Background
Twenty one countries across the Asia–Pacific region have set a target for eliminating malaria by 2030 [
1]. Whereas
Plasmodium falciparum case numbers have decreased significantly over the last decade, the success with
Plasmodium vivax infections has been much lower [
2,
3]. In contrast to
P. falciparum, P. vivax forms dormant liver stages (hypnozoites) that can reactivate weeks to months after the initial infection causing recurrent febrile illness and cumulative risk of morbidity and mortality. In most endemic areas, more than 60% of vivax malaria is attributable to relapsing infections [
4]. The 8-aminoquinolines primaquine (PQ) and tafenoquine (TQ) are the only available drugs capable of killing
P. vivax hypnozoites. While well tolerated in most patients, both can cause severe and potentially fatal drug induced hemolysis in patients with glucose-6-phosphate dehydrogenase (G6PD) deficiency [
5]. To improve its tolerability, PQ is administered over a prolonged 14 day course, but such a long treatment course is associated with poor adherence [
6‐
8]. The World Health Organization (WHO) recommends that all patients should be tested for G6PD deficiency (G6PDd) prior to administration of PQ or TQ, and deficient patients offered alternative treatment regimens [
9]. Shorter treatment regimens, including TQ or a 7-day high daily dose PQ regimens, are efficacious [
10,
11], but increase the risk of drug-induced haemolysis and thus the need for prior screening for G6PDd. Increased availability of G6PD testing is, therefore, essential for safe administration of radical cure treatments that will be required for the timely elimination of
P. vivax.
While the population at risk of malaria has declined over the last decade in Bangladesh, approximately 17 million population across 13 districts continue to be at risk of infection [
12,
13], with highest numbers reported from the multi-ethnic Chittagong Hill Tracts Districts (CHT) in the Southeast of the country. In parallel to the decline in overall malaria cases in the country, the proportion of malaria episodes due to
P. vivax has increased steadily, currently contributing approximately 20% of all cases [
14‐
17]. PQ is provided for the radical cure of
P. vivax, but patients are not tested for G6PDd prior to treatment. The prevalence of G6PDd is highly heterogeneous in Bangladesh [
18], and its diagnosis is challenging [
19]. The reference method is spectrophotometry but this requires a well-established laboratory infrastructure, which is rarely available in remote areas where the majority of malaria occurs [
20,
21]. The G6PD gene is X-linked. Whilst hemizygous males or homozygous females are severely deficient (< 30% enzyme activity), heterozygous females can have intermediate enzyme activity (30–80%) [
22,
23]. Qualitative point of care diagnostics are simple to use but categorize individuals above and below 30% enzyme activity and, therefore, do not identify females with intermediate deficiency who are also at risk of drug-induced haemolysis [
24]. A quantitative assay is a much safer and more equitable tool to guide radical cure treatment decisions. A novel handheld biosensor (STANDARD™ G6PD, SD Biosensor, Republic of Korea) [
25,
26], has been developed for use in point-of-care settings in endemic areas that could potentially address this shortfall.
The Bangladesh National Malaria Elimination Programme (NMEP) is currently considering novel approaches to G6PD diagnostics, supported by a dynamic research programme. If implemented, the biosensor would likely support the introduction of TQ or shorter PQ treatment courses. To assess the feasibility of implementing the biosensor in Bangladesh from the perspective of various users, a qualitative research study was undertaken to investigate user perspectives and practices of G6PD diagnostics [
27,
28]. However, considerations around specific treatment courses and
P. vivax case management as a whole were beyond the scope of the study, but have been the focus of another recent study [
29]. This paper reports the different considerations and practices at clinic, health worker and policy levels that influence the introduction ofG6PD diagnostics for new radical cure therapeutics.
Discussion
This study aimed to improve the understanding of user perspectives on the feasibility of implementing the biosensor for G6PD testing in Bangladesh, in the event that the NMEP switches from the current 14 days PQ policy without G6PD testing to shorter course radical cure regimens with G6PD testing. While study participants emphasized the life-saving potential of the biosensor, they were concerned about available resources, in particular staff time and workload, limited follow-up capacity for ensuring PQ adherence, documentation burden, supply chains and the impact of supply delays on shelf-life. They were also concerned about technical aspects of the device, including the sample collector, battery life and the devices storage conditions if kept at the community level.
Qualitative research shows that implementing diagnostics at the point of care is not straightforward, particularly in economically constrained health systems. It requires strong and well-funded health infrastructure and systems [
27], as well as providers who are undergoing continuous professional development and are actively engaged in the policy process [
35]. Such requirements were indeed raised by this study, both at the programme officer level (e.g. efforts required to raise funds and the need to make careful decisions about how and where to implement the biosensor) and at the LT level (e.g. training requirements, refresher trainings every season, etc.). Further, improvisations by health workers at every level of the healthcare system play a central and often structural role in fragmented health systems to cope with uncertainty, such as stock-outs, and need to be considered when implementing new technologies [
36]. The study participants pointed out the need to organize well the entire supply chain of biosensor devices, with proper management of quantities and timely deliveries. They also noted the need to engage a wide variety of healthcare professionals in training on the biosensor beyond those conducting it. Implicitly, this acknowledges the importance of functioning relationships between technicians and doctors or nurses, between providers for referral, and effective collaboration to ensure diagnostic processes at the point of care [
28].
