Clin Genet 2016 © 2016 John Wiley & Sons A/S.

Printed in Singapore. All rights reserved Published by John Wiley & Sons Ltd
CLINICAL GENETICS
doi: 10.1111/cge.12787

Short Report

Down syndrome in adults: a 27-year follow-up

of adaptive skills
Arvio M., Luostarinen L. Down syndrome in adults: a 27-year follow-up of M. Arvioa,b and L. Luostarinena
adaptive skills. a Päijät-Häme Joint Municipal Authority,
Clin Genet 2016. © John Wiley & Sons A/S. Published by John Wiley & Lahti, Finland and b Turku University
Sons Ltd, 2016 Hospital, Turku, Finland

In 1988, we assessed the adaptive skills of 45 adults with Down syndrome

(DS) (21 women and 24 men, age 20–58) with the Portage scale. Since then,
we have followed them and also screened for signs of clinical dementia with
the Present Psychiatric State – Learning Disabilities assessment. The mean
adaptive age (AA) of the study group decreased with increasing age; the age
of 35 being the turning point in the clinical course of DS. The mean AA was
4.4 years between ages 20 and 34, 3.4 years between ages 35 and 49, and
2.4 years between ages 50 and 66. Inter-individual variation was, however,
large. Between ages 20 and 25, the AA of the study subjects ranged from 2.3 Key words: aging – dementia – Down
to 6 years; and after the age of 50, from 0.3 to 4.8 years. By the end of the syndrome – follow-up
study, all subjects showed signs of clinical dementia. These appeared most
Corresponding author: Maria Arvio,
frequently as reduced self-care skills, loss of energy, forgetfulness, and Arvi Kariston katu 4b10, 13100
impaired understanding. We found no connection between apolipoprotein E Hämeenlinna, Finland.
genotype and the clinical course of DS. We recommend follow-up of Tel.: +358 40 083 5614;
adaptive skills and screening for dementia signs in adults with DS. fax: +358 38 19 2311;
e-mail: maria.arvio@phsotey.fi
Conflict of interest
Received 16 December 2015, revised
The authors declare that there is no conflict of interest. and accepted for publication 7 April
2016

Down syndrome (DS) is the most common genetic cause
for intellectual disability (ID). Those with DS repre-
sent 10–15% of the ID population (1), with abnormally
slow development in childhood and premature aging in
adulthood as their main characteristics (2, 3). The devel-
opmental age of adolescents and young adults usually
corresponds that of 4–6 year old healthy children (2).
Alzheimer’s disease (AD) neuropathology is observable
in virtually all individuals by the age of 40 years (4–6).
Apolipoprotein (APO) E4 genotype is considered as a
risk factor for early dementia and death (7, 8).
Adaptive behavior reflects an individual’s competence
in daily skills to meet the demands of life. In diagnosing
ID, professionals assess adaptive skills (communication,
daily living skills, socialization and motor skills), in
addition to intelligence quotient.
Individuals with ID need the aid of others throughout
their lives. Those with relatively good adaptive skills
can live quite independently in a group home and also
work, for instance, as assistants in a warehouse; while
those with poor skills need constant assistance around
the clock. We aimed to determine the clinical course of
adaptive skills and identify signs of clinical dementia in
individuals with DS in a longitudinal study for 27 years.
Patients and methods
Finland was divided into 16 state-supported regional
full-service districts, providing residential care, educa-
tion, rehabilitation, medical follow-up, shelter work, and
day activity for individuals with ID, back in the late
1960s. In 1988, the client register of the South Häme dis-
trict, with a total population of 340,000, included names
of 229 persons with DS; of these 184 were older than
19. The study subjects were invited according to the fol-
lowing criteria: DS without evidence of chromosomal
mosaicism, age 19+, and a follow-up appointment with
the first author during summer 1988. A total of 45 adults,
21 women and 24 men, meeting the inclusion criteria
were willing to participate.

