This is a sel -a chi ed e sion o an o iginal a icle. This e sion
may di e om he o iginal in pagina ion and ypog aphic de ails.
Au ho (s):
Ti le:
Yea :
Ve sion:
Copy igh :
Righ s:
Righ s u l:
Please ci e he o iginal e sion:
CC BY 4.0
h ps://c ea i ecommons.o g/licenses/by/4.0/
E ec s o an Indi idualized Ac i e Aging Counseling In e en ion on Mobili y and
Physical Ac i i y : Seconda y Analyses o a Randomized Con olled T ial
© The Au ho (s) 2020
Published e sion
Sil anen, Sini; Po egijs, E ja; Pynnönen, Ka ja; Hassand a, Ma y; Ran alainen,
Timo; Ka a i a, Lau a; Saajanaho, Milla J.; Ran anen, Taina
Sil anen, S., Po egijs, E., Pynnönen, K., Hassand a, M., Ran alainen, T., Ka a i a, L., Saajanaho,
M. J., & Ran anen, T. (2020). E ec s o an Indi idualized Ac i e Aging Counseling In e en ion on
Mobili y and Physical Ac i i y : Seconda y Analyses o a Randomized Con olled T ial. Jou nal o
Aging and Heal h, 32(10), 1316-1324. h ps://doi.o g/10.1177/0898264320924258
2020
A icle
Jou nal o Aging and Heal h
2020, Vol. 0(0) 1–9
© The Au ho (s) 2020
A icle euse guidelines:
sagepub.com/jou nals-pe missions
DOI: 10.1177/0898264320924258
jou nals.sagepub.com/home/jah
E ec s o an Indi idualized Ac i e Aging
Counseling In e en ion on Mobili y and
Physical Ac i i y: Seconda y Analyses o
a Randomized Con olled T ial
Sini Sil anen, MSc
1
, E ja Po egijs, PhD
1
, Ka ja Pynn ¨
onen, PhD
1
,
Ma y Hassand a, PhD
2
, Timo Ran alainen, PhD
1
, Lau a Ka a i a, PhD
1
,
Milla J. Saajanaho, PhD
1
, and Taina Ran anen, PhD
1

Abs ac
Objec i es: The aim o his s udy was o epo p eplanned seconda y analyses o he e ec s o a 12-mon h indi idualized
ac i e aging counseling in e en ion on six mobili y and physical ac i i y ou comes. Me hods: A wo-a m, single-blinded
andomized con olled ial was conduc ed among 75- and 80-yea -old communi y-dwelling people. The in e en ion g oup
(IG, n= 101) ecei ed counseling aimed a inc easing sel -selec ed, p ima ily ou -o -home ac i i y. The con ol g oup (CG, n=
103) ecei ed gene al heal h in o ma ion. Da a we e analyzed wi h gene alized es ima ing equa ions. Resul s: Physical
pe o mance imp o ed in he IG mo e han ha in he CG (g oup by ime p= .022), sel - epo ed physical ac i i y inc eased in
bo h g oups ( ime p= .012), and au onomy in ou doo mobili y declined in he IG and was enhanced in he CG (g oup by ime
p= .011). No change was obse ed o li e-space mobili y, p opo ion o pe sons pe cei ing di ficul y walking 2 km, o
moni o ed physical ac i i y. Discussion: Indi idualized counseling aiming a inc easing sel -selec ed ou -o -home ac i i y had
nonsys ema ic e ec s on mobili y and posi i ely a ec ed physical pe o mance only.
Keywo ds
meaning ul ac i i y, physical unc ion, li e-space, au onomy, andomized con olled ial
In oduc ion
Mobili y and physical ac i i y (PA) a e closely in e wined
wi h many e e yday ac i i ies o olde people, such as
making social isi s, a ending e en s, shopping, o unning
e ands (Tsai e al., 2016). Inc easing pa icipa ion in any
meaning ul ac i i y ou side he home will likely inc ease PA
and p omo e mobili y in e ms o ex ending li e space (Ba nes
e al., 2007;Saajanaho e al., 2014;Saajanaho e al., 2015). In
gene al, op imal mobili y e e s o he abili y o mo e onesel
sa ely om one place o ano he (Sa a iano e al., 2012) and
allows o pa icipa ion in di e en ac i i ies in a a ie y
o en i onmen s. I may be iewed om such di e se pe -
spec i es as (1) he abili y o pe o m a ious asks such as
walk gi en dis ances o climb s ai s (M¨
an y e al., 2007), (2)
he ex en o mo ing abou ei he on oo o ia anspo a ion
(Peel e al., 2005), and (3) pe cei ed abili y o decide whe e,
when, and how o mo e, ha is au onomy in ou doo mobili y
(Ca dol e al., 1999;Wilkie e al., 2006). These aspec s can
all be assessed by obse ing o moni o ing pa icipan s o
by sel - epo s, and combined, hey will p o ide a mo e
comp ehensi e unde s anding o a pe son’s o e all mobili y.
