Literature week 2
Continuities and discontinuities in psychopathology between childhood and adult life –
Michael Rutter, Julia Kim-Cohen, and Barbara Maughan
https://web-s-ebscohost-com.proxy-ub.rug.nl/ehost/pdfviewer/pdfviewer?vid=4&sid=63e02e81-1388-4734-8717-94729bbda7be%40redis
The possible mechanisms involved in continuities and discontinuities in psychopathology between
childhood and adult life are considered in relation to the findings from systematic, prospective,
longterm longitudinal studies. Findings on schizophrenia, neurodevelopmental disorders, emotional
disturbances, antisocial behaviour and substance abuse are used as conditions illustrating the key
issues. The overarching themes are then discussed in relation to heterotypic continuity and
psychopathologic progression, early age at onset and a range of possible mediating mechanisms –
including genetic mediation, ‘kindling’ effects, environmental influences, coping mechanisms and
cognitive processing of experiences. Some of the key research challenges that remain concern the
testing of competing hypotheses on mediating processes, the changes involved in adolescence, the
transition from prodromal phase to overt schizophrenia and the emergence of adolescent-limited
antisocial behaviour. Greater use needs to be made of genetic research strategies and of the testing
of possible cognitive processing mediation effects.
When the associations between mental disorders in childhood and adulthood were reviewed a
decade ago (Rutter, 1995), the main conclusion was that the findings raised key questions about
possible causal mechanisms – the circumstances in the biology or in the environment that make
continuities or discontinuities either more or less likely.
The possibility of examining mediating processes has come about partly as a result of the availability
of a range of high quality long-term longitudinal/ epidemiological studies (most with repeated,
multimethod assessments) extending from childhood into adult life, partly because of new thinking
on concepts and new findings on mechanisms, and partly because of a virtual revolution in research
and theoretical approaches to the topic of continuities and discontinuities between childhood and
adult life. A developmental perspective is mainstream.
This review primarily focuses on the possible mechanisms underlying continuities and discontinuities
over the life span, rather than seeking to quantify the overall level of association between
psychopathology in childhood and in adult life.
Childhood origins of schizophrenia
Traditionally, schizophrenia was conceptualised as a psychosis, usually beginning in late adolescence/
early adult life. Nevertheless, early case–control and follow-back studies (Rutter & Garmezy, 1983)
showed that children who later developed overt schizophrenia were more likely than controls to
show social, emotional and behavioural problems in childhood. Moreover, the longitudinal study of
children born to mothers with schizophrenia (as compared with children born to mothers with an
affective disorder, and with controls) has shown that those who develop a schizophrenia spectrum
disorder exhibit a stable pattern of impaired attention.
children who later show overt schizophrenia differ from other children in both socio-emotional
features and motor coordination. Findings on child psychiatric clinic attendees who go on to develop
either schizophrenia or bipolar disorder, nevertheless, suggest that there may be a degree of
,diagnostic specificity, with abnormal suspiciousness/sensitivity and relationship difficulties with peers
being particularly associated with later schizophrenia.
The earlier studies of special groups had given no strong indication that these childhood precursors
included any early manifestations of psychotic symptomatology. The findings from the Dunedin
Longitudinal Study, however, were important in providing the first evidence for continuity of
psychotic-like symptoms from childhood to adulthood. Psychotic-like symptoms from childhood to
adulthood. Self-reported symptoms about delusional beliefs and hallucinatory experiences at age 11
were significantly associated with an increase in the risk of developing a schizophrenia spectrum
disorder by the age of 26 years.
Three main queries have yet to be resolved with respect to the meaning of the findings on the
features in childhood and adolescence that predict the later development of a schizophrenic
psychosis. First, distinctions have usually been drawn between precursors and prodromal symptoms,
the difference being that the former constitute risk factors, whereas the latter constitute the early
manifestations of the disorder itself. The general notion is probably valid but over half of individuals
with prodromal features do not go on to develop schizophrenia. Accordingly, this necessarily raises
queries about what is meant by a prodromal phase
Second, all the findings raise the question of what it is that leads to the translation of precursors or
prodromata into overt schizophrenia. Three main possibilities have been considered. To begin with,
developmental changes in brain structure and function during late adolescence may be crucial.
Alternatively, as consistently shown by epidemiological/ longitudinal studies, heavy early use of
cannabis in adolescence is associated with a substantial increase in the risk of schizophrenia (maybe
because of the COMT-gene). The third possibility is that certain types of social adversity, such as
migration and isolation, may also be contributory.