1
Arvio and Luostarinen
Table 1. Clinical data of the study group at the baseline and at the end of the study

1988 2015
Women; n (%) Men; n (%) Women; n (%) Men; n (%)

21 (47) 24 (53) 8 (47) 9 (53)

Age in years
20–29 8 (38) 9 (37) 0 0
30–39 6 (29) 4 (17) 0 0
40–49 5 (24) 7 (29) 3 (37.5) 2 (22)
50–59 2 (9) 4 (17) 3 (37.5) 6 (67)
60+ 0 0 2 (25) 1 (11)
State of health
Alzheimer disease, confirmed?a ? ? 5 (63) 4 (44)
Visual impairment 2 (10) 0 3 (38) 0
Congenital heart defect 2 (10) 3 (13) 0 0
Diabetes type 1 1 (5) 2 (8) 1 (13) 1 (11)
Epilepsy 4 (20) 4 (17) 1 (13) 3 (33)
Hypothyroidism 6 (29) 6 (25) 3 (38) 4 (44)
Psychiatric disorder 2 (10) 3 (13) 2 (25) 2 (22)
Hypercholesterolemiab ? ? 0 2 (22)
Gout 0 0 0 2 (22)
Hirschprung disease, stoma 0 2 (8) 0 2 (22)
Urinary infection 0 0 2 (25) 0
No health problem 4 (20) 3 (13) 0 0
Accommodation
With relatives 11 (52) 11 (46) 1 (12) 0
Nursing home 10 (48) 13 (54) 7 (88) 9 (100)
Severity of ID at adolescence
Mild 1 (5) 0 1 (13) 0
Moderate 8 (38) 15 (63) 3 (37.5) 7 (78)
Severe 11 (43) 8 (29) 3 (37.5) 2 (22)
Profound 1 (14) 1 (8) 1 (13) 0

ID, intellectual disability.

a
Alzheimer disease diagnostics were considered only in late 1990s. The diagnosis was confirmed with brain magnetic resonance
image.
b Cholesterol levels were not systemically determined in 1988.

All assessments were carried by the same two special
nurses during home visits and by the following evalua-
tion methods:
In 1988: Health status determination, verification of the
level of ID at adolescence according to their records.
In 1988, 1992, 1995, 1999, 2003, and 2015: An
adaptive age (AA) assessment with the Portage scale (9).
This includes 527 questions which assess an individual’s
social, verbal, self-help, cognitive, and motor skills.
Portage is primarily intended as an early intervention
tool, but in Finland, this method has served in the
follow-up of individuals of all ages with a moderate,
severe, or profound ID (10).
In 2001: APO E genotype testing to 31 study subjects.
In 2001 and 2012: Screening for signs of clinical
dementia with Present Psychiatric State – Learning Dis-
abilities assessment (11–13).
In 2015: Health status determination of surviving study
subjects, and collection of the ages and causes of death
of deceased study subjects.
Relationship between AA and chronological age
was estimated using random-effects [generalized least
squares (GLS)] regression with quadratic term, subject
identifier as a random effect; this method accounts for
the correlated structure of dependent variables arising
from repeated measures over time, controlling for each
individual. Bootstrap estimation was used to derive a
95% confidence interval of curvilinear correlation.
The Pirkanmaa Hospital District’s ethics committee
reviewed and supported our study. Written informed
consent came from each of the study subjects or from
their caretaker or both.
Results
Clinical data of the 45 study subjects is shown in Table 1.
At baseline, 84%, and, at the end of the study, every
subject had at least one longstanding health problem
or impairment. During the follow-up period, the size
of the study group decreased. The caretakers of three
subjects refused participation in the final evaluation in
2015. Twenty-five (12 women and 13 men) died, median
age at death being 57 (range: 36–65).

