BackgroundAlthough unemployment and high levels of perceived stress have been associated in cross-sectional studies, the direction of causation is unknown. We prospectively examined if high levels of perceived everyday life stress increased the risk of subsequent unemployment and further if differences existed between socioeconomic status-groups.MethodsWe included 9335 18–64-year-old employed respondents of a health survey (North Denmark Health Profile 2010) in which Cohen’s Perceived Stress Scale was used to assess the level of perceived stress. Data were linked individually to national administrative registers. Cox proportional hazards model was used to investigate the association between perceived stress quintiles and risk of unemployment during 98 weeks of follow-up. Analyses were further performed in subgroups defined by education and income.ResultsIn total, 224 people (10.4%) of the high stress group became unemployed during follow-up, which was higher than the lower stress groups. After adjusting for gender, age, education and income, the risk of unemployment was 1.64 (95% CI: 1.28;2.11) in the high stress group compared to the low stress group. After adjusting for gender and age, a similar trend was observed across different education levels and among the lower income groups, but no higher risk of unemployment due to perceived stress was found among the higher income groups. However, there was no statistically significant interaction between perceived stress and income level (p = 0.841) or perceived stress and education level (p = 0.587).ConclusionPerceived everyday life stress nearly doubled the risk of subsequent unemployment in a working population. No statistically significant interactions between SES and perceived stress were found. This indicates that stress prevention among the working population should not solely focus on stress in the workplace but also include stress from everyday life.Electronic supplementary materialThe online version of this article (10.1186/s12889-018-5618-z) contains supplementary material, which is available to authorized users.

BackgroundFrequent attendance to primary care constitutes a large use of resources for the health care system. The association between frequent attendance and illness-related factors has been examined in several studies, but little is known about the association between frequent attendance and individual social capital. The aim of this study is to explore this association.MethodsThe analysis is conducted on responders to the North Denmark Region Health Profile 2010 (n = 23,384), individually linked with information from administrative registers. Social capital is operationalized at the individual level, and includes cognitive (interpersonal trust and norms of reciprocity) as well as structural (social network and civic engagement) dimensions. Frequent attendance is defined as the upper-quartile of the total number of measured consultations with a general practitioner over a period of 148 weeks.ResultsUsing multiple logistic regression, we found that frequent attendance was associated with a lower score in interpersonal trust [OR 0.86 (0.79–0.94)] and social network [OR 0.88 (0.79–0.98)] for women, when adjusted for age, education, income and SF12 health scores. Norms of reciprocity and civic engagement were not significantly associated with frequent attendance for women [OR 1.05 (0.99–1.11) and OR 1.01 (0.92–1.11) respectively]. None of the associations were statistically significant for men.ConclusionThis study suggests that for women, some aspects of social capital are associated with frequent attendance in general practice, and the statistically significant dimensions belonged to both cognitive and structural aspects of social capital. This association was not seen for men. This indicates a multifaceted and heterogeneous relationship between social capital and frequent attendance among genders.Electronic supplementary materialThe online version of this article (10.1186/s12889-018-5230-2) contains supplementary material, which is available to authorized users.