The amplification of target DNA by highly specific probes using the polymerase chain reaction (PCR) provides a highly sensitive and specific method for the detection of malaria infection. The use the of PCR in settings with varying endemicity within one survey area has not been investigated intensively. Therefore, a cross-sectional study was conducted in the districts of Kabarole and Bundibugyo in western Uganda using material from three villages with different epidemiologic situations regarding malaria and DNA primers for a PCR that had shown satisfactory sensitivity and specificity in previous trials. The sensitivity of the PCR varied significantly (P < 0.001) in the three survey villages (between 63.2% and 83.9% for the primer pair K1-14-1 and between 37.9% and 69.9% for the primer pair MSP-1) and was highly linked to geographic differences and social exchanges of the inhabitants with other areas of the district. According to the results of this investigation, it is advisable not to use a single primer pair in epidemiologic field studies for the detection of falciparum malaria. The use of combined primer pairs and the frequent confirmation of the results by microscopy are recommended.