J Youth Adolescence (2011) 40:12881301
DOI 10.1007/s10964-011-9667-1
EMPIRICAL RESEARCH
Pubertal Timing and Substance Use in Middle Adolescence:
A 2-Year Follow-up Study
Riittakerttu Kaltiala-Heino Anna-Maija Koivisto
Mauri Marttunen Sari Frojd
Received: 29 August 2010 / Accepted: 18 April 2011 / Published online: 30 April 2011
Springer Science+Business Media, LLC 2011
Abstract Earlier research has associated early puberty
with emotional and behavioral symptoms particularly
among girls, while among boys, findings have been con-
tradictory as to whether risks are associated with early or
late pubertal timing. We studied the association between
pubertal timing and substance use behaviors in middle
adolescence in a 2-year follow up study of 2,070 (mean
age 15.5 years, SD 0.36; 56.4% females) Finnish adoles-
cents. Pubertal timing was measured by age at menarche/
oigarche. Eleven years or less was classified as early,
1213 years as normative and 14 years or later as late
pubertal timing. Substance use behaviors were elicited by a
number of questions related to alcohol use patterns,
smoking and cannabis use. As factors that could explain the
association between pubertal timing and substance use, we
studied depressive symptoms, delinquency and aggression,
and parental monitoring. In boys, all these substance use
behaviors were the more common the earlier the puberty
and the associations persisted at age 17. Among girls, early
pubertal timing was similarly associated with substance use
behaviors at age 15, but no longer at age 17. The asso-
ciations between pubertal timing and substance use
R. Kaltiala-Heino (&)
Medical School, University of Tampere, Tampere, Finland
e-mail: merihe@uta.fi
R. Kaltiala-Heino
Department of Adolescent Psychiatry, Tampere University
Hospital, Tampere, Finland
A.-M. Koivisto
S. Frojd
Tampere School of Public Health, University of Tampere,
Tampere, Finland
M. Marttunen
National Institute for Health and Welfare, Helsinki, Finland
123
behaviors persisted when symptom dimensions and
parental monitoring were added into the models. Early
puberty is a risk factor for substance use particularly
among boys. Among girls, the impact of pubertal timing
already tempers off during adolescence.
Keywords Pubertal timing
Substance use
Depression

Delinquency
Aggression
Parental monitoring
Introduction
With the transition from childhood to adolescence, psy-
chological symptoms, problem behaviors and both inter-
nalizing and externalizing psychiatric disorders increase in
prevalence (Buchanan et al. 1992). The hormonal changes
brought about by puberty offer a biological explanation, as,
for example, depression in girls has been associated with
increased estradiol and testosterone levels Angold et al.
(1999), and aggressive behavior with testosterone in boys
and dehydroepiandrosterone and dehydroepiandrosterone
sulphate in girls (Graber et al. 2006). However, hormonal
changes alone are an insufficient explanation for adoles-
cents psychological symptoms and problem behaviors
(Buchanan et al. 1992). Along with the physical changes,
puberty brings about major psychological and social
changes that are likely to cause stress and susceptibility to
psychiatric disorders.
Among girls, early pubertal timing, i.e. that the girl
matures physically earlier than peers, has been shown to be
associated with a variety of psychiatric disorders such as
depression, anxiety disorders, eating disorders, conduct
disorders and substance use disorders, whereas late
pubertal timing appears to be protective against these dis-
orders (Mendle et al. 2007). Three hypotheses have been
J Youth Adolescence (2011) 40:12881301
presented about why pubertal timing per se is a risk factor
(Ge et al. 2003). The early timing (stage termination)
hypothesis suggests that girls entering puberty earlier than
their peers experience the biological changes too early,
without having had time to gain the necessary cognitive
and social skills to cope with the changes. The off-time
hypothesis assumes that entering puberty at a different rate
from the majority of peers, and thus without the support of
peers facing the same challenges, is stressful and causes
vulnerability. In this view, both early and late timing
should emerge as a risk factor for mental health problems.
According to the stressful change hypothesis, the stress and
susceptibility to disorders emerge from being in a phase of
change, regardless of when it occurs, and the impact of
timing is an artefact that vanishes in the long run when the
whole cohort has gone through puberty (Ge et al. 2003).
Most of the research on the association between pubertal
timing and psychiatric disorders among girls is cross-
sectional and cannot claim longitudinal associations.
However, Ge et al. (2003) found that the impact of early
pubertal timing on depression in girls persisted across early
adolescence. In a study from adolescence to adulthood, the
disadvantageous impact of early pubertal timing persisted
over years. Girls with early pubertal timing later presented
with excessive lifetime and current internalizing and
externalizing disorders in early adulthood (Graber et al.
2004). To summarize, previous research suggests that early
pubertal timing in girls is associated with both internalising
and externalising psychiatric symptoms and disorders, and
the associations are likely to persist well beyond the period
when pubertal changes are completed.
Among boys, less research on pubertal timing has been
carried out, and the findings have been less consistent.
Early puberty has been associated with better self-esteem,
positive self-image, popularity and achievement in boys
(Taga et al. 2006). Late maturation has been associated
with problematic psychosocial development in boys
(Graber et al. 1997, 2004). Hormonal influences of course
differ by gender, but the stressfulness of change and its
timing also could be related to psychosocial aspects. Par-
ticularly early maturing girls may find the bodily changes
with increasing adipose tissue undesirable and the social
role changes brought about by bodily changes frightening,
whereas boys may find the more mature masculine body
and role desirable, with stress emerging due to being late to
enter these changes. However, some studies have associ-
ated early pubertal timing with both internalizing and
externalizing disorders among boys (Kaltiala-Heino et al.
2003; Ge et al. 2001; Ge et al. 2003; Ge et al. 2006).
Follow-up studies are few and differ particularly as to
follow-up time. Ge et al. (2001) noted that the association
between early maturation and internalizing and external-
izing disorders in boys persists in follow-up during
1289
adolescence. Graber et al. (2004) suggested that particu-
larly young adult males who had been late maturers pre-
sented with serious psychopathology in young adulthood
(even though early maturers also differed negatively from
normative maturers in adulthood in their study). Taga et al.
(2006) presented findings that early maturing boys adjusted
better in middle adulthood, for example regarding carreer
success, marital satisfaction and health behaviors. The
findings on associations between pubertal timing and
internalizing and externalising symptoms and disorders
among boys are thus inconclusive, and neither is it known
whether possible early influences of pubertal timing persist
over time in boys.
Substance use behaviors increase across adolescence,
and progress from experimentations to regular use (DeWitt
et al. 2000; Buckstein et al. 2005; Chung et al. 2005;
Costello et al. 2007). The adolescent brain is more vulner-
able to the negative influences of substance use (Carpenter-
Hyland and Chandler 2007; Clark et al. 2008), and the
earlier the initiation of experiments and use, the greater is
the risk for substance abuse and substance use disorders
(Anthony and Petronis 1995; Breslau et al. 1993; DeWitt
et al. 2000; Lubman et al. 2007). Drinking, smoking and
substance use are more common among adolescents in more
advanced pubertal stages (Costello et al. 2007). Early
pubertal timing seems to increase the risk for substance use
during the adolescent years in both sexes (Dick et al. 2000;
Martin et al. 2002; Kaltiala-Heino et al. 2003; Ge et al.