G ea e li e-space mobili y, desc ibing he spa ial a ea
a pe son mo es h ough in daily li e (Bake e al., 2003),
co ela es wi h a highe le el o PA (Po egijs e al., 2015;Tsai
e al., 2015), be e physical pe o mance, g ea e au onomy
in ou doo mobili y (Po egijs, Ran akokko, Mikkola e al.,
2014), and ewe pe cei ed di ficul ies in walking longe
dis ances (Ran akokko e al., 2017). Viewed he o he way
ound, mobili y is a p e equisi e o main aining posi i e
social oles, a good quali y o li e, and independence in old
age (Pa la & Shumway-Cook, 1999;Ran akokko e al., 2013;
Ran anen, 2013). Fo example we obse ed ha inc eased
engagemen in any ou -o -home ac i i y can imp o e he
physical domain o quali y o li e, e en among olde peo-
ple wi h se e e mobili y limi a ions (Ran anen e al., 2015).
1
Uni e si y o Jy askyla, Finland
2
Uni e si y o Thessaly, G eece
Co esponding Au ho :
Sini Sil anen, Ge on ology Resea ch Cen e and Facul y o Spo and
Heal h Sciences, Uni e si y o Jy ¨
askyl¨
a, PO Box 35 (Vi eca 271) FI-40014
Jy ¨
askyl¨
a, Finland.
Email: sini.k. .sil anen@jyu.fi
Fu he mo e, ou p e ious s udies ha e shown ha s i ing o
ac i i y- ela ed goals is associa ed wi h highe li e-space
mobili y (Saajanaho e al., 2015) and p edic s g ea e exe -
cise ac i i y (Saajanaho e al., 2014). Because people o en
s i e o each hei goals, we expec ed mobili y and PA o be
p omo ed as a “side e ec ”o inc easing he pu sui o any
meaning ul and sel -selec ed ac i i y ha akes place ou side
he home.
This s udy epo s p eplanned seconda y analyses o
a andomized con olled ial o ac i e aging counseling
among communi y-dwelling olde people. The in e en ion,
which cen e ed on suppo ing he pa icipan s’au onomous
mo i a ion and goal se ing and inc easing hei awa eness o
desi able ou -o -home ac i i ies (Ran anen e al., 2019),
enhanced he pa icipan s’ac i e aging sco e, al hough he
e ec was small (Ran anen e al., 2020). The aim o he
p esen s udy was o es whe he he indi idualized coun-
seling in e en ion also a ec s physical pe o mance, pe -
cei ed di ficul ies in walking 2 km, li e-space mobili y,
pe cei ed au onomy in ou doo mobili y, and sel - epo ed
and objec i ely moni o ed PA. We expec ed posi i e changes
in he in e en ion g oup (IG) and no o smalle changes in
he con ol g oup (CG), which ecei ed gene al heal h in-
o ma ion by o dina y mail.
Me hods
Design and Pa icipan s
This s udy epo s p eplanned seconda y analyses o a com-
muni y-based wo-a m single-blinded andomized con-
olled ial (ISRCTN16172390), “indi idualized counseling
o ac i e aging—AGNES in e en ion.”The ial has been
desc ibed in-dep h in he s udy p o ocol (Ran anen e al.,
2019), and he p ima y ou comes ha e been epo ed else-
whe e (Ran anen e al., 2020). B iefly, he s udy pa icipan s
we e communi y-dwelling olde people li ing in he ci y o
Jy ¨
askyl¨
a in Cen al Finland. The ial comp ised wo pa allel
g oups: an IG and a CG wi h a 1:1 alloca ion a io. Pa ic-
ipan s we e ec ui ed om among he pa icipan s o he
AGNES coho s udy (Ran anen e al., 2018) be ween Oc-
obe 2017 and Augus 2018. The inclusion c i e ia o he
ial we e willingness o pa icipa e, age 75 o 80 yea s,
a baseline sco e be ween 52.3 and 90.0 on he Uni e si y o
Alabama a Bi mingham Li e-Space Assessmen (LSA)
(Bake e al., 2003), and a minimum sco e o 25 on he Mini-
Men al S a e Examina ion (Fols ein e al., 1975). In addi ion,
pa icipan s we e expec ed o be able o communica e. These
c i e ia we e chosen o include pa icipan s who ha e oom
o imp o emen in hei ac i i y le els and whose cogni i e
unc ion enables compliance wi h he in e en ion. Pe sons
pa icipa ing in ano he ongoing in e en ion we e excluded.
In acco dance wi h he powe calcula ions made o he
p ima y ou come (Ran anen e al., 2019), 101 pe sons we e
andomly alloca ed o he IG and 103 o he CG. The s udy
s a is ician gene a ed he andom alloca ion sequence wi h
S a a 15.0 s a is ical so wa e and sealed hem in en elopes.
Randomiza ion was s a ified by age and gende . A e he
p e ial da a collec ion was comple ed, he s udy counselo
opened he andomiza ion en elopes. The flowcha o he
s udy was p e iously published as pa o he p ima y ou come
a icle (Ran anen e al., 2019), and hus appended o his
a icle in Appendix A. Mobili y and PA we e assessed p e ial
(be o e andomiza ion) and pos ial (a 12 mon hs) by home
in e iews and ac i i y moni o ing in he ee-li ing en i-
onmen using accele ome e s. In e iewe s and assesso s
we e blinded o ea men g oup alloca ion. A o al o 17
pe sons (n= 10 in he IG and n= 7 con ols) d opped ou
du ing he ial. O hese, wo d opped ou immedia ely a e
andomiza ion and hus did no ecei e he in e en ion.
