Reporting delays of AIDS cases in France could depend on such factors as geographic area of diagnosis, risk group or time of diagnosis. In order to identify the main sources of heterogeneity and to obtain realistic estimates of the number of diagnosed AIDS cases in France, we take into account these reporting co-factors by using a regression model. Results show significant variations between the 4 geographical areas considered (Ile-de-France, Provence-Alpes-Côte d'Azur, 3 south-west regions, and the rest of metropolitan France). Reporting delay improved since 1986 in Ile-de-France and the 17 other areas of metropolitan France, with an earlier report of heterosexual cases. In the two other areas, no difference can be evidenced between reporting delays, nor improvement of such delay over time. The estimation discrepancies obtained are more noticeable over the last concerned year and will therefore have an impact on the epidemic growth previsions. The correction methods should therefore take into account these disparities in order to obtain more realistic predictions of the diagnosed AIDS cases.