Introduction
Parasitic disease, particularly Plasmodium falciparum malaria, remains a major health problem for schoolchildren in sub-Saharan Africa.1–4 In this population, malaria is an important contributor to anemia, malnutrition, and mortality.5,6 In addition, impaired cognition and poor school performance are important hidden burdens of malaria, which may prevent children from achieving their full educational potential.7–12
The health and educational burden of malaria is the result of both clinical disease and asymptomatic parasitemia, with the latter much more common than the former13; the consequences of asymptomatic parasitemia are likely to depend on the underlying intensity of malaria transmission. Studies conducted in settings characterized by seasonal transmission12 or moderate transmission10 have shown associations between asymptomatic parasitemia, poor cognitive function, and educational outcomes. Data on the impact of malaria on cognition among schoolchildren living in high transmission settings are surprisingly limited; one study conducted in western Kenya showed that three rounds of intermittent preventive treatment (IPT) significantly improved classroom attention of schoolchildren.7 This work investigated the association between asymptomatic Plasmodium infection and cognitive function among schoolchildren in eastern Uganda, where malaria transmission is perennial and high, and asymptomatic Plasmodium infections are common.2–4
Methods
Study site.
The study was conducted in February 2011 in Mulanda primary school located in Mulanda sub-county, Tororo District, eastern Uganda. This area is mainly dry savannah grassland interrupted by bare rocky outcrops and lower lying swamps, although natural vegetation has mostly been replaced by cultivated crops. Tororo District is characterized by a high intensity of malaria transmission,14 with an estimated entomological inoculation rate of 562 infective bites per person per year.15 Transmission is bimodal, with peaks associated with the two rainy seasons. Malaria in the area is predominantly caused by Plasmodium falciparum3 and Anopheles gambiae s.s., and to a lesser extent, Anopheles funestus are the main vectors.15 In the 5 years before this study, malaria control in Tororo District was typically limited to the promotion of IPT during pregnancy, distribution of insecticide-treated nets (ITNs) through antenatal care services and malaria case management with artemisinin-based combined therapy. In January 2011, however, a mass community-based free ITN distribution campaign was conducted throughout the sub-county.
Mulanda sub-county has eight primary schools, all of which are under the government supported universal primary education scheme. Mulanda primary school was purposively selected for the study because of its large student population (1,320 students) and close proximity (approximately 500 meters) to the main public health facility in the area (Mulanda Health Center IV).
Study procedures.
This study reports baseline data collected as part of a randomized placebo-controlled trial investigating the impact of IPT on malaria morbidity and cognitive function in Ugandan schoolchildren (Clinicaltrials.gov identifier NCT01231880).
All children with parental consent were screened for eligibility to join the study. Children were excluded if they had any of the following: 1) known allergy or prior adverse reaction to artemisinin-based regimens; 2) history of menarche; 3) fever (axillary temperature ≥ 37.5°C) or history of fever in previous 24 hours; 4) evidence of severe malaria or danger signs; or 5) ongoing antimalarial treatment.
At enrollment, a standardized questionnaire was administered by study personnel to children fulfilling the selection criteria to record data on socio- demographics, bed-net ownership, and bed-net use. A focused physical examination was conducted that included a general and abdominal examination as well as measurement of temperature and weight. A finger-prick blood sample was obtained for thick and thin blood smears to assess for Plasmodium infection, hemoglobin estimation, and for filter paper storage. Stool samples were also collected.
Laboratory evaluations.
All blood smears were labeled and air dried at the school and subsequently stained with 2% Giemsa for 30 minutes at the health facility at the end of each day. Parasite densities were determined from thick blood smears by counting the number of asexual parasites per 200 white blood cells (or per 500 if the count was < 10 parasites/200 white cells), assuming a white blood cell count of 8,000/mL. A smear was considered negative after reviewing 100 high-powered fields. Asymptomatic malaria was defined as a positive blood smear for Plasmodium parasites with no associated clinical symptoms. All positive thick blood smears had corresponding thin smears viewed for species identification. Gametocytaemia was also determined from thick blood smears. Two independent microscopists read the slides, with a third microscopist resolving discrepant results. Hemoglobin concentration was assessed using a portable hemoglobinometer (HemoCue Ltd., Angelholm, Sweden) and estimated to an accuracy of 1 g/dL. Stool samples were examined microscopically for the eggs of intestinal nematodes and Schistosoma mansoni (the sole cause of schistosomiasis in the study area), using the Kato-Katz technique.
Cognitive function testing.
Two aspects of cognition were assessed, including sustained attention assessed by a code transmission test and abstract reasoning tested using Ravens matrices. These two tests have been adapted for an African setting and used in a number of previous studies.7,16 The test for sustained attention was adapted from the Tests of Everyday Attention for Children (TEA-Ch)17 and was administered in groups of 15 or less in the local language, Japhadhola, which is also the language of school instruction for children in classes 1–3 in Mulanda sub-county. The test involved listening to a tape and identifying different numbers (code transmission). During the code transmission task, a list of digits was read out aloud at the speed of one per second. Children were required to listen out for a code—“the number 5 repeated twice consecutively”—and then record the two numbers that preceded the code. The test was administered at school over the weekend when there were no active classes at the school to reduce on external interference. Only children invited for the test were present at school on the testing days. Before the test, children were given three warm-up activities to familiarize them with the tape recorder, as well as assess their ability to count and write numbers. During the test of abstract reasoning, a set of geometric figures with a missing pattern was presented to the children and they were asked to identify the missing item that completes a pattern from a group of almost identical alternatives.18
Statistical analysis.
