Available via license: CC BY-NC
Content may be subject to copyright.
Received: 3 November 2020 Revised: 1 February 2021 Accepted: 30 July 2021
DOI: 10.1002/alz.12462
FEATURED ARTICLE
Attention-deficit/hyperactivity disorder and Alzheimer’s
disease and any dementia: A multi-generation cohort study in
Sweden
Le Zhang1Ebba Du Rietz1Ralf Kuja-Halkola1Maja Dobrosavljevic2
Kristina Johnell1Nancy L. Pedersen1Henrik Larsson1,2Zheng Chang1
1Department of Medical Epidemiology and
Biostatistics, Karolinska Institutet, Stockholm,
Sweden
2School of Medical Sciences, Örebro
University, Örebro, Sweden
Correspondence
Le Zhang and Zheng Chang, Department
of Medical Epidemiology and Biostatistics,
Karolinska Institutet, Nobels väg 12A, 171 65
Stockholm, Sweden.
E-mail: le.zhang@ki.se;zheng.chang@ki.se
Funding information
Swedish Council for Health, Working Life and
Welfare, Grant/AwardNumbers: 2019-00176,
2019-01172; Swedish Research Council,
Grant/AwardNumber: 2018-02599; Swedish
Brain Foundation,Grant/Award Number:
FO2018-0273; European Union’s Horizon
2020 research and innovation programme
under the 274 Marie Skłodowska-Curie,
Grant/AwardNumber: 1754285; Fredrik &
Ingrid Thurings Stiftelse; Karolinska Institutet
Research Foundation
Abstract
Introduction: We examined the extent to which attention-deficit/hyperactivity disor-
der (ADHD), a neurodevelopmental disorder, is linked with Alzheimer’s disease (AD)
and any dementia, neurodegenerative diseases, across generations.
Methods: A nationwide cohort born between 1980 and 2001 (index persons) were
linked to their biological relatives (parents, grandparents, uncles/aunts) using Swedish
national registers. We used Cox models to examine the cross-generation associations.
Results: Among relatives of 2,132,929 index persons, 3042 parents, 171,732 grand-
parents, and 1369 uncles/aunts had a diagnosis of AD. Parents of individuals with
ADHD had an increased risk of AD (hazard ratio 1.55, 95% confidence interval
1.26–1.89). The associations attenuated but remained elevated in grandparents and
uncles/aunts. The association for early-onset AD was stronger than late-onset AD. Sim-
ilar results were observed for any dementia.
Discussion: ADHD is associated with AD and any dementia across generations. The
associations attenuated with decreasing genetic relatedness, suggesting shared famil-
ial risk between ADHD and AD.
KEYWORDS
Alzheimer’s disease, dementia, epidemiology, family design, neurodevelopmental disorder
1INTRODUCTION
Attention-deficit/hyperactivity disorder (ADHD) is a common neu-
rodevelopmental disorder, characterized by impairing levels of poor
sustained attention, impaired impulse control, and hyperactivity.1
Follow-up studies have shown that the disorder often persists into
adulthood, affecting 3% of adults worldwide.2,3 Alzheimer’s disease
(AD) is a neurodegenerative disease characterized by aging-related
progressive deterioration in cognition and ability for independent liv-
ing, and it is the most common subtype of dementia. A meta-analytic
This is an open access article under the terms of the Creative Commons Attribution-NonCommercial License, which permits use, distribution and reproduction in any
medium, provided the original work is properly cited and is not used for commercial purposes.
© 2021 The Authors. Alzheimer’s & Dementia published by Wiley Periodicals LLC on behalf of Alzheimer’s Association
study estimated that the age-standardized prevalence of dementia in
those aged 60 and above ranged from 5% to 7% worldwide.4With the
life expectancy of individuals getting longer, dementia represents an
increasing public health concern.5
Very few studies, with limited sample size, have explored the
association between ADHD and AD, and conflicting results exist.6–8
An ecologic study using state-level hospitalization discharge data
from the United States found that antecedent ADHD significantly
predicted AD with incidence rate increased 15%.6On the other hand,
a case-control study interviewing individuals from a medical insurance
Alzheimer’s Dement. 2021;1–9. wileyonlinelibrary.com/journal/alz 1
2ZHANG ET AL.
group in Buenos Aires, Argentina, suggested that prior ADHD was not
associated with AD (non-significant increased odds 10%).7Similar non-
significant results were found in a retrospective matched-cohort study
using Taiwan’s Health Insurance Research Database.8So far, there are
no large-scale longitudinal studies that have explored the potential
association, as such a study would require very long follow-up from
diagnoses of ADHD, which was not commonly diagnosed until recent
decades,9to diagnoses of AD in older age. Even though the magnitude
of the association between these two conditions remains unclear, a few
potential mechanisms may explain an observed association.10 First,
ADHD could increase the risk of late-life dementia through adverse
health outcomes of ADHD. A review of meta-analyses identified seven
modifiable risk factors associated with dementia, including diabetes
mellitus, midlife hypertension, physical inactivity, depression, smoking,
and low educational attainment;11 each of these risk factors has
been shown to be a consequence of ADHD.12,13 Second, adult ADHD
may mimic cognitive symptoms of AD (including loss of memory and
inattention) as studies have suggested that ADHD is not adequately
and accurately identified in clinics in the context of late-life cognitive
disorders.14,15 Third, ADHD and AD may share genetic risk, as both
ADHD and AD are highly heritable and complex disorders. So far, the
largest genome-wide association studies (GWAS) have failed to detect
any genetic overlap between ADHD and AD,16,17 yet the result is
inconclusive as the genetic variants discovered so far only explained a
relatively small proportion of the heritability.
