These efforts included increased awareness campaigns towards uptake of the use of Long-Lasting Insecticide Treated Nets (LLINs), Indoor Residual Spraying (IRS), increased use of RDTs for diagnosis [
91] and introduction of artemisinin combination therapy (ACT) for treatment. Unfortunately, the projected target for elimination could not be achieved, because of challenges in detecting asymptomatic infections and the sporadic cases of malaria localized in the north, eastern, southeastern and parts of the southern regions [
92] (
Figure 2). The failure was partly because the surveillance strategy had been based on passive case detection which did not account for asymptomatic infections and non-falciparum malaria [
93]. The burden of malaria disease has traditionally been in the northern part of the country which is wetter and warmer compared to the southern part of the country. Occasionally the relatively wetter parts to the eastern, southeastern, and southern parts of the country also experience sporadic transmission, depending on the rainfall patterns [
92]. Rainfall is seasonal and occurs between November and May [
92,
94] so the transmission season occurs within the same period.
In 2008 and 2009, the foci of malaria disease burden were in three districts: Okavango, Chobe, and Ngami in the north (
Figure 2). By 2010, the burden in Okavango had diminished a bit with the foci localized in Ngami and Chobe (
Figure 2). At the same time a small node of transmission was beginning from Ngami towards Boteti. This became significant in 2011 (
Figure 2). By 2012, the entire stretch of land from the northeastern section towards the southeastern border regions were experiencing significant and sporadic malaria outbreaks. These outbreaks were seen despite the general reduction in malaria disease burden <1 per 1000 populations, with reference to
P. falciparum as the main causative parasite. It was a puzzling occurrence and after several discussions on the way forward, it was decided that an active survey was needed to unravel some elements of the puzzle. In 2012, considering that the target for malaria elimination was 2015, an active survey was done across the country, within the transmission period and species-specific nested PCR was used for detection of all Plasmodium species and asymptomatic infections. The result of this survey was profound as published previously [
88]. We affirmed that indeed
P. falciparum was mainly localized in the Kavango and Ngami regions (no samples were collected in Chobe at the time), with some asymptomatic infections. However, we observed for the first time that
P. vivax asymptomatic infections were present and localized at hotspots across the country [
88]. Areas where the asymptomatic
P. vivax burden were found mirrored the observed outbreaks, in a way that suggested that they were part of the epidemic outbreaks that were recorded. In addition, areas in the South (Kweneng), that had seen some occasional outbreaks but deemed not to be important (
Figure 2), were discovered as major foci of
P. vivax asymptomatic infections. These findings corroborated the observed characteristics of
P. vivax transmission, which is that
P. vivax asymptomatic infections significantly contribute to transmission and account for the majority in low endemic settings, as observed elsewhere [
95,
96]. In addition, they were highest in areas where
P. falciparum were not detected by nested PCR, indicating a reciprocal interaction between the two parasites as have been observed generally. Since samples were collected in the transmission season, we could not make a claim to relapsing
P. vivax hypnozoites. Two questions were on our minds: if these were partly hypnozoites, that relapsed a. what caused the triggers for activation; and b. were these relapses from liver hypnozoites only? Since the outbreaks were spread out over a large area, within the same period, we wondered if the trigger/triggers were something in the environment that led to a physiological response and activation or due to an innate biological clock that spontaneously induced an activation. We also wondered whether the parasites responses to a trigger were spontaneous within a giving geographical area (what we would term a ‘
Spoke wheel effect’) (
Figure 3) or relayed. Following the extension of the elimination agenda to 2018, in 2016, to insight into the nature of the observed
P. vivax parasites, we again, we embarked on another survey at the same sites as we did previously and included Chobe that we had previously missed. This time we collected samples all year round from August 2016 to October 2017, and added qPCR to increase the sensitivity of detection. The findings were particularly interesting. We noted that not only was
P. vivax still present in the sites we had previously seen, but also for some unknown reasons
P. falciparum had re-appeared in all the places that
P. vivax had been seen previously. In addition, with the return of
P. falciparum, the burden of
P. vivax diminished but did not disappear, allowing for coexistence within the communities. Interestingly, we identified
P. vivax in the non-transmission season, indicating a role of hypnozoites relapses in the population. The active survey reflected in the epidemic outbreak for the 2016/2017 period, showing an 80% increase in malaria transmission for that period [
97], (WHO, world Malaria report, 2018). In 2018, the NMP mapped out areas of active malaria transmission across the country, using a modelling approach. It was evident that the predicted active places were consistently in areas where
P. vivax asymptomatic infections had been detected. Not surprisingly, the elimination date was revised to 2025, following a review in 2021 by the Malaria Elimination Oversight Committee (MEOC) of the WHO, to include Botswana in the Elimination 2025 (E-2025) countries [
98].
Figure 2.
Annual patterns of malaria disease prevalence in Botswana: from 2008 to 2012 (Source: NMP Botswana).
Figure 3.
Asymptomatic Plasmodium infections in 2012 and 2016-2017.
Figure 4.
Predicted malaria prevalence 2018 (Source: NMP Botswana).