Latent class analysis of obesity-related characteristics and associations with body mass index among young children

A Latent Class Analysis of Family Eating Behaviors and Home Environment Habits on Preschool-Aged Children's Body Mass Index

Background: The etiology of obesity is multifaceted, with multiple risk factors occurring during early childhood (e.g., fast food frequency, eating dinner as a family, TV in the bedroom). Many past studies have largely considered obesity risk factors in isolation, when in reality, the risk factors likely cluster together. A latent class analysis can be used to identify patterns in child eating behaviors, parent feeding behaviors, and household habits among preschool-aged children and their families to identify distinct, heterogenous classes and to determine if classes are associated with overweight and obesity. Methods: We used data from a community-based study of 624 three- to five-year-old children and a parent in New Hampshire, from March 2014 to October 2015. Parent-reported data were used to determine frequency of eating behaviors and household habits. Height and weight were objectively measured. Results: Four classes were identified; Class 1: "Healthy/Mildly accommodating," Class 2: "Healthy/Accommodating," Class 3: "Moderately healthy/Moderately accommodating," and Class 4: "Least healthy/Least accommodating." Compared with Class 1, children in Class 4 had increased odds of being overweight or obese [adjusted odds ratio (aOR): 1.64, 95% confidence interval (CI): 1.13-2.15], whereas Classes 2 and 3 were not associated with BMI (Class 2: aOR: 1.24, 95% CI: 0.62-1.86; Class 3: aOR: 1.31, 95% CI: 0.81-1.81). Conclusion: Study findings highlight that child-parent interactions around meals differentially relate to children's weight status given the context of children's eating habits. Most important, our study findings confirm the importance of adapting multiple healthy habits within the home social and physical environment to offset obesity risk in young children.

Association of maternal physical and mental health characteristics with the hazard of having any medical condition or disability in Australian children: A 15-year birth cohort study

Any long-term medical condition or disability among children is a significant health issue. This study measured the incidence rate of any medical condition or disability among children from a nationally representative birth cohort, then used the random effect parametric survival regression model to assess whether the hazard of any medical condition or disability in children is associated with maternal physical and mental health characteristics (obesity, general health status, having a medical condition, stressful life events or mental illness). The study followed up 5019 children from the Longitudinal Study of Australian Children, assessing their time-to-event data from birth (2004) to 14 or 15 years of age (2018). The hazard rate of any medical condition or disability was 26.11 per 1000 person-years for all the children and 29.29 for the males-a noticeable gender difference. It was the highest (hazard rate: 62.90) among the children when their mothers had a medical condition, while the hazard rate was 22.40 per 1000 person-years among the children whose mothers had no medical conditions. The parametric panel regression results also suggested that the children of mothers with a medical condition during the 15-year study period were more likely to have a medical condition or disability (hazard ratio [HR]: 2.61, 95% confidence interval [CI]: 2.24-3.02) compared to the children of mothers with none. Similar trends were observed among children of mothers who had fair or poor general health (HR: 1.48, 95% CI: 1.15-1.91), obesity (HR: 1.40, 95% CI: 1.18-1.66) or experienced stressful life events (HR: 1.23, 95% CI: 1.06-1.43) over time compared to those whose mothers did not. These findings suggest that additional healthcare interventions targeting mothers with medical conditions, obesity, poor general health, or mental illness would help minimise the risk of medical conditions and disabilities among children.

Clustering of lifestyle and health behaviours in Australian adolescents and associations with obesity, self-rated health and quality of life

OBJECTIVE

The primary aim of this study was to identify clusters of lifestyle and health behaviours and explore their associations with health outcomes in a nationally representative sample of Australian adolescents.

METHODS

The study participants were 3127 adolescents aged 14-15 years who participated in the eighth wave of the birth cohort of the Longitudinal Study of Australian Children (LSAC). A latent class analysis (LCA) was performed to identify clusters based on the behaviours of physical activity, alcohol consumption, smoking, diet, eating disorders, sleep problems and weight consciousness. Multinomial logistic regression models were fitted to the following health outcome variables: obesity, self-rated general health and pediatric health-related quality of life, to investigate their associations with LCA clusters.

