Brain fog in menopause: a health-care professional's guide for decision-making and counseling on cognition.

Midlife women commonly experience changes in their cognitive function as they transition through menopause and express concern about whether these changes represent the initial stages of a more serious cognitive disorder. Health-care practitioners play an important role in counseling women on cognitive changes at midlife and normalizing women's experience. The aim of this commissioned International Menopause Society White Paper on cognition is to provide practitioners with an overview of data informing the clinical care of menopausal women and a framework for clinical counseling and decision-making. Among the topics presented are the specific cognitive changes occurring in menopause, the duration of such changes and their severity. The role of estrogen and menopause symptoms is reviewed. We present talking points for clinical counseling on the effects of hormone therapy on cognition and dementia risk in women, including discussion of absolute risk. Lastly, a brief review of modifiable risk factors for age-related cognitive decline and dementia is presented, with guidance for counseling patients on optimizing their brain health at midlife and beyond.

A research primer for studies of cognitive changes across the menopause transition.

There has been a proliferation of studies demonstrating important sex differences in cognitive aging and dementia, and with this an increased interest in the role of menopause and sex steroid hormones in women's brain health. Foundational longitudinal studies of cognitive changes from the premenopause to perimenopause stage have shown reliable declines in verbal memory, with variable findings in processing speed, attention/working memory and verbal fluency. Continued research is needed to advance understanding of the range of cognitive domains affected, the duration of cognitive changes, the generalizability of these changes across cultures, the factors that account for such changes and the factors that can improve cognition at this time. In this article, we briefly review and draw on findings from large longitudinal studies of cognitive changes across the menopause transition to inform the design of future studies on this topic. We focus on key issues such as objective versus subjective cognitive measures; cognitive domains and tests; staging menopause; study design; mediators of cognitive effects (including hormones and menopause symptoms); and consideration of key covariates. We suggest that a more uniform and evidence-based approach to the investigation of these issues can advance the quality of the science in menopause and cognition.

Cognitive profiles in perimenopause: hormonal and menopausal symptom correlates.

OBJECTIVE: Perimenopause is associated with declines in attention, working memory and verbal memory; however, there are significant individual differences. Further, the contributions of hormones and menopausal symptoms to domain-specific cognitive functions remain unknown. This longitudinal study aimed to determine whether there were distinct cognitive profiles in perimenopause and to identify factors associated with each profile. DESIGN: In a sample of 85 women evaluated over 400 bi-annual visits, we administered a comprehensive neuropsychological battery, assessed menopausal symptoms and measured 17ß-estradiol and follicle stimulating hormone. Multilevel latent profile analysis was used to identify cognitive profiles. Regressions were conducted to determine differences in hormones and symptoms by profile after adjusting for Stages of Reproductive Aging Workshop + 10 (STRAW + 10) stage and demographic factors. RESULTS: Perimenopausal cognitive profiles consisted of cognitively normal (Profile 1; n = 162), weaknesses in verbal learning and memory (Profile 2; n = 94), strengths in verbal learning and memory (Profile 3; n = 98) and strengths in attention and executive function (Profile 4; n = 61). Profile 2 was differentiated by less hormonal variability and more sleep disturbance than Profile 1 (p < 0.05). CONCLUSIONS: There is significant heterogeneity in cognition during perimenopause. While most women do not develop impairments, a significant minority experience weaknesses in verbal learning and memory. Profile analysis may identify at-risk populations and inform interventions.

School success in childhood and subsequent prodromal symptoms and psychoses in the Northern Finland Birth Cohort 1986.

BACKGROUND: Low IQ is a risk factor for psychosis, but the effect of high IQ is more controversial. The aim was to explore the association of childhood school success with prodromal symptoms in adolescence and psychoses in adulthood. METHODS: In the general population-based Northern Finland Birth Cohort 1986 (n = 8 229), we studied the relationship between teacher-assessed learning deficits, special talents and general school success at age 8 years and both prodromal symptoms (PROD-screen) at age 15-16 years and the occurrence of psychoses by age 30 years. RESULTS: More prodromal symptoms were experienced by those talented in oral presentation [boys: adjusted odds ratio (OR) 1.49; 95% confidence interval 1.14-1.96; girls: 1.23; 1.00-1.52] or drawing (boys: 1.44; 1.10-1.87). Conversely, being talented in athletics decreased the probability of psychotic-like symptoms (boys: OR 0.72; 0.58-0.90). School success below average predicted less prodromal symptoms with boys (OR 0.68; 0.48-0.97), whereas above-average success predicted more prodromal symptoms with girls (OR 1.22; 1.03-1.44). The occurrence of psychoses was not affected. Learning deficits did not associate with prodromal symptoms or psychoses. CONCLUSIONS: Learning deficits in childhood did not increase the risk of prodromal symptoms in adolescence or later psychosis in this large birth cohort. Learning deficits are not always associated with increased risk of psychosis, which might be due to, e.g. special support given in schools. The higher prevalence of prodromal symptoms in talented children may reflect a different kind of relationship of school success with prodromal symptoms compared to full psychoses.

