Background: While loneliness is a global public health problem, the literature lacks studies assessing loneliness predictors in low- and middle-income countries. Therefore, we aimed to analyze clinical and lifestyle predictors of loneliness. Methods: We conducted a 2-year longitudinal study in Brazil based on a snowball sample and online surveys (baseline: May 6 to June 6, 2020). We assessed clinical and lifestyle predictors of loneliness using multiple regression models. The analyses were adjusted for several sociodemographic variables and weighted for attrition and sampling procedures. Results: The study included a nationwide sample of 473 participants (18-75 years; 87.1% females). After adjusting for sociodemographic factors, we identified as risk factors: depressive symptoms (RR: 1.214; 95%CI: 1.08-1.36; p = 0.001), anxiety symptoms (RR:1.191; 95%CI: 1.04-1.35; p = 0.007), alcohol abuse (RR: 1.579; 95%CI: 1.32-1.88; p < 0.001), and cannabis use (RR: 1.750; 95%CI: 1.25-2.39; p < 0.001). More than 150 min/week of physical activity (RR: 0.177; 95%CI: 0.07-0.34; p < 0.001) and good/excellent quality of family relationships (RR: 0.73; 95%CI: 0.60-0.87; p < 0.001) and sleep (RR: 0.483; 95%CI: 0.39-0.59; p < 0.001) were protective factors. Conclusion: Several clinical factors (depression, anxiety, alcohol, and cannabis) have been identified as risk factors for loneliness, while lifestyle factors (physical activity, better quality of sleep, and family relationships) have been associated with a lower incidence of loneliness. Addressing clinical and lifestyle factors may therefore be essential to preventing loneliness.