Designing Virtual Coaching Solutions

Schlie e , Hannes e al.
A icle — Published Ve sion
Designing Vi ual Coaching Solu ions
Business & In o ma ion Sys ems Enginee ing
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: Schlie e , Hannes e al. (2024) : Designing Vi ual Coaching Solu ions, Business
& In o ma ion Sys ems Enginee ing, ISSN 1867-0202, Sp inge Fachmedien Wiesbaden GmbH,
Wiesbaden, Vol. 66, Iss. 3, pp. 377-400,
h ps://doi.o g/10.1007/s12599-024-00871-w
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/315769
S anda d-Nu zungsbedingungen:
Die Dokumen e au EconS o dü en zu eigenen wissenscha lichen
Zwecken und zum P i a geb auch gespeiche und kopie we den.
Sie dü en die Dokumen e nich ü ö en liche ode komme zielle
Zwecke e iel äl igen, ö en lich auss ellen, ö en lich zugänglich
machen, e eiben ode ande wei ig nu zen.
So e n die Ve asse die Dokumen e un e Open-Con en -Lizenzen
(insbesonde e CC-Lizenzen) zu Ve ügung ges ell haben soll en,
gel en abweichend on diesen Nu zungsbedingungen die in de do
genann en Lizenz gewäh en Nu zungs ech e.
Te ms o use:
Documen s in EconS o may be sa ed and copied o you pe sonal
and schola ly pu poses.
You a e no o copy documen s o public o comme cial pu poses, o
exhibi he documen s publicly, o make hem publicly a ailable on he
in e ne , o o dis ibu e o o he wise use he documen s in public.
I he documen s ha e been made a ailable unde an Open Con en
Licence (especially C ea i e Commons Licences), you may exe cise
u he usage igh s as speci ied in he indica ed licence.
h p://c ea i ecommons.o g/licenses/by/4.0/
RESEARCH PAPER
Designing Vi ual Coaching Solu ions
Design and E alua ion o a Digi al Heal h In e en ion o Rehabili a ion
Hannes Schlie e •Kai Gand •Thu e Geo g Weimann •Emanuel Sandne •
Ka l K eine •S e en Thoma •Jin Liu •Massimo Cap ino •Massimo Co bo •
Agnese Se egni •Peppino T opea •Rocio Del Pino •Juan Ca los Go
´mez Es eban •
Inigo Gabilondo •And eea Elena Lac a u •S e an Sebas ian Busna u
Recei ed: 30 June 2023 / Accep ed: 15 Ap il 2024 / Published online: 29 May 2024
The Au ho (s) 2024
Abs ac Especially olde pe sons a e p one o disabili ies
and ch onic diseases. These ch onic condi ions pose a
wo ldwide challenge, leading o de e io a ing heal h, eco-
nomic s ain, loss o li e, and a decline in he quali y o li e
(QoL). The e o e, heal hca e ins i u ions seek o enhance
hei s a egies o disease p e en ion and managemen o
uphold he well-being o he communi y. This leads o he
need o egain independence and imp o e QoL o p ope ly
ehabili a e he pa ien s. Vi ual Coaches (VCs) in he o m
o Embodied Con e sa ional Agen s a e seen as a ele an
digi al in e en ion o suppo he con inui y o ca e. The
pape a hand epo s on a Design Science Resea ch p ojec
abou implemen ing a VC solu ion o suppo olde
pa ien s’ home ehabili a ion. The s udy unde pins ou
pi o al design p inciples: Adap i i y, Coaching S a egy,
Mul i-use In e ace, and Sus ainable In as uc u e. The
inal a i ac was es ed wi h 80 pa ien s which we e sup-
po ed in con inuing hei inpa ien ehabili a ion a home
by using a VC. The e alua ion shows bo h posi i e esul s
o usabili y and accep ance o he in e en ion o ou
di e en use cases and a posi i e impac on he QoL. Gi en
he comp ehensi e clinical e alua ion, he sys em ep e-
sen s a sa e and appealing solu ion o ensu ing he con i-
nui y o medical ehabili a ion ca e and he access o
pe sonalized cogni i e and mo o unc ion ea men s.
Keywo ds Vi ual coaching Digi al heal h 
Rehabili a ion ca e Design science
1 In oduc ion
Non-communicable diseases (NCDs), including ch onic
condi ions like hea disease, cance , and ch onic kidney
disease, pose a majo public heal h issue. These diseases,
which a e no ansmi ed om pe son o pe son, lead o
signi ican economic s ain, p ema u e mo ali y, and a
educ ion in he quali y o li e (QoL) o indi iduals
a ec ed (Lopez e al. 2014). Olde adul s a e he mos
a ec ed in e ms o heal h, and hey a e also he g oup wi h
a g ea e need o suppo in ehabili a ion (De Biase e al.
2020). Recen ly, also he COVID-19 pandemic has
e ealed sho comings in heal hca e and he need o
implemen solu ions o coping wi h diseases om a dis-
ance o a home (Faghe azzi e al. 2020). In his sense,
pe suasi e sys ems may con ibu e o imp o ing he
Accep ed a e wo e isions by he edi o s o he Special Issue.
H. Schlie e K. Gand (&)T. G. Weimann
Resea ch G oup Digi al Heal h, Facul y o Business and
Economics, Technische Uni e si a
¨ D esden, Helmhol zs asse
10, 01069 D esden, Ge many
e-mail: [email p o ec ed]
E. Sandne K. K eine
Digi al Heal h In o ma ion Sys ems, Cen e o Heal h &
Bio esou ces, AIT Aus ian Ins i u e o Technology, 8020 G az,
Aus ia
S. Thoma J. Liu
In o ma ion P ocess Enginee ing, FZI Fo schungszen um
In o ma ik, 76131 Ka ls uhe, Ge many
M. Cap ino M. Co bo A. Se egni P. T opea
Depa men o Neu o ehabili a ion Sciences, Casa Di Cu a Igea,
20144 Milano, I aly
R. Del Pino J. C. Go
´mez Es eban I. Gabilondo
Neu odegene a i e Diseases G oup, Bioc uces Bizkaia Heal h
Resea ch Ins i u e, 48903 Ba akaldo, Spain
A. E. Lac a u S. S. Busna u
Depa men o Ca diology, Uni e si y o Medicine and
Pha macy ‘‘Ca ol Da ila’’, 050474 Bucha es , Romania
123
Bus In Sys Eng 66(3):377–400 (2024)
h ps://doi.o g/10.1007/s12599-024-00871-w
si ua ion. Pe suasi e sys ems a e designed and s udied in
In o ma ion Sys ems Resea ch (ISR) o in o m and suppo
pa ien s, change undesi able beha io s, o guide decision-
making (Cha e jee and P ice 2009; O ji and Mo a 2018).
Despi e he po en ial bene i s hese sys ems o e o he
elde ly, hei o en less ad anced digi al skills migh hinde
accep ance, highligh ing he need o ailo ed solu ions o
enhance hei accessibili y (Mannheim e al. 2019).
Also, on he policy le el, he use o digi al solu ions is
seen as an impo an means o disease p e en ion and
con ol. The e is s ong e idence ha p ope ehabili a ion
helps o egain pa ien s’ independence and QoL (Cli o d
2011; Be hge e al. 2014; Siegel and Do ne 2017). Despi e
i s impo ance, a common issue is ha he con inua ion o
ehabili a ion is o en dis up ed a e hospi al discha ge.
The ansi ion o pa ien s’ homes is agg a a ed, which has
a nega i e impac on adhe ence o ea men plans and he
long- e m success o he ea men . To add ess his gap,
Vi ual Coaches (VCs) a e being ecognized as p omising
in e en ions in he ealm o digi al heal h. They ha e he
po en ial o suppo he con inui y o ca e and allow o he
pa ial eplacemen o complemen o adi ional in-pe son
coaching, esp. o ehabili a ion a home (Ding e al. 2010;
Ky iazakos e al. 2020; G aßmann and Sche muly 2021;
Weimann e al. 2022). In pa icula , hey could mo i a e
he pa ien o adhe e o physical ac i i ies, se incen i es o
be ac i e, moni o i al signs, suppo heal h li e acy
(Ky iazakos e al. 2020), and be a companion o us ed
guidance (Allen e al. 2008). The e o e, VCs mus ade-
qua ely e lec bo h he needs o he pa ien s and he ele-
an clinical (p ocess) knowledge in e ms o clinical
pa hways (Gand e al. 2021b).
In ecen yea s, he in e ace o VCs has equen ly
aken he o m o so-called con e sa ional agen s (CAs)
ha a e au onomous sys ems designed o esemble a human
communica ion pa ne (Seege e al. 2021). Embodied
con e sa ional agen s (ECAs) ep esen a special ype o
CAs, cha ac e ized by compu e -gene a ed and o en
human-like a a a s in e ac ing wi h he use (Cu is e al.
2021). Gi en hei capaci y o also con ey non- e bal cues,
se e al s udies sugges hei po en ial sui abili y o heal h
coaching pu poses by enhancing use engagemen , building
a us ul and empa hic ela ionship, and p omo ing sys em
accessibili y (T opea e al. 2019; K ame e al. 2020; Jiang
e al. 2024). Howe e , u he explo a ions beyond he
cu en s a e-o -a a e equi ed in he ollowing a eas:
•Use o CAs in eal-wo ld ca e scena ios: E idence o
(E)CAs has o en been ga he ed in labo a o y se ings
wi h a c oss-sec ional cha ac e and heal hy subjec s
(e.g., e S al e al. (2020); Schachne e al. (2020)).
Real-wo ld aspec s like he long- e m use o ECAs in a
clinical se ing a e possibly o en neglec ed due o
complica ed and cos ly p ocesses (e.g., e hical
app o al, da a p o ec ion o medical de ice egula ions,
pa ien ec ui men , and pa ien aining). This is
especially ele an when in eg a ing senso de ices
eeding he VC, using pe sonaliza ion algo i hms, o
dealing wi h olde /impai ed use s o use s who lack
digi al skills.
•Design knowledge and impac o VCs: The e is only
sca ce e idence o he design o VCs and hei
ad an ageousness in he medical domain (T opea
e al. 2019; Bin Sawad e al. 2022). This includes he
absence o explici ly o mula ed and alida ed design
knowledge in e ms o me a- equi emen s, design
p inciples, and design ea u es, which a e a he hea
o con empo a y ISR (S ohmann e al. 2023). De ailed,
sys ema ically s uc u ed and e alua ed design knowl-
edge is c ucial o unde s anding and eplica ing VCs in
p ac ice. Wi hou igo ous clinical me hodology
applied in p ac i ione - esea che in e en ions, ISR
may s uggle o p oduce uly i - o -pu pose solu ions
(Baske ille e al. 2023).
•In es iga ion o i ual coaching pla o ms: To da e,
VCs ha e been p ima ily in es iga ed and de eloped
o speci ic disease scena ios (K ame e al. 2020; Ca
e al. 2020; Jiang e al. 2024). The de elopmen o
holis ic coaching pla o ms applicable o a ange o
medical indica ions, including he demons a ion o he
echnical easibili y and a igo ous e alua ion o
mul iple disease con ex s, has been a la gely unad-
d essed ield in he li e a u e so a . Such pla o ms,
wi h a scalable a chi ec u e, could po en ially ans o m
heal hca e deli e y by p o iding ailo ed and e icien
suppo o a as , di e se, and mo bid popula ion
(Fu
¨ s enau e al. 2023).
The aim o ad ancing knowledge on he design and use
o VCs o home ehabili a ion he e o e leads us o he
ollowing esea ch ques ions (RQs):
RQ1: Wha a e he design p inciples o implemen ing
o VC-based home ehabili a ion sys ems o suppo he
con inui y o ca e?