Menopausal symptoms, menopausal stage and cognitive functioning in black urban African women.

Objective: Studies, conducted largely in North America and Europe, demonstrate that menopausal symptoms and menopausal stage influence cognitive function. Here, we evaluate these associations in a large cohort of sub-Saharan African women, a population where these associations are understudied. We hypothesized that premenopausal women would show better cognitive performance than women later in the transition, and that menopausal symptoms would be inversely related to cognition.Methods: This cross-sectional study included 702 black urban South African women between the ages of 40 and 60 years from the Study of Women Entering and in Endocrine Transition. Participants completed the Symbol Digit Modalities Test, a measure of processing speed and incidental recall. Menopausal stage was ascertained using the Stages of Reproductive Aging Workshop+ 10 criteria and symptoms using the Menopause Rating Scale. Multivariable linear regression analyses were used to examine adjusted associations between menopausal stage and menopausal symptoms on cognitive performance.Results: In adjusted analyses, menopausal stage was not associated with processing speed (p = 0.35) or incidental recall (p = 0.64). However, more severe symptoms of hot flushes and anxiety were associated with slower processing speed (all p < 0.05), and more severe mood symptoms were associated with worse incidental recall (p = 0.008).Conclusion: Menopausal symptoms, but not menopausal stage, were associated with cognitive function in this cross-sectional study of sub-Saharan African women.

A pilot investigation of differential neuroendocrine associations with fronto-limbic activation during semantically-cued list learning in mood disorders.

BACKGROUND: Decreased volume and disrupted function in neural structures essential for memory formation (e.g. medial temporal lobe and prefrontal cortex) are common among individuals with depression. Hypothalamic-pituitary-axis function, as reflected by measurement of cortisol levels, is linked to neural activity during memory encoding in healthy people. However, it is not as well understood whether cortisol is associated with alterations in fronto-temporal recruitment during memory encoding in depression. METHODS: In this pilot study, we evaluated associations between cortisol and neural activation during memory encoding in 62 adults (18-65 years) with mood disorders (MD; n = 39, 66.7% female), including major depression (n = 28) and bipolar I disorder (n = 11), and healthy controls (HC; n = 23, 43.5% female). Participants provided salivary cortisol samples before and after completing a semantically-cued list-learning task during 3-Tesla fMRI. Links between pre-scan cortisol (and cortisol change) and activation during encoding were evaluated using block and event-related models. RESULTS: Overall, pre-scan cortisol level was positively associated with greater engagement of fronto-limbic activation during the encoding block. However, in MD, pre-scan cortisol was associated with attenuated activation during encoding in medial frontal, superior and middle temporal gyri, insula, lingual gyrus, and claustrum relative to HCs. Cortisol-related attenuation of activation in MD was also observed during encoding of words subsequently recalled in the ventral anterior cingulate, hypothalamus, and middle temporal gyrus. By and large, cortisol change (pre/post scan) predicted the same pattern of findings in both block and event-related contrasts. LIMITATIONS: Although analyses accounted for variations in scanner time of day, circadian alterations in cortisol may have introduced variability into the results. CONCLUSIONS: Pre-scan cortisol may selectively interfere with recruitment of important fronto-temporal memory circuitry in mood disorders. The inverted associations between cortisol and neural function in MD relative to HC also elucidate potentially unique pathophysiological markers of mood disorders.

Verbal and spatial working memory among drug-using HIV-infected men and women.

