Assessment of health state utilities associated with adult and pediatric acid sphingomyelinase deficiency (ASMD)

Ma za, Louis S. e al.
A icle — Published Ve sion
Assessmen o heal h s a e u ili ies associa ed wi h adul
and pedia ic acid sphingomyelinase de iciency (ASMD)
The Eu opean Jou nal o Heal h Economics
P o ided in Coope a ion wi h:
Sp inge Na u e
Sugges ed Ci a ion: Ma za, Louis S. e al. (2024) : Assessmen o heal h s a e u ili ies associa ed wi h
adul and pedia ic acid sphingomyelinase de iciency (ASMD), The Eu opean Jou nal o Heal h
Economics, ISSN 1618-7601, Sp inge , Be lin, Heidelbe g, Vol. 25, Iss. 8, pp. 1437-1448,
h ps://doi.o g/10.1007/s10198-023-01667-7
This Ve sion is a ailable a :
h ps://hdl.handle.ne /10419/315210
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/
Vol.:(0123456789)
The Eu opean Jou nal o Heal h Economics (2024) 25:1437–1448
h ps://doi.o g/10.1007/s10198-023-01667-7
ORIGINAL PAPER
Assessmen o heal h s a e u ili ies associa ed wi hadul andpedia ic
acid sphingomyelinase de iciency (ASMD)
LouisS.Ma za1 · Ka ieD.S ewa 1· Ma ieFou nie 2· DonnaRowen3· RobinLachmann4· Mau izioSca pa5·
EugenMengel6· T a isObe meye 7· E enAyik8· Fe nandoLa edo9· Ru hPuliko il‑Jacob10
Recei ed: 7 June 2023 / Accep ed: 21 Decembe 2023 / Published online: 27 Feb ua y 2024
© The Au ho (s) 2024
Abs ac
In oduc ion Acid sphingomyelinase de iciency (ASMD) ype B is a a e gene ic diso de leading o enla gemen o he
spleen and li e , pulmona y dys unc ion, and o he symp oms. Cos -u ili y analyses a e o en conduc ed o quan i y he
alue o new ea men s, and hese analyses equi e heal h s a e u ili ies. The e o e, he pu pose o his s udy was o es ima e
u ili ies associa ed wi h a ying le els o se e i y o adul and pedia ic ASMD ype B.
Me hods Se en adul and se en child heal h s a e igne es desc ibing ASMD we e de eloped based on published li e a u e,
clinical ial esul s, and in e iews wi h clinicians, pa ien s wi h ASMD, and pa en s o child en wi h ASMD. The heal h
s a es we e alued in ime ade-o in e iews wi h adul gene al popula ion esponden s in he UK.
Resul s In e iews we e comple ed wi h 202 pa icipan s (50.0% emale; mean age = 41.3yea s). The heal h s a e ep e-
sen ing ASMD wi hou impai men had he highes mean u ili y o bo h he adul and child heal h s a es (0.92/0.94), and
se e e ASMD had he lowes mean u ili y (0.33/0.45). E e y child heal h s a e had a signi ican ly g ea e u ili y han he
co esponding adul heal h s a e. Di e ences be ween adul /child pai ed s a es anged om 0.02 o 0.13. Subg oup analyses
explo ed he impac o pa en ing s a us on alua ion o child heal h s a es.
Discussion G ea e se e i y o ASMD was associa ed wi h lowe mean u ili y. Resul s ha e implica ions o alua ion o pedi-
a ic heal h s a es. The esul ing u ili ies may be use ul in cos -u ili y modeling es ima ing he alue o ea men o ASMD.
Keywo ds U ili y· Acid sphingomyelinase de iciency· Ra e disease· Time ade-o · TTO· Pedia ic u ili y
Fe nando La edo was an employee o Sano i a he ime o s udy
conduc .
* Louis S. Ma za
louis.ma za@e ide a.com
1 Pa ien -Cen e ed Resea ch, E ide a, 7101 Wisconsin
A enue, Sui e 1400, Be hesda, MD20814, USA
2 Sano i, Chilly-Maza in, F ance
3 School o Heal h andRela ed Resea ch, Uni e si y
o She ield, She ield, UK
4 Uni e si y College London Hospi als, London, UK
5 Cen o Coo dinamen o Regionale Mala ie Ra e, Azienda
Sani a ia Uni e si a ia del F iuli Cen ale, Udine, I aly
6 SphinCS-Ins i u e o Clinical Science o Lysosomal S o age
Diseases, Hochheim, Ge many
7 NNPDF, Cha lo e, NC, USA
8 F esno, CA, USA
9 Sano i, SãoPaulo, B azil
10 Sano i, Thames Valley Pa k, Reading, UK
1438 L.S.Ma za e al.
JEL Classi ica ion I. Heal h, Educa ion, and Wel a e· I00 Gene al· I1 Heal h· I10 Gene al
In oduc ion
Acid sphingomyelinase de iciency (ASMD; his o ically
known as Niemann–Pick disease ypes A, A/B, and B) is
a a e gene ic diso de ha esul s in an accumula ion o
sphingomyelin. Whe eas ASMD ype A is usually diag-
nosed du ing in ancy, wi h apid p og ession o dea h du -
ing ea ly childhood [1], ASMD ype B has a la e onse and
a slowe a e o p og ession [2–6]. Commonly epo ed
signs and symp oms o ASMD ype B include enla gemen
o he spleen and li e , pulmona y dys unc ion, dyspnea,
dyslipidemia, a igue, and pain [3, 6–8]. ASMD ype B is
also associa ed wi h limi ed physical ac i i y [9], g ow h
de ici s in child en, and declines in physical unc ion, such
as walking and s anding [6, 10], as well as impac on social
ela ionships [9], psychological unc ioning [9], and gene al
heal h- ela ed quali y o li e [11].
