Measuring the level of social support using latent class analysis

Area of Science:

• Epidemiology
• Psychosocial Health
• Mental Health Research

Background:

• The Medical Outcomes Study Social Support Scale (SSS-MOS) is widely used but lacks measures of social support levels related to health risks.
• Epidemiological studies often lack nuanced measures of social support, limiting understanding of its impact on health outcomes.

Purpose of the Study:

• To utilize latent class analysis (LCA) to identify subgroups with varying levels of perceived social support.
• To assess the association between these identified social support levels and the prevalence of Common Mental Disorders (CMD) among mothers.

Main Methods:

• A cross-sectional study involving 1013 mothers in Salvador, Brazil.
• Psychosocial data collected via home visits using the SSS-MOS and Self Reporting Questionnaire-20.
• Latent class analysis (LCA) employed to define social support classes; logistic regression used to analyze associations with CMD.

Main Results:

• LCA successfully distinguished two distinct classes of perceived social support: high and low.
• Mothers in the low social support class reported significantly lower probabilities (13.7-59.8%) of perceiving various support types compared to the high support class (84.3-98%).
• A higher prevalence of Common Mental Disorders (CMD) was observed among mothers classified as having low levels of social support.

Conclusions:

• Latent class analysis (LCA) is an effective tool for refining the measurement of perceived social support.
• Identifying distinct levels of social support, particularly low levels, is crucial for understanding increased CMD prevalence.
• This approach enhances the measurement of social support in epidemiological studies, linking it directly to mental health risks.