ORIGINAL ARTICLE Reciprocity and Depressive Symptoms in Belgian Workers A Cross-Sectional Multilevel Analysis Bart De Clercq, MSc, Els Clays, PhD, Heidi Janssens, MD, Dirk De Bacquer, PhD, Annalisa Casini, PhD, France Kittel, PhD, and Lutgart Braeckman, MD, PhD Objective: This study examines the multidimensional association between reciprocity at work and depressive symptoms. Methods: Data from the Belgian BELSTRESS survey (32 companies; N = 24,402) were analyzed. Multilevel statistical procedures were used to account for company-level as- sociations while controlling for individual-level associations. Results: Dif- ferent dimensions of individual reciprocity were negatively associated with depressive symptoms. On the company level, only vertical emotional reci- procity was negatively associated (β =−4.660; SE = 1.117) independently from individual reciprocity (β =−0.557; SE = 0.042). Complex interactions were found such that workplace reciprocity (1) may not uniformly benefit individuals and (2) related differently to depressive symptoms, depending on occupational group. Conclusions: This study extends the existing literature with evidence on the multidimensional, contextual, and cross-level interac- tion associations of reciprocity as a key aspect of social capital on depressive symptoms. G iven its wide societal impact, mental health problems are a major concern for public health and occupational medicine. 1,2 In a recent publication, Jung 3 outlined the rationale to study work- place social capital as a psychosocial determinant of mental health problems in occupational settings. This study contributes to this line of inquiry, with an empirical test of a model linking workplace reci- procity to depressive symptoms. RECIPROCITY, SOCIAL CAPITAL, AND DEPRESSIVE SYMPTOMS Reciprocity can be defined as a social dynamic whereby per- sons give, receive, and return. 4 Every donation contains the request to return what has been received and, therefore, implies reciprocity. Social relations at work are typically characterized by certain norms of reciprocity. For example, employees provide emotional support and help each other in solving problems, not knowing but assum- ing that someday someone will return the favor. This idea was al- ready formulated in the convoy model of Kahn and Antonucci 5 and Antonucci and Akiyama, 6 in which an individual is seen as a traveler through life, surrounded by members of his cohort who share expe- riences and life events (eg, a cohort of colleagues) and who provide support to one another reciprocally over time. As a part of the concept of social capital, reciprocity is misunderstood, undertheorized, and rarely measured. 7 From a “social network perspective,” reciprocity From the Department of Public Health (Mr De Clercq and Drs Clays, Janssens, De Bacquer, and Braeckman), Faculty of Medicine and Health Sciences, Ghent University, Ghent, Belgium; and School of Public Health (Drs Casini and Kittel), Universit´ e Libre de Bruxelles, Brussels, Belgium. The BELSTRESS study was funded by the Belgian Federal Public Service Employment, Labor and Social Dialogue and the European Social Fund. The authors declare no conflicts of interest. Address correspondence to: Bart De Clercq, MSc, Department of Public Health, Faculty of Medicine and Health Sciences, Ghent University, Academical Hospital, Block A-2, De Pintelaan, 185, B-9000 Ghent, Belgium (b.declercq@ugent.be). Copyright C  2013 by American College of Occupational and Environmental Medicine DOI: 10.1097/JOM.0b013e31828dc8cd is an objective network characteristic—analogous to other character- istics such as density, homogeneity, and duration—that describes the extent to which exchanges or transactions are reciprocal. 8 The more subjective “social cohesion perspective” assesses rather “perceived” norms of reciprocity as both an individual construct and an eco- logical construct. 9 Being a cognitive component of social capital, norms of reciprocity are seldom defined; however, it is often as- sumed that they refer to several forms of social support. 10,11 Social support is typically divided into subtypes: emotional, instrumental, appraisal, and informational support. 12 Emotional support involves the provision of empathy, trust, and caring; instrumental support refers to practical help; appraisal support relates to help in decision making and giving appropriate feedback; and informational support refers to the facility of advice and information leading to a solution to problems. Note that there is a considerable disagreement about how social capital and social support relate to each other: some re- searchers explicitly differentiate between both the constructs, while others conceptualize social support as a component of social capital. Two central arguments support the first view. One is that social capi- tal is a feature of the social structure, while social support is a feature of the individual actors within the social structure. 13 This duality al- ready emerged in the early writings of Bourdieu, 14 Coleman, 15 and Putnam 16 and persisted in two main schools of thought regarding the definition of social capital. 9,17 The second argument is that, among other mechanisms, social support is a pathway between social capi- tal and health. 18 In contrast, others acknowledge social support as a cognitive component of social capital 11,19 or, to cite Carpiano’s 20,21 resource-based theory, as a specific “form” of social capital. This study conceptualizes social capital via norms of reciprocity mea- sured both at the individual level and at the workplace level. This kind of collective treatment of social capital has been criticized for its failure to clearly separate the causes of social capital from its consequences. 22 In this respect, the provision of feedback (ap- praisal support), advise, or information (informational support) may be rather consequences of the more basic form of instrumental reci- procity. For example, Coleman 15 considers access to information as a consequence of the possession of social capital. Therefore, the theo- retical focus of the present study was on emotional and instrumental reciprocity. More generally, workplace social capital can be defined as the shared values, attitudes, and norms of trust and reciprocity as well as practices of collective action in the work unit. 23,24 Overall, the associations between social capital and mental health problems are well documented, 25,26 and the majority of these studies have adopted the social cohesion approach of social cap- ital focusing on contextual influences of geographic entities like countries, 27 states, 28 and communities. 29 Meanwhile, evidence on social capital in workplace settings remains sparse. 30 Only a small number of multilevel studies investigating the individual and con- textual relationships between workplace social capital and health exist, 23,24,31–39 of which only two 33,38 examined the multilevel asso- ciation between workplace social capital and depression. Kouvonen et al 33 found that low individual workplace social capital was as- sociated with the onset of depression, but this association was not confirmed on the contextual level via the work unit coworkers’ as- sessment of social capital. This finding is in contrast with the other Copyright © 2013 Lippincott Williams & Wilkins. Unauthorized reproduction of this article is prohibited. 824 JOEM  Volume 55, Number 7, July 2013