Reasons o d opping ou we e unwillingness o con inue (n=
13), heal h decline (n= 2), and dea h (n= 1). In addi ion, one
pa icipan ’s all ollow-up da a we e damaged.
In e en ion
The in e en ion’s aim was o inc ease sel -selec ed mean-
ing ul ac i i y in e e yday li e. Al hough he emphasis in
he counseling and suppo i e ma e ial was on inc easing
pa icipa ion in ou -o -home physical and social ac i i ies,
pa icipan s we e suppo ed in s i ing o any goals hey ound
impo an (Ran anen e al., 2019). To enable pe sonaliza ion
o he in e en ion, pa icipan s we e p ofiled acco ding o
hei baseline heal h s a us, social con ac s, and he le el o
well-being, and he counseling p o ocol hen was adjus ed o
hei p e e ed ac i i ies and goals. The counseling app oach
was based on wo majo mo i a ional heo ies: he sel -
de e mina ion heo y, which highligh s he impo ance o in-
insic and sel -de e mined a he han ex e nal and egula ed
mo i a ion behind ac ions (Deci & Ryan, 2000), and he heo y
o planned beha io , which emphasizes he ole o belie s and
in en ions as a basis o desi able beha io (Ajzen, 1985).
The in e en ion included a 90-minu es ace- o- ace
counseling session a he esea ch cen e a he beginning
o he s udy and ou sho e phone counseling sessions a
mon hs 1, 3, 6, and 9. The counseling sessions ollowed
a semis uc u ed p o ocol conce ning, o example pa ic-
ipan s’cu en ac i i ies, goals, and ac ion plans. Du ing
he fi s session, pa icipan s we e p o ided wi h suppo -
i e ma e ials, such as an ac i e aging in o ma ion bookle ,
a calenda , and a newsle e (Ran anen e al., 2019). The
newsle e , ea u ing in o ma ion on he ac i i ies a ailable in
he ci y o Jy ¨
askyl¨
a and s o ies o o he pa icipan s’success
in expe iencing an ac i e li e, was upda ed and sen o he
pa icipan s e e y h ee mon hs du ing he ial. The sub-
sequen phone counseling sessions p o ided social suppo ,
eedback, and encou agemen ela ed o pu suing he selec ed
goals. A ained counselo wi h p e ious expe ience in
counseling olde adul s implemen ed he in e en ion.
2Jou nal o Aging and Heal h 0(0)
Con ol G oup
Con ols we e mailed p in ed b ochu es and bookle s ela ed
o gene al heal h a mon hs 1, 3, 6, and 9. B ochu es and
bookle s we e ob ained om di e en na ional public heal h
associa ions and so ed in o ou hemes: (1) exe cise, (2)
nu i ion, (3) ca dio ascula diseases, and (4) ype II diabe es.
Measu emen s
Physical pe o mance. Physical pe o mance was assessed
wi h he Sho Physical Pe o mance Ba e y (SPPB), which
includes es s o s anding balance ( ee oge he , semi-
andem, and andem), no mal gai speed o e 3 m, and chai
ise ime (5 s ands) (Gu alnik e al., 1994;Ran anen e al.,
2018). Each es was sco ed om 0 (lowes ) o 4 (highes ).
The indi idual es sco es we e summed o o m a o al sco e
( ange 0–12) wi h highe sco es indica ing be e physical
pe o mance.
Pe cei ed walking di ficul ies. Pe cei ed walking di ficul ies
we e epo ed o a 2-km dis ance wi h a alida ed ques ion
(M¨
an y e al., 2007). Pa icipan s we e asked whe he hey a e
able o walk 2 km, and he esponse op ions we e “able
wi hou di ficul y,”“able wi h some di ficul y,”“able wi h
a g ea deal o di ficul y,”“unable wi hou he help o ano he
pe son,”and “unable o manage e en wi h help.”Fo bina y
logis ic modeling, he esponses we e ca ego ized in o “no
di ficul ies” e sus “walking di ficul ies,”when a leas some
di ficul ies in walking we e epo ed.
Li e-space mobili y. Li e-space mobili y was assessed wi h he
Uni e si y o Alabama a Bi mingham S udy o Aging LSA
(Bake e al., 2003), which eflec s he equency and in-
dependence o mobili y h ough di e en li e-space le els
du ing he p eceding 4 weeks. Li e-space le els s a om he
pe son’s bed oom and ex end o o he ooms, ya d, neigh-
bo hood, own, and beyond own. Pa icipan s we e asked
whe he hey ha e mo ed in hese li e-space a eas du ing he
p eceding 4 weeks, and i so, how o en and whe he hey
needed help om any de ices o ano he pe son. A composi e
sco e( ange0–120) was used in he p esen analyses wi h
highe sco es indica ing g ea e li e-space mobili y. The alidi y
and eliabili y o he measu e ha e been es ablished among
olde people in Finland (Po egijs, Iwa sson, Ran akokko e al.,
2014).
Au onomy in ou doo mobili y. Au onomy in ou doo mobili y
was measu ed wi h he “au onomy ou doo s”subscale o he
alida ed Impac on Pa icipa ion and Au onomy ques ion-
nai e (Ca dol e al., 2001;Ke s en e al., 2007). The ou doo s
subscale assesses he pe son’s sel - a ed possibili ies o (1)
isi ela i es and iends, (2) make ips and a el, (3) spend
leisu e ime, (4) mee o he people, and (5) li e li e as he/she
wan s. Responses a e gi en on a 5-poin Like scale anging
om e y good (0) o e y poo (4). A sum sco e ( ange 0–20)
was calcula ed wi h highe sco es indica ing poo e au on-
omy. One missing i em was allowed, and o al sco es we e
impu ed o wo pe sons (one a baseline and one a ollow-
up) based on he mean o hei exis ing alues a he same
ime poin .
