Rapid diagnostic tests for Plasmodium vivax malaria in endemic countries.

BACKGROUND: Plasmodium vivax (P vivax) is a focus of malaria elimination. It is important because P vivax and Plasmodium falciparum infection are co-endemic in some areas. There are asymptomatic carriers of P vivax, and the treatment for P vivax and Plasmodium ovale malaria differs from that used in other types of malaria. Rapid diagnostic tests (RDTs) will help distinguish P vivax from other malaria species to help treatment and elimination. There are RDTs available that detect P vivax parasitaemia through the detection of P vivax-specific lactate dehydrogenase (LDH) antigens.
OBJECTIVES: To assess the diagnostic accuracy of RDTs for detecting P vivax malaria infection in people living in malaria-endemic areas who present to ambulatory healthcare facilities with symptoms suggestive of malaria; and to identify which types and brands of commercial tests best detect P vivax malaria. SEARCH
METHODS: We undertook a comprehensive search of the following databases up to 30 July 2019: Cochrane Infectious Diseases Group Specialized Register; Central Register of Controlled Trials (CENTRAL), published in the Cochrane Library; MEDLINE (PubMed); Embase (OVID); Science Citation Index Expanded (SCI-EXPANDED) and Conference Proceedings Citation Index-Science (CPCI-S), both in the Web of Science. SELECTION CRITERIA: Studies comparing RDTs with a reference standard (microscopy or polymerase chain reaction (PCR)) in blood samples from patients attending ambulatory health facilities with symptoms suggestive of malaria in P vivax-endemic areas. DATA COLLECTION AND ANALYSIS: For each included study, two review authors independently extracted data using a pre-piloted data extraction form. The methodological quality of the studies were assessed using a tailored Quality Assessment of Diagnostic Accuracy Studies-2 (QUADAS-2) tool. We grouped studies according to commercial brand of the RDT and performed meta-analysis when appropriate. The results given by the index tests were based on the antibody affinity (referred to as the strength of the bond between an antibody and an antigen) and avidity (referred to as the strength of the overall bond between a multivalent antibody and multiple antigens). All analyses were stratified by the type of reference standard. The bivariate model was used to estimate the pooled sensitivity and specificity with 95% confidence intervals (CIs), this model was simplified when studies were few. We assessed the certainty of the evidence using the GRADE approach. MAIN
RESULTS: We included 10 studies that assessed the accuracy of six different RDT brands (CareStart Malaria Pf/Pv Combo test, Falcivax Device Rapid test, Immuno-Rapid Malaria Pf/Pv test, SD Bioline Malaria Ag Pf/Pv test, OnSite Pf/Pv test and Test Malaria Pf/Pv rapid test) for detecting P vivax malaria. One study directly compared the accuracy of two RDT brands. Of the 10 studies, six used microscopy, one used PCR, two used both microscopy and PCR separately and one used microscopy corrected by PCR as the reference standard. Four of the studies were conducted in Ethiopia, two in India, and one each in Bangladesh, Brazil, Colombia and Sudan. The studies often did not report how patients were selected. In the patient selection domain, we judged the risk of bias as unclear for nine studies. We judged all studies to be of unclear applicability concern. In the index test domain, we judged most studies to be at low risk of bias, but we judged nine studies to be of unclear applicability concern. There was poor reporting on lot testing, how the RDTs were stored, and background parasitaemia density (a key variable determining diagnostic accuracy of RDTs). Only half of the included studies were judged to be at low risk of bias in the reference standard domain, Studies often did not report whether the results of the reference standard could classify the target condition or whether investigators knew the results of the RDT when interpreting the results of the reference standard. All 10 studies were judged to be at low risk of bias in the flow and timing domain. Only two brands were evaluated by more than one study. Four studies evaluated the CareStart Malaria Pf/Pv Combo test against microscopy and two studies evaluated the Falcivax Device Rapid test against microscopy. The pooled sensitivity and specificity were 99% (95% CI 94% to 100%; 251 patients, moderate-certainty evidence) and 99% (95% CI 99% to 100%; 2147 patients, moderate-certainty evidence) for CareStart Malaria Pf/Pv Combo test. For a prevalence of 20%, about 206 people will have a positive CareStart Malaria Pf/Pv Combo test result and the remaining 794 people will have a negative result. Of the 206 people with positive results, eight will be incorrect (false positives), and of the 794 people with a negative result, two would be incorrect (false negative). For the Falcivax Device Rapid test, the pooled sensitivity was 77% (95% CI: 53% to 91%, 89 patients, low-certainty evidence) and the pooled specificity was 99% (95% CI: 98% to 100%, 621 patients, moderate-certainty evidence), respectively. For a prevalence of 20%, about 162 people will have a positive Falcivax Device Rapid test result and the remaining 838 people will have a negative result. Of the 162 people with positive results, eight will be incorrect (false positives), and of the 838 people with a negative result, 46 would be incorrect (false negative). AUTHORS'
CONCLUSIONS: The CareStart Malaria Pf/Pv Combo test was found to be highly sensitive and specific in comparison to microscopy for detecting P vivax in ambulatory healthcare in endemic settings, with moderate-certainty evidence. The number of studies included in this review was limited to 10 studies and we were able to estimate the accuracy of 2 out of 6 RDT brands included, the CareStart Malaria Pf/Pv Combo test and the Falcivax Device Rapid test. Thus, the differences in sensitivity and specificity between all the RDT brands could not be assessed. More high-quality studies in endemic field settings are needed to assess and compare the accuracy of RDTs designed to detect P vivax.

Background

Target condition being diagnosed

Malaria is a life‐threatening disease caused by Plasmodium species (Plasmodium spp.), transmitted by the bite of a female Anopheles mosquito. Currently, there are five established Plasmodium spp. that cause malaria in humans. The two most common are Plasmodium falciparum (P falciparum) and Plasmodium vivax (P vivax). P vivax malaria is a relapsing form, which is rarely fatal but can cause serious anaemia in children (White 2018). There has been an increased focus on P vivax, as malaria‐endemic settings that also have P falciparum have made progress in P falciparum control. In the World Health Organization (WHO) regions of the Americas, South‐East Asia and Eastern Mediterranean, P vivax is the predominant Plasmodium spp., and causes 64%, greater than 30%, and greater than 40% of all malaria cases, respectively, in these regions (WHO 2017a). People with malaria caused by P vivax can have relapses due to the dormant liver stage hypnozoites. People can carry hypnozoites ranging from a few weeks to more than 12 months before reporting symptoms again (Campo 2015). Primaquine is recommended additionally to standard malaria treatment for P vivax and Plasmodium ovale (P ovale) to clear these liver stage parasites. Due to this, it is important to have diagnostic tests that are highly sensitive and that can specifically detect P vivax from other Plasmodium spp.

Index test(s)

Rapid diagnostic tests (RDTs) (WHO 2003), detect parasite‐specific antigens in a drop of fresh blood through lateral flow immunochromatography (WHO 2006). Generally, RDTs do not require a laboratory, any special equipment, or specialized training. They are easy to use and can give results as a simple positive or negative result within 15 to 20 minutes based on the antibody affinity (referred to as the strength of the bond between an antibody and an antigen) and avidity (referred to as the strength of the overall bond between a multivalent antibody and multiple antigens) (Talman 2007; WHO 2006). Therefore, RDTs are, in general, suitable for remote areas with limited facilities and lack of laboratory expertise. They typically have a shelf life of 24 months and need to be kept dry and away from temperature extremes (greater than 40°C). They may fail to detect malaria where there are low levels of Plasmodium parasites (and antigens) in the blood and false positives are possible due to cross reactions with other disease conditions, presence of certain immunological factors, and gametocytaemia (Gatton 2018; Kakkilaya 2003).

There is strong evidence that storage conditions of the RDT affect their performance (Moonasar 2007). The parasite density of the blood sample can also affect the performance of the RDT. The WHO malaria RDT product‐testing programme report investigated the effect of parasite density by testing individual products under laboratory conditions using standardized blood samples at low and high parasite densities (200 and 2000 parasites/µL), and reported the ‘panel detection score' (WHO 2012). An existing Cochrane Review on non‐falciparum RDTs found that parasite density and storage conditions are often poorly reported in field studies (Abba 2014). Moreover, due to the lag period between when the RDT was evaluated by the WHO malaria RDT product testing programme to when the RDT is actually used in the field, manufacturers may have modified the RDT during this period.

Different types of RDT use different types of antibody or combination of antibodies to detect Plasmodium antigens. Some antibodies aim to detect a particular species while others are pan‐malarial, aiming to detect all types of Plasmodium spp. Currently, all commercial RDTs specific for P vivax use P vivax‐specific lactate dehydrogenase (LDH) antigens (WHO 2017b).

Clinical pathway

People of any age with malaria typically present to medical care with non‐specific symptoms of fever, headache, chills, or rigors. The RDTs are most commonly used at the point of presentation with these symptoms, most often in settings where quality microscopy is not available. Parasitological diagnosis is recommended prior to commencing on any treatment (WHO 2015a).

Prior test(s)

It is unlikely that patients would have had previous testing for their current infection prior to presentation to healthcare centres with symptoms of malaria. One key benefit of RDTs is the ease of use at point of care. For the purpose of this review, we did not address the sensitivity or specificity of P vivax‐specific RDTs for confirming efficacy of treatment as this is not recommended practice.

Role of index test(s)

Malaria is a common cause of fever in endemic regions. Given the non‐specific symptoms patients with malaria often present with, a parasitological test is recommended to make a formal diagnosis (WHO 2015b). Often people of any age or gender presenting to a healthcare clinic with a history of fever in a malaria‐endemic region will undergo a malaria test as part of a routine initial work‐up. As such, the population receiving the index test would be identified solely on the basis of the clinical history and physical examination. RDTs have a role in malaria diagnosis where there is no access to good quality microscopy services and in outbreak investigation or surveys of parasite prevalence. The pre‐test probability of clinical malaria is an important determinant of the RDT performance. In the absence of strong clinical suspicion of malaria, it may not be reliable to use an RDT, because the test results from this device could potentially be misleading or inaccurate. Reliable diagnosis of P vivax malaria with RDTs would not only benefit the individual by allowing treatment of the blood stage and latent hypnozoite stage, but also would have benefits at a population level by potentially reducing low‐level ongoing transmission due to relapsing disease. Widespread use of accurate RDTs can facilitate greater diagnosis and treatment rates of P vivax malaria in areas where there is inadequate access to high‐quality microscopy.

True positive results would allow effective treatment of active disease and facilitate prevention of relapse using drugs that target the liver stage hypnozoites such as primaquine or tafenoquine, thus effectively treating individuals and reducing the risk of onward transmission. True negative results facilitate accurate diagnosis by narrowing differential diagnoses of people presenting to care with fever and non‐specific symptoms. False positives would potentially lead to over treatment of individuals with primaquine, tafenoquine and either chloroquine or artemisinin combination therapies and would mean that patients are not treated for the actual cause of their symptoms. False negatives would lead to potential relapsing disease and potentially ongoing transmission at the population level.

Alternative test(s)

Microscopic examination of Giemsa‐stained thick and thin blood films remains the conventional laboratory method. Microscopic examination has good sensitivity and specificity, and it allows species and stage differentiations and quantification of parasites, all of which are important in assessing disease severity, monitoring response to treatment, and prescribing appropriate therapy. Intensive examination is more likely to reveal parasitaemia so the test is carried out with a fixed number of fields examined. Infections may be missed if slides are not examined carefully (Wongsrichanalai 2007). Very low parasitaemia may be missed even by good quality microscopy; the limit of detection of thick smear microscopy has been estimated at approximately four to 20 asexual parasites per µL, although a threshold of 50 to 100 asexual parasites per µL is more realistic under field conditions (Wongsrichanalai 2007). False positive results are also possible; if blood slides are not prepared carefully, artefacts may be formed, which can be mistaken for Plasmodium parasites (Wongsrichanalai 2007).

The polymerase chain reaction (PCR), a molecular method based on DNA amplification, is the most analytically sensitive method of detecting parasites in the blood. Compared to microscopy, PCR is less prone to observer error and more sensitive at low levels of parasitaemia (Han 2017; Snounou 1993). For PCR, the limit of detection may be as low as 0.004 asexual parasites per µL (Hänscheid 2002). This increased ability to detect low level parasitaemia is important as submicroscopic parasitaemiae may have clinical and public health significance and the prevalence of asymptomatic submicroscopic infection is high in some areas (Chen 2016). PCR is currently not widely available due to logistical constraints and the need for specially‐trained technicians and a well‐equipped laboratory. It is usually used only for research purposes.

