Mixed malaria infections: A critical review and Detect-Characterise-Act surveillance framework

Abstract

Mixed malaria infections, involving concurrent Plasmodium species and/or multiple genetically distinct parasite clones, remain under-measured in routine malaria programmes despite implications for diagnosis, treatment, drug-resistance surveillance, and elimination planning. This critical narrative review synthesises recent evidence, prioritising 2020–2026 literature, and proposes a Detect-Characterise-Act framework linking point-of-care detection, molecular or genomic confirmation, and clinical or surveillance responses. The review argues that mixed infections should be treated as operational signals of diagnostic sensitivity, species-specific treatment adequacy, parasite population connectivity, diagnostic escape, and antimalarial resistance. Microscopy and many rapid diagnostic tests often underestimate mixed infections, particularly when secondary species are low-density, HRP2-based tests are affected by pfhrp2/pfhrp3 deletions, or non-P. falciparum species are not assayed. Genomic approaches, including targeted amplicon sequencing, nanopore sequencing, and whole-genome sequencing, can resolve species identity, multiplicity of infection, minority resistance variants, and diagnostic-target integrity, but require rigorous thresholds for read depth, allele frequency, contamination control, and sample representativeness. The review contributes four clarifications: distinguishing mixed-species infection, within-species multiplicity, and within-host resistance-associated variation; integrating pfhrp2/pfhrp3-mediated diagnostic escape; clarifying the staged interpretation of minority resistance-associated variants in Africa; and reframing mixed infection as a marker of programme maturity. Countries able to detect, genotype, report, and respond to mixed infections are better positioned to protect case management, contain resistance, and sustain malaria-elimination gains.

Graphical abstract

Highlights

• •This review distinguishes mixed-species infection, polyclonality, and resistance variation.
• •Detect-Characterise-Act links diagnosis, genomics, treatment, and policy.
• •Minority variants and deletions in pfhrp2 and pfhrp3 require staged verification.
• •Tiered surveillance must balance evidence, feasibility, cost, and equity.

1. Introduction

Malaria remains one of the most persistent infectious-disease challenges, despite major gains in prevention, diagnosis, and treatment. A recent World Malaria Report estimates 282 million cases and 610,000 deaths in 2024, with the WHO African Region continuing to bear the greatest burden and antimalarial drug resistance now highlighted as a threat to recent progress (WHO, 2025a). These contemporary pressures make mixed malaria infections more important than their historical treatment as diagnostic anomalies suggests. As malaria programmes move from broad control toward precision surveillance, mixed infections expose whether systems can detect low-density parasites, recognise non-P. falciparum species, interpret discordant diagnostic results, and act on genomic signals of resistance or importation.

In high-transmission settings, repeated infectious bites can generate polyclonal P. falciparum infections and mixed-species infections involving P. vivax, P. malariae, P. ovale curtisi, P. ovale wallikeri (Souleiman et al., 2024). In parts of Southeast Asia, mixed-species dynamics may also involve zoonotic malaria, particularly P. knowlesi, whereas P. simium is relevant in parts of South America. These transmission systems are geographically and ecologically distinct from the predominantly human malaria transmission occurring in sub-Saharan Africa (Van De Straat et al., 2022). In lower-transmission or elimination settings, even rare mixed infections can reveal imported cases, residual foci, diagnostic blind spots, or hidden reservoirs capable of maintaining transmission (Silwamba et al., 2025).

The earlier literature on mixed malaria infections often treated them as diagnostic anomalies or descriptive prevalence findings. Thus, there is a need for a more explicit conceptual shift. Mixed infections should be understood as a convergence point for four contemporary concerns: diagnostic target failure, species-specific treatment gaps, parasite population connectivity, and genomic signals of drug resistance. HRP2-based rapid diagnostic tests remain central to malaria case management, yet pfhrp2/pfhrp3 deletions can cause false-negative P. falciparum results and complicate the interpretation of apparent non-P. falciparum or mixed infections (Poti et al., 2020; WHO, 2024a, 2024b). At the same time, artemisinin partial resistance has been confirmed in several African settings and is suspected in others, changing the clinical and surveillance significance of mixed-clone infections (Rosenthal et al., 2024; WHO, 2024c). A mixed infection that is incorrectly typed, under-treated, or excluded from genomic surveillance may therefore distort estimates of both transmission and therapeutic efficacy.

Mixed malaria infection is used in this review as an umbrella expression requiring further analytical specification. The review distinguishes infections involving more than one Plasmodium species from infections containing multiple genetically distinguishable clones of the same species and from infections in which resistance-associated alleles occur at different within-host frequencies. These phenomena arise at different biological levels, require different laboratory and analytical methods, and do not automatically imply the same clinical or surveillance response.

Accordingly, the central question guiding this critical narrative review is: How should mixed malaria infections be detected, molecularly or genomically characterised, interpreted, and translated into proportionate clinical or public health action? The review focuses on the analytical pathway from an initial diagnostic observation to species confirmation, within-species diversity assessment, resistance-marker or diagnostic-target investigation, and the selection of an appropriate response. Technologies are considered according to the question they can answer rather than according to technical novelty alone.

The objectives are to: (i) define and analytically separate the principal forms of mixed malaria infection; (ii) critically assess laboratory approaches for species confirmation, parasite quantification, multiplicity estimation, resistance-marker detection, diagnostic-target investigation, and recurrence classification; (iii) identify common sources of technical and interpretive error; and (iv) operationalise a Detect-Characterise-Act framework that distinguishes individual patient management from population-level surveillance and policy decisions.

In this review, multiplicity of infection is the preferred general term for the estimated number of genetically distinguishable lineages within a species. Complexity of infection is used only when referring to an analytical metric or source that specifically employs that term. Polyclonal infection is used descriptively for the biological presence of more than one lineage and should not be interpreted as a precise numerical estimate.

2. Evidence-synthesis approach and methodological rigour

This study was conducted as a critical narrative review rather than a formal systematic review. Its purpose was not to calculate pooled prevalence, diagnostic accuracy, or treatment effects, but to integrate heterogeneous evidence from malaria epidemiology, diagnostics, genomics, drug-resistance surveillance, clinical management, and programme implementation into a decision-oriented framework. Quantitative pooling was considered inappropriate because the relevant studies differed substantially in their definitions of mixed infection, diagnostic and genomic methods, study populations, transmission settings, sampling periods, and reported outcomes (Kotepui et al., 2020).

The review was informed by the Scale for the Assessment of Narrative Review Articles (SANRA), particularly its domains relating to justification of the review’s importance, statement of aims, description of the literature search, appropriate referencing, scientific reasoning, and presentation of relevant evidence (Baethge et al., 2019). To reduce selective interpretation, the synthesis explicitly separates strong evidence from plausible inference, identifies areas of uncertainty, and avoids presenting any single diagnostic or genomic technology as universally superior. This approach strengthens methodological rigour while preserving the flexibility required for a narrative review of a heterogeneous, multidisciplinary evidence base.

The evidence search was conducted for literature published from 1 January 2020 to 14 July 2026, with older foundational studies retained only when they established concepts that remain necessary for interpretation, such as molecular diagnosis, within-host parasite diversity, P. vivax relapse biology, or narrative-review quality standards. Searches were conducted in PubMed/MEDLINE, Scopus, Web of Science Core Collection, Google Scholar, WHO Institutional Repository for Information Sharing, WHO Global Malaria Programme resources, and journal/publisher websites for article verification. Reference lists of highly relevant reviews, genomic-surveillance studies, and WHO guidance documents were also screened to identify additional sources that may not have been captured by database searching.

Search terms were combined using Boolean logic and adapted to each database. Core terms included: “mixed malaria infection”, “mixed Plasmodium infection”, “Plasmodium falciparum vivax mixed infection”, “Plasmodium ovale”, “Plasmodium malariae”, “multiplicity of infection”, “complexity of infection”, “within-host diversity”, “malaria genomic surveillance”, “amplicon sequencing malaria”, “nanopore sequencing malaria”, “whole genome sequencing Plasmodium”, “pfhrp2/pfhrp3 deletion”, “HRP2 rapid diagnostic test failure”, “K13 artemisinin resistance Africa”, “therapeutic efficacy malaria genotyping”, “vivax relapse radical cure”, “G6PD primaquine tafenoquine”, and “malaria elimination surveillance”. Searches were intentionally broad at the first stage and then narrowed by relevance to mixed-species infection, multiplicity of infection, diagnostic escape, drug resistance, or programme response.

Studies and guidance documents were eligible for inclusion if they met at least one of the following criteria: they reported mixed Plasmodium species infection; examined multiplicity or complexity of infection within a Plasmodium species; evaluated diagnostic performance for detecting mixed or low-density infections; reported genomic or molecular markers relevant to resistance, diagnostic-target deletion, or transmission inference; addressed treatment implications for co-endemic P. falciparum, P. vivax, P. ovale, or P. malariae; or provided normative guidance relevant to malaria diagnosis, treatment, resistance surveillance, vaccine implementation, or elimination planning. Priority was given to peer-reviewed original studies, systematic reviews, high-quality narrative reviews, multicountry or sentinel surveillance studies, genomic epidemiology papers, and WHO policy documents.