Olipudase al a, an enzyme eplacemen he apy, is he
i s disease-modi ying he apy o non-cen al ne ous sys-
em mani es a ions o ASMD in child en and adul s [12].
Olipudase al a ea men is associa ed wi h imp o emen s in
o gan olumes, pulmona y unc ion, hema ologic pa ame-
e s, dyslipidemia, and in child en, ca ch-up g ow h [13–16].
As wi h any new ea men , cos -e ec i eness analyses may
be use ul in a ange o coun ies o examine he alue o
olipudase al a and in o m decision-making abou heal hca e
esou ce alloca ion [17–19]. Submissions o heal h echnol-
ogy assessmen (HTA) agencies ypically include a ype
o cos -e ec i eness model called a cos -u ili y analysis
(CUA). In a CUA, heal h- ela ed quali y o li e is quan i ied
in e ms o heal h s a e u ili ies, which a e alues on a scale
ancho ed o 1 ( ull heal h) and 0 (dead) ep esen ing he
s eng h o p e e ence o heal h s a es. HTA agencies o en
p e e ha CUAs inco po a e u ili ies de i ed om gene ic
p e e ence-based ins umen s such as he EQ-5D [20–22].
Fo a e diseases like ASMD, howe e , gene ic ins u-
men s comple ed by pa ien s may no be sensi i e o easi-
ble, and al e na e app oaches may be necessa y [23, 24]. To
de i e u ili ies om pa ien -comple ed ins umen s such as
he EQ-5D, a su icien ly la ge sample o pa ien s is needed
o ep esen he heal h s a es ha will be included in eco-
nomic models. Wi h a a e disease such as ASMD, i may
no be easible o ec ui a la ge enough sample o pa ien s
o ob ain a eliable u ili y es ima e o e e y heal h s a e
needed in a CUA [25, 26]. This is u he complica ed by
onse o ASMD in childhood because he e a e challenges
associa ed wi h ob aining heal h- ela ed quali y o li e da a
in child en ha a e consis en wi h adul da a [27, 28]. In
addi ion, gene ic ins umen s may no assess concep s ha
a e impo an o pa ien s wi h speci ic a e diseases like
ASMD [25]. Fo example, a ailable gene ic ins umen s lack
key aspec s o ASMD, such as de e io a ion in pulmona y
unc ion and symp oms ela ed o spleen olume, including
bleeding and b uising [6, 8, 11, 29].
When gene ic p e e ence-based ins umen s a e no
app op ia e, easible, o su icien , some HTA agencies
such as he Na ional Ins i u e o Heal h and Ca e Excel-
lence allow o he use o al e na i e me hods o es ima e
u ili ies [21]. Wi h a e diseases, igne e-based me hodol-
ogy is o en used o es ima e u ili ies o use in HTA submis-
sions [24, 30–37]. In hese s udies, heal h s a e igne es can
be d a ed based on bes a ailable e idence and alued in
p e e ence-based asks by gene al popula ion esponden s
[23, 37].
The pu pose o he cu en igne e-based s udy was o
es ima e u ili ies associa ed wi h a ying se e i y o ASMD
ype B (subsequen ly e e ed o as ASMD). Because his
condi ion ypically p esen s in childhood and pe sis s in o
adul hood [2, 8, 11, 38], CUAs o ea men s o ASMD
will equi e u ili ies associa ed wi h bo h adul and pedia ic
heal h s a es. The e o e, his s udy was designed o es ima e
u ili ies o wo pa allel se s o heal h s a es, one se o heal h
s a es desc ibing adul s wi h ASMD and ano he desc ib-
ing child en. These da a p o ide a unique oppo uni y o
examine u ili y di e ences be ween heal h s a e igne es
ep esen ing adul s and child en wi h he same disease.
Me hods
O e iew o s udy design
To de e mine whe he a igne e-based s udy would be nec-
essa y o es ima ing u ili ies associa ed wi h ASMD, analy-
ses we e conduc ed using da a om wo gene ic p e e ence-
based ins umen s (EQ-5D-5L, SF-6D) adminis e ed in a
clinical ial o ea men o ASMD [13, 15]. A baseline,
u ili ies de i ed wi h hese wo gene ic ins umen s did no
co espond o ASMD se e i y as assessed by measu emen s
o o gan olume and pulmona y unc ion, sugges ing ha
hese gene ic ins umen s we e no sensi i e o se e i y o
illness o impac on quali y o li e in pa ien s wi h ASMD.
Fu he mo e, EQ-5D-5L and SF-6D da a we e limi ed by
he ela i ely small sample size o a clinical ial conduc ed
in pa ien s wi h a a e disease.
Due o hese limi a ions, an al e na i e u ili y assessmen
app oach was necessa y [23]. The igne e-based me hod
was selec ed o his s udy because i is well-sui ed o es i-
ma ing u ili ies associa ed wi h he speci ic symp oms and
impai men obse ed in pa ien s wi h a e diseases [31–36].
1439
Assessmen o heal h s a e u ili ies associa ed wi hadul andpedia ic acid sphingomyelinase…
This app oach also o e comes he sample size limi a ions
associa ed wi h a e diseases because igne es can be al-
ued by gene al popula ion esponden s, a he han people
wi h he ele an disease.