As ADHD has only been commonly diagnosed in recent decades
and follow-up data of diagnosed individuals into later life is limited, we
performed a longitudinal multi-generation study to explore the cross-
generation (i.e., parents, grandparents, and uncles/aunts) familial co-
aggregation of ADHD with AD and any dementia, using data from the
linkage of Swedish national registers.
2METHODS
2.1 Data sources
We used data from the linkage of several Swedish nationwide regis-
ters through unique personal identification numbers.18 (1) The Med-
ical Birth Register contains information on all births in Sweden since
1973 and pregnancy related factors;19 (2) the Swedish Total Population
Register covers demographic information on all Swedish inhabitants
since 1968;20 (3) the Multi-generation Register contains information
on biological and adoptive relationships on individuals living in Sweden
since 1961;21 (4) the National Patient Register (NPR) contains data on
inpatient diagnoses since 197322 and outpatient diagnoses since 2001
based on the International Classification of Diseases (ICD) in its sev-
enth (ICD-7; before 1969), eighth (ICD-8; 1969–1986), ninth (ICD-9;
1987–1996), and tenth (ICD-10; since 1997) revisions; (5) the Cause
of Death Register (CDR) contains information on all deaths since 1952
(complete coverage since 1961) based on ICD codes, including under-
lying and contributing causes of death;23 (6) the Prescribed Drug Reg-
ister contains information on all prescribed medications dispensed at
pharmacies in Sweden since July 2005, with drug identity defined using
HIGHLIGHTS
∙This study suggests that attention-deficit hyperactivity
disorder (ADHD) is associated with Alzheimer’s disease
(AD) and any dementia across generations.
∙The associations attenuated with decreasing genetic relat-
edness (parents >grandparents and uncles/aunts), sug-
gesting shared familial risk between ADHD and AD.
∙The increased familial risk for early-onset AD, associated
with ADHD, was higher than that for late-onset AD.
RESEARCH IN CONTEXT
1. Systematic review: The authors searched titles and
abstracts indexed on PubMed. Only a few studies
with limited sample size have investigated the asso-
ciation between attention-deficit/hyperactivity disorder
(ADHD) and Alzheimer’s disease (AD). Findings of avail-
able studies are inconsistent.
2. Interpretation: In a large population-based study, we
found that ADHD is associated with AD and any demen-
tia across generations. The associations attenuated with
decreasing genetic relatedness (parents >grandpar-
ents and uncles/aunts), suggesting shared familial risk
between ADHD and AD. The increased familial risk for
early-onset AD, associated with ADHD, was higher than
that for late-onset AD.
3. Future directions: Research on underlying risk factors
contributing to both ADHD and AD are warranted,
including molecular genetic and family studies aiming
to identify attributing pleiotropic genetic variants and
family-wide environmental risk factors. Our study calls
attention to advancing the understanding of ADHD and
cognitive decline in older age, and, if verified, warrants
investigation of treatment of ADHD to prevent or delay
the development of neurodegenerative diseases in indi-
viduals with ADHD and their family members.
Anatomical Therapeutic Chemical (ATC) codes;24 and (7) the Migration
Register records all migration in and out of Sweden.
2.2 Study population
We identified all individuals born in Sweden between 1980 and 2001.
We excluded stillbirths, individuals missing key demographic informa-
tion, and individuals who died or migrated before their 12th birth-
day, by linking to the Total Population Register, Medical Birth Register,
Migration Register, and CDR, respectively. Each individual (referred
ZHANG ET AL.3
TAB L E 1 Descriptive characteristics of the three relative cohorts
Type of individuals Variables Overall Female Male
Index persons No. of index persons 2,132,929 1,037,385 1,095,544
Age, median (IQR)a23 (18–28) 23 (18–28) 23 (18–28)
ADHD, no. (%) 68,379 (3.21) 24,226 (2.34) 44,153 (4.03)
Onset age of ADHD, median (IQR) 16 (12–20) 18 (14–22) 15 (11–19)
Parents No. of parents 2,293,961 1,146,865 1,147,096
Age, median (IQR)a53 (47–59) 51 (46–58) 54 (48–61)
Alzheimer’s disease, no. (%) 3042 (0.13) 980 (0.09) 2062 (0.18)
Any dementia, no. (%) 3792 (0.17) 1114 (0.10) 2678 (0.23)
Onset age of Alzheimer’s disease, median (IQR)b61 (56–67) 57 (53–62) 63 (57–70)
Onset age of any dementia, median (IQR)b59 (54–65) 56 (52–60) 61 (56–67)
Grandparents No. of grandparents 2,518,669 1,275,202 1,243,467
Age, median (IQR)a82 (72–91) 80 (71–90) 83 (73–93)
Alzheimer’s disease, no. (%) 171,732 (6.82) 99,454 (7.80) 72,278 (5.81)
Any dementia, no. (%) 197,843 (7.86) 111,584 (8.75) 86,259 (6.94)
Onset age of Alzheimer’s disease, median (IQR)b78 (73–83) 79 (74–83) 78 (73–82)
Onset age of any dementia, median (IQR)b77 (71–81) 77 (72–82) 76 (70–80)
Uncles/aunts No. of uncles/aunts 933,263 475,793 457,470
Age, median (IQR)a53 (47–60) 53 (47–60) 53 (47–60)
Alzheimer’s disease, no. (%) 1369 (0.15) 743 (0.16) 626 (0.14)
Any dementia, no. (%) 1697 (0.18) 852 (0.18) 845 (0.18)
Onset age of Alzheimer’s disease, median (IQR)b62 (57–67) 62 (56–66) 62 (57–67)
Onset age of any dementia, median (IQR)b60 (55–65) 60 (55–64) 60 (55–65)
Abbreviations: AD, Alzheimer’s disease; ADHD, attention-deficit/hyperactivity disorder; IQR, interquartile range.