Documents were excluded when they were unrelated to human malaria, focused only on vector biology without relevance to mixed infection or surveillance response, lacked sufficient methodological information to judge diagnostic or genomic validity, duplicated data already reported in a more complete source, made claims unsupported by verifiable evidence, or could not be traced to a publisher, DOI, PubMed record, institutional repository, or authoritative organizational source. Conference abstracts, non-technical news items, and unverifiable online sources were not used to support substantive scientific claims. Preprints were not used as primary evidence where a peer-reviewed alternative was available. Publications in English were eligible for inclusion because the review team did not have resources for reliable translation of other languages.

Screening and selection were undertaken by both authors, with disagreements resolved through discussion. The final manuscript cites 36 sources, comprising 29 peer-reviewed publications and 7 WHO normative, technical, report, or institutional sources. Of the 29 peer-reviewed publications, 22 principal sources were included in the structured evidence matrix presented in Supplementary Table S1. The remaining seven peer-reviewed sources comprised six broader contextual malaria reviews and the SANRA methodological publication.

The 22 studies in Supplementary Table S1 were selected because they contributed directly to one or more core components of the synthesis: mixed-species detection, within-species multiplicity, within-host resistance-associated variation, diagnostic escape, recurrence, mobility, genomic surveillance, therapeutic efficacy, or programme response.

Evidence was appraised using five criteria: relevance to mixed-species infection, within-host diversity, or resistance mixtures; methodological transparency; diagnostic standard and limit of detection; species coverage; and direct policy relevance. Studies based only on microscopy were interpreted cautiously unless supported by molecular confirmation. PCR, qPCR, targeted amplicon sequencing, nanopore sequencing, and whole-genome sequencing were weighted more strongly for species and clone resolution but were assessed for contamination controls, read depth, allele-frequency thresholds, human-DNA background, and sample-selection bias.

To improve citation integrity, sources were excluded when they were unrelated to the review question, lacked sufficient methodological detail, could not be verified through a publisher, DOI, bibliographic database, or authoritative institutional source, or did not support the claim for which they were cited. Where evidence remained uncertain, the manuscript explicitly reports that uncertainty rather than presenting inference as established fact. Characteristics of the principal studies contributing to the synthesis were extracted into a structured evidence matrix. The extracted fields included geographical setting, study design or population, Plasmodium species, operational definition or analytical focus, diagnostic or genomic method, principal finding, methodological limitation, and relevance to the Detect-Characterise-Act framework. The core evidence is summarised in Supplementary Table S1.

3. Conceptual definitions and levels of analysis

The term “mixed malaria infection” is frequently used to describe biologically and analytically different phenomena. For the purposes of this review, three distinct levels are recognised: mixed-species infection, within-species multiplicity or complexity of infection, and within-host resistance-associated variation. Although these conditions may occur concurrently, they differ in the biological unit being measured, the laboratory methods required for their identification, the major sources of analytical error, and their implications for clinical management and surveillance.

3.1. Mixed-species infection

Mixed-species infection refers to the concurrent presence of two or more human-infecting Plasmodium species in the same host. Examples include simultaneous infection with P. falciparum and P. vivax, P. malariae, P. ovale curtisi, or P. ovale wallikeri. The principal biological unit is therefore the parasite species rather than the parasite clone or genotype. Mixed-species infections may have direct clinical relevance because different species differ in relapse biology, parasite-clearance patterns, disease progression, and treatment requirements (Kim et al., 2019; Price et al., 2020; Yan et al., 2023).

The detection of mixed-species infection depends strongly on the diagnostic method used. Microscopy and rapid diagnostic tests may fail to detect a low-density secondary species, particularly where one species dominates the blood-stage parasitaemia or where the diagnostic assay has restricted species coverage. Molecular methods generally detect a greater proportion of mixed infections, although their performance still depends on target selection, analytical sensitivity, contamination control, sample quality, and the range of species included in the assay (Kotepui et al., 2020; Holzschuh et al., 2022; Mediavilla et al., 2024). Consequently, the absence of a secondary species from a routine diagnostic result does not necessarily establish a true mono-species infection.

3.2. Multiplicity or complexity of infection

Multiplicity of infection refers to the presence or estimated number of genetically distinguishable parasite lineages within a single Plasmodium species. The related term “complexity of infection” is commonly used in genomic and population-genetic studies, although definitions and estimation methods vary among studies. In this review, multiplicity of infection is used as the preferred general term, while complexity of infection is retained when referring to a specific analytical method, metric, or cited study.

A polyclonal infection is the underlying biological condition in which more than one genetically distinct lineage of the same species occurs in a host. By contrast, multiplicity or complexity of infection is an estimate derived from molecular markers, amplicon sequencing, single-nucleotide polymorphism barcodes, microsatellites, antigen loci, whole-genome sequencing, or statistical mixture-inference methods. The estimated number of lineages may therefore vary according to marker diversity, sequencing depth, parasite density, allele-frequency thresholds, allele dropout, and the analytical model employed (MalariaGEN et al., 2021; De Cesare et al., 2024; Aranda-Díaz et al., 2025).

Multiplicity of infection may provide information about transmission intensity, repeated inoculation, superinfection, infection duration, population connectivity, or the interpretation of recurrent parasitaemia. However, it should not be treated as directly equivalent to mixed-species infection or as an independent indicator of clinical severity. A patient may have a mixed-species infection in which each species is represented by a single detectable lineage, or a mono-species infection containing several genetically distinguishable lineages.

3.3. Within-host resistance-associated variation

The expression “mixed-resistance infection” has sometimes been used descriptively for infections in which drug-sensitive and resistance-associated parasite genotypes coexist. However, this expression should not be treated as a formally established clinical category. In the present review, the more precise term “within-host resistance-associated variation” is used to describe the detection of wild-type and resistance-associated alleles, or multiple resistance-associated haplotypes, at different frequencies within the same sample.

The interpretation of such variation is technically and biologically complex. Detection of a low-frequency resistance-associated allele does not, by itself, demonstrate the presence of a clinically meaningful resistant parasite population. The signal may represent a genuine minority clone, polymerase or sequencing error, contamination, index misassignment, amplification bias, or an allele whose phenotypic association with resistance has not been adequately validated. Interpretation must therefore consider the specific mutation, the strength of evidence supporting its association with resistance, within-sample allele frequency, parasite density, sequencing depth, strand and read support, analytical error, reproducibility, and the relevant clinical and epidemiological context (Whitlock et al., 2021; Rosenthal et al., 2024; Aranda-Díaz et al., 2025).

For P. falciparum artemisinin partial resistance, a distinction should be maintained among: (i) an initial low-frequency sequence call; (ii) a technically confirmed within-host minority variant; (iii) a validated resistance-associated pfk13 mutation; (iv) repeated detection or evidence of clonal expansion; and (v) clinical evidence of delayed parasite clearance or reduced therapeutic efficacy. Evidence of artemisinin partial resistance in Africa has generally depended on integration of molecular findings with clinical, parasitological, in vitro, and population-level evidence rather than isolated detection of a sequence variant (Uwimana et al., 2020; Balikagala et al., 2021; Rosenthal et al., 2024; Henrici et al., 2025).

Accordingly, within-host resistance-associated variation should be treated as an operational surveillance concept requiring staged verification. A technically credible minority variant may justify confirmatory testing and enhanced surveillance, but it does not automatically justify modification of individual treatment or national drug policy.

Table 1 summarises the analytical distinction among mixed-species infection, multiplicity of infection, polyclonal infection, and within-host resistance-associated variation.

Table 1.

Analytical distinction among mixed-species infection, within-species multiplicity, and within-host resistance-associated variation.

Category	Biological unit measured	Operational definition	Principal laboratory or analytical approach	Common sources of misclassification	Clinical or surveillance interpretation
Mixed-species infection	Different Plasmodium species	Concurrent presence of two or more Plasmodium species in the same host	Expert microscopy, species-specific PCR/qPCR, validated multiplex assays, targeted sequencing or WGS	Low-density secondary species, incomplete species panels, morphological similarity, cross-reactivity, contamination and dominant-species masking	May affect species-specific treatment, radical-cure assessment, relapse prevention and species-distribution surveillance (Kim et al., 2019; Price et al., 2020; Holzschuh et al., 2022)
Multiplicity or complexity of infection	Genetically distinguishable parasite lineages within one species	Estimated number or diversity of parasite clones, strains or haplotypes within a sample	Microsatellite or antigen-marker genotyping, SNP barcodes, amplicon sequencing, targeted sequencing, WGS and mixture-inference models	Marker resolution, allele dropout, low parasite density, sequencing depth, multicopy loci, threshold selection and model assumptions	May inform transmission intensity, superinfection, recurrence classification and parasite connectivity, but is not directly equivalent to mixed-species infection or disease severity (MalariaGEN et al., 2021; De Cesare et al., 2024; Aranda-Díaz et al., 2025)
Polyclonal infection	Multiple genetically distinct lineages of the same species	Biological presence of more than one parasite lineage within a host	Identified through genotyping or sequencing methods capable of resolving within-host diversity	Inability of low-resolution markers to distinguish lineages, failure to detect minority clones	Descriptive biological condition underlying an MOI or COI estimate; should not be used as a precise numerical measure (MalariaGEN et al., 2021; Whitlock et al., 2021)
Within-host resistance-associated variation	Resistance-associated alleles or haplotypes occurring at different frequencies	Detection of wild-type and resistance-associated alleles, or more than one resistance profile, in the same sample	Allele-specific PCR, amplicon deep sequencing, targeted nanopore sequencing, digital PCR or WGS	Sequencing error, contamination, amplification bias, insufficient read depth, inappropriate allele-frequency thresholds and unvalidated markers	Requires technical verification and marker-specific interpretation; a low-frequency allele does not alone establish clinically significant resistance (De Cesare et al., 2024; Rosenthal et al., 2024; Aranda-Díaz et al., 2025)
Confirmed resistant parasite lineage	A parasite lineage supported by validated molecular and phenotypic evidence	Detection of a validated resistance marker accompanied by appropriate clinical, in vitro, ex vivo or epidemiological evidence	Validated molecular assays integrated with therapeutic efficacy, parasite clearance or phenotypic data	Treating candidate mutations as validated markers, absence of clinical correlation, inadequate temporal or spatial sampling	May justify enhanced surveillance and, when supported by population-level evidence, programme or treatment-policy review (Uwimana et al., 2020; Balikagala et al., 2021; Rosenthal et al., 2024; Henrici et al., 2025)

