Hepatitis A virus isolates from anti-HAV IgM positive sera of 70 hepatitis cases in two outbreaks and 216 other cases in Central America, 136 sporadic cases and 53 cases from an hyper-endemic region in Costa Rica, were compared by phylogenetic analyses within the VP1 region. The outbreaks in all 531 cases, in 1992 and 1999, respectively, were presumed water borne. In the first outbreak, HAV RNA could be detected in 70% of the cases sampled during 6 weeks after onset of jaundice. In the hyper-endemic region of San Ramón in Costa Rica, 1,932 cases were registered between 1972 and 1985. All isolates belonged to subtype 1A. Background isolates from Costa Rica and El Salvador tended to form separate subclusters in the phylogenetic tree construction and were mostly unrelated to subtype 1A strains from other parts of the world. Based on their amino acid sequences, four HAV strains, all related to CR326 sampled in Costa Rica in 1960, were found to have circulated in the area during the last three decades. However, on the basis of nucleotide variability the isolates from the outbreaks could be distinguished from the strains from sporadic cases and sequence analysis could confirm the epidemiological homogeneity of both outbreaks. In the hyper-endemic region, 16 different sequences were encountered forming one single subcluster. Thus, limited sequencing within the VP1 region proved useful to identify outbreaks of hepatitis A in a highly endemic area, where most strains were local and only one subtype was prevalent.