The diagnostic accuracy of three faecal egg count techniques (Kato-Katz, McMaster and FLOTAC) to assess albendazole efficacy against soil-transmitted helminth (STH) infections was compared.; The study is registered with Current Controlled Trials [identifier: ISRCTN90088840]. During September-November 2009, 304 school-aged children on Pemba Island, Tanzania, were screened and those infected with Ascaris lumbricoides, hookworm or Trichuris trichiura were treated with a single dose of albendazole (400 mg). Twenty-one days post-treatment, children provided a single stool sample which was examined using the same diagnostic methods. All stool samples were divided into two aliquots and one was fixed in 5% formalin and examined using FLOTAC and McMaster approximately 6 months after collection.; Using fresh stool samples, comparable prevalences were demonstrated for the three methods at baseline (90-92.2% for T. trichiura, 41.1-52.8% for hookworm, 32.9-37.2% for A. lumbricoides); FLOTAC was the most sensitive method at baseline and follow-up. Albendazole showed high cure rate (CR) against A. lumbricoides (90-97%), moderate CR against hookworm (63-72%) and very low CR against T. trichiura (6-9%), regardless of the technique used. Egg counts (eggs per gram) at baseline were similar for A. lumbricoides and for hookworm among the three methods, and higher using McMaster and Kato-Katz compared with FLOTAC for T. trichiura. All methods were similar for hookworm and A. lumbricoides egg reduction rate (ERR) estimation, but Kato-Katz indicated a significantly higher ERR than McMaster and FLOTAC for T. trichiura. Preserved stool samples revealed consistently lower FECs at baseline and follow-up for all STHs.; Further development and validation of standard protocols for anthelminthic drug efficacy evaluation must be pursued.