Diagnostic performance of malaria microscopy among laboratory professionals in Ethiopia: a systematic review

Abstract

Background

Malaria remains a major public health concern in Ethiopia, where light microscopy serves as the primary diagnostic method. Accurate microscopy performance among laboratory professionals is critical for appropriate case management, rational antimalarial use, and malaria control efforts. This systematic review synthesized available evidence on the diagnostic performance of malaria microscopy in Ethiopia.

Methods

This systematic review evaluated malaria microscopy performance in Ethiopia in accordance with PRISMA 2020 guidelines. Electronic databases and manual searches were used to identify cross-sectional studies published between 2014 and 2021 that evaluated malaria microscopy performance using External Quality Assessment (EQA) components. Extracted indicators included sensitivity, specificity, detection agreement, species identification agreement, and associated factors. Due to the absence of primary numerator and denominator data in most studies, quantitative meta-analysis was not feasible; therefore, findings were summarized descriptively using ranges and means.

Results

Twelve cross-sectional studies involving 1173 laboratory professionals from 533 public health laboratories were included. Sensitivity ranged from 62 to 94.7% (mean: 78.9%), while specificity ranged from 79.7 to 100% (mean: 90.8%). Detection agreement averaged 85.2% (range: 71.4–97.3%). Species identification agreement was lower, averaging 70.6% (range: 43.8–95.0%). Frequently reported factors associated with suboptimal performance included inadequate refresher training, limited EQA participation, poor smear preparation and staining quality, insufficient supportive supervision, and high workload.

Conclusion

Although malaria parasite detection and specificity by microscopy are relatively satisfactory in Ethiopia, sensitivity and species identification accuracy remain below WHO-recommended benchmarks. Strengthened competency-based training, structured EQA implementation, improved slide preparation practices, and enhanced supportive supervision are essential to improve diagnostic performance and support malaria elimination efforts.

Supplementary Information

The online version contains supplementary material available at https://doi.org/10.1186/s12936-026-06023-7.

Introduction

Malaria continues to pose a significant public health challenge in Ethiopia, where a substantial proportion of the population resides in malaria-endemic areas [1]. The disease remains a major contributor to outpatient visits, hospital admissions, and preventable deaths, particularly among vulnerable populations [2]. Effective malaria control and elimination strategies rely heavily on accurate and timely laboratory diagnosis to guide appropriate case management, prevent unnecessary treatment, and ensure rational use of antimalarial drugs [3].

Light microscopy remains the reference standard for malaria diagnosis in Ethiopia. Unlike rapid diagnostic tests, microscopy enables identification of Plasmodium species, quantification of parasite density, and monitoring of treatment response [4]. However, the accuracy of microscopy is highly dependent on technical competency, including proper blood smear preparation, staining quality, parasite identification skills, and microscope maintenance. Variability in any of these components may compromise diagnostic reliability. To ensure consistent performance, the World Health Organization recommends the implementation of comprehensive laboratory quality assurance systems [5].

External Quality Assessment (EQA) programs comprising panel testing, blinded slide rechecking, and onsite supervision are widely adopted mechanisms to monitor laboratory performance and provide corrective feedback [6]. Ethiopia has scaled up EQA activities as part of its national malaria control strategy [7]. Nevertheless, regional reports indicate differences in sensitivity, specificity, and agreement rates among laboratory professionals.

Given the critical role of laboratory diagnosis in malaria control and the observed variability in reported performance, a systematic synthesis of available evidence is essential. This review therefore aimed to assess the overall diagnostic performance of malaria microscopy among laboratory professionals in Ethiopia and to identify key factors contributing to poor performance.

Methods

Study design

This study employed a systematic review design guided by the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2020 framework [8]. The review sought to comprehensively identify, appraise, and synthesize published studies evaluating malaria microscopy performance in Ethiopia.

Protocol and registration

The review protocol was not prospectively registered in PROSPERO or another registry. However, the review was conducted according to PRISMA guidelines.