RQ2: Which design p inciples could be adop ed o
designing compa able sys ems?
RQ3: To wha ex en do VC-based ehabili a ion sys-
ems in luence ehabili a ion posi i ely conce ning use
expe ience and QoL?
The p esen s udy epo s on a comp ehensi e esea ch
p ojec ha buil and e alua ed a VC o suppo ing he
ehabili a ion o olde pa ien s a home. Me hodologically,
he p ojec ollowed a Design Science Resea ch (DSR)
app oach, embedding h ee build and e alua e cycles
(Kuechle and Vaishna i 2008; Sonnenbe g and om
B ocke 2012). The p ojec has aken he design p o o ype
123
378 H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024)
in o pilo es ing o add ess he abo e sho comings o VC
s udies. The e o e, in ol ing di e en p o essions like
clinicians and so wa e enginee s p o ided access o di -
e en s akeholde s, pa icula ly pa ien s, which a e essen-
ial o he design and e alua ion.
Fo p ac ice, he s udy con ibu es wi h he design a i-
ac i sel , including he equi emen s se , echnical speci-
ica ion, and he lea ning om h ee build-and-e alua e
cycles. Fo heo y, we jus i y a se o alida ed me a- e-
qui emen s, design p inciples, and design ea u es ha
could in o m u u e esea che s when designing compa a-
ble in e en ions.
The emainde o he pape is s uc u ed as ollows: In
chap e 2, we ou line he backg ound and heo e ical
ounda ions e lec ing he Compu e s-A e-Social-Ac o s
Pa adigm and he Social Response Theo y. In chap e 3,
he me hod is p esen ed. In chap e 4.1, we de i e he
ini ial design p inciples o VCs in ehabili a ion con ex s
based on li e a u e and summa ize he a i ac design.
Chap e 4.2 consolida es he e alua ion cycles. We con-
clude wi h implica ions o heo y and p ac ice, he limi-
a ions, and an ou look o u u e esea ch.
2 Theo y and Backg ound
2.1 Vi ual Coaches and Rela ed Sys em Classes
Due o he no el y o VCs, he e has long been a lack o
e minological cla i y in he li e a u e conce ning how hey
ela e o o he concep s such as i ual assis an s o CAs.
Fu he mo e, a a ie y o synonymous e ms exis in he
li e a u e (e.g., ‘‘e-coach’’ o ‘‘ i ual aine ’’; T opea e al.
2019). A holis ic and in eg a i e iew o VCs and ela ed
sys em classes was ecen ly gi en by Weimann e al.
(2022). In his wo k, VCs we e de ined as adap i e so -
wa e sys ems ha a ge goal-di ec ed ans o ma ions o
he coachee’s (i.e., pa ien ’s) cogni ion, a ec ion, and
beha io o achie e imp o emen s on he indi idual and
socie al le el. Vi ual assis an s, in con as , aim o sim-
pli y he use ’s li e and a he unc ion as se an s. O e all,
VCs could suppo human skill de elopmen in many
domains (Cha e jee e al. 2021). Rega ding heal hca e,
VCs ha e been p e iously s udied o ain physicians o
nu sing s a (Cook e al. 2010; Richa ds e al. 2020) bu
e en mo e equen ly as pa ien -suppo ing applica ions o
p omo e physical, men al, and/o social heal h by
ins uc ing exe cises, impa ing disease- ela ed knowledge,
and se ing as a long- e m mo i a ional companion (T o-
pea e al. 2019). The li e a u e e iew by T opea e al.
u he highligh s se e al examples o VCs wo king wi h
heal hy indi iduals (e.g., physical ac i i y p omo ion o
p e en obesi y), bu poin s ou a no able de iciency in hei
use o elde ly pa ien s’ ehabili a ion.
F om a concep ual iew, bo h he human coach and
coachee can a y in hei digi al p esence on a con inuum.
A one end o his con inuum, he ul ima e o m o he
coach’s digi al e olu ion can be ega ded as he eplace-
men by an au onomous sys em (Kool e al. 2013, p. 27).
Ye , his classi ica ion app oach ends o blu he dis inc-
ions be ween how ca e is deli e ed (in-pe son, emo ely,
and/o ia s andalone au onomous sys ems, e.g., Tuckson
e al. (2017)) wi h he di e ing le els o he coach and
coachee’s digi al en ichmen . Ins ead, ano he pe spec i e
posi s ha all h ee modes o ca e deli e y can be com-
plemen ed by he widesp ead adop ion o digi al ech-
nologies (e.g., using senso s, decision suppo sys ems, o
i ual eali y ools du ing in-pe son coaching; Philpo
e al. (2023)). While emo e and s andalone deli e y modes
can be summa ized as VC in a b oade sense, p io li e -
a u e a gues ha VCs go beyond me e compu e -media ed
communica ion ools by being a leas semi-au onomous
so wa e agen s (Weimann e al. 2022). Fu he , VCs
emphasize he con e sa ional cha ac e and con ex o he
coaching p ocess. This allows o mo e in elligen beha io
by adap ing he coaching ac ions depending on he use ’s
cu en con ex (Kampho s 2017). Gi en hei con e sa-
ional cha ac e , i seems unsu p ising ha cu en VCs a e
associa ed wi h CAs o he in e ace. Howe e , ecen
li e a u e e iews on ECAs as heal h coaches highligh a
lack o con ex -based pe sonaliza ion and o e all e idence,
unde sco ing he need o u he ad ancemen s and clini-
cal ials (Lyzwinski e al. 2023; Jiang e al. 2024).
Gi en he socio- echnical na u e o VCs, di e en pe -
spec i es a e o in e es when e alua ing he VC in medical
ca e. These could be assessing he VC’s clinical e i-
cacy and e ec i eness by conside ing psychological,
beha io al, and physiological ou come pa ame e s (e.g.,
QoL, daily s eps, blood p essu e), usabili y o use -expe-
ience me ics, and he e alua ion o he gene al echnol-
ogy implemen a ion along wi h economic measu es (Cu is
e al. 2021; Jiang e al. 2024). Gi en he long- e m use o
he VC applica ion, ensu ing high use expe ience coupled
wi h app oaches o p omo e adhe ence in a a ge ed manne
(e.g., se ious games o gami ica ion) is c ucial o achie -
ing clinical imp o emen s and plays a signi ican ole in he
design p ocess o VCs (McCallum 2012; Oinas-Kukkonen
2013).
2.2 Vi ual Coaches as Social Ac o s
When CAs a e used o he on -end o he VC, design
aspec s such as he a a a ’s isual appea ance, oice, o
ges u es may igge emo ional esponses (e.g., enjoymen )
and hus (posi i ely) impac use expe ience (Lo eys e al.
123
H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024) 379
2020; e S al e al. 2020). The gene al idea o delibe a ely
including such social cues (e.g., gi ing he VC a name,
selec ing pa icula clo hing) in he sys em design is
empi ically g ounded in he so-called ‘‘Compu e s-A e-
Social-Ac o s’’ Pa adigm and he Social Response Theo y
based on i (Nass and Moon 2000). Acco ding o his
undamen al heo y, compu e s can in luence cogni ion,
a ec ion, and beha io simila o humans. Conside ing he
high media ichness o ECAs, a a ie y o di e en social
cues can be implemen ed ha may inc ease he pe suasi e
cha ac e is ics o he VC and he likeliness ha he use
ollows he VC’s ad ice (Feine e al. 2019). Howe e ,
because o his la ge design space, he design o ECAs is
mo e challenging han ha o simple disembodied cha bo s.
Consequen ly, an un a o able design may lead o mis-
ma ches o social cues ha nega i ely a ec use expe i-
ence. The so-called heo y o ‘‘uncanny alley’’ (Mo i e al.
2012) om obo ic esea ch pos ula es ha inc easing he
humanness o he VC can inc ease accep ance o a ce ain
le el. S ill, he accep ance apidly d ops when he e a e
impe ec ions be ween human-like appea ance and he
sys em’s ac ual beha io . Consequen ly, using ECAs as
heal h coaches can be bene icial o pe suading he use o
adop a speci ic beha io bu may lead o ad e se e ec s
wi h an un a o able design (Venning e al. 2021).
2.3 G ounding he Design o Vi ual Coaches
in Beha io Change Theo y
P omo ing beha io al changes o imp o e he disease a-
jec o y o p e en diseases can be conside ed he p ima y
in en ion o coaching in medical ca e (Olsen and Nesbi
2010; Passmo e and Lai 2020). The e o e, e idence om
beha io al medicine, social psychology, and beha io al IS
esea ch p o ides a ich design knowledge base o VCs.
Mainly, wo heo e ical amewo ks a e e e enced in his
ega d and ha e been used in he pas o design digi al
heal h in e en ions (Wang e al. 2019): The Beha io
Change Technique Taxonomy by Ab aham and Michie
(2008) and Michie e al. (2013) ( om he ield o beha -
io al medicine), and he Pe suasi e Sys em Design Model
by Oinas-Kukkonen and Ha jumaa (2009) (o igina ed om
ISR). The Pe suasi e Sys em Design Model is, due o i s
o igin, one s ep close o he so wa e de elopmen pe -
spec i e by desc ibing conc e e design p inciples along
wi h exempla y equi emen s and implemen a ions. Bo h
amewo ks a e g ounded on mul iple es ablished psycho-
logical heo ies, such as goal-se ing heo y (Locke and
La ham 1990), heo y o planned beha io and i s p ede-
cesso (Ajzen 1991), o he in o ma ion-mo i a ion-be-
ha io al skills model (Fishe and Fishe 1992). Howe e ,
bo h amewo ks a y in hei a angemen and numbe o
p oposed echniques, espec i ely, design p inciples. The
Pe suasi e Sys em Design Model classi ies 28 design
p inciples ela ed o p ima y ask suppo , dialog suppo ,
sys em c edibili y suppo , and social suppo . In con as ,
he Beha io Change Technique Taxonomy p oposes 93
echniques di ided in o 16 g oups (e.g., goals and planning,
eedback and moni o ing, social suppo ). Despi e some
simila i ies (e.g., sel -moni o ing, social suppo ), he
amewo ks complemen each o he and, he e o e, se e as
a holis ic undamen o designing i ual coaching in e -
en ions (Asbjø nsen e al. 2022). Howe e , o he bes o
ou knowledge, he amewo ks ha e so a no been used
o implemen and es a VC in he home ehabili a ion
domain o mul iple disease con ex s.
3 Me hod
3.1 Case Desc ip ion
The s udy was pa o he EU- unded Ca e p ojec .
1
This
p o ided access o ou clinical e e ence si es and hei
s akeholde s, as well as o esou ces o implemen ing and
e alua ing a VC solu ion ega ding echnical easibili y,
usabili y, and e ec i eness in a eal-wo ld se ing. Fol-
lowing he DSR pa adigm, he p ojec me hodology
included h ee p ima y join clinical- echnical de elopmen
phases (Ky iazakos e al. 2020) co esponding o h ee
dis inc build and e alua e cycles (Kuechle and Vaishna i
2008; Sonnenbe g and om B ocke 2012). The de elop-
men eam included 12 in e disciplina y membe s wi h
clinical, echnical, and exploi a ion backg ounds. In mid-
2017, he de elopmen o he VC solu ion s a ed; he inal
a i ac was eleased wi hin he hi d phase in mid-2022
(see also Fig. 2below o a p ocedu al o e iew). E hical
app o al we e g an ed o he s udy si es.
The p ojec ocused on home ehabili a ion o wo
neu ological (s oke and Pa kinson’s disease) and wo
ca diological (hea ailu e and ischemic hea disease)
pa hologies. Clinical si es in I aly (Lomba dy), Spain
(Basque Coun y) and Romania (Bucha es egion) ha e
been in ol ed. The clinical aim was o suppo pa ien s o
con inue hei inpa ien ehabili a ion a home using a VC.