Additionally, the study shows how a context of elimination and a disappearing disease add to these dynamics and perspectives in four important ways. First, the remoteness of many malaria endemic sites, as well as changing malaria hotspots require flexibility and improvisation by policymakers and health workers. Eliminating malaria is aiming at a constantly moving target. This heterogeneity makes it more difficult to introduce any new diagnostic including G6PD testing, and to budget for related supplies, human resources, or decide on deployment. As others have argued before, malaria control must be managed locally, in everyday practices [
37] and cannot be achieved through commodities alone [
38].
Second, declining patient numbers in elimination settings raise issues regarding healthcare providers having a low index of suspicion of malaria. These data show that as the numbers of malaria patients decline [
32], malaria testing can fall from the practice of both health workers and patients. Concurrently communities continue to be exposed to parasites, and elimination becomes difficult. Ensuring that malaria diagnostics are integrated into a package of diagnostic services for other febrile illnesses, so that a negative malaria test result guides further diagnostic steps rather than being a final diagnostic conclusion [
39], would increase the relevance of testing to community members given the low incidence of malaria. In the case of human African trypanosomiasis (HAT), anthropologists found that a diagnostic reflex among healthcare workers is stronger when they are in the habit of testing and have time and space to consider symptoms and that this reflex also exists among communities and patients. A diagnostic reflex can be trained in times of elimination by treating detection of events as important learning opportunities and ensuring multidimensional access to diagnostics across the health system [
40]. These data show that in order to make new technologies such as the biosensor accessible, malaria needs to be made and kept visible. The results of this study point to different strategies towards achieving this, such as: demonstrating that malaria can be eliminated, mobilizing political will and different actors (NGOs, government clinics, border control guards, police, tribal community leaders, Ministry of Health department), for instance, to find out why a potential resurgence exists and what is happening to testing practices, creating partnerships, generating research-based evidence and communicating on the risks and benefits of PQ treatment and severe reactions to PQ.
Third, the biosensor requires new forms of evidence to justify its introduction into policy. The study revealed current perceptions that there are simultaneously too few and too many
P. vivax patients to implement G6PD testing owing to challenges of funding, workload and complex testing infrastructure. Efficient communication about the need to eliminate and properly treat
P. vivax, radical cure treatment, and G6PDd is vital to ensure that the relevance of the biosensor is appreciated and its value recognized. Generating such evidence and initiating these discussions therefore becomes an important part in the decision-making process of implementing the biosensor and shorter course radical cure treatment to support
P. vivax elimination. It would also fuel ongoing debates about priorities at policy and practice levels (“
P. vivax is benign”, versus “malaria is malaria” and “all forms are equally relevant”; “radical cure is a societal not an individual concern”), which were reported in this study and which persist in Bangladesh without consensus [
41].
Fourth, declining patient numbers combined with the relative complexity and cost of the test –in comparison with simple malaria RDTs for example- complicate the process of deployment. Should G6PD testing and radical cure with shorter course regimens be centralized with qualified LTs and doctors at UHC and DH level, or rather be done at community level and in remote areas, closer to the malaria patients? The study findings provide various considerations to answer this question and highlight barriers that would need to be addressed. One important aspect is the difficulty in referring patients to the UHC or DH level in the absence of any severe symptoms and to manage the health staff’s workload if the biosensor was implemented at that level. In areas pushing for malaria elimination, additional resources towards reinforced surveillance, for instance during rainy seasons, may be required to support and complement the ongoing efforts of the staff overseeing routine care. More generally, the results show that any decisions about deployment must take into account the technical aspects, including throughput, packaging and shelf life, of the device as well as infrastructure, epidemiological, behavioural and ecological factors of the settings.
Questions around deployment and implementation strategies need to address the fact that diagnostics have different meanings for different actors (e.g. at programme level) and users (e.g. in clinics, laboratories and at the POC). The study results indicate that implementing and using the biosensor can mean offering better care and facilitating elimination of
P. vivax malaria, while strengthening one’s expertise and professional role in the healthcare system, it can help to mobilize support/funding for an NGO, it can be perceived as an indicator of good quality care, but also as a tool that increases workload in already strained work environments. Research on malaria RDTs, for example, shows the importance that such meanings and values take on for adhering to test results, and how local knowledge and understanding of good clinical practice interact with the availability of drugs and contexts of scarcity in determining the meaning of these diagnostics for health workers and patients [
36,
42‐
44]. For the policymaker, the biosensor signifies extra work obtaining funding and setting up implementation plans, guidelines and workflows. For the health worker, the biosensor may add workload too but also allows a safer prescription of shorter course PQ or TQ.
Study limitations
This study focused on LTs and a small number of programme officers and decision-makers around training workshops. Future research could assess how user expectations and concerns that were highlighted compare with reality after a few months of using the assay. Future research should include community level providers, NGOs as well as patients which were not included in this study. Without doubt, similar studies involving healthcare staff at other levels (i.e. community health workers, doctors, nurses, NGOs) as well as patients would complement the findings of this study to support the design of efficient implementation strategies for G6PD testing and new radical cure approaches, and to improve care of P. vivax patients and move towards elimination of this malaria parasite species.
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.