Premium
Latent class analysis of need descriptors within an I rish youth mental health early intervention program toward a typology of need
Author(s) -
Peiper Nicholas,
Illback Robert J.,
O'Reilly Aileen,
Clayton Richard
Publication year - 2017
Publication title -
early intervention in psychiatry
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 1.087
H-Index - 45
eISSN - 1751-7893
pISSN - 1751-7885
DOI - 10.1111/eip.12213
Subject(s) - latent class model , typology , multinomial logistic regression , psychological intervention , mental health , psychology , descriptive statistics , intervention (counseling) , comorbidity , logistic regression , clinical psychology , cluster (spacecraft) , psychiatry , medicine , sociology , mathematics , machine learning , anthropology , computer science , statistics , programming language
Aim Significant overlap and comorbidity has been demonstrated among young people with mental health problems. This paper examined demographic characteristics, heterogeneity of need descriptors and services provided among young people (12–25 years) engaging in brief interventions at J igsaw in the R epublic of I reland. Method Between 1 J anuary 2013 and 31 D ecember 2013, a total of 2571 young people sought help from 1 of 10 J igsaw sites. Of these, 1247 engaged in goal‐focused brief interventions, typically consisting of one to six face‐to‐face sessions. Descriptive statistics were used to summarize social and demographic factors. Latent class analysis was used to cluster young people into relevant typologies of presenting issues. Multinomial logistic regression was then performed to determine significant predictors of class membership. Results The most common age of young people was 16. More women (59.6%) than men engaged in brief interventions, 56% attended school, 74% lived with their family of origin or with one parent, and 54.2% came from families where parents were married. Using established fit criteria, four relevant typologies emerged: Developmental (26.8%), Comorbid (15.8%), Anxious (42.7%) and Externalising (14.6%). Predictors varied by class membership, but general family problems and lack of adult support emerged as the strongest predictors for all classes. Conclusion This study demonstrated that the mental health needs of young people in I reland are significant and diverse. Because J igsaw favours a more descriptive approach to problem identification, the four typologies suggest a need to determine program capacity in engaging youth with heterogeneous presenting issues and to tailor brief interventions to each group's clinical profiles.