2006; Costello et al. 2007). Some research has demon-
strated that early pubertal timing is associated with
increased substance use behaviors not only during the ado-
lescent transition years but also in adulthood (Hayatbakhsh
et al. 2009; Graber et al. 2004), but one long-term follow-up
study (Taga et al. 2006) found that early maturing boys
fared psychosocially better in adulthood and did not display
increased substance use behaviors compared to normative
maturers. Thus, research concerning longitudinal associa-
tions between pubertal timing and substance use is scarce
and so far inconclusive.
Substance use in adolescence is associated with depres-
sion and conduct disorder (McMahon 1999; McArdle 2008;
Buckstein et al. 2005) that have also both been associated
with early pubertal timing (see above). On the other hand,
changes in parental supervision have been suggested as a
possible explanation for the association between early
pubertal timing and substance use behaviors (Costello et al.
2007; Hayatbakhsh et al. 2009). Parents may allow more,
even too much, freedom to an adolescent who appears
physically mature, and this may lead to problem behaviors
such as substance use, perhaps via affiliation with older
delinquent peers (Ge et al. 2002). Less parental involvement
in an adolescents life is also associated with increased
internalizing and externalizing symptoms and disorders
123
1290
during middle adolescence (Frojd et al. 2007). To better
understand how pubertal timing is associated with substance
use behaviors, it is important to control for the influence of
disorders or symptom dimensions associated with both
pubertal timing and substance use. Such symptom dimen-
sions are depression and conduct disorder. Parental super-
vising behaviors are another feature the previous literature
suggests associated both with pubertal development and
substance use.
The Current Study
We studied the impact of pubertal timing on adolescent
substance use behaviors during middle adolescence in a
2-year follow-up study in a population sample, bringing into
the analyses the role of depressive symptoms, aggression
and delinquency. We also controlled for the impact of
parental involvement in the adolescents social lives. Given
that the previous literature is more consistent about the
negative impact of early pubertal timing on the development
of girls, we hypothesised that more negative and persistent
associations would be found among girls. Our study
hypotheses were that early pubertal timing would be asso-
ciated with substance use among both boys and girls at age
15, that the association would persist at age 17, and that they
would be clearer among girls. We expected that depressive
symptoms, aggression and delinquency as well as parental
monitoring would partially explain the associations between
early pubertal timing and substance use so that when taking
into account these factors, the risk for substance use
according to pubertal timing would diminish.
Method
The present study is part of an ongoing prospective follow-
up study entitled the Adolescent Mental Health Cohort
Study. Ninth grade students from all Finnish-speaking
secondary schools in two Finnish cities, Tampere (200,000
inhabitants) and Vantaa (180,000 inhabitants), filled in a
person-identifiable questionnaire during a school lesson
supervised by a teacher. The questionnaires were returned
in sealed envelopes and collected by a member of the
research team. Questionnaires with two reminders were
mailed to students absent from school on the day of the
survey. The final baseline (T1) sample comprised 1,609
girls and 1,669 boys, with a mean age of 15.5 (SD 0.39).
The response rate was 94%.
Subjects responding to the first survey were contacted
through their current educational institutions, by post and
finally through the Internet to invite their participation in
the 2-year follow-up survey (T2). A total of 2,082
123
J Youth Adolescence (2011) 40:12881301
responses were received. Two participants responded
twice, once at the educational institution and once on the
Internet, and so their internet responses were excluded. Ten
respondents were excluded as they were judged by the
researchers to have responded facetiously. The final T2
sample (2,070 respondents) represented 63% of the ado-
lescents who responded to the first survey, with a mean age
of 17.6 (SD 0.41).
Drop-out from Follow-up
Altogether, 28% of the girls and 46% of the boys
responding to the baseline survey dropped out in the fol-
low-up. Responding was not associated with pubertal
timing in girls, whereas non-normative pubertal timing was
associated with a lower probability of responding in boys
(early 50%, normative 59%, late 49%; p = 0.001). Higher
levels of substance use behaviors at T1 were associated with
a lower probability of responding (cumulative experiences
of drunkenness: 52% vs. 65%; frequent alcohol use: 58%
vs. 67%, frequent drunkenness: 53% vs. 66%, smoking:
60% vs. 69%, cannabis use: 55% vs. 65%, p \ 0.001 for all
associations; first time drunk by age 13: 55% vs. 60%,
p = 0.024). Higher age, higher levels of delinquency and
lower levels of parental monitoring were associated with a
lower probability of responding (p \ 0.001), whereas
aggressiveness was not associated with the probability of
responding.
Measures
Pubertal Timing
Pubertal timing was measured by age at menarche/oigarche
by asking: How old were you when you first had your
periods/experienced ejaculation? The response alternatives
were 10 years or less/11/12/13/14/15 or more years/I have
not yet. In the analyses pubertal timing was classified
according to age at menarche/oigarche to early (11 years or
earlier), normative (1213 years) and late (14, 15 or later
and not yet).
Menarche is an accepted milestone in measuring
pubertal maturation (Rimpela and Rimpela 1993; Dick
et al. 2000), although self-reported timing of menarche has
also been criticised (Dorn et al. 1999). Onset of ejacula-
tions can be considered a corresponding way of measuring
puberty in boys in survey studies (Carlier and Steeno 1985;
Kulin et al. 1989), even if there is more room for errors. In
self-report studies, young boys may be uncertain whether
they ejaculate or not even if it occurred spontaneously at
night. Another method of assessing oigarche is to measure
morning spermaturia (Schaefer et al. 1990). However, even
after reaching oigarche, sperm cannot be observed in all the
J Youth Adolescence (2011) 40:12881301
urine samples of males. Many studies on the impact of
pubertal timing on mental health and psychosocial prob-
lems have used Tanner stages (Tanner 1962) as self-report
(Graber et al. 2004; Costello et al., 2007; Hayatbakhsh
et al., 2009). Studies using age at menarche/oigarche and
Tanner stages give similar results in this field.
Substance Use Behaviors
Substance use behaviors were measured at baseline (T1)
and at follow-up (T2). The questions concerning substance
use have been widely used in The School Health Promotion
Study, an annual survey among Finnish teenagers about
their health, health behavior and school experiences carried
out since 1995 (Torikka et al. 2001). Substance use vari-
ables were all elicited as categorical (Table 1), and they
were in the analyses dichotomised so that the analyses
focus on heaviest end of substance use in the present data.
Cumulative Experiences of Drunkenness
At baseline the adolescents were asked how many times so
far they had been really drunk. The response alternatives
were none/once/23 times/410 times/more than 10 times.
In the analyses cumulative experiences of drunkenness
were dichotomised to more than 10 times versus less or not
at all. Self-reports of drinking to the point of intoxication
have shown reasonable validity and reliability among
Finnish adolescents (Lintonen et al. 2004).
Frequent Alcohol Use
Frequency of drinking alcoholic beverages was elicited:
How often do you drink alcoholic beverages? (once a Heino et al. 1999). In the present data, the depression scale
week or more often/approximately once or twice a month/
less often/never). In the analyses we dichotomised alcohol
use frequency to once a week or more often versus less
frequently.
Age When Drunk for the First Time
At T1 the adolescents were asked: How old were you
when you were really drunk for the first time? Age in
years was recorded and dichotomised in the analyses to
13 years or younger versus 14 years or older.
Smoking
At baseline the adolescents were asked How many ciga-
rettes, pipes or portions of snuff have you used so far?
The response alternatives none/one/250/more than 50
were dichotomised to having never (none) versus ever (any
other alternative) smoked. At follow-up smoking was
1291
elicited as What best describes your current smoking?
