This thesis investigated work-related well-being from seven rarely studied angles, e.g., the role of negative life events and pre-employment resource losses, and work engagement were explored. The data sets were a three-wave 35-year follow-up questionnaire data (N = 532), a questionnaire data based on the staff of a large educational organization (N = 3365), and qualitative interviews of the 22 most burned out participants in the 35-year prospective study. The main results of the study were: 1) Adverse socio-economic and individual conditions in childhood were negatively associated with educational achievements, which in turn exposed to jobs with less resources, and hence, led to burnout symptoms and furthermore to poor health and increased intentions of early retirement. 2) The instability of the work career during 13 years of follow-up was positively associated with burnout and negatively with life satisfaction. 3) The role of negative life events, family-to-work conflict and personality factors (strong sense of responsibility and sense of coherence) in the burnout process was small compared with the role of working conditions and work-to-family conflict. 4) however, work and non-work stressors, as well as work and personality factors had some joint effects on burnout and life satisfaction. 5) Burnout could be interpreted in accordance with Hobfoll's conservation of resources (COR) theory as a loss spiral of resources, while at the same time the qualitative data analysis made it possible to refine some of the general assumptions of the COR theory. Strong initial motivation or enthusiasm seemed to be a prerequisite for burnout only in the case of some of the interviewed employees. 6) CFA confirmed the factorial validity of the Finnish version of the UWES. Work engagement was positively related to health, work ability, and job satisfaction, and negatively to intentions of quitting one's job and early retirement. Women, those with fixed-term work contracts, those with less than 5 years or more than 30 years' tenure in the present job, as well as those with long working hours, were more engaged than their counterparts. 7) the hypothesized Job Demands - Resources model was partly supported.