2

Men
8
Curvilinear correlation = 0.50 (95% CI: 0.26 to 0.67)
6
Adaptive age in years
0
20 25 30 35 40 45 50 55 60
Chronological age in years
Fig. 1. Adaptive ages according to the Portage scale of 45 adults with Down syndrome by chronological age, 1–6 assessments/individual, in total 199
assessments. Non-linear quadratic models with 95% confidence intervals. Difference between gender p = 0.17.
Figure 1 shows AAs of 45 study subjects (1–6
AAs/subject, total of 199 AAs) in terms of chronolog-
ical age. Inter-individual variation was large. Between
ages 20 and 25, the AA ranged from 2.3 to 6 years;
and after the age of 50 from 0.3 to 4.8 years. The
mean AA of the study group decreased with increas-
ing age; age 35 being the mean turning point (Fig. 1).
The mean AA was 4.4 years between ages 20 and 34,
3.4 years between ages 35 and 49, and 2.4 years between
ages 50 and 66.
In APO genotype testing, 19 (61%) subjects had
E3/E3, 9 (29%) E4/E3, 2 (6%) E4/E4, and 1 (3%) E3/E2.
Figure S1, Supporting information shows AAs of 31
study subjects (2–6 AAs/subject, total of 159 AAs) rep-
resenting different APO genotypes. There was no con-
nection between the clinical course of adaptive skills
and APO genotype. Of the 25 deceased study subjects,
12 were alive in 2001 and gave a sample for genotype
testing. The median age at death of those 8 with E3/E3
was 58 (range: 45–65). The ages at death of the 4 with
E4/E3 were 36, 58, 58, and 62. The cause of death was
paralytic ileus associated with diagraphic hernia in a
36 year old man (APO E4/E3, normal brain MRI at the
age of 33; minor changes suggesting AD in neuropatho-
logical post-mortem); congenital heart defect associated
with diabetes mellitus type 1 in a 39 year old man (no
APO genotype testing); cortical multi-infarction related
to Moyamoya in a 45 year old woman (APO E3/E3); and
laryngeal cancer in a 57 year old man (APO E3/E3). For
21 study subjects, the cause of death was AD. Of these,
12 had undergone brain imaging, and 1 neuropathologi-
cal examination showing AD-characteristic findings, and
4 a routine autopsy.
Down syndrome in adults
Women
Curvilinear correlation = 0.59 (95% CI: 0.44 to 0.71)
65 20 25 30 35 40 45 50 55 60 65
Chronological age in years
In 19 study subjects (9 women and 10 men) for
whom we obtained follow-up data during 2001–2012,
Table 2 shows signs of clinical dementia. These signs
appeared most frequently as reduced self-care skills, loss
of energy, forgetfulness, and impaired understanding.
The adolescent level of ID and the number of dementia
signs showed no correlation.
Discussion
The health problems common in elderly people are
becoming more frequent in daily medical ID practice
as the number of senior-aged citizens with ID continues
growing (14). A fact long known is that DS is a risk factor
for AD and for Moyamoya-related vascular dementia (4,
15, 16). In two recent studies, the median age of demen-
tia development was 55.5 in those with DS (17, 18). This
age appears high, because as neuroradiological and neu-
ropathological studies report AD-characteristic findings
already evident at the age of 40 (4–6). In our longitu-
dinal study, age 35 was the turning point in the clinical
course of DS, in accordance with radiological and labo-
ratory findings. Further, in our study the signs of clinical
dementia were frequent already at the mean age of 41
but continued increasing with age (Table 2). Dementia
signs in our study were quite the same: reduced self-care
skills, loss of energy, forgetfulness, and impaired under-
standing, as in a recent study, although those Italian
colleagues used other expressions like ‘deficit in verbal
comprehension, along with social isolation’, ‘loss of
interest’, and ‘greater fatigue in daily tasks’ (18). The
British study reported ‘frontal lobe’-related dementia
symptoms like general slowness, language problems,
loss of interest in activities, social withdrawal in adults
3
Arvio and Luostarinen
Table 2. Signs indicating dementia in 19 study subjects accord- On the other hand, in the diagnostic workup the fact
ing to Present Psychiatric State – Learning Disabilities checklist that those with ID have undergone several psychological
tested in 2001 and 2012 evaluations, and most of them, also brain imaging in
2001 2012
childhood and adolescence is helpful.
Dementia and ID are both signs reflecting disturbed
Age, range, mean 34–54, 41 45–65, 52 cortical function. ID is diagnosed in childhood and
Sign n (%) n (%) dementia usually in adulthood. In those with ID, the
Autonomic anxiety 4 (21) 7 (37) signs of clinical dementia may quite rapidly have a
Change in appetite 4 (21) 4 (21) huge impact upon a person’s life, resulting in personal
Changed sleep pattern 3 (16) 6 (32) discomfort and permanent placement in a nursing home.