aThe age of individuals by the end of study.
bTime of disease onset was estimated as 3 years before the first diagnosis, or 5 years before death (primary and contributing causes), whichever came first.
to as index persons) was linked to their biological relatives—parents,
grandparents, uncles and aunts—through the Medical Birth Register
and Multi-generation Register. All relatives were followed from the
date they turned 50 years of age to onset of dementia, date of first
migration, date of death, or December 31, 2013 (end of study follow-
up), whichever came first. Thus, we generatedthree cohorts of relatives
representing different levels of genetic relatedness: parents who share
50% of their segregating genes with index persons; grandparents who
share 25% of their segregating genes with index persons; uncles and
aunts who share 25% of their segregating genes with index persons. To
ensure the comparability between parents and uncles/aunts cohorts,
for each index person we included one uncle/aunt who had at least one
child and whose birth date was nearest to that of parents of index per-
sons.
2.3 Identification of ADHD
We used information from the NPR to identify ADHD diagnoses among
index persons. In sensitivity analyses, we additionally used information
on prescriptions of ADHD medication from the Prescribed Drug Regis-
ter for case identification (ICD and ATC codes in Table S1 in supporting
information). Our approach of using ADHD medication for case identi-
fication is consistent with prior research.25,26
2.4 Identification of AD and any dementia
We used validated diagnoses from the NPR and CDR to iden-
tify dementia cases among relatives of index persons, including
two definitions: AD and any dementia (including AD).27,28 In line
with previous studies, time of disease onset was estimated as
three years before the first diagnosis (ascertained in the NPR) or
5 years before death (ascertained in the CDR), whichever came
first.27–29 In sensitivity analyses, we additionally used information
on prescriptions of medications for AD, and the time of disease
onset was estimated as the date of the first relevant prescription
recorded in the Prescribed Drug Register, in keeping with previous
studies.29,30
2.5 Statistical analysis
Cox proportional hazards models were used to examine the associa-
tion between ADHD and dementia (AD or any dementia) in each of the
three relative cohorts (parents, grandparents, and uncles/aunts), with
attained age of relatives as the underlying timescale.31–33 Theriskof
having dementia in relatives of individuals with ADHD was compared
to the risk in relatives of individuals without ADHD, and hazard ratios
4ZHANG ET AL.
FIGURE 1 Flowchart of the inclusion of index persons and their biological relatives
(HRs) were estimated with 95% confidence intervals (CIs). Robust
standard errors were used to account for non-independence of data
due to the repeat of individuals in index person–relative pairs. In
adjusted models, HRs were adjusted for birth year of index persons,
birth year of relatives, sex of index persons, and sex of relatives. The
analyses were further stratified by sex of relatives and sex of index
persons.
To test whether the association is moderated by onset age of demen-
tia, HRs were estimated for early-onset dementia (onset before 65
years) and late-onset dementia (onset after 65 years).34 We fitted a
Cox model allowing the HRs to be different for before and after age
65, corresponding to early- and late-onset dementia, in each relative
cohort.
Sensitivity analyses were performed to examine whether the results
were robust with different case and cohort identifications. First, to
improve the coverage of ADHD and AD, we additionally used prescrip-
tions of ADHD and AD medications to identify individuals with ADHD
or AD. Second, we stratified the estimates by birth year of index per-
sons (1980–1989 and 1990–2001). This was done to assess potential
bias due to differences in register coverage and follow-up time for rel-
atives. Third, we additionally adjusted for ADHD status in the relatives
in each of the three relative cohorts. If the HRs remain significant after
adjustment, the contribution of common familial risk factors to ADHD
and AD would be further supported (see further explanation in Figure
S2 in supporting information).
Data management was performed using SAS version 9.4 (SAS Insti-
tute, Inc.) and all analyses were performed using R version 3.6.1.
3RESULTS
We identified 2,224,189 individuals born between 1980 and 2001
from the Medical Birth Register, and 2,132,929 individuals were iden-
tified as eligible index persons after applying the exclusion criteria (Fig-
ure 1). After linking index persons to their biological relatives, the eli-
gible study cohorts contained 2,293,961 parents, 2,518,669 grand-
parents, and 933,263 uncles/aunts, which created 4,246,182 index
person–parent pairs, 7,548,861 index person–grandparent pairs, and
1,838,520 index person–uncle/aunt pairs. Among the index persons,
68,379 (3.21%) were diagnosed with ADHD (44,153 [4.03%] in men,
24,226 [2.34%] in women, Table 1). The relatives were followed for
a median of 8.0 years in parents, 25.0 years in grandparents, and 8.5
years in uncles/aunts. By the end of follow-up, 3042 (0.13%) of parents,
171,732 (6.82%) of grandparents, and 1369 (0.15%) of uncles/aunts
had a diagnosis of AD. The numbers for any dementia were 3792
(0.17%) for parents, 197,843 (7.86%) for grandparents, and 1697
(0.18%) for uncles/aunts (Table 1).
