Much emphasis has been placed on the severe forms of malaria due to their high mortality and resulting sequelae. Due to its role in malaria transmission, asymptomatic malaria, until recently has not been considered of much clinical concern, but it is now gaining important consideration [27]. High rates of asymptomatic malaria have been observed in African children with rates as high as 80 % in the wet seasons [27, 43]. Two studies have so far evaluated the effects of asymptomatic malaria on cognition in children.

In a study to assess the effect of asymptomatic malaria on cognition, Al Serouri et al. [1] compared 445 Yemeni children with asymptomatic malaria to 142 children matched for grade and school on tests of verbal fluency, working memory, sustained attention, psychomotor speed, fine motor coordination and visual recognition [1]. Two weeks later, 150 children still asymptomatic were compared to 150 children whose parasitemia cleared on the above tests. Regression analysis controlling for age, socioeconomic status and nutrition found fine motor coordination to be the only cognitive ability affected by asymptomatic malaria. Parasite density predicted fine motor coordination scores at baseline and improvement in visual recall after parasite clearance. No significant change in cognitive scores was noticed 2 weeks later between those who were still asymptomatic and those whose parasites cleared. Other factors associated with cognitive test performance were anemia, wealth factor and age. The authors concluded that parasitemia may affect fine motor skills and visual recall.

In a Ugandan study, Nankabirwa et al. [50] carried out a study to assess the association between anemia, asymptomatic malaria and cognition in Ugandan school children aged 6–14 years [50]. They assessed sustained attention and abstract reasoning in 740 children as well as, sociodemographic variables, helminth infection and haemoglobin. About 30 % of the children had asymptomatic malaria. Multiple regression analyses controlling for sex, age, weight for age Z score, helminth infection, anemia, socioeconomic status and maternal education showed an association between asymptomatic malaria and sustained attention (adjusted mean difference: −1.6, 95 % CI: −2.49 to −0.81) and abstract reasoning (−0.6, 95 % CI: −1.01 to −0.21). Higher parasitemia was associated with lower attention scores. Unlike the study by Al Serouri et al. [1], anemia was not associated with any cognitive outcome in this study. They concluded that effect of asymptomatic malaria on cognition could be a result of repeated infections or an immunological pathway.

These studies suggest that asymptomatic malaria affects cognition in children though it is not clear how this occurs since both studies reported slightly differing results. In the Yemeni study, being anemic (haemoglobin ≤90 g/l) was associated with lower cognitive scores which was not the case in the Ugandan study. Nankabirwa et al. [50] showed that higher parasitemia affected sustained attention scores while Al Serouri et al. [1] reported that asymptomatic children who became parasite free 2 weeks later did not show an improvement in cognitive scores, implying that parasite presence is mostly likely not the main cause for low cognitive scores.

Fernando and colleagues have carried out a series of studies looking at the effect of uncomplicated malaria on cognitive outcome and school performance in children. They followed up 571 children aged 1–8 years for 6 years from 1992 and recorded the number of malaria episodes [26]. Children’s scores in mathematics and language in 1997 were associated with the number of malaria episodes. Those who experienced three or less episodes scored 15 % higher than those with more episodes. These findings were replicated in a cross sectional study where they assessed the effect of repeated malaria attacks on measures of writing, reading skills, letter reading, sentence structure and mathematics in 325 children aged 4–6 years [25]. Children were tested at one time point and number of malaria episodes suffered in their life time recorded. Total number of malaria attacks was associated with letter reading and languages scores, where those not experiencing any malaria episode had scores 19 % higher than those who had five or more episodes for letter reading and 4.3 % for language. They later investigated the short term impact of an acute malaria episode on tests of mathematics and language where 199 children with malaria and 144 children with non malaria fever were tested at presentation to a health facility and 2 weeks later after treatment of the malaria bout [24]. A control group of 305 healthy children were also given the same tests. Children with malaria performed significantly poorer in mathematics and language at both time points compared to the controls. Those with non malaria fever also had poorer scores at baseline but not at follow-up. Contrary to expected results, higher haemoglobin was associated with poorer scores in mathematics and language.

Recently Thuilliez and colleagues followed up 227 Malian children for 8 months recording malaria episodes, socioeconomic variables and cognition [55]. Children who had not experienced any malaria episode had better scores than those who had malaria in the follow-up period. In addition, higher parasitemia was associated with lower cognitive scores. The effect of anemia on cognition was however not conclusive. Malaria was also the most common cause of absenteeism from school.

Contrary to all the above findings, Halliday and colleagues did not observe any effect by either malaria infection or anemia during malarial illness on sustained attention, literacy, cognition or numeracy in 2,400 Kenyan children [29]. Other non-health related variables like age, gender, socioeconomic status, parental education and school environment were associated with cognitive outcomes suggesting health status (single malaria episode or anemia) may not be predictive of cognition in children in moderate malaria transmission areas after non-severe malaria.

