In Canada, nasopharyngeal swabs are preferred to oropharyngeal ones. We aimed to compare the sensitivity of NP swabs and saliva for SARS-CoV-2 detection in hospitalized patients. In the study NP swab and saliva were collected on the day of enrollment in patients who were confirmed to have COVID-19 with an NP, mid turbinate, or nasal swab. For saliva specimens, patients were asked to spit 5 mL of saliva into a sterile container and then was tested by RT-PCR. In 61% of the patients, both NP swab and saliva were positive, 28% only the NP swab was positive, and in 11% only saliva was positive (P = 0.02). Using NP swabs only would have detected 89% patients with at least 1 positive specimen and using saliva only would have detected 72% patients with at least 1 positive specimen NP swabs only would have detected 94%, 89%, and 82% patients in their first, second, and third week of illness, respectively. Using saliva only would have detected 88%, 66% and 71% patients in their first, second, and third/fourth week of illness. These results prove that NP swabs are more sensitive for SARS-CoV-2 detection. However, collection of NP swabs are uncomfortable for patients and may pose a risk to healthcare workers. Saliva, in contrast, can be easily self-collected. More data is needed to assess testing and sensitivity of different specimen types in asymptomatic patients or those whose illness does not require hospitalization.