Malaria is a vector-borne disease caused by parasites of Plasmodium species that are transmitted to people through the bites of infected female Anopheles mosquitoes and is one of the major public health problems worldwide. In 2021, there were 247 million estimated cases of malaria worldwide, and 619,000 estimated malaria-related deaths. The World Health Organization (WHO) African Region carries a disproportionately high share of the global malaria burden (1). After decades of efforts, China has not reported indigenous malaria cases since 2017 (2) and was certified malaria-free by the WHO on the 30th of June 2021. However, a large number of imported malaria cases have been reported annually in China (3).

Timely detection and identification of all malaria cases, the first line of defense against malaria in China, is essential to prevent transmission and deaths (4, 5). Methods used for malaria diagnosis in health facilities include microscopy and rapid diagnostic tests (RDTs), and RDTs were usually based on the detection of Plasmodium falciparum lactate dehydrogenase (pf-LDH) and Plasmodium lactate dehydrogenase (pan-pLDH) antigens. Since 2011, the China malaria diagnosis reference laboratory network based on provincial Centers for Disease Control and Prevention (CDCs) and the Institute of Parasitic Diseases (IPDs) has been set up in stages to review and assess malaria cases using microscopy and molecular diagnostic tools (6). The provincial malaria diagnosis reference laboratories are primarily responsible for case reviews, malaria diagnostic capacity training, and quality assurance of the performance of malaria case identification in health facilities (7). A number of measures to control the internal quality assessment of malaria diagnosis, such as routine reviews of blood samples, and provincial blind sample assessment, have been carried out continuously. Furthermore, technique competitions of parasitic diseases including malaria were held to promote diagnostic capacity in health facilities in the province (8). Likewise, the provincial reference laboratory diagnosis provides guidance for epidemiological investigation and standardized treatment of reported cases.

Anhui, Henan, and Hubei provinces are historically malaria-endemic provinces in central China (9–11). Guangxi Zhuang Autonomous Region and Zhejiang Province, located in the south and southeast of China, respectively, are the regions with large number of imported malaria cases (12, 13). In order to assess malaria diagnosis and explore strategy to improve diagnostic capacity, we analyzed in this study the accuracy of malaria diagnosis in the five provincial-level administrative divisions (PLADs) of Anhui, Henan, Hubei, Guangxi, and Zhejiang from 2014 to 2021.

The malaria diagnosis reference laboratory network has provided technical support for identifying malaria cases during its elimination and prevention of re-transmission of imported malaria in China. Malaria diagnosis in China still faces many challenges, including improvement of microscopy competency at lower levels (7) and promotion awareness of malaria diagnosis in an elimination setting (16).

Microscopy is still the gold standard method for malaria diagnosis (17), although it is time-consuming and requires extensive technical experience. Malaria RDTs have been widely used in primary health facilities due to their simplicity and convenience (18). In fact, more than 78% of the total diagnoses were made using both microscopy and RDTs in the five PLADs, and microscopy showed better sensitivity than RDTs. Malaria RDTs are usually most effective in cases of P. falciparum (19). Jean et al. revealed that the sensitivity of RDTs was 93% (90.6–94.2%) for P. vivax in French Amazonia (20); whereas RDTs still showed low sensitivity and specificity for P. ovale and P. malariae in various other diagnostic contexts (21). In addition, malaria RDTs based on P. falciparum histidine-rich protein 2 (HRP2) raised malaria public health concerns because HRP2 deletions have been reported in different parts of the world (22). In view of their limitations, malaria RDTs are suggested to be complemented by microscopy to detect malaria, rather than being used as an alternative to microscopy in non-endemic areas (23, 24).