Rationale

P vivax is becoming increasingly important, especially in regions targeting malaria elimination. In areas of co‐endemicity, P vivax malaria is increasing disproportionally compared to P falciparum malaria. Moreover, treatment for P vivax and P ovale malaria differs from treatments for other types of malaria. Therefore, it is important that the RDT correctly distinguish P vivax from other species. Geographically, P vivax has a much wider infection range compared to other Plasmodium spp. This may increase over time due to climate change (Culleton 2012). Historically, autochthonous transmission of P vivax also occurred in temperate climates, such as that of England (Dobson 1994). Autochthonous transmission is referred to as the spread of a disease from one individual and received by another individual from the same place. An existing Cochrane Review assessing RDTs for diagnosing uncomplicated non‐falciparum malaria was conducted in 2014 (Abba 2014). A subset of this review included RDTs that diagnosed P vivax. This review only assesses the diagnostic accuracy of RDTs that specifically detect P vivax with Pvivax‐specific LDH) antigens.

Objectives

To assess the diagnostic accuracy of RDTs for detecting P vivax malaria parasitaemia in people living in malaria‐endemic areas who present to ambulatory healthcare facilities with symptoms suggestive of malaria, and to identify which types and brands of commercial tests best detect P vivax malaria.

Methods

Criteria for considering studies for this review

Types of studies

We included retrospective or prospective cohort or cross‐sectional studies that assessed the accuracy of an RDT, or compared the accuracy of two or more RDTs, in the same study population (i.e. comparative accuracy studies). We excluded case‐control studies because they are known to overestimate test accuracy (Whiting 2011). Eligible studies included a consecutive series of patients, or a randomly selected series of patients. If the study did not explicitly state that the sampling was consecutive or random, the study was considered unclear but was still included. We excluded studies if they did not present sufficient data to allow us to extract or deduce the number of true positives, false positives, false negatives, and true negatives (i.e. 2 x 2 table data). We also excluded studies published in predatory journals, which is referred to as journals that accept articles for publication for a fee without providing peer‐review or quality checks for plagiarism or ethical approval.

Participants

Studies recruiting people living in P vivax‐endemic areas attending ambulatory healthcare settings with symptoms of uncomplicated malaria were eligible.

We excluded studies if participants:

had travelled from non‐malarious region to malarious regions, e.g. travellers or displaced populations;

had been previously treated for their current malaria infection or the test was performed to assess whether treatment was successful, or both;

had symptoms of severe malaria as defined by the WHO clinical definition (WHO 2014);

did not have symptoms of malaria as defined by history of fever, headache, or chills/rigors; or

were recruited through active case finding (for example, door to door surveys).

Index tests

Studies evaluating any immunochromatography‐based RDT specifically designed to detect P vivax malaria. We only included RDTs that met the WHO malaria RDT performance criteria (WHO 2017b).

Target conditions

Studies aimed at detecting P vivax malaria.

Reference standards

Studies that diagnosed P vivax malaria using at least one of the following two reference standards:

Conventional microscopy of thick blood smears and thin blood smears. Presence of asexual parasites of any density is regarded as a positive smear. Once the diagnosis is established – usually by detecting parasites in the thick smear – the laboratory technician can examine the thin smear to determine the malaria species and the parasitaemia, or the percentage of the patient’s red blood cells that are infected with malaria parasites. The thin and thick smears are able to provide all three of these vital pieces of information. Ideally, blood smears would be examined independently and in duplicate with more than 100 high‐power fields;

PCR, including quantitative PCR (qPCR), nested PCR (nPCR), and real‐time PCR (rPCR). We also included studies that used loop‐mediated isothermal amplification (LAMP). Most PCR‐based assays for P vivax are only available as laboratory‐developed tests, which means they are rarely used clinically outside of research projects where P vivax malaria is endemic. They are especially useful for diagnosing asymptomatic people as the assays have high sensitivity. Molecular diagnostics theoretically have a lower limit of detection than both RDTs and microscopy depending on the training of microscopists and quality of samples analysed. Significant variation exists between molecular diagnostics developed including type of input material (DNA, RNA, or whole blood), target gene, (number of) species detected, primer/probe composition and concentration, amplification technique (PCR or isothermal), read‐out (gel‐electrophoresis, fluorescence detection, lateral flow), and whether it is qualitative or quantitative. However, no important differences have been found in the accuracy of these tests (Roth 2016).

For studies that used both reference standards, we extracted 2 x 2 data for each reference standard and stratified the analyses by reference standard.

Search methods for identification of studies

We attempted to identify all relevant studies regardless of language or publication status (published, unpublished, in press, and in progress).

Electronic searches

We searched the following databases up to 30 July 2019 using the search terms and strategy described in Appendix 1: Cochrane Infectious Diseases Group Specialized Register; Central Register of Controlled Trials (CENTRAL), published in the Cochrane Library (issue 7, 2019); MEDLINE (PubMed, from 1966); Embase (OVID, from 1947); Science Citation Index Expanded (SCI‐EXPANDED) and Conference Proceedings Citation Index‐ Science (CPCI‐S), both in the Web of Science, from 1900; LILACS (BIREME).

We also searched the WHO International Clinical Trials Registry Platform (WHO ICTRP; www.who.int/ictrp/en/) and ClinicalTrials.gov (clinicaltrials.gov/ct2/home) for trials in progress, using "vivax malaria", "Plasmodium vivax", and "rapid diagnostic test*" or RDT* as search terms.

Searching other resources

We checked the reference lists of studies identified by the above methods.

Data collection and analysis

Selection of studies

Three review authors (RA, LC and SJ) independently assessed the study eligibility by examining the title and abstract of each article identified by the literature search and excluded obviously irrelevant studies. If a review author considered the abstract to be potentially eligible, we obtained the full‐text article. Three review authors independently assessed each full‐text article against the predefined inclusion and exclusion criteria, as stated in the ‘Criteria for considering studies for this review' section, and resolved any disagreements by discussion. All articles that were excluded after full‐text assessment are listed with reasons for exclusion in the ʽCharacteristics of excluded studies' table. We illustrated the study selection process with a PRISMA flow diagram (Figure 1).

1

Study flow diagram.

Data extraction and management

Three review authors (RA, LC and SJ) independently extracted data using a pre designed data extraction form.

We extracted the following data.

Authors, publication year, and journal.

Study design.

Study start date.

Characteristics study participants (age, gender, co morbidities, and pregnancy).

Study inclusion/exclusion criteria.

Study setting.

Malaria species in study setting.

Malaria prevalence and endemicity in study setting.

Reference standard.

Index test (brand name, target antigen, and batch numbers).

Additional tests (and their results).

RDT and reference standard setting.

Lot testing of RDT used.

Transport and storage conditions of RDTs.

Training level of person performing index test.

Training level of person performing reference standard (and if available the WHO certified training level of the microscopist).

Number of high power fields observed in microscopy.

Parasite density of microscopy positive cases or PCR.

Observers or repeats used.

Number of indeterminate, missing or unavailable test results.

Number of true positives, false positives, false negatives, and true negatives.

Type of molecular amplification assay.

Volume of blood samples.

Limit of detection for PCR.

We resolved any discrepancies in data extraction by discussion. We contacted the authors of primary studies when we could not resolve any disagreements.

Assessment of methodological quality

We used the revised tool for the Quality Assessment of Diagnostic Accuracy Studies (QUADAS‐2) to assess the risk of bias and applicability of included studies (Whiting 2011). We tailored the tool to the context of the review as shown in Appendix 2. Three review authors (RA, LC and SJ) independently assessed methodological quality using the tailored QUADAS‐2 tool. We resolved any disagreements through consensus. We used both graphics and text to summarize the results.

Statistical analysis and data synthesis

We stratified all analyses by the type of reference standard used. We plotted estimates of sensitivity and specificity from the included studies in forest plots and in receiver operating characteristic (ROC) space using the software, Review Manager 5 (RevMan 5) (RevMan 2014). We planned to perform meta‐analysis using the bivariate model to estimate summary sensitivities and specificities (summary points) (Chu 2006; Macaskill 2010; Takwoingi 2015b). However, due to sparse data or few studies, we simplified the models to univariate random effects logistic regression models to pool sensitivity and specificity separately (Takwoingi 2015a). We performed meta‐analyses using the 'meqrlogit' command in Stata (STATA 2015). Due to the limited number of included studies we did not perform meta‐analyses to compare the accuracy of different RDT brands as planned. However, we summarized individual study estimates from head‐to‐head comparisons of brands in a table.

Investigations of heterogeneity

We intended to investigate any heterogeneity from the pooled analyses with pre‐specified factors, as stated in our secondary objective. Due to the limited number of studies, we were unable to investigate heterogeneity as planned.

Sensitivity analyses

We did not have sufficient data for sensitivity analyses.

Assessment of the certainty of the evidence

We assessed the certainty of the evidence for comparisons where there were sufficient studies enabling meta‐analyses (i.e. quality of evidence or confidence in effect estimates) using the GRADE approach and GRADEpro Guideline Development Tool software (GRADE 2013; GRADEpro GDT 2015). In the context of a systematic review, the ratings of the certainty of the evidence reflect the extent of our confidence that the estimates of test accuracy are correct. As recommended, we rated the certainty of the evidence as either high (not downgraded), moderate (downgraded by one level), low (downgraded by two levels), or very low (downgraded by more than two levels) for four domains: risk of bias, indirectness, inconsistency, and imprecision. For sensitivity and specificity, the certainty of the evidence initially started as high when there were high‐quality cross‐sectional or cohort studies that enrolled participants with diagnostic uncertainty. If we found a reason for downgrading the certainty of the evidence, we classified the reason as either serious (downgraded by one level) or very serious (downgraded by two levels).

Three review authors (RA, LC and SJ) discussed judgments and reached a consensus. We applied GRADE in the following way.

Risk of bias: we used QUADAS‐2 to assess risk of bias.

Indirectness: we considered indirectness from the perspective of test accuracy. We used QUADAS‐2 to assess applicability concerns and looked for important differences between the populations studied (for example, in the transmission intensity as defined by the WHO World Malaria Report or WHO malaria country profiles for the corresponding year), the setting, and the review question.

Inconsistency: GRADE recommends downgrading for unexplained inconsistency in sensitivity and specificity estimates.

Imprecision: we considered the width of the confidence intervals (CIs), and asked ourselves, “would we make a different decision if the lower or upper limit of the 95% confidence interval (CI) represented the truth?” In addition, we calculated absolute numbers of true positives, false negatives, false positives, and true negatives, as well as ranges for these values based on the CIs of the pooled estimates of sensitivity and specificity for various prevalences of P vivax malaria; we also made judgements on imprecision using these calculations. We also calculated positive and negative likelihood ratios with their 95% CIs.

Assessment of reporting bias

We did not assess publication bias due to the uncertainty about the determinants of publication bias for diagnostic accuracy studies, and the inadequacy of tests for detecting funnel plot asymmetry (Deeks 2005).

Results

Results of the search

We identified and screened 768 reports through the database searches conducted on 30 July 2019. We excluded 706 of these reports based on their title or abstract alone. We considered the remaining 62 articles for full‐text screening, along with the 37 studies included in the non‐falciparum malaria review by Abba 2014. Of the 109 articles, we excluded 99 for various reasons as reported in the Characteristics of excluded studies section, shown in Figure 1. We included 10 studies, of which five studies (Alam 2011; Chanie 2011; Mekonnen 2010;Singh 2010; Sharew 2009) were also included in the review by Abba 2014. The 10 studies assessed six different RDT brands (CareStart Malaria Pf/Pv Combo test, Falcivax Device Rapid test, Immuno‐Rapid Malaria Pf/Pv test, SD Bioline Malaria Ag Pf/Pv test, OnSite Pf/Pv test and Test Malaria Pf/Pv rapid test). One study directly compared the accuracy of two RDT brands (Falcivax Device Rapid test and OnSite Pf/Pv test) (Alam 2011). The six RDT brands detect P vivax as part of a mixed infection with P vivax‐specific LDH antigens. The tests have two test lines, an HRP‐2 line to detect P falciparum and an pLDH line to detect P vivax. For our analysis we only considered the presence of the pLDH line.