Working memory (WM) is a critical component of many neurocognitive functions. The literature has demonstrated consistently that WM impairment is more frequent and severe among substance-dependent individuals (SDIs) infected with HIV compared with uninfected SDIs; however, the SDIs who participated in these previous studies were primarily male. There are few published data on WM performance among HIV+ women with or without substance use disorders, and essentially no direct comparisons of WM performance between HIV+ men and women, regardless of substance use. We investigated potential sex and serostatus effects on WM among a sample of 360 SDIs (114 with HIV; 66% female) verified abstinent from alcohol and drugs of abuse at testing and generally comparable on substance use and comorbid characteristics. Participants were tested with the n-back task, a well-established WM measure that is sensitive to HIV-associated cognitive impairment. HIV+ men and women performed spatial and verbal versions of the n-back significantly less accurately compared with HIV- participants. Women showed slower response times compared with men on both versions, regardless of HIV serostatus. Individuals dependent on cocaine showed faster RTs compared with non-dependent users, but this effect was not apparent among opioid- or alcohol-dependent groups. Findings on n-back accuracy are consistent with our previous proposal that WM impairment represents a signature deficit among HIV+ SDIs; however, WM impairment appears less common among HIV+ women without a substance use history. The pattern of sex differences in response speed but serostatus effects on response accuracy is comparable to a recent report by our group of sex differences in learning speed but serostatus effects on delayed recall.

Smokin' hot: adolescent smoking and the risk of psychosis.

OBJECTIVE: Daily smoking has been associated with a greater risk of psychosis. However, we are still lacking studies to adjust for baseline psychotic experiences and other substance use. We examined associations between daily smoking and psychosis risk in a 15-year follow-up while accounting for these covariates in a prospective sample (N = 6081) from the Northern Finland Birth Cohort 1986. METHODS: Self-report questionnaires on psychotic experiences (PROD-screen), tobacco smoking and other substance use were completed when the cohort members were 15-16 years old. Tobacco smoking was categorized into three groups (non-smokers, 1-9 cigarettes and ≥10 cigarettes/day). Psychosis diagnoses were obtained from national registers until the age of 30 years. RESULTS: Subjects in heaviest smoking category were at increased risk of subsequent psychosis (unadjusted HR = 3.15; 95% CI 1.94-5.13). When adjusted for baseline psychotic experiences the association persisted (HR = 2.87; 1.76-4.68) and remained significant even after adjustments for multiple known risk factors such as cannabis use, frequent alcohol use, other illicit substance use, parental substance abuse, and psychosis. Furthermore, number of smoked cigarettes increased psychosis risk in a dose-response manner (adjusted OR = 1.05; 1.01-1.08). CONCLUSION: Heavy tobacco smoking in adolescence was associated with a greater risk for psychosis even after adjustment for confounders.

Considering sex differences clarifies the effects of depression on facial emotion processing during fMRI.

BACKGROUND: Sex differences in emotion processing may play a role in women's increased risk for Major Depressive Disorder (MDD). However, studies of sex differences in brain mechanisms involved in emotion processing in MDD (or interactions of sex and diagnosis) are sparse. METHODS: We conducted an event-related fMRI study examining the interactive and distinct effects of sex and MDD on neural activity during a facial emotion perception task. To minimize effects of current affective state and cumulative disease burden, we studied participants with remitted MDD (rMDD) who were early in the course of the illness. In total, 88 individuals aged 18-23 participated, including 48 with rMDD (32 female) and 40 healthy controls (HC; 25 female). RESULTS: fMRI revealed an interaction between sex and diagnosis for sad and neutral facial expressions in the superior frontal gyrus and left middle temporal gyrus. Results also revealed an interaction of sex with diagnosis in the amygdala. LIMITATIONS: Data was from two sites, which might increase variability, but it also increases power to examine sex by diagnosis interactions. CONCLUSIONS: This study demonstrates the importance of taking sex differences into account when examining potential trait (or scar) mechanisms that could be useful in identifying individuals at-risk for MDD as well as for evaluating potential therapeutic innovations.

Effects of sex and HIV serostatus on spatial navigational learning and memory among cocaine users.

Spatial learning and memory are critically dependent on the integrity of hippocampal systems. Functional MRI and neuropathological studies show that hippocampal circuitry is prominently affected among HIV-seropositive individuals, but potential spatial learning and memory deficits have not been studied in detail in this population. We investigated the independent and interactive effects of sex and HIV serostatus on performance of a spatial learning and memory task in a sample of 181 individuals with a history of cocaine dependence. We found that men showed faster times to completion on immediate recall trials compared with women and that delayed recall was significantly poorer among HIV-infected compared with HIV-uninfected participants. Additionally, a sex × serostatus effect was found on the total number of completed learning trials. Specifically, HIV-infected men successfully completed more learning trials compared with HIV-infected women. Results are discussed in the context of recent reports of sex and HIV serostatus effects on episodic memory performance.