Parents of index persons with ADHD had an increased risk of
AD compared to the parents of index persons without ADHD (HR
1.55, 95% CI 1.26–1.89; Table 2). The associations attenuated with
decreasing genetic relatedness, that is, the association with AD in
grandparents attenuated (1.11, 1.08–1.13), and the association in
uncles/aunts was similar to grandparents but not statistically signifi-
cant (1.15, 0.85–1.56). A similar pattern was observed for any demen-
tia with an increased risk in parents (1.34, 1.11–1.63) and grandpar-
ents (1.10, 1.08–1.12), and a nonsignificant association in uncles/aunts
ZHANG ET AL.5
TAB L E 2 Association between ADHD and Alzheimer’s disease and any dementia in three relative cohorts
Relaves of index
persons
No. of event
(ADHD 0)
Person-years
(ADHD 0)
No. of event
(ADHD 1)
Person-years
(ADHD 1)
Crude hazard rao Adjusted hazard rao
(95% CI)
(95% CI)
Alzheimer’s disease
Parents 3569 17,398,062 110 365,803 1.51 (1.24–1.85) 1.55 (1.26–1.89)
1.77 (1.23–2.54)1.66 (1.16–2.38)141,129357,175,4751128Mother
Father 2441 4,338,978 75 96,100 1.44 (1.13–1.84) 1.46 (1.15–1.86)
1.11 (1.08–1.13)1.08 (1.06–1.11)4,699,59611,665163,361,512464,503Grandparents
Grandmother 257,621 83,608,787 6323 2,392,283 1.09 (1.06–1.12) 1.12 (1.09–1.16)
1.08 (1.05–1.12)1.08 (1.05–1.11)2,307,313534279,752,726206,882Grandfather
Uncles/aunts 1976 8,811,647 46 194,967 1.12 (0.83–1.51) 1.15 (0.85–1.56)
1.13 (0.75–1.72)1.09 (0.72–1.67)98,867254,472,6691091Aunt
Uncle 885 4,338,978 21 96,100 1.15 (0.75–1.77) 1.18 (0.77–1.82)
Anydemena
Parents 4467 17,384,751 120 365,526 1.31 (1.08–1.59) 1.34 (1.11–1.63)
1.45 (1.00–2.13)1.30 (0.89–1.90)98,830324,469,5971266Mother
Father 3201 4,334,740 88 96,017 1.29 (1.03–1.61) 1.31 (1.05–1.63)
1.10 (1.08–1.12)1.05 (1.03–1.07)4,665,58813,590162,075,129537,135Grandparents
Grandmother 289,315 82,952,329 7114 2,375,513 1.05 (1.03–1.08) 1.11 (1.08–1.14)
1.08 (1.05–1.11)1.05 (1.03–1.08)2,290,075647679,122,800247,820Grandfather
Uncles/aunts 2452 8,804,337 51 194,847 0.98 (0.74–1.30) 1.04 (0.79–1.39)
0.96 (0.63–1.48)0.91 (0.59–1.40)98,830244,469,5971241Aunt
Uncle 1211 4,334,740 27 96,017 1.05 (0.72–1.53) 1.13 (0.77–1.65)
Note: No. of events and person-years are shown for index person–relative pairs (not individuals); adjusted hazard ratios were derived from Cox proportional
models adjusted for birth year of index persons, birth year of relatives, sexof index persons, and sex of relatives.
Abbreviation: ADHD, attention-deficit/hyperactivity disorder; CI, confidence interval.
(1.04, 0.79–1.39). Kaplan-Meier curves of time to AD/any dementia
development for relatives of indexpersons with and without ADHD are
shown in Figure S1 in supporting information.
The estimates stratified by sex of the relatives showed that moth-
ers of index persons with ADHD had higher risks for AD (1.77, 1.23–
2.54) than fathers (1.46, 1.16–1.84), though the difference was not sta-
tistically significant (Table 2). Analysis stratified by sex of index per-
son showed similar estimates in parents for male index persons (1.53,
1.18–1.99) and female index persons (1.55,1.16–2.09; Table S2 in sup-
porting information). When considering age of onset, the risk of hav-
ing early-onset AD in relatives of index persons with ADHD was higher
than the risk of having late-onset AD in relatives of index persons with
ADHD (Table 3). In parents, the association with early-onset AD (1.69,
1.34–2.13) was stronger than the association with late-onset AD (1.20,
0.82–1.77). This was also the case with grandparents, that is, the asso-
ciation with early-onset AD (1.27, 1.19–1.36) was stronger than the
association with late-onset AD (1.09, 1.07–1.11). Similar patterns were
observed in uncles/aunts, and for the associations with any dementia.
Sensitivity analyses further included information on prescribed
medications of ADHD and AD for case identification, increased sta-
tistical power generated robust results, that is, the coaggregation of
ADHD and AD in index person–uncle/aunt pairs became significant
(1.28, 1.02–1.61; Table 4). We observed similar results in the subco-
horts of index persons born between 1980 and 1989 and between
1990 and 2001, although parents in the subgroup of index persons
born between 1990 and 2001 were not significant due to low number
of AD cases. Further, the results adjusting for ADHD status of relatives
were similar to the main analyses across relative types.
4DISCUSSION
To the best of our knowledge, this is the first study to explore the
association of ADHD,a neurodevelopmental disorder, with AD and
any dementia, neurodegenerative diseases, across generations. We
observed that relatives of individuals with ADHD had an increased
risk of developing AD and any dementia compared to relatives of indi-
viduals without ADHD. The associations attenuated with decreasing
genetic relatedness (parents >grandparents and uncles/aunts). The
risk of having early-onset AD associated with ADHD was higher than
that for late-onset AD.
The observed familial coaggregation of ADHD in index persons
and AD in relatives can be explained by several potential mechanisms
or their combination. One mechanism is that the observed familial
6ZHANG ET AL.