Abbreviations: MOI, multiplicity of infection; COI, complexity of infection; SNP, single-nucleotide polymorphism; WGS, whole-genome sequencing.

Note: Mixed-species infection, multiplicity of infection, and within-host resistance-associated variation are not equivalent categories. Each represents a different biological level and requires a different evidentiary basis for clinical or programme interpretation.

These distinctions guide the organisation of the remainder of the review. Epidemiological studies of mixed-species infection are evaluated primarily according to species coverage and diagnostic sensitivity, whereas studies of multiplicity are assessed according to marker resolution and mixture-inference methods. Resistance-associated findings are interpreted using a more stringent hierarchy that distinguishes an initial sequence signal from a confirmed minority variant, a validated resistance marker, evidence of clonal expansion, and clinically demonstrated changes in parasite clearance or therapeutic efficacy (Uwimana et al., 2020; Balikagala et al., 2021; Rosenthal et al., 2024). This analytical separation prevents prevalence estimates, genomic observations, and treatment implications from being combined as though they represented the same biological phenomenon.

4. Novelty and original contribution of this study

The principal novelty of this review is the integration of epidemiology, diagnostics, genomics, treatment, and elimination response into a single decision pathway. Previous work has established that mixed infections occur and may be clinically important, but prevalence alone is a weak endpoint for contemporary malaria programmes (Kotepui et al., 2020). In contrast, this paper argues that mixed infections should be positioned within a decision pathway that begins with detection, proceeds through species and clone resolution, incorporates genomic markers of resistance and diagnostic escape, and ends with treatment or public-health action. This framing is timely because malaria programmes are increasingly expected to integrate case management, molecular surveillance, vaccine implementation, and elimination planning under constrained budgets (WHO, 2022, 2025b).

The review contributes four original clarifications. By separating mixed-species infection from multiplicity of infection and mixed-resistance infection, the review clarifies why a single term cannot support all clinical and surveillance decisions. A P. falciparum + P. vivax infection raises radical-cure questions; a polyclonal P. falciparum infection raises transmission and recurrence-classification questions; and a minority resistance-associated pfk13 variant in a polyclonal infection raises early warning questions for artemisinin partial resistance. These are not interchangeable phenomena.

The second contribution is the integration of diagnostic escape into mixed-infection interpretation. In areas using HRP2-based RDTs, apparent non-P. falciparum or RDT-negative malaria cannot be interpreted safely without considering pfhrp2/pfhrp3 deletion, low parasite density, and assay design. WHO’s updated response plan and surveillance protocol emphasise the need to identify where false-negative RDTs due to pfhrp2/pfhrp3 deletions are sufficiently frequent to affect RDT policy (Poti et al., 2020; WHO, 2024a, 2024b). Mixed-infection studies that ignore diagnostic target failure may misclassify both species distribution and programme performance.

The third contribution is resistance-aware interpretation. Recent evidence from Rwanda and Uganda indicates that Africa can experience independent emergence and local spread of artemisinin partial-resistance-associated parasites (Uwimana et al., 2020; Balikagala et al., 2021; Rosenthal et al., 2024; Henrici et al., 2025). In polyclonal infections, resistance-associated variants may initially occur at minority frequencies. The review therefore argues that mixed-infection genomics should report allele-frequency thresholds, read-depth criteria, and the uncertainty around minority variant calls rather than treating sequencing as a black-box confirmation tool.

The fourth contribution is the proposed operational maturity lens. A malaria programme that can move from detection to characterisation and action is likely to have stronger laboratory networks, better sample systems, more informative therapeutic efficacy studies, and more responsive elimination surveillance. Mixed infection is therefore not only a biological observation; it is a stress test for malaria-system readiness.

The originality of the synthesis lies in linking mixed infections to actionable thresholds. A descriptive review might end by stating that better diagnostics are needed. This review goes further by asking which infections require molecular confirmation, which samples should be prioritised for sequencing, which genomic outputs should be reported, and how results should alter treatment or surveillance. This moves the manuscript from a catalogue of technologies to a policy-relevant framework for malaria control programmes.

5. Contemporary epidemiology of mixed malaria infections

The global distribution of mixed malaria infections reflects the overlap of parasite species, vector ecology, human mobility, immunity, diagnostic intensity, and treatment history. In sub-Saharan Africa, P. falciparum predominates, but P. malariae, P. ovale curtisi, P. ovale wallikeri, and increasingly recognised P. vivax infections occur in particular ecological and population contexts (Alum et al., 2024, 2025; White et al., 2025). In the Americas, the coexistence of P. falciparum and P. vivax is central to case management because P. vivax relapse biology requires radical cure (Price et al., 2020). In Southeast Asia and the Western Pacific, sympatric P. falciparum and P. vivax transmission, forest exposure, migrant labour, and residual foci make mixed infections especially important for elimination (Price et al., 2020; Yan et al., 2023). Recent genomic work in Guyana demonstrates that sympatric P. falciparum and P. vivax can exhibit contrasting population structures and transmission dynamics even when sampled within the same broad epidemiological setting (Schwabl et al., 2024). Co-endemicity therefore does not imply identical epidemiology, and mixed infections must be interpreted through species-specific biology.

Mixed-species infections are frequently underestimated. Microscopy can miss low-density secondary species, and many routine reporting systems record only the dominant species or the species targeted by the diagnostic platform. A systematic review and meta-analysis found that mixed Plasmodium infections can be associated with severe malaria and that mixed infections are often underrecognised by routine diagnostic methods (Kotepui et al., 2020). The clinical significance of this finding is not that all mixed infections are more severe than mono-infections; rather, it is that they cannot be dismissed as benign or incidental. The direction of association depends on parasite densities, host immunity, age, species combination, treatment history, and whether the outcome is parasitaemia, gametocyte carriage, anaemia, recrudescence, relapse, or mortality.

Within-host interactions are heterogeneous and may involve competition, facilitation, immune modulation, or shared exposure. Holzschuh et al. (2022) used molecular assays to examine co-infection among the four major human malaria species in Papua New Guinea and showed that heterogeneity in exposure to mosquito bites partly explained excess co-infection. They also reported species-specific interaction patterns, including altered P. malariae density in the presence of other species. These findings caution against overly simple assumptions that one parasite species always suppresses or protects against another. They also illustrate why molecular quantification, not merely categorical species detection, is necessary to infer biologically meaningful interactions.

Multiplicity of infection adds a second epidemiological layer. In high-transmission settings, P. falciparum infections are often polyclonal, and within-host diversity may reflect repeated inoculation, superinfection, long infection duration, or local connectivity among transmission foci (MalariaGEN et al., 2021). Complexity of infection can be used as a population-level proxy for transmission intensity, but it must be interpreted carefully because sampling depth, marker choice, parasite density, and prior treatment can change apparent diversity. In elimination settings, polyclonal infection may indicate importation from higher-transmission areas or persistent local transmission. Therefore, studies of mixed infections should report both species-level and clone-level complexity whenever possible.

Human mobility, climate variability, land-use change, and vector adaptation further complicate the epidemiology. Forest work, mining, agricultural expansion, displacement, and cross-border movement can expose individuals to multiple vector populations and parasite species (Msellemu et al., 2024). Climate-sensitive changes in vector distribution may alter the spatial overlap of species and the seasonality of mixed infections. These mechanisms are not unique to mixed malaria, but mixed infections may be particularly sensitive indicators of ecological transition because they require overlapping transmission opportunities. For this reason, mixed-infection surveillance should be integrated with geospatial risk mapping, travel history, occupational exposure, and vector surveillance rather than interpreted as a purely clinical laboratory finding.