All data were double-entered and cross-checked in a bespoke Microsoft access database (Microsoft Corp., Seattle, WA). Consistency checks were performed and all discrepancies and queries verified against original paper forms. Plasmodium infection was defined on the basis of expert microscopy results and the proportion of children with asymptomatic Plasmodium parasitemia was calculated as the number of children with any parasites (irrespective of species) on thick smear divided by the total number of children enrolled. Parasite density was categorized as above and below 1,000 parasites/μL. Anemia was defined using the World Health Organization (WHO) age-specific thresholds for hemoglobin (< 11.5 g/dL for children 6 to < 12 years of age and < 12.0 g/dL for those 12–14 years of age).19,20 The anthropometric index z-score weight-for-age was calculated using the egen Stata function for standardizing anthropometric measures in children and adolescents.21 Children were classified as underweight if they were less than two standard deviations below the reference mean. A household wealth index was created using principal component analysis of data on household possessions, utilities, and housing construction for each student.22 Households were ranked according to their distribution along the index, which was then divided into quartiles and classified as an ordinal variable for use in multivariate analysis models. Information on maternal education was also included in the analysis, because of its key influence on children's cognition.
All statistical analyses were carried out using Stata version 12.0 software (STATA Corporation, College Station, TX). The outcomes of interest were Plasmodium infection and scores in the code transmission and Raven's matrices tests. Ninety-five percent binomial confidence intervals (CI) were estimated for proportions and standard deviations presented for means. Univariable associations between Plasmodium infection and potential risk factors were assessed using logistic regression and all variables showing an association at a 20% significance level were included into a multivariable logistic regression model. Logical model building using both forward and backward elimination was used to generate minimum adequate model using a 5% significance level; however, bed-net use, socioeconomic group, and anemia were retained as fixed terms in the model regardless of statistical significance because of their known association to Plasmodium infection among school-aged children.3,23
To investigate the association between asymptomatic Plasmodium infection and cognitive function, two sets of analyses, one for abstract reasoning (score 0–20) and the second for sustained attention (score 0–20) were undertaken. The effect of explanatory variables was quantified by mean differences in test performance scores using univariable and multivariable linear regression. Case re-sampling bootstrapping was used to account for non-normality of the scores. Variables identified as significant (P < 0.2) in univariate analysis were considered for multivariable analysis. Logical model building using both forward and backward elimination was used to generate minimum adequate models; however, helminth infection, maternal education, and anemia were retained as fixed terms in all models regardless of statistical significance because of their known effect on cognition.24,25 Interaction was assessed on the basis of likelihood ratio test and included in the final model if P ≤ 0.05. Sensitivity analysis explored the influence of parasite density on cognition, with densities categorized as uninfected, infected with 1–999 parasites/μL, and infected with ≥ 1000 parasites/μL.
Ethics.
The study protocol was approved by the Makerere University School of Medicine Research and Ethics Committee (#2010-016) and the Uganda National Council of Science and Technology (#HS 865). Before the start of the study, investigators met with elected government representatives and community leaders to inform them of the study and explain the methodology. Written informed consent was obtained from the parents/guardians of all the children included in the study and written assent was obtained for children 8 years of age and above.
Results
A total of 740 children 6 to 14 years of age were enrolled (Table 1). There were slightly more girls (53.2%) than boys, and the mean age (SD) of the children in the study was 9.8 (2.2) years. The overall prevalence of Plasmodium infection was 30.1% (223 of 740 (95% CI, 29.4–30.7]) with P. falciparum the most prevalent species at 91.5% (204 of 223). Most of parasite densities were relatively light, with 61.0% (136 of 223) infected children having parasite densities < 1000 parasites/μL. Overall, 89.5% (662 of 740) of children reported that their household owned a bed-net, but only 262 (39.6%) of these children reported sleeping under a net the previous night. The mean test score in the sustained attention test was 9.4, whereas the score in the abstract reasoning test was 8.2.
Characteristics of the 740 primary schoolchildren participating in the study
| Characteristic | N = 740 |
|---|---|
| Gender (n female, %) | 394 (53.2) |
| Mean age (SD, years) | 9.8 (2.2) |
| Mean weight (SD, kg) | 28.5 (8.2) |
| WAZ (z-scores < −2) (n, %) | 105 (14.2) |
| Household (HH) owning at least one bed-net (n, %) | 662 (89.5) |
| Bed-net use in last 24 hours (n, %) | 262 (35.4) |
| Asymptomatic Plasmodium infection (n, %) | 223 (30.1) |
| Plasmodium falciparum | 204 |
| Plasmodium malarie | 11 |
| Plasmodium ovale | 4 |
| Plasmodium vivax | 4 |
| Gametocytes present (%) | 22 (3.0) |
| Helminths present (%) | 56 (7.6) |
| Ascaris lumbricoides | 38 |
| Hookworms | 15 |
| Trichuris trichiura | 3 |
| Mean hemoglobin (g/dL, SD) | 11.7 (1.0) |
| Household > 500 m from health facility (n, %) | 666 (90.0) |
| Mean score sustained attention test (SD) | 9.4 (3.4) |
| Mean score abstract reasoning test (SD) | 8.2 (2.6) |
Factors associated with Plasmodium infection.
Male sex, younger age group (6–10 years), and being underweight were significantly associated with Plasmodium infection (Table 2). In multivariable analysis, Plasmodium infection differed significantly by age group with a 71% (95% CI: 57–80%, P < 0.001) reduction in the odds of infection in children aged 11–14 years compared with younger children. Girls had significantly lower odds of infection than boys (odds ratio [OR] = 0.70, 95% CI: 0.50–0.98) and children who were underweight were 1.61 (95% CI: 0.71–1.41.) times more likely to have Plasmodium infection than children with normal weight. Interestingly, bed-net use was not significantly associated with lower odds of Plasmodium infection (OR = 0.99, 95% CI: 0.70 –1.39).
Univariate and multivariable analysis of factors associated with asymptomatic Plasmodium infection among the 740 primary schoolchildren in Tororo, Uganda*
| Variables | n/N with Plasmodium infection | Univariate analysis | Multivariable analysis | ||
|---|---|---|---|---|---|
| Crude odds ratio (95% CI) | P value | Adjusted odds ratio (95% CI) | P value | ||
| Sex | |||||
| Male | 117/346 | ||||
| Female | 106/394 | 0.72 (0.53–0.99) | 0.032 | 0.70 (0.50–0.98) | 0.040 |
| Age group (years) | |||||
| 6–10 | 175/448 | ||||
| 11–14 | 48/292 | 0.31 (0.21–0.44) | < 0.001 | 0.29 (0.20–0.43) | < 0.001 |
| WAZ (z-scores) | |||||
| Not wasted | 182/635 | ||||
| Wasted | 41/105 | 1.59 (1.04–2.45) | 0.033 | 1.61 (1.02–2.53) | 0.040 |
| Bed-net use in last 24 hours | |||||
| No | 139/478 | ||||
| Yes | 84/262 | 1.15 (0.83–1.59) | 0.398 | 0.99 (0.71–1.41) | 0.987 |
| Helminth infection | |||||
| None | 204/684 | ||||
| Ascaris lumbricoides | 11/38 | 0.95 (0.47–1.96) | 0.908 | ||
| Hookworms | 6/15 | 1.56 (0.55–4.46) | 0.399 | ||
| Trichuris trichiura | 2/3 | 4.70 (0.42–52.1) | 0.207 | ||
| Anemia | |||||
| No | 132/420 | ||||
| Yes | 91/320 | 0.86 (0.63–1.19) | 0.380 | 0.89 (0.64–1.25) | 0.498 |
| Distance to HF | |||||
| ≤ 500 m | 21/74 | ||||
| > 500 m | 202/666 | 1.09 (0.64–1.87) | 0.729 | ||
| Socioeconomic group | |||||
| Poorest | 63/196 | ||||
| Poor | 66/241 | 0.79 (0.52–1.20) | 0.279 | 0.85 (0.55–1.31) | 0.457 |
| Less poor | 32/120 | 0.77 (0.46–1.27) | 0.303 | 0.80 (0.48–1.36) | 0.418 |
| Least poor | 62/183 | 1.08 (0.70–1.66) | 0.719 | 1.23 (0.79–1.94) | 0.347 |
| Maternal education | |||||
| None | 21/71 | ||||
| Primary | 187/622 | 1.02 (0.59–1.75) | 0.932 | ||
| Above primary | 15/47 | 1.12 (0.50–2.48) | 0.787 | ||
CI = confidence interval; WAZ = weight for age z-score; HF = health facility.
Factors associated with poor cognition.
In univariable analysis, Plasmodium infection, female gender, young age (6–11 years), socioeconomic group, and infection with either Ascaris lumbricoides or hookworm were associated with low scores in the abstract reasoning (Table 3). In multivariable analysis, children with Plasmodium infection scored significantly lower in the abstract reasoning test than uninfected children (mean difference −0.6 95% CI: −1.01 to −0.21). Sensitivity analysis of the influence of parasite density suggested that mean differences did not increase with increasing parasite density: 8.5 (SD = 2.8) among uninfected children; 7.5 (2.0) among children with 1–999 parasites/μL; and 7.4 (2.1) among children with ≥ 1,000 parasites/μL. The multivariable analysis also showed that children infected with A. lumbricoides had lower scores than children uninfected, younger children (6–10 years) performed worse than older children (11–14 years), and scores improved with increasing socioeconomic status. Although hookworm infection and female gender were associated with lower scores in the abstract reasoning test in univariate analysis, these associations were not significant after adjusting for potential confounders.
Factors associated with poor performance in the abstract reasoning (Raven's test) among the 740 primary schoolchildren in Tororo, Uganda
| Variables | n | Mean score 1–20 (SD) | Un-adjusted Mean difference between test performance | P value | Adjusted mean difference between test performance | P value |
|---|---|---|---|---|---|---|
| Plasmodium infection status | ||||||
| Not infected | 517 | 8.5 (2.8) | ||||
| Infected | 223 | 7.5 (2.0) | −1.0 (−1.33, −0.67) | < 0.001 | −0.6 (−1.01, −0.21) | 0.001 |
| Sex | ||||||
| Male | 346 | 8.5 (3.1) | ||||
| Female | 394 | 7.9 (2.2) | −0.6 (−0.94, −0.21) | 0.002 | −0.2 (−0.61, 0.15) | 0.234 |
| Age group (years) | ||||||
| 6–10 | 448 | 7.4 (2.1) | ||||
| 11–14 | 292 | 9.4 (2.9) | 1.9 (1.59, 2.39) | < 0.001 | 2.4 (1.68, 3.10) | < 0.001 |
| WAZ (z-scores) | ||||||
| Not wasted | 635 | 8.23 (2.7) | ||||
| Wasted | 105 | 8.26 (2.4) | 0.03 (−0.45, 0.51) | 0.897 | 0.2 (−0.30, 0.61) | 0.613 |
| Bed-net use in last 24 hours | ||||||
| No | 478 | 8.3 (2.5) | ||||
| Yes | 262 | 8.1 (2.8) | −0.2 (−0.65, 0.23) | 0.347 | ||
| Helminth infection | ||||||
| None | 684 | 8.3 (2.7) | ||||
| Ascaris lumbricoides | 38 | 7.4 (2.1) | −0.9 (−1.66, −0.11) | 0.025 | −1.4 (−2.11, −0.74) | < 0.001 |
| Hookworm | 15 | 7.1 (2.0) | −1.2 (−2.18, −0.17) | 0.022 | −0.1 (−0.82, 0.64) | 0.810 |
| Trichuris trichiura | 3 | 8.0 (2.6) | −0.3 (−2.82, 2.19) | 0.809 | 1.6 (−3.04, 6.16) | 0.506 |
| Anemia | ||||||
| No | 420 | 8.3 (2.8) | ||||
| Yes | 320 | 8.1 (2.5) | −0.2 (−0.61, 0.13) | 0.203 | −0.01 (−0.44, 0.41) | 0.954 |
| Distance to HF | ||||||
| ≤ 500 m | 74 | 8.5 (2.6) | ||||
| > 500 m | 666 | 8.2 (2.6) | −0.3 (−0.93, 0.29) | 0.304 | ||
| Socioeconomic group | ||||||
| Poorest | 196 | 7.7 (2.3) | ||||
| Poor | 241 | 8.1 (2.5) | 0.4 (−0.02, 0.87) | 0.059 | 0.3 (−0.12, 0.69) | 0.162 |
| Less poor | 120 | 8.4 (2.8) | 0.7 (0.13, 1.31) | 0.017 | 0.7 (0.17, 1.25) | 0.010 |
| Least poor | 183 | 8.8 (3.1) | 1.1 (0.60, 1.65) | < 0.001 | 1.03 (0.51, 1.54) | < 0.001 |
| Maternal education | ||||||
| None | 71 | 8.3 (2.7) | ||||
| Primary | 622 | 8.2 (2.6) | −0.1 (−0.73, 0.44) | 0.623 | −0.2 (−0.83, 0.26) | 0.301 |
| Above primary | 47 | 8.7 (3.3) | 0.4 (−0.70, 1.47) | 0.487 | −0.1 (−1.14, 0.99) | 0.892 |
Univariate and multivariable mean differences in test scores with 95% confidence intervals (CI).
Age confounded Plasmodium infection and helminth infection. Socioeconomic group confounded of weight for age z-score (WAZ) scores, helminth infection status and maternal education.
Interaction between age and sex, as well as sex and helminth infection.
Results for the analysis of associations between sustained attention and several child level variables are presented in Table 4. In both univariable and multivariable analyses, Plasmodium infection was significantly associated with lower scores in the sustained attention test (adjusted mean difference: −1.6 CI −2.49 to −0.81). Sensitivity analysis indicated that mean scores decreased with increasing parasite density: 9.8 (SD = 3.5) among uninfected children; 8.7 (3.1) among children with 1–999 parasites/μL; and 7.7 (2.3) among children with ≥ 1,000 parasites/μL. The adjusted mean difference was −1.4 (−2.24, −0.60) and −1.9 (−1.23, −0.90), respectively. In addition, younger age (6–10 years) was significantly associated with lower mean scores compared with the older age (11–14years) in this test. Although girls had lower scores than boys in univariate analysis, this association was not significant after multivariable analysis. Scores in the sustained attention test was only marginally associated with socioeconomic status. Infection with helminths, anemia, and maternal education were found not to be associated with scores in the sustained attention test.
Factors associated with poor performance in the sustained attention (code transmission test) among the 740 primary schoolchildren in Tororo, Uganda*
| Variables | n | Mean score 1–20 (SD) | Unadjusted mean difference between test performance | P value | Adjusted mean difference between test performance | P value |
|---|---|---|---|---|---|---|
| Plasmodium infection | ||||||
| Not infected | 517 | 9.8 (3.5) | ||||
| Infected | 223 | 8.3 (2.8) | −1.5 (−2.01, −0.99) | < 0.001 | −1.6 (−2.40, −0.81) | < 0.001 |
| Sex | ||||||
| Male | 346 | 9.6 (3.6) | ||||
| Female | 394 | 9.1 (3.1) | −0.5 (−1.04, −0.03) | 0.038 | −0.5 (−1.20, 0.13) | 0.115 |
| Age group (years) | ||||||
| 6–10 | 448 | 8.4 (2.8) | ||||
| 11–14 | 292 | 10.8 (3.6) | 2.4 (1.88, 2.83) | < 0.001 | 1.6 (0.94, 2.22) | < 0.001 |
| WAZ (z-scores) | ||||||
| Not wasted | 635 | 9.4 (3.4) | ||||
| Wasted | 105 | 9.0 (3.2) | −0.4 (−1.11, 0.21) | 0.186 | −0.2 (−0.76, 0.34) | 0.461 |
| Bed-net use in last 24 hours | ||||||
| No | 478 | 9.5 (3.5) | ||||
| Yes | 262 | 9.1 (3.1) | −0.4 (−0.92, 0.06) | 0.089 | ||
| Helminth infection | ||||||
| None | 684 | 9.4 (3.4) | ||||
| Ascaris lumbricoides | 38 | 8.8 (2.4) | −0.6 (−1.29, 0.07) | 0.080 | −0.2 (−0.87, 0.48) | 0.573 |
| Hookworms | 15 | 7.8 (3.2) | −1.6 (−3.18, −0.04) | 0.044 | −0.7 (−2.51, 1.08) | 0.437 |
| Trichuris trichiura | 3 | 9.6 (2.3) | 0.2 (−2.14, 2.53) | 0.873 | 1.9 (−0.03, 3.81) | 0.047 |
| Anemia | ||||||
| No | 420 | 9.5 (3.4) | ||||
| Yes | 320 | 9.2 (3.3) | −0.30 (−0.67, 0.15) | 0.191 | −0.1 (−0.54, 0.36) | 0.709 |
| Distance to HF | ||||||
| ≤ 500 m | 74 | 9.48 (3.2) | ||||
| > 500 m | 666 | 9.40 (3.4) | −0.1 (−0.95, 0.79) | 0.855 | ||
| Socioeconomic group | ||||||
| Poorest | 196 | 8.9 (3.4) | ||||
| Poor | 241 | 9.4 (3.2) | 0.4 (−0.16, 0.97) | 0.163 | 0.2 (−0.37, 0.75) | 0.513 |
| Less poor | 120 | 9.8 (3.7) | 0.8 (−0.01, 1.52) | 0.053 | 0.7 (−0.01, 1.49) | 0.051 |
| Least poor | 183 | 9.7 (3.5) | 0.7 (−0.09, 1.43) | 0.086 | 0.6 (−0.82, 1.19) | 0.088 |
| Maternal education | ||||||
| None | 71 | 9.8 (3.8) | ||||
| Primary | 622 | 9.4 (3.3) | −0.4 (−1.33, 0.52) | 0.386 | −0.4 (−1.14, 0.39) | 0.346 |
| Above primary | 47 | 9.7 (3.3) | −0.04 (−1.39, 1.32) | 0.957 | −0.3 (−1.38, 0.75) | 0.560 |
Univariable and multivariable mean differences in test scores with 95% confidence intervals (CI).
Age confounded Plasmodium infection and Helminth infections. Socioeconomic group confounded maternal education.
Interaction between sex and Plasmodium infection and sex with age.
Discussion
Targeting interventions to improve the educational performance of schoolchildren requires an understanding of factors affecting health and cognition, such as malaria. There is limited evidence on the impact of asymptomatic Plasmodium infection on cognition among school children and to our knowledge our study is one of the few that have investigated this relationship in a high malaria transmission setting in Africa. Results from this study show that asymptomatic Plasmodium infections are strongly associated with poor performance in tests of abstract reasoning and sustained attention. Interestingly, the adjusted mean difference is greater for sustained attention than abstract reasoning. Additionally, there was evidence of a dose–response relationship between parasite density and sustained attention, but not between parasite density and abstract reasoning. By contrast, the association of scores with socioeconomic status was greater for abstract reasoning than for sustained attention.
The observed association of asymptomatic Plasmodium infection and cognitive scores may be a consequence of repetitive infections having direct effect on cognition8,26,27; another mechanism that has been suggested in the literature that could explain cognitive impairment is through the effect of iron deficiency anemia28–31 after repeated episodes of asymptomatic infections. However, in our study, anemia was not associated with poor cognition, suggesting that the effects of the infection may be more direct, possibly involving an immunological pathway.32
Our results corroborate those from studies in Yemen10 and Mali,12 where parasitemic school-aged children performed worse than non-infected children in the tests of cognitive function, including attention, and from studies that report a cognitive improvement following malaria control.7,33 The findings also support studies, which showed that severe and/or cerebral malaria is associated with cognitive impairment.34–37 For example, John and others35found that cerebral malaria is associated with cognitive impairment at 2 years follow-up. Interestingly, Bangirana and others37 in a subsequent study noted that malaria neurological involvement results in behavior problems at 3 months follow-up, but only affected sustained attention minimally, suggesting cognitive impacts of malaria may be gradual with less effect in the short term compared with the long term. Such a suggestion may help interpret the observed dose–response relationship between parasite density and sustained attention scores, whereby denser (and hence assumed more chronic) infection were associated with lower scores.
It is additionally noteworthy that in our study there was evidence of an association between socioeconomic status and scores of abstract reasoning, however there was little evidence of an association between socioeconomic status and scores of sustained attention. Previous authors including John and others35 have suggested that attention measures tend to be more sensitive to proximal brain/behavior effects from malaria, whereas executive functions, such as abstract reasoning, are more affected by distal effects of poverty.
Our findings differ from recent findings from coastal Kenya16 where there was no observed association between P. falciparum and performance in the same tests of cognitive function as used in this study. It is possible that these differing results are a consequence of differences in underlying intensity of malaria: moderate transmission in coastal Kenya versus high transmission in Uganda. Another possible reason for the differing results is a stronger influence of poverty on cognitive outcomes in coastal Kenya, where poverty and educational indicators are the lowest in Kenya.38 In such a setting, any effect of Plasmodium infection on cognitive is overshadowed by distal effects of poverty.
The results also show potentially important variation in cognitive scores between age groups and by helminth infection status. The observed association between low scores in the abstract reasoning test and A. lumbricoides infection corroborates previous findings where infection was strongly related to cognitive function in schoolchildren.39–42 The effect on cognition could be explained by the parasites leading to under-nutrition and therefore poorer growth rates of the infected children.41,43,44 The lack of observed association between helminth infections and sustained attention may not necessarily reflect an absence of effect but may be caused by the cross-sectional design of the study, such that we were unable to capture the effects of chronic infection over an extended period.41
Our findings must be viewed in light of limitations inherent to a cross-sectional study design and issues related to determining causality. However, we believe that reverse causality is unlikely to explain our findings because cognitive ability is unlikely to lead to Plasmodium infection. A further problem with interpretation of these findings arises from the potential for confounding, as any association found in this study may have resulted from a codependence of cognitive function on a third variable. Our study was designed to minimize this problem by sampling a wide range of potential confounders including age, gender, helminth infection, anemia, maternal education, and socioeconomic status, which we controlled for in the final analysis.
In conclusion, our study results provide evidence of a strong association between asymptomatic Plasmodium infection and poor cognition among primary schoolchildren in a high transmission setting, with a greater apparent effect on sustained attention than on abstract reasoning. Future analyses of the main trial will report the impact of averting malaria on the health and cognition of these children.
ACKNOWLEDGMENTS
We are thankful for the hard work of all the field study staff; Pauline Amuge, Nandera Immaculate, Adriko Godfrey, Owino Daniel, Odoi Godfrey, Onyango Christine, Twaha Yiga, Asau Victor, and Sarah Nabwire. We also thank the primary school students who participated in the study, their teachers, the village leaders, and the health centre staff of Mulanda health centre IV. Special thanks also go to Philip Rosenthal and Grant Dorsey who provided comments on the manuscript and to the Health and Literacy Intervention (HALI) project, including Kate Halliday, Peggy Dubeck, and Matthew Jukes, in coastal Kenya for providing training in the administration of cognitive tests.
- 1.↑
Snow RW, Guerra CA, Noor AM, Myint HY, Hay SI, 2005. The global distribution of clinical episodes of Plasmodium falciparum malaria. Nature 434: 214–217.
- 2.↑
Nankabirwa J, Cundill B, Clarke S, Kabatereine N, Rosenthal PJ, Dorsey G, Brooker S, Staedke SG, 2010. Efficacy, safety, and tolerability of three regimens for prevention of malaria: a randomized, placebo-controlled trial in Ugandan schoolchildren. PLoS ONE 5: e13438.
- 3.↑
Pullan RL, Bukirwa H, Staedke SG, Snow RW, Brooker S, 2010. Plasmodium infection and its risk factors in eastern Uganda. Malar J 9: 2.
- 4.↑
Kabatereine NB, Standley CJ, Sousa-Figueiredo JC, Fleming FM, Stothard JR, Talisuna A, Fenwick A, 2011. Integrated prevalence mapping of schistosomiasis, soil-transmitted helminthiasis and malaria in lakeside and island communities in Lake Victoria, Uganda. Parasit Vectors 4: 232.
- 5.↑
Brooker S, Clarke S, Njagi JK, Polack S, Mugo B, Estambale B, Magnusen P, Cox J, 2004. Spatial clustering of malaria and associated risk factors during an epidemic in a highland area of western Kenya. Trop Med Int Health 9: 757–766.
- 6.↑
Lalloo DG, Olukoya P, Olliaro P, 2006. Malaria in adolescence: burden of disease, consequences, and opportunities for intervention. Lancet Infect Dis 6: 780–793.
- 7.↑
Clarke SE, Jukes MC, Njagi JK, Khasakhala L, Cundill B, Otido J, Crudder C, Estambale BA, Brooker S, 2008. Effect of intermittent preventive treatment of malaria on health and education in schoolchildren: a cluster-randomized, double-blind, placebo-controlled trial. Lancet 372: 127–138.
- 8.↑
Holding PA, Snow RW, 2001. Impact of Plasmodium falciparum malaria on performance and learning: review of the evidence. Am J Trop Med Hyg 64 (1–2 Suppl): 68–75.
- 9.
Kihara M, Carter JA, Newton CR, 2006. The effect of Plasmodium falciparum on cognition: a systematic review. Trop Med Int Health 11: 386–397.
- 10.↑
Al Serouri AW, Grantham-McGregor SM, Greenwood B, Costello A, 2000. Impact of asymptomatic malaria parasitemia on cognitive function and school achievement of schoolchildren in the Yemen Republic. Parasitology 121: 337–345.
- 11.
Thuilliez J, 2010. Fever, malaria and primary repetition rates amongst school children in Mali: combining demographic and health surveys (DHS) with spatial malariological measures. Soc Sci Med 71: 314–323.
- 12.↑
Thuilliez J, Sissoko MS, Toure OB, Kamate P, Berthelemy JC, Doumbo OK, 2010. Malaria and primary education in Mali: a longitudinal study in the village of Doneguebougou. Soc Sci Med 71: 324–334.
- 13.↑
Fernando SD, Fau-Rodrigo C, Rodrigo C, Fau-Rajapakse S, Rajapakse S, 2010. The ‘hidden’ burden of malaria: cognitive impairment following infection. Malar J 9: 366.
- 14.↑
Hay SI, Guerra CA, Gething PW, Patil AP, Tatem AJ, Noor AM, Kabaria CW, Manh BH, Elyazar IR, Brooker S, Smith DL, Moyeed RA, Snow RW, 2009. A world malaria map: Plasmodium falciparum endemicity in 2007. PLoS Med 6: e1000048.
- 15.↑
Okello PE, Van Bortel W, Byaruhanga AM, Correwyn A, Roelants P, Talisuna A, D'Alessandro U, Coosemans M, 2006. Variation in malaria transmission intensity in seven sites throughout Uganda. Am J Trop Med Hyg 75: 219–225.
- 16.↑
Halliday EK, Karanja P, Turner LE, Okello G, Njagi K, Dubeck MM, Allen E, Jukes CHM, Brooker JS, 2012. Plasmodium falciparum, anaemia and cognitive and educational performance among school children in an area of moderate malaria transmission: baseline results of a cluster randomized trial on the coast of Kenya. Trop Med Int Health 17: 532–549.
- 17.↑
Manly T, Robertson IH, Anderson V, Nimmo-Smith I, 1999. Test of Everyday Attention for Children: TEA-Ch. Bury St. Edmunds, UK: Thames Valley Test Company.
- 18.↑
Raven JC, Styles I, Raven MA, 1998. Raven's Progressive Matrices. Oxford, UK: Oxford Psychologists Press.
- 19.↑
WHO, 2001. Iron Deficiency Anaemia: Assessment, Prevention and Control. A guide for programme managers. Geneva: World Health Organization.
- 20.↑
WHO 2008. Worldwide prevalence of anaemia 1993–2005. de Benoist B, McLean E, Egli I, Cogswell M, eds. WHO Global Database on Anaemia. Geneva: World Health Organization. 7 pp.
- 21.↑
Vidmar Suzanna, Carlin John, Kylie H, Tim C, 2004. Standardizing anthropometric measures in children and adolescents with new functions for egen. Stata J 4: 50–55.
- 22.↑
Filmer D, Fau-Pritchett LH, Pritchett LH, 2001. Estimating wealth effects without expenditure data–or tears: an application to educational enrollments in states of India. Demography 38: 115–132.
- 23.↑
Gitonga CW, Edwards T, Karanja PN, Noor AM, Snow RW, Brooker SJ, 2012. Plasmodium infection, anaemia and mosquito net use among school children across different settings in Kenya. Trop Med Int Health 17: 858–870.
- 24.↑
Nokes C, Grantham-McGregor SM, Sawyer AW, Cooper ES, Robinson BA, Bundy DA, 1992. Moderate to heavy infections of Trichuris trichiura affect cognitive function in Jamaican school children. Parasitology 104: 539–547.
- 25.↑
Falkingham M, Abdelhamid A, Curtis P, Fairweather-Tait S, Dye L, Hooper L, 2010. The effects of oral iron supplementation on cognition in older children and adults: a systematic review and meta-analysis. Nutr J 9: 4.
- 26.↑
Fernando D, Wickremasinghe R, Mendis KN, Wickremasinghe AR, 2003. Cognitive performance at school entry of children living in malaria-endemic areas of Sri Lanka. Trans R Soc Trop Med Hyg 97: 161–165.
- 27.↑
Fernando SD, Gunawardena DM, Bandara MR, De Silva D, Carter R, Mendis KN, Wickremasinghe AR, 2003. The impact of repeated malaria attacks on the school performance of children. Am J Trop Med Hyg 69: 582–588.
- 28.↑
Kurtzhals JA, Addae MM, Akanmori BD, Dunyo S, Koram KA, Appawu MA, 1999. Anaemia caused by asymptomatic Plasmodium falciparum infection in semi-immune African schoolchildren. Trans R Soc Trop Med Hyg 93: 623–627.
- 29.
Halterman JS, Kaczorowski JM, Aligne CA, Auinger P, Szilagyi PG, 2001. Iron deficiency and cognitive achievement among school-aged children and adolescents in the United States. Pediatrics 107: 1381–1386.
- 30.
Grantham-McGregor S & Ani C 2001. A review of studies on the effect of iron deficiency on cognitive development in children. J Nutr 131: 649S–666S; discussion 66S–68S.
- 31.↑
Agaoglu L, Torun O, Unuvar E, Sefil Y, Demir D, 2007. Effects of iron deficiency anemia on cognitive function in children. Arzneimittelforschung 57: 426–430.
- 32.↑
John CC, Panoskaltsis-Mortari A, Opoka RO, Park GS, Orchard PJ, Jurek AM, Idro R, Byarugaba J, Boivin MJ, 2008. Cerebrospinal fluid cytokine levels and cognitive impairment in cerebral malaria. Am J Trop Med Hyg 78: 198–205.
- 33.↑
Jukes MC, Pinder M, Grigorenko EL, Smith HB, Walraven G, Bariau EM, Sternberg RJ, Drake LJ, Milligan P, Cheung YB, Greenwood BM, Bundy DA, 2006. Long-term impact of malaria chemoprophylaxis on cognitive abilities and educational attainment: follow-up of a controlled trial. PLoS Clin Trials 1: e19.
- 34.↑
Carter JA, Mung'ala-Odera V, Neville BG, Murira G, Mturi N, Musumba C, Newton CR, 2005. Persistent neurocognitive impairments associated with severe falciparum malaria in Kenyan children. J Neurol Neurosurg Psychiatry 76: 476–481.
- 35.↑
John CC, Bangirana P, Byarugaba J, Opoka RO, Idro R, Jurek AM, Wu B, Boivin MJ, 2008. Cerebral malaria in children is associated with long-term cognitive impairment. Pediatrics 122: e92–e99.
- 36.
Vitor-Silva S, Reyes-Lecca RC, Pinheiro TR, Lacerda MV, 2009. Malaria is associated with poor school performance in an endemic area of the Brazilian Amazon. Malar J 8: 230.
- 37.↑
Bangirana P, Musisi S, Boivin MJ, Ehnvall A, John CC, Bergemann TL, Allebeck P, 2011. Malaria with neurological involvement in Ugandan children: effect on cognitive ability, academic achievement and behavior. Malar J 10: 334.
- 38.↑
RTI International, 2008. Early Grade Reading Kenya: Baseline Assessment: Analyses and Implications for Teaching Interventions. Final Report. Research Triangle Park, NC: RTI International.
- 39.↑
Sakti H, Nokes C, Hertanto WS, Hall A, Bundy DA, Satoto, 1999. Evidence for an association between hookworm infection and cognitive function in Indonesian school children. Trop Med Int Health 4: 322–334.
- 40.
Nokes C, McGarvey ST, Shiue L, Wu G, Wu H, Bundy DA, Olds GR, 1999. Evidence for an improvement in cognitive function following treatment of Schistosoma japonicum infection in Chinese primary schoolchildren. Am J Trop Med Hyg 60: 556–565.
- 41.↑
Jukes MC, Nokes CA, Alcock KJ, Lambo JK, Kihamia C, Ngorosho N, Mbise A, Lorri W, Yona E, Mwanri L, Baddeley AD, Hall A, Bundy DA, 2002. Heavy schistosomiasis associated with poor short-term memory and slower reaction times in Tanzanian schoolchildren. Trop Med Int Health 7: 104–117.
- 42.↑
Ezeamama AE, Friedman JF, Acosta LP, Bellinger DC, Langdon GC, Manalo DL, Olveda RM, Kurtis JD, McGarvey ST, 2005. Helminth infection and cognitive impairment among Filipino children. Am J Trop Med Hyg 72: 540–548.
- 43.↑
Stoltzfus RJ, Albonico M, Tielsch JM, Chwaya HM, Savioli L, 1997. School-based deworming program yields small improvement in growth of Zanzibari school children after one year. J Nutr 127: 2187–2193.
- 44.↑
Bellisle F, 2004. Effects of diet on behavior and cognition in children. Br J Nutr 92 (Suppl 2): S227–S232.