Of the 10 included studies, six used microscopy (Chanie 2011; Costa 2019; Hailu 2014; Mekonnen 2010; Sharew 2009; Singh 2010), one used PCR (Mussa 2019), two used both microscopy and PCR separately (Alam 2011; Saha 2017), and one used microscopy corrected by PCR (Mendoza 2013) as the reference standard. Four of the studies were conducted in Ethiopia (Chanie 2011; Hailu 2014; Mekonnen 2010; Sharew 2009), two in India (Saha 2017; Singh 2010), and one each in Bangladesh (Alam 2011), Brazil (Costa 2019), Colombia (Mendoza 2013), and Sudan (Mussa 2019).

There was a lack of detail on how the RDTs were stored and whether RDT lots were quality‐controlled prior to testing. Key study characteristics that may affect the performance of RDTs (e.g. training level of person performing the RDT, storage conditions, and parasite density of microscopy‐positive cases or PCR) are summarised in Table 2.

1

Summary of key study characteristics

Study	Country	Sample size	Sex	Age	RDT brand	Personnel performing RDT	Storage conditions of RDT	Reference standard	Personnel performing reference standard	Parasite density of positive cases
Alam 2011	Bangladesh	338	49.7% male50.3% female	Median (range): 14 years (18 months to 82 years)	OnSite Pf/Pv test (CTK Biotech Inc, USA)Falcivax Device Rapid test for malaria Pv/Pf (Zephyer Biomedicals, Goa)	An experienced medical technologist	Unclear, although study stated that the instructions of the manufacturers were followed.	PCR and microscopy (separately)	Slides assessed by two independent microscopists	Of 21 P vivax positive slides, parasite count ranged from 32 to 25,120 parasites/μL of blood, with a median of 5,040 (IQR 520 to 17,160) parasites/μL blood.
Chanie 2011	Ethiopia	1092	51.4% male 48.6% female	Mean (SD): 22 (12.8) years	CareStart Malaria Pf/Pv Combo test (Access Bio Inc, Somerset, NJ)	Experienced malaria technicians	Kept at the local temperature of the region without any controlling system of the storage temperature during data collection	Microscopy	Experienced technicians examined the slides	Not reported
Costa 2019	Brazil	181	64.1% male 35.9% female	Mean (SD): 41.7 (14.4) years	Immuno‐Rapid Malaria Pf/Pv test (Wama Diagnostica, Sao Paulo, Brazil)	Hospital laboratory staff	According to manufacturer's instructions (2ºC to 30ºC until the expiration date)	Microscopy	Experienced microscopists then examined the slides	Mean parasitaemia detected by TBS for P vivax malaria was 1,206.5 parasites/mm3 blood
Hailu 2014	Ethiopia	398	44.2% male 55.8% female	Range: 1 to 70 years	CareStart Malaria Pf/Pv Combo test (Access Bio Inc, Somerset, NJ)	Not reported	Stored at room temperature according to manufacturer's instructions	Microscopy	Two experienced malaria technologists performed the microscopy	Not reported
Mekonnen 2010	Ethiopia	240	57.5% male42.5% female	Mean (range): 25 years (1 to 60 years)	CareStart Malaria Pf/Pv Combo test (Access Bio Inc, Somerset, NJ)	Experienced malaria technicians	According to manufacturer's instructions	Microscopy	Three experienced technicians examined the slides	Not reported
Mendoza 2013	Colombia	383	52.5% male 47.5% female	Range: 6 to 92 years	SD Bioline Malaria Ag Pf/Pv test (Standard Diagnostics Inc)	Conducted by a trained person	According to manufacturer’s recommendations (1ºC to 40ºC)	Microscopy corrected with PCR	Blood films were examined by two experienced readers	Parasitemia for P vivax ranged from 40 to 40,000 parasites/µL
Mussa 2019	Sudan	59	45.8% male 54.2% female	Not reported	Test Malaria Pf/Pv rapid test (Alltest Biotech, China)	Not reported	Unclear, although study stated that instructions of the manufacturer were followed	PCR	Not reported	Not reported
Saha 2017	India	200	56.0% male 44.0% female	Mean: 34.6 years11 to 20 years: 20.5%21 to 60 years: 68.5%<10 years: 2.5%> 61 years: 8.5%	SD Bioline Malaria Ag Pf/Pv test (Standard Diagnostics Inc)	Microscopy, RDT and PCR done by different technicians	Unclear, although study stated that instructions of the manufacturer were followed	PCR and Microscopy (separately)	Blood films were examined by two microscopists having >15 years of experience	Not reported
Sharew 2009	Ethiopia	668	54.0% males46.0% females	Range: 6 months to 75 years	CareStart Malaria Pf/Pv Combo test (Access Bio Inc, Somerset, NJ)	Experienced malaria technicians	Stored according to manufacturer's instructions	Microscopy	Thick and thin smears determined by two experienced malaria technicians	Not reported
Singh 2010	India	372	Not reported	Mean (SD): 15 (14.1) years	Falcivax Device Rapid test for malaria Pv/Pf (Zephyer Biomedicals, Goa)	Two research assistants	Detailed storage information provided	Microscopy	Blood films examined by an experienced microscopist	Not reported

PCR = polymerase chain reaction; RDT = rapid diagnostic test; SD = standard deviation; TBS =thick blood smear

Methodological quality of included studies

The results of the risk of bias and applicability assessment are summarised in Figure 2. One study was judged to be at low risk of bias in all four domains of the QUADAS‐2 tool (Saha 2017). This study assessed the SD Bioline Malaria Ag Pf/Pv test.

2

Risk of bias and applicability concerns summary: review authors' judgements about each domain for each included study

Patient selection

Six (60%) studies were at unclear risk of bias in the patient selection domain because the method of participant recruitment (random or consecutive) was unclear (five studies), and/or the exclusion criteria were unclear (five studies). All studies were of low concern regarding applicability as they were all conducted in settings endemic with P vivax. However, Saha 2017 and Mussa 2019 did not report the prevalence of P vivax malaria. The remaining eight studies reported P vivax malaria or malaria in general as prevalent or endemic, but it was unclear to what degree.

Index test

We judged eight (80%) studies to be at low risk of bias in this domain because the results of the RDTs were interpreted without knowledge of the results of the reference standard. We judged the risk of bias for the remaining two studies to be unclear (Alam 2011; Mussa 2019). We judged the applicability of eight studies to be unclear, as poor reporting of the storage conditions or lot testing hampered the assessment. Singh 2010 provided thorough detail of how their RDT was stored, but it was unclear whether these conditions followed the instructions of the manufacturer. Applicability in this study was thus unclear. This study tested the temperature stability of the tests (see Table 2). Chanie 2011 evaluated the CareStart Malaria Pf/Pv Combo test. This was the only study considered to be of low applicability concern, because lot testing was reported.

Reference standard

We judged five studies to be at low risk of bias in the reference standard domain (Mendoza 2013; Saha 2017; Hailu 2014; Mekonnen 2010; Sharew 2009), while we judged one to be at high risk of bias (Singh 2010). We judged the remaining four studies to be at unclear risk of bias in this domain. It was unclear for two studies whether the results of the reference standard were interpreted without knowledge of the RDT results (Alam 2011, Mussa 2019), and it was unclear for two studies whether the results of the reference standard could classify the target condition. Costa 2019 and Chanie 2011 did not provide enough information on the reference standard to deduce if at least two microscopists independently examined the same slides from microscopy. We deemed Singh 2010 to be at high risk of bias because the second microscopist did not verify all of the reference standard results.

Flow and timing

We judged all 10 studies to be at low risk of bias in the flow and timing domain. All studies avoided partial verification, differential verification and incorporation bias, and reasons for any withdrawals were recorded. Nine studies appeared to have no uninterpretable results because the number of participants enrolled matched the number in the analysis. The remaining study reported two invalid RDT results, which were retested with the same test kits by taking fresh blood from the patients (Hailu 2014). However, it was unclear whether the same blood sample was used for the reference standard.

Test comparison

Although the QUADAS‐2 tool does not specifically address risk of bias in a test comparison, we additionally considered the potential for such bias in a study that directly compared two RDT brands (OnSite Pf/Pv test and Falcivax Device Rapid test) (Alam 2011). It was unclear whether the results of one RDT brand were interpreted without knowledge of the results of the other brand. The study used both microscopy and PCR as two separate reference standards, but it was unclear whether the conduct and interpretation of the results from these two reference standards were done independently of each other.

Findings

Verified by PCR

Three studies (Alam 2011; Mussa 2019; Saha 2017) evaluated the accuracy of four different brands of RDTs against PCR (Figure 3; Table 3). One of the studies had no cases of P vivax malaria, so sensitivity was not estimable (Mussa 2019). The sensitivities of the RDTs ranged between 77% and 86% and the specificities ranged between 93% and 100%.

3

Forest plot of brands of rapid diagnostic tests verified against PCR or microscopy corrected with PCR

2

Comparison of microscopy and PCR reference standards for P vivax

RDT brand

Microscopy

PCR

Microscopy corrected with PCR

Number of studies

Number of participants (P vivax malaria cases)

Sensitivity (95% CI) (%)

Specificity (95% CI) (%)

Number of studies

Number of participants (P vivax malaria cases)

Sensitivity (95% CI) (%)

Specificity (95% CI) (%)

Number of studies

Number of participants (P vivax malaria cases)

Sensitivity (95% CI) (%)

Specificity (95% CI) (%)

CareStart Malaria Pf/Pv Combo test (Access Bio Inc, Somerset, NJ)

4

2398 (251)

99% (94% to 100%)

99% (99% to 100%)

0

‐

‐

‐

0

‐

‐

‐

Falcivax Device Rapid test for malaria Pv/Pf (Zephyer Biomedicals, Goa)

2

710 (89)

77% (53% to 91%)

99% (98% to 100%)

1

338 (26)

77% (56% to 91%)

100% (99% to 100%)

0

‐

‐

‐

Immuno‐Rapid Malaria Pf/Pv test (Wama Diagnostica, Sao Paulo, Brazil)

1

181 (95)

99% (94% to 100%)

100% (96% to 100%)

0

‐

‐

‐

0

‐

‐

‐

SD Bioline Malaria Ag Pf/Pv test (Standard Diagnostics Inc)

1

200 (4)

75% (19% to 99%)

98% (95% to 99%)

1

200 (7)

86% (42% to 100%)

99% (97% to 100%)

1

383 (79)

92% (84% to 97%)

100% (99% to 100%)

OnSite Pf/Pv test (CTK Biotech Inc, USA)

1

338 (21)

90% (70% to 99%)

99% (97% to 100%)

1

338 (26)

77% (56% to 91%)

99% (97% to 100%)

0

‐

‐

‐

Test Malaria Pf/Pv rapid test (Alltest Biotech, China)

0

‐

‐

‐

1

59 (0)

Not estimable

93% (84% to 98%)

0

‐

‐

‐

PCR = polymerase chain reaction; RDT = rapid diagnostic test.

Verified by microscopy

Eight studies conducted in four different countries evaluated the accuracy of RDTs against microscopy (Figure 4). Five different RDT brands were assessed: CareStart Malaria Pf/Pv Combo test (four studies), Falcivax Device Rapid test (two studies), Immuno‐Rapid Malaria Pf/Pv test (one study), OnSite Pf/Pv test (one study), and SD Bioline Malaria Ag Pf/Pv test (one study).

4

Forest plot of brands of rapid diagnostic tests verified against microscopy, within each brand sorted by sensitivity and specificity

In the four CareStart Malaria Pf/Pv Combo test studies (251 P vivax malaria cases, 2398 patients), the sensitivity ranged from 95% to 100% and specificity ranged from 98% to 100%. The pooled sensitivity (95% CI) was 99% (94% to 100%) and the pooled specificity (95% CI) was 99% (99% to 100%) (Figure 5). The positive likelihood ratio (95% CI) was 141.09 (68.18 to 292.00) and the negative likelihood ratio (95% CI) was 0.01 (0.00 to 0.06).

5

Summary ROC plot for CareStart Malaria Pf/Pv Combo test verified against microscopy. The size of each study point was scaled by the sample size of the diseased and non‐diseased groups used to estimate the study's sensitivity and specificity respectively, and reflects the precision of sensitivity and specificity in the study relative to other study points.The solid circle (summary point) represents the summary estimate of sensitivity and specificity. The summary point is not surrounded by a 95% confidence region because the bivariate model was simplified to univariate models.

