The global pandemic of SARS-CoV-2 (severe acute respiratory syndrome coronavirus 2) infection and associated illness (coronavirus disease or COVID-19) have emerged very quickly, challenging traditional systems of clinical and public health response [1,2]. There is broad consensus that adequate testing for SARS-CoV-2 is imperative as a cornerstone of public health efforts to control the spread of the virus [3-5]. The response in the United States to the SARS-CoV-2 pandemic has been hampered by a slow implementation of screening programs and by a variety of factors that have limited the extent of SARS-CoV-2 testing. Following initial problems with quality control of reagents [6], the US government allowed more liberal policies for the development of laboratory developed tests (LDTs) [7], which allowed for a massive expansion of capacity in terms of availability of equipment and staff capacity at commercial laboratories.

However, other factors now limit the reach and volume of testing for SARS-CoV-2 infection. Currently, the gold standard specimen for testing for SARS-CoV-2 RNA is a provider-administered nasopharyngeal swab (NPS) [8]; there are currently supply chain challenges, including shortages of rigid swabs for NPS collection, personal protective equipment (PPE) required for health care workers to collect NPS specimens, and viral transport media required for transporting specimens [9-12]. There are also important questions about testing sites for SARS-CoV-2 infection—should people with mild symptoms who may or may not have COVID-19 come into clinics for testing if they do not require immediate clinical care? How can people who have been quarantined receive testing to document viral shedding to guide the time of release from quarantine, without coming into places where patients are congregating? Finally, we have essentially no epidemiologic data about asymptomatic infection, which could be answered with serology data. Decisions about when to end “stay at home” curfews should likely be based, in part, on the prevalence of antibodies (and, perhaps, immunity) among populations. All these applications will require mechanisms to collect specimens that minimize the need for PPE and allow flexibility of where specimens are collected. Patient-collected samples have appeal in terms of minimizing PPE requirements and enabling the possibility of epidemiologic studies to characterize the infection and immune status of populations.

The regulatory environment governing both the development of LDTs and specifically the testing of patient-collected samples, either in the office of the health care provider or at home, is complicated [13,14]. In response to the urgency of COVID-19 in the United States, the Food and Drug Administration (FDA) has relaxed the regulatory pathways for the development of LDTs to increase the national laboratory capacity [15]. With respect to specimen types, the FDA has developed two pathways for approval for SARS-CoV-2 infection testing: (1) develop laboratory data documenting validation or performance to submit to the FDA for review as an Emergency Use Authorization (EUA); or (2) use state regulatory mechanisms when the test is developed under the jurisdiction of the state in which the lab resides. In the latter case, the state regulators take responsibility for COVID-19 testing by laboratories in their state.

Patient-collected samples have slightly different requirements. The FDA has issued clarifications that the ability of states to oversee the validation of LDTs with patient-collected samples is not covered by the general policy for the development of LDTs, and that assays with patient-collected samples will be required to submit data for review through the FDA for an EUA application [15]. The FDA recently updated guidance to allow collection of patient-collected mid-nasal turbinate samples collected in the provider office, but specifically noted that this approval did not extend to patient-collected samples collected at home [16]. Guidelines for the validation of patient-collected serology specimens do not appear to be explicitly addressed under current FDA guidance, which has focused on the collection of samples for viral detection.

Outside of the realm of SARS-CoV-2 diagnosis, the FDA has reviewed and approved patient-collected samples for a wide variety of laboratory assays, including HIV serology through dried blood spot (DBS) specimens [17]. In other fields, there is a long history of using patient-collected samples under research protocols to develop data on acceptability and to provide clinical services (eg, STI [sexually transmitted infection] testing) as part of research studies [18,19].

We anticipate that there will be biological, immunologic, and temporal aspects that will be important to consider in the design of validation studies for alternative specimen types and results interpretation. When directly comparing provider-collected and patient-collected samples of the same type (eg, provider-collected versus patient-collected oropharyngeal swab [OPS]), it is appropriate to compare the cycle threshold (C_t) for the paired samples to assess comparability. However, RNA concentrations may differ between different specimen types because of differences in RNA shedding at the two sites, such that a direct comparison of C_t results between nasopharyngeal and oropharyngeal swabs might not be an appropriate comparison. Similarly, the timing of onset and waning of IgM (immunoglobulin M) titers in patients and detection of IgA (immunoglobulin A) in saliva or serum following infection with SARS-CoV-2 are not well understood. Finally, it is unclear whether RNA might persist for variable lengths of time after infection in different specimen types. For example, To et al [20] document the presence of RNA in saliva for nearly 2 weeks post hospitalization. Ultimately, all of these questions need to be explored to make the most evidence-based recommendations for specimen screening types and collection methods in specific time phases of the infection cycle.

In this paper, we lay out a protocol for describing the sufficiency of patient-collected samples for SARS-CoV-2 infection testing in OPS and saliva, and for immune response to SARS-CoV-2 in DBS and saliva. We consider two aspects of assessment:

Providers will observe participants as they collect specimens through a telehealth video session on a HIPPA (Health Insurance Portability and Accountability Act of 1996)-compliant video conference service. After establishing a secure connection and confirming the identity of participants using a study identifier, the provider will direct the participant to use the provided instructions to collect the specimen; the provider will identify their role as that of an observer. The provider will document on case report forms (CRFs) that they observed the participant collecting the specimens (see Multimedia Appendix 1 for CRFs). The provider will also record their assessment of whether the specimen collected is suitable for laboratory testing, and, as secondary assessments, complete checklists of steps in the correct collection of the specimen (eg, did not smoke or drink while collecting saliva specimen, inverted closed saliva tube 20 times as directed, dropped blood on the DBS card rather than touching the card).

We will tabulate the provider impressions of specimen suitability (primary outcome); exploratory analyses for provider observation will include enumeration of how common certain errors in self-collection were. For assessment of the sufficiency of RNA-PCR specimens, we will tabulate the proportion of OPS and saliva samples that had C_t values for RNAse P<30. As a secondary analysis, we will examine whether the RNAse P C_t values for the patient-collected OPS were different from the C_t values from a historical set of provider-collected OPS tested with the same reagents and laboratory equipment in the study laboratory. These analyses will involve comparing the means of the two groups using a t test.