Following ecen ly published ecommenda ions on
igne e de elopmen and alua ion [23, 37], 14 heal h
s a e igne es we e de eloped o ep esen ASMD based
on bes a ailable e idence, including inpu om clinicians,
pa ien s, and ca egi e s. U ili ies o hese heal h s a es we e
elici ed in a ime ade-o (TTO) ask wi h a 10-yea ime
ho izon. All in e iews we e conduc ed in pe son by ained
in e iewe s wi h adul gene al popula ion esponden s in
Ma ch 2022 in London, England. All pa icipan s p o ided
in o med consen be o e comple ing he s udy p ocedu es.
All s udy p ocedu es and ma e ials we e app o ed by an
independen ins i u ional e iew boa d (E hical and Inde-
penden Re iew Se ices; S udy Numbe 21178-01).
Heal h s a e de elopmen
Heal h s a e igne es desc ibed a ying le els o ASMD
se e i y o adul s and child en. These heal h s a es we e
de eloped based on a e iew o clinical ial da a, consul-
a ion wi h clinicians specializing in ASMD, in e iews
wi h adul s wi h ASMD, and in e iews wi h ca egi e s o
child en wi h ASMD. A a ge ed li e a u e sea ch was pe -
o med o in o m he de elopmen o he clinician, pa ien ,
and ca egi e in e iew guides and o iden i y symp oms and
a eas o impac associa ed wi h ASMD o be included in he
heal h s a es [2–11, 39–42].
To in o m con en o he heal h s a es, mul iple ounds o
in e iews we e conduc ed wi h h ee clinicians om Eng-
land, Ge many, and I aly, including wo physicians specializ-
ing in me abolic diseases (one in adul s and one in pedia ics)
and a physician specializing in lysosomal s o age diso de s
in bo h adul s and child en. Each clinician had mo e han
20yea s o clinical expe ience in managing pa ien s wi h
ASMD. In he ini ial in e iews, he ad iso s we e asked
o desc ibe ypical pa ien expe iences wi h ASMD. These
desc ip ions we e used o de elop he i s d a o he heal h
s a es, which we e hen e iewed and e ined in subsequen
in e iews o ensu e he desc ip ions we e clea and accu a e
ep esen a ions o he ypical pa ien expe ience.
In e iews we e also conduc ed wi h six adul s wi h
ASMD and wo ca egi e s o child en wi h ASMD. These
pa ien s and ca egi e s p o ided eedback on he heal h s a e
desc ip ions. Each esponden was asked o e iew and com-
men on he heal h s a e mos closely ma ching hei own
cu en condi ion (o hei child’s cu en condi ion), as well
as any heal h s a es co esponding o hei p e ious expe i-
ences wi h ASMD. Heal h s a es we e e ised based on hei
inpu , and hese e isions we e discussed wi h and app o ed
by he clinicians.
The inal se o 14 heal h s a es included se en adul and
se en pedia ic igne es desc ibing ypical expe iences wi h
ASMD a a ying le els o se e i y. Al hough ASMD is
associa ed wi h a a ie y o symp oms [8, 11], he heal h
s a es ocused on pulmona y dys unc ion, enla gemen o
he spleen and li e , and he impac o hese impai men s
on quali y o li e. Pulmona y dys unc ion and spleen/li e
enla gemen we e p io i ized because hey a e key mani es-
a ions o ASMD ha a e associa ed wi h no able impac on
pa ien s’ li es [6, 8, 11, 29, 39–41].
The se en heal h s a es in he adul and pedia ic se s
included combina ions o pulmona y dys unc ion se e i y
and spleen/li e enla gemen (summa ized in Table1). These
combina ions o symp om se e i y we e selec ed based on
discussions wi h he clinicians and e iew o clinical ial
da a o iden i y he combina ions o symp oms commonly
expe ienced by pa ien s. Fo example, he e was no heal h
s a e ep esen ing se e e pulmona y dys unc ion wi hou
spleen/li e enla gemen because he clinicians epo ed
ha hey do no see his combina ion in hei pa ien s. This
clinical opinion was suppo ed by baseline da a om wo
clinical ials [13, 15] and da a om a na u al his o y s udy
o ASMD [43]. Among he 84 pa ien s wi h ASMD ac oss
hese h ee s udies, none had se e e pulmona y dys unc ion
wi hou spleen/li e enla gemen , and only 2.4% had se e e
spleen/li e enla gemen wi hou pulmona y dys unc ion.
The heal h s a es we e p esen ed o esponden s on indi-
idual ca ds wi h bulle poin desc ip ions o ganized in o
sec ions wi h headings o “B ea hing,” “Spleen and Li e ,”
and “Quali y o Li e.” The “Quali y o Li e” sec ion included
subheadings o “ac i i ies,” “in ec ions and hospi aliza-
ions,” “appea ance,” and “emo ional impac .” The inal
heal h s a es a e p esen ed in he Supplemen a y Ma e ial.
Pa icipan s
This s udy was conduc ed wi h a sample o gene al popula ion
esponden s, consis en wi h HTA p e e ences o he gene al
popula ion pe spec i e in u ili y elici a ion [21]. Pa icipan s
we e ec ui ed using digi al social media ma ke ing (e.g.,
ia Facebook, Twi e , Google). Po en ial pa icipan s who
esponded wi h in e es we e sc eened by phone o eligibili y.
