Associations between air pollution and chronic kidney disease (CKD) have been reported, but studies at low exposure levels and relevant exposure time windows are still warranted. This study investigated clinical CKD at low air pollution levels in the Swedish Malmö Diet and Cancer Cohort in different exposure time windows. This study included 30,396 individuals, aged 45-74 at enrollment 1991-1996. Individual annual average residential outdoor PM2.5, PM10, nitrogen oxides (NOx), and black carbon (BC) were assigned using dispersion models from enrollment to 2016. Diagnoses of incident CKD were retrieved from national registries. Cox proportional hazards models were used to obtain hazard ratios (HRs) for CKD in relation to three time-dependent exposure time windows: exposure at concurrent year (lag 0), mean exposure in the 1-5 or 6-10 preceding years (lag 1-5 and lag 6-10), and baseline exposure. During the study period, the average annual residential exposures were 16μg/m3 for PM10, 11μg/m3 for PM2.5, 26μg/m3 for NOx, and 0.97μg/m3 for BC. For lag 1-5 and lag 6-10 exposure, significantly elevated HRs for incident CKD were found for total PM10:1.13 (95% CI: 1.01-1.26) and 1.22 (1.06-1.41); NOx: 1.19 (1.07-1.33) and 1.13 (1.02-1.25) and BC: 1.12 (1.03-1.22) and 1.11 (1.02-1.21) per interquartile range increase in exposure. For total PM2.5 the positive associations of 1.12 (0.97-1.31) and 1.16 (0.98-1.36) were not significant. For baseline or lag 0 exposure there were significant associations only for NOx and BC, not for PM. Residential exposure to outdoor air pollution was associated with increased risk of incident CKD at relatively low exposure levels. Average long-term exposure was more clearly associated with CKD than current exposure or exposure at recruitment. Our findings imply that the health effects of low-level air pollution on CKD are considerable.