Sel - epo ed PA. Sel - epo ed PA was assessed wi h he
second pa o he Yale Physical Ac i i y Su ey, which is an
in e iew-adminis e ed ques ionnai e on (1) igo ous ac i i y,
(2) walking, (3) gene al mo ing, (4) s anding, and (5) si ing
(Dipie o e al., 1993). Fo he p esen s udy, we calcula ed
a To al Ac i i y Summa y Index ( ange 0–137), wi h highe
sco es indica ing highe PA. The alidi y and eliabili y o he
scale a e mode a e (Schule e al., 2001). One pa icipan did
no ha e da a on sel - epo ed PA a ei he baseline o ollow-
up and hus was excluded om he analysis.
Moni o ed PA. Pa icipan s we e asked o wea a iaxial
accele ome e (13-bi ± 16 g, UKK RM42, UKK Te ey-
spal elu Oy, Tampe e, Finland) con inuously o 7–10 days
p e ial (Po egijs e al., 2019) and o 6 days pos ial im-
media ely ollowing he home in e iew. Only hose who
pa icipa ed in he p e ial moni o ing (n= 139, 68% o he
o al sample) we e in i ed o pa icipa e in he pos ial
moni o ing. The senso was a ached on he an e io aspec o
he dominan high (i.e. he ake-o leg) wi h a wa e p oo
sel -adhesi e film. PA, exp essed as mean 24-h accele a ion
(millig a i y, mg)(Rowlands, 2018), was compu ed as he
mean high-pass fil e ed ec o magni ude o nono e lapping
5-s epochs (Van Hees e al., 2013). A e age accele a ion
summa izes all mo emen wi hou an in ensi y h eshold,
combines bo h he in ensi y and du a ion o ac i i y in o
a single measu e, and p oduces alues ha a e di ec ly
compa able in he same wea loca ion. Highe alues indica e
a g ea e o al olume o ac i i y. A minimum o h ee ull
days o da a we e equi ed a bo h assessmen poin s.
Backg ound cha ac e is ics. Backg ound cha ac e is ics in-
cluded ca ego ical a iables o age, gende , pe cei ed heal h,
ma i al s a us, li ing alone, and le el o educa ion. Age and
gende we e d awn om he na ional popula ion egis e and
o he a iables sel - epo ed. Pe cei ed heal h was dicho-
omized as good o e y good heal h e sus mode a e o
poo e heal h. The le el o educa ion was ca ego ized as
ollows: high (high school diploma o uni e si y deg ee),
in e media e (middle school, olk high school, oca ional
school, o seconda y school), and low (p ima y school o less).
S a is ical Me hods
Pa icipan s’backg ound cha ac e is ics we e examined
sepa a ely in he in e en ion and CGs, and be ween-g oup
di e ences we e es ed wi h he chi-squa e es (χ
2
). In
compliance wi h he p inciples o in en ion- o- ea analysis
Sil anen e al. 3
(McCoy, 2017), he in e en ion’s e ec s on he di e en
mobili y and PA ou comes we e es ed wi h gene al es i-
ma ing equa ion (GEE) analysis wi h an uns uc u ed
wo king co ela ion ma ix. GEE analysis is a semipa ame ic
me hod designed o wo k wi h co ela ed da a and does no
assume a no mal dis ibu ion o a iables (Liang & Zege ,
1986). GEE can also use in o ma ion om incomple e pai s
o obse a ions and hus is sui able o use in cases o missing
da a in longi udinal da ase s (Zhang e al., 2014). Linea
models we e used o con inuous ou comes and bina y lo-
gis ic models o bina y ou comes. We es ed he main e ec s
o g oups and ime and he in e ac ions be ween hese. All
ou comes we e analyzed in sepa a e GEE models.
I a s a is ically significan g oup by a ime e ec was
obse ed in any o he ou comes, we calcula ed a change
sco e by sub ac ing he baseline sco e om he ollow-up
sco e. We used hese change sco es and hei s anda d de-
ia ions o calcula e e ec sizes acco ding o Cohen’sd o -
mula (Cohen, 1992). Confidence in e als (CIs) o Cohen’s
dwe e calcula ed using he o mula by Lee (2016). In ad-
di ion, we calcula ed ela i e imp o emen sco es (pe cen ual
posi i e change) and es ed be ween-g oup di e ences wi h
he independen samples - es and wi hin-g oup di e ences
wi h he pai ed sample - es . All analyses we e pe o med
wi h IBM SPSS S a is ics e sion 24.0 o Windows.
Resul s
Backg ound Cha ac e is ics
Pa icipan s’baseline cha ac e is ics by ea men g oup al-
loca ion a e p esen ed in Table 1. O he subsample o 139
olun ee pa icipan s included in he PA moni o ing a
baseline, 67 we e andomized in o he IG and 72 in o he CG.
The cha ac e is ics o his subsample we e ound o be simila
in bo h g oups (χ
2
= .58–.81, Table 1).
