Longitudinal Studies on Loneliness and Smoking Behaviours

Cross-sectional studies cannot disentangle the temporal relationship between loneliness and smoking. In this study, we focused on the longitudinal relationship from loneliness to smoking behaviour among adolescents and young adults. So far, only a few longitudinal studies have assessed this longitudinal relationship in these populations. According to Park (2009), loneliness in school predicting smoking initiation among Korean elementary-school students (4th grade onwards) in bivariate analyses, while the findings were not robust in multivariate analysis controlling for other factors. Qualter et al. (2013), examined different trajectories of loneliness (age 7 to 17) but did not find support for any trajectory predicting smoking status (yes/no). Although these studies provide valuable insights into possible longitudinal associations between loneliness and smoking initiation/status among adolescents/young adults, methodological limitations (e.g. self-reporting, significant drop-outs, bias by uncontrolled confounding) might have impacted the findings. Consequently, more longitudinal studies, across different cultures, are needed to confirm the findings from these two existing longitudinal studies.

Other available studies focused on older samples, e.g. working adults (Stenlund et al., 2021) or older adults (> 50 years; Luo & Waite, 2014; Philip et al., 2022), or utilized other study designs. For instance, a Mendelian randomization study, where genetic variants are utilized as instrumental variables to assess potential causal effects (Richmond & Davey Smith, 2022), did report weak evidence for increased loneliness leading to higher likelihood of initiating smoking, smoking more cigarettes, and a lower likelihood of quitting smoking in adults (Wootton et al., 2021).

Taking this together, limited research showed no or weak evidence for a longitudinal relationship from loneliness to smoking (in adolescents or adults), and as far as we know, no study has specifically focused on young adults (e.g. university students). Additionally, it is important to disentangle how loneliness might be longitudinally associated with different smoking behaviours, i.e. smoking initiation, being a current smoker (smoking status), or smoking frequency (among smokers). Consequently, additional longitudinal studies are needed to shed new light on the possible longitudinal relationship from loneliness to smoking among adolescents and young adults, considering different smoking behaviours. These insights are essential, as longitudinal studies can yield new insights that can facilitate the development of interventions that address loneliness and reduce smoking behaviour among youth (Stickley et al., 2014).