The response alternatives were that one smokes daily/
weekly/less than weekly/has stopped smoking/has never
smoked, and in the analyses they were dichotomised to
smoking daily versus less or not at all.
Illegal Drug Use
Cannabis use was elicited by asking Have you ever tried
or used hash? (At follow-up the question was: During the
past 2 years, have you tried or used hash?). Other illegal
drug use was similarly elicited: Have you ever tried or
used ecstasy, heroine, cocaine, amphetamine or any other
hard drug? (At follow-up similarly as eliciting about
hash). The response alternatives were never/once/24
times/5 times or more. In the analyses cannabis use and
hard drug use were combined to illegal drug use, and this
variable was dichotomised to never versus ever. Possessing
and selling any of the above mentioned substances is illegal
in Finland.
Confounders
Depressive Symptoms
Depressive symptoms were measured by a Finnish modi-
fication of the 13-item Beck Depression Inventory (Beck
et al. 1974; Raitasalo 2007). The 13 items indicating dif-
ferent feelings, cognitions, and physical symptoms related
to depression were scored 03 and the scores were sum-
marized to a continuous depressive symptoms score. The
Finnish modification of the 13-item BDI has demonstrated
good psychometric properties among adolescents (Kaltiala-
displayed good internal consistency with Cronbachs alpha
of 0.84.
Delinquency
Delinquency was measured by the delinquency scale of the
Youth Self Report (Achenbach 1991). Aggression was
similarly measured by the Aggression Scale of Youth Self
Report. Both were used in the analyses as continuous
variables. In the present data, the Cronbachs alpha for the
delinquency scale was 0.67, and for the aggression scale it
was 0.86.
Parental Knowledge of the Adolescents Friends
and Whereabouts
The adolescents were asked two questions about their
parents knowing their friends and whereabouts (scored as
in parenthesis): Do your parents know most of your
123
1292 J Youth Adolescence (2011) 40:12881301

Table 1 Timing of oigarche/

Boys T1 Boys T2 Girls T1 Girls T2
menarche reported at age 15,
(age 15) (age 17) (age 15) (age 17)
and substance use behaviors
n = 903 n = 903 n = 1,167 n = 1,167
at age 15 (T1) and 2 years
later (T2) among Finnish Timing of oigarche/menarche 4.8 2.7
adolescents (%)
10 or less 12.7 17.0
11 28.3 35.6
12 29.2 27.7
13 13.5 13.3
14 3.2 2.1
15 or later 4.0 1.3
Has not occurred yet missing 4.2 0.3
How many times has been really drunk
Not at all 50.6 45.7
Once 10.9 10.5
23 times 14.0 18.3
410 times 11.1 13.0
More than 10 times 13.3 12.1
Missing 0.2 0.4
How often drinks alcohol
Weekly or more often 10.9 26.0 9.1 15.9
Twice a month 15.6 22.3 19.5 26.3
Once a month 9.4 12.4 13.7 16.0
Less than monthly 31.6 21.0 29.4 25.4
Not at all 32.3 17.4 27.9 16.1
Missing 0.2 0.9 0.3 0.4
First time really drunk at age
11 or less 4.1 1.9
12 6.9 6.2
13 13.2 15.0
14 15.2 22.5
15 9.1 9.0
Has not been drunk 50.5 44.8
Missing 1.1 0.7
Smoking
Daily 19.7 20.4 25.6 25.4
Less or not at all 79.1 78.7 73.5 74.1
Missing 1.2 0.9 0.9 0.4
Any cannabis use * *
Never 89.0 82.5 88.6 84.7
Once 4.0 5.9 5.0 5.7
24 times 3.5 5.1 3.5 5.3
5 or more times 2.4 5.6 2.9 4.1
Missing 0.9 0.1 0.2
Any hard drug use * *
Never 99.0 97.2 96.2 96.1
Once 0.9 1.2 2.8 2.0
24 times 0.3 0.4 0.8 1.0
5 times or more 0.7 1.1 0.2 0.7
Missing 0.1 0.4 0.3
* During past 2 years

123
J Youth Adolescence (2011) 40:12881301
friends? (Both know = 3/only mother knows = 2/only
father knows = 2/neither of them knows = 1), and Do
your parents know where you spend Friday and Saturday
nights? (always = 3/sometimes = 2/generally not = 1).
The scores of the two questions were summarised to a
continuous score of parental knowledge of the adolescents
friends and whereabouts (for brevity called parental
knowledge in the Tables). We have previously demon-
strated that parental knowledge of an adolescents friends
and whereabouts is associated with less internalizing and
externalizing symptoms, and more symptoms emerge when
parents do not know about these aspects of the adolescents
life (Frojd et al. 2007).
Age
Even if the participants were very homogenous as to age,
age was controlled for because during the rapid changes of
adolescent development, even small differences in age may
have an impact on both psychiatric disorders and problem
behaviors. Age was calculated from dates of birth and of
responding, and is expressed in years and used in the
analyses as continuous variable.
Ethical Considerations
The study received approval of the Ethics Committee of
Pirkanmaa Hospital District. The subjects gave written
informed consent after being informed about the study and
the voluntariness of participation. Their parents were
informed in advance by letter, but parental consent was not
obtained because the Finnish legislation on medical
research allows minors from 15 years to consent alone.
Results
Drinking, Smoking and Substance Use
Of the boys, 13.3%, and of the girls, 12.1%, had been really
drunk for more than 10 times at age 15. The proportion of
Table 2 Depressive symptoms, delinquency, aggression and parental knowledge of the adolescents friends and whereabouts according to
timing of oigarche/menarche among 15-year-old Finnish boys and girls (mean (SD))
Boys age
11 or less 1213 14 or more
Mean (SD) Mean (SD) Mean (SD)
Depressive symptoms 2.08 (3.7) 1.78 (3.0) 2.31 (4.4)
Delinquency 6.03 (2.9) 5.22 (2.5) 4.69 (2.8)
Aggression 9.08 (6.4) 7.50 (5.6) 6.78 (5.4)
Parental knowledge 5.10 (1.0) 5.24 (0.9) 5.42 (0.9)
1293
those using alcohol weekly or at least monthly increased
among both sexes clearly from age 15 to age 17. However,
daily smoking and illegal drug use did not essentially
increase from age 15 to age 17 (Table 1).
The Associations Between Depressive Symptoms,
Delinquency, Aggression and Parental Knowledge
with Pubertal Timing and with Substance Use Variables
The association between pubertal timing and the symptom
dimensions as well as between pubertal timing and
parental knowledge of the adolescents friends and
whereabouts were studied using analysis of variance.
Early pubertal timing was among boys associated with
increased delinquency, aggression and decreased parental
knowledge of the adolescents friends and whereabouts
and among girls with increased delinquency and aggres-
sion (Table 2). The associations between depressive
symptoms, delinquency and aggression as well as parental
knowledge about the adolescents friends and whereabouts
at T1, and substance use variables at T1 and T2 were
studied using t-test. Depressive symptoms, delinquency
and aggression at age 15 were associated with all the
studied substance use variables at age 15 at significance
level p \ 0.001 among both boys and girls. Among boys,
depressive symptoms, delinquency and aggression at age
15 were associated with all the substance use variables
studied at age 17 (p \ 0.001 in all analyses except for
depressive symptoms at age 15 9 weekly alcohol use at
age 17, p = 0.001, and depressive symptoms at age
15 9 smoking daily at age 17, p = 0.001). Among girls,
delinquency and aggression at age 15 were associated with
frequent alcohol use, daily smoking and illegal drug use at
age 17, all these associations at level p \ 0.001. Depres-
sive symptoms at age 15 were among girls also associated
with weekly alcohol use (p = 0.02), daily smoking
(0 \ 0.001) and illegal drug use (p \ 0.001) at age 17. All
the studied substance use variables at T1 and at T2 were
among both boys and girls associated with less parental
knowledge of friends and whereabouts at age 15
(p \ 0.001 in all associations).