Eligibility criteria

Studies were considered eligible if they met the following criteria: they used a cross-sectional design; were conducted within Ethiopia; evaluated the diagnostic performance of malaria microscopy among laboratory professionals; and utilized recognized EQA components such as panel testing, slide rechecking, or onsite evaluation. Studies reporting malaria microscopy performance in terms of sensitivity, specificity, or detection agreement were included in our study. Studies were excluded if they focused exclusively on rapid diagnostic tests (RDTs), were review articles or case reports, were conducted outside Ethiopia, or failed to report measurable diagnostic performance indicators.

Search strategy

A comprehensive and systematic literature search was conducted to identify relevant studies assessing the diagnostic performance of malaria microscopy among laboratory professionals in Ethiopia. The search was performed using multiple electronic databases, including PubMed, Scopus, and Google Scholar, to ensure broad coverage of published literature.

The search covered studies published from January 1 st 2014 to December 31, 2021. The final search was performed on January 15, 2026.

The search strategy combined terms related to malaria microscopy, diagnostic performance, and Ethiopia using Boolean operators (AND/OR). Search terms were adapted appropriately for each database. In addition to electronic searches, manual screening of reference lists of eligible articles was performed to identify additional relevant studies. Duplicate records were removed prior to screening.

Data study selection

Titles and abstracts were screened for relevance, followed by full-text assessment based on predefined eligibility criteria. After systematic screening and eligibility assessment, only twelve studies fulfilled the predefined eligibility criteria, including study design, Ethiopian setting, malaria microscopy performance evaluation, and reporting of measurable diagnostic indicators were included in the final synthesis (Fig. 1).

Fig. 1.

PRISMA flow diagram of article selection for the systematic review of diagnostic performance of malaria microscopy among laboratory professionals in Ethiopia

Data extraction

Data were extracted using a structured template to ensure consistency. Extracted information included authorship, publication year, study region, sample size, EQA method, results/outcomes, and factors associated with diagnostic performance. Data extraction was performed carefully to ensure completeness and accuracy (Table 1).

Table 1.

Summary of included articles in systematic review of diagnostic performance of Malaria Microscopy Among laboratory professionals in Ethiopia

S/N	Author et al., Year	Study design	Facility/region	Sample size	EQA method	Results/Outcomes	Factors for poor performance
1	Ayalew F. et al. 2014	Cross-sectional	11 public and private health laboratories Hawassa Town	80 laboratory professionals	Panel testing	Sensitivity: 82%; Specificity: 96.5%; Detection agreement: 88% (K = 0.76); Species ID agreement: 74.3% (K = 0.63)	Lack of training
2	Alemu M. et al. 2017	Cross-sectional	22 health laboratories Tigray, North Ethiopia	46 laboratory professionals	Onsite evaluation and panel testing	Sensitivity: 62%, Specificity: 95.7%, Detection agreement: 79% (K = 0.62), Species identification agreement:87% (K = 0.58)	Lack of refresher training, no participation in malaria epidemic investigation
3	Atalele K. et al. 2018	Cross-sectional	14 Health laboratories, Bahir Dar city, NW Ethiopia	87 laboratory professionals	Panel testing	Sensitivity: 83%; Specificity: 97%; Detection agreement: 88.5% (K = 0.78); Species identification agreement: 72% (K = 0.47)	Lack of in-service training
4	Challi S. et al. 2020	Cross-sectional	86 districts, Oromia	174 laboratory professionals	Panel testing	Sensitivity: 69.7%, Specificity: 80.4% Detection agreement: 74.3% (K = 0.45); Species identification agreement: 71.2% (K = 0.40)	Low overall competence, lack of in-service training and supportive supervision
5	Yitbarek T. et al. 2016	Cross-sectional	16 health laboratories, Addis Ababa	60 laboratory professionals	Panel testing	Sensitivity: 65.7%; Specificity: 100%; Detection agreement: 71.4% (K = 0.4); Species identification 51.1% (K = 0.04)	limited exposure in non-malarious areas, poor skills
6	Tegegne B. et al. 2020	Cross-sectional	30 health laboratories, West Amhara	6689 slides	Slide rechecking	Detection agreement: 97.31%; Species identification agreement: 94.6%	workload, microscope condition, laboratory poor quality reagent, limited supportive supervision
7	Jaleta F. et al., 2020	Hospital-based cross-sectional	46 public hospitals, Eastern & Central Oromia	46 laboratory professionals	Onsite evaluation and Panel testing	Sensitivity: 77.63%, Specificity: 96.41%, Detection agreement: 80.45% (K = 0.79); Species ID agreement: 63.03% (K = 0.38)	Lack of training, low work experience
8	Gebrekidan L. et al. 2019	Cross-sectional	50 health laboratories, Addis Ababa	211 laboratory professionals	Panel testing	Sensitivity: 94.69%; Specificity: 79.71%; Detection agreement: 91.7%; Species ID agreement: 67.63%;	Lack of training, low participation in external quality assessment
9	Gidey B. et al. 2021	Cross-sectional	102 health laboratories, Nationwide from all regions	202 laboratory professionals	Onsite evaluation, slide rechecking and panel testing	In panel testing: Sensitivity: 92.7%, Specificity: 89.1%, Detection agreement: 91.8% (K = 0.79); Species ID agreement: 57% (K = 0.18); In slide re-checking: Sensitivity: 78.1%, specificity: 80.7%, Detection agreement 76% (K = 0.41)	limited mentorship, lack of in-service training
10	Nega D. et al., 2020	Cross-sectional	106 health laboratories from 20 districts	237 laboratory professionals	Onsite evaluation and panel testing	Sensitivity: 83.2%, Specificity: 90.1%, Detection agreement: 84.6% (K = 0.6); Species ID agreement: 43.8% (K = 0.11)	Poor quality of supply, low skill, poor laboratory set up
11	Sori G. et al. 2018	Cross-sectional	30 public health laboratories, East Wolega Zone, Oromia	30 laboratory professionals	Onsite evaluation, panel testing and slide rechecking	In panel testing; Sensitivity: 77%, Specificity83.3%, Detection agreement: 78%(K = 0.5) In slide rechecking; Sensitivity:78%, Specificity: 83.7% Detection agreement: 82%(K = 0.62)	Poor smear and staining quality, lack of in-service training
12	Hailu H.A. et al. 2017	Cross-sectional	20 health laboratories, West Amhara	3148 slides	Slide rechecking	Detection agreement: 96.62% Species identification agreement:95.01%	Limited participation in EQA