6. Genomic characterisation: From detection to transmission intelligence

Genomic characterisation extends mixed-infection analysis from detection to inference about population structure, connectivity, resistance, diagnostic escape, and recurrence. Whole-genome sequencing offers the broadest view of parasite variation, and large resources such as MalariaGEN’s open P. falciparum dataset demonstrate how genome variation can support global malaria genomic epidemiology (MalariaGEN et al., 2021). For mixed infections, genomics is especially valuable because it can distinguish true mixed-species infection from assay cross-reactivity, identify minority clones, detect resistance mutations, and estimate whether parasites are locally acquired or imported (Aranda-Díaz et al., 2025). However, genomic interpretation in mixed infections is technically demanding. Low parasite density, high human DNA background, allele-frequency mixtures, copy-number variation, and uneven representation of co-infecting species can bias variant calling and haplotype reconstruction (MalariaGEN et al., 2021; De Cesare et al., 2024).

6.1. Matching molecular and genomic methods to the analytical question

The analytical question should be defined before a molecular or genomic technology is selected. Species confirmation, parasite quantification, multiplicity-of-infection estimation, minority resistance-variant detection, recurrence classification, and population-connectivity inference are distinct analytical tasks. They require different markers, quality controls, thresholds, sampling designs, and interpretation criteria. Consequently, a platform capable of generating extensive sequence data is not necessarily appropriate for every clinical or surveillance question.

Species confirmation requires validated species-specific targets and appropriate positive, negative, and contamination controls. The assay should include all locally plausible human-infecting Plasmodium species and should distinguish closely related taxa where this distinction affects epidemiological or clinical interpretation. Molecular methods generally provide greater sensitivity than microscopy for detecting low-density secondary species, but their performance remains dependent on target selection, DNA quality, parasite density, and laboratory quality assurance (Holzschuh et al., 2022; Mediavilla et al., 2024).

Parasite quantification requires a validated quantitative method, suitable reference standards, and clear reporting of the relationship between assay output and parasite density. Sequencing read proportions should not be interpreted automatically as precise estimates of relative parasite density because amplification efficiency, target copy number, hybrid-capture performance, and sequence-mapping efficiency may differ among species and loci (De Cesare et al., 2024; Aranda-Díaz et al., 2025).

Multiplicity or complexity of infection requires polymorphic markers capable of distinguishing genetically different parasite lineages and an explicit analytical model for resolving within-host mixtures. Estimates may vary according to marker diversity, parasite density, allele dropout, sequencing depth, within-sample allele-frequency thresholds, and the mixture-inference method used. Multiplicity estimates derived from different marker panels or analytical pipelines should therefore not be assumed to be directly comparable (MalariaGEN et al., 2021; De Cesare et al., 2024; Aranda-Díaz et al., 2025).

Minority-variant detection requires pre-specified reporting thresholds, adequate sequence depth, characterisation of platform-specific error, assessment of strand and read support, contamination control, and, where possible, replicate or orthogonal confirmation. The threshold used to identify a minority variant should reflect the sequencing platform, library-preparation procedure, amplification method, genomic locus, parasite density, and intended surveillance application (De Cesare et al., 2024; Aranda-Díaz et al., 2025). Detection of very low-frequency variants or low-density secondary species may require deep or ultra-deep sequencing, but increasing read depth does not eliminate amplification bias, contamination, allele dropout, or low-template stochastic error. Ultra-deep sequencing should therefore be used selectively and accompanied by defined controls, replicate testing, and pre-specified interpretation criteria.

Recurrence classification requires paired parasite samples obtained at enrolment and recurrence and the application of sufficiently informative genotyping markers. Interpretation must account for transmission intensity, marker diversity, missing genotype data, polyclonal infection, and the probability of reinfection. For relapsing species, particularly P. vivax and P. ovale, recurrence may additionally reflect activation of dormant liver-stage parasites, making the distinction among relapse, recrudescence, and reinfection particularly difficult (Walker et al., 2023; Yan et al., 2023).

Inference of population connectivity or parasite importation requires representative sampling across locations and time, informative genomic markers, accurate epidemiological metadata, and appropriate population-genetic analysis. The detection of multiple genotypes in a single host does not, by itself, establish parasite movement, local transmission connectivity, or importation. Such conclusions require comparison with parasite populations from plausible source and destination locations (MalariaGEN et al., 2021; Schwabl et al., 2024).

Whole-genome sequencing provides the broadest view of parasite diversity, but its cost, infrastructure requirements, and need for sufficient parasite DNA can limit routine use. Targeted amplicon sequencing and hybrid-capture panels offer more scalable alternatives for surveillance, particularly when the policy question concerns defined resistance markers, diagnostic targets, vaccine antigens, or barcode markers. De Cesare et al. (2024) showed that targeted nanopore sequencing from dried blood spots can generate policy-relevant information on drug resistance, diagnostic test failure, vaccine targets, and within-sample diversity. This is a major advance for mixed-infection surveillance because dried blood spots are already widely used in therapeutic efficacy studies and field surveys. Aranda-Díaz et al. (2025) reported a modular amplicon sequencing approach capable of generating data on diversity, drug and diagnostic resistance, CSP vaccine targets, and non-P. falciparum species detection. These platforms are not simply technical advances; they make it more plausible for surveillance programmes to combine therapeutic efficacy monitoring, pfhrp2/pfhrp3 surveillance, multiplicity-of-infection estimation, and vaccine-antigen monitoring within one sampling architecture.

A genomic surveillance framework for mixed malaria should include at least five analytic outputs. First, species confirmation should be performed using molecular or sequencing-based markers capable of resolving P. ovale curtisi and P. ovale wallikeri where relevant. Secondly, parasite density or read-depth proxies should be reported because minority species may be clinically relevant even at low density. Thirdly, multiplicity of infection should be estimated using standardised methods and reported separately from species co-infection. Fourthly, resistance markers should include validated or candidate loci appropriate to local drug policy, including validated and candidate mutations in pfk13 relevant to artemisinin partial resistance, pfcrt, pfmdr1, plasmepsin 2/3 copy number where piperaquine pressure is relevant, pfdhfr/pfdhps for antifolate resistance. For P. vivax, surveillance panels may include pvmdr1 and pvcrt-o as putative markers investigated in relation to chloroquine response, and pvdhfr and pvdhps for antifolate resistance. However, the genotype-phenotype relationship for chloroquine resistance remains less firmly established than for several major P. falciparum markers. These loci should therefore be interpreted alongside therapeutic efficacy and clinical recurrence data rather than as definitive stand-alone predictors (Noisang et al., 2020). Fifthly, diagnostic target loci, particularly pfhrp2 and pfhrp3, should be assessed in settings where HRP2-based RDTs are used (Poti et al., 2020; WHO, 2024a, 2024b).

Technical feasibility does not ensure routine programmatic usability. Sequencing workflows require validated pipelines for demultiplexing, contamination detection, mapping, variant calling, copy-number analysis, minority-allele classification, and quality reporting. These functions may require specialist bioinformatic expertise, computational infrastructure, software maintenance, and reproducible version control. Transfer to national malaria programmes is further complicated when analytical pipelines were developed for research datasets rather than routine samples with low parasitaemia, variable DNA quality, incomplete metadata, and irregular batch sizes. Public-health implementation therefore requires simplified and validated analytical workflows, transparent quality metrics, local data-governance arrangements, staff training, external quality assurance, and mechanisms for translating genomic outputs into predefined programme recommendations. Research on integrated malaria molecular surveillance similarly emphasises that genomic data must be contextualised and linked to national programme priorities rather than generated as an isolated laboratory output (Dada et al., 2024).

6.2. Interpretation of minority resistance-associated variants

The emergence and local spread of artemisinin partial resistance in parts of Africa increase the importance of detecting resistance-associated parasite lineages before they become widely established. Evidence from Rwanda and Uganda demonstrates that resistance interpretation is strongest when molecular findings are integrated with parasite-clearance measurements, therapeutic efficacy observations, phenotypic evidence, temporal trends, and population-genetic analysis (Uwimana et al., 2020; Balikagala et al., 2021; Rosenthal et al., 2024; Henrici et al., 2025).

A low-frequency sequence call should not automatically be interpreted as evidence of a clinically meaningful resistant parasite population. Such a call is initially an analytical observation that requires evaluation of base quality, sequence depth, strand representation, mapping quality, amplification bias, contamination, and platform-specific error. When adequate quality criteria are met, and the result is reproducible, it may be described as a technically confirmed within-host minority variant. Its resistance significance must then be evaluated according to the validation status of the specific mutation and the relevant clinical and epidemiological evidence.

The following distinctions should therefore be maintained: an initial low-frequency sequence call; a technically confirmed within-host minority variant; a mutation recognised as a candidate resistance marker; a validated resistance-associated marker; repeated detection of the same marker across patients, places, or time; evidence of local clonal expansion; and clinical or parasitological evidence of altered parasite clearance or reduced therapeutic efficacy. These evidence levels are related but not interchangeable. A technically credible minority pfk13 variant may justify confirmatory testing, enhanced sampling, and linkage with therapeutic efficacy surveillance, but it does not independently demonstrate treatment failure in the individual patient or justify immediate modification of national treatment policy (Rosenthal et al., 2024).