In e en ion E ec s
In he GEE analyses, g oup, ime, and g oup-by- ime e ec s
we e no s a is ically significan o li e-space mobili y,
moni o ed PA (a e age accele a ion), o pe cei ed walking
di ficul ies (Table 2). The ime e ec was significan o sel -
epo ed PA (p= .012), indica ing ha PA inc eased bo h in
he con ol and IGs. The only s a is ically significan g oup-
by- ime e ec s we e obse ed o physical pe o mance (p=
.022) and pe cei ed au onomy (p= .011). A baseline, he IG
had poo e physical pe o mance han he CG, bu du ing he
in e en ion, i eached he same le el o pe o mance as he
CG and hus demons a ed g ea e imp o emen (+5.0%
s +1.8%). In addi ion, whe eas au onomy in he CG im-
p o ed du ing he 1-yea ial, au onomy in he IG declined
Table 1. Backg ound Cha ac e is ics o he Pa icipan s by T ea men G oup Alloca ion.
All (n= 204) PA Moni o ing (n= 139)
Cha ac e is ic In e en ion, N(%) Con ol, N(%) In e en ion, N(%) Con ol, N(%)
Age (yea s)
75 75 (74) 77 (75) 49 (73) 55 (76)
80 26 (26) 26 (25) 18 (27) 17 (24)
Gende
Female 61 (60) 63 (61) 41 (61) 46 (64)
Male 40 (40) 40 (39) 26 (39) 26 (36)
Pe cei ed heal h
Good o e y good 54 (54) 64 (62) 35 (52) 41 (57)
Mode a e o poo e 47 (47) 39 (38) 32 (48) 31 (43)
Ma i al s a us
Ma ied 61 (60) 68 (67) 46 (69) 51 (72)
No ma ied 40 (40) 34 (33) 21 (31) 20 (28)
Li ing alone
Yes 37 (37) 33 (32) 20 (30) 20 (28)
No 64 (63) 70 (68) 47 (70) 52 (72)
Le el o educa ion
High 28 (28) 32 (31) 18 (27) 21 (29)
In e media e 48 (48) 56 (54) 35 (52) 39 (54)
Low 24 (24) 15 (15) 14 (21) 12 (17)
No e. PA = physical ac i i y.
4Jou nal o Aging and Heal h 0(0)

(+1.7% s 3.0%, Figu e 1). The e ec size o physical
pe o mance was d= .31 (95% CI .02–.59) and o pe cei ed
au onomy d= .36 (95% CI .07 –.64).
Discussion
The indi idualized ac i e aging counseling in e en ion had
no sys ema ic e ec s on mobili y o PA among olde people.
Ins ead, he e ec s we e inconsis en depending on he
aspec o mobili y s udied, as physical pe o mance im-
p o ed and sense o au onomy declined in he IG. Sel -
epo ed PA inc eased in bo h g oups. No ea men e ec s
we e obse ed o li e-space mobili y, pe cei ed di ficul ies
in walking 2 km, o moni o ed PA. Al hough he po en ial
associa ion o s i ing o pa icipa ion in meaning ul ac-
i i y wi h g ea e li e-space mobili y and PA has been
es ablished in p e ious obse a ional s udies (Saajanaho e
al. 2015), inc eased ac i e aging (Ran anen e al., 2020)did
no ansla e o enhanced li e-space mobili y in he p esen
andomized con olled ial.
Compa ed wi h con ols, a small bu s a is ically signifi-
can imp o emen was obse ed in physical pe o mance in
he IG bu no change in pe cei ed di ficul y in walking 2 km.
Compa ed wi h he bina y pe cei ed walking di ficul y
a iable, he SPPB sco e is mo e sensi i e o change (Os i
e al., 2002) and cap u es smalle imp o emen s. Ou an-
ecdo al da a om he counseling sessions sugges ha some
pa icipan s in he IG s a ed esis ance aining o a -home
exe cises. These physical ac i i ies may no be ully cap u ed
by accele ome e s o PA and li e-space mobili y ques-
ionnai es bu will likely imp o e lowe ex emi y pe o -
mance (Ran anen, 2013).
Table 2. Means o Mobili y and Physical Ac i i y Va iables by T ea men G oup Alloca ion a Baseline and a 12-Mon h Follow-Up,
and p-Values o G oup, Time, and G oup-By-Time In e ac ion E ec s Tes ed wi h Gene al Es ima ing Equa ion Analysis.
Baseline Follow-Up G oup Time G oup × Time
Ou come Mean (SD) Mean (SD)p-Value p-Value p-Value
Physical pe o mance .010 .060 .022
In e en ion 10.6 (1.6) 11.2 (1.3)
Con ol 11.0 (1.2) 11.2 (1.1)
Li e-space mobili y .482 .807 .409
In e en ion 74.4 (9.2) 76.3 (14.5)
Con ol 74.7 (9.3) 74.9 (13.6)
Au onomy in mobili y .043 .204 .011
In e en ion 4.2 (3.1) 4.8 (3.3)
Con ol 4.7 (3.5) 4.3 (2.5)
Sel - epo ed PA
a
.512 .012 .462
In e en ion 58.9 (24.0) 65.6 (24.3)
Con ol 60.3 (20.5) 65.1 (22.6)
A e age accele a ion (mg)
b
.401 .782 .603
In e en ion 23.9 (5.9) 24.0 (7.4)
Con ol 25.6 (8.4) 25.3 (9.9)
2 km walking di ficul y, no .841 .994 .738
In e en ion 81% 78%
Con ol 84% 84%
No e. All ou comes we e analyzed in sepa a e models. PA = physical ac i i y, mg= millig a i y, SD = s anda d de ia ion. N= 204 o o he models.
a
n= 203.
b
n= 139.