The sensitivities of the Falcivax Device Rapid test from the two studies (89 P vivax malaria cases, 710 patients) were 66% (95% CI 54% to 77%) and 90% (95% CI 70% to 99%), and specificities were 99% (95% CI 97% to 100%) and 100% (95% CI 98% to 100%). The pooled sensitivity (95% CI) was 77% (53% to 91%) and the pooled specificity (95% CI) was 99% (98% to 100%). The positive likelihood ratio (95% CI) was 120.31 (43.10 to 335.87) and the negative likelihood ratio (95% CI) was 0.23 (0.10 to 0.53).

The sensitivities of the three remaining RDT brands ranged between 75% and 99% and the specificities ranged between 98% and 100% (Table 3; Figure 4).

Verified by microscopy corrected with PCR

Mendoza 2013 evaluated the accuracy of SD Bioline Malaria Ag Pf/Pv test against microscopy corrected with PCR (Figure 3). When there were discordant results between microscopy and PCR, the result of the PCR was taken, except in those in which the thick drop showed parasitic forms and the PCR was negative. The study reported a sensitivity (95% CI) of 92% (84% to 97%) and a specificity (95% CI) of 100% (99% to 100%).

Comparison between RDT brands

Alam 2011 directly compared the accuracy of Falcivax Device Rapid test and OnSite Pf/Pv test with PCR and microscopy as the reference standards. There was no evidence to suggest a difference in the sensitivity and specificity of the two brands (Table 4). Using microscopy as the reference standard, the absolute difference in sensitivity (95% CI) was 0 percentage points (‐17.8 to 17.8 percentage points) and the absolute difference in specificity (95% CI) was 0.9 percentage points (‐0.4 to 2.3 percentage points). Using PCR as the reference standard, the differences in sensitivity and specificity were similar.

3

Direct comparisons between OnSite Pf/Pv test and Falcivax Device Rapid test

Study	Reference standard	Sensitivity (true positives/malaria cases) (%)		Difference (95% CI) (percentage points)	P value	Specificity (true negatives/non‐cases) (%)		Difference (95% CI) (percentage points)	P value
		OnSite Pf/Pv test (CTK Biotech Inc, USA)	Falcivax Device Rapid test for malaria Pv/Pf (Zephyer Biomedicals, Goa)			OnSite Pf/Pv test (CTK Biotech Inc, USA)	Falcivax Device Rapid test for malaria Pv/Pf (Zephyer Biomedicals, Goa)
Alam 2011	Microscopy	90 (19/21)	90 (19/21)	0 (‐17.8 to 17.8)	P = 1.00	99 (313/317)	100 (316/317)	0.9 (‐0.4 to 2.3)	P = 0.18
Alam 2011	PCR	77 (20/26)	77 (20/26)	0 (‐22.9 to 22.9)	P = 1.00	99 (309/312)	100 (312/312)	1.0 (‐0.1 to 2.0)	P = 0.08

PCR = polymerase chain reaction.

Discussion

Summary of main results

This systematic review included 10 studies conducted in six different countries (Bangladesh, Brazil, Colombia, Ethiopia, India, and Sudan). The studies assessed six different RDT brands: CareStart Malaria Pf/Pv Combo test (four studies), Falcivax Device Rapid test for malaria Pv/Pf (three studies), Immuno‐Rapid Malaria Pf/Pv test (one study), SD Bioline Malaria Ag Pf/Pv test (three study), OnSite Pf/Pv test (two studies), and Test Malaria Pf/Pv rapid test (one study). However, only one study directly compared the accuracy of two brands.

The main findings of the review are summarised in Table 1, together with illustrations of what the findings mean. We assume median prevalences of ranges that would be classified as moderate, low, and very low transmission areas for P vivax (20%, 5%, and 0.5% respectively) in a hypothetical cohort of 1000 people suspected of having Pvivax malaria (WHO 2017c). The CareStart Malaria Pf/Pv Combo test had a pooled sensitivity (95% CI) and specificity (95% CI) of 99% (94% to 100%) and 99% (99% to 100%) when microscopy was the reference standard. For a prevalence of 20%, about 206 people will have a positive CareStart Malaria Pf/Pv Combo test result and the remaining 794 people will have a negative result. Of the 206 people with positive results, eight will be incorrect (false positives), and of the 794 people with a negative result, two would be incorrect (false negative). The potential consequence of false positive results is unnecessary initiation of treatment and over‐treatment of individuals with primaquine and either chloroquine or artemisinin combination therapies, and that patients are not treated for the actual cause of their symptoms. The consequences of false negative results are potential relapsing disease and continued risk of transmission of P vivax malaria at population level.

Summary of findings 1

Summary of findings table for RDTs for diagnosing P vivax malaria

Population: people presenting with symptoms of uncomplicated malaria
Prior testing: none
Setting: ambulatory healthcare settings in P vivax endemic areas
Index tests: immunochromatography‐based rapid diagnostic tests (RDTs) for P vivax malaria that meet the WHO malaria RDT performance criteria (WHO 2017b)
Reference standards: conventional microscopy, polymerase chain reaction (PCR)
Target condition:P vivax malaria
Importance: accurate and fast diagnosis of P vivax from other malaria species allows appropriate treatment to be provided quickly
Study design: retrospective or prospective cohort or cross‐sectional
Findings: 10 studies of six different RDT brands (CareStart Malaria Pf/Pv Combo test, Falcivax Device Rapid test, Immuno‐Rapid Malaria Pf/Pv test, SD Bioline Malaria Ag Pf/Pv test, OnSite Pf/Pv test and Test Malaria Pf/Pv rapid test) for P vivax malaria were included. Only two brands (CareStart Malaria Pf/Pv Combo test and Falcivax Device Rapid test) were evaluated against the same reference standard by more than one study.
Limitations: a small number of studies were included in the analyses and meta‐analyses were only possible for two RDT brands. Studies often did not report how patients were selected, the blinding of the RDT results to the reference standard and the storage conditions and lot testing of RDTs.
Outcome	№ of studies	№ of patients	Numbers in a cohort of 1000 patients tested (95% CI)a			Certainty of the evidence (GRADE)b
			Prevalence of 0.5%	Prevalence of 5%	Prevalence of 20%
Test (reference standard): CareStart Malaria Pf/Pv Combo test (microscopy), pooled sensitivity (95% CI) = 99% (94% to 100%) and pooled specificity (95% CI) = 99% (99% to 100%), positive likelihood ratio (95% CI) = 141.09 (68.18 to 292.00) and negative likelihood ratio (95% CI) = 0.01 (0.00 to 0.06)
True positives(patients with P vivax malaria)	4	251	5 (5 to 10)	50 (47 to 50)	198 (188 to 200)	⊕⊕⊕⊝MODERATE1
False negatives(patients incorrectly classified as not having P vivax malaria)			0 (0 to 0)	0 (0 to 3)	2 (0 to 12)
True negatives(patients without P vivax malaria)		2147	985 (980 to 995)	941 (941 to 950)	792 (792 to 800)	⊕⊕⊕⊝MODERATE1
False positives (patients incorrectly classified as having P vivax malaria)			10 (0 to 10)	9 (0 to 9)	8 (0 to 8)
Test (reference standard): Falcivax Device Rapid test (microscopy), pooled sensitivity (95% CI) = 77% (53% to 91%) and pooled specificity (95% CI) = 99% (98% to 100%), positive likelihood ratio (95% CI) = 120.31 (43.10 to 335.87) and negative likelihood ratio (95% CI) = 0.23 (0.10 to 0.53)
True positives(patients with P vivax malaria)	2	89	4 (3 to 5)	39 (27 to 46)	154 (106 to 182)	⊕⊕⊝⊝LOW1,2
False negatives(patients incorrectly classified as not having P vivax malaria)			1 (0 to 2)	11 (4 to 23)	46 (18 to 94)
True negatives(patients without P vivax malaria)		621	985 (975 to 995)	941 (931 to 950)	792 (784 to 800)	⊕⊕⊕⊝MODERATE1
False positives (patients incorrectly classified as having P vivax malaria)			10 (0 to 20)	9 (0 to 19)	8 (0 to 16)

aMedian values were chosen from ranges of prevalence considered to be moderate, low, and very low transmission settings for P vivax (WHO 2017c).
bMethods are lacking to assess the determinants and extent of publication bias for diagnostic studies. However, in this table, we considered publication bias ‘undetected'.
1Downgraded for risk of bias by one.
2Downgraded for imprecision by two due to wide confidence intervals.

GRADE certainty of the evidence.

High: we are very confident that the true effect lies close to that of the estimate of the effect.
Moderate: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different.
Low: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect.
Very low: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect.

The Falcivax Device Rapid test had a pooled sensitivity and specificity of 77% (53% to 91%) and 99% (98% to 100%) when microscopy was the reference standard. For a prevalence of 20%, about 162 people will have a positive Falcivax Device Rapid test result and the remaining 838 people will have a negative result. Of the 162 people with positive results, eight will be incorrect (false positives), and of the 838 people with a negative result, 46 would be incorrect (false negative). A study that verified the results of the Falcivax Device Rapid test against PCR (Alam 2011), had a similar sensitivity and specificity of 77% (56% to 91%) and 100% (99% to 100%).

Strengths and weaknesses of the review

It is possible that some studies eligible for the inclusion in the review were missed by our search strategy. DTA studies are known to be poorly indexed, thus liable to be missed despite a broad literature search (Whiting 2009). However, our search was systematic, included studies published in all languages, and identified eligible studies from a previous review (Abba 2014). We also corresponded with study authors, when necessary, to obtain additional and unpublished data.

The main limitation of the review was the small number of studies included in the analyses. The meta‐analysis of the Falcivax Device Rapid test verified by microscopy included only two studies. Thus, the pooled estimate of sensitivity, and in general from analyses containing a small number of studies, should be interpreted with caution. Comparative accuracy studies are known to be typically scarce (Takwoingi 2013). Only one of the included studies compared the accuracy of two RDT brands, so we were unable to conduct comparative meta‐analyses to determine which brands were more sensitive and/or more specific. We intended to investigate any heterogeneity from the pooled analyses with pre‐specified factors, as stated in our secondary objective, but this was not possible due to the small number of studies included in the analyses.

For the diagnostic test accuracy of RDTs, there is a lack of a 'perfect reference standard'. PCR is often seen as the gold standard for malaria diagnosis, because it is less prone to observer error and more sensitive at low levels of parasitaemia (Han 2017; Snounou 1993). On the other hand, it is too analytically sensitive to be a gold standard, because it detects subclinical infections (e.g. in patients with partial immunity). Furthermore, PCR sometimes has poor sensitivity for the detection of mixed infections (Shokoples 2009). A small sample of the cases in our review are mixed infections using PCR as the reference standard (Alam 2011; Mendoza 2013), so the analysis may be flawed.

PCR is currently not widely available due to logistical constraints, namely the need for specially‐trained technicians and a well‐equipped laboratory. It is thus mostly used for research purposes and is less applicable in clinical settings. Thus, microscopy in the correct clinical setting, with well‐trained microscopists, remains the acceptable reference standard. This method is less costly than PCR, but infections can be missed if the slides are not examined carefully (Wongsrichanalai 2007). This raises the possibility that in some cases, the RDT results may in fact have been correct and the microscopy results incorrect. Alam 2011 verified RDT results against both microscopy and PCR separately, giving similar results of high specificity but lower sensitivity when verified against PCR. As mentioned previously, microscopy is more prone to observer error and is less sensitive at low levels of parasitaemia in comparison to PCR.

As reported in the Methodological quality of included studies, there was a high number of ’unclear’ evaluations of risk of bias and applicability due to poor reporting of study methods and characteristics. Nine studies (90%) did not provide enough information for us to adequately assess the selection of patients. Eight studies (80%) used an adequate reference standard, which was likely to have classified the target condition, but only four studies (40%) reported that readers of the reference standard were blinded to the results of the RDTs.

Applicability of findings to the review question

Due to the small number of studies included in this review, it is doubtful that the results obtained here can be considered to be generally applicable. Nevertheless, the findings show that the CareStart Malaria Pf/Pv Combo test verified by microscopy appeared to be both highly sensitive (missing 1% of cases) and highly specific (incorrectly classifying 1% of non‐cases as positives) in detecting P vivax alone or as part of a mixed infection. In contrast, the Falcivax Device Rapid test, verified by microscopy, appeared to be less sensitive (missing 23% of cases), but was similarly highly specific. This result should be interpreted with caution because only two studies were used to obtain the pooled estimates.