Particular caution is required when candidate pfk13 mutations are reported. Not every nonsynonymous alteration in pfk13 has the same degree of evidence supporting an association with artemisinin partial resistance. Interpretation should therefore specify the mutation, within-sample allele frequency, sequence depth, analytical threshold, confirmation method, marker-classification status, geographical setting, treatment history, and any available parasite-clearance or therapeutic efficacy data (Uwimana et al., 2020; Balikagala et al., 2021; Angwe et al., 2024; Rosenthal et al., 2024).

Evidence of clonal expansion requires more than repeated identification of the same mutation. It generally requires assessment of the surrounding haplotype, genetic relatedness among parasites, spatial and temporal distribution, and increasing frequency in the sampled population. Similarly, clinical evidence of artemisinin partial resistance should be distinguished from partner-drug failure and from complete treatment failure because delayed parasite clearance does not necessarily imply failure of an effective artemisinin-based combination therapy (Balikagala et al., 2021; Rosenthal et al., 2024). Accordingly, this review uses the term “within-host resistance-associated variation” rather than treating “mixed-resistance infection” as a formal clinical category. The term describes an operational molecular observation whose significance must be established through staged technical, biological, clinical, and epidemiological evaluation.

Published amplicon-sequencing platforms demonstrate that minority alleles may be detected at within-sample frequencies near 1% under favourable conditions, particularly in high-density samples with adequate depth and validated controls. However, a 1% analytical detection limit should not be treated automatically as a programme reporting threshold. False-positive minority calls may still occur near this range, especially in low-parasitaemia samples, and performance varies by platform, locus, amplification design, sequencing depth, replicate concordance, and bioinformatic filtering. Some surveillance applications may therefore adopt a more conservative threshold, such as 5%, unless lower-frequency calls are supported by replicate testing or an orthogonal method. The appropriate threshold should be established during assay validation using negative controls, defined parasite mixtures, replicate samples, locus-specific error estimates, and clinically relevant reporting objectives (Aranda-Díaz et al., 2025).

To support consistent interpretation and prevent low-frequency molecular signals from being overclassified as clinically meaningful resistance, Table 2 presents a staged evidence hierarchy distinguishing an uncertain sequence call from a confirmed minority variant, a validated resistance marker, evidence of clonal expansion, and clinically significant resistance requiring programme-level response (Uwimana et al., 2020; Balikagala et al., 2021; Rosenthal et al., 2024; Henrici et al., 2025).

Table 2.

Evidence hierarchy for interpreting resistance-associated variants in mixed malaria infections.

Evidence level	Interpretation	Appropriate response
Low-frequency sequence call without adequate quality confirmation	Uncertain analytical signal	Review sequence quality, contamination, depth and analytical threshold
Reproducible minority allele with adequate read support	Confirmed within-host minority variant	Report allele frequency and uncertainty; consider orthogonal confirmation
Confirmed variant at a candidate resistance marker	Possible molecular resistance signal	Increase sampling and interpret with caution
Confirmed variant at a validated resistance marker	Credible resistance-associated molecular signal	Link with treatment, clearance and therapeutic efficacy data
Repeated detection in several patients or locations	Possible emergence or spread	Intensify molecular and therapeutic efficacy surveillance
Shared surrounding haplotype and increasing frequency	Evidence consistent with clonal expansion	Conduct population-genomic and epidemiological investigation
Validated marker with delayed clearance or reduced efficacy	Clinically and epidemiologically important resistance evidence	Escalate to national programme assessment and resistance-response planning

Notes: A resistance-associated allele should not be interpreted solely according to its presence or absence. Sequence quality, within-host frequency, marker-validation status, clinical phenotype, therapeutic efficacy, spatial distribution, and evidence of clonal expansion should be considered together (Uwimana et al., 2020; Balikagala et al., 2021; Rosenthal et al., 2024; Henrici et al., 2025).

7. Diagnostics: Critical appraisal and operational implications

Diagnostic choice should be matched with the decision being made. For immediate case management, WHO-recommended microscopy and RDTs remain indispensable because they support rapid treatment in peripheral settings. However, their limitations must be acknowledged. Microscopy remains useful where well-trained microscopists, quality assurance, and adequate parasite densities are present, but it is weak for low-density mixed infections and for distinguishing morphologically similar species. RDTs are operationally essential because they enable rapid treatment in peripheral settings, yet most are not designed to quantify parasite density or resolve multiple species reliably. HRP2-based RDTs are particularly vulnerable to pfhrp2/pfhrp3 deletions, which WHO identifies as a biological threat because deleted parasites can evade the most commonly used P. falciparum RDTs (Poti et al., 2020; WHO, 2024b).

An HRP2-based rapid diagnostic test-negative result should not be equated directly with pfhrp2 or pfhrp3 deletion. Discordant diagnostic findings can arise from low parasite density, incorrect testing procedure, poor test storage, test deterioration, prozone effects at very high parasitaemia, previous antimalarial treatment, errors in microscopy, limitations in the comparator assay, or differences in antigen production and persistence. A negative HRP2-based RDT result accompanied by independent evidence of P. falciparum infection should therefore be classified initially as a suspected false-negative RDT result rather than as confirmed gene deletion (Poti et al., 2020; WHO, 2024a, 2024b).

Investigation should begin by confirming the presence of P. falciparum using expert microscopy, an alternative antigen target, or a validated molecular assay. Parasite density should be considered because low-density infections may produce insufficient circulating antigen for reliable RDT detection. The quality, storage conditions, expiry status, lot performance, and use procedure of the RDT should also be reviewed before a parasite genetic explanation is assigned (Poti et al., 2020; WHO, 2024a).

Failure to amplify pfhrp2 or pfhrp3 does not, by itself, confirm deletion. Amplification failure may result from inadequate parasite DNA, low parasite density, DNA degradation, primer-binding variation, PCR inhibition, or laboratory error. Adequate parasite DNA should therefore be demonstrated through successful amplification of appropriate single-copy control loci before absence of the diagnostic-target gene is interpreted as deletion. Relevant exons and, where required by the validated protocol, flanking regions should be assessed using appropriate positive and negative controls. Repeated testing or confirmation using an alternative molecular approach should be considered where results are uncertain (WHO, 2024a). The revised framework distinguishes three evidentiary levels:

• •Suspected diagnostic escape refers to an HRP2-based RDT-negative result in a sample for which independent evidence supports P. falciparum infection.
• •Laboratory-confirmed deletion refers to reproducible evidence that the relevant pfhrp2 or pfhrp3 target is absent in a sample with adequate parasite DNA and satisfactory control-target amplification.
• •Population-level evidence relevant to policy refers to standardised and representative surveillance demonstrating that false-negative HRP2-based RDT results attributable to gene deletions occur sufficiently frequently to compromise diagnostic performance in the target population.

These distinctions are important because the required action differs at each level. A single discordant result warrants diagnostic confirmation and appropriate patient treatment. A laboratory-confirmed deletion should be reported and may justify enhanced surveillance. However, modification of national RDT policy requires systematic population-level evidence collected according to standardised surveillance methods and interpreted with national and WHO guidance (WHO, 2024a, 2024b).

The principal programme-relevant outcome is not merely the prevalence of pfhrp2 or pfhrp3 deletion among molecularly tested samples. It is the proportion of symptomatic, confirmed P. falciparum infections that would generate false-negative HRP2-based RDT results because of deletion. Samples selected only because they were RDT-negative may overestimate the population-level frequency of diagnostic escape and should not be used alone to infer the need for an RDT-policy change (WHO, 2024a, 2024b).

Molecular assays provide greater sensitivity and species resolution but are not a single solution. Conventional PCR, real-time PCR, reverse-transcription PCR, loop-mediated isothermal amplification, and multiplex bead-based assays each have strengths and limitations. PCR and qPCR are valuable for reference laboratories and surveys, but the risk of contamination, the need for cold-chain reagents, and the requirement for trained personnel can limit routine implementation (Mediavilla et al., 2024). LAMP can be more field-adaptable but may still require careful species-panel design. Multiplex platforms can detect parasite antigens and host serological markers simultaneously, which is useful for surveillance but may not be feasible for immediate case management. Sequencing-based methods can resolve species, clones, and resistance markers but are best deployed as sentinel or reference tools rather than universal frontline diagnostics (De Cesare et al., 2024; WHO, 2025b). The diagnostic hierarchy should therefore be tiered, not technology maximalist.

A practical tiered algorithm is proposed. At the point of care, programmes should use WHO-recommended RDTs or microscopy according to local policy and should treat immediately when malaria is confirmed. At district or regional level, recurrent infections, severe cases, unexpected RDT-negative microscopy-positive cases, suspected mixed infections, and therapeutic efficacy study samples should undergo molecular confirmation. At national reference level, a subset of molecularly confirmed cases should be sequenced for resistance markers, pfhrp2/pfhrp3 status, barcode diversity, and species confirmation. This tiered approach balances feasibility with scientific rigour and prevents the unrealistic expectation that every peripheral facility must perform sequencing.

Diagnostic reporting should also be improved. A result labelled simply as “mixed malaria” is insufficient for clinical or policy action. Reports should specify the species detected, the method used, whether the finding was qualitative or quantitative, the approximate density or cycle-threshold value when available, and whether the result has treatment implications such as need for primaquine or tafenoquine eligibility assessment for P. vivax radical cure. Research reports should disclose limits of detection, primer targets, cross-reactivity testing, sample storage conditions, and whether assays detect gametocyte stages. Such reporting standards would make prevalence estimates more comparable and reduce the risk of overclaiming novelty from poorly characterised mixed infections. Without this detail, mixed-infection prevalence estimates are difficult to compare and may not be clinically actionable.