Figu e 1. Mean ela i e imp o emen s (%) wi h s anda d e o s
in physical pe o mance (Sho Physical Pe o mance Ba e y) and
pe cei ed au onomy in ou doo mobili y (Impac on Pa icipa ion
and Au onomy ou doo s subscale) du ing he 12-mon h ial by
ea men g oup alloca ion.
∗
Significan be ween-g oup di e ence,
+
significan wi hin-g oup di e ence, p< .05.
Sil anen e al. 5
Au onomy in ou doo mobili y declined in he IG bu was
sligh ly enhanced in he CG. This was coun e o ou ex-
pec a ions because he counseling app oach was au onomy-
suppo ing and we used beha io al change echniques, such
as p oblem sol ing and ac ion planning, which aimed a
helping pa icipan s ob ain social and p ac ical suppo in
pe o ming hei selec ed ac i i ies. No es ablished cu poin
exis s o a meaning ul change in he p esen au onomy sco e;
howe e , he small decline obse ed may ha e been mean-
ing ul as i eflec s pa icipan s’own pe cep ions o hei
e e yday li e. The decline may po en ially be explained by
pa icipan s becoming awa e o a ailable ac i i ies du ing
he in e en ion bu no ecei ing p ac ical help in engag-
ing in hem (B and s ¨
ad e , 2009). Thus, we may ha e un-
in en ionally c ea ed an imbalance in people’s aspi a ions o
ac i i y ela i e o hei esou ces.
Al hough no ea men e ec s we e ound on li e-space
mobili y o objec i ely moni o ed PA, sel - epo ed PA in-
c eased in bo h g oups. Compa ed wi h hose pa icipa ing in
he sel - epo s only, hose also in ol ed in he PA moni o ing
epo ed ewe dep essi e symp oms and a highe le el o PA
(Ran anen e al., 2019), hus po en ially possessing less oom
o imp o emen in hei ac i i y le els. Howe e , sel -
epo ed PA inc eased simila ly among bo h subg oups,
making selec ion bias an unlikely explana ion o his finding.
Ins ead, i has been ound ha sel - epo s o PA a e ul-
ne able o bias ela ed o social desi abili y and app o al
(Adams e al., 2005) and may lead o o e es ima ions o
ac i i y (S eene-Johannessen e al., 2016). The simila change
in he in e en ion and CGs may be due o he Haw ho ne
e ec , ha is he ac ha people o en ac di e en ly simply
because hey a e being s udied a he han because o he
in e en ion hey a e ecei ing (Becke e al., 2003).
O e all, hese findings imply ha changing e e yday
beha io s is challenging. Acco ding o anecdo al da a om
he counseling sessions, many o ou pa icipan s p e e ed
s i ing o main ain hei cu en si ua ion and ac i i y a he
han se ing new ac i i y goals. This main enance a he han
g ow h o ien a ion in goal pu sui is common in old age and
co ela es wi h well-being (Ebne e al., 2006) bu may ha e
led o in a iabili y in he p esen ou comes. In addi ion, al-
hough he counseling was cen e ed on p omo ing ou -o -
home ac i i y, some people a he se goals ela ed o mo e
seden a y and a -home ac i i ies. Fu he mo e, as we de ec ed
only small inc eases in he ial’s p ima y ou come a iable o
o e all ac i i y (Ran anen e al., 2020), i is unde s andable
ha he changes in mobili y and PA ou comes we e also
modes . Finally, ec ui men o he p esen ial was based on
a popula ion-based p obabili y sample, which should educe
selec ion bias. Howe e , ac i i y s udies o his kind end o
a ac heal hy, ac i e, and in e es ed people (Po egijs e al.,
2019), which, in u n, may lead o pa icipan s wi h highe
han expec ed le els o unc ioning and ac i i y.
S eng hs o he S udy
The s eng hs o his s udy include he andomized con olled
ial design and he use o alida ed and es ablished measu es.
Unlike in many ea lie s udies, we conside ed a ious mo-
bili y and PA ou comes, including bo h subjec i e and ob-
jec i e measu es. In addi ion, ou popula ion-based sample
con ained bo h men and women a ages ha a e ulne able o
unc ional decline. Finally, we had ba ely any missing da a
and low a i ion, as only a ew pa icipan s d opped ou o he
a he long ial. Fu he mo e, we applied GEE analysis,
which also akes unpai ed obse a ions in o accoun (Zhang
e al., 2014) in examining he causal associa ions.
