BackgroundAccording to the Common-Sense Model of Illness Representations, illness beliefs, such as causal attributions, can influence the way people assess and cope with their illness and vice versa. To date, causal attributions in people with depressive symptoms have been studied mainly cross-sectionally, quantitatively and independently. The purpose of this study is to examine the causal attributions of people with depressive symptoms in terms of their stability over time, dependence on treatment experience, and differentiation of causal concepts.MethodsIn a population-based prospective sample, people with at least mild depressive symptoms (PHQ-9 Score ≥ 5) were interviewed via telephone at T0 and twelve months later (T1). Causal attributions were assessed using the Brief Illness Perception Questionnaire. After the open responses were qualitatively analysed using a deductive-inductive approach, stability over time was assessed for causal attributions and concepts by comparing answers between the two time points. Subsequent exploratory quantitative analyses were conducted using chi-square tests, t-tests, and logistic regression analyses.ResultsA total of 471 individuals (age M = 53.9, 53.6% female) with a mean PHQ-9 Score of 8.4 were included in the analyses. Causal attributions related to participants’ social environment, workplace, and past are the most stable over time. However, individuals with and without a time-stable causal concept showed no differences in terms of sociodemographic characteristics, severity of depressive symptoms, risk of comorbidity, and treatment experiences. Overall, the causal concepts of people with depressive symptoms appear to be very diverse. Those with treatment experience (M = 2.21, SD = 0.80) named significantly more causal attributions compared to people without treatment experience (M = 1.98, SD = 0.81, t(471) = -3.060, p < 0.01). In addition, logistic regression analyses revealed that treatment-experienced respondents were more likely to attribute “childhood/youth/parental home” and “predisposition”.ConclusionsOur study reveals that people with treatment experience tend to report treatment-congruent causal attributions, such as childhood and family environment, as well as predisposition, more frequently. Understanding how causal attributions and concepts are formed and change can be helpful for addressing causal attributions in treatment. Future studies should take into account the benefits of employing qualitative survey methods for exploring causal attributions.