The menopausal hot flush: a review.

The hot flush is the most characteristic and often the most distressing symptom of the menopause. It is a unique feature and yet the mechanism and health implications are still not fully understood. This review summarizes some of the current thoughts on factors contributing to flushing, the physiological, vascular and neuroendocrine changes associated with flushing and the possible cardiovascular and other health implications for women experiencing hot flushes. Therapy is not discussed.

Schizotypal and affective traits in the offspring of antenatally depressed mothers - Relationship to family history of psychosis in the Northern Finland 1966 Birth Cohort.

BACKGROUND: Maternal depression is relatively common during pregnancy. However, follow-ups of the adult offspring of antenatally depressed mothers are scarce. Previously we found the risk of schizophrenia to be higher in the adult offspring with antenatally depressed mothers and parents with psychosis than in subjects with only one or neither of these risk factors. The aim was to study whether the risk of schizotypal or affective traits differ among adult offspring with antenatally depressed mothers with or without a parental history of psychosis when compared with offspring without antenatally depressed mothers and without parental psychosis. METHODS: In the general population-based Northern Finland 1966 Birth Cohort (NFBC 1966), the mothers of the cohort members were asked at mid-gestation whether they felt depressed. Parental psychosis (Familial Risk, FR) was detected using the Finnish Care Register for Health Care. In the 31-year field study, seven psychometric questionnaires surveyed schizotypal and affective traits in the offspring. The final sample included 4928 individuals (2203 males). RESULTS: There were no statistically significant differences in mean scores on the schizotypal and affective scales between offspring with and without antenatally depressed mothers, or between subjects with and without parental psychosis. The scores were not highest in the subjects with both maternal antenatal depressed mood and FR. CONCLUSION: Surprisingly, maternal depressed mood during pregnancy was unlikely to increase the risk of schizotypy or affective traits in adult offspring, and not even with parental psychosis (FR) in this general population-based birth cohort with about 5000 subjects.

A comparison of the cumulative incidence and early risk factors for psychotic disorder in young adults in the Northern Finland Birth Cohorts 1966 and 1986.

AIMS: Few studies have compared time trends for the incidence of psychosis. To date, the results have been inconsistent, showing a decline, an increase or no significant change. As far as we know, no studies explored changes in prevalence of early risk factors. The aim of this study was to investigate differences in early risk factors and cumulative incidences of psychosis by type of psychosis in two comparable birth cohorts. METHODS: The Northern Finland Birth cohorts (NFBCs) 1966 (N = 12 058) and 1986 (N = 9432) are prospective general population-based cohorts with the children followed since mother's mid-pregnancy. The data for psychoses, i.e. schizophrenia (narrow, spectrum), bipolar disorder with psychotic features, major depressive episode with psychotic features, brief psychosis and other psychoses (ICD 8-10) were collected from nationwide registers including both inpatients and outpatients. The data on early risk factors including sex and place of birth of the offspring, parental age and psychosis, maternal education at birth were prospectively collected from the population registers. The follow-up reached until the age of 27 years. RESULTS: An increase in the cumulative incidence of all psychoses was seen (1.01% in NFBC 1966 v. 1.90% in NFBC 1986; p < 0.001), which was due to an increase in diagnosed affective and other psychoses. Earlier onset of cases and relatively more psychoses in women were observed in the NFBC 1986. Changes in prevalence of potential early risk factors were identified, but only parental psychosis was a significant predictor in both cohorts (hazard ratios ≥3.0; 95% CI 1.86-4.88). The difference in psychosis incidence was not dependent on changes in prevalence of studied early risk factors. CONCLUSIONS: Surprisingly, increase in the cumulative incidence of psychosis and also changes in the types of psychoses were found between two birth cohorts 20 years apart. The observed differences could be due to real changes in incidence or they can be attributable to changes in diagnostic practices, or to early psychosis detection and treatment.

A New Agility Test for Adults: Its Test-Retest Reliability and Minimal Detectable Change in Untrained Women and Men Aged 28-55.

Manderoos, SA, Vaara, ME, Mäki, PJ, Mälkiä, EA, Aunola, SK, and Karppi, S-L. A new agility test for adults: its test-retest reliability and minimal detectable change in untrained women and men aged 28-55. J Strength Cond Res 30(8): 2226-2234, 2016-The aims of this study were to present a new Agility Test for Adults (ATA), to investigate its test-retest reliability and to quantify minimal detectable change at the 95% confidence interval (MDC95). Both the relative and absolute reliabilities were evaluated. Altogether 52 healthy untrained volunteers (25 women: age 43.3 ± 6.6 years; 27 men: age 42.8 ± 7.2 years) were recruited into the study. The subjects performed 3 ATA tests repeated after 2 different intervals: the first test session was baseline, session 2 was a week later, and session 3 was half an hour after session 2. The intraclass correlation coefficient and the SEM of the performance time of ATA were 0.91 and 0.27 seconds (same day), 0.94 and 0.20 seconds (1 week) for women, and 0.95, 0.13 seconds, and 0.94, 0.19 seconds for men, respectively. MDC95 was 0.76 seconds (same day) and 0.56 seconds (1 week) for women, and respectively 0.37 and 0.51 seconds for men. The results showed that ATA is stable and reliable when evaluating agility characteristics in untrained adults. The properties of ATA make it appropriate for screening people to find early signs of declined agility and allow possibility to clinicians and physical trainers to monitor true changes in performance time at agility test by applying the knowledge of MDC95 coefficient. Furthermore, ATA can give tips for planning appropriate exercise programes to prevent clumsiness and falls with more serious consequences among aging people.

Childcare use and overweight in Finland: cross-sectional and retrospective associations among 3- and 5-year-old children.

BACKGROUND: Different types of non-parental childcare have been found to associate with childhood overweight in several, but not all studies. Studies on the matter are mainly North American. OBJECTIVES: The objective of our study was to examine associations between childcare use and overweight in Finland. METHODS: The cross-sectional and partly retrospective data consists of 1683 3- and 5-year-old children participating in the Child Health Monitoring Development project (LATE-project) conducted in 2007-2009 in Finland. Children were measured at health check-ups and information on child's age when entering childcare, the number of childcare places the child has had, current type of childcare (parental, informal, [group] family childcare, childcare centre) and the current amount of childcare (hours) were gathered. Parents' body mass indices, family educational level, family structure, maternal smoking during pregnancy and child's birth weight were treated as covariates. RESULTS: Beginning childcare before age 1 (adjusted model: odds ratio [OR] 2.53, 95% confidence interval [CI] 1.41-4.52) and, for girls only, number of childcare places (adjusted model: OR 1.33, 95% CI 1.11-1.60), were associated with an increased risk of overweight. The current type of childcare or the time currently spent in childcare was not associated with overweight. CONCLUSION: Beginning childcare before age 1, which is quite rare in Finland, and having attended several childcare places were associated with overweight even when adjusting for family socioeconomic status and other family background variables. The significance of these findings needs to be further studied.

Individually modifiable risk factors to ameliorate cognitive aging: a systematic review and meta-analysis.

A number of health and lifestyle factors are thought to contribute to cognitive decline associated with age but cannot be easily modified by the individual patient. We identified 12 individually modifiable interventions that can be implemented during midlife or later with the potential to ameliorate cognitive aging. For ten of these, we used PubMed databases for a systematic review of long-duration (at least 6 months), randomized, controlled trials in midlife and older adults without dementia or mild cognitive impairment with objective measures of neuropsychological performance. Using network meta-analysis, we performed a quantitative synthesis for global cognition (primary outcome) and episodic memory (secondary outcome). Of 1038 publications identified by our search strategy, 24 eligible trials were included in the network meta-analysis. Results suggested that the Mediterranean diet supplemented by olive oil and tai chi exercise may improve global cognition, and the Mediterranean diet plus olive oil and soy isoflavone supplements may improve memory. Effect sizes were no more than small (standardized mean differences 0.11-0.22). Cognitive training may have cognitive benefit as well. Most individually modifiable risk factors have not yet been adequately studied. We conclude that some interventions that can be self-initiated by healthy midlife and older adults may ameliorate cognitive aging.

Interaction between parental psychosis and early motor development and the risk of schizophrenia in a general population birth cohort.

BACKGROUND: Delayed motor development in infancy and family history of psychosis are both associated with increased risk of schizophrenia, but their interaction is largely unstudied. AIM: To investigate the association of the age of achieving motor milestones and parental psychosis and their interaction in respect to risk of schizophrenia. METHODS: We used data from the general population-based prospective Northern Finland Birth Cohort 1966 (n=10,283). Developmental information of the cohort members was gathered during regular visits to Finnish child welfare clinics. Several registers were used to determine the diagnosis of schizophrenia among the cohort members and psychosis among the parents. Altogether 152 (1.5%) individuals had schizophrenia by the age of 46 years, with 23 (15.1%) of them having a parent with psychosis. Cox regression analysis was used in analyses. RESULTS: Parental psychosis was associated (P<0.05) with later achievement of holding the head up, grabbing an object, and walking without support. In the parental psychosis group, the risk for schizophrenia was increased if holding the head up (hazard ratio [HR]: 2.46; degrees of freedom [df]=1; 95% confidence interval [95% CI]: 1.07-5.66) and touching the thumb with the index finger (HR: 1.84; df=1; 95% CI: 1.11-3.06) was later. In the group without parental psychosis, a delay in the following milestones increased the risk of schizophrenia: standing without support and walking without support. Parental psychosis had an interaction with delayed touching thumb with index finger (HR: 1.87; df=1; 95% CI: 1.08-3.25) when risk of schizophrenia was investigated. CONCLUSIONS: Parental psychosis was associated with achieving motor milestones later in infancy, particularly the milestones that appear early in a child's life. Parental psychosis and touching the thumb with the index finger had a significant interaction on risk of schizophrenia. Genetic risk for psychosis may interact with delayed development to raise future risk of schizophrenia, or delayed development may be a marker of other risk processes that interact with genetic liability to cause later schizophrenia.

Difficulty in making contact with others and social withdrawal as early signs of psychosis in adolescents--the Northern Finland Birth Cohort 1986.

AIM: Social withdrawal is among the first signs of the prodromal state of psychosis seen in clinical samples. The aim of this prospective study was to find out whether difficulty in making contact with others and social withdrawal precede first episode psychosis in the young general population. METHODS: The members of the Northern Finland Birth Cohort 1986 (n=6274) completed the PROD-screen questionnaire in 2001-2002. The Finnish Hospital Discharge Register was used to detect both new psychotic and non-psychotic disorders requiring hospitalisation during 2003-2008. RESULTS: Twenty-three subjects developed psychosis and 89 developed a non-psychotic mental disorder requiring hospitalisation during the follow-up. Of those who developed psychosis, 35% had reported difficulty or uncertainty in making contact with others and 30% social withdrawal in adolescence. In hospitalised non-psychotic disorder, the corresponding precentages were 10 and 13% and in the control group without hospital-treated mental disorder 9 and 11%. The differences between psychotic and non-psychotic hospitalised subjects (P<0.01) as well as controls (P<0.001) were statistically significant regarding difficulty or uncertainty in making contact with others. CONCLUSIONS: In this general population-based sample self-reported difficulty or uncertainty in making contact with others in adolescence preceded psychosis specifically compared to hospitalised non-psychotic mental disorders and controls.

Longitudinal associations between childhood and adulthood externalizing and internalizing psychopathology and adolescent substance use.

BACKGROUND: Emotional and behavioral problems are commonly associated with substance use in adolescence but it is unclear whether substance use precedes or follows mental health problems. The aim was to investigate longitudinal associations between externalizing and internalizing psychopathology and substance use in a prospective population study design. METHOD: The sample was the Northern Finland Birth Cohort 1986 Study (NFBC 1986; n = 6349; 3103 males). Externalizing and internalizing mental health problems were assessed at age 8 years (Rutter scales), substance use and externalizing and internalizing problems [Youth Self-Report (YSR)] at age 15-16 years, and hospital diagnoses for internalizing disorders (age 25) and criminal offences (age 20) from nationwide registers in adulthood. RESULTS: Externalizing problems at age 8 were associated with later substance use. After adjustment for sociodemographic factors, parental alcohol use and psychiatric disorders, and earlier externalizing and internalizing problems, substance use predicted criminality, especially among males, with the highest odds ratio (OR) for cannabis use [adjusted OR 6.2, 95% confidence interval (CI) 3.1-12.7]. Early internalizing problems were not a risk for later substance use. Female adolescent cannabis (OR 3.2, 95% CI 1.4-7.3) and alcohol (OR 2.1, 95% CI 1.1-4.2) use predicted internalizing disorders in adulthood. CONCLUSIONS: Externalizing problems precede adolescent substance use in both genders, whereas, among boys, substance use also precedes criminal offences. Internalizing problems may follow substance use in females. These associations were robust even when taking into account previous mental health problems.