Implementation also requires an explicit specimen and information pathway. Programmes must define who identifies eligible samples, obtains consent where required, prepares and stores dried blood spots or whole-blood specimens, completes clinical and epidemiological metadata, arranges transport, monitors turnaround time, receives results, and initiates follow-up. Although dried blood spots can reduce some cold-chain requirements, other reagents and confirmatory workflows may still depend on controlled storage and reliable procurement. Improved reporting of cycle-threshold values, assay targets, limits of detection, and quality controls will require retraining of laboratory personnel, clinicians, surveillance officers, and health-information teams. Data fields must be standardised and incorporated into existing reporting systems so that technically detailed results do not remain disconnected from case investigation and programme action. Targeted or whole-genome sequencing should generally be deployed through sentinel or reference systems rather than as universal frontline testing. Dried blood spots may reduce some transport and storage requirements, but sequencing workflows still depend on reliable procurement, equipment maintenance, laboratory quality assurance, trained personnel, and appropriate reagent storage. Programmes should therefore assess the full specimen-to-decision pathway rather than considering sequencing cost in isolation.

8. Treatment implications and drug resistance

Treatment of mixed malaria infections must address all clinically relevant species and life stages. For uncomplicated P. falciparum malaria, WHO-recommended artemisinin-based combination therapies remain the mainstay of treatment, but their continued effectiveness depends on surveillance for artemisinin partial resistance and partner-drug resistance (WHO, 2022, 2025b). In confirmed or suspected P. vivax or P. ovale infections, blood-stage treatment alone is insufficient because hypnozoites can cause relapse (Yan et al., 2023). Radical cure requires an 8-aminoquinoline, such as primaquine or tafenoquine, after appropriate assessment for glucose-6-phosphate dehydrogenase deficiency. A mixed P. falciparum + P. vivax infection that is diagnosed only as P. falciparum malaria may therefore be treated for the acute episode but remain vulnerable to relapse.

Mixed infections can also complicate therapeutic efficacy interpretation. In polyclonal P. falciparum infection, treatment can clear sensitive clones while resistant or tolerant clones persist or recrudesce. Without genotyping, recurrent parasitaemia may be misclassified as reinfection or recrudescence. In co-endemic P. falciparum + P. vivax settings, a later P. vivax episode may represent relapse rather than a new infection. Walker et al. (2023) modelled malaria treatment evaluation in the presence of multiple species and showed why mixed-species treatment strategies and radical cure can alter future incidence, although model outputs depend on assumptions about glucose-6-phosphate dehydrogenase (G6PD) testing, adherence, relapse, and coverage. In co-endemic areas, efficacy studies should therefore collect dried blood spots at enrolment and recurrence, apply standardised genotyping, and report mixed-species and multiplicity patterns.

The resistance implications are increasingly important in Africa. WHO’s strategy to respond to antimalarial drug resistance in Africa emphasises improved detection, delayed emergence, and containment of resistant parasites where resistance is confirmed (WHO, 2022). Polyclonal infections are relevant to all three objectives. They can obscure early resistance detection if minority variants are missed; they can increase opportunities for recombination if resistant parasites enter high-transmission settings; and they can complicate containment because resistant and sensitive lineages may coexist within mobile populations. Genomic surveillance linked to therapeutic efficacy data is therefore essential (Balikagala et al., 2021; Angwe et al., 2024; Rosenthal et al., 2024; Henrici et al., 2025).

Drug choice in mixed infections should therefore be guided by species composition, severity, patient factors, national policy, and local resistance patterns. Severe malaria should be treated promptly according to WHO guidelines, regardless of mixed status, because delays increase mortality. For uncomplicated disease, ACT choice should follow national policy but should be informed by therapeutic efficacy studies and molecular surveillance. Where P. vivax or P. ovale is detected, radical cure should be considered after G6PD testing. Where P. malariae is detected, clinicians should recognise that it may persist at low density and contribute to chronic infection or anaemia. Where zoonotic species such as P. knowlesi is plausible, particularly in Southeast Asia, rapid clinical progression must be considered, although detailed zoonotic malaria is outside the primary scope of this review.

Elimination-oriented treatment should distinguish three therapeutic objectives: clearance of asexual blood-stage infection; prevention of relapse from dormant liver stages; and reduction of onward transmission. Blood-stage treatment is required for all clinically relevant species. Anti-relapse treatment, or radical cure, is relevant to confirmed or suspected P. vivax and P. ovale infection because these species form hypnozoites. Primaquine or tafenoquine use requires assessment of G6PD activity according to the selected regimen and national guidance, because therapeutic 8-aminoquinoline exposure can cause clinically important haemolysis in individuals with G6PD deficiency. WHO’s updated guidance includes near-patient qualitative and semiquantitative G6PD testing to support anti-relapse treatment decisions (WHO, 2025b).

Transmission-reducing treatment is a separate objective. In low-transmission settings and areas affected by artemisinin resistance, WHO recommends adding a single low dose of primaquine to effective ACT for P. falciparum to reduce transmissibility, subject to national policy and patient-specific contraindications. This gametocytocidal use should not be described as radical cure because P. falciparum lacks hypnozoites (WHO, 2025b).

Mixed P. falciparum + P. vivax or P. falciparum + P. ovale infections may therefore require consideration of blood-stage therapy, species-appropriate anti-relapse treatment, and, where recommended, transmission-reducing therapy. These interventions are not universally interchangeable and should be selected according to species, G6PD status, epidemiological setting, national policy, contraindications, and the evidence supporting transmission benefit.

Species interactions may complicate interpretation of intervention effects. Apparent increases in P. vivax following reductions in P. falciparum may reflect relapse biology, altered treatment exposure, improved diagnostic sensitivity, ecological change, or release from within-host competition. Current evidence does not support the assumption of a universal suppressive effect of P. falciparum on P. vivax; longitudinal studies with species-specific quantification are required before causal attribution (Holzschuh et al., 2022).

Primaquine and tafenoquine are used for relapsing malaria by targeting P. vivax and P. ovale hypnozoites, subject to appropriate G6PD assessment and national guidance. Primaquine also has a separate gametocytocidal role in P. falciparum malaria. Where recommended, a single low dose may be added to effective blood-stage therapy to reduce onward transmission. These uses differ in purpose, dose, duration, and safety requirements and should not be described interchangeably.

9. Public health and elimination implications

Mixed infections have implications beyond the individual patient. They can sustain transmission if one species is missed and untreated, if gametocyte carriage persists, or if relapse biology is ignored. They can also bias surveillance indicators. A programme that relies heavily on HRP2-based RDTs may underestimate P. falciparum in areas with pfhrp2/pfhrp3 deletions and overestimate the proportional burden of non-P. falciparum malaria. Conversely, a programme that does not test for non-P. falciparum species may overattribute fever to P. falciparum and miss radical-cure opportunities. Both errors weaken elimination planning (Poti et al., 2020; WHO, 2024a, 2024b).

A further consideration is whether expanded malaria-vaccine implementation may alter the epidemiological patterns captured by species-specific and genomic surveillance. Malaria-vaccine implementation is relevant to this review primarily where surveillance platforms are used to examine changes in P. falciparum infection patterns or variation in vaccine-target antigens alongside resistance and diagnostic markers. Mixed-species surveillance may also help determine whether the proportional contribution of non-P. falciparum species changes as P. falciparum disease is reduced. However, evidence that vaccine implementation directly alters mixed-infection epidemiology remains limited. Vaccine monitoring is therefore considered a potential application of integrated surveillance rather than a central component of the Detect-Characterise-Act framework (De Cesare et al., 2024; WHO, 2024d).

For elimination programmes, mixed infections should be linked to response systems. A confirmed mixed infection in a low-transmission district may trigger household screening, travel-history investigation, vector assessment, or focal response. In border regions, mixed infection may indicate importation from areas with different species ecology or resistance profiles. In humanitarian settings, where disrupted health systems and population displacement increase malaria vulnerability, mixed infections may signal diagnostic gaps and treatment interruptions. These public-health uses require timely reporting and interoperable data systems.

Capacity building is therefore not optional. Molecular and genomic surveillance must be embedded in endemic-country institutions rather than outsourced indefinitely. Training should cover sample collection, dried blood spot storage, DNA extraction, assay interpretation, sequencing quality control, bioinformatics, ethical data sharing, and translation of results into policy. Regional reference laboratories can support quality assurance, but national programmes need sufficient in-country capacity to interpret findings quickly. Targeted sequencing has reduced barriers; the remaining challenge is sustainable implementation, not proof of technological possibility (De Cesare et al., 2024).

The operational value of molecular and genomic surveillance depends not only on analytical performance but also on affordability, logistical feasibility, workforce capacity, and the probability that the resulting information will change policy or practice. Capital equipment, maintenance, reliable electricity, reagent procurement, sample transport, cold-chain requirements, quality assurance, data storage, and bioinformatic support can make routine molecular confirmation substantially more expensive than frontline microscopy or rapid diagnostic testing. These constraints are particularly important in remote endemic settings, where specimen transport and equipment downtime may extend turnaround times beyond the period in which results can influence individual case management.

