	UK.:	
Cambridge	 University	 Press,	 pp.	 1-306.	
https://doi.org/10.1017/CBO9780511983
726	 	

Shephard,	 R.J.	 (1979).	 Endurance	 Fitness,	 2nd	 ed.	
Toronto,	ON,		University	of	Toronto	Press,		
pp.	1-380.	

Shephard,	 R.J.	 (1980).	 Work	 physiology	 and	
activity	 patterns.	 In	 F.A.	 Milan	 (Ed.),	 The	
humnan	 biology	 of	 circumpolar	
populations.	 Cambridge,	 UK,	 Cambridge	
University	Press,	pp.	305-338.		

Shephard,	 R.J.	 (1996).	 Worksite	 fitness	 and	
exercise	 programs:	 A	 review	 of	
methodology	 and	 health	 impact.	 Am	 J	
Health	 Promot,	 10(6),	 436-452.	
https://doi.org/10.4278/0890-1171-
10.6.436	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	95	

Shephard,	 R.J.	 (2018).	 Does	 it	 matter	 if	 I	 am	
overweight?	 2.	 Psycho-social	
consequences.	 Health	 Fitness	 J	 Canada,	
11(3),	 22-66.		
https://doi.org/10.14288/hfjc.v11i2.263		

Shephard,	R.J.	(2019).	Can	regular	physical	activity	
prevent	 obesity?	 1.	 Potential	 factors	
reducing	 the	 efficacy	 of	 physical	 activity.	
Health	Fitness	J	Canada,	In	press.	

Shephard,	R.J.,	&	Åstrand,	P.-O.	(2000).	Endurance	
in	 Sport,	 2nd	 ed.	 Oxford,	 UK,	 Blackwell	
Scientific,	 pp.	 1-987	
https://doi.org/10.1002/9780470694930	

Shephard,	 R.J.,	 Kavanagh,	 T.,	 Conway,	 S.,	 Thomas,	
M.,	 &	 Anderson,	 G.H.	 (1975).	 Nutritional	
demands	of	sub-maximum	work:	Marathon	
and	 Trans-Canadian	 events.	 Proceedings,	
III.	International	Symposium	on	Sportsmen	
Nutrition,	Warsaw.	Warsaw,	Poland,	Polish	
Sports	Federation.		

Shephard,	R.J.,	&	Pimm,	P.	(1975).	Physical	fitness	
of	 Canadian	 physical	 education	 students,	
with	 a	 note	 on	 international	 differences.	
Brit	 J	 Sports	 Med,	 9,	 165-174.	
https://doi.org/10.1136/bjsm.9.4.165	

Smolarczyk,	 M.,	 Wiśniewski,	 A.,	 Czajkowska,	 A.,	
Kȩska,	 A.,	 Tkaczyk,	 J.,	 Milde,	 K.,	 .	 .	 .	
Majchrzak,	 A.	 (2012).	 The	 physique	 and	
body	 composition	 of	 students	 studying	
physical	 education:	 a	 preliminary	 report.	
Pediatr	Endocrinol	Diabet	Metab,	18(1),	27-
32.	
https://www.ncbi.nlm.nih.gov/pubmed/2
2525688		

Soteriades,	E.S.,	Hauser,	R.,	Kawachi,	I.,	Liarokapis,	
D.,	 Christiani,	 D.C.,	 &	 Kales,	 S.N.	 (2005).	
Obesity	 and	 cardiovascular	 disease	 risk	
factors	in	firefighters.	A	prospective	cohort	
study.	 Obes	 Res,	 13(10),	 1756-1763.	
https://doi.org/10.1038/oby.2005.214		

Sprynarová,	 Ṥ.,	&	Parizková,	 J.	 (1971).	Functional	
capacity	 and	 body	 composition	 in	 top	
weight	 lifters,	 swimmers,	 runners	 and	
skiers.	Int	Z	angew	Physiol,	29(2),	184-194.	
https://doi.org/10.1007/BF00698026		

Stachoń,	 A.,	 &	 Pietraszweska,	 J.	 (2013).	 Body	
composition	 in	 male	 physical	 education	
university	 students	 in	 view	 of	 their	
physical	 activity	 level.	 Hum	 Movement,	
14(3),	 205-209.	
https://doi.org/10.2478/humo-2013-
0024		

Tarnus,	E.,	&	Bourdon,	E.	(2006).	Anthropometric	
evaluations	 of	 body	 composition	 of	

undergraduate	students	at		the	University	
of	 La	 Réunion.	 Adv	 Physiol	 Educ,	 30(4),	
248-253.	
https://doi.org/10.1152/advan.00069.20
05		

Tremblay,	 M.S.,	 Barnes,	 J.D.,	 Copeland,	 J.L.,	 &	
Esliger,	 D.W.	 (2005).	 Conquering	
childhood	 inactivity.	 Is	 the	answer	 in	 the	
past?	 Med	 Sci	 Sports	 Exerc,	 37(7	 ),	 1187-
1194	
https://doi.org/10.1249/01.mss.0000170
079.71522.a6	

Tremblay,	M.S.,	Esliger,	D.W.,	Copeland,	J.L.,	Barnes,	
J.D.,	 &	 Bassett,	 D.R.	 	 (2008).	 Moving	
forward	by	looking	back:	lessons	learned	
from	 long	 lost	 lifestyles.	 Appl	 Physiol	
Nutr	 Metab,	 33(4),	 836-842.		
https://doi.org/10.1139/H08-045		

Tryon,	 W.W.,	 Goldberg,	 J.L.,	 &	 Morrison,	 D.F.	
(1992).	 Activity	 decreases	 as	
PMid:1326489	