Girls age
11 or less 1213 14 or more p
p Mean (SD) Mean (SD) Mean (SD)
0.18 3.44 (4.4) 3.09 (4.0) 2.84 (4.3) 0.32
\0.001 6.00 (2.7) 5.41 (2.7) 4.96 (2.5) \0.001
0.001 9.85 (5.0) 8.91 (4.9) 8.35 (5.0) 0.006
0.004 5.21 (0.9) 5.21 (0.9) 5.29 (0.9) 0.52
123
1294 J Youth Adolescence (2011) 40:12881301

Table 3 Drinking, smoking and other substance use behaviors among Finnish boys and girls at age 15 (T1), and 2 years later (T2), according to
age at oigarche/menarche (%, n/N)
Age at menarche/oigarche
11 years or less 1213 years 14 or more years p
%, n/N %, n/N %, n/N

Boys T1
More than ten times drunk 22.2 (35/158) 14.1 (73/519) 4.8 (9/187) \0.001
Uses alcohol weekly or more often 16.5 (26/258) 11.0 (57/520) 5.9 (11/187) 0.007
First time drunk by age 13 39.2 (62/158) 26.3 (137/520) 6.4 (12/187) \0.001
Smokes daily 30.6 (48/157) 22.0 (113/513) 7.0 (1/186) \0.001
Any illegal drug use 20.3 (32/158) 11.2 (58/520) 5.9 (11/187) \0.001
Boys T2
Uses alcohol weekly or more often 31.6 (50/158) 27.3 (141/517) 18.0 (33/183) 0.01
Smokes daily 36.7 (58/158) 27.7 (143/516) 12.0 (22/184) \0.001
Any illegal drug use past 2 years 24.1 (38/158) 18.5 (96/515) 9.1 (17/187) 0.001
Girls T1
More than ten times drunk 19.2 (44/229) 11.8 (87/737) 5.2 (10/194) \0.001
Uses alcohol weekly or more often 14.4 (33/229) 9.1 (67/736) 3.1 (6/195) \0.001
First time drunk by age 13 38.7 (89/230) 21.4 (158/739) 11.3 (22/195) \0.001
Smokes daily 29.4 (6/228) 20.0 (146/731) 12.8 (2/195) \0.001
Any illegal drug use 19.1 (44/230) 12.3 (91/739) 4.1 (8/195) \0.001
Girls T2
Uses alcohol weekly or more often 18.0 (41/228) 15.9 (117/736) 13.8 (27/195) 0.5
Smokes daily 31.4 (72/229) 24.7 (182/736) 22.2 (43/194) 0.06
Any illegal drug use past 2 years 20.9 (48/230) 16.0 (118/738) 12.3 (24/195) 0.05

Bivariate Associations Between Pubertal Timing
and Substance Use
The associations between pubertal timing with baseline and
follow-up substance use behaviors were first studied with
cross-tabulations and chi-square statistics. At age 15, all
the substance use behaviors studied were most common
among both sexes among those with early pubertal timing,
and least common among those with late pubertal timing.
The association persisted among boys at follow-up, but
among girls, only smoking was at T2 less common among
those with late puberty, otherwise the associations between
pubertal timing and substance use behaviors had disap-
peared (Table 3).
Multivariate Associations Between Pubertal Timing
and Substance Use
Multivariate associations between pubertal timing and
substance use behaviors were studied using logistic
regression. The substance use behaviors were entered each
in turn as the dependent variable. Pubertal timing and age
were first entered as independent variables. When studying
substance use behaviours at age 17, use of the substance in
question at age 15 was then added into the models. Finally,
symptom dimensions and parental knowledge of the ado-
lescents friends and whereabouts was also added into the
models.
Among boys, the age-controlled risk for all the sub-
stance use behaviors studied increased among those with
normative pubertal timing, and further increased among
those with early pubertal timing as compared to late
maturers. Age itself showed no statistically significant
association to substance use behaviors. The increasing risk
for substance use behaviors by normative and particularly
early puberty compared to late pubertal timing persisted
among 15-year-old boys also when depressive symptoms,
delinquency, aggression, and parental knowledge of the
adolescents friends and whereabouts were added to the
model. Adding symptom dimensions and parental knowl-
edge did not cause essential changes in the ORs by pubertal
timing, All the studied substance use behavior were also
predicted by delinquency with ORs varying from 1.5 to 1.8.
At follow-up, when the subjects aged 17, substance use
behaviors were similarly predicted among boys by early
pubertal timing when controlling for age, and the associ-
ations persisted when use of the substance in question at
age 15 was added into the model. Use at age 15 was in each
model a strong predictor of use at age 17. The association
between early puberty and daily smoking, and between

123
J Youth Adolescence (2011) 40:12881301
early puberty and illegal drug use also persisted when
symptom dimensions and parental knowledge of the ado-
lescents friends and whereabouts were added into the
models. Weekly use of alcohol, daily smoking and illegal
drug use at age 17 were also predicted by use at age 15 and
delinquency. All the models showed adequate fit based on
Hosmer-Lemeshow goodness of fit test (Table 4).
Among girls aged 15, risk for all the substance use
behaviors studied increased from late to normative and to
early maturers when controlling for age, and this finding
persisted when depressive symptoms, delinquency,
aggression and parental knowledge of the adolescents
friends and whereabouts were added to the model (for daily
smoking, only early puberty statistically significant).
Increasing age was significantly associated with having
been drunk more than 10 times, and with daily smoking
and illegal drug use. At follow-up, daily smoking and
illegal drug use were statistically significantly associated
with early puberty when age was controlled for. Adding
use at age 15 into the model leveled out the associations
between pubertal timing and substance use at age 17. Use
at age 15 was a strong predictor of use at age 17. This
persisted when symptom dimensions and parental knowl-
edge of the adolescents friends and whereabouts were
added. Frequent drinking, daily smoking and illegal drug
use were also all associated with delinquency. Chrono-
logical age revealed no significant association with sub-
stance use behaviors at follow-up. Most of the models
showed adequate fit based on Hosmer-Lemeshow goodness
of fit test (Table 5).
Discussion
The presents study adds to the literature by analyzing the
associations between pubertal timing and substance use
behaviors when controlling for both internalizing (depres-
sive symptoms) and externalizing (aggression and delin-
quency) symptoms and also at the same time for parental
knowledge of the adolescents friends and whereabouts,
which was used as a proxy for parental supervision (see
below). Among boys, all the substance use behaviors
studied were the more common the earlier the puberty.
This finding was in accordance with our hypotheses based
on previous research (Kaltiala-Heino et al. 2003; Ge et al.
2001; Ge et al. 2003; Ge et al. 2006). However, unlike what
we had hypothesized, the association between pubertal
timing and substance use behaviors was not explained by
depressive symptoms, aggression and delinquency, or
parental knowledge of friends and whereabouts. Among
boys, the association between pubertal timing and sub-
stance use behaviors persisted across middle adolescence,
from age 15 to 17, which is in accordance with the findings
1295
of Ge et al. (2001), but differ from the findings of Graber
et al. (2004) who also suggested that problems associated
with pubertal timing persist across adolescence among
boys, but it is late maturation that is the risk factor, no
early. Earlier use and delinquency were also predictive of
substance use behaviors at age 17.