Coarsening of personality 2 (11) 9 (47) We therefore suggest regular follow-up of adaptive skills
Confusion 2 (11) 5 (26) and paying special attention to a healthy lifestyle and to
Delusions 2 (11) 5 (26) nutrition in adults with DS.
Diurnal mood variation 6 (32) 8 (42)
Forgetfulness 4 (21) 14 (7)
Forgetting people’s names 1 (5) 3 (16) Supporting Information
Geographical disorientation 2 (11) 7 (37)
Impaired understanding 5 (26) 15 (79) Additional supporting information may be found in the online
Irritability 4 (21) 10 (53) version of this article at the publisher’s web-site.
Loss of concentration 4 (21) 11 (58)
Loss of energy 6 (32) 15 (79)
Loss of literacy skills 1 (5) 5 (26) Acknowledgements
Misery 6 (32) 8 (42) We thank Jorma Koskinen and Liseli Louhiala for help in the clinical
Onset of or increase in other 3 (16) 9 (47) evaluation of the patients, Hannu Kautiainen for statistical analysis,
maladaptive behavior and Päijät-Häme Joint Municipal Authority for financial support.
Onset of or increase in physical 4 (21) 7 (37)
aggression
Onset of or increase in verbal 1 (5) 3 (16) References
aggression 1. Stromme P. Aetiology in severe and mild mental retardation: a
Onset of or increased fearfulness 4 (21) 8 (42) population-based study of Norwegian children. Dev Med Child Neurol
Reduced quantity of speech 4 (21) 13 (68) 2000: 42: 76–86.
Reduced self-care skills 4 (21) 17 (89) 2. Tsao R, Kindelberger C. Variability of cognitive development in children
with Down syndrome: relevance of good reasons for using the cluster
Social withdrawal/reduced social 1 (5) 7 (37) procedure. Res Dev Disabil 2009: 30: 426–432.
interaction 3. Hawkins BA, Eklund SJ, James D, Foose AK. Adaptive behavior and
Tearfulness 4 (21) 7 (37) cognitive function of adults with Down’s syndrome: modeling change
Temporal disorientation 4 (21) 8 (42) with age. Ment Retard 2003: 41: 7–28.
Weight change 3 (16) 8 (42) 4. Margallo-Lana ML, Moore PB, Kay DW et al. Fifteen-year follow-up
Worry 5 (26) 9 (47) of 92 hospitalized adults with Down’s syndrome: incidence of cognitive
decline, its relationship to age and neuropathology. J Intellect Disabil
Res 2007: 51 (6): 463–477.
5. Prasher V, Cumella S, Natarajan K, Rolfe E, Shah S, Hague MS. Mag-
netic resonance imaging, Down’s syndrome and Alzheimer’s disease:
with DS. These symptoms were common also in our
research and clinical implications. J Intellect Disabil Res 2003: 47:
study subjects (19). Some studies suggest APO E4 90–100.
genotype as an additional AD risk factor in those with 6. Roth GM, Sun B, Greensite FS et al. Premature aging in persons
DS (4, 7, 8). In our small sample, the clinical course of with Down syndrome: MR findings. Am J Neuroradiol 1996: 17:
adaptive skills (Fig. S1) and also ages at death of those 1283–1289.
7. Rohn TT, McCarty KL, Love JE, Head E. Is apolipoprotein E4 an
with E4 genotype were the same as in subjects with other important risk factor for dementia in persons with Down syndrome?
genotypes. J Parkinsons Dis Alzheimers Dis 2014: 1 (1): pii: 7.
The strength of our study is the long follow-up time 8. Del Bo R, Comi GP, Bresolin N et al. The apolipoprotein E4 allele causes
of adult patients and thorough analysis of adaptive a faster decline of cognitive performances in Down’s syndrome subjects.
skills assessed by the same professionals. The weak- J Neurol Sci 1994: 145: 87–91.
9. Bluma S, Shearer M, Frohman A, Hilliard J. Portage: steps to
nesses of our study were the small sample and the early childhood education. In: Tiilikka P, Hautamäki J, eds.
fact that the study members were not of the same age Portaat-varhaiskasvatusohjelma (in Finnish). Helsinki: Kehitys-
at the study’s initiation. However, as the researches vammaliitto, 1986.
also become older, prospective follow-up studies of 10. Arvio M, Hautamäki J, Tiilikka P. Reliability and validity of the
Portage assessment scale for clinical studies of mentally handicapped
adult patients ones lasting for several decades, may be populations. Child Care Health Dev 1993: 19: 89–98.
impossible. 11. Cooper S. Psychiatric symptoms of dementia among elderly peo-
Dementia diagnostics in the ID population is chal- ple with learning disabilities. Int J Geriatr Psychiatry 1997: 12:
lenging (20). This population consists of individuals of 622–666.
great variety; firstly, because of their multiple etiologies 12. Mölsä P. Dementia and Intellectual Disability (in Finnish). Finn Med J
2001: 13: 1495–1497.
(genetic, acquired, or multifactorial), secondly because 13. Arvio M, Ajasto M, Koskinen J, Louhiala L. Middle-aged people
of large heterogeneity in the severity of ID, and thirdly with Down syndrome receive good care in home towns (in Finnish).
because of neuropsychiatric comorbidities like autism. Finn Med J 2013: 15: 1108–1112.

