The real challenge of COVID-19 containment is to break the transmission chain by preventing individuals from getting infected and preventing those infected from transmitting to others [1]. The gold standard of COVID-19 containment has shifted to mass vaccination, with the belief that vaccines will break the chain of transmission. It is reported that current vaccines are close to 100% effective in reducing hospitalizations and deaths [2]. These significant efficacy results from conducted studies may however be limited given that these are just 2 of the endpoints of vaccine effectiveness. In reality, the effectiveness of available vaccines in preventing infections is yet to be fully understood in many countries. The duration of immunity of current vaccines is also a critical yet unanswered question [3]. It becomes even more challenging to measure the above endpoints amid emerging new variants [4-8] since this requires constant knowledge renewal on viral genomic epidemiology [9]. New variants of SARS-CoV-2, including B.1.1.7, B.1.351, and P.1, have been reported in the United Kingdom, South Africa, and Brazil, respectively. The emergence of these new variants—when the global economy is yet to recover from the devastating effects of the previous variants still in circulation—is particularly challenging. More so, the fact that these new variants of concern [10,11] are not only more transmissible but appear to be more asymptomatic and resistant to polymerase chain reaction detection, flags the overriding importance of waiting time in the detection of those infected. Transmissibility of the virus appears to vary by context as factors reported to determine transmission include the capacity for the virus to replicate, the exhibition of symptoms (cough), and the individual associated environmental factors (including behavior) [12].

It is reported that a major factor that can contribute to up to 50% of infected persons being overlooked during symptom screening is the inability to detect the virus during the asymptomatic stage by programs designed to diagnose cases when the disease is quite advanced [13]. Apart from a group of people who deliberately might not declare they have symptoms, the first 2 days of the 12-day infectious period of the virus, during which infected persons do not show any symptoms, remain an important proportion in the transmission mechanism of a deadly disease. The inability to effectively manage the transmission mechanism in low- and middle-income countries (LMICs) due to limited resources prompts the need for low-cost rapid tests in these countries. While some remain skeptical about the diagnostic safety and effectiveness of rapid tests [14,15], others believe that the currently available rapid tests are equally accurate enough in detecting some of the new variants [16-18]. The current evidence of what these different tests are and their diagnostic safety [19] within the context of LMIC is limited. In addition, standard guidelines for their use [20,21] seldom exist in these settings, making their effective deployment more challenging.

Key questions to ask when planning to identify infections promptly include, among others, the following: (1) When to test? (2) Where to test? (3) When to get the results? (4) How safe? and (5) How regular? As the global economy learns to cope with the more silent new variant infections, it is apparent that low-cost regular rapid mass testing might assume an important role [22,23], but their diagnostic safety [24] may be an issue particularly in LMICs. According to a 2020 report by the World Bank, low-income countries are those with a gross national income per capita of ≤US $1045, and middle-income countries are those with a gross national income per capita of US $1046 to US $12,695 [25]. About 500 million tests have been estimated to be needed in LMICs in 2021 [26], and the adoption of low-cost rapid tests might lead to up to 50% cost savings [27,28]; however, their fitness for purpose in these countries is yet to be fully understood. This study seeks to systematically review the evidence-based rapid mass testing options for SARS-CoV-2 that can be delivered at point of care in LMICs.

The specific objectives of this study are as follows: (1) to investigate the impact of the different rapid mass testing options for SARS-CoV-2 that have been delivered at point of care in LMICs and (2) to evaluate the diagnostic accuracy of rapid mass testing for SARS-CoV-2 in LMICs.

Given that our review was geared toward going beyond a quantitative evidence synthesis, the research questions were framed following the Perspective, Setting, Phenomenon, Environment, Comparison, Timing, and Findings (PerSPEcTiF) framework to incorporate the (1) setting, (2) time, and (3) place [29] for objective 1, and using the population, intervention, comparison, outcome (PICO) approach for objective 2 [30].

This is a systematic review of both qualitative and quantitative evidence synthesis. A qualitative approach will take the form of a metasynthesis while the quantitative approach will take the form of a meta-analysis.

Searched records will be screened using the inclusion and exclusion criteria. Given the important role played by both contextual variables and time in the detection of SARS-CoV-2 to break the transmission chain, we believed that rapid mass testing interventions were complex ones [31], not only following a systems approach but underpinned by both contextual variability and time scale [29]. The eligibility criteria for the first objective follows the PerSPEcTiF framework, while that for the second objective follows the PICO framework, as seen in Table 1.

Data items to be extracted will include study title, year, author, country, study objective, conflicts of interest, funding, study design, population (age, sex, and ethnicity), setting, sample size, eligibility criteria, type of test, the time interval between the index and reference tests, the sample collected, statistical analysis, test status, time to results (turnaround time), test accuracy, and study limitations.