S udy Limi a ions
A no able limi a ion in his s udy was he use o only wo
assessmen poin s: baseline and 12 mon hs he ea e . Thus,
we do no know how mobili y o PA migh ha e changed in
be ween hese ime poin s. Fo example i is possible ha he
in e en ion posi i ely a ec ed mobili y and/o PA imme-
dia ely a e he ace- o- ace counseling, bu as ace- o- ace
con ac was eplaced by phone calls and became less e-
quen , such e ec s diminished. In addi ion, we may ha e
ailed o ec ui enough people wi h ea ly phase decline in
heal h and ac i i y, as such indi iduals would likely ha e had
mo e oom and mo i a ion o inc easing hei le el o ac-
i i y. Finally, i should be no ed ha PA pa e ns and li e-
space mobili y may be highly a iable due o he no mal
a ia ion in e e yday li e (Te wee e al., 2010)o o he
ac o s such as wea he condi ions (Po egijs, Iwa sson,
Ran akokko e al., 2014) ha make i ha de o de ec
change in longi udinal da a wi h momen a y assessmen s o
mobili y.
Conclusion
We ound ha indi idualized counseling cen e ed on in-
c easing sel -selec ed meaning ul ac i i y ou side he home
has inconsis en e ec s on mobili y and PA. Al hough
modes posi i e changes we e seen in physical pe o -
mance in he IG, pe cei ed au onomy in ou doo mobili y
declined and no di e gen changes we e obse ed in he
o he ou come a iables. This sugges s ha p omo ing
mobili yandPAisnoeasy askinpopula ion-baseds udies
a ge ing inc eased pa icipa ion in a a ie y o ac i i ies.
Because ea lie s udies on indi idualized PA counseling
ha e epo ed mo e posi i e esul s o physical unc ion
(M¨
an y e al., 2009)andPA(Rasinaho e al., 2012), we may
deduce ha u u e in e en ions ailo ed specifically o
inc easing pa icipa ion in mobili y- ela ed and physical
ac i i ies could yield g ea e posi i e e ec s han hose
epo ed he e.
6Jou nal o Aging and Heal h 0(0)
Appendix A. Flowcha o he S udy (O iginally Published in Ran anen e al. (2020)).
Acknowledgmen s
The au ho s wish o acknowledge CSC—IT Cen e o Science,
Finland, o compu a ional esou ces. The Ge on ology Resea ch
Cen e is a join e o be ween he Uni e si y o Jy ¨
askyl¨
a and he
Uni e si y o Tampe e.
Decla a ion o Conflic ing In e es s
The au ho s decla ed no po en ial conflic s o in e es wi h espec o
he esea ch, au ho ship, and/o publica ion o his a icle.
Funding
The au ho s disclosed eceip o he ollowing financial suppo o
he esea ch, au ho ship, and/o publica ion o his a icle: This s udy
was financially suppo ed by he Eu opean Resea ch Council (g an
numbe 693045 o TaRa), he Academy o Finland (g an numbe
310526 o TaRa, 321336, and 328818 o TiRa), and he Finnish
Minis y o Educa ion and Cul u e ( o EP). The p epa a ion o his
a icle (MS) was also unded by he Academy o Finland (g an
numbe 323541). The unde s had no ole in designing and im-
plemen ing he s udy o epo ing he analyses. In addi ion, he
Sil anen e al. 7
con en o his a icle does no eflec he o ficial opinion o he
Eu opean Union. Responsibili y o he in o ma ion and iews
exp essed in he a icle lies en i ely wi h he au ho s.
E hical S a emen
All pa icipan s ga e hei e bal and w i en in o med consen
be o e da a collec ion, and he ial ollowed he p inciples o he
Decla a ion o Helsinki. E hical s a emen o he s udy was p o ided
by he Cen al Finland Hospi al Dis ic on Augus 23, 2017.
ORCID iD
Sini Sil anen h ps://o cid.o g/0000-0003-2054-1341
Re e ences
Adams, S. A., Ma hews, C. E., Ebbeling, C. B., Moo e, C. G.,
Cunningham, J. E., Ful on, J., & Hebe , J. R. (2005). The e ec
o social desi abili y and social app o al on sel - epo s o
physical ac i i y. Ame ican Jou nal o Epidemiology,161(4),
389-398.
Ajzen, I. (1985). F om in en ions o ac ions: A heo y o planned
beha io . In J. Kuhl, & J. Beckmann (Eds.), Ac ion con ol
(pp. 11-39). Be lin: Sp inge .
Bake , P. S., Bodne , E. V., & Allman, R. M. (2003). Measu ing li e-
space mobili y in communi y-dwelling olde adul s. Jou nal o
he Ame ican Ge ia ics Socie y,51(11), 1610-1614. doi: 10.
1046/j.1532-5415.2003.51512.x
Ba nes, L. L., Wilson, R. S., Bienias, J. L., Mendes de Leon, C. F.,
Kim H.-J. N., Buchman, A. S., Benne , D. A., & Benne , D. A.
(2007). Co ela es o li e space in a olun ee coho o olde
adul s. Expe imen al Aging Resea ch,33(1), 77-93.
Becke , H., Robe s, G., & Voelmeck, W. (2003). Explana ions o
imp o emen in bo h expe imen al and con ol g oups. Wes e n
Jou nal o Nu sing Resea ch,25(6), 746-755.
B and s ¨
ad e , J. (2009). Goal pu sui and goal adjus men : Sel -
egula ion and in en ional sel -de elopmen in changing
de elopmen al con ex s. Ad ances in Li e Cou se Resea ch,
14(1-2), 52-62.
Ca dol,M.,deHaan,R.J.,deJong,B.A., andenBos,G.A.M.,
& de G oo , I. J. M. (2001). Psychome ic p ope ies o he
impac on pa icipa ion and au onomy ques ionnai e. A -
chi es o Physical Medicine and Rehabili a ion,82(2),
210-216.
