This study compared SARS-CoV-2 RNA loads at different anatomical sites, and the impact of self-swabbing and food intake. Adult symptomatic patients with SARS-CoV-2 or non-SARS-CoV-2 respiratory tract infection were included between 2021 and 2022. Patients performed a nasal and buccal swab before a professionally collected nasopharyngeal/oropharyngeal swab (NOPS). Buccal swabs were collected fasting and after breakfast in a subgroup of patients. SARS-CoV-2 RNA loads were determined by nucleic acid testing. Swabbing convenience was evaluated using a survey. The median age of 199 patients was 54 years (interquartile range 38−68); 42% were female and 52% tested positive for SARS-CoV-2. The majority of patients (70%) were hospitalized. The mean SARS-CoV-2 RNA load was 6.6 log10 copies/mL (standard deviation (SD), ±1.5), 5.6 log10 copies/mL (SD ± 1.9), and 3.4 log10 copies/mL (SD ± 1.9) in the professionally collected NOPS, and self-collected nasal and buccal swabs, respectively (p < 0.0001). Sensitivity was 96.1% (95% CI 90.4−98.9) and 75.3% (95% CI 63.9−81.8) for the nasal and buccal swabs, respectively. After food intake, SARS-CoV-2 RNA load decreased (p = 0.0006). Buccal swabbing was the preferred sampling procedure for the patients. In conclusion, NOPS yielded the highest SARS-CoV-2 RNA loads. Nasal self-swabbing emerged as a reliable alternative in contrast to buccal swabs. If buccal swabs are used, they should be performed before food intake.