In sum, 80 pa ien s we e in ol ed. The use s ob ained a
holis ic VC solu ion ha uses se ious games (games ha go
beyond me e en e ainmen , he e o be e heal h condi-
ions, and use elemen s like s i ing o new high sco es o
achie e he goals) in on o a came a wi h body ecog-
ni ion o mo o ic exe cises, o able -based games o
cogni i e exe cises. The VC p o ides mo i a ional eed-
back, con ols he o e all ehabili a ion p ocedu e, shows
1
See he p ojec websi e o de ails (las accessed: 04–03-2024):
h ps:// ca e-p ojec .eu/.
123
380 H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024)

exe cise esul s and i al signs o e iews, and pa ly
compensa es o he absence o human ca egi e s. A ea-
soning componen allows he VC o adap o he condi ion
and p e e ences o he pa ien s and, hus, allows a pe -
sonaliza ion o he ehabili a ion p ocedu es eplica ing
clinical p ocess knowledge and human adap i i y (e.g.,
adap ing physical exe cises o hea a e excesses). Va ious
senso s (e.g., o s ep coun , hea a e, o posi ion in he
home) a e in eg a ed and used as he da a basis and o
decisions o in e ence mechanisms. Clinical p ocedu al
knowledge is ep esen ed and e e ed o o o e all con ol
and scheduling o all coaching measu es in e ms o clin-
ical pa hways as concep ual p ocedu al models (Gand e al.
2021b).
3.2 Resea ch App oach
DSR has e ol ed as a undamen al p esc ip i e pa adigm
in ISR, con ibu ing o heo y and p ac ice o sol e eal-
wo ld p oblems ( om B ocke e al. 2020). Mo eo e , DSR
p o ides dis inc guidance o help o mula e esea ch aims
and s uc u e wo k (He ne e al. 2004; G ego and He ne
2013; om B ocke e al. 2020).
The pape ou lines a high-le el design app oach ha
makes use o a ious design science me hods. Based on he
p oblem ( he ele ance ou lined by he clinical expe s in
wo kshops and in e iews; uppe le pa in Fig. 1), a
p oblem-diminishing solu ion is sugges ed. The knowledge
base in e ms o exis ing li e a u e is he p ima y sou ce o
e idence o he ini ial de i a ion o he design (uppe igh
pa in Fig. 1; He ne 2007)). Wi h hese inpu s, he (me a-)
equi emen s o add essing he p oblem a e de i ed, guid-
ing he de elopmen o he a i ac and eeding i s e alua ion
(uppe middle pa in Fig. 1; Pe e s e al. 2007; Kuechle
and Vaishna i 2008).
Also, he expe s ac i ely pa icipa ed in all echnical
de elopmen s and always accompanied he equi emen s
(as a as con en - ela ed clinical and no pu ely echnical
aspec s we e conce ned; uppe le pa in Fig. 1). This hen
ed u he execu ions o he (design) cycles. In his sense,
he p ojec is a mode o con e gence be ween he p oblem
space ( he need o home ehabili a ion) and he solu ion
space (pe sonalized i ual coaching) ha no only leads o
his con e gence i sel , bu also con ibu es o he unde -
s anding o how his con e gence akes place (solu ion
design knowledge o k-knowledge acco ding o D echsle
and He ne (2018)).
A e heo e ically de i ing he design p inciples o
VCs in ehabili a ion con ex s in gene al ( ocal le el 2 o
DSR con ibu ion ypes in Fig. 1), we de eloped an
ins an ia ion du ing he p ojec , which we hen e alua ed in
a h ee old es se ing (le el 1 in Fig. 1). The es ing
phases we e c ucial in de e mining whe he he VC
solu ion could achie e he a ge ed clinical ou comes and
in con i ming he e ec i eness o he applied p esc ip i e
knowledge.
The p ojec me hodology included a di ision in o h ee
p ima y de elopmen phases (Ky iazakos e al. 2020) ha
co espond o h ee dis inc design cycles ((Kuechle and
Vaishna i 2008; Sonnenbe g and om B ocke 2012); see
Fig. 2). O e all, a mixed me hod app oach was used o
ecei e eedback on he p oposed solu ion (ex-an e), o
p oo s o he echnical p o o ype (ex-pos ). Wi h each sub-
cycle, he ocus shi ed om echnical easibili y o
usabili y and clinical ou comes (Ky iazakos e al. 2020):
(1) Tech Lab (DSR Cycle 1): The echnical easibili y o
he solu ion in e ms o a o ma i e e alua ion was
assessed. Based on he clinical needs, he echnical
p ojec pa ne s de eloped a p o o ype o es in a
labo a o y se ing, examining he sys em in eg a ion.
Also, he clinicians simula ed he pa ien s o con i m
he p o o ype’s compliance wi h he clinical equi e-
men s and he ex e nal alidi y by assessing i s
compa ibili y wi h he needs o he clinical si es. In
e ms o an a i icial e alua ion (Venable e al.
2012), he i o he ac ual implemen a ion wi h he
de ined echnical ( unc ional and non- unc ional)
equi emen s was in es iga ed.
(2) Li ing Lab (DSR Cycle 2): A con olled en i on-
men in e ms o he clinical acili ies was used o
assess he clinical easibili y o he VC solu ion. The
in e ac ion wi h he a i ac was pe o med by he
pa ien s, bu no in a eal-li e en i onmen in o de o
be e con ol and assess he e ec s o and on his
in e ac ion. Wi h he echnical easibili y es ab-
lished, now he pe cep ion by he use s was in
p ima y ocus. Also, some mino echnical issues
and use expe ience laws we e de ec ed du ing he
ac ual usage o he solu ion.
(3) Pilo Phase (DSR Cycle 3): A e assessing bo h he
echnical and clinical easibili y o he VC solu ion,
i s e ec on he clinical ou comes had o be inally
assessed. This was done by using s anda d clinical
scales and quan i a i e assessmen s (like o QoL) o
e alua e he e ec o he solu ion on he pa ien s.
Finally, he use pe cep ion, mo i a ion, and accep-
ance o he echnology was assessed. Ten pa ien s
pe pa hology we e included in a con ol g oup, and
ano he en pa ien s in a es g oup
Fo each o he h ee phases, we e alua ed he app o-
p ia eness o he design gi en he espec i e equi emen s
(see he e alua ion esul s in Sec . 4.2, he espec i e lab
adap a ion o e iew in Appendix A (a ailable online ia
h p://link.sp inge .com), and he inal equi emen se as
deduced in Sec . 4.1.1). Figu e 3o e s an o e all
123
H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024) 381
o e iew o he e alua ion app oach. I is wi h in en ion
ha all aspec s a e co e ed in e ms o a holis ic e alua ion
concep , including dis inc pa icipa o y design ools
(Ha s e al. 2021).
The high deg ee o in e ac ion be ween heal h p o es-
sionals and pa ien s (especially in he Li ing Lab) also
poses a isk o he eliabili y o he e alua ion esul s. In
pa icula , con i ma ion bias can lead o dis o ion (Nick-
e son 1998). To coun e ac his, a s udy p o ocol wi h
Fig. 1 Cyclic DSR app oach
(uppe pa ; (He ne 2007)) and
espec i e DSR con ibu ion
ypes (lowe pa ; building upon
G ego and He ne (2013);
D echsle and He ne (2018))
Fig. 2 O e iew o he design phases ( op: imeline/phases o he p ojec ; le : gene ic Design Science Resea ch Cycle acco ding o Kuechle
and Vaishna i (2008); igh : pa ly eusing he Build and E alua e phases acco ding o Sonnenbe g and om B ocke (2012))
123
382 H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024)
p ecisely de ined quali y c i e ia and scales o he indi-
idual s eps was de eloped in ad ance o all h ee s udy
phases (pa allel o he de i a ion o he equi emen s). This
had o be s ic ly adhe ed o. Thus, bias could a leas be
educed, al hough pe son- ela ed biases may be di icul o
ully exclude. Fu he mo e, he app oach o se e al pilo
loca ions was no only chosen o ensu e pa hological as
well as geog aphical/cul u al di e si y. Ra he , con adic-
o y esul s may also ha e been unco e ed in o e ly
he e ogeneous se ings and majo bias in one o he loca-
ions. The di e en loca ions a e, he e o e, also hei own
co ec ions in e ms o mu ual lea ning o ailu e
p e en ion.
The esul s o he li e a u e e iew (knowledge base in
Fig. 1, op igh ) a e p esen ed in chap e 4.1. A pa ici-
pa o y design app oach (Ha s e al. 2021) was chosen o
he ele ance pa (Fig. 1, op le ) ha simila ly con-
ibu ed o he de i a ion o he equi emen s o he design
o he VC solu ion. The esul s o he pa icipa o y design
app oach can be ound in he e alua ion o each phase (see
chap e 4.2 o he o e all e alua ion esul s and Appendix
A o he adap ions pe es phase). The basic p ocedu e o
he pa icipa o y design app oach and he sub-s eps pe
phase a e summa ized in Fig. 4. Mo e de ails a e gi en in
Appendix C. Pa icula ly o he i s pa , we ex ensi ely
ga he ed use iewpoin s. Also, he basic shape o he VC
solu ion was based on he needs and espec i e discussions
wi h he clinicians. These ini ial equi emen s we e e ined
and amended in discussions wi h he echnical p ojec
pa ne s, he pa icipa o y design elemen s and also aking
in o accoun he li e a u e, esul ing in he inal se o
equi emen s as discussed in Sec . 4.1.1. Fo example, he
o iginal equi emen s analysis en isaged moni o ing use s’
i al signs and ac i i y con inuously. This was in ended o
ensu e an op imal da abase and he bes possible basis o
clinical decision-making. Howe e , his was changed as
pa o he pa icipa o y design. C i icism was oiced by
bo h pa ien s and expe s in he Tech Lab (see Appendix
C). Fo example, use s said ha he e should also be
unobse ed imes (‘‘s eal h mode’’). Pa ien s’ au onomy
should no be unnecessa ily es ic ed. Expe s also
exp essed conce ns abou da a p o ec ion ega ding ull-
ime moni o ing. Fo his eason, an easy- o-ac i a e
s andby mode was de eloped wi hin he VC solu ion, in
which he a a a does no ‘‘lis en’’ o is inac i e as a as
he echnical moni o ing componen s a e conce ned. Ul i-
ma ely, he inal e sion o he equi emen is e lec ed
below in DP1 o DF1-5, 7, 9, 11, 13–19, which can be
ac i ely es ic ed by he use h ough he s andby mode.
The es esul s pe phase a e discussed in Sec . 4.2. The
gene alizable lea nings o he o e all design p ocess a e
summa ized below. We also p o ide mic o-lea nings om
each build-e alua e cycle in e ms o pa icula design
adap ions (see Sec . 4.2 and Appendix A).
4 Resul s
4.1 A i ac Design
In he ollowing, we epo on he de i ed me a equi e-
men s (MRs), design p inciples (DPs), design ea u es
(DFs), and hei echnical ins an ia ion in he inal design
a i ac . The comple e se o DFs can be ound in Appendix
D.
Fig. 3 O e iew o he e alua ion du ing he di e en phases (on he le ), con as ed o he choice wi hin he DSR E alua ion F amewo k by
Venable e al. (2012) on he igh ( he e ical spanning o he phases on he le and he e alua ion ypes on he igh is ma ched)
123
H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024) 383
The design knowledge is he esul o he h ee DSR
cycles illus a ed in Fig. 2. Based on he ecommenda ions
by B adbu y e al. (2014), we combined an analysis o he
li e a u e and a pa icipa o y design in ol ing bo h clinical
domain expe s and pa ien s (i.e., deduc i e and induc i e).
