BackgroundEffective methods are needed to collect fecal samples from children for large-scale microbiota studies. Stool collected on fecal occult blood test (FOBT) cards that can be mailed provides an effective solution; however, the quality of sequencing resulting from this method is unknown.The aim of this study is to compare microbiota metrics of 16S ribosomal RNA (rRNA) gene sequencing from stool and meconium collected on FOBT cards with stool collected in an Eppendorf tube (ET) under different conditions.MethodsEight stool samples from children in diapers aged 0 month–2 years and three meconium samples were collected and stored as follows: (1) ≤ 2 days at room temperature (RT) in an ET, (2) 7 days at − 80 °C in an ET, (3) 3–5 days at RT on a FOBT card, (4) 7 days at RT on a FOBT card, and (5) 7 days at − 80 °C on a FOBT card. Samples stored at − 80 °C were frozen immediately. Each specimen/condition underwent 16S rRNA gene sequencing with replicates on the Illumina MiSeq. Alpha and beta diversity measures and relative abundance of major phyla were compared between storage conditions and container (ET vs. FOBT card), with pairwise comparison between different storage conditions and the “standard” of 7 days at − 80 °C in an ET and fresh stool in an ET.ResultsStool samples clustered mainly by individual diaper (P < 10−5, Adonis), rather than by storage condition (P = 0.42) or container (P = 0.16). However, meconium samples clustered more by container (P = 0.002) than by individual diaper (P = 0.009) and storage condition (P = 0.02).Additionally, there were no differences in alpha diversity measures and relative abundance of major phyla after Bonferroni correction between stool stored on a FOBT card at RT for 7 days with stool stored in an ET tube at − 80 °C; differences in alpha diversity were seen however when compared to fresh stool in an ET. Overall, based on the intraclass correlation coefficient (ICC), the different storage containers/conditions are reliable in preserving the microbial memberships and slightly less reliable in preserving the alpha diversity and relative microbial composition of infant stool.ConclusionsAcknowledging certain limitations, FOBT cards may be a useful tool in large-scale stool microbiota studies in children requiring outpatient follow-up where only small amounts of stool can be obtained, but should not be used when studying meconium.Electronic supplementary materialThe online version of this article (10.1186/s40168-017-0333-z) contains supplementary material, which is available to authorized users.