TAB L E 3 Association between ADHD and Alzheimer’s disease and any dementia stratified by onset age of dementia
Relaves of index
Age at onsetpersons
No. of event
(ADHD =0)
Person-years
(ADHD =0)
No. of event
(ADHD =1)
Person-years Adjusted hazard rao
(95% CI)
(ADHD =1)
Alzheimer’s disease
Parents Early-onset 2562 16,574,683 85 349,912 1.69 (1.34–2.13)
1.20 (0.82–1.77)15,89225823,3791007Late-onset
Grandparents Early-onset 29,950 92,295,850 1147 2,911,247 1.27 (1.19–1.36)
1.09 (1.07–1.11)1,788,34910,51871,065,662434,553Late-onset
Uncles/aunts Early-onset 1431 8,239,251 34 183,840 1.16 (0.81–1.66)
1.13 (0.65–1.98)11,12712572,396545Late-onset
Anydemena
1.40 (1.13–1.74)349,7339716,565,8473489Early-onsetParents
Late-onset 978 818,904 23 15,793 1.15 (0.77–1.70)
1.35 (1.29–1.42)2,907,230203592,199,74851,855Early-onsetGrandparents
Late-onset 485,280 818,904 11,555 15,793 1.06 (1.04–1.08)
1.09 (0.80–1.48)183,757438,234,3831971Early-onsetUncles/aunts
Late-onset 481 569,953 8 11,090 0.86 (0.44–1.71)
Note: No. of events and person-years are shown for index person–relative pairs (not individuals); adjusted hazard ratios were derived from Cox proportional
models adjusted for birth year of index persons, birth year of relatives, sexof index persons, and sex of relatives.
Abbreviation: ADHD, attention-deficit/hyperactivity disorder; CI, confidence interval.
TAB L E 4 Sensitivity analyses of familial coaggregation between ADHD and Alzheimer’s disease
Analysis Type of relatives No. of pairs No. of events Adjusted hazard ratio (95% CI)
Including medication in addition to diagnosis
and cause of death for case identification
Parents 4,246,182 5951 1.44 (1.21–1.71)
Grandparents 7,548,861 571,025 1.09 (1.07–1.11)
Uncles/aunts 1,838,520 2849 1.28 (1.02–1.61)
Subgroup of index persons born 1980–1989 Parents 1,903,623 4128 1.56 (1.25– 1.96)
Grandparents 3,437,638 318,249 1.08 (1.05–1.11)
Uncles/aunts 842,114 1755 1.10 (0.76–1.58)
Subgroup of index persons born 1990–2001 Parents 2,342,559 982 1.48 (0.96–2.28)
Grandparents 4,111,223 193,608 1.13 (1.10– 1.16)
Uncles/aunts 996,406 360 1.30 (0.78–2.15)
Additionally adjust for ADHD diagnosis in the
relatives of the index person
Parents 4,246,182 4400 1.53 (1.25–1.87)
Grandparents 7,548,861 510,759 1.11 (1.08–1.13)
Uncles/aunts 1,838,520 2115 1.15 (0.85–1.56)
Note: No. of events are shown for index person-relative pairs (not individuals); adjusted hazard ratios were derived from Cox proportional models adjusted
for the birth year of index persons, the birth year of relatives, sex of index persons, and sex of relatives.
Abbreviation: ADHD, attention-deficit/hyperactivity disorder; CI, confidence interval.
coaggregation is attributable to familial risk factors shared by the two
conditions within families, including pleiotropic genetic variants and
family-wide environmental factors affecting both conditions. Although
the largest GWAS studies on ADHD and AD to date have failed to
detect any genetic variant in common,16,17 there have been studies
suggesting that the genes SORCS2 and SORCS3 may be implicated in
both ADHD and AD, with amyloid precursor protein (APP) processing,
neuronal development, and plasticity being altered.35,36 Furthermore,
as common genetic variants only explain a moderate proportion of the
heritability of both disorders,16,37 it is possible that there exist other
ZHANG ET AL.7
genetic factors, including rare genetic variants, contributing to both
traits.38 Apart from pleiotropic genetic factors shared within families,
family-wide environmental risk factors such as familial socioeconomic
status may influence the development of both AD39 and ADHD.40
The shared familial risk factors would not only explain the observed
coaggregation across generations, but also indicate an individual-level
association between ADHD and AD. A graphical representation of the
relationship is explained in Figure S2 (path a).
ADHD in the index persons and AD in their relatives could also be
associated through the mediation of genetic or/and environmental risk
factors of ADHD. It is worth noting that these risk factors may rep-
resent factors unique to ADHD (independent of AD), which in turn
increase the risk of ADHD in both the index persons and their relatives,
and further increase the risk of AD via adverse health outcomes trig-
gered by ADHD (path b in Figure S2). For example, ADHD in both chil-
dren and adults has been shown to be associated with obesity/being
overweight,41 while midlife obesity has been suggested to increase
the risk of dementia later in life.42 This explanation also indicates an
individual-level association of two conditions through adverse health
outcomes of ADHD, and is supported by the results from the US study
using hospitalization discharge data, in which a significant individual-
level association between ADHD and AD was observed in the unad-
justed model, while the association disappeared after adjusting for dia-
betes and obesity.6
A third possible mechanism could be implicated by the direct effect
of ADHD in the relatives on ADHD in the index person, as well as
on their own risk of AD.43 For example, poor parenting style of indi-
viduals with ADHD may influence the behavior of one’s child, which
may explain the observed coaggregation in the parent cohort (path c in
Figure S2). However, the cross-generation associations remained after
adjusting for ADHD status in relatives (Table 4), suggesting that such an
effect, if it exists, is unlikely to be the main explanation of the observed
familial coaggregation. Another possible mechanism is indicated by the
direct effect of ADHD in the index person on the development of AD in
his/her relatives. Although there is no direct evidence for this, previous
studies have shown that parents of a child with ADHD may feel power-
less and experience extensive psychological distress.44 This could pos-
sibly contribute to the association in parents (path d in Figure S2), but
is less likely to explain the association in more distant relatives (i.e.,
grandparents and uncles/aunts).