The final query concerns the possibility that, despite the early neurodevelopmental abnormalities,
there are further changes in both cognitive function and brain structure and function that take place
during the course of the schizophrenia spectrum disorder in adult life either as a result of the disease
process or of the drugs used in its treatment. The evidence so far is not decisive but it does suggest
that further changes may take place after the development of the psychosis and, therefore, that
these may be relevant with respect to continuity in psychopathology.
Neurodevelopmental disorders
In recent years there has been a tendency for certain early-onset disorders to be grouped together
under the concept of ‘neurodevelopmental disorders’. Such disorders have eight main features. First,
they are manifest by a delay/deviance in maturationallyinfluenced psychological features (i.e., the
skills cannot develop unless the necessary neural structure is available). Secondly, the course of the
disorder is not marked by the remissions and relapses that are characteristic of most multifactorial
mental disorders. Third, there is a general tendency for the impairment associated with the disorder
to lessen with age, but this goes in parallel with a tendency for it to persist into adulthood. In other
words, the disorder is not just a normal variation. Fourth, the disorders all involve some degree of
specific or general cognitive impairment. Fifth, there is a tendency for overlap among the different
neurodevelopmental disorders. That is, although each disorder has some important specificities they
also have substantial overlap with other neurodevelopmental disorders. Sixth, in almost all cases, the
genetic influences on individual differences and liability are quite strong. Seventh, despite this,
, environmental influences are probably also contributory. Finally, the disorders all show a marked
male preponderance.
Autism spectrum disorders
The heritability of autism exceeds 90% but it is a multifactorial disorder, the environmental risk
factors remaining unknown at the present time. Home movies have shown that abnormalities may
be detectable in the first year of life but most children with autism do not show readily identifiable
abnormalities until about 18 months of age, with reliable and valid diagnoses being possible only
after the age of 2 years in many case.
Prognosis for those with an initial non-verbal IQ in childhood of less than 50 was uniformly poor, with
none achieving independence in adult life and all showing continuing handicapping autistic problems.
In keeping with earlier findings, the best predictor of outcome was the IQ level and the presence of
useful language by the age of 5 years. There is some indication that more extensive and appropriate
help with independent living and employment in adult life may make a real difference.
The milder varieties of autism, now usually described in terms of a so-called broader phenotype,
show a much higher proportion of those that do achieve independent functioning in adult life,
although what the proportion is remains uncertain, and it does appear that, despite their good
functioning in many aspects, they continue to show important autistic features.
Specific language impairments (SLI)
Traditionally, concepts of developmental language disorders (specific language impairment: SLI) have
assumed that these represent a relatively pure deficit in language. The Clegg, Hollis, Mawhood, and
Rutter (2005) follow-up from childhood into mid-adult life has cast considerable doubt on that
concept. All of the individuals studied had gained a reasonable level of communicative language, but
their scores on tests of language functioning were all far below those of both their siblings and the
IQ-matched control group (the latter two groups not differing significantly on any of the outcomes
assessed). Not surprisingly, the specific language impairment group was also significantly impaired in
adult life on both reading and spelling.
What was much more unexpected, however, was that only just over one in six of the SLI group had
been continually in paid employment, as compared with some 94% of their siblings. Although subtle
language deficits continued right into midadult life, the main impairment in adulthood concerned
social functioning and social relationships, rather than language. The SLI group showed significant
impairment in theory of mind task performance.
Overall, in keeping with the findings from other studies (Bishop & Norbury, 2002; Botting &
ContiRamsden, 2003; Stothard, Snowling, Bishop, Chipcase, & Kaplan, 1998), four findings stand out.
First, although there are children who are very delayed in their onset of spoken language who
nevertheless catch up and appear to show normal functioning from the time of school entry, at least
as high a proportion go on to show language difficulties that persist right into adult life. In that
respect, it is clear that the disorder represents far more than the end of a normal continuum. Second,
although there are important differences between autism and SLI, there is also substantial overlap,
especially in those whose language deficit includes pragmatic problems. Third, genetic influences
seem to be substantially greater in the case of persistent language impairment resulting in clinical
care, as compared with transient delays in language development (Bishop, Price, Dale, & Plomin,
2003). Fourth, although SLI is defined in terms of a language impairment, the adult outcome indicates
