Using Latent Class Analysis to Characterize Sickness Signaling in Relation to Familism and Public and Private...

In response to contracting an infection, individuals usually display a suite of external signs (including sickness behavior) as an outward indication of illness. This context-dependent phenomenon seems to weigh the benefits and costs of eliciting sympathy by indicating sickness versus hiding signs of illness to avoid exposing others to potential infection. In a dynamically social species like humans, non-kin may be as likely to respond to these signs with care as family members, particularly fellow church members. We explore the relative contributions of religiosity and familism in shaping self-reported sickness signaling styles as two dimensions central to human altruism using latent class analysis (LCA).

LCA was used to characterize the signaling styles of the study participants. Data come from a large 2018 survey (n = 1259) of sickness and health behaviors among US adults. We used denomination public (church attendance) and private religiosity (time spent in prayer, meditation, etc.) and the God Locus of Health Control scale to assess the impact of God on health. Sickness signaling style was assessed with the SicknessQ and three additional items. Covariates included age, gender, education, and income.

We identified four classes (Familiar, Moderate, Gregarious, and Stoic) tied to signaling styles. The Familiar Signaling class displayed sickness verbally to familiar others, were the oldest, and were least guided by an internal sense of religion. The Moderate Signaling class was younger and had lower public and private religiosity (except regarding health issues) than the Stoic and Gregarious Signaling classes. The Gregarious class signaled to both close others and strangers and scored highest in familism and religiosity. The Stoic class did not verbally signal but indicated sickness worsening when around both close others and strangers, were less likely to be married or endorse private religiosity, and were least likely to recall recent illness as severe.

The signaling classes strongly resembled aspects of the introvert-ambivert-extrovert spectrum. We conclude that variation is important at multiple levels, including personality types, and potentially prevents the loss of immunological diversity.