In addition to the main findings, our study found that the familial
coaggregation of ADHD with early-onset AD was stronger than that
with late-onset AD. This could be due to the differences in the etiology
of early-onset versus late-onset AD, as early-onset AD is more strongly
genetically influenced (heritability estimated around92% to 100%45)
than late-onset AD (heritability estimated around61% to 79%45,46).
Previous research findings have suggested that early-onset AD may
have a stronger load of autosomal dominant and recessive causes,45
while late-onset dementia is more of a polygenetic disease.47 Thus,
there might be different genetic factors involved in the association of
ADHD with early-onset and late-onset dementia, which requires fur-
ther investigation.
Our study highlights the importance of advancing the understand-
ing of ADHD in older adults. With further triangulation with other stud-
ies, ADHD could be considered a potential modifiable risk factor for AD
and dementia. A recent meta-analysis found that a considerable num-
ber of older adults presented ADHD symptoms, and only less than half
of older adults with clinically diagnosed ADHD received treatment.48
Research has shown that ADHD medications are effective in reduc-
ing ADHD symptoms as well as adverse outcomes of ADHD, such as
substance abuse, physical injury, and low educational achievement,49
which are well-documented risk factors for AD. Further studies are
warranted to examine whether ADHD medications could alleviate the
risk of AD associated with ADHD.
This is, to our knowledge, the first study explored the familial asso-
ciation of ADHD with AD and any dementia using a large population-
based sample. The use of multi-generational design enables follow-up
from younger elderly to older elderly in different relatives, and pro-
vides insight into the genetic and environmental contribution to the
associations of ADHD with AD and any dementia. The diagnoses of
ADHD and AD were made separately in the index person and rela-
tives, preventing biases from symptom misclassification or preconcep-
tions of patients or clinicians. Our results should also be interpreted in
light of several limitations. First, despite the multi-generation design,
we were only able to follow most of the parents and uncles/aunts until
their sixties; however, the onset of dementia usually peaks around
80 years of age.34 Nonetheless, results from the subcohort of index
persons born between 1980 and 1989 showed similar estimates with
those born between 1990 and 2001, suggesting the length of follow-
up did not bias the estimates. Second, we were not able to exam-
ine the association with specific dementia subtypes other than AD
(e.g., vascular dementia, dementia with Lewy bodies, and frontotem-
poral dementia) due to insufficient statistical power. Future studies
are needed to explore whether the associations are differential across
subtypes as well as if there exist underlying biological mechanisms
specific to associations of certain subtypes. Third, although prior val-
idation studies have reported the diagnoses of dementia in the NPR
and CDR with specificity of 99%, the sensitivity is only 63%.27,28 Such
misclassifications of dementia cases would attenuate the associations
toward the null and thus our estimates are likely to be conservative.
Nonetheless, we additionally used information on prescriptions for AD
in case identification in sensitivity analysis and generated consistent
results. Fourth, using anti-dementia drugs for AD case identification
in the sensitivity analysis would misclassify some cases, because anti-
dementia drugs are also prescribed for other dementias, for exam-
ple, Lewy body dementia and Parkinson’s disease dementia.50 How-
ever, this misclassification should be nondifferential and not substan-
tially influence the association between ADHD and AD, as Lewy body
dementia and Parkinson’s disease dementia constitute only a small pro-
portion of total dementia cases.50 Finally, although the prevalence of
ADHD diagnosis and ADHD medication prescription in Sweden is simi-
lar to many European countries, it is lower than other countries, such as
the United States,51,52 meaning individuals with ADHD in Sweden may
represent more severe cases than those countries. Thus, replications in
8ZHANG ET AL.
other countries are needed to examine to what extent our results gen-
eralize to other settings.
5CONCLUSIONS
We found that ADHD coaggregated with AD and any dementia within
families, and the strength of association attenuated with decreasing
degree of genetic relatedness. Molecular genetic and family studies
aiming to identify pleiotropic genetic variants and family-wide envi-
ronmental risk factors contributing to both conditions are warranted.
GWAS of ADHD and AD based on a larger sample size could provide
the potential for unraveling shared genetic mechanisms between the
disorders through linkage disequilibrium score regression and network
analyses.53 In addition, our study highlights the importance of advanc-
ing the understanding of ADHD and cognitive decline in older age and,
if verified, calls for investigation of early-life psychiatric prevention on
the development of neurodegenerative diseases in older age.
ACKNOWLEDGMENTS
The study was supported by grants from the Swedish Council for
Health, Working Life and Welfare (2019-00176; 2019-01172), the
Swedish Research Council (2018-02599), the Swedish Brain Founda-
tion (FO2018-0273), the European Union’s Horizon 2020 research and
innovation programme under the Marie Skłodowska-Curie (754285),
the Fredrik & Ingrid Thurings Stiftelse, and the Karolinska Institutet
Research Foundation.
CONFLICTS OF INTEREST
EDR has served as a consultant for Shire Sweden AB; HL has served
as a speaker for Evolan Pharma and Shire/Takeda and has received
research grants from Shire/Takeda, all outside the submitted work. All
other authors have no competing interests.
ORCID
Le Zhang https://orcid.org/0000-0002-7290-5103
REFERENCES
1. Faraone SV, Asherson P, Banaschewski T, et al. Attention-
deficit/hyperactivity disorder. Nat Rev Dis Primers. 2015;1:15020.
2. Asherson P, Buitelaar J, Faraone SV, Rohde LA. Adult attention-
deficit hyperactivity disorder: key conceptual issues. Lancet Psychiatry.