Furthermore, the RDTs are heterogeneous in terms of quality. The devices can give ambiguous test results, are prone to drying out in low‐humidity climates, resulting in lack of fluid migration. They are often not tested after they have been exposed to field conditions (Maltha 2013). In January 2020, the CareStart Malaria Pf/Pv Combo test produced by Access Bio Inc. was issued a WHO notice of concern due to their manufacturing quality assurance processes, which in turn could impact on patient safety (WHO 2020). Thus, in addition to considering results of test accuracy in published reports, end‐users must be attuned to outcomes of periodic monitoring procedures of regulatory authorities and WHO prequalification.

Comparison with previous systematic reviews

An existing Cochrane Review assessing RDTs for diagnosing uncomplicated non‐falciparum malaria was conducted in 2014 (Abba 2014). A subset of the review included RDTs that diagnosed P vivax. Our review only assessed the diagnostic accuracy of RDTs that specifically detect P vivax with Pvivax‐specific LDH antigens, however all the RDTs included in this review are combo tests that are used to detect P falciparum as well as P vivax. We included 10 studies, of which five studies (Alam 2011; Chanie 2011; Mekonnen 2010; Singh 2010; Sharew 2009) were included in the review by Abba 2014. Four studies were published following the review by Abba 2014 (Costa 2019; Hailu 2014; Mussa 2019; Saha 2017). One study (Mendoza 2013) was excluded by Abba 2014 because non‐English language studies were excluded due to resource constraints. We included studies published in all languages.

Authors' conclusions

Differentiating between Plasmodium species is particularly important in areas of co‐endemicity whereby P vivax malaria is increasing proportionally, compared to P falciparum malaria. The main analysis included in this review was CareStart Malaria Pf/Pv Combo test against microscopy as the reference standard, and this RDT was found to be both highly sensitive and specific. Owing to concerns regarding methodological quality, these findings should be interpreted with caution. Only two RDT brands were assessed by more than one study in this review, so we could not assess differences in sensitivity and specificity between RDT brands. Studies often did not report on transport, storage conditions and quality control practices for RDTs such as lot testing prior to use, therefore damage to RDTs in transit or during the study period cannot be excluded and may have negatively impacted on test results. Studies also often did not report on the background parasitaemia density. This is an important variable which influences the performance of the RDTs.

More high‐quality studies are needed to assess and compare the accuracy of RDTs designed to detect P vivax. The studies should clearly report their sampling methods, if exclusion criteria were used and whether the results of index tests and reference standards were blinded from each other. Studies should also report the background parasitaemia density, if and how RDTs were quality assured prior to use, including details of transport, storage conditions, and lot testing.

In the future, the RDTs studied here may no longer be available. The quality of those that remain may be improved by the manufacturers. Thus, this review will require updating.

History

Protocol first published: Issue 2, 2019
Review first published: Issue 11, 2020

Search set	MEDLINE (PubMed)
1	Malaria, vivax [MeSH]
2	Plasmodium vivax [MeSH]
3	“Plasmodium vivax” or “P vivax” or “vivax malaria” or “non‐falciparum Malaria” Field: Title/Abstract
4	1 or 2 or 3
5	Exp Reagent kits, diagnostics [MeSH]
6	"Diagnostic Tests, Routine"[Mesh]
7	rapid diagnostic test* Field: Title/Abstract
8	RDT* Field: Title/Abstract
9	Dipstick* Field: Title/Abstract
10	“Rapid diagnostic device*” Field: Title/Abstract
11	MRDD Field: Title/Abstract
12	OptiMal Field: Title/Abstract
13	“Binax NOW” or “NOW‐ICT‐Malaria” or “NOW‐Malaria‐ICT” Field: Title/Abstract
14	ParaSight or Parascreen or ParaHIT Field: Title/Abstract
15	“SD Bioline” or Carestart or Falcivax or Malascan Field: Title/Abstract
16	Immunochromatograph* or Immuno‐chromatograph* Field: Title/Abstract
17	“Antigen detection” Field: Title/Abstract
18	“Rapid malaria antigen test*” Field: Title/Abstract
19	“Combo card test*” Field: Title/Abstract
20	Immunoassay [MeSH]
21	Chromatography [MeSH]
22	Enzyme‐linked immunosorbent assay [MeSH]
23	“Rapid test*” Field: Title/Abstract
24	“Card test*” Field: Title/Abstract
25	Rapid AND (detection* or device* or test* or kit*) Field: Title/Abstract
26	5 or 6 or 7 or 8 or 9 or 10 or 11 or 12 or 13 or 14 or 15 or 16 or 17 or 18 or 19 or 20 or 21 or 22 or 23 or 24 or 25
27	4 and 26

Search set	Web of Science
# 6	#5 AND #1Indexes=SCI‐EXPANDED, CPCI‐S Timespan=All years
# 5	#4 OR #3 OR #2Indexes=SCI‐EXPANDED, CPCI‐S Timespan=All years
# 4	TOPIC: (("alere trueline" or "Rapigen biocredit" or "SD bioline" or "standard Q" or VISITECT* or PALUTOP*)) ORTOPIC: (((necviparum or "one step" or meriscreen or "onsite malaria" or paraHIt* or Quickprofile)))Indexes=SCI‐EXPANDED, CPCI‐S Timespan=All years
# 3	TOPIC: (("ADVANCED QUALITY ONE STEP" or Tri‐line or BIOCREDIT or Biosynex or BioTracer or Carestart or Aspenmal)) ORTOPIC: (("combo RDT" or careUS or Coretests* or EGENS or EzDx or Falcivax or "first response" or Humasis or Karwa or KHB* or "malaria Pf (HRPII)/ PV"))Indexes=SCI‐EXPANDED, CPCI‐S Timespan=All years
# 2	TOPIC: ((("rapid diagnostic test*" or RDT* or dipstick or MRDD) OR (“Binax NOW” or “NOW‐ICT‐Malaria” or “NOW‐Malaria‐ICT”))) ORTOPIC: (((ParaSight or Parascreen or ParaHIT or “SD Bioline” or Carestart or Falcivax or Malascan))) ORTOPIC: (: ((Immunochromatograph* or Immuno‐chromatograph* or "card test" or chromatography)))Indexes=SCI‐EXPANDED, CPCI‐S Timespan=All years
# 1	TOPIC: (("plasmodium vivax" or "vivax malaria"))Indexes=SCI‐EXPANDED, CPCI‐S Timespan=All years

Database :	LILACS
Search on :	vivax malaria [Words] and "rapid test$" or PCR or diagnosis [Words]

Domain	Patient selection	Index test	Reference standard	Flow and timing
Description	Methods of patient selection	How index test was conducted and reported	How reference standard was conducted and reported	Describe patients that did not receive and time interval between index test or reference standard
Signalling questions (yes, no, or unclear)	Consecutive or random sample of patients? ‘Yes' if the study reported consecutive enrolment or random sampling of patients presenting with uncomplicated malaria symptoms. ‘No' if patients were purposefully selected, for example based on previous test results (such as using Rapid diagnostic tests (RDTs) only on those who tested positive for P vivax by microscopy/PCR). ‘Unclear' if the study did not explicitly state consecutive enrolment or random sampling, and it was unclear how patients were sampled.	Index test results interpreted without knowledge of the results of reference standard? ‘Yes' if RDT was performed fully blinded to reference standard result. ‘No' if reference standard result was known prior to interpretation of RDT result. ‘Unclear' if blinding was no explicitly stated.	PCRPCR likely to correctly classify the target condition?We will answer this question as ‘yes' for all studies because PCR is an objective test with binary outcomes. Thus, there is no room for subjective interpretation of test results or poor performance of the test leading to false negatives or false positives.‘Yes' if reference standard was PCR.MicroscopyMicroscopy likely to correctly classify the target condition?‘Yes' if microscopy was performed for one sample by two independent trained microscopist examining 100 high‐power fields.‘No' if microscopy was performed: by insufficiently trained individuals; by one individual only; with inadequate equipment; by viewing less than 100 microscopic fields before declaring negative. ‘Unclear' if insufficient information was provided.	Was there an appropriate interval between index test and reference standard? ‘Yes' if samples for RDT and microscopy or PCR were taken at the same time. We felt this was important given the transient parasitaemia associated with malaria. ‘No' if the samples for RDT and microscopy or PCR were taken at different times. ‘Unclear' if insufficient or no information on the time interval.
	Was a case‐control design avoided?This will always be ‘yes' because case control studies will be excluded from this review.	Pre‐specified threshold used?As the threshold is prespecified by the manufacturer in all RDTs, we will answer this question ‘yes' for all studies.	Reference standard results interpreted without knowledge of the results of index test?We will answer this question ‘yes' for all studies using only PCR as the reference standard because PCR is an objective test with binary outcomes. Thus, there is no room for subjective interpretation of test results. ‘Yes' if results of microscopy were interpreted without knowledge of RDT results ‘No' if results of microscopy were interpreted with knowledge of RDT results ‘Unclear' if there is insufficient information on whether or not microscopy results were interpreted with knowledge of RDT results	Did all patients receive a reference standard? ‘Yes' if all participants received a microscopy or PCR. ‘No' if one or more participants did not receive microscopy or PCR. Or if the reference standard was applied depending on index test results ‘Unclear' if there is insufficient information to determine whether or not all patients received microscopy/PCR.
	Did the study avoid inappropriate exclusions? ‘Yes' if no patients were excluded after inclusion in the study or if exclusions are adequately described. ‘No' if specific populations were excluded (for example, pregnant patients, children or immunocompromised patients), ‘Unclear' if unreported or insufficient information given to make a decision.			Did all patients receive the same reference standard? We will answer this question ‘yes' if all participants in the study or a subset of participants in the study received the acceptable reference standard (microscopy, PCR, or both), which we specified as a criterion for inclusion in the review. ‘No' if participants did not receive the same reference standard. ‘Unclear' if there is insufficient information to determine whether or not all patients received the same reference standard.
				Were all patients included in the analysis? ‘Yes' if the number of participants in the two‐by‐two table matches the number of participants recruited into the study or if sufficient explanation was provided for any discrepancy. ‘No' if some participants recruited into the study were unaccounted for. ‘Unclear' if unreported or insufficient information given to make a decision.
Risk of bias (high, low, or unclear)	Could the selection of patients have introduced bias?	Could the conduct or interpretation of the index test have introduced bias?	Could the reference standard, its conduct, or its interpretation has introduced bias?	Could the patient flow have introduced bias?
Applicability concerns (high, low, or unclear)	Not applicable	Are there concerns that the index test, its conduct, or interpretation differs from the review question? ‘High' if the study describes inappropriate storage conditions for the index test, or if the index test has not been lot tested ‘Low' if the study describes suitable storage conditions for the index test that meet manufacturer's requirements and if the study has reported the index test has been lot tested ‘Unclear' if insufficient information to make a decision	Are there concerns that the target condition as defined by the reference standard does not match the review question?We will answer this question ‘low' for all studies because P vivax diagnosed by light microscopy or PCR does match the review question	Not applicable

Tests

Data tables by test

Test	No. of studies	No. of participants
1 CareStart Malaria Pf/Pv Combo test (Access Bio Inc, New Jersey, USA) (Microscopy)	4	2398
2 Falcivax Device Rapid test for malaria Pv/Pf (Zephyer Biomedicals Goa) (Microscopy)	2	710
3 Immuno‐Rapid Malaria Pf/Pv test (Wama Diagnostica, Sao Paulo, Brazil) (Microscopy)	1	181
4 OnSite Pf/Pv test (CTK Biotech Inc, USA) (Microscopy)	1	338
5 SD Bioline Malaria Ag Pf/Pv test (Microscopy)	1	200
6 Falcivax Device Rapid test for malaria Pv/Pf (Zephyer Biomedicals Goa) (PCR)	1	338
7 OnSite Pf/Pv test (CTK Biotech Inc, USA) (PCR)	1	338
8 SD Bioline Malaria Ag Pf/Pv test (PCR)	1	200
9 Test Malaria Pf/Pv rapid test (Alltest Biotech, China) (PCR)	1	59
10 SD Bioline Malaria Ag Pf/Pv test (Microscopy corrected by PCR)	1	383

