BACKGROUND: Obtaining a nasal swab (NS) from a child for human rhinovirus (HRV) RNA detection is simple and well tolerated even for repeated sampling, but only few studies have compared them qualitatively and quantitatively with other sampling methods. OBJECTIVES: Real-time PCR was used to study the stability of HRV genomes in swabs, and to compare different swabs and induced sputum specimens with nasopharyngeal aspirates (NPAs). STUDY DESIGN: Replicate swabs in a dry test tube were stored at room temperature or mailed to the laboratory before freezing, and compared to freshly frozen specimens. To compare sampling methods, paediatric patients had NPA, NS and throat swab collected. In paired sputum and NPA specimens, viral load was correlated to the amount of beta-actin mRNA. RESULTS: Specimens were stable at room temperature for at least 4 days and survived mailing without loss of HRV detectability. As compared to NPA, NS had an equal diagnostic sensitivity, with no significant quantitative difference using flocked nylon swabs and a 2.2-fold drop in the average copy number using cotton swabs. The diagnostic sensitivity of cotton swab-collected throat specimens was 97%, with a 26-fold lower mean copy number. Sputum specimens had higher HRV RNA (2.3-fold) and beta-actin mRNA (1.6-fold) copy numbers than NPAs, but there was a poor correlation between HRV RNA and beta-actin mRNA. CONCLUSION: HRV remains well detectable by PCR in specimens mailed to the laboratory. The diagnostic efficacy of NPA can be obtained with NS, quantitative comparison and patient comfort favouring flocked nylon-tipped over cotton-tipped swabs.