All pa icipan s we e equi ed o be o e 18 yea s old, a esi-
den o he UK, able o unde s and he assessmen s as judged
by he in es iga o , able and willing o gi e elec onic con-
sen , and able o comple e he p o ocol equi emen s. E o s
we e made o ec ui a sample o esponden s e lec i e o
he UK gene al popula ion wi h ega d o gende , age, acial/
e hnic backg ound, and a e o unemploymen . Fo he sa e y
o in e iewe s and pa icipan s, all s udy pe sonnel and pa -
icipan s we e equi ed o be accina ed agains COVID-19
and masked du ing he in e iew.
1440 L.S.Ma za e al.
Pilo s udy
A pilo s udy was conduc ed wi h 25 gene al popula ion
pa icipan s in London, England. The pu pose o his pilo
s udy was o e alua e he heal h s a es and he u ili y assess-
men p ocedu es o ensu e hey we e clea o esponden s.
Pa icipan s we e asked o comple e he u ili y alua ion and
p o ide eedback. Heal h s a es and p ocedu es we e e ised
based on his eedback o imp o e cla i y and ease o unde -
s anding in he la ge alua ion s udy. The esul s om pilo
in e iews we e no included in he subsequen analysis.
U ili y in e iew p ocedu es andsco ing
The heal h s a es and p ocedu es inalized in he pilo s udy
we e used o elici u ili ies in he main s udy. In e iewe s
ained in TTO me hods conduc ed one-on-one in e iews
ollowing a semi-s uc u ed in e iew guide using TTO
bookle s and heal h s a e ca ds. Pa icipan s we e and-
omized o begin wi h ei he he adul o pedia ic heal h
s a es i s , ollowed by he o he se .
Pa icipan s we e i s in oduced o he heal h s a es
and shown a backg ound desc ip ion o ei he adul o
pedia ic ASMD, acco ding o hei heal h s a e o de
assignmen . Then, he i s se o se en heal h s a es (i.e.,
ei he adul o child) was p esen ed in andom o de , and
pa icipan s we e asked o ank he se en heal h s a es in
o de o p e e ence (i.e., om mos p e e able o leas p e -
e able). Heal h s a es we e labeled wi h le e s o acili a e
discussion du ing he in e iews, bu he o de o he le -
e s did no co espond o he se e i y o he heal h s a es.
A e anking he heal h s a es and explaining hei p e -
e ences, pa icipan s alued he heal h s a es in he TTO
ask. Then, pa icipan s comple ed he anking and TTO
asks o he second se o heal h s a es.
To elici u ili ies o he adul heal h s a es, pa ici-
pan s alued he heal h s a es in a TTO ask wi h a 10-yea
ime ho izon ollowing p ocedu es desc ibed ex ensi ely
in p e ious publica ions [17, 44]. The speci ied amoun
o ime in he heal h s a e (i.e., he ime ho izon) a ies
ac oss TTO u ili y elici a ion s udies, and no speci ic ime
ho izon is uni e sally p e e ed o conside ed o be mo e
“co ec ” [44–47]. The mos commonly used TTO ime
ho izon appea s o be 10yea s, which was used in he
in luen ial Measu emen and Valua ion o Heal h s udy
ha iden i ied u ili ies o EQ-5D heal h s a es [48, 49],
he EQ-VT p o ocol o alua ion o EQ-5D-5L s a es,
he in e na ional alua ion p o ocol o aluing pedia ic
heal h s a es de i ed om he EQ-5D-Y [50], and a wide
ange o igne e-based s udies [47, 51–56]. To maximize
compa abili y wi h p e ious esea ch, he 10-yea ime
ho izon was selec ed o he cu en s udy.
Fo each heal h s a e, pa icipan s we e o e ed a se ies
o choices be ween a 10-yea pe iod in he heal h s a e
Table 1 Lis o heal h s a es
ASMD acid sphingomyelinase de iciency, DLco di using lung capaci y, MN mul iples o no mal
Adul
heal h
s a es
Child
heal h
s a es
Desc ip ion Pulmona y unc ion Spleen and li e size
A1 C1 ASMD wi hou symp oms o impai men No impai men
DLco ≥ 80% No mal
Spleen olume < 6 MN
Li e olume no mal
A2 C2 ASMD wi h mild- o-mode a e pulmona y dys unc ion Mild- o-mode a e impai men
DLco ≥ 40% o 79% No mal
Spleen olume < 6 MN
Li e olume no mal
A3 C3 ASMD wi h mode a e spleen/li e enla gemen No impai men
DLco ≥ 80% Mode a ely enla ged
Spleen olume ≥ 6 o 15 MN
Mode a e li e enla gemen
A4 C4 Mild- o-mode a e ASMD (combina ion o mode a e
impai men desc ibed in A2/C2 and A3/C3) Mild- o-mode a e impai men
DLco ≥ 40% o 79% Mode a ely enla ged
Spleen olume ≥ 6 o 15 MN
Mode a e li e enla gemen
A5 C5 ASMD wi h mild- o-mode a e pulmona y dys unc ion
wi h se e e spleen/li e enla gemen Mild- o-mode a e impai men
DLco ≥ 40% o 79% Ve y enla ged
Spleen olume > 15 MN
Se e e li e enla gemen
A6 C6 ASMD wi h se e e pulmona y dys unc ion wi h mode -
a e spleen/li e enla gemen Se e e impai men
DLco < 40% Mode a ely enla ged
Spleen olume ≥ 6 o 15 MN
Mode a e li e enla gemen
A7 C7 Se e e ASMD (combina ion o se e e impai men
desc ibed in A5/C5 and A6/C6) Se e e impai men
DLco < 40% Ve y enla ged
Spleen olume > 15 MN
Se e e li e enla gemen

1441
Assessmen o heal h s a e u ili ies associa ed wi hadul andpedia ic acid sphingomyelinase…
being a ed and a ying amoun s o ime in ull heal h.