This way, bo h he DSR en i onmen and knowledge base
con ibu ed o he inal design (see Fig. 5, le ). The
deduc i ely de i ed MRs we e subs an ia ed and alida ed
h ough empi ical s udies wi h clinical domain expe s and
pa ien s. The li e a u e e iew ocused on pape s concep-
ualizing he domain o VCs and ela ed concep s (esp.
CAs, pe sonaliza ion) as well as pape s on he applica ion
o beha io change echniques in he medical ields ele-
an o ou p ojec (Pa kinson’s disease, s oke, hea
ailu e, ischemic hea disease). Se e al li e a u e e iews
we e conduc ed as pa o he Ca e p ojec (see T opea
e al. (2019), Philipp e al. (2019), and Weimann e al.
(2022)), which se ed o iden i y essen ial concep s in he
domain o VCs and key publica ions. The indings o hese
e iews also se ed as guidance o unsys ema ically sea ch
o u he li e a u e e iews o su eys add essing speci ic
aspec s in mo e de ail (e.g., beha io change echniques).
Likewise, dedica ed pape s eme ged wi hin he con ex o
he p ojec ha discussed some equi emen s and hei
implemen a ion aspec s in mo e de ail (esp. Benedic e al.
2019; K op e al. 2020; Gand e al. 2021a,b). Conse-
quen ly, he o e all li e a u e sea ch p ocess can be
ega ded as semi-sys ema ic. Fo de i ing he design
p inciples (DPs) based on he elabo a ed me a- equi e-
men s, we ollowed Mo
¨lle e al. (2020) and G ego e al.
(2020). Addi ionally, he se o DFs was collec ed based on
hese s udies o speci y he solu ion space (see Fig. 5).
4.1.1 Me a-Requi emen s and Design P inciples
Se e al p e ious esea ch e o s ha e signi ican ly con-
ibu ed o unde s anding he ‘‘ana omy’’ o VCs. Fo
example, Kampho s (2017) pos ula ed se e al equi e-
men s o VCs: social abili ies, c edibili y, con ex awa e-
ness, ailo ing and lea ning abili ies, he abili y o in e ace
Fig. 4 Pa icipa o y Design App oach (Ha s e al. 2021) as pa o he DSR en i onmen /Rele ance cycle (He ne 2007)
Fig. 5 Mapping p inciple o MRs, DPs and DFs (based on S ohmann e al. (2023))
123
384 H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024)
Table 2). Mos pa ien s epo ed g adually inc easing
con idence wi h he echnology in subsequen sessions,
enabling hem o use i independen ly a home (e.g., see
quo es (1), (2), (6) in Table 2). In only a ew ins ances,
addi ional ins uc ion and suppo we e deemed necessa y;
pa ien s would ha e app ecia ed use manuals o b ie
demons a ion ideos as eminde s on how o ope a e he
sys em (e.g., see quo es (2), (7) in Table 2). In some cases,
pa ien s exp essed a desi e o a mo e imme si e gaming
en i onmen and highe complexi y le els in he games o
encou age mo e cons an ac i i y (e.g., see quo es (10),
(12) in Table 2).
The echnical componen s as such ha e al eady been
p o ed in he p eceding Tech Lab phase. Only some
op imiza ions in he un ime beha io (i e a i e imp o e-
men s o he app) o addi ions o he scope o unc ions
we e necessa y. Thus, he ocus mo ed mo e o a p oces-
sual o pa hway- ela ed le el (see Gand e al. (2021a,b) o
mo e de ails). Also, he e we e new equi emen s o he
ehabili a ion ca e pa hways ha esul ed om discussions
wi h clinicians du ing eal sys em ope a ions. This is
because he i s se o pa hway empla es was de i ed
p ima ily om pape wo k wi hou ha ing an ac ual wo k-
ing sys em. These changes a e summa ized in Appendix A,
Table A.2. As he changes ega ding he pa hway con en
and me hodology we e limi ed, a well-conduc ed ini ial
equi emen s analysis can be assumed.
4.2.3 Pilo Phase – DSR Cycle 3 – Focus on Clinical
Ou comes
The Pilo Phase was conduc ed subsequen o sys em
e inemen , inco po a ing eedback ecei ed om use s
du ing ea lie s ages. The p ima y echnical wo k was
abou ins alling he needed de ices a he pa ien s’ homes,
main aining he de ices, and aining he pa ien s on how o
use hem. Also, wi hin he P o essional Po al, o e iews
o he measu ed i al pa ame e s and ques ionnai e esul s
om dis an pa ien s ha e been added. Some clinical
pa hways had again o be e ined, and he machine lea ning
algo i hms had been ini ia ed/ ed wi h i s pa ien da a.
The Pilo Phase was planned as a small-scale pilo
andomized ial (Cocks and To ge son 2013), whe ein he
de eloped sys em unde goes clinical alida ion wi hin he
use ’s home en i onmen o e alua e he VC’s impac . The
pa icipan s ha e been ec ui ed om he pool o pa ien s
who ha e been ea ed a he clinical si es and who mee
some clinical inclusion c i e ia. Ten pa icipan s pe
pa hology ollowed a pe sonalized ehabili a ion p og am
a home o a pe iod o up o h ee mon hs. Du ing he
s udy, he pa icipan s used he VC app on a able ha sen
eminde s o p esc ibed exe cises ( o physical and cog-
ni i e ac i a ion and ehabili a ion), allowed moni o ing o
i al and mo emen pa ame e s, and p o ided hin s and
e-lea ning ma e ial ia he VC’s a a a in e ace ( ollowing
he p ecep s as se by he pa hway empla es). Also, a se
o conduc ing physical and cogni i e se ious games a
home (in on o a sc een, wi h a 3D dep h came a o
mo ion sensing) was ins alled o hem. Fo each clinical
si e, a u he en pa ien s we e en olled as a con ol g oup
( ecei ing con en ional ehabili a ion, i.e., clinical ecom-
menda ions a discha ge; see Table 3 o an o e iew o
he pa icipan s).
I is acknowledged ha he inal ou comes sligh ly di e
based on he pa hology and speci ic indica o s assessed.
None heless, o e all, he goal o es o ing ac i e and
independen li ing a home, in e ms o QoL, has been
achie ed h ough imp o emen s in con inuous ca e and
access o pe sonalized cogni i e ( o neu ological cases),
mo o exe cises ( o all pa ien g oups), and comp ehensi e
coaching ad ice p o ided by he VC. The EQ5D scale
4
has
been used as a s anda d ool o compa e QoL be o e and
a e singula /sho e in e en ions (see Table 4), gi en he
ela i ely small sample size o he Pilo s udy (Gandhi e al.
2017). The a ge imp o emen o 10% o he QoL alues
s ems om he clinical expe s as an empi ical alue o be
conside ed e y good and compe i i e compa ed o o he
in e en ions. I is, he e o e, a consolida ed expe
assessmen .
Figu e 11 p o ides a compa ison o he summa i e QoL
scales p e and pos -in e en ion. E.g., he mobili y sco e
o he s oke case could ha e been dec eased om an
a e age o 2,0 o 1,7 (p e- o pos -in e en ion), leading o
an imp o emen o abou 17%. O e all, he e is a clea
endency owa d imp o emen , gi en ha he scale is seen
as a i s indica o o he ad an ageousness and usabili y o
he VC solu ion.
Figu e 12 depic s he EQ5D isual analog scale (VAS)
a ings assessing he pe cei ed heal h s a us o pa ien s in
he VC-guided in e en ion (expe imen al) and he adi-
ional in e en ion (con ol). Gi en he small sample size
bu simila ea men needs and ehabili a ion p og ams, we
decided o pool he da a (neu ological and ca diological
pa ien s) o analyzing he o e all changes. The esul s
sugges ha he VC-guided in e en ion was associa ed
wi h a leas compa able bu e en mo e posi i e e ec s
ega ding he pe cei ed heal h s a us, pa icula ly o VC-
guided ca diological p og ams (see pe cen age imp o e-
men compa ed o T0).
Posi i e ou comes we e also obse ed in educing isk
ac o s, including a signi ican inc ease in daily s eps and
ime de o ed o exe cises o e-lea ning sessions. The
adhe ence o he ehabili a ion plan ega ding access o he
4
See o de ails and a demo he e (las accessed: 04–03-2024): h ps://
eu oqol.o g/eq-5d-ins umen s/.
123
H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024) 391

pla o m, sys em in e ac ions, o he o al numbe o imes
pa ien s pe o med a sugges ed ac i i y was in line wi h he
expec a ions. Posi i e indica ions we e also no ed con-
ce ning he iabili y o he an icipa ed pe sonaliza ion and
heal h p omo ion e o s. Howe e , he olume o da a
collec ed was somewha less han expec ed which in u n
limi ed he (au oma ed) e inemen s o he ehabili a ion
he apy. Despi e his, he VC managed o adjus he pa h-
ways in ce ain ins ances.
The use expe ience a ings we e mainly posi i e o he
hedonic dimensions (s imula ion and no el y) ac oss he
ou pa hologies (see Fig. 13 o an o e iew). In con as ,
a ings o he p agma ic dimensions (pe spicui y, e i-
ciency, and dependabili y) we e a he neu al o e en
Table 3 O e iew o he Pilo Phase pa icipan s
S oke Pa kinson disease Hea ailu e Ischemic hea disease
Expe imen al
g oup
7 male, 3 emale, age:
72.50 ±9.00,
usage ime o he solu ion
(days): 40.43 ±14.18
7 male, 3 emale, age
64.50 ±7.90,
usage ime o he solu ion
(days): 106.40 ±36.88
6 male, 4 emale, age:
60.81 ±10.87,
usage ime o he solu ion
(days): 37.10 ±34.16
10 male, 0 emale, age:
58.10 ±7.10,
usage ime o he solu ion
(days): 32.90 ±32.84
Con ol g oup 7 male, 3 emale, age:
72.20 ±11.80
7 male, 3 emale, age:
69.10 ±3.50
6 male, 4 emale, age:
59.60 ±7.99
7 male, 3 emale, age:
58.00 ±7.21
Table 4 Imp o emen s in QoL p e-pos in e en ion, sub-domains o he EQ5D QoL scale, p ojec ambi ion was an imp o emen by a leas
10%
S oke Pa kinson disease (PD) Hea ailu e (HF) Ischemic hea disease (IHD)
4Mobili y = 17%
8Sel -ca e = 0%
8Pain/discom o = 8%
8Daily ac i i ies = 0%
4Anxie y/ Dep ession = 25%
4Mobili y = 54%
4Sel -ca e = 60%
4Pain/discom o = 67%
4Daily ac i i ies = 57%
4Anxie y/ Dep ession = 50%
4Mobili y = 10%
4Sel -ca e = 17%
4Pain/discom o = 30%
4Daily ac i i ies = 18%
4Anxie y/ Dep ession = 23%
4Mobili y = 28%
4Sel -ca e = 23%
4Pain/discom o = 20%
4Daily ac i i ies = 20%
4Anxie y/ Dep ession = 28%
Fig. 11 Compa ison o he EQ5D QoL scales p e (T0) and pos (T1) in e en ion as a basic indica o o he ad an ageousness o he VC
solu ion; The sco es ange om 0 (no p oblems) o 4 (unable o walk, no being able o pe o m daily ac i i ies, e c.)
123
392 H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024)
nega i e, pa icula ly o he ca diological p og ams. One
explana ion o his esul may be he mo e complex se ing
in he ca diological p og ams (addi ional blood p essu e
and weigh scale de ices; mo e delays due o he pan-
demic). Pa ien s conside ed ha he echnical aspec s in
e ms o moni o ing de ices could be imp o ed by e ining
he sys em and elimina ing di icul ies hey encoun e ed.