Alam 2011

Study characteristics
Patient Sampling	Study design: cross‐sectional studyRecruitment: did not state consecutive or random samplingStudy period: May 2009 to August 2010Population: 338 febrile patients referred for microscopy to diagnose malaria diagnosis at a health facilityInclusion and exclusion criteria: not reported
Patient characteristics and setting	Sex: 49.7% male, 50.3% femaleAge: median = 14 years, range 18 months to 82 yearsSetting: Matiranga Upazila Health Complex (UHC), in Matiranga Upazila (sub‐district) of Khagrachari district, south‐eastern part of BangladeshMalaria transmission: perennial transmission of malaria with 2 peaks in pre‐monsoon (March to May) and post‐monsoon (September to November) periods
Index tests	RDT brand(s): OnSite Pf/Pv test (CTK Biotech Inc, USA) and Falcivax Device Rapid test for malaria Pv/Pf (Zephyer Biomedicals, Goa)Batch number: not reportedLot testing: not reportedStorage conditions: unclear, reported manufacturer's instructions were followed for useBlinding: not reported
Target condition and reference standard(s)	Target condition(s):P falciparum and P vivaxReference standard(s): PCR and microscopyMicroscopy details: 200 high powered fields Two microscopists independently examined each microscopic slide; one of which was employed by the study and the other was posted at Matiranga UHC. Slide considered positive only when the two microscopists were in agreement. Discrepancies were resolved by a third microscopist. PCR details: Did not report who performed PCR. Detection limit of 5‐10 parasites/μL. Blinding: not reported
Flow and timing	Appropriate interval between index test and reference standard: one blood sample taken from each patient.Invalid test results: None reported.
Comparative
Notes
Methodological quality
Item	Authors' judgement	Risk of bias	Applicability concerns
DOMAIN 1: Patient Selection
Was a consecutive or random sample of patients enrolled?	Unclear
Was a case‐control design avoided?	Yes
Did the study avoid inappropriate exclusions?	Unclear
Could the selection of patients have introduced bias?		Unclear risk
Are there concerns that the included patients and setting do not match the review question?			Low concern
DOMAIN 2: Index Test (All tests)
Were the index test results interpreted without knowledge of the results of the reference standard?	Unclear
If a threshold was used, was it pre‐specified?	Yes
Could the conduct or interpretation of the index test have introduced bias?		Unclear risk
Are there concerns that the index test, its conduct, or interpretation differ from the review question?			Unclear
DOMAIN 3: Reference Standard
Is the reference standards likely to correctly classify the target condition?	Yes
Were the reference standard results interpreted without knowledge of the results of the index tests?	Unclear
Could the reference standard, its conduct, or its interpretation have introduced bias?		Unclear risk
Are there concerns that the target condition as defined by the reference standard does not match the question?			Low concern
DOMAIN 4: Flow and Timing
Was there an appropriate interval between index test and reference standard?	Yes
Did all patients receive the same reference standard?	Yes
Were all patients included in the analysis?	Yes
Could the patient flow have introduced bias?		Low risk

Chanie 2011

Study characteristics
Patient Sampling	Study design: cross‐sectional studyRecruitment: consecutiveStudy period: December 2009 and January 2010Population: 1092 febrile patients who had clinical symptoms of malaria and visited the outpatient department (OPD) of three health facilities.Inclusion and exclusion criteria: no exclusion criteria, unless the patient or the guardians of children less than 18 years old did not consent to participate.
Patient characteristics and setting	Sex: 51.4% male, 48.6% femaleAge: median = 22.3 years, SD: 12.8Setting: 75 (61.81%), 238 (21.8%) and 179 (16.4%) patients were respectively from Melkawerer Health Centre, Gewane Health Centre and Dubti Hospital, in the Afar Region, Northeast EthiopiaMalaria transmission: The study reported the transmission of malaria as unstable, with some areas of perennial transmission.Other patient characteristics: 12.5% of the patients had anti‐malarial therapy in the preceding month.
Index tests	RDT brand(s): CareStart Malaria Pf/Pv Combo test (Access Bio Inc, Somerset, NJ)Batch number: not reportedLot testing: Lot No H38 IV and Lot No H28 IVStorage conditions: Test kits were kept at the local temperature of the region without any controlling system of the storage temperature during data collection. The quality of the package desiccant was checked before use.Blinding: Microscopy and RDT were performed independently by malaria technicians, results were recorded separately.
Target condition and reference standard(s)	Target condition(s):P falciparum and P vivaxReference standard(s): microscopyMicroscopy details: minimum of 100 high powered fields microscopy performed by experienced malaria technicians. The study did not explicitly state the number of observers or repeats. 20% of the positive and 10% of the negative slides and discordant results between CareStartTM Malaria Pf/Pv Combo test and those of microscopy were examined by another well experienced technician. Blinding: Microscopy and RDT was performed independently by malaria technicians, results were recorded separately.
Flow and timing	Appropriate interval between index test and reference standard: one blood sample taken from each patient.Invalid test results: None reported.
Comparative
Notes
Methodological quality
Item	Authors' judgement	Risk of bias	Applicability concerns
DOMAIN 1: Patient Selection
Was a consecutive or random sample of patients enrolled?	Yes
Was a case‐control design avoided?	Yes
Did the study avoid inappropriate exclusions?	Yes
Could the selection of patients have introduced bias?		Low risk
Are there concerns that the included patients and setting do not match the review question?			Unclear
DOMAIN 2: Index Test (All tests)
Were the index test results interpreted without knowledge of the results of the reference standard?	Yes
If a threshold was used, was it pre‐specified?	Yes
Could the conduct or interpretation of the index test have introduced bias?		Low risk
Are there concerns that the index test, its conduct, or interpretation differ from the review question?			Low concern
DOMAIN 3: Reference Standard
Is the reference standards likely to correctly classify the target condition?	Unclear
Were the reference standard results interpreted without knowledge of the results of the index tests?	Yes
Could the reference standard, its conduct, or its interpretation have introduced bias?		Unclear risk
Are there concerns that the target condition as defined by the reference standard does not match the question?			Low concern
DOMAIN 4: Flow and Timing
Was there an appropriate interval between index test and reference standard?	Yes
Did all patients receive the same reference standard?	Yes
Were all patients included in the analysis?	Yes
Could the patient flow have introduced bias?		Low risk

Costa 2019

Study characteristics
Patient Sampling	Study design: prospective cross‐sectional studyRecruitment: consecutiveStudy period: November 2016 and April 2017Population: 181 febrile patients were recruited based on the hospital's regular admissionsInclusion and exclusion criteria: not reported
Patient characteristics and setting	Sex: 64.1% male, 35.9% femaleAge: median = 41.7 years, SD: 14.4Setting: Tertiary health unit at Fundação de Medicina Tropical Dr. Heitor Vieira Dourado in Western Brazilian AmazonMalaria transmission: study reported P vivax as "the most prevalent species in Brazil and extra‐Afican areas", but did not specify to what degree.Other patient characteristics: 93.3% of the patients had up to three previous episodes of malaria.
Index tests	RDT brand(s): Immuno‐Rapid Malaria Pf/Pv test (Wama Diagnostica, Sao Paulo, Brazil)Batch number: not reportedLot testing: not reportedStorage conditions: reported manufacturer's instructions were followed (2oC‐30oC until the expiry date)Blinding: The hospital laboratory staff who performed the RDT and thick blood smear analysis were blinded.
Target condition and reference standard(s)	Target condition(s):P vivax and P falciparumReference standard(s): microscopyMicroscopy details: number of high powered fields not reported Experienced microscopists examined the slides, however the study did not explicitly state the number of observers or repeat. Blinding: The hospital laboratory staff who performed the RDT and thick blood smear analysis were blinded.
Flow and timing	Appropriate interval between index test and reference standard: RDT and microscopy were performed on admission.Invalid test results: None reported
Comparative
Notes
Methodological quality
Item	Authors' judgement	Risk of bias	Applicability concerns
DOMAIN 1: Patient Selection
Was a consecutive or random sample of patients enrolled?	Yes
Was a case‐control design avoided?	Yes
Did the study avoid inappropriate exclusions?	Unclear
Could the selection of patients have introduced bias?		Unclear risk
Are there concerns that the included patients and setting do not match the review question?			Low concern
DOMAIN 2: Index Test (All tests)
Were the index test results interpreted without knowledge of the results of the reference standard?	Yes
If a threshold was used, was it pre‐specified?	Yes
Could the conduct or interpretation of the index test have introduced bias?		Low risk
Are there concerns that the index test, its conduct, or interpretation differ from the review question?			Unclear
DOMAIN 3: Reference Standard
Is the reference standards likely to correctly classify the target condition?	Unclear
Were the reference standard results interpreted without knowledge of the results of the index tests?	Yes
Could the reference standard, its conduct, or its interpretation have introduced bias?		Unclear risk
Are there concerns that the target condition as defined by the reference standard does not match the question?			Low concern
DOMAIN 4: Flow and Timing
Was there an appropriate interval between index test and reference standard?	Yes
Did all patients receive the same reference standard?	Yes
Were all patients included in the analysis?	Yes
Could the patient flow have introduced bias?		Low risk

Hailu 2014

Study characteristics
Patient Sampling	Study design: cross‐sectional studyRecruitment: did not state consecutive or random samplingStudy period: patients recruited in December 2011Population: 398 febrile patients who visited the outpatient department of a health centreInclusion and exclusion criteria: Patients with acute febrile illnesses (body temperature > 37.5 degrees C) or a history of fever during the last 2 weeks at the date of data collection were included. Patients who took antimalarials within the last 2 weeks before the data collection date or refused participation were excluded.
Patient characteristics and setting	Sex: 44.2% male, 55.8% femaleAgerange: 1 and 70 yearsSetting: Felegeselam Health Center in Pawe Special Woredam, Northwest EthiopiaMalaria transmission: malaria transmission takes place throughout the year and that P falciparum and P vivax are co‐endemic.
Index tests	RDT brand(s): CareStart Malaria Pf/Pv Combo test (Access Bio Inc, Somerset, NJ)Batch number: not reportedLot testing: not reportedStorage conditions: RDT was stored based at room temperature and the quality of package and expiration date was checked before use.Blinding: The results of the RDT were determined earlier than microscopic results, with strict blinding to microscopic examination.
Target condition and reference standard(s)	Target condition(s):P falciparum and P vivaxReference standard(s): microscopyMicroscopy details: number of high powered fields for microscopy not reported Two experienced malaria technologists conducted the microscopic examination independently and blindly. Results of their observation were recorded for later comparison and all discordant results were repeated and rechecked by the principal investigator who was also experienced. Blinding: RDT with strict blinding to microscopic examination.
Flow and timing	Appropriate interval between index test and reference standard: Blood sample was collected from each patient. Two out of 398 patients with invalid test results from RDT were retested by taking fresh blood. Unclear whether the same blood sample was used for microscopy for these patients.Invalid test results: Two patients with invalid test results.
Comparative
Notes
Methodological quality
Item	Authors' judgement	Risk of bias	Applicability concerns
DOMAIN 1: Patient Selection
Was a consecutive or random sample of patients enrolled?	Unclear
Was a case‐control design avoided?	Yes
Did the study avoid inappropriate exclusions?	Yes
Could the selection of patients have introduced bias?		Unclear risk
Are there concerns that the included patients and setting do not match the review question?			Low concern
DOMAIN 2: Index Test (All tests)
Were the index test results interpreted without knowledge of the results of the reference standard?	Yes
If a threshold was used, was it pre‐specified?	Yes
Could the conduct or interpretation of the index test have introduced bias?		Low risk
Are there concerns that the index test, its conduct, or interpretation differ from the review question?			Unclear
DOMAIN 3: Reference Standard
Is the reference standards likely to correctly classify the target condition?	Yes
Were the reference standard results interpreted without knowledge of the results of the index tests?	Yes
Could the reference standard, its conduct, or its interpretation have introduced bias?		Low risk
Are there concerns that the target condition as defined by the reference standard does not match the question?			Low concern
DOMAIN 4: Flow and Timing
Was there an appropriate interval between index test and reference standard?	Yes
Did all patients receive the same reference standard?	Yes
Were all patients included in the analysis?	Yes
Could the patient flow have introduced bias?		Low risk