Among girls, substance use behaviors at age 15, as like
among boys, were the more common the earlier the
pubertal timing. This was to be expected given the
knowledge of the associations between pubertal timing and
psychiatric disorders and problem behaviors in girls
(Mendle et al. 2007). The associations similarly proved
independent of depressive symptoms, aggression, delin-
quency and parental knowledge of the adolescents friends
and whereabouts. In girls, chronological age also showed
some positive association with substance use behaviors,
even if the group was very homogenous regarding age.
During the adolescent transition, even slight differences in
age may be significant in relationship to such behaviors as
substance use. However, contrary to our hypotheses based
on previous literature, the impact of pubertal timing on
substance use behaviors disappeared by age 17 among
girls. The girls had perhaps grown out of the period during
which pubertal timing differentiates susceptibility to
problem behaviors. This concurs with the findings of Dick
et al. (2000), who, in a large, population-based and age-
standardized Nordic sample, showed that the higher levels
of substance use observed among early maturing girls
persisted at the age of 16 years but diminished by the age
of 18.5. Graber et al. (2004) reported that numerous
unfavourable outcomes persisted into young adulthood
among early maturing girls, but substance use was, how-
ever, not among these.
Not unexpectedly, substance use was associated with
delinquency in all the models. Parental knowledge of the
adolescents friends and whereabouts was inversely asso-
ciated with substance use behaviors, although this did not
always reach statistical significance in multivariate models.
For all the studied substance use behaviours at age 17, use
at age 15 was a strong predictor, particularly for daily
smoking.
Research so far has mainly associated early puberty with
symptoms and problem behaviors among girls, whereas
among boys early puberty has been suggested to actually
be protective and late puberty a risk factor. However, early
puberty has also in boys been associated with vulnerability
to externalizing problems such as substance use. In an
earlier Finnish study, it was also found that early puberty
was associated with a number of problems including
problem drinking among boys, and that late puberty was
protective (Kaltiala-Heino et al. 2003). Given these find-
ings, we hypothesized that the associations between
pubertal timing and substance use would be to the same
123
1296 J Youth Adolescence (2011) 40:12881301

Table 4 Risk (OR, 95% CI) and Hosmer-Lemeshow goodness of fit test (p-value) for alcohol use, smoking and substance use among boys at age
15 (T1) and 2 years later (T2), according to timing of oigarche
Dependent variable
Predictors More than 10 Uses alcohol weekly First time drunk Smokes Any illegal
times drunk or more often by age 13 daily drug use
OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI)

T1
Model 1: controlling for age
Age at oigarche
11 or less 5.6 (2.612.1) 3.1 (1.56.6) 9.5 (4.918.5) 5.9 (3.011.3) 4.1 (2.08.4)
1213 3.2 (1.66.6) 2.0 (1.03.9) 5.2 (2.89.7) 3.8 (2.16.9) 2.0 (1.03.9)
14 or more Ref Ref Ref Ref Ref
Age 1.0 (0.61.8) 1.6 (1.02.6) 0.7 (0.41.1) 1.2 (0.81.9) 1.1 (0.62.0)
pa 0.618 0.412 0.137 0.598 0.399
Model 2. controlling for age, depressive symptoms, delinquency, aggression and parental knowledge of the adolescents friends and whereabouts
at age 15
Age at oigarche
11 or less 6.2 (2.316.9) 2.6 (1.16.4) 9.6 (4.421.2) 6.0 (2.514.4) 3.6 (1.58.7)
1213 4.3 (1.711.0) 2.2 (1.05.0) 5.8 (2.812.0) 4.8 (2.110.7) 2.1 (0.95.0)
14 or more Ref Ref Ref Ref Ref
Age 1.2 (0.62.3) 2.1 (1.23.7) 0.6 (0.41.1) 1.3 (0.82.3) 1.2 (0.72.2)
Depressive symptoms 1.0 (0.91.1) 1.0 (0.91.1) 1.0 (0.91.0) 1.0 (0.91.1) 1.0 (1.01.1)
Delinquency 1.7 (1.52.0) 1.5 (1.31.7) 1.5 (1.41.7) 1.8 (1.62.1) 1.4 (1.31.6)
Aggression 10. (0.91.0) 1.0 (1.01.1) 1.0 (0.91.0) 1.0 (0.91.0) 1.0 (1.01.0)
Parental knowledge 1.2 (0.91.5) 1.1 (0.81.4) 0.8 (0.71.0) 0.7 (0.71.2) 0.7 (0.50.9)
pa 0.244 0.817 0.301 0.668 0.359
T2
Model 1: controlling for age
Age at oigarche
11 or less 2.1 (1.33.5) 4.3 (2.57.4) 3.2 (1.75.9)
1213 1.7 (1.12.6) 2.8 (1.74.6) 2.1 (1.33.9)
14 or more Ref Ref Ref Ref Ref
Age 1.5 (1.02.2) 1.4 (1.02.2) 1.5 (1.02.4)
pa 0.991 0.727 0.466
Model 2: controlling for age and use at age 15
Age at oigarche
11 or less 1.7 (1.02.9) 2.3 (1.24.4) 2.2 (1.14.3)
1213 1.6 (1.02.4) 1.7 (1.03.0) 2.0 (1.13.6)
14 or more Ref Ref Ref
Age 1.3 (0.92.0) 1.6 (1.02.7) 1.6 (1.02.5)
Use at age 15 7.6 (4.712.2) 27.1 (17.342.2) 10.2 (6.616.2)
pa 0.758 0.907 0.093
Model 3. controlling for age, use at age 15 and depressive symptoms, delinquency, aggression and parental knowledge of the adolescents friends
and whereabouts at age 15
Age at oigarche
11 or less 1.4 (0.82.5) 2.5 (1.14.5) 1.9 (0.93.8)
1213 1.5 (0.92.3) 1.8 (1.03.3) 2.0 (1.23.7)
14 or more Ref Ref Ref
Age 1.5 (0.92.3) 1.6 (1.02.6) 1.6 (0.92.6)
Use at age 15 3.7 (2.26.2) 16.1 (9.826.5) 5.6 (3.49.4)
Depressive symptoms 1.0 (0.91.0) 1.0 (0.91.0) 1.0 (0.91.0)

123
J Youth Adolescence (2011) 40:12881301
Table 4 continued
Dependent variable
Predictors More than 10 Uses alcohol weekly
times drunk or more often
OR (95% CI) OR (95% CI)
Delinquency 1.3 (1.21.5)
Aggression 1.0 (1.01.0)
Parental knowledge 1.1 (0.91.3)
pa 0.204
a
Hosmer-Lemeshow goodness of fit test (p-value)
direction in both sexes but more pronounced and persistant
in girls. However, it was found that the association between
early puberty and substance use persisted across middle
adolescence among boys. It is possible that cultural dif-
ferences, especially gender role expectations, affect the
impact of pubertal timing on problem behaviors. In Fin-
land, gender equality is rather advanced relative to many
other Western societies, and this may modify the associa-
tions between pubertal timing and problem behaviours
such as substance use.