Tsismenakis,	 A.J.,	 Chriatophi,	 C.A.,	 Burress,	 J.W.,	
Kinney,	 A.M.,	 Kim,	 M.,	 &	 Kales,	 S.N.	
(2009).	The	obesity	epidemic	and	future	
emergency	responders.	Obesity	17,	1648-
1650.	
https://doi.org/10.1038/oby.2009.63		

Tudor-Locke,	 C.,	 Ainsworth,	 B.E.,	 Whitt,	 M.C.,	
Thompson,	R.W.,	Addy,	C.L.,	&	Jones,	D.A.	
(2001).	 The	 relationship	 between	
pedometer-dtermined	 ambulatory	
activity	and	body	composition	variables.	
Int	 J	 Obesity,	 25(11),	 1571-1578.	
https://doi.org/10.1038/sj.ijo.0801783	

Tudor-Locke,	C.,	Brashear,	M.M.,	 Johnson,	W.D.,	&	
Katzmarzyk,	 P.T.	 (2010).	 Accelerometer	
profiles	 of	 physical	 activity	 and		
inactivity	 in	 normal	 weight,	 overweight	
and	 obese	 U.S.men	 and	 women.	 Int	 J	
Behav	 Nutr	 Phys	 Activ,	 7:60.	
https://doi.org/10.1186/1479-5868-7-
60		

Upton,	S.J.,	Hagan,	R.D.,	Rosentswieg,	J.,	&		Gettman,	
L.R.	 (1983).	 Comparison	 of	 the	
physioogical	 profile	 of	 middle-aged	
women	 distance	 runners	 and	 sedentary	
women.	Res	Quart	Exerc	Sport,	54(1),	83-
87.	
https://doi.org/10.1080/02701367.1983.
10605277	

Vaccaro,	 P.,	 Morris,	 A.F.,	 &	 Clarke,	 D.H.	 (1981).	
Physiologic	 characteristics	 of	 Masters	
female	 distance	 runners.	 Phys	 Sportsmed,	
9(7),	 105-108.	



Can	Physical	Activity	Prevent	Obesity?	Part	2	

 

Health	&	Fitness	Journal	of	Canada,	ISSN	1920-6216,	Vol.	12,	No.	2	⋅	June	30,	2019	⋅	96	

https://doi.org/10.1080/00913847.1981.
11711122	

Vernillo,	G.,	Schena,	F.,	Berardelli,	C.,	Rosa,	G.,	lvani,	
C.,	 Maggioni,	 M.,	 .	 .	 .	 La	 Torre,	 A.	 (2013).	
Anthropometric	 characteristics	 of	 top-
class	 Kenyan	 marathon	 runners.	 J	 Sports	
Med	 Phys	 Fitness,	 53,	 403-408.	
https://www.ncbi.nlm.nih.gov/pubmed/2
3828288		

Wanner,	M.,	Martin,	B.W.,	Autenrieth,	C.,	Schaffner,	
E.,	 Meier,	 F.,	 Brombach,	 C.,	 .	 .	 .	 Probst-
Hensch	 N.	 (2016).	 Associations	 between	
domains	 of	 physical	 activity,	 sitting	 time	
and		 different	measures	of	overweight	
and	 obesity.	 Prev	 Med	 Rep,	 3,	 177-184.	
https://doi.org/10.1016/j.pmedr.2016.01.
007	

Washburn,	 R.A.,	 Smith,	 K.W.,	 Goldfield,	 S.R.W.,	 &	
McKinlay,	J.B.	(1991).	Reliability		 and	
physiological	 correlates	 of	 the	 Harvard	
Alumni	 Acrivity	 Survey	 in	 a	 general	
population.	J	Clin	Epidemiol,	44(12),	1319-
1326.	 https://doi.org/10.1016/0895-
4356(91)90093-O	

Wasiluk,	A.,	Saczuk,	 J.,	Czeczelewski,	 J.,	Eucolecka,	
B.,	 &	 	 Czeczelewska,	 E.	 (2013).	
Morphological	 characteristics	 of	 female	
students	 of	 physical	 education	 compared	
to	students	of	other	subjects	at	the	Faculty	
of	 Physical	 Education	 and	 Sport	 in	 Biala	
Podalska.	Pol	J	Sport	Tourism,	20,	279-282.	
https://doi.org/10.2478/pjst-2013-0026	

Wilmore,	J.H.,	&	Brown,	C.H.	(1974).	Physiological	
profiles	 of	 women	 distance	 runners.	 Med	
Sci	 Sports,	 6(3),	 178-181.	
https://doi.org/10.1249/00005768-
197423000-00010	

Wilmore,	 J.H.,	Brown,	C.H.,	&	 	Navis,	 J.A	 .	 (1977).	
Body	 physique	 and	 composition	 of	 the	
female	distance	runner.	Ann	NY	Acad	Sci,	
301(1),	 764-776.	
https://doi.org/10.1111/j.1749-
6632.1977.tb38245.x	

Young,	 T.K.	 (1994).	 Human	 obesity	 and	 arctic	
adaptation;	 epidemiological	 patterns,	
metabolic	 effects	 and	 evolutionary	
implications.	 D.Phil.dissertation,	 Oxford	
University,	Oxford,	UK,	pp.	1-251.				

Zanovec,	M.,	Lakkakula,	A.P.,	Johnson,	L.G.,	&	Turri,	
G.	(2009).	Physical	activity	is	associated	
with	 percent	 body	 fat	 and	 body	
composition	 but	 not	 body	 mass	 inexin	
white	 and	 black	 college	 students.	 Int	 J	
Exerc	 Sci,	 2(3),	 175-185.	

https://www.ncbi.nlm.nih.gov/pmc/arti
cles/PMC4739486/