Mekonnen 2010

Study characteristics
Patient Sampling	Study design: cross‐sectional studyRecruitment: did not state consecutive or random samplingStudy period: October 2007 and December 2008Population: 240 febrile patients who were clinically suspected of malaria and visited the outpatient department of a health cent er.Inclusion and exclusion criteria: not reported.
Patient characteristics and setting	Sex: 57.5% male, 42.5% femaleAge: mean = 25 years, range 1 and 60 yearsSetting: Serbo health cent er in Jimma zone, southern EthiopiaMalaria transmission: Study reported that P falciparum and P vivax were both prevalent.
Index tests	RDT brand(s): CareStart Malaria Pf/Pv Combo test (Access Bio Inc, Somerset, NJ)Batch number: not reportedLot testing: not reportedStorage conditions: manufacturer’s instruction was followed and the quality of package desiccant was checked before use.Blinding: The RDT and microscopy were performed by three experienced malaria technicians independently. The results of the RDT were determined before microscopic results with strict blinding to microscopic examination.
Target condition and reference standard(s)	Target condition(s):P falciparum and P vivaxReference standard(s): microscopyMicroscopy details: At least 300 high powered fields Three experienced technicians examined the slides independently. Results of their observation were recorded for later comparison and discordant results between microscopy and RDT were repeated. Blinding: RDT with strict blinding to microscopic examination
Flow and timing	Appropriate interval between index test and reference standard: Blood sample was collected from each patient.Invalid test results: None reported.
Comparative
Notes
Methodological quality
Item	Authors' judgement	Risk of bias	Applicability concerns
DOMAIN 1: Patient Selection
Was a consecutive or random sample of patients enrolled?	Unclear
Was a case‐control design avoided?	Yes
Did the study avoid inappropriate exclusions?	Unclear
Could the selection of patients have introduced bias?		Unclear risk
Are there concerns that the included patients and setting do not match the review question?			Low concern
DOMAIN 2: Index Test (All tests)
Were the index test results interpreted without knowledge of the results of the reference standard?	Yes
If a threshold was used, was it pre‐specified?	Yes
Could the conduct or interpretation of the index test have introduced bias?		Low risk
Are there concerns that the index test, its conduct, or interpretation differ from the review question?			Unclear
DOMAIN 3: Reference Standard
Is the reference standards likely to correctly classify the target condition?	Yes
Were the reference standard results interpreted without knowledge of the results of the index tests?	Yes
Could the reference standard, its conduct, or its interpretation have introduced bias?		Low risk
Are there concerns that the target condition as defined by the reference standard does not match the question?			Low concern
DOMAIN 4: Flow and Timing
Was there an appropriate interval between index test and reference standard?	Yes
Did all patients receive the same reference standard?	Yes
Were all patients included in the analysis?	Yes
Could the patient flow have introduced bias?		Low risk

Mendoza 2013

Study characteristics
Patient Sampling	Study design: retrospective cross‐sectional studyRecruitment: consecutiveStudy period: November 16 to December 2, 2010 in Córdoba and from June 14 to 25, 2011 in Chocó, ColombaPopulation: 383 patients who attended one of three clinics with microscopy for diagnosis of malaria.Inclusion and exclusion criteria: patient was considered as a probable case of malaria and at least 6 years of age (with consent), were included. Probable case defined as patients presenting with current or recent fever within 72 hours, who came from an endemic area in the last 15 days and who may or may not have an epidemiological relationship with diagnosed cases. The study excluded patients who did not consent to participation, lack of diligence of the clinical epidemiological record or who presented with symptoms of complicated malaria.
Patient characteristics and setting	Sex: 52.5% male, 47.5% femaleAge range: 6 and 92 yearsSetting: 233 patients came from Córdoba, of which 121 were from Tierralta and 112 from Puerto Libertador. The remaining 150 patients were recruited in the department of Chocó, in the municipality of Quibdo.Malaria transmission: The study reported Córdoba had the highest prevalence of P vivax, unclear for Chocó.Other patient characteristics: 7.8% of the patients received treatment for malaria in the previous month of recruitment.
Index tests	RDT brand(s): SD Bioline Malaria Ag Pf/Pv test (Standard Diagnostics Inc)Batch number: not reportedLot testing: not reportedStorage conditions: reported manufacturer's instructions were followed (1ºC‐40ºC)Blinding: The results of the RDT were determined and kept separate so it does not interfere with the reference standard results.
Target condition and reference standard(s)	Target condition(s):P falciparum and P vivaxReference standard(s): microscopy corrected by PCRMicroscopy details: Number of high powered fields not reported. Blood films were examined by two experienced readers independently and blindly. Discordant results were checked by a third reader. PCR details: Did not report who performed PCR. Detection limit not reported. When there were discordant results between microscopy and PCR, the result of the PCR was taken, except in those in which the thick drop showed parasitic forms and the PCR was negative. Blinding: The results of the RDT were determined and kept seperate so it does not interfere with the reference standard results.
Flow and timing	Appropriate interval between index test and reference standard: Multiple blood samples were taken at the same time for each patient.Invalid test results: None reported.
Comparative
Notes
Methodological quality
Item	Authors' judgement	Risk of bias	Applicability concerns
DOMAIN 1: Patient Selection
Was a consecutive or random sample of patients enrolled?	Yes
Was a case‐control design avoided?	Yes
Did the study avoid inappropriate exclusions?	Yes
Could the selection of patients have introduced bias?		Low risk
Are there concerns that the included patients and setting do not match the review question?			Low concern
DOMAIN 2: Index Test (All tests)
Were the index test results interpreted without knowledge of the results of the reference standard?	Yes
If a threshold was used, was it pre‐specified?	Yes
Could the conduct or interpretation of the index test have introduced bias?		Low risk
Are there concerns that the index test, its conduct, or interpretation differ from the review question?			Unclear
DOMAIN 3: Reference Standard
Is the reference standards likely to correctly classify the target condition?	Yes
Were the reference standard results interpreted without knowledge of the results of the index tests?	Yes
Could the reference standard, its conduct, or its interpretation have introduced bias?		Low risk
Are there concerns that the target condition as defined by the reference standard does not match the question?			Low concern
DOMAIN 4: Flow and Timing
Was there an appropriate interval between index test and reference standard?	Yes
Did all patients receive the same reference standard?	Yes
Were all patients included in the analysis?	Yes
Could the patient flow have introduced bias?		Low risk

Mussa 2019

Study characteristics
Patient Sampling	Study design: cross‐sectional studyRecruitment: did not state consecutive or random samplingStudy period: Not reportedPopulation: 59 suspected patients with P falciparum infection from different clinical centers were recruited.Inclusion and exclusion criteria: not reported
Patient characteristics and setting	Sex: 45.8% male, 54.2% femaleAge: Not reportedSetting: different clinics in Omdurman, SudanMalaria transmission: Not reported
Index tests	RDT brand(s): Test Malaria Pf/Pv rapid test cassette (Alltest Biotech, China)Batch number: not reportedLot testing: not reportedStorage conditions: Unclear, reported manufacturer's instructions were followed for useBlinding: not reported
Target condition and reference standard(s)	Target condition(s):P falciparumReference standard(s): PCRPCR details: Did not report who performed PCR. Did not report detection limit. Blinding: not reported
Flow and timing	Appropriate interval between index test and reference standard: one blood sample taken from each patient.Invalid test results: None reported.
Comparative
Notes	Contacted author specifically for P vivax results.
Methodological quality
Item	Authors' judgement	Risk of bias	Applicability concerns
DOMAIN 1: Patient Selection
Was a consecutive or random sample of patients enrolled?	Unclear
Was a case‐control design avoided?	Yes
Did the study avoid inappropriate exclusions?	Unclear
Could the selection of patients have introduced bias?		Unclear risk
Are there concerns that the included patients and setting do not match the review question?			Low concern
DOMAIN 2: Index Test (All tests)
Were the index test results interpreted without knowledge of the results of the reference standard?	Unclear
If a threshold was used, was it pre‐specified?	Yes
Could the conduct or interpretation of the index test have introduced bias?		Unclear risk
Are there concerns that the index test, its conduct, or interpretation differ from the review question?			Unclear
DOMAIN 3: Reference Standard
Is the reference standards likely to correctly classify the target condition?	Yes
Were the reference standard results interpreted without knowledge of the results of the index tests?	Unclear
Could the reference standard, its conduct, or its interpretation have introduced bias?		Unclear risk
Are there concerns that the target condition as defined by the reference standard does not match the question?			Low concern
DOMAIN 4: Flow and Timing
Was there an appropriate interval between index test and reference standard?	Yes
Did all patients receive the same reference standard?	Yes
Were all patients included in the analysis?	Yes
Could the patient flow have introduced bias?		Low risk

Saha 2017

Study characteristics
Patient Sampling	Study design: cross‐sectional studyRecruitment: consecutiveStudy period: Not reportedPopulation: 200 febrile patients in whom clinicians suspected malaria and raised the investigations.Inclusion and exclusion criteria: Patients having a fever with chills and rigor in the absence of any obvious cause such as upper respiratory tract infection. All patients diagnosed and/or treated with antimalarial drugs within the past six months were excluded.
Patient characteristics and setting	Sex: 56.0% male, 44.0% femaleAge: mean = 34.6 years, <10 years: 2.5%, 11‐20 years: 20.5%, 21‐60 years: 68.5%, >61 years: 8.5%Setting: tertiary care hospital setting at the outpatient department of Kasturba Hospital, Sewagram, Wardha in Central IndiaMalaria transmission: Not reported
Index tests	RDT brand(s): SD Bioline Malaria Ag Pf/Pv test (Standard Diagnostics Inc)Batch number: not reportedLot testing: not reportedStorage conditions: Unclear, although reported manufacturer's instructions were followed for use.Blinding: The RDT, microscopy and PCR were performed by different technicians and results of all three tests were kept blind.
Target condition and reference standard(s)	Target condition(s):P falciparum and P vivaxReference standard(s): PCR and microscopyMicroscopy details: Number of high powered fields not reported Two microscopists having >15years of experience independently examined the slides. If there was discordance, this was resolved by a third reader (microbiologists). PCR details: Different techinicans conducted the microscopy and PCR. Limit detection not reported. Blinding: The RDT, microscopy and PCR were performed by different technicians and results of all three tests were kept blind.
Flow and timing	Appropriate interval between index test and reference standard: one blood sample taken from each patient.Invalid test results: None reported.
Comparative
Notes	Contacted author specifically for P vivax results.
Methodological quality
Item	Authors' judgement	Risk of bias	Applicability concerns
DOMAIN 1: Patient Selection
Was a consecutive or random sample of patients enrolled?	Yes
Was a case‐control design avoided?	Yes
Did the study avoid inappropriate exclusions?	Yes
Could the selection of patients have introduced bias?		Low risk
Are there concerns that the included patients and setting do not match the review question?			Low concern
DOMAIN 2: Index Test (All tests)
Were the index test results interpreted without knowledge of the results of the reference standard?	Yes
If a threshold was used, was it pre‐specified?	Yes
Could the conduct or interpretation of the index test have introduced bias?		Low risk
Are there concerns that the index test, its conduct, or interpretation differ from the review question?			Unclear
DOMAIN 3: Reference Standard
Is the reference standards likely to correctly classify the target condition?	Yes
Were the reference standard results interpreted without knowledge of the results of the index tests?	Yes
Could the reference standard, its conduct, or its interpretation have introduced bias?		Low risk
Are there concerns that the target condition as defined by the reference standard does not match the question?			Low concern
DOMAIN 4: Flow and Timing
Was there an appropriate interval between index test and reference standard?	Yes
Did all patients receive the same reference standard?	Yes
Were all patients included in the analysis?	Yes
Could the patient flow have introduced bias?		Low risk

