The present study investigates whether there is an excess incidence of thyroid cancer among people living in the vicinity of the nuclear sites in Belgium. Adjusted Rate Ratios were obtained from Poisson regressions for proximity areas of varying sizes. In addition, focused hypothesis tests and generalized additive models were performed to test the hypothesis of a gradient in thyroid cancer incidence with increasing levels of surrogate exposures. Residential proximity to the nuclear site, prevailing dominant winds frequency from the site, and simulated radioactive discharges were used as surrogate exposures. No excess incidence of thyroid cancer was observed around the nuclear power plants of Doel or Tihange. In contrast, increases in thyroid cancer incidence were found around the nuclear sites of Mol-Dessel and Fleurus; risk ratios were borderline not significant. For Mol-Dessel, there was evidence for a gradient in thyroid cancer incidence with increased proximity, prevailing winds, and simulated radioactive discharges. For Fleurus, a gradient was observed with increasing prevailing winds and, to a lesser extent, with increasing simulated radioactive discharges. This study strengthens earlier findings and suggests increased incidences in thyroid cancer around two of the four Belgian nuclear sites. Further analyses will be performed at a more detailed geographical level.