That substance use at age 15 was in both sexes the more
common the earlier the puberty seems to support the early
timing hypothesis of the role of puberty in vulnerability to
psychiatric disorders. Considerations of why early timing
would negatively influence mental health comprise the
stage termination hypothesisthat adolescents entering
puberty very early would not have enough time to gain the
cognitive and social skills to cope with the changesand,
like the off-time hypothesis, that meeting the bodily and
social changes without the support of peers in a similar
situation is stressful and causes susceptibility to disorders
and problem behaviors. In the light of knowledge gained in
modern neuroimaging studies, the stage termination
hypothesis may be problematic. The hormonal changes of
puberty (Buchanan et al. 1992) cause the neurobiological
changes in the brain that are a prerequisite of the qualita-
tive changes in cognitive processes (Luna et al. 2004; Paus
et al. 2008). Therefore, the cognitive skills needed to cope
with more complex social roles brought about by puberty
cannot be acquired without puberty itself. We therefore
conclude that it is the lack of developmental support by
having peers sharing the same situation that is stressful for
early developing adolescents and produces the risk for
problem behaviors such as substance use. Adolescents who
mature late, on the other hand, have the benefit of being
able to observe the changes in the lives of their peers before
facing them themselves, and thus late maturation, even if
deviance from the normative like early maturation, does
not pose a risk as does early maturation.
1297
First time drunk Smokes Any illegal
by age 13 daily drug use
OR (95% CI) OR (95% CI) OR (95% CI)
1.2 (1.11.3) 1.1 (1.01.3)
1.0 (1.01.0) 1.0 (1.01.1)
0.9 (0.71.2) 0.8 (0.61.0)
0.858 0.548
To some extent, substance use in adolescence is nor-
mative (Hibell et al. 2007). Another explanation of our
findings could be that early maturers could be more popular
among their peers, particularly among older peers; in the
case of girls, in relation to older boyfriends, and in the case
of boys, early maturation allows access to older males
through sports participation, etc. Accordingly, it may be
through this route that early maturers are more likely to be
engaging in substance use. It may have nothing to do with
stress at all. However, this would rather apply if we had
studied lesser substance use, such as compared totally
abstinent adolescents with those experimenting with sub-
stances. Because we focused in heaviest end of substance
use, and were also able to demonstrate that the level of use
we studied is associated with symptoms of both internal-
izing and externalizing psychiatric disorders, this line of
argument seems less plausible.
On the other hand, our data showed that, in girls, the
impact of pubertal timing on substance use behaviors lev-
eled out towards the end of middle adolescence. This lends
support to the stressful change hypothesis. Boys mature at a
slower rate than girls. Our data did not allow us to ascertain
whether the impact of pubertal timing on substance use
behaviors also levels out in boys as they terminate the
phase of middle adolescence. A longer follow-up would be
needed to study the issue further among boys.
Among both sexes at age 15, as expected, delinquency
and aggression were the more common the earlier the
puberty. Among girls, a linear, albeit not statistically sig-
nificant, increase in depressive symptom scores occurred
from the late to the early pubertal timing group. Regarding
boys, parental knowledge of the adolescents friends and
whereabouts decreased from late to early maturers, but
among girls, parental knowledge was not associated with
the girls pubertal timing. Parental knowledge of aspects of
an adolescents life is, of course, not solely a matter of
parental behavior, or effectiveness of parenting. Adoles-
cents are selective in what they disclose to their parents
(Kerr and Stattin 2000; Stattin and Kerr 2000). In the
123
1298 J Youth Adolescence (2011) 40:12881301

Table 5 Risk (OR, 95% CI) and Hosmer-Lemeshow goodness of fit test (p-value) for alcohol use, smoking and substance use among girls at age
15 (T1) and 2 years later (T2), according to timing of menarche
Dependent variable
Predictors More than 10 Uses alcohol weekly First time drunk Smokes Any illegal
times drunk or more often by age 13 daily drug use
OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI) OR (95% CI)

T1
Model 1: controlling for age
Age at menarche
11 or less 4.6 (2.29.5) 5.4 (2.213.3) 4.9 (2.98.3) 2.9 (1.74.8) 5.7 (2.612.5)
1213 2.6 (1.35.1) 3.2 (1.47.5) 2.1 (1.33.4) 1.7 (1.12.7) 3.4 (1.67.0)
14 or more Ref Ref Ref Ref Ref
Age 2.0 (1.23.2) 1.5 (0.82.6) 0.8 (0.61.3) 1.5 (1.02.1) 1.6 (1.02.5)
pa 0.438 0.984 0.611 0.563 0.952
Model 2: controlling for age, depressive symptoms, delinquency, aggression and parental knowledge of the adolescents friends and whereabouts
at age 15
Age at menarche
11 or less 3.7 (1.68.3) 3.6 (1.49.7) 4.7 (2.68.5) 2.2 (1.24.0) 4.9 (2.012.1)
1213 2.6 (1.25.5) 2.7 (1.16.7) 2.1 (1.23.5) 1.5 (0.92.6) 3.3 (1.47.9)
14 or more Ref Ref Ref Ref Ref
Age 2.1 (1.23.8) 1.4 (0.72.9) 0.7 (0.41.2) 1.4 (0.82.4) 1.7 (1.02.9)
Depressive symptoms 1.0 (1.01.1) 1.0 (1.01.1) 1.0 (0.91.0) 1.0 (1.01.0) 1.0 (1.01.1)
Delinquency 1.4 (1.31.6) 1.8 (1.62.0) 1.5 (1.41.6) 1.7 (1.51.8) 1.7 (1.61.9)
Aggression 1.0 (1.01.1) 0.9 (0.91.0) 1.0 (0.91.0) 1.0 (1.01.0) 1.0 (0.91.0)
Parental knowledge 0.8 (0.61.0) 1.0 (0.81.3) 0.8 (0.60.9) 0.8 (0.71.0) 1.0 (0.81.3)
pa 0.068 0.610 0.520 0.095 0.018
T2
Model 1: controlling for age
Age at menarche
11 or less 1.4 (0.82.3) 1.6 (1.02.5) 1.9 (1.13.2)
1213 1.2 (0.71.8) 1.2 (0.81.7) 1.3 (0.82.2)
14 or more Ref Ref Ref
Age 1.0 (0.61.6) 1.0 (0.71.5) 0.9 (0.61.4)
pa 0.826 0.201 0.920
Model 2: controlling for age and use at age 15
Age at menarche
11 or less 1.1 (0.61.9) 0.9 (0.51.5) 1.1 (0.61.9)
1213 1.0 (0.61.6) 0.8 (0.51.3) 0.9 (0.61.6)
14 or more Ref Ref Ref
Age 1.0 (0.61.6) 0.8 (0.51.3) 0.6 (0.41.1)
Use at age 15 4.9 (3.27.6) 30.3 (20.744.4) 14.3 (9.521.5)
pa 0.378 0.905 0.990
Model 3 controlling for age, depressive symptoms, delinquency, aggression and parental knowledge of the adolescents friends and whereabouts
at age 15
Age at menarche
11 or less 1.0 (0.51.7) 0.7 (0.91.3) 1.1 (0.62.0)
1213 1.0 (0.61.5) 0.7 (0.41.2) 1.0 (0.61.7)
14 or more Ref Ref Ref
Age 0.9 (0.51.6) 0.7 (0.41.3) 0.6 (0.31.1)
Use at age 15 2.3 (1.43.8) 16.4 (10.824.9) 6.8 (4.310.7)
Depressive symptoms 1.0 (0.91.0) 1.0 (1.01.1) 1.0 (1.01.0)

123
J Youth Adolescence (2011) 40:12881301
Table 5 continued
Dependent variable
Predictors More than 10 Uses alcohol weekly
times drunk or more often
OR (95% CI) OR (95% CI)
Delinquency 1.3 (1.21.4)
Aggression 1.0 (0.91.0)
Parental knowledge 0.9 (0.71.0)
pa 0.013
a
Hosmer-Lemeshow goodness of fit test (p-value)
present study, the information whether the parents know
the adolescents friends and whereabouts came from the
adolescents and thus illustrates how closely the adolescents
perceive themselves to be monitored, which is likely to
influence their own behavior. Gender difference in parental
knowledge of the adolescents friends and whereabouts
seems inappropriate because decreased knowledge is
associated with increased substance use in both sexes.