Sharew 2009

Study characteristics
Patient Sampling	Study design: cross‐sectional studyRecruitment: did not state consecutive or random samplingStudy period: November and December 2008Population: 668 febrile patients who were clinically suspected of malaria and visited the outpatient department of two health centers.Inclusion and exclusion criteria: not reported.
Patient characteristics and setting	Sex: 54.0% male, 46.0% femaleAge range: 6 months and 75 yearsSetting: Bussa and Kella health centers in Wondo Genet area, southern EthiopaMalaria transmission: Study reported that P falciparum and P vivax were both prevalent.
Index tests	RDT brand(s): CareStart Malaria Pf/Pv Combo test (Access Bio Inc, Somerset, NJ)Batch number: not reportedLot testing: not reportedStorage conditions: manufacturer’s instruction was followed and the quality of package desiccant was checked before use.Blinding: The RDT and microscopy were performed by two experienced malaria technicians independently. The results of the RDT were determined before microscopic results with strict blinding to microscopic examination.
Target condition and reference standard(s)	Target condition(s):P falciparum and P vivaxReference standard(s): microscopyMicroscopy details: At least 100 high powered fields Two experienced technicians examined the slides independently, which was checked by the team leader who is also experienced. Discordant results between microscopy and RDT were repeated. Blinding: RDT with strict blinding to microscopic examination
Flow and timing	Appropriate interval between index test and reference standard: Blood sample was collected from each patient.Invalid test results: None reported.
Comparative
Notes
Methodological quality
Item	Authors' judgement	Risk of bias	Applicability concerns
DOMAIN 1: Patient Selection
Was a consecutive or random sample of patients enrolled?	Unclear
Was a case‐control design avoided?	Yes
Did the study avoid inappropriate exclusions?	Unclear
Could the selection of patients have introduced bias?		Unclear risk
Are there concerns that the included patients and setting do not match the review question?			Low concern
DOMAIN 2: Index Test (All tests)
Were the index test results interpreted without knowledge of the results of the reference standard?	Yes
If a threshold was used, was it pre‐specified?	Yes
Could the conduct or interpretation of the index test have introduced bias?		Low risk
Are there concerns that the index test, its conduct, or interpretation differ from the review question?			Unclear
DOMAIN 3: Reference Standard
Is the reference standards likely to correctly classify the target condition?	Yes
Were the reference standard results interpreted without knowledge of the results of the index tests?	Yes
Could the reference standard, its conduct, or its interpretation have introduced bias?		Low risk
Are there concerns that the target condition as defined by the reference standard does not match the question?			Low concern
DOMAIN 4: Flow and Timing
Was there an appropriate interval between index test and reference standard?	Yes
Did all patients receive the same reference standard?	Yes
Were all patients included in the analysis?	Yes
Could the patient flow have introduced bias?		Low risk

Singh 2010

Study characteristics
Patient Sampling	Study design: cross‐sectional studyRecruitment: consecutiveStudy period: August and December 2009Population: 372 febrile patients with clinical suspicion of malaria who visited field clinicsInclusion and exclusion criteria: excluded pregnant women and patients who took antimalarials.
Patient characteristics and setting	Sex: Not reportedAge: mean = 15 years, SD: 14.1Setting: Bajag Primary Health Centre (PHC) of district Dindori and Satanwada PHC of district Shivpuri, IndiaMalaria transmission: study reported that both P falciparum and P vivax as co‐endemic in the study area.
Index tests	RDT brand(s): Falcivax Device Rapid test for malaria Pv/Pf (Zephyer Biomedicals, Goa)Batch number: not reportedLot testing: not reportedStorage conditions: "For testing temperature stability of the tests, RDTs were stored at 25°C on receipt in the study sites, then allocated to separate groups for storage at 35°C & 45°C for 90 days, at 60°C for 48 hours, and at ‐10°C for 60 minutes before testing [21]. At the start of the study, the incubators were stabilized at the required temperature for three days before the RDTs to be tested were placed inside. RDTs were removed from storage to reach room temperature for 2 hours before testing and comparisons were made with control RDTs kept at 25°C until use."Blinding: Microscopy examination was conducted without reference to the results of RDTs.
Target condition and reference standard(s)	Target condition(s):P falciparum and P vivaxReference standard(s): microscopyMicroscopy details: 100 high powered fields Only one experienced microscopist conducted the examination in the laboratory. Any discrepancies between the reference standard or index test were re‐examined by another expert technician who was blinded to the results of microscopy and RDT. Blinding: microscopy conducted without reference to the results of RDTs.
Flow and timing	Appropriate interval between index test and reference standard: Multiple samples were taken at the same time.Invalid test results: None reported.
Comparative
Notes
Methodological quality
Item	Authors' judgement	Risk of bias	Applicability concerns
DOMAIN 1: Patient Selection
Was a consecutive or random sample of patients enrolled?	Yes
Was a case‐control design avoided?	Yes
Did the study avoid inappropriate exclusions?	Yes
Could the selection of patients have introduced bias?		Low risk
Are there concerns that the included patients and setting do not match the review question?			Low concern
DOMAIN 2: Index Test (All tests)
Were the index test results interpreted without knowledge of the results of the reference standard?	Yes
If a threshold was used, was it pre‐specified?	Yes
Could the conduct or interpretation of the index test have introduced bias?		Low risk
Are there concerns that the index test, its conduct, or interpretation differ from the review question?			Unclear
DOMAIN 3: Reference Standard
Is the reference standards likely to correctly classify the target condition?	No
Were the reference standard results interpreted without knowledge of the results of the index tests?	Yes
Could the reference standard, its conduct, or its interpretation have introduced bias?		High risk
Are there concerns that the target condition as defined by the reference standard does not match the question?			Low concern
DOMAIN 4: Flow and Timing
Was there an appropriate interval between index test and reference standard?	Yes
Did all patients receive the same reference standard?	Yes
Were all patients included in the analysis?	Yes
Could the patient flow have introduced bias?		Low risk

Study	Reason for exclusion
Abdelraheem 2016	Insufficient data
Adams 2015	Wrong patient population
Adnan 2017	Insufficient data
Ageep 2013	Wrong target condition
Andrade 2010	Wrong index test
Arvind 2015	Predatory journal
Ashton 2010	Wrong index test
Ayorinde 2016	Wrong target condition
Ba 2017	Wrong index test
Bahk 2018	Wrong index test
Barber 2013	Wrong index test
Bell 2001	Wrong index test
Bendezu 2010	Wrong index test
Berhane 2017	Wrong study design
Berzosa 2018	Wrong index test
Bharti 2008	Wrong index test
Bharti 2013	Wrong study design
Bhide 2014	Wrong index test
Birhanie 2016	Wrong index test
Bisoffi 2014	Wrong study design
Britton 2016	Wrong index test
Chayani 2004	Wrong index test
Cho 2016	Wrong index test
Dahesh 2015	Full text not available
Dash 2013	Insufficient data
Deida 2019	Wrong index test
DeKoninck 2017	Wrong patient population
Dev 2004	Wrong index test
Dinzouna‐Boutamba 2014	Wrong index test
Dzakah 2014	Wrong patient population
Ehtesham 2015	Wrong patient population
Eibach 2013	Wrong index test
Elahi 2013	Wrong index test
Endeshaw 2012	Wrong index test
Falade 2016	Wrong target condition
Fernando 2004	Wrong index test
Fernando 2004a	Wrong index test
Foster 2014	Wrong study design
Fransisca 2015	Wrong index test
Gabrielli 2016	Wrong patient population
Ghai 2016	Wrong index test
Gupta 2018	Wrong index test
Harani 2006	Wrong index test
Hawash 2019	Wrong patient population
Jabeen 2016	Insufficient data
Jahan 2019	Wrong index test
Joseph 2018	Wrong index test
Karimov 2013	Full text not available
Kim 2013	Wrong index test
Kolaczinski 2004	Wrong index test
Kosack 2013	Wrong index test
Kumari 2014	Predatory journal
Liu 2013	Full text not available
Mallepaddi 2019	Wrong index test
Metzger 2011	Wrong index test
Moges 2012	Wrong index test
Mohon 2012	Wrong index test
Olasehinde 2018	Wrong index test
Pakalapati 2013	Wrong index test
Pattanasin 2003	Wrong index test
Puri 2013	Wrong index test
Rakotonirina 2008	Wrong index test
Ranjan 2016	Insufficient data
Ratsimbasoa 2007	Wrong index test
Ratsimbasoa 2008	Wrong index test
Samane 2010	Wrong index test
Selimuzzaman 2010	Wrong index test
Shaikh 2013	Insufficient data
Sharma 2014	Wrong study design
Singh 2000a	Wrong index test
Singh 2003	Wrong index test
Singh 2013	Wrong study design
Siwal 2018	Wrong patient population
Stijnberg 2013	Conference abstract
Strom 2014	Wrong target condition
Thongdee 2014	Wrong index test
Tjitra 1999	Wrong index test
Trouvay 2013	Wrong index test
Valecha 2003	Wrong index test
van den Broek 2006	Wrong index test
Vohra 2014	Insufficient data
Wang 2014	Full text not available
Wongsrichanalai 2003	Wrong index test
Woyessa 2013	Wrong study design
Xiaodong 2013	Wrong index test
Yan 2013	Wrong index test

Boni 2015

Patient Sampling	Patients were recruited from two provinces in Central Vietnam between January and August 2015. The sampling method and inclusion/exclusion criteria were not reported in the abstract.
Patient characteristics and setting	The prevalence, number of patients recruited and characteristics of patients were not reported in the abstract.
Index tests	RDT brand was not reported in abstract. No information on blinding, batch number of RDT, lot testing or storage conditions in the abstract.
Target condition and reference standard(s)	The target conditions were P falciparum and P vivax. The reference standard was microscopy examined by at least two expert microscopists. No information on number of high powered field or blinding in the abstract.
Flow and timing	Unclear whether the index test and reference standard were performed at the same time and if blood sample was taken at the same time for the tests.
Comparative
Notes	Unable to deduce the number of true positives, false positives, false negatives and true negatives for P vivax. Contacted authors for more details on methodology and results.

Cheng 2013

Patient Sampling	This was a cross‐sectional study, patients were recruited in 2008 and in 2011. The study did not explicitly state consecutive or random sampling. Inclusion and exclusion criteria were not reported. Unclear how the febrile patients were recruited, i.e. whether they presented themselves to a health centre.
Patient characteristics and setting	202 febrile patients (49 patients in 2008 and 153 in 2011) with fever of unknown origin in Kachine Myanmar and in Yunnan, China were recruited. 13 healthy patients were also recruited in Beijing, China, however they were not used in the analysis of the RDT. The study did not describe the characteristics of patients recruited. The study reported malaria as endemic in study area.
Index tests	CareStart Malaria HRP2/pLDH combo test (Access Bio Inc., Somerset, NJ). In the study there was no information on batch number, lot testing or storage conditions, however the study stated that the RDT was done according to manufacturer's protocol. The study did not mention blinding.
Target condition and reference standard(s)	The target conditions were P falciparum and P vivax. The reference standards were PCR and microscopy. The study reported a minimum of 100 high powered fields for microscopy. Two professional microscopists conducted the microscopic examination independently. Unclear how discrepancies for microscopy results between the two microscopists were handled, if any. With PCR, unable to deduce the number of true positives, false positives, false negatives and true negatives.
Flow and timing	The tests were performed at different times, but the blood sample was taken at the same time. All samples were frozen and stored at ‐80 oC but 2008 samples suffered freeze–and–thaw cycles. Only the 2011 samples were tested using the RDT but this was stated as 143 patients in the results rather than 153.
Comparative
Notes	Unclear whether the RDT used by the study is eligible for this review because the specific brand name was not mentioned in the main study publication, contacted authors for further information.

Reda 2016

Patient Sampling	This was a cross‐sectional study. Patients were recruited between November and December 2014 in two health centres in Adam and Amaya, Oromia region, Ethiopia. Inclusion/ exclusion criteria or sampling method were not reported.
Patient characteristics and setting	Febrile patients with symptoms of malaria who visited the two health facilities were recruited. The study abstract did not include the prevalence of malaria or the characteristics of patients recruited."A total of 547 febrile patients were diagnosed, of which 127 were microscopy positive for Pf (n=38) and Pv (n=85)".
Index tests	CareStart Malaria Ag Pf/Pv combo test (Access Bio Inc., Somerset, NJ) and SD BIOLINE malaria AG PF/PV test. The RDTs were performed following manufacturer's instructions. In the study abstract, there was no information on batch number or lot testing. The study abstract did not mention blinding.
Target condition and reference standard(s)	The target conditions were P falciparum and P vivax. The reference standard was microscopy. Microscopic examination was done under 100x magnifications. No information on blinding or who performed the microscopic examination in the abstract.
Flow and timing	Unclear whether the index test and reference standard were performed at the same time and if blood sample was taken at the same time for the tests.
Comparative
Notes	Unable to deduce the number of true positives, false positives, false negatives and true negatives for P vivax. Contacted authors for more details on methodology and results.