4

14. Glasson EJ, Sullivan SG, Hussain R, Petterson BA, Montgomery P,
Bittes BD. The changing survival profile of people with Down’s syn-
drome: implications for genetic counselling. Clin Genet 2002: 62 (5):
390–393.
15. Wisniewski K, Howe J, Williams DG, Wisniewski HM. Precocious aging
and dementia in patients with Down’s syndrome. Biol Psychiatry 1978:
13 (5): 619–627.
16. Devenny DA, Krinsky-McHale SJ, Sersen G, Silverman WP. Sequence
of cognitive decline in dementia in adults with Down’s syndrome.
J Intellect Disabil Res 2000: 44: 654–665.
17. Head E, Silverman W, Patterson D, Lott IT. Aging and Down syndrome.
Curr Gerontol Geriatr Res 2012: 2012: 412536.
Down syndrome in adults
18. Ghezzo A, Salvioli S, Solimando MC et al. Age-related changes of adap-
tive and neuropsychological features in persons with Down Syndrome.
PLoS One 2014: 9 (11): e113111.
19. Deb S, Hare M, Prior L. Symptoms of dementia among adults with
Down’s syndrome: a qualitative study. J Intellect Disabil Res 2007: 51:
726–739.
20. Prasher VP, Sachdeva N, Tarrant N. Diagnosing dementia in adults with
Down’s syndrome. Neurodegener Dis Manag 2015: 5 (3): 249–256.
DOI: 10.2217/nmt.15.8.
5
 Graphical Abstract

Men Women
8
Curvilinear correlation = 0.50 (95% CI: 0.26 to 0.67) Curvilinear correlation = 0.59 (95% CI: 0.44 to 0.71)
7

Adaptive age in years

5

0
20 25 30 35 40 45 50 55 60 65 20 25 30 35 40 45 50 55 60 65
Chronological age in years Chronological age in years

Relationships of adaptive age and chronological age in men and women.

Non-linear quadratic models with 95 per cent confidence intervals. Difference between gender p=0.17

Relationships of adaptive age and chronological age in men and women. Non-linear quadratic models with 95% confidence intervals. Difference between
gender p = 0.17.