Fo he adul heal h s a es, he TTO ask was p esen ed
in he con ex o he esponden ’s own li e (“Which li e
would you p e e ? You can choose Li e 1, Li e 2, o hey
can be equal.”). Choices we e p esen ed in 6-mon h (5%)
inc emen s, al e na ing be ween longe and sho e pe i-
ods o ime in ull heal h. The inc emen o 6mon hs was
selec ed o allow o a easonable le el o sensi i i y o
di e ences be ween he heal h s a es, wi hou assuming a
po en ially un ealis ic deg ee o p ecision ha would be
associa ed wi h smalle inc emen s.
The e is no consensus ega ding he bes app oach o
elici ing u ili ies o pedia ic heal h s a es [28, 57, 58].
S udies a y wi h ega d o he ype o sample aluing he
heal h s a es (e.g., pa en s, gene al popula ion adul s, o
child en) as well as he aming o he u ili y alua ion ask
(e.g., aluing heal h s a es o one’s own child, you sel
as a child, o a hypo he ical unknown child). The cu en
s udy used an app oach simila o a ecen s udy in ol -
ing EQ-5D-Y heal h s a es [50]. Gene al popula ion adul
esponden s we e asked o imagine a child o a speci ic
age, bu he child’s iden i y was no speci ied. The age o
8yea s old was selec ed o he cu en s udy because i
was he mean age o pedia ic pa icipan s a baseline in
he ASCEND-Peds ial o ea men o ASMD [13]. The
TTO ask was amed as ollows: “Imagine an 8-yea -old
child wi h a a e gene ic diso de . Fo his child, which
li e do you hink would be p e e able? Li e 1, Li e 2, o
a e Li e 1 and Li e 2 abou equal?” O he han his e-
aming, he TTO p ocedu es o pedia ic heal h s a es
we e he same as he p ocedu es desc ibed abo e o he
adul heal h s a es. A e comple ing he TTO ask, pa -
icipan s we e asked i hey we e hinking abou a speci ic
child du ing he TTO.
Each heal h s a e pe cei ed o be be e han “dead”
ecei ed a u ili y alue (u) on a scale wi h ancho s o dead
(0) and ull heal h (1) based on he poin o indi e ence
be ween y yea s in he heal h s a e being alued and x
yea s in ull heal h ( ollowed by dead). U ili y was calcu-
la ed as u = x/y. When pa icipan s p e e ed “dead” o e a
heal h s a e, he ask and sco ing p ocedu es we e al e ed
as desc ibed in p e ious li e a u e [59]. Pa icipan s we e
o e ed a choice be ween dead (choice 1) and a 10-yea li e
span (choice 2) beginning wi h a ying amoun s o ime in
he heal h s a e being a ed, ollowed by ull heal h o he
emainde o he li e span. The esul ing nega i e u ili y
sco es we e calcula ed wi h he bounded sco ing app oach
(u = − x/10, whe e x is he numbe o yea s in ull heal h).
S a is ical analysis p ocedu es
S a is ical analyses we e comple ed wi h SAS ( e sion 9.4,
SAS Ins i u e, Ca y, NC, USA). Desc ip i e s a is ics we e
used o summa ize demog aphic da a, heal h s a e ank-
ings, and u ili ies. Ca ego ical a iables a e summa ized
as equencies and pe cen ages, while means and s anda d
de ia ions a e epo ed o con inuous a iables. Pai ed
es s we e conduc ed o es whe he he e we e s a is i-
cally signi ican pai wise di e ences be ween hea h s a e
u ili ies (e.g., he u ili y o he heal h s a e ep esen ing
ASMD wi hou impai men s. se e e ASMD). Independ-
en es s we e conduc ed o es o di e ences in u ili y
by age, gende , and pa en ing s a us.
Table 2 Demog aphic and clinical cha ac e is ics
SD s anda d de ia ion
a O he ma i al s a us includes di o ced (n = 19), sepa a ed (n = 4),
widowed (n = 1), and o he (no speci ied) (n = 3)
b O he employmen s a us includes s uden (n = 20), unemployed
(n = 19), e i ed (n = 14), homemake (n = 7), and o he (no speci ied)
(n = 10)
Cha ac e is ics Desc ip i e s a is ics
(n = 202)
Age, mean yea s (SD) 41.3 (14.4)
Gende , n (%)
Male 99 (49.0%)
Female 101 (50.0%)
Nonbina y 2 (1.0%)
E hnic/ acial backg ound, n (%)
Asian/Asian B i ish 20 (9.9%)
Black/A ican/Ca ibbean/Black B i ish 5 (2.5%)
Whi e 167 (82.7%)
Mixed/mul iple e hnic g oups 7 (3.5%)
O he 3 (1.5%)
Ma i al s a us, n (%)
Single 114 (56.4%)
Ma ied/cohabi a ing/li ing wi h a pa ne 61 (30.2%)
 O he a27 (13.4%)
Do you ha e any child en?