The e o equi ed o echnical adjus men s and a educed
a ini y o he echnology o olde pa ien s may, he e o e,
ha e been unde es ima ed o some ex en . Howe e , he
esul s sugges ha , pa icula ly o neu ological pa ien s,
he use expe ience was sa is ac o y, and no such ad e se
e en s occu ed.
When aking he esul s o he SUS e alua ion in o
accoun , he accep abili y h eshold was me o all
pa hologies wi h he excep ion o pa ien s wi h ischemic
hea disease (see Fig. 14). The e alua ion o he pe cei ed
use ulness and ease o use, p oposed as key ac o s o
echnology accep ance (see Da is (1989)), u he
unde sco es his inding (see Fig. B.2 in Appendix B). I
can, he e o e, be concluded ha he de eloped VC ep-
esen s a sa e, engaging, and aes he ically pleasing sys em
o ensu ing he con inui y o medical ehabili a ion ca e in
he pa ien ’s home en i onmen . Thus, he design cycle can
be ega ded as success ul, suppo ing he b oade applica-
bili y o he elabo a ed DPs.
De ailed insigh s in o he medical e icacy o he VC
wi h a ocus on he sepa a e medical use cases can be ound
in Del Pino e al. (2023), Se egni e al. (2022), Busna u
e al. (2022) and La
˘c a u e al. (2023).
5 Discussion
In con as o he p io s udies in es iga ing single DPs in
an a i icial expe imen se ing, we ha e ocused on hei
in eg a ion in o a ull- ea u ed VC sys em in a na u al
en i onmen . We we e able o in es iga e he impac o a
Fig. 12 Changes in he EQ5D-VAS Scale (Pe cei ed Heal h S a us) o he expe imen al g oup and he con ol g oup o he ca diological and
neu ological ehabili a ion p og am
Fig. 13 Use Expe ience Ques ionnai e (UEQ) a ings o all pa ien g oups a e pilo phase; UEQ a ings abo e 0.8 a e conside ed as posi i e
e alua ion
123
H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024) 393
VC om bo h he clinical and human–compu e in e ac ion
pe spec i es (see Fig. 15 o an o e iew o he main
indings). E.g., he mobili y sco e o he s oke case could
ha e been dec eased om an a e age o 2,0 o 1,7 (p e- o
pos -in e en ion), leading o an imp o emen o abou
17%; pa ien s gene ally had posi i e use expe iences wi h
he VC sys em, pa icula ly as ega ds he s imula ion and
no el y (hedonic UEQ dimensions). Also, see he posi i e
use eedback as exempla ily shown in Table 2. The s udy
was able o show how he p inciples o DSR we e used in a
eal-li e con ex o in e wine MRs and DPs. The VC
solu ion is p o en in a p ac ically ele an con ex and is
bene icial in e ms o use expe ience and clinical aspec s.
Mo eo e , he indings p o ide eal-wo ld e idence o he
use ulness o DPs in he ield o VC solu ions. By e alu-
a ing clinical ou comes and use expe ience- ela ed aspec s
oge he , he s udy na ows down he cu en e alua ion
gap.
F om a clinical iewpoin , he s udy lays he ounda ions
o la ge-scale clinical ials (RCTs) o alida e he esul s
and ule ou u he con ounde s. F om an ISR iewpoin ,
he s udy delinea es i s he complexi y o he design and
implemen a ion o digi al in e en ions in heal hca e.
Second, he indings b oaden he design knowledge
ega ding VCs o home ehabili a ion and clinical esea ch
in gene al. Wi h his s udy, we p o ide ou DPs as guid-
ance and a de ailed se o DFs which help o implemen
hose p inciples. The s udy was conduc ed wi hin a di e se
na u alis ic se ing, whe e he dimensions o ele ance and
igo encoun e a subs an ially b oade ange o in luencing
ac o s o e alua ion han in a con olled, labo a o y-based
expe imen al con ex . The s udy elucida es he manne in
Fig. 14 SUS-Sco es o Li ing Lab and Pilo Phase
MR
En i onmen
Knowledge Base
Rele ance
Cycle
Design
Cycle
Kn
Rigou
Cycle
Build
MR1:
In e en ion
Adap i i y
MR2:
Coaching
S a egy
MR4:
Sus ainable
Technical
In as uc u e
MR3: Use
In e ace
DP
DP1 - Adap i i y:
adap a ion mechanisms o
enable pe sonalized
coaching and hus imp o e
engagemen and ou comes
DP2 - Coaching s a egy:
s a egy and echniques o
beha io change o le
pa ien s adap owa ds an
inc ease o QoL
DP3 - Mul i-use in e ace:
modali ies o add ess bo h
he needs o he coachee
(pa ien ) and human coach
(heal h p o essional)
DP4 - Sus ainable
in as uc u e: enable
di e en business
oppo uni ies o exploi a ion
and echnical openness
Clinical: QoL imp o emen s
Mobili y
✓ ✓
Sel -ca e
✓
Pain /
discom o
✓
Daily
ac i i ies
✓
Anxie y /
Dep ession
✓ ✓
ISR: Sys em accep ance & usabili y
TAM
✓ ✓
UEQ
✓
SUS
✓ ✓
Fig. 15 Main indings and p ima y esul s o he build-e alua e cycles (double ick = ully ul illed; single ick = mino weakness, bu o e all
ul illed)
123
394 H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024)
which comp ehensi e e alua ion and documen a ion
e o s can acili a e he inco po a ion o complex in e -
en ions in o he ISR knowledge co pus. The speci ic
implica ions o heo y and p ac ice and limi a ions a e
delinea ed in he subsequen sec ions.
Rega ding RQ1 (Wha a e he design p inciples o
implemen ing o VC-based home ehabili a ion sys ems o
suppo he con inui y o ca e?), ou s udy in oduces a
comp ehensi e amewo k consis ing o igo ously de el-
oped me a- equi emen s and de ailed sub- equi emen s.
This is achie ed by employing i e a i e build-e alua e
cycles o VCs o ehabili a ion pu poses, which is sub-
s an ia ed by posi i e ou comes du ing hei e alua ion.
The con ibu ion o ou esea ch is guided by ou pi o al
DPs: Adap i i y, Coaching S a egy, Mul i-use In e ace,
and Sus ainable In as uc u e. These p inciples guide he
c ea ion o VCs, ensu ing hey a e no only e ec i e in
ehabili a ion se ings bu also g ounded on obus heo-
e ical ounda ions. Fu he , we elabo a e on a de ailed lis
o design ea u es, illus a ing how each con ibu es o he
ealiza ion o he DPs.
In add essing RQ2 (Which design p inciples could be
adop ed o designing compa able sys ems?), ou in es i-
ga ions p o ide p esc ip i e knowledge o he de elop-
men o dis inc VC solu ions, he eby illus a ing he
usabili y and ans e abili y o design knowledge ac oss
a ious con ex s. Showing he ins an ia ion o he p e-
sc ip i e knowledge in he solu ion o he Ca e p ojec no
only acili a es he c ea ion o ailo ed VC in e en ions bu
also con ibu es o he o e a ching discou se on he u i-
liza ion o VCs in he ealm o digi al heal hca e.
Finally, he s udy p o ides insigh s in o RQ3 (To wha
ex en do VC-based ehabili a ion sys ems in luence eha-
bili a ion posi i ely conce ning use expe ience and
QoL?): Ou analysis un eils ha he deploymen o ou VC
solu ion signi ican ly enhances clinical ou comes and
exhibi s commendable esul s in e ms o use expe i-
ence (including sys em accep ance and usabili y). The
e alua ion highligh s ha pa icipan s expe ienced no e-
wo hy imp o emen s in hei QoL, subs an ia ing he po-
en ial clinical e icacy and e ec i eness o ou VC
solu ion. Following an in eg a ed e alua ion design based
on he amewo k o Venable e al. (2012), as explica ed in
Sec . 3.2, ou s udy ein o ces he clinical and p ac ical
alue o VCs in ehabili a ion. Due o i s cos ly design and
in e disciplina y cha ac e , his kind o s udy has been
a ely conduc ed so a .
5.1 Implica ions o Theo y
This s udy’s con ibu ions embody he essen ial c i e ia o
a nascen design heo y (G ego and Jones (2007); G ego
and He ne (2013); see Table 5). G ounded in a obus
heo e ical ounda ion, i ex ends exis ing knowledge wi h
igo ous heo e ical de elopmen by conduc ing an i e a-
i e p ocess o design and de elop VCs o ehabili a ion
backed by a posi i e clinical ial (Pu pose and Scope). We
desc ibed he VC sys em’s key echnical pa s and he
a ied s akeholde s in ol ed, om ca egi e s o solu ion
designe s (Cons uc s). The VC’s DPs a e de i ed om
exis ing heo ies and expe s’ knowledge, o guide medical
p o essionals in VC design and use (P inciple o Fo m and
Func ion). The VC design is lexible and adap able, hus
i ing a ious clinical scena ios and o e ing cus omiza ion
and pe sonaliza ion (A i ac Mu abili y). The VC’s e ec-
i eness can be measu ed h ough clinical ials, whe e
s udy esul s ha e shown p omise (Tes able P oposi ions).
The design buil is oo ed in co e VC design heo ies, and
in eg a es pas esea ch and p ac ical app oaches (Jus i i-
ca o y Knowledge). The esea ch p o ides a angible VC
design, which has been ope a ionalized and se in o mo ion
(P inciples o Implemen a ion). A VC o ehabili a ion has
been designed, while main aining design consis ency
(Exposi o y Ins an ia ion). Table 5p o ides mo e de ails o
he componen s.
5.2 Implica ions o P ac ice
As men ioned, he e is a lack o la ge-scale VC solu ions
ha conside he clinical ou ines and pa ien needs while
complemen ing he exis ing ace- o- ace ca e. The e o e,
he esea ch esul s p o ide a mul i ude o con ibu ions o
p ac ice. Fi s ly, i p o ides e idence ha he employed
DPs can acili a e p oduc de elopmen . We demons a ed
he gene ic applicabili y o he DPs ac oss di e se disease
con ex s, as e idenced by hei ins an ia ion in a scalable
pla o m ailo ed o he equi emen s o bo h ca diological
and neu ological pa ien s. The dis inc coaching sys em o
he p ojec could unc ion as a baseline o o he VC sys-
ems since i is buil as a modula and dis ibu ed sys em. I
also lays he concep ual ounda ion o a VC ecosys em.
Second, he DPs wi h he de ailed se o DFs allow o
he de elopmen o sub-p oduc s. Speci ically, a i s s ep
owa ds he deli e y o VC solu ions a ge ing mul iple
mo bidi ies has been aken, which is a ising issue in
heal hca e (Skou e al. 2022). Fo example, ca dio ascula
diseases can be associa ed wi h obesi y, and bo h condi-
ions equi e a specialized coaching p og am. I is c ucial
o a oid he need o pa ien s o in e ac wi h sepa a e
so wa e applica ions in o de o enhance accessibili y,
seamless use expe ience, and long- e m echnology
adop ion, as well as o a oid con adic ions o incompa ible
coaching p og ams.
Thi d, new business models in he ield o VCs could
eme ge. S anda diza ion and in e ope abili y a e he main
d i e s ha enable a VC ecosys em whe e di e en
123
H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024) 395
endo s ha e a ‘‘co-ope i ion’’ o he bes solu ion, se -
ices, and business models (Eisenmann e al. 2008).
Fou h and inally, a VC solu ion may change ole
models and modali ies o ehabili a ion and ela ed in o -
ma ion sys ems. P essu e could be aken o ca egi e s
wi hou eplacing hem. Ra he , i supplemen s he emo e
physician, o example, by gi ing he apy ecommenda-
ions and cons an ly collec ing i al pa ame e s. Wi h AI,
mo e ad anced pe sonaliza ion and adap a ion mechanisms
a e likely o be ealis ic soon (Weimann and Gißke 2024).
