Organizational predictors and health consequences of changes in burnout: A 12-year cohort study MICHAEL P. LEITER 1 * , JARI J. HAKANEN 2 , KIRSI AHOLA 2 , SALLA TOPPINEN-TANNER 2 , AKI KOSKINEN 2 AND ARI VÄÄNÄNEN 2 1 Acadia University, Wolfville, NS, Canada 2 Finnish Institute of Occupational Health, Helsinki, Finland Summary We investigated job burnout and job characteristics, including decision authority, skill discretion, predict- ability, and information flow, among Finnish forestry workers (N = 4356) in a longitudinal study. We linked these responses individually with data on the participants’ subsequent prescriptions for psychotropic drugs including antidepressants. We aim to study the antecedents of changes in burnout levels over four years time and their health-related consequences in an eight-year follow-up. The results showed that inconsistency among the levels of the Maslach Burnout Inventory subscales (e. g., high scores in exhaustion and low cynicism or vice versa) at baseline identified patterns that were prone to change in burnout four years later. Information flow predicted the direction of this change for the exhaustion and cynicism aspects of burnout, whereas skill discretion and predictability did so for reduced professional efficacy. Change toward burnout predicted future risk of psychotropic drug use. It seems that adverse changes in burnout are influenced by poor organizational resources, and change toward burnout is likely to elevate the risk of poor mental health. Copyright © 2012 John Wiley & Sons, Ltd. Keywords: burnout; health; longitudinal; change Introduction Job burnout, a syndrome of chronic exhaustion, cynicism, and inefficacy, is an ongoing problem in work life that distresses employees and weakens the financial performance of organizations. Losing the capacity to be energetically involved in work, something that normally strengthens self-efficacy, imposes huge costs on personal fulfillment and organizational productivity. Extensive research has established clear connections between health outcomes and burnout (Halbesleben & Buckley, 2004; Maslach, Schaufeli, & Leiter, 2001; Shirom & Melamed, 2005). Further understanding of these connections would benefit from focusing on the actual process of burning out and using more objective measures of health. The present study addresses these points. A relevant part of the challenge to decrease burnout and its detrimental consequences concerns the chronic nature of burnout. Indeed, research on job burnout has found the underlying constructs—energy, involvement, and efficacy—to be remarkably stable over time (Hakanen & Schaufeli, 2012; Hakanen, Schaufeli, & Ahola, 2008; Maslach et al., 2001; Schaufeli, Leiter, & Maslach, 2009). Ironically, despite burnout’s considerable stability, models of the syndrome emphasize the multidimensional process that evolves over time (Leiter, 1993; Maslach et al., 2001), and longitudinal, regression-based research designs are an established standard for research on burnout. However, in a large sample, the people who are in flux are easily lost against the backdrop of stability. Thus, examining the sample as a whole does not foster understanding of burnout as a dynamic process. A method for identifying people who are about to change would be valuable. In order to gain an insight into the process of burnout, it is essential to have an a priori rationale for separating more volatile people from stable ones and to identify the dynamics of change. The process tested *Correspondence to: Michael P. Leiter, Acadia University, Wolfville, NS Canada B4P 2R6. E-mail: michael.leiter@acadiau.ca Copyright © 2012 John Wiley & Sons, Ltd. Received 17 November 2011 Revised 19 August 2012, Accepted 24 August 2012 Journal of Organizational Behavior, J. Organiz. Behav. 34, 959–973 (2013) Published online 20 September 2012 in Wiley Online Library (wileyonlinelibrary.com) DOI: 10.1002/job.1830 Research Article