No, n (%) 151 (74.8%)
Yes, n (%) 51 (25.2%)
I yes, how many? Mean (SD) 1.7 (0.8)
I yes, how many o hese child en s ill
depend on you o ca e o hem? Mean
(SD)
0.8 (0.8)
Employmen s a us, n (%)
Full- ime wo k 92 (45.5%)
Pa - ime wo k 40 (19.8%)
 O he b70 (34.7%)
Educa ion le el, n (%)
Uni e si y deg ee 146 (72.3%)
No uni e si y deg ee 56 (27.7%)
1442 L.S.Ma za e al.
Resul s
Sample cha ac e is ics
A o al o 221 pa icipan s we e scheduled, and 203 a ended
hei in e iews. One o he 203 had di icul y unde s and-
ing he u ili y p ocedu es and was unable o p o ide alid
da a. The e o e, he analyses we e conduc ed wi h a sample
o 202 pa icipan s. The sample epo ed gende as 50.0%
emale, 49.0% male, and 1.0% nonbina y, wi h a mean age
o 41.3yea s (Table2). Abou one-qua e o he sample
epo ed ha ing child en (25.2%). The mos commonly
epo ed heal h condi ions we e anxie y (33.2%), dep ession
(27.2%), as hma (10.9%), and a h i is (8.9%). No pa ici-
pan s epo ed ha ing ASMD, and one pa icipan (0.5%)
epo ed knowing someone who has been diagnosed wi h
ASMD.
Heal h s a e ankings andp e e ences
In he in oduc o y ask, pa icipan s anked he se en adul
heal h s a es (A1 o A7) and he se en child heal h s a es (C1
o C7) sepa a ely in o de o p e e ence. Rankings, which
anged om 1 (mos p e e able heal h s a e) o 7 (leas p e -
e able heal h s a e), ollowed he same pa e n o he adul
and child heal h s a es. The heal h s a es desc ibing ASMD
wi hou impai men (A1, C1) we e mos p e e ed o all
pa icipan s. The heal h s a es desc ibing se e e ASMD
we e leas p e e ed by nea ly all pa icipan s (99.5% o
A7; 100% o C7).
P e e ences o he in e media e heal h s a es also ol-
lowed a logical o de . The heal h s a es wi h only one symp-
om a mild- o-mode a e se e i y (A2, A3, C2, C3) we e
a ed as second and hi d in p e e ence by nea ly all pa -
icipan s (99.5%), al hough he o de a ied be ween hese
pai s o heal h s a es (i.e., A2 s. A3 and C2 s. C3). Mild-
o-mode a e ASMD was anked ou h by nea ly all pa ici-
pan s (99.5% o bo h A4 and C4). The heal h s a es wi h
one mild- o-mode a e symp om and one se e e symp om
(A5, A6, C5, C6) we e anked as i h and six h by nea ly all
pa icipan s (99.5%), wi h a ying o de be ween hese pai s
o heal h s a es (i.e., A5 s. A6 and C5 s. C6).
Heal h s a e u ili ies
Heal h s a e u ili ies a e p esen ed in Table3 and Fig.1.
Mean u ili ies ollowed a logical o de , wi h u ili y dec eas-
ing as symp om se e i y inc eased. ASMD wi hou symp-
oms o impai men had he highes u ili y o bo h he adul
(A1) and child heal h s a es (C1) (0.92 and 0.94, espec-
i ely). Se e e ASMD had he lowes u ili y o bo h he
adul (A7) and child heal h s a es (C7) (0.33 and 0.45,
espec i ely). E e y child heal h s a e had a signi ican ly
g ea e u ili y han he co esponding adul heal h s a e. Di -
e ences be ween adul /child pai s anged om 0.02 (A1 s.
C1) o 0.13 (A6 s. C6), and P alues o hese se en com-
pa isons anged om 0.04 o < 0.0001.
To es ima e disu ili y associa ed wi h a ious se e i y
le els o ASMD symp oms, u ili y di e ences we e calcu-
la ed o all heal h s a es wi h symp oms (i.e., A2 o A7
and C2 o C7) compa ed o ASMD wi hou symp oms o
impai men (A1 and C1). These disu ili ies a e p esen ed in
Table3. U ili y di e ences o he adul heal h s a es anged
om − 0.07 o mild- o-mode a e b ea hing impai men
(A2) o − 0.58 o se e e ASMD (A7). Disu ili ies o he
child heal h s a es anged om − 0.06 o mild- o-mode a e
b ea hing impai men (C2) o − 0.49 o se e e ASMD (C7).
Pai ed es s ound ha u ili ies o all heal h s a es wi h
symp oms we e signi ican ly lowe han he u ili y o ASMD
wi hou impai men (all P < 0.0001).
Willingness o ade ime a ied by heal h s a e se e i y.
Mos pa icipan s we e no willing o ade ime o a oid he
adul (69.8%) and pedia ic (68.3%) heal h s a es desc ibing
ASMD wi hou symp oms o impai men (A1, C1). How-
e e , mo e han hal o he pa icipan s we e willing o ade
ime o a oid all o he heal h s a es, wi h o e 90% willing o
ade ime o a oid he se e e ASMD heal h s a es (A7, C7).
Mos pa icipan s a ed all adul and child heal h s a es
as p e e able o dead, esul ing in posi i e u ili ies. Se e e
ASMD heal h s a es A7 and C7 had he highes a es o
nega i e u ili ies, wi h 13.4% and 8.9% pe cei ing hese
heal h s a es o be wo se han dead.
