H2 "Developmental specificity of risk factors for substance use disorders over the life span"

Submitted by: Dr. rer. nat. Silke Behrendt

Duration: 10/2009 - 10/2017

Background: Epidemiological research shows that substance use disorders (SUD) are prevalent in all age groups from adolescence to old adulthood. However, epidemiological research also reveals differences between age groups in terms of incidence and prevalence rates and type of SUD consequences. For example, SUD incidence is more likely in adolescence than in old adulthood.
To understand the etiology of mental disorders (here: SUD) and to be able to inform health care providers and public health decision makers about possible targets for prevention and intervention, it is of importance to identify risk factors for the incidence, stability, and recurrence of the disorder. For SUD, past research has mostly focused on identifying SUD predictors in young samples, because the first main incidence phases of SUD are in the second and third decade of life. In contrast, studies on SUD prediction in other age groups as middle-aged and older adults are scarce. However, because of differences in resources and life circumstances between age groups, generalization from risk factors identified among adolescents and young adults to older adulthood appears unwarranted. Moreover, some developmental periods as adolescence are characterized by large developmental changes. Here, the timing of onset of a risk factor within a developmental period may be of importance for SUD development. However, whether the age of onset of a risk factor within one developmental period is related to SUD risk remains understudied to date. There is also limited knowledge on putative age-related differences in SUD symptomatology to date, which is an important prerequisite for comparing risk factors between different age groups and with different ages of onset.
Thus, in terms of understanding the pathogenesis of SUD, it remains unclear to date, whether predictors of SUD are specific for certain ages and developmental stages. This research gap has important implications: to date, predictors of SUD in older subjects remain poorly understood. This knowledge is lacking in treatment and preventive efforts for this growing segment of ageing western populations. It remains unclear whether non-normative timing of onset of a risk factor is relevant for SUD risk within adolescence and young adulthood and which putative specific and shared mechanisms might underlie SUD risk factors at different ages.
Aims: Given this background, the overarching aim of this thesis is to investigate the developmental specificity of risk factors of SUD by 1) investigating age-related differences in the symptomatology of SUD in adolescents/young adults and in older adults, 2) contributing to the understanding of the developmental specificity of risk factors within the developmental period of adolescence/young adulthood, 3) providing information on similarities and differences in the role of single mental disorders in the prediction of SUD in adolescence/young adulthood in comparison to middle-aged/older adulthood and by 4) providing information on similarities and differences in the role of psychopathology risk profiles in the prediction of SUD in adolescence/early adulthood in comparison to middle- aged/older adulthood.

Methods: Data from three studies were used in the analyses. The prospective-longitudinal Early Developmental Stages of Psychopathology (EDSP) study includes a representative community sample of adolescents and young adults from municipal Munich (n= 3021 at baseline) that was followed over up to four assessment waves in up to ten years (baseline age range: 14-24 years; age range at ten-year follow-up: 21-34 years). The DEGS1-MH study is a cross-sectional epidemiological study with a sample representative for the German national population aged 18 to 79 years (n=5303). In the analysis conducted for this thesis, data from the subsample aged 55 years and older (n=2045) was used. The ELDERLY-study is an international multicenter randomized controlled clinical trial in which the effects of two short psycho-social interventions for alcohol use disorders (AUD) among the elderly are investigated. The baseline study sample consists of n=693 treatment seeking subjects aged 60 and older and diagnosed with DSM-5 12-month AUD. Due to high rates of missing information from the New Mexican study site, data from this site was excluded from the analysis presented here leading to n=544. In all three studies, diagnoses of SUD and other mental disorders were achieved with established instruments for classificatory diagnoses, the Munich Composite International Diagnostic Interview (DIA-X/M-CIDI), an adapted version thereof (DEGS-CIDI), and the MINI International Neuropsychiatric Interview (M.I.N.I.). Substance use (SU) patterns were either assessed with the CIDI (EDSP and DEGS1- MH study) or the well-established Form-90 (ELDERLY-Study). The latter instrument follows a time-line follow-back approach. Statistical methods applied were, for example, Cox- Regression Analysis and Latent Class Analysis with auxiliary variables.