Professionals working with adolescents health and edu-
cation could advise parents to not decrease involvement in
their adolescent offsprings whereabouts and social net-
works too quickly, particularly when the adolescent
matures early.
Weekly or more frequent alcohol use increased in fre-
quency both among boys and among girls from age 15 to
age 17, whereas the proportion of those who smoked daily
increased only little, and the proportion of those who had
used cannabis or hard drugs between age 15 and 17 was
somewhat lower or about the same as the proportion of
those who had done so by age 15. This suggests that if
smoking initiation can be postponed over early adoles-
cence, it may not occur at all. Alcohol use, on the contrary,
increases with age across middle adolescence. In Finland,
buying or possessing alcohol or tobacco productions is not
allowed for minors under age 18, and cannabis as well as
hard drugs are totally illegal.
Methodological Considerations
The study was based on a large population sample. The
coverage of compulsory comprehensive school until age 16
in Finland is more than 99%. The cohort may thus be
considered representative of Finnish adolescents except for
those with mental handicaps or severe sensory deficits. The
material was, however, collected exclusively in urban
areas, thus the results may not be generalizable to rural
populations. The response in the Adolescent Mental Health
Cohort baseline survey was acceptable. However, among
those who already dropped out at baseline, psychosocial
and health problems may be more common than among
1299
First time drunk Smokes Any illegal
by age 13 daily drug use
OR (95% CI) OR (95% CI) OR (95% CI)
1.3 (1.21.4) 1.2 (1.11.4)
1.0 (1.01.0) 1.0 (1.01.0)
1.0 (0.81.2) 0.8 (0.60.9)
0.577 0.405
participants. Thus, their absence may result in the reported
prevalence rates of substance use behaviors being under-
estimates at T1. The response rate in the Adolescent
Mental Health Cohort follow-up was satisfactory. Not
responding in the follow-up was associated with increased
substance use behaviors in the baseline survey. It is thus
likely that in the follow-up, the prevalences of substance
use behaviors were somewhat underestimated. Not
responding in the follow-up was also among boys associ-
ated with reporting off-time maturation at T1. This signifies
that some caution needs to be taken in drawing firm con-
clusions of the findings, given also that drop-out in general
was greater among boys. However, even high levels of
attrition may not necessarily affect the associations studied
in health surveys van Loon et al. (2003).
In evaluating validity of information received with
logistic regression analyses, which we used to study mul-
tivariate associations, it is also important to consider
goodness of fit of the models, not only the significance of
the odds ratios for the different variables. Most of the
logistic regression models showed adequate fit based on
Hosmer-Lemeshow goodness of fit test, except two models
for girls. Because our main aim was to evaluate, if selected
predictors were statistically significantly associated to
dependent variables, not to create the optimal model
beyond the chosen independent variables, we also decided
to interpret these models. In future studies it would be
interesting to focus more on finding what combinations of
variables, in relation to pubertal timing, produce the best
fitting explaining models.
Conclusion
Early puberty is a associated with substance use behaviors
among both boys and girls, but only among boys does the
risk persist across middle adolescence. The association
between earlier pubertal timing and substance use behav-
iors is independent of emotional (depressive) and behav-
ioral (delinquency and aggression) symptoms, even if
123
1300
delinquency and aggression are the more common the
earlier the puberty. Nor is the association explained by
decreasing parental monitoring, even if parental monitoring
per se has an inverse association with substance use
behaviors. Parents and health and social authorities should
pay attention to early maturing adolescents, particularly
boys, in order to target interventions to prevent adolescent
substance use which is a risk factor for subsequent sub-
stance abuse and substance use disorders.
References
Achenbach, T. M. (1991). Integrative guide for the 1991 CBL/418,
YSR, and TRF profiles. Burlington, VT: University of Vermont.
Angold, A., Costello, E., Erkanli, A., & Worthmann, C. (1999).
Pubertal changes in hormone levels and depression in girls.
Psychological Medicine, 29, 10431053.
Anthony, J., & Petronis, K. (1995). Early onset drug use and risk of
later drug problems. Drug and Alcohol Dependence, 40, 915.
Beck, A. T., Rial, W. Y., & Rickels, K. (1974). Short form of
depression inventory: Cross-validation. Psychological Reports,
34, 11841186.
Breslau, N., Fenn, N., & Peterson, E. (1993). Early smoking initiation
and nicotine dependence in a cohort of young adults. Drug and
Alcohol Dependence, 33, 129137.
Buchanan, C., Eccles, J., & Becker, J. (1992). Are adolescents the
victims of raging hormones: Evidence for activational effects of
hormones on moods and behavior at adolescence. Psychological
Bulletin, 111, 62107.
Buckstein, O. G., Bernet, W., Arnold, V., Beitchman, J., Shaw, J.,
Benson, R. S., et al. (2005). Practice parameter for the
assessment and treatment of children and adolescents with
substance use disorders. Journal of the American Academy of
Child and Adolescent Psychiatry, 44, 609621.
Carlier, J., & Steeno, O. (1985). Oigarche: The age at first ejaculation.
Andrologia, 17, 104106.
Carpenter-Hyland, E., & Chandler, J. (2007). Adaptive plasticity of
NDMA receptors and dendritic spines: Implications for enhanced
vulnerability of the adolescent brain to alcohol addiction.
Pharmacology, Biochemistry and Behavior, 86, 200208.
Chung, H., Park, Y. S., & Lanza, S. (2005). Latent transition analysis
with covariates: Pubertal timing and substance use behaviors in
adolescent females. Statistics in Medicine, 24, 28952910.
Clark, D. B., Thatcher, D. L., & Tapert, S. F. (2008). Alcohol,
psychological dysregulation, and adolescent brain development.
Alcoholism, Clinical and Experimental Research, 32, 375385.
Costello, E., Sung, M., Worthman, C., & Angold, A. (2007). Pubertal
maturation and the development of alcohol use and abuse. Drug
and Alcohol dependence, 88S, S50S59.
DeWitt, D., Adlaf, E., Offord, D., & Ogborne, A. (2000). Age at first
alcohol use: A risk factor for the development of alcohol
disorders. American Journal of Psychiatry, 157, 745750.
Dick, D., Rose, D., Viken, R., & Kaprio, J. (2000). Pubertal timing
and substance use: Associations between and within families
across late adolescence. Developmental Psychology, 36,
180189.
Dorn, L., Nottelmann, E., Susman, E., Inoff-Germain, G., Cutler, G.,
Jr., & Chrousos, G. (1999). Variability in hormone concentra-
tions and self-reported menstrual histories in young adolescents:
Menarche as an integral part of a developmental process. Journal
of Youth and Adolescence, 28, 283304.