The adul and child heal h s a es each ecei ed a o al
o 1414 alua ions (i.e., se en heal h s a es alued by 202
esponden s). Fo he adul heal h s a es, he e we e a o al
o 63 nega i e alua ions (4.5%) and 27 (1.9%) alua ions
esul ing in a u ili y o 1. Fo he child heal h s a es, he e
we e a o al o 42 nega i e alua ions (3.0%) and 26 (1.8%)
alua ions esul ing in a u ili y o 1.
Subg oup analyses
The e we e no signi ican be ween-g oup di e ences in u il-
i y by gende o ei he he adul o child heal h s a es. In
addi ion, he e we e no signi ican age di e ences o u ili-
ies o he adul heal h s a es (i.e., olde s. younge sub-
g oups, ca ego ized based on median spli ). Howe e , he e
we e di e ences by age o he child heal h s a es. Fo all
child heal h s a es, he olde subg oup (n=99) had g ea e
mean u ili ies han he younge subg oup (n=103), and
his be ween-g oup di e ence was s a is ically signi ican o
ou heal h s a es (C2, C3, C4, C7), wi h he g ea es di e -
ence o he se e e ASMD heal h s a e (0.120, P = 0.047).
These age di e ences o he child heal h s a es may be
ela ed o whe he he esponden s had child en. Among he
1443
Assessmen o heal h s a e u ili ies associa ed wi hadul andpedia ic acid sphingomyelinase…
olde subg oup, 44 o 99 esponden s (44.4%) epo ed ha -
ing child en. In con as , only se en o he 103 esponden s
(6.8%) in he younge subg oup epo ed ha ing child en.
To examine he po en ial associa ion be ween u ili ies and
pa en ing s a us, es s we e conduc ed o compa e u ili ies
be ween he subg oups wi h (n = 51) and wi hou (n = 151)
child en (Table4). The e we e no signi ican be ween-g oup
di e ences o he adul heal h s a es. Howe e , he g oups
di e ged in alua ions o he child heal h s a es. The sub-
g oup wi h child en ga e highe u ili ies o e e y child
heal h s a e. The be ween-g oup di e ence was s a is ically
signi ican o i e o he se en heal h s a es (C3 o C7), and
his di e ence inc eased wi h he mo e se e e heal h s a es.
The se e e ASMD heal h s a e had he g ea es u ili y di e -
ence (be ween-g oup di e ence = 0.21, P = 0.002).
Child conside ed when aluing hepedia ic heal h
s a es
Fo he pedia ic heal h s a es, esponden s we e asked o
hink abou he heal h s a es in he con ex o “an 8-yea -
old child wi h a a e gene ic diso de .” A e comple ing he
TTO u ili y elici a ion, esponden s we e asked whe he hey
we e hinking o a speci ic child. Responses a ied widely
(Table5). App oxima ely hal o he pa icipan s who had
child en (51.0%) epo ed hinking abou hei own child
when aluing he pedia ic heal h s a es. In con as , mos
(64.2%) o he pa icipan s wi hou child en hough abou
a hypo he ical o gene ic child a he han a speci ic child.
Discussion
U ili ies ollowed he expec ed pa e n wi h g ea e se e i y
o ASMD associa ed wi h lowe mean u ili y o bo h he
adul and pedia ic heal h s a es. Consis en wi h he b oad
ange o disease se e i y in hese heal h s a es, u ili ies a -
ied widely. Sco es anged om 0.92/0.94 (i.e., adul /child)
o ASMD wi hou symp oms o impai men o 0.33/0.45
o ASMD wi h se e e impai men . While he highe u ili-
ies a e in he ange ypically obse ed o ela i ely heal hy
indi iduals, he lowe u ili ies a e in a ange simila o o he
diseases wi h subs an ial symp om se e i y and impai -
men , such as p og essi e lymphoma [60] and sho bowel
synd ome wi h daily in a enous nu i ion supplemen a ion
[61].
Table 3 Heal h s a e u ili ies and disu ili ies
ASMD acid sphingomyelinase de iciency, CI con idence in e al, SD s anda d de ia ion
a Disu ili y o each symp om/impai men is compu ed by sub ac ing heal h s a e 1 (ASMD wi hou impai men ) om each o he o he heal h
s a es (ASMD wi h impai men )
Heal h s a e U ili y Disu ili y o symp oms/impai men
(di e ence om ASMD wi hou
impai men )a
Mean (SD) 95% CI Mean (SD) 95% CI
Adul ASMD heal h s a es
A1: No symp oms o impai men 0.918 (0.184) 0.892 o 0.943 – – –
A2: Mild- o-mode a e pulmona y dys unc ion 0.847 (0.215) 0.817 o 0.876 − 0.071 (0.110) − 0.086 o − 0.056
A3: Mode a e spleen/li e enla gemen 0.837 (0.219) 0.806 o 0.867 − 0.081 (0.116) − 0.097 o − 0.065
A4: Mild- o-mode a e ASMD 0.770 (0.246) 0.736 o 0.804 − 0.148 (0.166) − 0.171 o − 0.125
A5: Mild- o-mode a e pulmona y dys unc ion wi h se e e
spleen/li e enla gemen
0.532 (0.423) 0.474 o 0.591 − 0.385 (0.381) − 0.438 o − 0.332
A6: Se e e pulmona y dys unc ion wi h mode a e spleen/li e
enla gemen
0.527 (0.402) 0.472 o 0.583 − 0.390 (0.359) − 0.440 o − 0.340
A7: Se e e ASMD 0.333 (0.468) 0.268 o 0.398 − 0.584 (0.440) − 0.645 o − 0.523
Child ASMD heal h s a es
C1: No symp oms o impai men 0.939 (0.104) 0.924 o 0.953 – – –
C2: Mild- o-mode a e pulmona y dys unc ion 0.882 (0.146) 0.862 o 0.903 − 0.056 (0.092) − 0.069 o − 0.044
C3: Mode a e spleen/li e enla gemen 0.867 (0.191) 0.840 o 0.893 − 0.072 (0.120) − 0.088 o − 0.055
C4: Mild- o-mode a e ASMD 0.809 (0.241) 0.775 o 0.842 − 0.130 (0.182) − 0.155 o − 0.104