The accuracy for species identification initially diagnosed in the five PLADs was 90.51% from 2014 to 2021, which was higher than the 62–66% accuracy reported in China in 2015 (7), but similar to the 88.8% reported in Jiangsu Province in 2017 (25) and the 92.6% reported in Yunnan Province from 2013 to 2018 (26). Moreover, the accuracy of P. falciparum diagnosis remained above 94% in all five PLADs, but showed poorer accuracy for species identification of P. ovale and P. malariae. This may be due to the distribution of imported malaria species reported in China over recent years (27, 28), with P. falciparum as the predominant species, increased cases of P. ovale, and decreased cases of P. vivax. Consequently, the diagnostic capacity of health facilities for P. falciparum was maintained at a high level, whereas the capacity for non-falciparum species declined. Further comparison of species identification between the five PLADs showed that the accuracy of malaria diagnosis in Guangxi Zhuang Autonomous Region was higher than that in the other four provinces. This may be related to migrant workers returning from Africa (i.e., the Republic of Ghana) to Shanglin County of Guangxi in clusters which resulted in a significant increase in the number of malaria cases in 2013 (12). Therefore, the high prevalence of malaria cases at these might be advantageous in maintaining the microscopic ability to identify malaria in a stable state.

Case detection and accurate identification of Plasmodium species directly affect the implementation of malaria prevention and control measures (29). According to the national technical program for malaria elimination, the provincial reference laboratories for malaria diagnosis are required to provide timely feedback on results of confirmatory diagnosis to the administrative county from where malaria cases were reported. The results showed that the time interval from cases being reported to samples being received by the reference laboratories was shorter in 2018 to 2021 than that in 2017. Given the risk of local transmission caused by P. vivax (30) due to suitable vectors and conditions in historically malaria-endemic areas in China, timely detection of P. vivax for foci investigation and disposal to prevent re-transmission is of crucial importance. As shown in this study, the time interval was shorter for P. vivax than that for non-vivax species. Another reason to pay attention to P. vivax is that P. vivax forms dormant in the liver (hypnozoites) that can cause relapses and requires primaquine for radical cure (31), which remains a challenge to diagnosis and treatment.

The inconsistency between microscopy and nPCR confirmation at provincial reference laboratories was mainly in determining malaria positivity or negativity. Explanations for this included low parasite density affecting the ability to detect malaria by microscopy (32), some patients with illness history self-medicated after their return from overseas resulting in low detectability of viable parasites in the bloodstream by microscopy, or non-optimal sampling time of blood smears and samples for diagnosis (33). In addition, some differences in reference laboratory confirmatory diagnosis showed inconsistencies between mixed and single infections, as well as between P. ovale and P. vivax. Therefore, it is necessary to strengthen training to distinguish the morphology of Plasmodium species mentioned above and improve the detection ability of microscopists in samples with low parasitemia.

Our study has two main limitations. First, the study was retrospective; hence, some information may have gone missing during the study period, such as the time interval of samples received by the reference laboratories from 2014 to 2016. Secondly, the database only included reported malaria-positive cases; those initially diagnosed as malaria with confirmed malaria-negative results were not included in the analysis.

The findings of this study showed that the accuracy of species identification by health facilities was higher than 90% in the five PLADs from 2014 to 2021, but fell short in the detection of P. ovale and P. malariae. Capacity training for the identification of Plasmodium species in health facilities needs to be strengthened. Multiple test approaches including microscopy and nPCR have enabled correct diagnosis of malaria, thus contributing to malaria elimination in China. The provincial malaria diagnosis reference laboratories have played an important role in ensuring the accuracy and reliability of malaria diagnosis in the elimination phase, as well as in the consolidation phase of malaria elimination.

The original contributions presented in the study are included in the article/supplementary material, further inquiries can be directed to the corresponding authors.

Ethical approval was not required for the study involving humans in accordance with the local legislation and institutional requirements. Written informed consent to participate in this study was not required from the participants or the participants’ legal guardians/next of kin in accordance with the national legislation and the institutional requirements.

SZL, WR, and DQW conceived and designed the study. JJJ, YS, HY, JX, YL, LCS, XXW, and SNL contributed to data collection, finding interpretation, and the writing. XZ and JM performed the analyses. XZ wrote the manuscript. All authors contributed to the article and approved the submitted version.