RESULTS

Based on the prevalence of health behaviour related characteristics, LCA identified gender based distinct clusters of adolescents with certain outward characteristics. There were five clusters for male and four clusters for female participants which are named as: healthy lifestyle, temperate, mixed lifestyle, multiple risk factors, and physically inactive (male only). Adolescents in the healthy lifestyle and temperate clusters reported low and moderately active health risk behaviours, for example, low physical activity, inadequate sleep and so on, while these behaviours were prevailing higher among adolescents of other clusters. Compared to adolescents of healthy lifestyle clusters, male members of physically inactive (OR = 3.87, 95% CI: 1.12 - 13.33) or mixed lifestyle (OR = 5.57, 95% CI: 3.15 - 9.84) clusters were over three to five times more likely to have obesity; while for female adolescents, members of only multiple risk factors clusters (OR = 3.61, 95% CI: 2.00 - 6.51) were over three time more likely to have obesity compared to their counterpart of healthy lifestyle clusters. Adolescents of physically inactive (b = -9.00 for male only), mixed lifestyle (b = -2.77 for male; b = -6.72 for female) or multiple risk factors clusters (b = -6.49 for male; b = -6.59 for female) had a stronger negative association with health-related quality of life scores compared to adolescents of healthy lifestyle clusters.

CONCLUSION

The study offers novel insights into latent class classification through the utilisation of different lifestyles and health-related behaviours of adolescents to identify characteristics of vulnerable groups concerning obesity, general health status and quality of life. This classification strategy may help health policy makers to target vulnerable groups and develop appropriate interventions.

Risk Behavioral Contexts in Adolescence of Obese Adults

PURPOSE

Previous research suggests that poor nutrition, physical activity, sleep, and social/emotional climate are associated with weight gain. However, few empirical studies have examined how these factors relate to each other in adolescents who are later obese. Are these factors uniformly present, or do some co-occur or occur independently? This study seeks to identify subgroups of obese individuals at ages 24-32 years who exhibited unique, co-occurring behavioral and emotional contexts for obesity at ages 14-17 years.

METHODS

To identify subgroups of behavioral and contextual profiles in adolescence, the study applies latent class analysis to a sample of individuals who were obese in the fourth wave of the National Longitudinal Study of Adolescent to Adult Health (Add Health, N = 1,889). The study then explored covariates (e.g., gender, race) of class membership.

RESULTS

Considerable heterogeneity exists in risk profiles of adolescents obese as adults. For example, 21.1 percent of the sample is in a class with no differentiating risk factors, whereas two classes containing 22.1 percent of the sample exhibit high levels of depression, and nearly all the emotional factors are considered. Although some covariates are predictive of class membership, clear patterns are difficult to discern. However, poor physical health is clearly predictive of membership in the classes exhibiting a high risk of depression.

DISCUSSION

Clinicians should be aware that at younger ages, people who are ultimately obese display a range of factors linked to obesity. Although some exhibit behaviors such as high screen time and processed food consumption, others exhibit mainly poor social/emotional climate.

Identifying Excessive Intake of Oil and Salt to Prevent and Control Hypertension: A Latent Class Analysis

Introduction

To identify health hazard behaviors and provide a basis for targeted management and intervention for patients with hypertension, we classified their health-related behaviors.

Methods

A multi-stage random sampling method was used to conduct an on-site questionnaire survey among residents aged ≥15 years in a certain urban area of Taiyuan City, Shanxi Province, China. A latent class analysis was used to classify the lifestyle behaviors of patients with hypertension. The lifestyle behavior characteristics of different types of patients with hypertension and their awareness of hypertension were assessed.

Results

The prevalence of hypertension in Taiyuan City was 19.5%. Patients with hypertension were classified into three clusters according to their lifestyle patterns: smoking and drinking (13.35%), excessive edible oil and salt intake (68.27%), and healthy behavior (18.38%). Comparing the three latent classes of lifestyle, the distribution of age, sex, marital status, and education level was different (P < 0.05). The awareness of hypertension and the rate of control among the three classes were also different (P < 0.05).

Conclusion

The lifestyle behaviors of patients with hypertension have evident classification characteristics. Approximately two-thirds of the patients with hypertension have an excessive intake of oil and salt. Therefore, targeted and precise intervention measures should be taken to control the intake of oil and salt in this cohort.