Consequen ly, ou s udy pa es he way o addi ional
in es iga ions in o he use o AI o cus omize heal h- ela ed
cues and aining, he eby enhancing he o e all e ec o
he in e en ion. Ne e heless, while he DPs unde sco e
he impo ance o in eg a ing pe sonaliza ion ( ia AI)
wi hin he comp ehensi e solu ion, hey do no u nish
de ailed speci ica ions o he dis inc AI componen s.
5.3 Limi a ions and Oppo uni ies o Fu u e Resea ch
The selec ion o pa hologies is one o he limi a ions o he
cu en s udy. The design decision and adap a ions (see
Appendix A) migh ha e been di e en i applied o, e.g., a
younge popula ion o people wi h s onge cogni i e
impai men s. We ied o mi iga e his e ec by including
di e en pa hologies wi h high social impac in di e en
coun ies.
A u he design- ela ed limi a ion is he choice o
echnology. Since he s udy las ed o e a pe iod o i e
yea s, we canno ule ou ha al e na i e se ices could
ha e shown a be e pe o mance. Howe e , o inc ease he
Table 5 Design heo y componen s (G ego and Jones 2007; G ego and He ne 2013)
Componen Desc ip ion
Pu pose and scope VCs o home ehabili a ion inc ease clinical e ec i eness and gain high use expe ience sco es. We p opose ou
DPs and es ed hem in an i e a i e e alua ion p ocess
Cons uc s •P ima y ( echnical) componen s o he VC sys em (chap e 4.1.2) o conside he con ex and allow pe sonaliza ion
(knowledge laye , senso laye ), linkage and o ches a e o sub-elemen s (middlewa e laye ), go e n and o e iew
he ehabili a ion p ocess (pa hway laye ), se he dis inc coaching ac i i ies (coaching laye ), display he VC
solu ion and link i o o he sys ems (UI/exploi a ion laye )
•S akeholde s using he VC solu ion (ca egi e s, pa ien s, ela i es, and solu ion designe s)
•Resea ch needs, design objec i es, a i ac equi emen s: chap e 4.1
P inciple o o m and
unc ion
On he basis o exis ing li e a u e on heo e ical VC ounda ions and expe s’ knowledge, we de i ed ou DPs aiming
o design o implemen VC solu ions in medical se ings. These DPs ha e been es ed in h ee design-e alua e cycles:
DP1—Adap i i y: adap a ion mechanisms o enable pe sonalized coaching and hus imp o e engagemen and
ou comes
DP2—Coaching s a egy: s a egy and echniques o beha io change o enhance QoL
DP3—Mul i-use in e ace: modali ies o add ess bo h he needs o he coachee (pa ien ) and human coach (heal h
p o essional)
DP4—Sus ainable in as uc u e: enable di e en business oppo uni ies o exploi a ion and echnical openness
A i ac mu abili y The a i ac is highly mu able as changes in he a i ac mani es a ion a e pe se included and an icipa ed (see DP1),
shown by i s success ul usage in di e en clinical con ex s/pa hologies using di e en pa hway con igu a ions. Inne
mu abili y is included, gi en he pe sonaliza ion capabili ies and he se ice-o ien ed a chi ec u e app oach
Tes able p oposi ions Posi i e ou comes (p ima y goal: enhance pa ien ou comes in e ms o QoL and use expe ience) om he clinical
ials using he VC solu ion ep esen a leas p elimina y u h s a emen s o he p oposed DPs and o e all app oach
Mo e la ge-scale ials would hen inally e alua e whe he he p oposed DPs success ully mee he o mula ed me a-
equi emen s
The p ima y ou comes o e alua ion include (chap e 4.2):
•Clinical/QoL: EQ5D
•Use expe ience: UEQ (hedonic and p agma ic dimensions), SUS, TAM
Jus i ica o y knowledge The co e cons uc s and heo ies ha in o m he design o VCs a e syn hesized in chap e 2. The jus i ica o y
knowledge is mainly g ounded on he Compu e s-A e-Social-Ac o s Pa adigm, he Social Response Theo y, he
Beha io Change Technique Taxonomy and he Pe suasi e Sys em Design Model
P inciples o implemen -
a ion
We p o ide an example o he DPs and o e all design app oach o VC in home ehabili a ion. I was ins an ia ed as a
dis inc coaching sys em in he p ojec . This can se e as a baseline o o he VC sys ems
Exposi o y ins an ia ion We ha e designed he VC ins an ia ion using he gi en p esc ip i e knowledge. The a i ac ’s consis ency wi h he
design p ecep s has been shown by equi emen s ul ilmen in e ms o he posi i ely e alua ed clinical ials.
Sec ion 4.1.2 ou lines his alignmen
123
396 H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024)

s abili y and compa abili y o he e alua ion, he main
componen s had o pe sis o e he p ojec phases a e
inal app o al.
Limi a ions in he e alua ion phase also esul om
choosing speci ic scales o he clinical accep abili y o he
a i ac . While his does no imply ha al e na i e scales
a e inhe en ly mo e meaning ul o alid, he eason o he
selec ion o scales was in luenced by he a ailabili y in he
equi ed languages (essen ial o di ec use wi h olde
pa ien s and o an in e na ional pe spec i e) and he
exis ing knowledge o clinical esea che s.
Fu he mo e, he e alua ion was pa ly designed like a
classical RCT (wi h in e en ion and con ol g oup) wi h a
a he explo a o y in en ion han a con i ma o y scope
(explo a o y design), which esul ed in a limi ed numbe o
pa icipan s. As pa o his explo a ion, we we e able o
conduc a easibili y s udy in a clinical-scien i ic se ing
wi h eal pa ien s, p o iding esul s on accep ance and
e icacy. This small-scale s udy in a na u alis ic se ing
builds he ounda ion o de elop a la ge-scale RCT wi h a
mo e medical ocus on he con i ma ion o he al eady
gained esul s.
Fu u e esea ch should also add ess cu en ends, such
as digi al ubiqui y o gene a i e AI (esp. la ge language
models) (Feue iegel e al. 2024). This would po en ially
esul in be e and non-in asi e con ex awa eness and,
o e all, in a be e pe sonaliza ion o coaching
in e en ions.
Finally, esea ch should also add ess he ques ion o a
mul i-VC app oach a he han a single-VC app oach (i.e.,
mul iple a a a s; Beinema e al. (2021)). Depending on he
applica ion scena io, a specialized coach (di e en da a-
bases, di e en echnical equipmen ) could hen appea and
in e ac wi h he pa ien (e.g., o physical ac i i y, nu i-
ion, social heal h e c.).
6 Summa y & Ou look
Ou in es iga ion del es in o he u iliza ion o VCs as a
specialized o m o ECAs wi hin a clinical se ing o
enhance he con inui y o ca e, esp. in home ehabili a ion.
The p ima y aim o his s udy was o b oaden ou unde -
s anding ega ding he deploymen o clinical VCs, pa -
icula ly in add essing he e iden sca ci y o
comp ehensi e in e en ions, e alua ing he o e a ching
in luence o VCs, and ec i ying he sho all o e idence
ha is deeply oo ed in p ac ical applica ion.
Fu he esea ch should con ibu e o he de elopmen o
home-based ehabili a ion h ough VC solu ions, wi h a
pa icula ocus on pe sonaliza ion and pe sonaliza ion
algo i hms aiming o imp o e he adap a ion quali y o
ehabili a ion p og ams o he indi idual needs and
condi ions o pa ien s. Ensu ing pa ien sa e y wi hin hese
pe sonalized in e en ions eme ges as a c i ical esea ch
di ec i e, necessi a ing he de elopmen o obus ame-
wo ks ha moni o and mi iga e po en ial isks associa ed
wi h au oma ic adap a ions.
The design and implemen a ion o a VC ecosys em also
aises a numbe o esea ch ques ions. This includes he
ques ion o how o in eg a e low-/no-code capabili ies o
acili a e a mo e accessible and adap able en i onmen o
VC-based ea men p og ams, as well as he ques ion
ega ding sui able business models o he di e en s ake-
holde s in he ecosys em.
Fu he mo e, la ge clinical ials (RCTs) should be
included in u he esea ch o ex end he s a is ical e i-
dence. In addi ion o adi ional expe imen al and con ol
g oups o a s a is ically alid p oo o e ec , i would be
desi able o o m u he subg oups. This could p o ide
insigh s in o he bene i s o speci ic DPs and DFs.
All in all, ou wo k con ibu es o he e idence o VC
solu ions in he clinical domain and o he esea ch on
pe suasi e sys ems (Cha e jee and P ice 2009). We aspi e
o ad ance owa d a new e a cha ac e ized by he seamless
in eg a ion o digi al heal h solu ions.
Supplemen a y In o ma ion The online e sion con ains
supplemen a y ma e ial a ailable a h ps://doi.o g/10.1007/s12599-
024-00871-w.
Acknowledgemen s We wan o hank all he p ojec eam membe s
o p o iding hei expe ise and o hei comple e dedica ion
h oughou he p ojec ’s li e ime. This p ojec has ecei ed unding
om he Eu opean Union’s Ho izon 2020 esea ch and inno a ion
p og am unde g an ag eemen No 769807.
Funding Open Access unding enabled and o ganized by P ojek
DEAL.
Open Access This a icle is licensed unde a C ea i e Commons
A ibu ion 4.0 In e na ional License, which pe mi s use, sha ing,
adap a ion, dis ibu ion and ep oduc ion in any medium o o ma , as
long as you gi e app op ia e c edi o he o iginal au ho (s) and he
sou ce, p o ide a link o he C ea i e Commons licence, and indica e
i changes we e made. The images o o he hi d pa y ma e ial in his
a icle a e included in he a icle’s C ea i e Commons licence, unless
indica ed o he wise in a c edi line o he ma e ial. I ma e ial is no
included in he a icle’s C ea i e Commons licence and you in ended
use is no pe mi ed by s a u o y egula ion o exceeds he pe mi ed
use, you will need o ob ain pe mission di ec ly om he copy igh
holde . To iew a copy o his licence, isi h p://c ea i ecommons.
o g/licenses/by/4.0/.
Re e ences
Ab aham C, Michie S (2008) A axonomy o beha io change
echniques used in in e en ions. Heal h Psychol 27:379–387.
h ps://doi.o g/10.1037/0278-6133.27.3.379
Ajzen I (1991) The heo y o planned beha io . O gan Beha Hum
Decis P ocess 50:179–211
123
H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024) 397
Allen M, Iezzoni LI, Huang A e al (2008) Imp o ing pa ien -
clinician communica ion abou ch onic condi ions: desc ip ion
o an in e ne -based nu se E-coach in e en ion. Nu s Res
57:107–112. h ps://doi.o g/10.1097/01.NNR.0000313478.