The outcome of interest for the first objective is the test impact (patients, health personnel, and the general public). The outcome of the intervention for the second objective based on PICO is the status of SARS-CoV-2/COVID-19, which involves how well the test can identify those with the target condition and reject those without the conditions. Studies comparing the index and reference tests regarding the above will be prioritized.

The assessment of the risk of bias of included studies will be conducted using the most relevant tools and based on study design [49]. Randomized controlled trials will be assessed using the revised version of the Cochrane Risk of Bias (RoB 2) assessment tool [50,51]. The items for risk-of-bias assessment using the RoB 2 tool will be organized into 5 domains including (1) bias due to randomization, (2) bias due to deviation from intended intervention, (3) bias due to missing data, (4) bias arising from outcome measurements, and (5) bias in the selection of reported results. Studies will be categorized into “low,” “some concerns,” and “high” for randomized controlled trials, as shown in Table S1 of Multimedia Appendix 2. Nonrandomized intervention studies will be assessed using the 8-item tool developed by the Evidence Project, suitable for a variety of study designs [52]. This tool assesses the risk of bias under 3 domains, including (1) study design, (2) representation of participants, and (3) comparison group of equivalence. Following the view not to attribute a summary risk-of-bias score for studies evaluated using the Evidence Project tool [52], the risk of bias of nonrandomized studies will be reported based on Table S2 of Multimedia Appendix 2.

Publication bias will be reduced through the inclusion of gray literature [53,54]. Additionally, in the event of a meta-analysis, publication bias will be assessed by inspecting the funnel plot. Any indication of missing data through funnel plot asymmetry will be adjusted using the “trim and fill” approach [55]. A positive difference between the corrected and uncorrected diagnostic accuracy values will be indicative of an overestimation of diagnostic accuracy due to missing data.

The search strategy will follow the PRISMA search guidelines [66]. The quality of database search will be improved by searching beyond Medical Subject Headings (MeSH) terms by including “AND” and “OR” strings to the MeSH terms. Retained studies will be assessed for risk of bias using the most appropriate guidelines and checklists. The overall body of qualitative evidence in the first objective will be assessed using a critical reflection as highlighted [56], and the Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) Confidence in the Evidence from Reviews of Qualitative research (GRADE-CERQual) [67]. If applicable, the GRADE will be used to assess the overall body of evidence regarding objective 2 [68,69]. This will consider the risk-of-bias assessments, precision in reported results, consistency, directness, and publication bias.

This study does not warrant any ethical approval. This protocol has been registered with the International Prospective Register of Systematic Reviews (PROSPERO; CRD42022283776) [45], in line with the standards for undertaking systematic reviews. Review results will be disseminated through conferences and their peer-reviewed publication in a relevant journal.

This protocol is based on the diagnostic usefulness and safety of rapid diagnostic tests in LMICs. A point-of-care test is one that can be delivered at the point of sample collection or near the patient, with results made available almost immediately. By “near the patient,” we refer to settings as specified by the Centers for Disease Control and Prevention [36] and the Medicines and Healthcare products Regulatory Agency [73]. A range of rapid tests esteemed to be useful in surveillance has been reported in the literature [18,74,75], broadly categorized into diagnostic, screening, and public health surveillance tests [76]. Moreover, 36 lateral flow devices believed to have the minimum performance requirements have been published by the UK government [37]. Their usefulness is however yet to be fully understood within the context of LMICs.

A search for reviews evaluating rapid point-of-care test against the gold standard on asymptomatic carriers in Cochrane library found 3 reviews [38,77,78]. Only 1 of the reviews [38] evaluated the diagnostic accuracy of rapid diagnostic tests. The review concluded that evidence for the testing of asymptomatic carriers was limited and highlighted the overriding need for the evaluation of rapid tests in the settings for which they are intended. This review will be the first, to the best of our knowledge, to address this knowledge gap in LMICs from a public health perspective.

Although rapid mass testing programs came under serious criticism early in the pandemic [79-82], it is still obvious that the symptom-based testing alone is not comprehensive enough as it failed to bring the virus under control prior to the approval of vaccines; this is partly because even the gold standard reverse-transcription polymerase chain reaction test is also prone to false negatives [83]. One possible way of ensuring that people are detected as soon as they become infected is by making a rapid mass testing regime available at point of care and service points [84].

This protocol presents many strengths starting with its registration with PROSPERO. It is the first protocol that seeks to investigate both qualitative and quantitative evidence regarding the diagnostic safety of rapid mass tests in LMICs. The protocol’s specific objectives, search strategy, and reporting are all in line with the guidelines for the undertaking of systematic reviews.

This review highlights the role of context in infection prevention and control using rapid mass testing. It flags the overriding need to involve users and providers in the evaluation of such tests in the settings for which they are intended. This will be the first review, to the best of our knowledge, to generate both qualitative and quantitative evidence regarding rapid mass testing specific to LMICs.