Ca dol, M., de Haan, R. J., an den Bos, G. A. M., de Jong, B. A., &
de G oo , I. J. M. (1999). The de elopmen o a handicap
assessmen ques ionnai e: The impac on pa icipa ion and
au onomy (IPA). Clinical Rehabili a ion,13(5), 411-419. doi:
10.1191/026921599668601325
Cohen, J. (1992). A powe p ime . Psychological Bulle in,112(1),
155, 159.
Deci, E. L., & Ryan, R. M. (2000). The “Wha ”and “Why”o goal
pu sui s: Human needs and he sel -de e mina ion o beha io .
Psychological Inqui y,11(4), 227-268.
Dipie o, L., Caspe sen, C. J., Os eld, A. M., & Nadel, E. R. (1993).
A su ey o assessing physical ac i i y among olde adul s.
Medicine & Science in Spo s & Exe cise, 26(6), 350.
Ebne , N. C., F eund, A. M., & Bal es, P. B. (2006). De elop-
men al changes in pe sonal goal o ien a ion om young o la e
adul hood: F om s i ing o gains o main enance and p e-
en ion o losses. Psychology and Aging,21(4), 664, 678.
Fols ein, M. F., Fols ein, S. E., & McHugh, P. R. (1975). “Mini-
men al s a e”.Jou nal o Psychia ic Resea ch,12(3),
189-198.
Gu alnik, J. M., Simonsick, E. M., Fe ucci, L., Glynn, R. J.,
Be kman, L. F., Blaze , D. G., Sche , P. A., & Wallace, R. B.
(1994). A sho physical pe o mance ba e y assessing lowe
ex emi y unc ion: Associa ion wi h sel - epo ed disabili y and
p edic ion o mo ali y and nu sing home admission. Jou nal
o Ge on ology,49(2), 85-94.
Ke s en, P., Ca dol, M., Geo ge, S., Wa d, C., Sibley, A., & Whi e,
B. (2007). Validi y o he impac on pa icipa ion and au onomy
ques ionnai e: A compa ison be ween wo coun ies. Disabili y
and Rehabili a ion,29(19), 1502-1509.
Lee, D. K. (2016). Al e na i es o p alue: Confidence in e al and
e ec size. Ko ean Jou nal o Anes hesiology,69(6), 555.
Liang, K.-Y., & Zege , S. L. (1986). Longi udinal da a analysis using
gene alized linea models. Biome ika,73(1), 13-22.
Man y, M., Heinonen, A., Leinonen, R., To makangas, T., Hi ensalo,
M., Kallinen, M., Saka i, R., on Bonsdo , M. B., Heikkinen,
E., & Ran anen T. (2009). Long- e m e ec o physical ac-
i i y counseling on moili y limi a ion among olde people:
A andomized con olled s udy. The Jou nals o Ge on ol-
ogy Se ies A: Biological Sciences and Medical Sciences,
64A(1), 83-89.
M¨
an y, M., Heinonen, A., Leinonen, R., T¨
o m ¨
akangas, T.,
Saka i-Ran ala, R., Hi ensalo, M., on Bonsdo , M. B.,
& Ran anen, T. (2007). Cons uc and p edic i e alidi y o
a sel - epo ed measu e o p eclinical mobili y limi a ion.
A chi es o Physical Medicine and Rehabili a ion,88(9),
1108-1113.
McCoy, E. (2017). Unde s anding he in en ion- o- ea p inciple in
andomized con olled ials. Wes e n Jou nal o Eme gency
Medicine,18(6), 1075-1078.
Os i , G. V., Volpa o, S., F ied, L. P., Cha es, P., & Gu alnik, J. M.
(2002). Reliabili y and sensi i i y o change assessed o
a summa y measu e o lowe body unc ion. Jou nal o Clinical
Epidemiology,55(9), 916-921.
Pa la, A. E., & Shumway-Cook, A. (1999). Dimensions o mobili y:
Defining he complexi y and di ficul y associa ed wi h com-
muni y mobili y. Jou nal o Aging and Physical Ac i i y,
7(1), 7-19.
Peel, C., Bake , P. S., Ro h, D. L., B own, C. J., Bodne , E. V., &
Allman, R. M. (2005). Assessing mobili y in olde adul s: The
UAB s udy o aging li e-space assessmen . Physical The apy,
85(10), 1008-1019.
Po egijs, E., Iwa sson, S., Ran akokko, M., Viljanen, A., & Ran-
anen, T. (2014). Li e-space mobili y assessmen in olde people
in Finland; measu emen p ope ies in win e and sp ing. BMC
Resea ch No es,7(1), 323.
Po egijs, E., Ka a i a, L., Saajanaho, M., Ran alainen, T., &
Ran anen, T. (2019). Assessing physical pe o mance and
physical ac i i y in la ge popula ion-based aging s udies:
Home-based assessmen s o isi s o he esea ch cen e ? BMC
Public Heal h,19(1), 1570.
Po egijs, E., Ran akokko, M., Mikkola, T. M., Viljanen, A., &
Ran anen, T. (2014). Associa ion be ween physical pe o mance
and sense o au onomy in ou doo ac i i ies and li e-space
8Jou nal o Aging and Heal h 0(0)