Despite these findings from Halliday et al. [29], the preceding studies provide strong evidence that uncomplicated malaria affects cognition and academic performance in children with number of malaria episodes being the best predictor of outcome. The mechanisms for the effect on cognition are not clearly understood.

Primary prevention interventions will target prevention of malaria infection in order to remove the risk of these cognitive deficits. Malaria prevention approaches include; vaccination, vector control (insecticide-treated bed net use, indoor residual spraying), chemoprevention and chemoprophylaxis [58]. Chemoprevention reduces the incidence of malaria, parasitemia and anemia both of which are independently associated with poor cognitive outcome [51]. In Gambia, Jukes et al. assigned 1,190 children under 5 years of age to either chemoprophylaxis or placebo for 4 years during the malaria transmission season and followed up 14 years later [39]. No significant differences were seen in mental development scores between the treatment groups. Children who received chemoprevention achieved just over half a grade higher in school, a significant difference compared to controls. Clarke et al. assigned 6,758 children aged 5–18 years to intermittent preventive treatment (IPT) or a placebo [18]. At 12 months post intervention, the IPT group had a lower prevalence of anaemia and had higher attention scores than the placebo group. No trials have evaluated the effect of the other malaria prevention interventions on cognition.

Secondary prevention of cognitive outcomes in children infected with malaria involves administration of effective treatments and or using adjunct therapies to combat disease processes associated with cognitive impairment. In severe malaria, artesunate is more effective than quinine in reducing mortality and also reducing seizures and coma depth, two clinical features associated with poor cognitive outcome [20]. There was no significant difference in neurological outcome at 28 days between the groups. However trials for adjunct therapies for severe malaria have not been successful in reducing mortality, coma or seizures [38].

Children surviving severe malaria are at risk for cognitive impairment with 1 in 4 going on to have long term cognitive impairment [36]. Trials using computerised cognitive rehabilitation training in African children surviving CM have shown benefits in cognition and behaviour suggesting that interventions carried out after the disease may be beneficial [3, 5]. In these trials, children were assigned to 16 sessions, each last 45 min delivered over 8 weeks (two sessions per week). Children completed cognitive exercises training memory, attention, visuomotor and reasoning. These studies however did not have an active control group that was also exposed to computers, had no long term follow-up and it is not known which of the four categories of cognitive exercises are more effective or whether benefit can be obtained by training one skill as has been done successfully for ADHD [44]. Despite these limitations, they do improve cognition in children surviving severe malaria.

The earliest studies on the effect of HIV on African children’s cognition were carried out in the pre-ART era. Msellati and colleagues in Rwanda compared neurodevelopmental outcome (gross motor, fine motor, social contact and language) of perinatally HIV infected children to two control groups; HIV negative children born to HIV positive mothers (HIV exposed) and HIV negative children born to uninfected mothers (HIV unexposed) [48]. Tests were done at 6, 12, 18 and 24 months of age. The HIV positive children had a higher frequency of abnormal neurodevelopmental assessments than the control groups at 6 months (16 vs. 3.3, p = 0.0001), 12 (31 vs. 7.7, p = 0.0001), 18 (40 vs. 7.9, p = 0.0001) and 24 (15 vs. 2.2, p = 00001). Gross motor was the most affected area which had significantly poorer scores at all the three time points. The HIV exposed and unexposed groups did not differ in any of the neurodevelopmental areas tested. No risk factors for abnormal neurodevelopmental assessment were identified though the frequency of an abnormal assessment was much higher in children with clinical AIDS (ranging from 22 % to 87.5 % at the four time points).

A study by Boivin et al. in Zaire (now Democratic Republic of Congo) assessed two groups of children for cognition [10]. In the first sub-study among children below 2 years, 14 perinatally infected children were compared to 20 exposed and 16 unexposed children on tests of gross motor, fine motor, language and personal-social skills. Testing was done at 3, 6, 9, 12 and 18 months of age. Repeated measures analysis of variance showed the HIV group had poorer gross motor, fine motor and personal-social skills than the other groups. Similar to the previous study by Msellati and colleagues [48], there were no differences in outcome between the exposed and unexposed groups. In the second sub-study with children over 2 years of age, 11 children perinatally infected were compared to 15 exposed and 15 unexposed children on tests of cognition, language, motor, sequential processing, simultaneous processing, nonverbal reasoning and mental processing composite. The HIV infected group had poorer scores in motor, sequential processing, simultaneous processing, nonverbal reasoning and mental processing composite. Contrary to the earlier studies above, the exposed group performed poorer on many of the neurodevelopmental outcomes compared to the unexposed group.