Cost-effectiveness should therefore be assessed at the level of the surveillance strategy rather than the sequencing platform alone. Relevant comparisons include routine confirmation of all malaria-positive cases, targeted testing of diagnostically discordant or recurrent cases, sequencing of therapeutic efficacy study samples, periodic sentinel sampling, and focused testing in border or elimination settings. Outcomes should include the cost per actionable resistance signal detected, diagnostic-policy change appropriately informed, treatment failure averted, imported focus investigated, or outbreak prevented. Although targeted nanopore and amplicon approaches may reduce per-sample sequencing and infrastructure requirements, their economic value remains dependent on sample volume, batching, transport, staff time, repeat-testing rates, data-analysis capacity, and whether results are linked to predefined programme decisions (De Cesare et al., 2024). Recent work demonstrates that targeted nanopore surveillance can reduce some platform and infrastructure barriers, but these technical and platform-specific findings should not be interpreted as evidence of cost-effectiveness across health systems (De Cesare et al., 2024).

Programmes should consequently adopt a staged implementation model. Initial use may focus on clearly defined sentinel applications, supported by regional reference laboratories and standardised bioinformatic pipelines. Scale-up should occur only after evaluation of analytical performance, turnaround time, recurrent costs, representativeness, workforce requirements, data use, and incremental public-health benefit relative to existing surveillance. Economic evaluation should be planned prospectively alongside implementation rather than added after a genomic platform has already been selected.

Routine health-facility sampling is convenient but may not represent the wider parasite population. It disproportionately captures symptomatic individuals who seek care and can underrepresent asymptomatic, sub-patent, geographically remote, mobile, or privately treated infections. The extent of this bias varies by transmission intensity, healthcare access, age, immunity, diagnostic practices, and treatment-seeking behaviour. Surveillance objectives should therefore determine the sampling frame. Facility-based sampling may be appropriate for therapeutic efficacy or clinical resistance monitoring, whereas community surveys, reactive case detection, border-site sampling, or targeted occupational surveillance may be required for reservoir estimation, importation analysis, or elimination planning. Sample size and site selection should be based on expected prevalence, spatial heterogeneity, desired precision, and cluster design rather than laboratory convenience alone (Mayor et al., 2023).

10. Proposed integrated framework

To translate these analytical distinctions into proportionate clinical and public-health responses, this review proposes an integrated Detect-Characterise-Act framework by linking common diagnostic or surveillance triggers with the required level of characterisation, appropriate individual-level action, and the additional evidence needed before programme-level intervention is considered (summarised in Table 3). The Detect phase uses appropriate frontline diagnostics and flags cases requiring confirmatory testing: severe disease, treatment failure, recurrent parasitaemia, discordant RDT and microscopy results, suspected non-P. falciparum infection, and cases from elimination or border settings. The Characterise phase applies molecular or sequencing tools to resolve species, quantify density when possible, estimate multiplicity of infection, identify resistance markers, and assess diagnostic-target integrity. The Act phase translates findings into clinical treatment, RDT policy, therapeutic efficacy monitoring, focal response, or cross-border surveillance. The framework is intentionally tiered so that advanced tools are used where they add decision value rather than as universal requirements. For mixed P. falciparum + P. vivax or P. falciparum + P. ovale infection, provide species-appropriate blood-stage treatment, provide primaquine or tafenoquine where the patient is eligible and G6PD status permits, and consider transmission-reducing single low-dose primaquine for P. falciparum where recommended by national policy. The framework distinguishes an analytical finding from a verified biological finding and from evidence sufficient to support programme action. A low-frequency resistance-associated sequence call first requires technical verification before it can be described as a within-host minority variant. A confirmed minority variant must then be interpreted according to the validation status of the marker and the available clinical and epidemiological evidence. Similarly, an HRP2-RDT-negative result is initially a discordant diagnostic finding, not confirmation of pfhrp2/pfhrp3 deletion. Diagnostic-policy review requires representative population-level evidence rather than an isolated patient result. This staged interpretation reduces the risk of converting analytical uncertainty into inappropriate clinical or policy action (Poti et al., 2020; Rosenthal et al., 2024; WHO, 2024a, 2024b).

Table 3.

Operational Detect-Characterise-Act framework for mixed malaria infections.

Stage	Trigger	Characterisation required	Individual-level action	Programme-level action
Detect	Positive RDT or microscopy result	Record species reported, test type, severity, previous treatment, travel and epidemiological context	Treat promptly according to national guidance	Include in routine surveillance
Detect	Suspected mixed-species infection or unclear species identification	Species-specific PCR/qPCR or validated multiplex assay where available	Ensure treatment covers confirmed clinically relevant species; assess radical-cure eligibility where applicable	Monitor frequency and geographical distribution of confirmed mixed infections
Detect	HRP2-RDT negative but P. falciparum suspected or independently detected	Repeat/alternative diagnostic testing, parasite-density assessment and validated deletion investigation if indicated	Treat confirmed malaria according to clinical guidance	Aggregate confirmed cases; initiate standardised deletion surveillance if recurrent or clustered
Detect	Recurrent parasitaemia or suspected treatment failure	Paired enrolment and recurrence genotyping; resistance-marker testing; adherence and drug-quality assessment where appropriate	Manage recurrence according to national protocol	Strengthen therapeutic efficacy surveillance when recurrent patterns or validated markers emerge
Characterise	Confirmed mixed-species infection	Quantitative or species-resolving molecular testing where relevant	Species-appropriate treatment and follow-up	Review diagnostic coverage and species-specific reporting
Characterise	Polyclonal infection requiring epidemiological interpretation	MOI/COI estimation with explicit markers and analytical model	Usually no change in immediate treatment solely because infection is polyclonal	Interpret with transmission intensity, recurrence and connectivity data
Characterise	Low-frequency resistance-associated sequence call	Technical verification, replicate or orthogonal confirmation, validated marker classification	Do not infer treatment failure from the molecular result alone	Intensify sampling and link with therapeutic efficacy data
Characterise	Confirmed validated resistance marker in repeated cases	Temporal, spatial and haplotype assessment plus clinical efficacy data	Continue guideline-concordant care unless national guidance changes	Enhanced TES and molecular surveillance; assess evidence of local emergence or spread
Characterise	Confirmed molecular case selected through TES, recurrent parasitaemia, unusual resistance signal, diagnostic-target investigation, or elimination/importation inquiry	Targeted sequencing for predefined resistance, diagnostic-target, species, or barcode loci; WGS only where broader population structure, relatedness, novel variants, or source attribution is required	Usually no immediate treatment change from sequencing alone	Integrate with clinical and epidemiological evidence for resistance, connectivity, or importation assessment
Act	Confirmed infection in low-transmission or elimination setting	Case classification, travel history, household or focal investigation as locally indicated	Complete treatment and follow-up	Investigate importation, local transmission and potential focal response
Act	Cases linked to border crossing, displacement or mobile occupations	Travel, occupational and geospatial characterisation	Ensure continuity of diagnosis and treatment	Cross-border data sharing and coordinated surveillance when clusters or shared transmission routes are supported
Act	Confirmed P. falciparum infection in a setting where transmission-reducing treatment is nationally recommended	Review contraindications and national eligibility criteria	Add single low-dose primaquine to effective ACT where recommended; do not describe this as radical cure	Monitor implementation, safety, coverage, and transmission objectives
Act	Representative evidence that deletions cause clinically important false-negative HRP2-RDT results	Standardised population-level surveillance	Use available alternative diagnosis for affected patients	National diagnostic-policy review according to WHO and local evidence
Act	Validated markers plus declining clinical efficacy or clonal expansion	Integrated molecular, clinical and epidemiological assessment	Follow updated national treatment guidance	National or regional resistance response and potential treatment-policy evaluation

Abbreviations: RDT, rapid diagnostic test; qPCR, quantitative polymerase chain reaction; MOI, multiplicity of infection; COI, complexity of infection; TES, therapeutic efficacy study; G6PD, glucose-6-phosphate dehydrogenase; WGS, whole genome sequencing.

Notes: No universal sequencing, allele-frequency, or programme-escalation threshold is applicable across all settings; thresholds should be pre-specified according to assay performance, transmission setting, surveillance objective, and current national or WHO guidance. Individual-level findings should not be interpreted automatically as sufficient grounds for programme-level policy change. A discordant RDT result, minority resistance-associated variant, or single confirmed gene deletion may justify verification, reporting, or enhanced surveillance, whereas revision of diagnostic or treatment policy requires systematic and representative population-level evidence interpreted in accordance with national and WHO guidance (Poti et al., 2020; Rosenthal et al., 2024; WHO, 2024a, 2024b).

