ObjectivesThe high prevalence of burnout among medical residents and specialists raises concerns about the stressful demands in healthcare. This study investigated which job demands and job resources and personal resources are associated with work engagement and burnout and whether the effects of these demands and resources differ for medical residents and specialists.DesignIn a survey study among residents and specialists, we assessed job demands, job resources, personal resources, work engagement and burnout symptoms using validated questionnaires (January to December 2017). Results were analysed using multivariate generalised linear model, ordinary least squares regression analyses and path analyses.SettingFive academic and general hospitals in the Netherlands.ParticipantsA total number of 124 residents and 69 specialists participated in this study. Participants worked in the fields of pediatrics, internal medicine and neurology.ResultsThe associations of job and personal resources with burnout and work engagement differed for residents and specialists. Psychological capital was associated with burnout only for specialists (b=−0.58, p<0.001), whereas psychological flexibility was associated with burnout only for residents (b=−0.31, p<0.001). Colleague support (b=0.49, p<0.001) and self-compassion (b=−0.33, p=0.004) were associated with work engagement only for specialists.ConclusionThis study suggests that particularly personal resources safeguard the work engagement and lessen the risk of burnout of residents and specialists. Both residents and specialists benefit from psychological capital to maintain optimal functioning. In addition, residents benefit from psychological flexibility, while specialists benefit from colleague support. Personal resources seem important protective factors for physicians’ work engagement and well-being. When promoting physician well-being, a one-size-fits-all approach might not be effective but, instead, interventions should be tailored to the specific needs of specialists and residents.