Summary The present meta‐analysis investigates the associations between leadership, followers' mental health, and job performance by taking into account different groups of leadership constructs including transformational leadership, relations‐oriented leadership, task‐oriented leadership, destructive leadership, and leader–member exchange. Six categories of mental health‐related outcomes are considered representing both negative and positive mental health states of followers, namely, affective symptoms, burnout, stress, well‐being, psychological functioning, and health complaints. Meta‐analytic models are used to estimate the association between these categories of leadership and mental health. Our results reveal that transformational leadership, a high quality of relations‐oriented and task‐oriented leadership behavior, as well as a high quality of leader–follower interaction are positively associated with mental health. In contrast, destructive leadership is strongly negatively associated with mental health. In addition, the mediation effects of leadership on job performance via mental health are estimated. Results partially support the mediating role of mental health concerning the relationship between leadership and job performance. Our results emphasize the importance of leadership as an occupational health factor, and they may serve as basis for the planning and designing of occupational health policies and interventions despite existing research limitations. Copyright © 2016 John Wiley & Sons, Ltd.

Immigrants 'bring along' their suicide risk, at least for the initial period they spend in the immigration country. Health-care planners and providers need to be aware of this 'imported risks'. However, most immigrant groups do not have an increased suicide risk relative to the local-born population; some may even experience substantially lower risks.

BackgroundStillbirth is a sensitive indicator for access to, and quality of health care and social services in a society. If a particular population group e.g. migrants experiences higher rates of stillbirth, this might be an indication of social deprivation or barriers to health care. This study examines differences in risk of stillbirth for women of different regions of origin compared to women from Germany in order to identify high risk groups/target groups for prevention strategies.MethodsWe used the BQS dataset routinely compiled to examine perinatal outcomes in Germany nationwide. Participation of hospitals and completeness of data has been about 98% in recent years. Data on all live births and stillbirths were obtained for the period 2004 to 2007 (N = 2,670,048). We calculated crude and stratified mortality rates as well as corresponding relative mortality risks.ResultsA significantly elevated stillbirth rate was found for women from the Middle East and North Africa (incl. Turkey) (RR 1.34, CI 1.22-1.55). The risk was slightly attenuated for low SES. An elevated risk was also found for women from Asia (RR 1.18, CI 1.02-1.65) and from Mediterranean countries (RR 1.14, CI 0.93-1.28). No considerable differences either in use and timing of antenatal care or preterm birth and low birthweight were observed between migrant and non-migrant women. After stratification for light for gestational age, the relative risk of stillbirth for women from the Middle East/North Africa increased to 1.63 (95% CI 1.25-2.13). When adjusted for preterm births with low birthweight, women from Eastern Europe and the Middle East/North Africa experienced a 26% (43%) higher risk compared with women from Germany.ConclusionsWe found differences in risk of stillbirth among women from Middle East/North Africa, especially in association with low SES and low birthweight for gestational age. Our findings suggest a need for developing and evaluating socially and culturally sensitive health promotion and prevention programmes for this group. The findings should also stimulate discussion about the quality and appropriateness of antenatal and perinatal care of pregnant women and newborns with migrant backgrounds.