Results: 1) SUD symptomatology in different age groups: Overall, prevalence rates of AUD symptoms in different age groups did not reveal distinct age-related symptomatology patterns for AUD in adolescents/young adults compared to older adults. However, for specific symptoms as hazardous use differential endorsement patterns were noted in different age groups.
2) Developmental specificity of risk factors within the developmental period of adolescence and early adulthood: an earlier onset of specific but not all mental disorders as major depression within the developmental period of adolescence and young adulthood was associated with the risk of incident AUD. An earlier age at onset of cannabis use (CU) was associated with a higher risk of transition to a cannabis use disorder (CUD) independent of an earlier onset of alcohol and nicotine use and of externalizing disorders and parental SUD. 3) Risk factor identification in different age groups: among adolescents/young adults, a range of different mental disorders predicted AUD even after controlling for other mental disorders. In contrast, among older adults, mental disorders were largely unpredictive of problematic alcohol use (AU). 4) Risk profile identification in different age groups: in both young and old subjects, psychopathology risk profiles characterized by high probabilities of male gender and low mental disorder probabilities as well as risk profiles related to SUD were predictive of AUD respectively problematic alcohol use. Of note, risk profiles characterized by internalizing disorders were only related to alcohol-related outcomes among adolescents/young adults.
Conclusions: Results on symptomatology indicate that AUD in adolescence/early adulthood respectively old adulthood have shared features and similarities as well as age-specific characteristics regarding the report of certain symptoms. Overall however, using the example of AUD, there was no evidence that SUD in youth may differ substantially from those late in life or even constitute a distinct disorder entity. Furthermore, results on risk factors within the developmental period of adolescence/early adulthood indicated that development-related factors are of relevance for the association between specific mental disorders and AUD and between early onset of CU and CUD risk. Risk factors differ by age group which may indicate different underlying mechanisms due to age-related changes in resources and life-circumstances. The more similar results regarding SUD-related risk profiles in different age groups may be due to the stability of underlying psychopathology vulnerability dimensions through life.
In conclusion, it may be assumed that the pathogenesis of SUD in terms of its clinical presentation and predictors of risk is subject to age-specific as well as -unspecific factors. Research on risk factors in SUD etiology has to take into consideration not only different substances and diagnoses but also different ages and age groups in order to deepen the understanding of SUD as they occur over the lifespan. In future research, epidemiological studies with sufficient statistical power to address chronicity and different age groups among the elderly as well as high methodological comparability for between age group comparisons are needed.

Publications

1. Are clinical features of alcohol use disorders age-specific?

• Publication 1: Behrendt, S., Buehringer, G., Perkonigg, A., Lieb, R., Beesdo-Baum, K.,

  2013. Characteristics of developmentally early alcohol use disorder symptom reports: a prospective-longitudinal community study. Drug and Alcohol Dependence 131, 308- 315.

• Publication 2: Behrendt, S., Braun, B., Bilberg, R., Bühringer, G., Søgaard-Nielsen, A., Bogenschutz, M., Andersen, K., submitted. Clinical characteristics of older adults seeking outpatient-treatment for DSM-5 alcohol use disorder in two countries: a multicenter study. Journal of Substance Abuse Treatment.

2. Are risk factors for substance use disorders within adolescence/young adulthood development-specific?

• Publication 3: Behrendt, S., Beesdo-Baum, K., Zimmermann, P., Höfler, M., Perkonigg,

  A., Bühringer, G., Lieb, R., Wittchen, H.U., 2011. The role of mental disorders in the risk and speed of transition to alcohol use disorders among community youth. Psychological Medicine 41, 1073–1085.

• Publication 4: Behrendt, S., 2017. The role of age in the onset and further development of cannabis use disorders. In: Preedy, V. (Ed.), The Handbook of Cannabis and Related Pathologies: Biology, Diagnosis, Treatment, and Pharmacology. Academic Press, London. 138-150.

• Publication 5: Behrendt, S., Beesdo-Baum, K., Höfler, M., Perkonigg, A., Bühringer, G., Lieb, R., Wittchen, H.-U., 2012. The relevance of age at first alcohol and nicotine use for initiation of cannabis use and progression to cannabis use disorders. Drug and Alcohol Dependence 123, 48-56.

3. Are psychopathology risk factors in adolescents/young adults and in middle-aged/older adults development-specific?

• Publication 3 (see point 2)

• Publication 6: Behrendt, S., Buehringer, G., Beesdo-Baum, K., Strehle, J., Jacobi, F.,

  Wittchen, H.-U., accepted for publication. Mental disorders, gender, and partnership

  in problem alcohol use risk at age 55+. SUCHT.

4. Are psychopathology risk profiles in adolescents/young adults and in middle-aged/older adults development-specific?

• Publication 6: (see point 3)

• Publication 7: Behrendt, S., Bühringer, G., Höfler, M., Lieb, R., Beesdo-Baum, K., 2017.

  Prediction of incidence and stability of alcohol use disorders by latent internalizing psychopathology risk profiles in adolescence and young adulthood. Drug and Alcohol Dependence 179, 32-41.