123
J Youth Adolescence (2011) 40:12881301
Frojd, S., Kaltiala-Heino, R., & Rimpela, M. (2007). The association
of parental monitoring and family structure with diverse
maladjustment outcomes in middle adolescent boys and girls.
Nordic Journal of Psychiatry, 61, 296303.
Ge, X., Conger, R., & Elder, G. (2001). The relationship between
pubertal transition and psychological distress among adolescent
boys. Journal of Res Adolescence, 11, 4970.
Ge, X., Conger, R., Simons, R., Brody, G., & McBride Murry, V.
(2002). Contextual amplification of pubertal transition effects on
deviant peer affiliation and externalizing behavior among African
American Children. Developmental Psychology, 38, 4254.
Ge, X., Kim, I., Conger, R., Brody, G., Simons, R., Gibbons, F., et al.
(2003). Its about timing and change: Pubertal transition effects
on symptoms of major depression among African American
youths. Developmental Psychology, 39, 430439.
Ge, X., Brody, G., Conger, R., & Simons, R. (2006). Pubertal
maturation and African American childrens internalizing and
externalizing symptoms. Journal of Youth and Adolescence, 35,
531540.
Graber, J., Lewinsohn, P., Seeley, J., & Brooks-Gunn, J. (1997). Is
psychopathology associated with the timing of pubertal devel-
opment? Journal of the American Academy of Child and
Adolescent Psychiatry, 36, 17681776.
Graber, J., Seeley, J., Brooks-Gunn, J., & Lewinsohn, P. (2004). Is
pubertal timing associated with psychopathology in young
adulthood? Journal of the American Academy of Child and
Adolescent Psychiatry, 43, 718726.
Graber, J., Brooks-Gunn, J., & Warren, M. (2006). Pubertal effects on
adjustment in girls: Moving from demonstrating effects to identi-
fying pathways. Journal of Youth and Adolescence, 35, 413423.
Hayatbakhsh, M., Najman, J., McGee, T., Bor, W., & OCallaghan,
M. (2009). Early pubertal maturation in the prediction of early
adult substance use: a prospective study. Addiction, 104, 5965.
Hibell, B., Guttormsson, U., Ahlstrom, S., Balakireva, O., Bjarnason,
Y., Kokkevi, A., & Kraus, J. (2007) The 2007 ESPAD report.
Substance use among students in 35 European countries. The
European school survey project on alcohol and other drugs.
www.espad.org.
Kaltiala-Heino, R., Rimpela, M., Rantanen, P., & Laippala, P. (1999).
Finnish modification of the 13-item Beck Depression Inventory
in screening an adolescent population for depressiveness and
positive mood. Nordic Journal of Psychiatry, 53, 453457.
Kaltiala-Heino, R., Marttunen, M., Rantanen, P., & Rimpela, M.
(2003). Early puberty is associated with mental health problems
in middle adolescence. Social Science and Medicine, 57,
10551064.
Kerr, M., & Stattin, H. (2000). What parents know, how they know it,
and several forms of adolescent adjustment: Further support for a
reinterpretation of monitoring. Developmental Psychology, 36,
366380.
Kulin, H., Frontera, M., Demers, L., Bartholomew, M., & Lloyd, T.
(1989). The onset of sperm production in pubertal boys.
Relationship to gonadotropin excretion. American Journal of
Diseases of Children, 143, 190193.
Lintonen, T., Ahlstrom, S., & Metso, L. (2004). The reliability of self-
reported drinking in adolescence. Alcohol and Alcoholism, 39,
362368.
Lubman, D. I., Hides, L., Yucel, M., & Toumbourou, J. W. (2007).
Intervening early to reduce developmentally harmful substance
use among youth populations. Medical Journal of Australia,
187(7 Suppl), S22S25.
Luna, B., Graver, K. E., Urban, T. A., Lazar, N. A., & Sweeney, J. A.
(2004). Maturation of cognitive processes from late childhood to
adulthood. Child Development, 75, 13571372.
Martin, C., Kelly, T., Rayens, M., Brogli, B., Brentzl, A., Smith, J.,
et al. (2002). Sensation seeking, puberty, and nicotine, alcohol
J Youth Adolescence (2011) 40:12881301
and marijuana use in adolescence. Journal of the American
Academy of Child and Adolescent Psychiatry, 41, 14951502.
McArdle, P. (2008). Alcohol abuse in adolescents. Archives of
Diseases of Children, 93, 527.
McMahon, R. (1999). Child and adolescent psychopathology as risk
factors for subsequent tobacco use. Nicotine & Tobacco
Research, S2, S45S50.
Mendle, J., Turkheimer, E., & Emery, R. (2007). Detrimental
psychological outcomes associated with early pubertal timing
in adolescent girls. Developmental Review, 27, 151171.
Paus, T., Keshavan, M., & Giedd, J. N. (2008). Why do many mental
disorders emerge during adolescence? Nature reviews. Neuro-
science, 9, 947957.
Raitasalo, R. (2007). Mood questionnaire. Finnish modification of the
short form of the Beck Depression Inventory measuring depres-
sion symptoms and self-esteem. Helsinki: The Social Insurance
Institution of Finland.
Rimpela, A., & Rimpela, M. (1993). Towards an equal distribution of
health? Socioeconomic and regional differences of the secular
trend of the age of menarche in Finland from 19791989. Acta
Paediatrica, 82, 8790.
Schaefer, F., Marr, J., Aeidel, C., Tilgen, W., & Scharer, K. (1990).
Assessment of gonadal maturation by evaluation of spermaturia.
Achives of Diseases of Childhood, 65, 12051207.
Stattin, H., & Kerr, M. (2000). Parental monitoring: A reinterpreta-
tion. Child Development, 71, 10721085.
Taga, K., Markey, C., & Friedman, H. (2006). A longitudinal
investigation of associations between boys pubertal timing and
adult behavioral health and well-being. Journal of Youth and
Adolescence, 35, 401411.
Tanner, J. M. (1962). Growth at adolescence. Oxford: Blackwell.
Torikka, A., Kaltiala-Heino, R., Rimpela, M., & Rantanen, P. (2001).
Depression, drinking and substance use among 14- to 16-year-
old Finnish adolescents. Nordic Journal of Psychiatry, 55,
351358.
1301
Van Loon, A., Tijhuis, M., Picavet, S., Surtees, P., & Ormel, J.
(2003). Survey non-response in the Netherlands: Effects on
prevalence estimates and associations. Annals of Epidemiology,
13(2), 105110.
Author Biographies
Riittakerttu Kaltiala-Heino is professor of adolescent psychiatry in
University of Tampere, and director of the Department of Adolescent
Psychiatry in Tampere University Hospital. She received her doctorate
in University of Tampere. Her research interests include epidemiology
of psychiatric disorders in adolescence and adolescent psychiatric
health services research. Special topics include pubertal timing and
mental health, and school bullying and mental health.
Anna-Maija Koivisto is statistician in Tampere School of Public
Health. She acts as methodological consultant in a variety of research
projects in health research and medicine.
Mauri Marttunen is Professor of Adolescent Psychiatry at the
University of Helsinki and Helsinki University Central Hospital. His
major research interests include epidemiology, comorbidity, course
and predictors of youthful psychiatric disorders, particularly of
depression and suicidal behavior.
Sari Frojd is a Senior Assistant Professor at the University of
Tampere. She received her doctorate in social psychiatry at the
University of Tampere. Her major research interests include family
determinants of adolescent mental health, and comorbidity issues in
adolescence.
123
Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.