Clinical recommendation	Evidence rating	Comments
The pretest probability of COVID-19 should be based on the patient's exposure to someone with a confirmed or probable case, signs or symptoms of COVID-19, local or population-specific COVID-19 prevalence, and presence of an alternative diagnosis.8,25,27 A leaf plot can aid in visualizing how pretest probability and test characteristics impact posttest probability.	C	Expert opinion
A negative molecular or antigen test result might not rule out SARS-CoV-2 infection when pretest probability is high.13,25,27 Therefore, clinicians should recommend isolation precautions despite a negative test result when pretest probability is high.	C	Expert opinion, one systematic review of low-quality studies with inconsistent results
Use a symptom-based approach for discontinuing isolation precautions for most patients with COVID-19.21,32 Reverse transcriptase polymerase chain reaction detection of viral RNA does not necessarily correlate with infectivity.	C	Consensus and disease-oriented evidence
Antibody tests may help identify past SARS-CoV-2 infection if performed two to four weeks after symptom onset.36,37 Antibody test results should not yet be used to infer immunity to SARS-CoV-2 infection or inform decisions to discontinue social distancing or use of face masks or personal protective equipment.37	B	One systematic review of low-quality studies; consensus and disease-oriented evidence