2016;3(6):568-578.
3. Simon V, Czobor P, Balint S, Meszaros A, Bitter I. Prevalence and cor-
relates of adult attention-deficit hyperactivity disorder: meta-analysis.
Br J Psychiatry. 2009;194(3):204-211.
4. Prince M, Bryce R, Albanese E, Wimo A, Ribeiro W, Ferri CP. The
global prevalence of dementia: a systematic review and metaanalysis.
Alzheimers Dement. 2013;9(1):63-75. e62.
5. Livingston G, Sommerlad A, Orgeta V, et al. Dementia prevention,
intervention, and care. Lancet. 2017;390(10113):2673-2734.
6. Fluegge K, Fluegge K. Antecedent ADHD, dementia, and metabolic
dysregulation: a U.S. based cohort analysis. Neurochem Int.
2018;112:255-258.
7. Golimstok A, Rojas J, Romano M, Zurru M, Doctorovich D, Cristiano E.
Previous adult attention-deficit and hyperactivity disorder symptoms
and risk of dementia with Lewy bodies: a case–control study. Eur J Neu-
rol. 2011;18(1):78-84.
8. Tzeng N-S, Chung C-H, Lin F-H, et al. Risk of dementia in adults with
ADHD: a nationwide, population-based cohort study in Taiwan. J Atten
Disord. 2019;23(9):995-1006.
9. Rydell M, Lundström S, Gillberg C, Lichtenstein P, Larsson H. Has
the attention deficit hyperactivity disorder phenotype become more
common in children between 2004 and 2014? Trends over 10 years
from a Swedish general population sample. J Child Psychol Psychiatry.
2018;59(8):863-871.
10. Callahan BL, Bierstone D, Stuss DT, Black SE. Adult ADHD: risk factor
for dementia or phenotypic mimic?. Front Ag ing Neuros ci. 2017;9:260.
11. Norton S, Matthews FE, Barnes DE, Yaffe K, Brayne C. Potential for
primary prevention of Alzheimer’s disease: an analysis of population-
based data. Lancet Neurol. 2014;13(8):788-794.
12. Nigg JT. Attention-deficit/hyperactivity disorder and adverse health
outcomes. Clin Psychol Rev. 2013;33(2):215-228.
13. Spencer TJ, Faraone SV, Tarko L, McDermott K, Biederman J.
Attention-deficit/hyperactivity disorder and adverse health outcomes
in adults. J Nerv Ment Dis. 2014;202(10):725-731.
14. Fischer BL, Gunter-Hunt G, Steinhafel CH, Howell T. The identifica-
tion and assessment of late-life ADHD in memory clinics. J Atten Disord.
2012;16(4):333-338.
15. Ivanchak N, Fletcher K, Jicha GA. Attention-deficit/hyperactivity dis-
order in older adults: prevalence and possible connections to mild cog-
nitive impairment. Curr Psychiatry Rep. 2012;14(5):552-560.
16. Demontis D, Walters RK, Martin J, et al. Discoveryof the first genome-
wide significant risk loci for attention deficit/hyperactivity disorder.
Nat Genet. 2019;51(1):63-75.
17. Jansen IE, Savage JE, Watanabe K, et al. Genome-wide meta-analysis
identifies new loci and functional pathways influencing Alzheimer’s
disease risk. Nat Genet. 2019;51(3):404-413.
18. Ludvigsson JF, Otterblad-Olausson P, Pettersson BU, Ekbom A. The
Swedish personal identity number: possibilities and pitfalls in health-
care and medical research. Eur J Epidemiol. 2009;24(11):659-667.
19. National Board of Health and Welfare. The Swedish medical birth regis-
ter: a summary of content and quality. Stockholm, Sweden: National
Board of Health and Welfare; 2003.
20. Ludvigsson JF, Almqvist C, Bonamy A-KE, et al. Registers of the
Swedish total population and their use in medical research. Eur J Epi-
demiol. 2016;31(2):125-136.
21. Ekbom A. The Swedish Multi-generation Register. Methods Mol Biol.
2011;675:215-220.
22. Ludvigsson JF, Andersson E, Ekbom A, et al. External review and val-
idation of the Swedish national inpatient register. BMC Public Health.
2011;11(1):450.
23. Brooke HL, Talbäck M, Hörnblad J, et al. The Swedish cause of death
register. Eur J Epidemiol. 2017;32(9):765-773.
24. Wettermark B, Hammar N, Fored CM, et al. The new Swedish Pre-
scribed Drug Register–opportunities for pharmacoepidemiological
research and experience from the first six months. Pharmacoepidemiol
Drug Saf. 2007;16(7):726-735.
25. Ghirardi L, Brikell I, Kuja-Halkola R, et al. The familial co-aggregation
of ASD and ADHD: a register-based cohort study. Mol Psychiatry.
2018;23(2):257-262.
26. Sun S, Kuja-Halkola R, Faraone SV, et al. Association of psychiatric
comorbidity with the risk of premature death among children and
adults with attention-deficit/hyperactivity disorder. JAMA Psychiatry.
2019;76(11):1141-1149.
27. Rizzuto D, Feldman AL, Karlsson IK, Dahl Aslan AK, Gatz M, Pedersen
NL. Detection of dementia cases in two Swedish Health Registers: a
validation study. J Alzheimers Dis. 2018;61(4):1301-1310.
28. Jin Y-P, Gatz M, Johansson B, Pedersen NL. Sensitivity and speci-
ficity of dementia coding in two Swedish disease registries. Neurology.
2004;63(4):739-741.
ZHANG ET AL.9
29. Eriksson LI, Lundholm C, Narasimhalu K, et al. Hospitalization,
surgery, and incident dementia. Alzheimers Dement. 2019;15(4):534-
542.
