The review concluded that in two- to five-year-olds, daily iron supplementation increased haemoglobin and ferritin but there was a concerning lack of data on clinically important outcomes such as anaemia, iron-deficiency anaemia, iron deficiency and cognitive development. These conclusions highlight the limited clinical relevance of many of the trial results and appear likely to be reliable.

To evaluate the safety and benefit of daily iron supplementation with regard to haematologic, growth and cognitive parameters in two- to five-year-old children.

Several electronic databases, including MEDLINE and EMBASE, were searched to April 2012 with no language restrictions. A search strategy was reported as an appendix. Reference lists of relevant articles were searched.

Randomised or quasi-randomised controlled trials of oral iron supplementation (five or more days per week) for children aged two to five years were eligible. Studies of parenteral iron and of fortification were excluded. Studies of children with severe anaemia, children of low birth weight, pre-term infants and those with conditions affecting iron metabolism were excluded.

Iron was given as ferrous sulphate (between 10mg and 65mg) in around half of the studies (details were not reported in several studies). Some studies included cointerventions (such as vitamins, malaria treatment and anti-worm treatment). Most comparator treatments were placebo or vitamin supplementation. Treatment durations ranged from one to 15 months. Some children had anaemia, some did not and in some studies this was unclear. Most studies were performed in Africa or Asia. Studies were published between 1978 and 2004. The review authors prespecified seven primary outcomes.

Two reviewers screened full papers for inclusion. Disagreements were resolved by discussion.

Study quality was assessed using the Cochrane risk of bias tool. Studies were considered to be at low risk of bias if they were low risk for both selection and allocation bias and one of detection, performance and reporting bias.

The authors did not state how many reviewers performed the assessments.

Data were extracted in order to calculate mean differences (MD), standardised mean differences (SMD) or risk ratios, each with 95% confidence intervals (CI).

Two reviewers extracted data. Disagreements were resolved by discussion.

Meta-analyses were performed to calculate pooled effect estimates, with 95% confidence intervals, using a random-effects model. Statistical heterogeneity was assessed using the Ι² statistic. Eight subgroup analyses were prespecified. One subgroup analysis (total overall iron dose) was performed post hoc. The authors presented justification for incorporating uncorrected data from cluster randomised trials into the meta-analyses by referencing a study suggesting intra-class correlation coefficients for haemoglobin and ferritin were between <0.001 to 0.051.

Fifteen studies (4,149 patients, range 44 to 1,382) were included; three were cluster-randomised trials. All studies were rated as being at high risk of bias (most studies gave little or no information about randomisation and allocation concealment methods). One reference appeared to combine data from two separate trials.

Children who received iron supplementation had significantly more haemoglobin (MD 6.97g/L, 95% CI 4.21 to 9.72; nine studies, 12 treatment arms; Ι²=82%) and ferritin (MD 11.64µg/L, 95% CI 6.02 to 17.25; five studies, six treatment arms; Ι²=48%) than controls. Iron supplementation also significantly improved cognitive development (SMD 0.25, 95% CI 0.06 to 0.45; two studies) but not height or weight (seven studies each).

No trials reported the effects of iron supplementation on iron-deficiency or iron deficiency anaemia. One trial reported anaemia as an outcome, with no significant difference found between groups. There were no significant differences between groups for other haematologic parameters.

The effect on haemoglobin and ferritin was highly significant in children who were iron-deficient at baseline but was not significant in children who were iron-replete at baseline; this appeared to explain the statistical heterogeneity. Further subgroup analyses were reported.

In two- to five-year-olds, daily iron supplementation increased haemoglobin and ferritin. There was a concerning lack of data on the effect of iron supplementation on clinically-important outcomes including anaemia, iron-deficiency anaemia, iron-deficiency and cognitive development. Additional interventional studies in this age group are needed.

The review addressed a clear question and was supported by reproducible eligibility criteria. Attempts to identify all relevant studies in any language were undertaken by searching many electronic databases and by checking references. The authors noted that too few studies were available to allow a meaningful assessment of publication bias. It appeared that independent duplicate processes were used to reduce the risks of reviewer error and bias during the review; no process details were reported for the risk of bias assessments.

Risk of bias in the included trials was assessed adequately, with the results noted as being a limitation of the review findings. Sufficient study details were provided and appropriate methods were used to pool data and to assess and investigate heterogeneity.

The review findings had limited clinical relevance. The authors' conclusions reflected this appropriately and appear likely to be reliable.

Practice: The authors stated that where facilities to screen for anaemia or iron status are available, directed intervention with iron given exclusively to anaemic/iron deficient children is likely to achieve optimal benefit. They added that haemoglobin improvement may be optimal with 12.6mg to 30mg of iron daily over one to three months. Higher doses or longer treatment periods may be required for maximal improvement in ferritin. The review findings are relevant to lower- and middle-income countries.

Research: The authors stated a need for studies to evaluate clinically and functionally important outcomes, side-effects and adherence.

Government of Victoria, Australia; Royal Australasian College of Physicians; University of Melbourne, Australia.

This is a critical abstract of a systematic review that meets the criteria for inclusion on DARE. Each critical abstract contains a brief summary of the review methods, results and conclusions followed by a detailed critical assessment on the reliability of the review and the conclusions drawn.