Risk of bias assessment

Although all included studies were cross-sectional competency evaluations, formal quality appraisal was performed using a modified checklist for cross-sectional diagnostic performance studies. Most studies clearly described sampling procedures and EQA methodology; however, variability in reporting completeness and lack of standardized reference slide validation were noted (Supplementary file 1).

Data synthesis

Due to absence of primary numerator/denominator data, pooled meta-analysis was not feasible. Diagnostic performance indicators were summarized descriptively using reported ranges and unweighted arithmetic means.

Result

Characteristics of included studies

The review included twelve (n = 12) cross-sectional studies conducted across various regions of Ethiopia between 2014 and 2021. These studies represented diverse geographic settings, including Oromia, Amhara, Addis Ababa, Tigray, South Ethiopia, and nationwide assessments. Studies included a total of 1173 laboratory professionals from 533 public health laboratories across Ethiopia.

The EQA methodologies employed varied across studies, including panel testing exercises designed to evaluate individual competency, blinded slide rechecking programs assessing routine diagnostic accuracy, and onsite evaluations examining laboratory practices and infrastructure.

Malaria microscopy performance

Sensitivity and specificity

Ten studies reported sensitivity values, which ranged from 62 to 94.69%. The calculated mean sensitivity across studies was 78.93%. The specificity was also reported and ranged from 79.7 to 100%, with a mean of 90.82% [9–18]. This pattern reflects relatively strong overall diagnostic accuracy, but also highlights that sensitivity (detecting true positives) remains more variable and may need targeted improvement.

Detection and species identification agreement

All twelve studies assessed detection agreement, which ranged from 71.4 to 97.31%, with a mean of 85.18% [9–20]. However, eleven studies assessed species identification agreement which ranged from 43.8 to 95.01% with a mean of 70.6% [9–16, 18–20].This indicates that while detection was fairly reliable, accurately distinguishing the species was more challenging.

Associated factors for poor performance

Multiple studies reported factors contributing to suboptimal microscopy performance. The most consistently cited determinant was inadequate in-service or refresher training [9–15]. Limited participation in EQA programs, such as irregular engagement in panel testing, slide rechecking, and supportive supervision, tended to be associated with poor performance [11–13, 19, 20]. Additionally, operational constraints, including high workload [20], poor technical skill [11, 16–18], poor smear preparation and inconsistent staining quality [17, 20] were also frequently reported challenges.

Discussion

This systematic review synthesized evidence from twelve cross-sectional studies evaluating malaria microscopy performance in Ethiopia using External Quality Assessment (EQA) approaches. Overall, the findings demonstrate moderate diagnostic performance, with relatively higher specificity and detection agreement compared to sensitivity and species identification accuracy.

According to WHO malaria microscopy quality assurance guidelines, laboratories are expected to achieve high levels of parasite detection and species identification accuracy, ideally ≥ 90% agreement [5]. In this review, the detection agreement ranged from 71.4% to 97.31%, with an average of 85.18% and species identification agreement ranged from 43.8% to 95.01% with an average of 70.6%. These findings suggested that although few individual studies achieved or exceeded the standard, the overall average detection agreement and species identification agreement falls below the recommended standard. However, the overall detection agreement level of this study is comparable to findings from studies in other endemic countries like Kenya and Tanzania [21, 22].

In the present review, the mean specificity (90.82%) aligns with WHO expectations, suggesting relatively good control of false-positive results in Ethiopia. However, the pooled mean sensitivity (78.93%) falls below optimal WHO targets [23]. From a public health perspective, suboptimal sensitivity is particularly concerning because missed diagnoses may result in delayed treatment and continued transmission [24].

A comparison with the Ugandan study conducted in private health facilities in Entebbe reveals a contrasting pattern. In Uganda, microscopy demonstrated very high sensitivity (95.8%) and good specificity (90.1%), closely meeting WHO performance benchmarks. Unlike Ethiopia, where the mean sensitivity was 78.93%, the Ugandan findings suggest that near-optimal performance is achievable in similar African settings [25].

The findings of this review identified inadequate in-service training, limited participation in External Quality Assessment (EQA), high workload, poor smear preparation, inconsistent staining quality, and insufficient technical skill as major contributors to suboptimal microscopy performance. Similarly, one of the strongest and most consistent determinants of poor performance in Nigeria and Ghana is inadequate or outdated training [26, 27]. Technical deficiencies in smear preparation and staining were also another major factor consistently associated with poor performance in Uganda and Côte d’Ivoire [25, 28]. In addition, some studies highlighted the absence or weakness of structured quality assurance systems (EQA) as a contributing factor for poor malaria microscopy performance [22, 25].

Limitations

This review had some limitations. First, a pooled meta-analysis could not be performed because of substantial heterogeneity among the included studies. Second, the studies used variable EQA methodologies (panel testing, slide rechecking and onsite evaluation), which may have affected the comparability of findings across different settings. Third, inconsistent reporting of numerator and denominator data in some studies limited the ability to conduct comprehensive quantitative analysis and direct comparisons.

Conclusion

Malaria microscopy performance among laboratory professionals in Ethiopia demonstrates acceptable malaria detection and specificity, but sensitivity and species identification accuracy remain suboptimal. Key factors contributing to suboptimal performance include inadequate refresher training, limited participation in External Quality Assessment programs, inconsistent supervision, and operational challenges such as high workload and poor slide preparation.

Strengthening laboratory capacity through regular competency-based training, routine EQA participation, and improved quality assurance practices is essential to enhance diagnostic accuracy, align with WHO benchmarks, and support malaria control and elimination initiatives in Ethiopia.

Author contributions

BM conceived and designed the study, conducted literature search, data extraction, analysis, and drafted the manuscript. KM contributed to study design, methodological review, and critical revision of the manuscript. ZH contributed to data interpretation and manuscript editing. All authors read and approved the final manuscript.

Funding

The authors received no specific funding for this work.

Data availability

No datasets were generated or analysed during the current study.

Declarations

Ethics approval and consent to participate

Not applicable. This study was based exclusively on previously published data and did not involve human participants or identifiable personal data.

Competing interests

The authors declare no competing interests.