C5: Mild- o-mode a e pulmona y dys unc ion wi h se e e
spleen/li e enla gemen
0.594 (0.374) 0.542 o 0.646 − 0.344 (0.335) − 0.391 o − 0.298
C6: Se e e pulmona y dys unc ion wi h mode a e spleen/li e
enla gemen
0.654 (0.346) 0.606 o 0.702 − 0.285 (0.301) − 0.326 o − 0.243
C7: Se e e ASMD 0.450 (0.431) 0.390 o 0.510 − 0.489 (0.396) − 0.544 o − 0.434
1444 L.S.Ma za e al.
This s udy p o ides se en heal h s a e u ili ies co e-
sponding o a ious se e i ies o ASMD in child en and
ano he se en u ili ies o se e i ies o ASMD in adul s.
These 14 alues may be used o ep esen he heal h s a us
o pa ien s wi h ASMD in cos -u ili y modeling. Fo exam-
ple, a u ili y o 0.450 may be used in a model o ep esen a
child wi h se e e ASMD p io o ecei ing ea men . I a
clinical ial demons a es ha pedia ic pa ien s wi h se e e
ASMD ypically imp o e o mild- o-mode a e ASMD wi h
ea men , he u ili y o 0.809 may be used o ep esen he
hypo he ical child a endpoin . U ili ies o bo h child en and
adul s a e p esen ed in Table3.
Because o he s udy design, he cu en s udy o e s a
unique oppo uni y o compa e u ili y elici a ion o a pa al-
lel se o adul and child heal h s a es, alued by he same
esponden s. The di e ence be ween adul and child heal h
s a es a ied ac oss pa icipan s. Some esponden s had highe
u ili ies o adul heal h s a es, while o he s had highe u ili-
ies o child heal h s a es. Howe e , o all se en heal h s a e
pai s, i was mo e common o esponden s o ade mo e
ime om hei own li es han om he li e o a child, esul -
ing in lowe u ili ies o he adul heal h s a e (see Fig.1 o
an illus a ion o his end). Fo example, o heal h s a es
A6 and C6, 59.9% o he sample had a highe u ili y o he
child heal h s a e (C6), 25.7% had a highe u ili y o he adul
heal h s a e (A6), and 14.4% had he same u ili y o bo h.
This u ili y di e ence be ween adul and child heal h
s a es is consis en wi h esul s om p e ious s udies in
which EQ-5D-3L and EQ-5D-Y heal h s a es we e alued
by adul gene al popula ion esponden s [62, 63]. In hese
p e ious s udies, u ili ies ended o be lowe when espond-
en s we e asked o imagine hemsel es as adul s li ing in he
EQ-5D heal h s a es, compa ed wi h conside ing a child li -
ing in he same heal h s a es. The cu en s udy is he i s o
show ha his end occu s wi h u ili ies o disease-speci ic
adul and child heal h s a e igne es alued by he same
esponden s. In addi ion, he cu en esul s sugges ha he
di e ence be ween u ili ies o adul and child heal h s a es
may end o inc ease as he disease becomes mo e se e e and
u ili y alues dec ease.
Follow-up analyses highligh subg oup di e ences in
e alua ion o child heal h s a es. P e ious s udies ha e ound
ha being a pa en o ca egi e o a child has an impac
on TTO alua ions, ending o educe he amoun o ime
people a e willing o ade, which leads o highe u ili ies
[64, 65]. Resul s o mos o he adul heal h s a es in he
cu en s udy ollowed his pa e n, bu wi h ela i ely small
and non-signi ican di e ences be ween subg oups wi h and
wi hou child en (Table4). Fo he child heal h s a es, his
be ween-g oup di e ence was mo e consis en , g ea e in
magni ude, and s a is ically signi ican o he i e mos
se e e heal h s a es.
0.92
0.85 0.84
0.77
0.53 0.53
0.33
0.94
0.88 0.87
0.81
0.59
0.65
0.45
0
0.1
0.2
0.3
0.4
0.5
0.6
0.7
0.8
0.9
1
ASMD wi hou
symp oms o
impai men
ASMD wi h mild- o-
mode a e pulmona y
dys uncon
ASMD wi h mode a e
spleen/li e
enla gemen
Mild- o-mode a e
ASMD
ASMD wi h mild- o-
mode a e pulmona y
dys uncon and se e e
spleen/li e
enla gemen
ASMD wi h se e e
pulmona y dys uncon
and mode a e
spleen/li e
enla gemen
Se e e ASMD
Mean Uli y
Heal h S a e Vignees Desc ibing ASMD Type B
Adul Heal h S a es (A1 o A7)
Child Heal h S a es (C1 o C7)
A1 C1 A2 C2 A3 C3 A4 C4 A5 C5 A6 C6 A7 C7
Fig. 1 Heal h s a e u ili ies