47379.98
Asbjø nsen RA, Hjelmesæ h J, Smeds ød ML e al (2022) Combining
pe suasi e sys em design p inciples and beha io change
echniques in digi al in e en ions suppo ing long- e m weigh
loss main enance. JMIR Hum Fac o s 9:e37372. h ps://doi.o g/
10.2196/37372
Bango A, Ko um P, Mille J (2009) De e mining wha indi idual
SUS sco es mean: adding an adjec i e a ing scale. J Usabili y
S ud 4:114–123
Baske ille R, om B ocke J, Ma hiassen L, Scheepe s H (2023)
Clinical esea ch om in o ma ion sys ems p ac ice. Eu J In
Sys 32:1–9. h ps://doi.o g/10.1080/0960085X.2022.2126030
Beinema T, op den Akke H, an Velsen L, He mens H, (2021)
Tailo ing coaching s a egies o use s’ mo i a ion in a mul i-
agen heal h coaching applica ion. Compu Hum Beha
121:106787. h ps://doi.o g/10.1016/j.chb.2021.106787
Benedic M, Schlie e H, Gißke C (2019) Managemen Digi ale
O
¨kosys eme mi de DREEM-Me hode am Beispiel eine
Vi ual-Coaching-Pla o m. HMD P ax Wi sch 56:857–875.
h ps://doi.o g/10.1365/s40702-019-00538-1
Be hge M, on G oo e P, Gius ini A, Gu enb unne C (2014) The
wo ld epo on disabili y: a challenge o ehabili a ion
medicine. Am J Phys Med Rehabil 93:S4-11. h ps://doi.o g/
10.1097/PHM.0000000000000016
Bin Sawad A, Na ayan B, Alne aie A e al (2022) A sys ema ic
e iew on heal hca e a i icial in elligen con e sa ional agen s
o ch onic condi ions. Senso s 22:2625. h ps://doi.o g/10.3390/
s22072625
Boecking L, Philipp P (2020) Recommending sa e ac ions by lea ning
om sub-op imal demons a ions. In: P oceedings o he 5 h
In e na ional Wo kshop on Heal h Recommende Sys ems co-
loca ed wi h 14 h ACM Con e ence on Recommende Sys ems
(Heal hRecSys’20). h ps://ceu -ws.o g/Vol-2684/6-pagina ed.
pd
B adbu y K, Wa s S, A den-Close E e al (2014) De eloping digi al
in e en ions: a me hodological guide. E id Based Complemen
Al e na Med 2014:1–7. h ps://doi.o g/10.1155/2014/561320
B ooke J (1996) SUS: A ‘‘quick and di y’’ usabili y scale. In: Jo dan
PW e al (eds) Usabili y e alua ion in indus y. CRC P ess,
London, pp 189–194
Busna u S, Pana MA, Lac a u AE e al (2022) Pa ien pe cep ion
when ansi ioning om classic o emo e assis ed ca diac
ehabili a ion. Diagnos ics 12(4):926. h ps://doi.o g/10.3390/
diagnos ics12040926
Ca LT, Dhinaga an DA, Kyaw BM e al (2020) Con e sa ional
agen s in heal h ca e: scoping e iew and concep ual analysis.
J Med In e ne Res 22:e17158. h ps://doi.o g/10.2196/17158
Cha e jee S, P ice A (2009) Heal hy li ing wi h pe suasi e
echnologies: amewo k, issues, and challenges. J Am Med
In o m Assoc JAMIA 16:171–178. h ps://doi.o g/10.1197/
jamia.M2859
Cha e jee A, Ge des M, P inz A, Ma inez S (2021) Human coaching
me hodologies o au oma ic elec onic coaching (ecoaching) as
beha io al in e en ions wi h in o ma ion and communica ion
echnology: sys ema ic e iew. J Med In e ne Res 23:e23533.
h ps://doi.o g/10.2196/23533
Cheng C, Beauchamp A, Elswo h GR, Osbo ne RH (2020) Applying
he elec onic heal h li e acy lens: sys ema ic e iew o elec-
onic heal h in e en ions a ge ed a socially disad an aged
g oups. J Med In e ne Res 22:e18476. h ps://doi.o g/10.2196/
18476
Ci illo F, Solmaz G, Be z EL e al (2019) A s anda d-based open
sou ce io pla o m: FIWARE. IEEE In e ne Things Mag
2:12–18. h ps://doi.o g/10.1109/IOTM.0001.1800022
Cli o d J (2011) The UN disabili y con en ion and i s impac on
Eu opean equali y law. The Equal Righ s Re 6(11):15
Cocks K, To ge son DJ (2013) Sample size calcula ions o pilo
andomized ials: a con idence in e al app oach. J Clin Epi-
demiol 66:197–201. h ps://doi.o g/10.1016/j.jclinepi.2012.09.
002
Cook DA, E win PJ, T iola MM (2010) Compu e ized i ual pa ien s
in heal h p o essions educa ion: a sys ema ic e iew and me a-
analysis. Acad Med 85:1589–1602. h ps://doi.o g/10.1097/
ACM.0b013e3181ed e13
Cu is RG, Ba el B, Fe guson T e al (2021) Imp o ing use
expe ience o i ual heal h assis an s: scoping e iew. J Med
In e ne Res 23:e31737. h ps://doi.o g/10.2196/31737
Da is FD (1989) Pe cei ed use ulness, pe cei ed ease o use, and
use accep ance o in o ma ion echnology. MIS Qua
1(13):319–340
De Biase S, Cook L, Skel on DA e al (2020) The COVID-19
ehabili a ion pandemic. Age Ageing 49:696–700. h ps://doi.
o g/10.1093/ageing/a aa118
Del Pino R, De Eche a ı
´a AO, Dı
´ez-Ci a da M e al (2023) Vi ual
coach and ele ehabili a ion o Pa kinsons disease pa ien s:
Ca e sys em. J Public Heal h. h ps://doi.o g/10.1007/s10389-
023-02082-1
De e ding S, Dixon D, Khaled R, Nacke L (2011) F om game design
elemen s o game ulness: de ining ‘‘gami ica ion.’’ In: P oceed-
ings o he 15 h In e na ional Academic MindT ek Con e ence -
MindT ek ’11. ACM P ess, Tampe e, p 9
Di Ruscio D, Kolo os D, de La a J e al (2022) Low-code
de elopmen and model-d i en enginee ing: wo sides o he
same coin? So w Sys Model 21:437–446. h ps://doi.o g/10.
1007/s10270-021-00970-2
Ding D, Liu H-Y, Coope R e al (2010) Vi ual coach echnology o
suppo ing sel -ca e. Phys Med Rehabil Clin 21:179–194. h ps://
doi.o g/10.1016/j.pm .2009.07.012
D echsle A, He ne AR (2018) U ilizing, p oducing, and con ibu -
ing design knowledge in DSR p ojec s. In: Cha e jee S e al
(eds) Designing o a digi al and globalized wo ld. Sp inge ,
Cham, pp 82–97
Eisenmann TR, Pa ke G, Van Als yne MW (2008) Opening
pla o ms: how, when and why? Social Science Resea ch
Ne wo k, Roches e
Faghe azzi G, Goe zinge C, Rashid MA e al (2020) Digi al heal h
s a egies o igh COVID-19 wo ldwide: challenges, ecom-
menda ions, and a call o pape s. J Med In e ne Res 22:e19284.
h ps://doi.o g/10.2196/19284
Feine J, Gnewuch U, Mo ana S, Maedche A (2019) A axonomy o
social cues o con e sa ional agen s. In J Hum-Compu S ud
132:138–161. h ps://doi.o g/10.1016/j.ijhcs.2019.07.009
Felsbe g DT, Mahe JP, Rhea CK (2019) The s a e o beha io change
echniques in i ual eali y ehabili a ion o neu ologic popu-
la ions. F on Psychol 10:979. h ps://doi.o g/10.3389/ psyg.
2019.00979
Feue iegel S, Ha mann J, Janiesch C, Zschech P (2024) Gene a i e
AI. Bus In . Sys Eng 66:111–126. h ps://doi.o g/10.1007/
s12599-023-00834-7
Fishe JD, Fishe WA (1992) Changing AIDS- isk beha io . Psychol
Bull 111:455–474. h ps://doi.o g/10.1037/0033-2909.111.3.455
Fu
¨ s enau D, Ge sch M, Sch ei e S (2023) Digi al he apeu ics
(DTx). Bus In Sys Eng. h ps://doi.o g/10.1007/s12599-023-
00804-z
Gand K, Bo
¨cking L, K eine K, e al (2021a) Technically ep esen ing
clinical knowledge o ehabili a ion ca e. In: P oceedings o
MIE 2021
123
398 H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024)
Gand K, S a k J, Schlie e H, e al (2021b) Using clinical pa hways o
i ual coach pa ien s o home ehabili a ion. In: ICIS 2021
P oceedings
Gandhi M, Xu Y, Luo N, Cheung YB (2017) Sample size
de e mina ion o EQ-5D-5L alue se s udies. Qual Li e Res
In J Qual Li e Asp T ea Ca e Rehabil 26:3365–3376. h ps://
doi.o g/10.1007/s11136-017-1685-3
G aßmann C, Sche muly CC (2021) Coaching wi h a i icial in elli-
gence: concep s and capabili ies. Hum Resou De Re
20:106–126. h ps://doi.o g/10.1177/1534484320982891
G ego S, He ne AR (2013) Posi ioning and p esen ing design
science esea ch o maximum impac . MIS Q 37:337-A6
G ego S, Jones D (2007) The ana omy o a design heo y. J Assoc In
Sys 8:312–335
G ego S, Chand a K use L, Seidel S (2020) Resea ch pe spec i es:
he ana omy o a design p inciple. J Assoc In Sys 21:2
Ha s L, Wollschlaege B, Bi ns ein J e al (2021) E alua ion is key:
p o iding app op ia e e alua ion measu es o pa icipa o y and
use -cen ed design p ocesses o heal hca e IT. In J In eg Ca e
21:24. h ps://doi.o g/10.5334/ijic.5529
Hennessy EA, Johnson BT, Acabchuk RL e al (2020) Sel - egula ion
mechanisms in heal h beha io change: a sys ema ic me a-
e iew o me a-analyses, 2006–2017. Heal h Psychol Re
14:6–42. h ps://doi.o g/10.1080/17437199.2019.1679654
He ne AR (2007) A h ee cycle iew o design science esea ch.
Scand J In Sys 19:87–92
He ne AR, Ma ch ST, Pa k J, Ram S (2004) Design science in
in o ma ion sys ems esea ch. Manag In Sys Q 28:75–106
Howes SC, Cha les DK, Ma ley J e al (2017) Gaming o heal h:
sys ema ic e iew and me a-analysis o he physical and
cogni i e e ec s o ac i e compu e gaming in olde adul s.
Phys The 97:1122–1137. h ps://doi.o g/10.1093/p j/pzx088
Huebsche MC, McCann JA (2008) A su ey o au onomic compu -
ing – deg ees, models, and applica ions. ACM Compu Su
40:1–28. h ps://doi.o g/10.1145/1380584.1380585
Jiang Z, Huang X, Wang Z e al (2024) Embodied con e sa ional
agen s o ch onic diseases: scoping e iew. J Med In e ne Res
26:e47134. h ps://doi.o g/10.2196/47134
Kampho s BA (2017) E-coaching sys ems: wha hey a e, and wha
hey a en’ . Pe s Ubiqui ous Compu 21:625–632. h ps://doi.o g/
10.1007/s00779-017-1020-6
Kool L, Timme J, an Es Q (2013) Keuzes oo de e-coach:
maa schappelijke agen bij de au oma ise ing an de coach-
ingsp ak ijk. Ra henau Ins i uu , Den Haag
K ame LL, e S al S, Mulde BC e al (2020) De eloping embodied
con e sa ional agen s o coaching people in a heal hy li es yle:
scoping e iew. J Med In e ne Res 22:e14058. h ps://doi.o g/
10.2196/14058
K a h J, Schu
¨ mann L, on Ko lesch HFO (2021) Re ealing he
heo e ical basis o gami ica ion: a sys ema ic e iew and
analysis o heo y in esea ch on gami ica ion, se ious games
and game-based lea ning. Compu Hum Beha 125:106963.
h ps://doi.o g/10.1016/j.chb.2021.106963
K op J, Hunge la
¨nde NA, Gand K, Schlie e H (2020) T ans o ming
ehabili a ion o i ually suppo ed ca e - he Ca e p ojec .
