Browsing by Subject "Latent profile analysis"

Eating disorders pose a serious threat to the physical and mental health of those who suffer from them. Given the impact of these disorders and difficulty treating them, it is important to understand the nature of them and factors involved in their development and maintenance. The empirical investigation of eating disorders is made difficult by the extreme heterogeneity seen within current DSM-IV diagnostic categories. Research on emotion regulation in eating disorders is rising, yet scant in terms of identifying specific difficulties and points of intervention. The proposed study focused on the construct of emotion regulation and its relation to the eating disorders by: (1) empirically identifying groups of eating disorder participants based on symptoms and (2) examining specific difficulties in emotion regulation capacities in each of the identified groups of eating disorder participants and (3) identifying whether difficulties in emotion regulation contribute to eating disorder symptom severity. A clinical sample of individuals with eating disorders was classified into subgroups based on symptom frequency using latent profile analysis. The most parsimonious and best fitting model was a four-profile solution which resulted in four distinct subgroups. Profile 1 consisted of individuals who endorsed moderate restriction and occasional binge eating and vomiting, all at a subclinical level. Profiles 2, 3, and 4 all met criteria for bulimia nervosa and consisted of individuals who engaged in restriction, binge eating, and purging though in varying degrees. When these groups were compared to a sample of college aged healthy control participants using multivariate analysis of variance, results indicate that individuals in profile 1, which comprised 71% of the sample, experience greater difficulty with emotion regulation in the areas of awareness, nonacceptance, and perceived access to strategies to help them feel better. Results of three hierarchical regression analyses showed that difficulties in emotion regulation did not significantly contribute to symptom severity as anticipated..

Background: Globally, traumatic brain injury (TBI) accounts for the highest burden of deaths and disabilities globally. Studies suggest a complex relationship between injury, mental health, and alcohol. Though hazardous alcohol use and TBI exert heavy burdens in Tanzania, their interaction with mental health is largely unknown. This study aims to explore the mental health and alcohol use profiles of TBI patients in a low-income country.

Methods: Secondary data analysis of a registry of adults (≥ 18) with TBI of any severity presenting to the Kilimanjaro Christian Medical Center Emergency Department (ED) within 24 hours of injury. Patient data were collected at ED arrival and at three months follow-up. Variables included measures of functional independence, psychiatric health, quality of life, and alcohol use. Hazardous alcohol use was defined as an Alcohol Use Disorder Identification Test (AUDIT) score greater than seven. We conducted a latent profile analysis (LPA) to determine pre-injury mental health profiles of patients and logistic regression to assess association of patient profile with hazardous drinking at three months after injury.

Results: Of 190 participants, 51 (26.8%) were hazardous drinkers. The majority of the sample was male (83.7%) and the median age was 29.5 years. The LPA model with the strongest fitness revealed five profiles of mental health and drinking patterns. The “Poor Mental Health Drinkers” (9.4%) profile had worse quality of life and higher depression and hazardous drinking scores. The “Disabled Non-drinkers” (11.4%) profile had worse motor functional independence and low hazardous drinking scores. The “Non-drinkers” (53.5%) had good quality of life, little to no depression, good functional independence and low hazardous drinking scores. The “Drinkers” were similar to the “Non-drinkers” profile, except with high hazardous drinking scores. Predictors of hazardous drinking three months post-injury included disability and being a hazardous drinker before injury.

Conclusions: This study provides insight into the possible mental health and drinking pattern profiles for TBI patients. The categorization of patients may help in resource allocation of alcohol interventions for those who are at the highest risk for hazardous alcohol use. Limitations included recall bias for pre-injury information.