Abstract BACKGROUND One hospital in the city of Cali, Colombia, of the ECLAMC (Latin‐American Collaborative Study of Congenital Malformations) network, reported the unusual occurrence of four cases of sirenomelia within a 55‐day period. METHODS An ECLAMC routine for cluster evaluation (RUMOR) was followed that included: calculations of observed/expected ratios, site visits, comparison with comprehensively collected local, South American, and worldwide data, cluster analysis, and search for risk factors. RESULTS All four Cali sirenomelia cases were born to mothers living in a 2 km 2 area, in neighboring communes, within the municipality of Cali. Considering the total births of the city of Cali as the denominator, and based on ECLAMC baseline birth prevalence rates (per 100,000) for sirenomelia (2.25, 95% CI: 2.66, 3.80), the cluster for this congenital abnormality was unlikely to have occurred by chance (observed/expected ratio = 5.77; 95% CI: 1.57–14.78; p = .002). No consistent common factor was identified, but vicinity to an open landfill as the cause could not be rejected. Another ECLAMC hospital in San Justo, Buenos Aires, Argentina, reported three further cases but these did not seem to constitute a nonrandom cluster. CONCLUSIONS The methodology used to evaluate the two possible clusters of sirenomelia determined that the Cali sirenomelia cluster was unlikely to have occurred by chance whereas the sirenomelia cluster from San Justo seemed to be random. Birth Defects Research (Part A) 2009. © 2008 Wiley‐Liss, Inc.