30. Jørgensen JT, Hansen J, Westendorp RG, et al. Shift work and inci-
dence of dementia: a Danish Nurse Cohort study. Alzheimers Dement.
2020;16(9):1268-1279.
31. Hernán MA. The hazards of hazard ratios. Epidemiology. 2010;21(1):13.
32. Tolles J, Lewis RJ. Time-to-event analysis. JAMA. 2016;315(10):1046-
1047.
33. Stensrud MJ, Hernán MA. Why test for proportional hazards?. JAMA.
2020;323(14):1401-1402.
34. McMurtray A, Clark DG, Christine D, Mendez MF. Early-onset demen-
tia: frequency and causes compared to late-onset dementia. Dement
Geriatr Cogn Disord. 2006;21(2):59-64.
35. Reitz C, Tokuhiro S, Clark LN, et al. SORCS1 alters amyloid precursor
protein processing and variants may increase Alzheimer’s disease risk.
Ann Neurol. 2011;69(1):47-64.
36. Alemany S, Ribasés M, Vilor-Tejedor N, et al. New suggestive genetic
loci and biological pathways for attention function in adult attention-
deficit/hyperactivity disorder. Am J Med Genet B Neuropsychiatr Genet.
2015;168(6):459-470.
37. Ridge PG, Hoyt KB, Boehme K, et al. Assessment of the genetic vari-
ance of late-onset Alzheimer’s disease. Neurobiol Aging. 2016;41:200.
e213-200. e220.
38. Manolio TA, Collins FS, Cox NJ, et al. Finding the missing heritability of
complex diseases. Nature. 2009;461(7265):747-753.
39. Sattler C, Toro P, Schönknecht P, Schröder J. Cognitive activity, educa-
tion and socioeconomic status as preventive factors for mild cognitive
impairment and Alzheimer’s disease. Psychiatry Res. 2012;196(1):90-
95.
40. Russell AE, Ford T, Williams R, Russell G. The association between
socioeconomic disadvantage and attention deficit/hyperactivity dis-
order (ADHD): a systematic review. Child Psychiatry Hum Dev.
2016;47(3):440-458.
41. Cortese S, Moreira-Maia CR, St Fleur D, Morcillo-Penalver C, Rohde
LA, Faraone SV. Association between ADHD and obesity: a systematic
review and meta-analysis. Am J Psychiatry. 2016;173(1):34-43.
42. Fitzpatrick AL, Kuller LH, Lopez OL, et al. Midlife and late-life obe-
sity and the risk of dementia: cardiovascular Health Study. Arch Neurol.
2009;66(3):336-342.
43. D’Onofrio BM, Lahey BB, Turkheimer E, Lichtenstein P. Critical need
for family-based, quasi-experimental designs in integrating genetic
and social science research. Am J Public Health. 2013;103 Suppl 1(Suppl
1):S46-55.
44. Glatz T, Stattin H, Kerr M. Parents’ reactions to youths’ hyperac-
tivity, impulsivity, and attention problems. J Abnorm Child Psychol.
2011;39(8):1125-1135.
45. Wingo TS, Lah JJ, Levey AI, Cutler DJ. Autosomal recessive causes
likely in early-onset Alzheimer disease. Arch Neurol. 2012;69(1):59-64.
46. Gatz M, Reynolds CA, Fratiglioni L, et al. Role of genes and envi-
ronments for explaining Alzheimer disease. Arch Gen Psychiatry.
2006;63(2):168-174.
47. Jiang T, Yu J-T, Tian Y, Tan L. Epidemiology and etiology of Alzheimer’s
disease: from genetic to non-genetic factors. Curr Alzheimer Res.
2013;10(8):852-867.
48. Dobrosavljevic M, Solares C, Cortese S, Andershed H, Larsson
H. Prevalence of attention-deficit/hyperactivity disorder in older
adults: a systematic review and meta-analysis. Neurosci Biobehav Rev.
2020;118:282-289.
49. Chang Z, Ghirardi L, Quinn PD, Asherson P, D’Onofrio BM, Lars-
son H. Risks and benefits of attention-deficit/hyperactivity disorder
medication on behavioral and neuropsychiatric outcomes: a qualita-
tive review of pharmacoepidemiology studies using linkedprescription
databases. Biol Psychiatry. 2019;86(5):335-343.
50. Johnell K, Religa D, Eriksdotter M. Differences in drug therapy
between dementia disorders in the Swedish dementia registry: a
nationwide study of over 7,000 patients. Dement Geriatr Cogn Disord.
2013;35(5-6):239-248.
51. Faraone SV, Sergeant J, Gillberg C, Biederman J. The worldwide
prevalence of ADHD: is it an American condition?. World Psychiatry.
2003;2(2):104.
52. Raman SR, Man KK, Bahmanyar S, et al. Trends in attention-
deficit hyperactivity disorder medication use: a retrospective obser-
vational study using population-based databases. Lancet Psychiatry.
2018;5(10):824-835.
53. Patel H, Dobson RJ, Newhouse SJ. A meta-analysis of Alzheimer’s
disease brain transcriptomic data. J Alzheimers Dis. 2019;68(4):1635-
1656.
SUPPORTING INFORMATION
Additional supporting information may be found in the online version
of the article at the publisher’s website.
How to cite this article: Zhang Le, Rietz EDu, Kuja-Halkola R,
et al. Attention-deficit/hyperactivity disorder and Alzheimer’s
disease and any dementia: A multi-generation cohort study in
Sweden. Alzheimer’s Dement. 2021;1-9.
https://doi.org/10.1002/alz.12462
