Rationale The present study aims to investigate the pattern of cross-lagged relationships between job insecurity and self-rated health over a period of five years. While health complaints are usually seen as one of the detrimental outcomes of job insecurity, the question of the direction of the job insecurity-health relationship has not yet been fully resolved. Only a few longitudinal studies have explicitly aimed to test the possibility of reciprocal or reverse effects, and even fewer studies have used multi-wave designs to examine the pattern of these relationships. Objective The current study aims to address this gap by testing how cross-lagged relationships between job insecurity and self-rated health status unfold over time. Method We conducted this study with a sample of the working population in Switzerland (N = 928), using the data from five consecutive measurement occasions, each separated by a one year lag. Cross-lagged structural equation modelling was performed to examine the direction of the effects. Results The results revealed an interchangeable direction of the relationship between job insecurity and health over time. T1 job insecurity predicted lower ratings of health at T2, which then predicted job insecurity at T3, which, in turn, was related to lower health at T4. The only exception was observed in the last follow-up (i.e., T4 to T5), where no evidence of cross-lagged relationships between job insecurity and self-rated health was found. Conclusions These findings contribute to the literature suggesting that, not only may job insecurity predict later health impairment, but that in some cases, the reverse may be possible too. Researchers and policy makers need to consider this important message because the observed lagged reciprocal effects between job insecurity and health seem to form a negative cycle over time, thereby implying a dual process in the development of workplace vulnerabilities.