Companion Publica ion o he 2020 In e na ional Con e ence on
Mul imodal In e ac ion. ACM, Vi ual E en Ne he lands,
pp 261–264
Kuechle B, Vaishna i V (2008) On heo y de elopmen in design
science esea ch: ana omy o a esea ch p ojec . Eu J In Sys
17:489–504. h ps://doi.o g/10.1057/ejis.2008.40
Ky iazakos S, Schlie e H, Gand K e al (2020) A no el i ual
coaching sys em based on pe sonalized clinical pa hways o
ehabili a ion o olde adul s – equi emen s and implemen a ion
plan o he Ca e p ojec . F on Digi Heal h. h ps://doi.o g/10.
3389/ dg h.2020.546562
La
˘c a u A-E, S e an-S B, Pana
˘M-A e al (2023) Assessing he
e icacy o a i ual assis an in he emo e ca diac ehabili a ion
o hea ailu e and ischemic hea disease pa ien s. In J En i on
Res Public Heal h 20:3937. h ps://doi.o g/10.3390/
ije ph20053937
Laugwi z B, Held T, Sch epp M (2008) Cons uc ion and e alua ion
o a use expe ience ques ionnai e. In: Holzinge A (ed) HCI and
usabili y o educa ion and wo k. Sp inge , Heidelbe g, pp 63–76
Locke EA, La ham GP (1990) A heo y o goal se ing & ask
pe o mance. P en ice-Hall
Lopez AD, Williams TN, Le in A e al (2014) Remembe ing he
o go en non-communicable diseases. BMC Med 12:200.
h ps://doi.o g/10.1186/s12916-014-0200-8
Lo eys K, Seba a nam G, Saga M, B oadben E (2020) The e ec o
design ea u es on ela ionship quali y wi h embodied con e -
sa ional agen s: a sys ema ic e iew. In J Soc Robo
12(6):1293–312
Lyzwinski LN, Elgendi M, Menon C (2023) Con e sa ional agen s
and a a a s o ca diome abolic isk ac o s and li es yle- ela ed
beha io s: scoping e iew. JMIR MHeal h UHeal h 11:e39649.
h ps://doi.o g/10.2196/39649
Mannheim I, Schwa z E, Xi W e al (2019) Inclusion o olde adul s
in he esea ch and design o digi al echnology. In J En i on
Res Public Heal h 16:3718. h ps://doi.o g/10.3390/
ije ph16193718
McCallum S (2012) Gami ica ion and se ious games o pe sonalized
heal h. Pheal h 177:85–96
Michie S, Richa dson M, Johns on M e al (2013) The beha io
change echnique axonomy ( 1) o 93 hie a chically clus e ed
echniques. Ann Beha Med 46:81–95. h ps://doi.o g/10.1007/
s12160-013-9486-6
Mo
¨lle F, Guggenbe ge TM, O o B (2020) Towa ds a me hod o
design p inciple de elopmen in in o ma ion sys ems. In:
Ho mann S, e al (eds) Designing o digi al ans o ma ion.
Co-c ea ing se ices wi h ci izens and indus y. Sp inge , Cham,
pp 208–220
Mo i M, MacDo man K, Kageki N (2012) The uncanny alley [ om
he ield]. IEEE Robo Au om Mag 19(2):98–100. h ps://doi.o g/
10.1109/MRA.2012.2192811
Nahum-Shani I, Smi h SN, Sp ing BJ e al (2018) Jus -in- ime
adap i e in e en ions (JITAIs) in mobile heal h: key compo-
nen s and design p inciples o ongoing heal h beha io suppo .
Ann Beha Med 52:446–462
Nass C, Moon Y (2000) Machines and mindlessness: social esponses
o compu e s. J Soc Issues 56:81–103. h ps://doi.o g/10.1111/
0022-4537.00153
Nicke son RS (1998) Con i ma ion bias: a ubiqui ous phenomenon in
many guises. Re Gen Psychol 2:175–220
Ochoa SF, Gu ie ez FJ (2018) A chi ec ing e-coaching sys ems: a
i s s ep o dealing wi h hei in insic design complexi y.
Compu e 51:16–23. h ps://doi.o g/10.1109/MC.2018.1731079
Oinas-Kukkonen H (2013) A ounda ion o he s udy o beha io
change suppo sys ems. Pe s Ubiqui ous Compu 17:1223–1235.
h ps://doi.o g/10.1007/s00779-012-0591-5
Oinas-Kukkonen H, Ha jumaa M (2009) Pe suasi e sys ems design:
key issues, p ocess model, and sys em ea u es. Commun Assoc
In Sys 24:28
Olsen JM, Nesbi BJ (2010) Heal h coaching o imp o e heal hy
li es yle beha io s: an in eg a i e e iew. Am J Heal h P omo
25:e1–e12. h ps://doi.o g/10.4278/ajhp.090313-LIT-101
O ji R, Mo a K (2018) Pe suasi e echnology o heal h and
wellness: s a e-o - he-a and eme ging ends. Heal h In o ma -
ics J 24:66–91
Passmo e J, Lai Y (2020) Coaching psychology: explo ing de ini ions
and esea ch con ibu ion o p ac ice. In: Passmo e J, Tee D (eds)
Coaching Resea ched, 1s edn. Wiley, pp 3–22
123
H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024) 399
Pe e s K, Tuunanen T, Ro henbe ge MA, Cha e jee S (2007) A
design science esea ch me hodology o in o ma ion sys ems
esea ch. J Manag In Sys 24:45–77
Philipp P, Me kle N, Gand K, Gißke C (2019) Con inuous suppo o
ehabili a ion using machine lea ning. IT - In Technol
61:273–284. h ps://doi.o g/10.1515/i i -2019-0022
Philpo LM, Dugani SB, Singla A e al (2023) Digi al ca e ho izon: a
amewo k o ex ending heal h ca e h ough digi al ans o ma-
ion. Mayo Clin P oc Digi Heal h 1:210–216. h ps://doi.o g/10.
1016/j.mcpdig.2023.05.005
Richa ds JP, Done AJ, Ba be SR e al (2020) Vi ual coach: he nex
ool in unc ional endoscopic sinus su ge y educa ion. In Fo um
Alle gy Rhinol 10:97–102. h ps://doi.o g/10.1002/al .22452
Sahay A, Indamu sa A, Di Ruscio D, Pie an onio A (2020) Suppo ing
he unde s anding and compa ison o low-code de elopmen
pla o ms. In: 2020 46 h Eu omic o Con e ence on So wa e
Enginee ing and Ad anced Applica ions. IEEE, pp 171–178
Salehie M, Tah ilda i L (2009) Sel -adap i e so wa e: Landscape
and esea ch challenges. ACM T ans Au on Adap Sys 4:1–42.
h ps://doi.o g/10.1145/1516533.1516538
Schachne T, Kelle R, Wangenheim V, F, (2020) A i icial in elli-
gence-based con e sa ional agen s o ch onic condi ions: sys-
ema ic li e a u e e iew. J Med In e ne Res 22:e20701. h ps://
doi.o g/10.2196/20701
Seege A-M, P ei e J, Heinzl A (2021) Tex ing wi h humanlike
con e sa ional agen s: designing o an h opomo phism. J Assoc
In Sys 22:8
Se egni A, T icomi E, T opea P e al (2021) Vi ual coaching o
ehabili a ion: he pa icipa o y design expe ience o he Ca e
p ojec . F on Public Heal h. h ps://doi.o g/10.3389/ pubh.2021.
748307
Se egni A, T opea P, Re R e al (2022) Vi ual coaching sys em o
con inui y o ca e and ehabili a ion in pa ien s wi h s oke.
Resul s o he pilo s udy in he home scena io. Gai Pos u e
97:29–30. h ps://doi.o g/10.1016/j.gai pos .2022.09.051
Siegel C, Do ne TE (2017) In o ma ion echnologies o ac i e and
assis ed li ing –In luences o he quali y o li e o an ageing
socie y. In J Med In 100:32–45. h ps://doi.o g/10.1016/j.
ijmedin .2017.01.012
Sim I (2019) Mobile de ices and heal h. N Engl J Med 381:956–968.
h ps://doi.o g/10.1056/NEJM a1806949
Skou ST, Mai FS, Fo in M e al (2022) Mul imo bidi y Na Re Dis
P ime 8:48. h ps://doi.o g/10.1038/s41572-022-00376-4
Sonnenbe g C, om B ocke J (2012) E alua ion pa e ns o design
science esea ch a e ac s. Eu opean Design Science Sympo-
sium. Sp inge , Heidelbe g, pp 71–83
S ohmann T, Siemon D, Elshan E, Gnewuch U (2023) Design
p inciples in in o ma ion sys ems esea ch: ends in cons uc ion
and o mula ion. In: AMCIS 2023 P oceedings
e S al S, K ame LL, Tabak M e al (2020) Design ea u es o
embodied con e sa ional agen s in eheal h: a li e a u e e iew.
In J Hum-Compu S ud 138:102409. h ps://doi.o g/10.1016/j.
ijhcs.2020.102409
T opea P, Schlie e H, S e pi I e al (2019) Rehabili a ion, he g ea
absen ee o i ual coaching in medical ca e: scoping e iew.
J Med In e ne Res 21:e12805. h ps://doi.o g/10.2196/12805
Tuckson RV, Edmunds M, Hodgkins ML (2017) Teleheal h. N Engl J
Med 377:1585–1592. h ps://doi.o g/10.1056/NEJMs 1503323
Venable J, P ies-Heje J, Baske ille R (2012) A comp ehensi e
amewo k o e alua ion in design science esea ch. In: Design
science esea ch in in o ma ion sys ems. Ad ances in heo y and
p ac ice. Sp inge , pp 423–438
Venning A, He d MC, Oswald TK e al (2021) Explo ing he
accep abili y o a digi al men al heal h pla o m inco po a ing a
i ual coach: he good, he bad, and he oppo uni ies. Heal h
In o ma ics J 27:146045822199487. h ps://doi.o g/10.1177/
1460458221994873
om B ocke J, Win e R, He ne A, Maedche A (2020) Special issue
edi o ial – accumula ion and e olu ion o design knowledge in
design science esea ch: a jou ney h ough ime and space.
J Assoc In Sys . h ps://doi.o g/10.17705/1jais.00611
Wang Y, Fadhil A, Lange J-P, Rei e e H (2019) In eg a ing
axonomies in o heo y-based digi al heal h in e en ions o
beha io change: a holis ic amewo k. JMIR Res P o oc
8:e8055. h ps://doi.o g/10.2196/ esp o .8055
Weimann TG, Schlie e H, B endel AB (2022) Vi ual coaches:
backg ound, heo ies, and u u e esea ch di ec ions. Bus In
Sys Eng 64:515–528. h ps://doi.o g/10.1007/s12599-022-
00757-9
Weimann T, Gißke C (2024) Unleashing he Po en ial o Rein o ce-
men Lea ning o Pe sonalizing Beha io al T ans o ma ions
wi h Digi al The apeu ics: A Sys ema ic Li e a u e Re iew: In:
P oceedings o he 17 h In e na ional Join Con e ence on
Biomedical Enginee ing Sys ems and Technologies. SCITE-
PRESS - Science and Technology Publica ions, Rome, I aly,
pp 230–245
Win e SJ, Shea s JL, King AC (2016) The use o beha io change
echniques and heo y in echnologies o ca dio ascula disease
p e en ion and ea men in adul s. P og Ca dio asc Dis
58:605–612. h ps://doi.o g/10.1016/j.pcad.2016.02.005
Zhang C, Lakens D, IJssels eijn WA (2021) Theo y in eg a ion o
li es yle beha io change in he digi al age. J Med In e ne Res
23:e17127. h ps://doi.o g/10.2196/17127
123
400 H. Schlie e e al.: Designing Vi ual Coaching Solu ions, Bus In Sys Eng 66(3):377–400 (2024)