The Detect-Characterise-Act pathway is summarised in Fig. 1 by linking five common mixed-malaria scenarios with the minimum confirmatory steps and the corresponding individual- and programme-level responses. The framework has practical implications. A febrile child in a high-transmission African district with P. falciparum detected by RDT may not need sequencing for routine care, but samples from periodic sentinel surveys should be genotyped to monitor validated or candidate pfk13 mutations, pfhrp2/pfhrp3, and parasite diversity. A miner returning from a forested co-endemic area with P. falciparum + P. vivax infection should receive treatment that addresses P. falciparum malaria and, if eligible, P. vivax radical cure. A microscopy-positive, HRP2-RDT-negative result should first trigger confirmation of P. falciparum, review of parasite density and RDT performance, and, where sample quality is adequate, investigation of possible pfhrp2/pfhrp3 deletion using validated controls. A low-transmission district reporting mixed infections after months of zero cases should investigate importation and residual foci. In each scenario, mixed infection becomes an actionable signal rather than a descriptive label.

Fig. 1.

Detect-Characterise-Act pathway for mixed malaria infections. The figure presents five operational scenarios: (1) mixed-species infection; (2) within-species multiplicity; (3) resistance-associated variation; (4) suspected diagnostic-target failure; and (5) importation or cross-border infection. Blue panels indicate detection, orange panels indicate characterisation, and green panels indicate action. The pathway emphasises that molecular or genomic testing should be selected according to the analytical question and that individual findings should not automatically trigger programme-level policy change. Species-specific treatment and radical cure should follow national guidelines and appropriate G6PD assessment. A minority resistance-associated variant, a single discordant HRP2-RDT result, or one confirmed pfhrp2/pfhrp3 deletion may justify verification, reporting, or enhanced surveillance, but diagnostic or treatment-policy changes require representative population-level evidence interpreted with clinical, epidemiological, and molecular data (Poti et al., 2020; Rosenthal et al., 2024; WHO, 2024a, 2024b). Abbreviations: AF, allele frequency; COI, complexity of infection; G6PD, glucose-6-phosphate dehydrogenase; HRP2, histidine-rich protein 2; MOI, multiplicity of infection; PCR, polymerase chain reaction; Pf, Plasmodium falciparum; Po, Plasmodium ovale; PQ, primaquine; Pv, Plasmodium vivax; qPCR, quantitative polymerase chain reaction; RDT, rapid diagnostic test; Rx, treatment; TQ, tafenoquine; WGS, whole-genome sequencing.

Translation from genomic evidence to national policy requires more than an analytically valid result. Decision-makers must consider representativeness, precision, replication across sites, clinical consequences, feasibility, equity, budget impact, supply-chain requirements, workforce readiness, stakeholder acceptability, and the availability of alternative interventions. Programme action may therefore require integration of genomic evidence with diagnostic performance, therapeutic efficacy, epidemiological, operational, and economic data.

11. Research gaps and priorities

Several gaps limit progress. First, prevalence estimates remain difficult to compare because studies use different definitions of mixed infection, different assays, and different sampling populations. Future studies should explicitly distinguish mixed-species infection, multiplicity of infection, and minority resistance variants. Secondly, few studies integrate clinical outcomes, parasite density, gametocyte carriage, resistance markers, and treatment follow-up in the same cohort. Thirdly, non-P. falciparum species remain underrepresented in genomic surveillance, despite their importance for relapse, chronic infection, and elimination. Fourthly, most studies are cross-sectional, limiting inference about whether species interactions alter duration of infection, onward transmission, or treatment response (Kotepui et al., 2020; Holzschuh et al., 2022).

A high-impact research agenda should include prospective multicountry cohorts in co-endemic settings. These studies should enrol symptomatic and asymptomatic participants, use standardised molecular panels, quantify parasite density and gametocyte carriage, sequence resistance and diagnostic-target loci, and follow patients after treatment. Where full multicountry prospective cohorts are not feasible, programmes may begin with harmonised sentinel studies, nested sampling within therapeutic efficacy studies, standardised minimum datasets, shared molecular panels, and regional quality-assurance networks. This staged approach can improve comparability while reducing the initial logistical and financial burden. In P. vivax-endemic settings, follow-up must include radical-cure eligibility, G6PD status, and relapse monitoring. In high-transmission African settings, studies should link polyclonal infection to therapeutic efficacy and surveillance of validated and candidate pfk13 mutations. In elimination settings, studies should combine genomics with travel history and geospatial mapping to distinguish importation from residual transmission.

Reporting standards should be strengthened. Study reports should state whether the term mixed infection refers to species, clones, or resistance genotypes. Diagnostic methods should be described with target genes, limits of detection, quality controls, and sample handling procedures. Genomic studies should report read depth, human DNA contamination, within-sample diversity thresholds, haplotype reconstruction methods, and data-sharing arrangements. Clinical studies should report treatment regimen, adherence assessment, recurrence classification, and whether radical cure was given when appropriate. These improvements would make the literature more reproducible and more useful to policy.

Economic evaluation is another priority. Sequencing every malaria-positive case is neither necessary nor affordable in most endemic settings. The research question should be which sentinel designs provide the greatest policy value per dollar, and which triggers justify sequencing. Potential designs include sequencing all therapeutic efficacy study samples, sequencing RDT-negative microscopy-positive samples, targeted sequencing in border districts, or periodic sampling in vaccine implementation areas. Cost-effectiveness models should include not only laboratory costs but also costs avoided through earlier detection of resistance, appropriate RDT switching, better radical-cure targeting, and prevention of outbreaks.

Transition from routine case-counting to molecular or genomic surveillance requires institutional change rather than laboratory procurement alone. Implementation may require retraining of laboratory personnel, clinicians, epidemiologists, surveillance officers, data managers, and policy teams, revision of reporting forms and electronic information systems, development of referral and confirmatory-testing pathways, and agreement on who is responsible for interpreting and acting on results. A phased sentinel approach is therefore more realistic than immediate nationwide implementation.

12. Study limitations

This critical narrative review is subject to selection bias because study inclusion involved relevance-based judgement rather than a fully systematic screening process. Publication bias may also have favoured studies reporting unusual mixed infections, successful molecular detection, or significant genomic findings. Although several databases and authoritative sources were searched, relevant studies indexed only in regional databases or published in languages outside the eligibility criteria may have been missed. No formal study-level risk-of-bias tool was used because the included evidence comprised heterogeneous diagnostic, genomic, epidemiological, modelling, and policy sources. Instead, studies were appraised according to methodological transparency, diagnostic validity, species coverage, analytical quality, and relevance to the review question. Because the search was restricted to English-language publications, relevant evidence from francophone, lusophone, Spanish-speaking, and other malaria-endemic settings may have been missed.

A second limitation is that genomic evidence is unevenly distributed across species. Plasmodium falciparum has the strongest genomic-surveillance infrastructure, partly because of its high mortality burden and the policy urgency of artemisinin resistance. Plasmodium vivax genomics has advanced substantially, but relapse biology makes epidemiological inference more difficult because a blood-stage infection may originate from a recent mosquito inoculation or activation of dormant hypnozoites. Plasmodium malariae and P. ovale remain comparatively neglected, despite evidence that they are frequently missed by microscopy and can occur in mixed infections. This imbalance means that recommendations derived from P. falciparum genomic surveillance cannot be transferred uncritically to all species of Plasmodium.

A third limitation is diagnostic verification. Many published studies use nested PCR or qPCR as the reference standard, but molecular assays vary in limits of detection and may differ in their ability to detect low-density species. Sequencing-based methods add additional thresholds related to read depth, allele frequency, contamination, and haplotype reconstruction. Therefore, the review avoids claiming that any single technology is universally superior. Instead, it argues for a tiered diagnostic and genomic strategy matched to decision need.

13. Conclusions

Mixed malaria infections should not be treated as a single biological or diagnostic category. Mixed-species infection, within-species multiplicity, and within-host resistance-associated variation involve different units of analysis, different sources of error, and different implications for care and surveillance. Their interpretation therefore requires question-specific methods and proportionate evidence. An understanding of mixed malaria infections must therefore be integrative. Epidemiology should be linked with molecular diagnosis, molecular diagnosis should be linked with genomics, genomics should be linked with treatment policy, and treatment policy should be linked with surveillance response. The strongest future studies will not merely report that mixed infections exist. They will identify which mixed infections matter, under what ecological and clinical conditions, and what programmes should do differently when they are detected. The Detect-Characterise-Act framework proposed here offers a practical route from improved detection to policy-relevant action. Countries able to detect, characterise, report, and respond to mixed infections will be better positioned to protect case management, contain resistance, and sustain progress toward malaria elimination.

CRediT authorship contribution statement

Matthew Chibunna Igwe: Conceptualisation, Methodology, Data curation, Formal analysis, Writing - review & editing. Esther Ugo Alum: Methodology, Data curation, Formal analysis, Writing - original draft, Writing - review & editing, Supervision, Validation.

Data availability

The data supporting the conclusions of this article are included within the article and its supplementary file.

Statement on the use of AI-assisted technologies

During the preparation of this manuscript, the authors used QuillBot and Grammarly tools for language editing and grammar correction. After using these tools, the authors reviewed and edited the content as needed and take full responsibility for the content of the published article.

Funding

This research did not receive any specific grant from funding agencies in the public, commercial, or not-for-profit sectors.

Declaration of competing interests

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Appendix A. Supplementary data

The following is the Supplementary data to this article: