Description of the condition
Anaemia is a public health problem that affects people worldwide. Between 1993 and 2005, an estimated 1.62 billion people worldwide had anaemia, which corresponded to 24.8% of the world's population (WHO 2008). The reported global prevalence was 47.4% in children aged under five. Children in Africa and South-East Asia carried the highest reported burden of anaemia: 67.6% and 65.5% respectively (WHO 2008). Causes of anaemia are multifactorial and include poor nutritional status, micronutrient deficiencies (especially iron deficiency, but also vitamin A, vitamin B, and folic acid), intestinal helminth infection, HIV infection and haemoglobinopathies. However, malaria is probably the most important cause of anaemia in malaria-endemic countries (Antony 2008; Balarajan 2011; Crawley 2004). Anaemia is also more common in children from low income and illiterate families, compared to those coming from wealthier households (Balarajan 2011).
Malaria causes anaemia mostly by destruction of red blood cells (haemolysis) (Looareesuwan 1987) but also by causing an increase in the splenic pool of red blood cells and decreased production of red blood cells (Crawley 2004; Phillips 1992). Acute loss of red blood cells may lead to severe anaemia. Chronic anaemia can slow growth and result in learning difficulties and behavioral changes in affected children (Lozoff 1991; Grantham-McGregor 2001).
The symptoms of anaemia vary according to the severity, the age of the affected person, and whether the anaemia is acute or chronic. People with anaemia report fatigue, shortness of breath and palpitations. Clinical signs include paleness of the mucosal linings, such as the tongue, conjunctiva, palm and nail bed (Kalter 1997). Although palm pallor is commonly used for classification of disease in children (Meremikwu 2009), diagnosis of anaemia is based on laboratory tests. The World Health Organization (WHO) has defined anaemia in pre-school aged children as a haemoglobin (Hb) concentration of less than 11 g/dL (WHO 2008) and severe anaemia, often a complication of severe malaria, as a Hb concentration of less than 5 g/dL (WHO 2000). Researchers estimate that severe anaemia probably accounts for more than half of all childhood deaths from malaria in Africa (Crawley 2004). Children who are affected may need to be admitted to hospital and may need blood transfusions (Obonyo 2007).
Antimalarial interventions like insecticide residual spraying (IRS) and insecticide-treated nets (ITN) can be useful in preventing anaemia. A recent Cochrane review has shown that ITN use was highly effective in reducing childhood mortality and morbidity from malaria and had a positive effect on anaemia in children (Lengeler 2009). These vector control strategies are a core component of the malaria control programmes globally and especially in Africa (WHO 2012a). Other measures to prevent anaemia include prompt and effective treatment of malaria infections, intestinal helminths and HIV, increased use of measures to prevent mother-to-child transmission of HIV and provision of micronutrient supplementation (Balarajan 2011; Crawley 2004).
Description of the intervention
The intervention this review is considering is a short course of intermittent presumptive treatment in children who are anaemic. The aim of the intervention is to give them protection from malaria and allow their haemoglobin to recover to normal values.
Intermittent preventive treatment (IPT) is the administration of a full course of antimalarial treatment to a population at risk of malaria during a specific time period, regardless of whether or not they are known to be infected (Greenwood 2006). IPT policies were first implemented in pregnant women (IPTp) living in areas with a high rate of seasonal malaria transmission. This treatment consisted of a single dose of sulphadoxine/pyrimethamine (SP) given two or three times during the pregnancy, and was introduced as an alternative to chemoprophylaxis with chloroquine (CQ), due to the increasing CQ resistance and unpopularity of the drug (Greenwood 2010). The WHO also recommends that intermittent preventive treatment in infants (IPTi) up to the age of 12 months, should be administered together with the second and third diphtheria-pertussis-tetanus (DPT) and measles vaccination of infants in areas that have a moderate to high transmission rate of malaria (WHO 2010; WHO 2012a).
IPT was first made available for children (IPTc) after it had been shown that most children in highly seasonal malaria areas suffer from malaria and its related complications during the rainy season (Dicko 2011). Two recent systematic reviews have demonstrated that IPTc reduces episodes of clinical malaria in areas with a high rate of seasonal malaria transmission (Meremikwu 2012; Wilson 2011). Currently, the WHO recommends seasonal malarial chemoprevention (SMC) or IPTc, in seasonal malarial areas during the transmission season (WHO 2012b). This consists of a complete treatment course of SP and amodiaquine (AQ), given to children between 3-59 months, at monthly intervals, during the high risk period of malaria transmission. Children may receive up to four doses of this antimalaria treatment during the malaria transmission season with the aim of maintaining therapeutic drug levels during the period of high transmission. This strategy excludes areas with SP resistance outbreak (WHO 2013).
How the intervention might work
A recent Cochrane review reported that IPTc in community studies increased haemoglobin levels of children (Meremikwu 2012). They also concluded that there is moderate quality evidence that children given IPTc were less likely to have moderately severe anaemia at follow-up (Hb < 8 g/dL) compared to placebo (risk ratio (RR) 0.71, 95% confidence interval (CI) 0.52 to 0.98). There was, however, substantial variation between trials across 8805 participants making it unclear if this is a consistent effect. The review authors also found that substantively fewer children had severe anaemia (Hb < 5 g/dL) with IPTc (RR 0.24, 95% CI 0.06 to 0.84) but demonstrated no positive effect for mild anaemia (Hb < 11 g/dL) (Meremikwu 2012). It is unclear whether IPTc can effectively treat children that are identified as anaemic at a clinic or who are being discharged from hospital.
Treatment of anaemia in children often requires an integrated approach in order to address the multifactorial causes of the condition.
Children with severe anaemia, for whom routine management like blood transfusions and hematinics is insufficient to improve the haemoglobin level, might benefit from IPTc, since it has shown to augment the effect of hematinics on haemoglobin recovery when administered together in anaemic children (Akech 2008; Phiri 2011; Verhoef 2002). In addition, IPTc enables hematological recovery by preventing and treating new malaria infections (White 2004). Combining the effect of IPTc, ITN, and other programs like deworming and iron supplementation might add significant benefit in reducing the burden of anaemia in pre-school aged children. Iron supplementation is often recommended for children with anaemia, although there are concerns about an association between iron supplementation and increased malaria morbidity and mortality (WHO 2006). However, a recent Cochrane review concluded that there is high quality evidence that iron supplementation, even when given together with antimalarial treatment, does not increase the risk of clinical malaria morbidity or mortality (Okebe 2011).
Why it is important to do this review
The prevalence of anaemia in pre-school aged children remains high, especially in children living in Africa and South-East Asia. A published Cochrane review of IPTc in areas with seasonal transmission of malaria showed promising effects on prevention and treatment of anaemia in children (Meremikwu 2012). Although the review included all pre-school aged children living in malaria endemic regions, it did not examine the effects of IPTc on children diagnosed with anaemia.
Since the two systematic reviews on IPTc for malaria (Meremikwu 2012; Wilson 2011) have conflicting results on the effect of IPTc on anaemia, combining the results of existing studies systematically can provide physicians, policy makers and researchers with reliable evidence on the use of IPT in anaemic children living in malaria endemic areas with a high seasonal transmission rate.
To assess intermittent preventive antimalarial treatment for children with anaemia living in malaria-endemic areas.
Criteria for considering studies for this review
Types of studies
Randomized controlled trials and cluster randomized controlled trials.
Types of participants
Children aged below five years with anaemia (Hb < 11 g/dL; WHO 2008) living in malaria-endemic areas.
Types of interventions
Intermittent preventive treatment for malaria in children (IPTc).
No intermittent preventive treatment for malaria in children (IPTc).
Co-interventions, such as hematinics or insecticide treated nets, should be identical in both intervention and control groups.
Types of outcome measures
- Mean haemoglobin at follow-up (g/dL)
- Mean change in haemoglobin from baseline at follow-up
- Complete recovery from severe anaemia (Hb > 5 g/dL)
- Complete recovery from moderate anaemia (Hb > 8 g/dL)
- Blood transfusions
- All-cause mortality
- Admission due to severe anaemia
Search methods for identification of studies
We will attempt to identify all relevant studies regardless of the language and publication status (published, unpublished, in press and ongoing).
We will search the following databases using the search terms and strategy described in Appendix 1: Cochrane Infectious Diseases Group Specialized Register, Cochrane Central of Controlled Trials (CENTRAL), published in The Cochrane Library; MEDLINE; EMBASE; and LILACS. We will also search the WHO International Clinical Trial Registry Platform and metaRegister of Controlled Trials (mRCT) for ongoing trials using "anaemia", "children", "intermittent preventive treatment" and "malaria" as search terms.
Searching other resources
We will search the following conference proceedings for relevant abstracts: MIM conference abstract booklets (2005 and 2009);
ASTMH conference (2010 and 2011); and the European Congress of Tropical Medicine and Hygiene (2011).
Organizations and pharmaceutical companies
We will contact organizations, including the WHO, UNICEF and Centers for Disease Control and Prevention (CDC), for eligible studies. We will also contact pharmaceutical companies, including Novartis, Sanofi, Holley Cotec Pharmaceuticals, Roche, GlaxoSmithKline, Shangai Pharmaceutical Industries Company, Kinapharma Ltd. and Pfizer for ongoing and unpublished trials.
We will check the reference lists of all included studies for relevant studies.
Data collection and analysis
Selection of studies
Two authors (MA and ACR) will independently screen the results of the literature search for potentially eligible studies. We will retrieve full text articles of relevant studies and two authors (MA and AR) will independently assess eligibility using an eligibility form. We will contact authors in case of missing or unclear information. We will resolve discrepancies through discussion or alternatively through consulting the third author. We will ensure that multiple publications of the same trial are only included once. We will list excluded studies, together with the reasons for exclusion, in table format.
Data extraction and management
Two authors (MA and ACR) will extract data independently using pre-piloted, electronic data extraction forms. In cases of disagreement in the extracted data, we will resolve any disagreements through discussion. We will consult a third author if necessary (AMK). If data are missing, we will contact the study authors for clarification.
We will extract the following data: study details, including participant details, details regarding the intervention, details regarding the control intervention; outcome details, including which outcomes were measured, how they were measured; and results. For randomized controlled trials (RCTs), we will extract the number of participants randomized to each treatment arm and the number of participants monitored for each outcome of interest. For dichotomous data, we will extract the number of events in each of the treatment arms. For continuous data, we will extract the arithmetic mean, standard deviations and the number of participants in each group. For cluster RCTs that are adjusted for clustering,we will extract the measure of effect for each outcome (odds ratio, risk ratios (RRs) or mean difference values, with confidence intervals (CIs) or standard deviations).
For studies that are not adjusted for clustering, we will extract number of clusters randomized or the mean cluster size, the intra-cluster correlation coefficient, and the outcome data.
Assessment of risk of bias in included studies
Two authors (MA and ACR) will independently assess risk of bias for each included study by using the Cochrane Collaboration's tool to assess risk of bias (Higgins 2011). We will contact study authors in case of missing or unclear data. We will resolve discrepancies through discussion or consultation with the third author (AMK).
We will classify risk of bias judgements as either low, high or unclear risk of bias. We will assess the following components for risk of bias in each included trial as follows:
We will regard a study as having: low risk of bias if the sequence generation was truly random (for example, computer-generated table of random numbers, tossing a coin); high risk of bias if sequence generation contained a non-random component (for example, alternate randomization, randomization by birth date); or unclear risk of bias if the study authors did not clearly describe the randomization process.
We will regard studies as having: low risk of selection bias if allocation was truly concealed (for example, central allocation of participants, use of sequentially numbered, opaque, sealed envelopes); high risk of bias if the allocation process was not concealed (for example, open randomization, unsealed or non-opaque envelopes); or unclear risk of bias if the study authors did not describe the process of allocation concealment in sufficient detail.
Blinding of participants and personnel
We will determine whether blinding was present, who was blinded and the methods used to blind study participants and personnel. We will regard a study as having: low risk of bias if blinding was present, or if the absence of blinding was unlikely to affect the outcomes; high risk of bias if blinding was absent and likely to affect the results; or at unclear risk of bias if blinding was not clearly described.
Blinding of outcome assessors
We will describe whether blinding of outcome assessors was present and how they were blinded. We will regard a study as having: low risk of detection bias if they were blind to knowledge about which intervention the participants received; high risk of bias if blinding was absent; and unclear risk if blinding was not clearly described.
Incomplete outcome data
We will regard studies as having: low risk of attrition bias if there are no missing data or if missing data are balanced across groups; high risk of bias if there are missing data or if missing data are more prevalent in one of the groups; or unclear risk of bias if study authors do not clearly state whether outcome data is missing.
Selective outcome reporting
We will regard a study as having low risk of reporting bias if it is evident that all pre-specified outcomes were reported on; high risk of bias if it is evident that not all pre-specified outcomes were reported on; or unclear risk of bias if it is unclear whether all outcomes have been reported on.
We will describe any important feature of included trials that could have affected the result.
In addition to the above, we will assess the following for each included cluster RCT:
We will describe whether participants were recruited before or after randomization of clusters. We will regard studies as having low risk of recruitment bias if participants were recruited before randomization of clusters; high risk of bias if they were recruited after randomization; or unclear risk of bias if information about the timing of recruitment is unclear.
We will describe any baseline imbalances between individuals and clusters.
Loss of clusters
We will describe the number of clusters lost and reasons for attrition.
We will describe whether an analysis was adjusted for clustering.
Compatibility with RCTs randomized by individuals
We will describe whether the intervention effects differ systematically from individually RCTs (whether it was likely that the effect size was overestimated or underestimated).
Measures of treatment effect
We will compare dichotomous data using RRs. For continuous data summarized by arithmetic means and standard deviations, we will present mean difference values. For continuous data summarized by geometric means, we will report geometric mean ratios. We will present medians and ranges in table format. We will present all results with their associated 95% CIs.
Unit of analysis issues
When a multi-arm study contributes multiple comparisons to a particular meta-analysis, we will either combine treatment groups or split the 'shared' group as appropriate. We will take precautions to avoid inclusion of data from the same patient more than once in the same analysis.
If the included cluster RCTs have sufficiently accounted for the cluster design, we will include the effect estimates in the meta-analysis, combining them with individually randomized trials. If clustering has not been addressed, we will attempt to adjust the data for clustering by multiplying the standard errors by the square root of the design effect (Higgins 2011). We will then include the data in the meta-analysis.
Dealing with missing data
In case of missing data, we will apply available case analysis and only include data on the known results. The denominator will be the total number of participants who had data recorded for the specific outcome.
For outcomes with no missing data, we will undertake analyses on an intention-to-treat basis. We will include all participants randomized to each group in the analyses and we will analyse participants in the group to which they were randomized.
Assessment of heterogeneity
We will inspect forest plots for overlapping CIs and will assess statistical heterogeneity in each meta-analysis using the I² statistic and Chi² statistic. We will regard heterogeneity as moderate if I² statistic values are between 30% and 60%; substantial if they are between 50% and 90%; and considerable if they are between 75% and 100%. We will regard a P value of 0.10 or less indicative of statistically significant heterogeneity.
Assessment of reporting biases
If we include 10 or more studies in the meta-analysis, we will investigate reporting biases (such as publication bias) using funnel plots. We will assess funnel plot asymmetry visually and use formal tests for funnel plot asymmetry (Egger 1997 for continuous outcomes; Harbord 2006 for dichotomous outcomes). If we detect asymmetry in any of these tests or by a visual assessment, we will explore the reasons for asymmetry.
We will analyse the data using Review Manager 5 (RevMan) software (RevMan 2011). We will use fixed-effect meta-analysis to combine data statistically if heterogeneity is absent. If considerable heterogeneity is present, we will combine data using random-effects meta-analysis and we will report an average treatment effect if this is considered to be clinically meaningful. If the average treatment effect is not clinically meaningful, we will not combine data but instead present individual study results in a table.
We will combine results from cluster RCTs with individually RCTs if the study authors have adjusted for clustering in their analyses. We will present results using forest plots. If clustering is not adjusted for in the included RCTs, we will attempt to adjust data before combining it with data from individually RCTs. Alternatively, we will not included cluster RCTs that have not adjusted for clustering in the meta-analysis but present results in a separate table.
We will assess the quality of evidence using the GRADE approach (Guyatt 2011). We will rate each outcome as either high (we are very confident that the true effect lies close to that of the estimate of the effect); moderate (we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect);low (our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect); or very low quality of evidence (we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect) (Balshem 2011).
RCTs are regarded as high quality evidence but can be downgraded within the following five categories: risk of bias, imprecision, inconsistency, indirectness and publication bias. Studies can also be upgraded if there is a large effect, a dose-response effect, or if all plausible residual confounding would reduce a demonstrated effect or would suggest a spurious effect if no effect was observed (Balshem 2011). We will summarize our findings in a 'Summary of findings table'.
Subgroup analysis and investigation of heterogeneity
We will group the data by the drug used for IPTc. We will perform subgroup analyses if we detect substantial heterogeneity. We plan to carry out the following subgroup analyses:
- IPTi versus IPTc
- Additional interventions to treat anaemia (such as, hematinics or folic acid )
- The use of long lasting insecticide treated nets (LLINs) or not
We will perform subgroup analyses on primary outcomes, namely complete recovery from severe anaemia and complete recovery from moderate anaemia. We will assess differences between subgroups using the Chi
We will perform sensitivity analysis on primary outcomes to determine the effect if we exclude studies with high risk of bias (for allocation concealment and incomplete outcome data) on overall results; to assess the effect of excluding cluster-RCTs as the effects being evaluated might be different from individually RCTs; and to see what effect missing data has on results.
We thank Paul Garner, Taryn Young and Vittoria Lutje for their comments and support. The editorial base of the Cochrane Infectious Diseases Group is funded by UKaid from the UK Government for the benefit of developing countries.
Appendix 1. Search strategy
Contributions of authors
MA, AR and AMK developed the protocol. MA, ACR and AMK contributed to the protocol background and objectives. MA and ACR contributed to the protocol methods section.
Declarations of interest
The authors do not have any known conflicts of interest.
Sources of support
- Centre for Evidence-based Health Care, Stellenbosch University, South Africa.
- Ifakara Health Institute, Tanzania.
- Effective Health Care Research Consortium, UK.