Objectives: Reported associations of physical exposures during work (eg, lifting, trunk flexion or rotation) and low-back pain (LBP) are rather inconsistent. Mechanical back loads (eg, moments on the low back) as a result of exposure to abovementioned risk factors have been suggested to be important as such loads provide a more direct relationship with tissue failure and thus LBP. Since information on the effect of such load metrics with LBP is lacking yet, we aimed to assess this effect in a prospective study. Methods: Of 1131 workers, categorised into 19 groups, LBP was prospectively assessed over 3 years. Video and hand force recordings of 4-5 workers per group (93 in total) were used to estimate mechanical low-back loads ( peak load and three cumulative load metrics, ie, linear weighted load, squared weighted load and load weighted to the tenth power) during manual materials handling (MMH) tasks using a video analysis method. These data were combined with static mechanical load estimates based on structured observation of non-MMH tasks. Associations of mechanical loads and LBP were tested using generalised estimating equations. Results: Significant effects on LBP were found for cumulative low-back moments (linear and squared weighted; both p<0.01 and ORs of 3.01 and 3.50, respectively) but not for peak and cumulative moments weighted to the tenth power. Conclusions: Results of this first prospective study on the effect of mechanical low-back load on LBP support a LBP aetiology model of cumulative loads, potentially due to accumulation of microdamage or fatigue. Therefore, prevention of LBP should focus on reducing cumulative low-back loads, especially in highly exposed occupational groups, for example, by reducing handling of heavy loads and working in awkward body postures.