Skip navigation

Iron deficiency

Bandolier was struck by an article in JAMA on the prevalence of iron deficiency in the USA, which hit its desk at the same time as a report called The Hunger Within on child poverty and nutrition in the UK. Naively, Bandolier thought that iron deficiency was of only historical interest in Britain, though important in developing countries with poor diets. Wrong again, apparently.

Iron deficiency in the USA

The US study [1] was conducted on just under 25,000 people aged one year or more as part of an ongoing health and nutrition survey. This collects data in household interviews and standard physical examinations in mobile examination centres. Biochemical information was available on 79% of the interviewed sample. Iron deficiency was defined as two or more abnormal results for free erythrocyte protoporphyrin, transferrin saturation or serum ferritin. Iron deficiency anaemia was defined as iron deficiency plus a low haemoglobin.


  • For children (ages 1 to 11 years), 9% of those aged 1 to 2 years were iron deficient, and 3% had iron deficiency anaemia.
  • For women of all ages the figures were 5 to 11% and 2 to 5% respectively.
  • Only men aged 50 years or more had significant rates of 2 to 4% and 1 to 2% respectively.

And the UK?

The Hunger Within [2] pointed out that iron deficiency and low haemoglobin in children has been shown to be common in the UK. Two papers were impressive [3,4].

The first of these [3] was a randomised trial of iron supplementation in children aged 17 to 19 months in Birmingham. Parents of every child in this age group in four health centres in central Birmingham were sent an invitation to attend for routine surveillance, including a haemoglobin estimation. Four hundred and seventy children attended. Of these, eight (2%) had haemoglobin levels below 80 g/L, and were treated immediately. A further 116 (25%) had haemoglobin values of 80 to 110 g/L, and were invited to attend an anaemia clinic.

In the clinic further clinical investigation was done, and eventually 110 children were randomised to iron 24 mg plus vitamin C 10 mg daily, or an identical vitamin C preparation, for two months. Assessments before and after included physical measures (height, weight), psychomotor development, and haemoglobin and other biochemical variables.


Iron supplementation over two months, compared with no iron supplementation, resulted in significant improvements. The number of children with a haemoglobin of at least 110 g/L was larger, with a number needed to treat (NNT) of 1.8 (1.4 to 2.6). The number of children increasing their psychomotor skills by six points (the average over this period in this age group) increased, with a NNT of 5.3 (2.9 - 33). The number of children whose weight velocity was 7 grams per day or above (the average for this age group) increased, with a NNT of 3.8 (3.2 to 14).

This beautifully presented paper puts these results into perspective, and summarises four previous studies of iron supplementation on psychomotor development in children. All involved small numbers of children, but results of iron supplementation were impressive in three of them. A systematic review of all these trials would make interesting reading.

How to do it

Bandolier is always attracted to papers which have the word implementation in their title. A study from Bristol [4] reported results from screening children aged 13 to 24 months, two years apart. The incidence of low haemoglobin (less than 105 g/L) in the first screen was 25%. After a combined approach by midwives, health visitors, treatment room nurses and the primary health care team to improve dietary advice, screening a similar cohort two years later reduced the incidence of low haemoglobin to 8%.

What was interesting was the simple way in which the haemoglobin screening had been included in an immunisation programme for measles, or measles, mumps and rubella. The uptake of immunisation was 93%, and that of screening 90%.


This paper is a nice example of a problem, with an effective treatment, and a way of doing it. But the authors themselves point out that by measuring haemoglobin and mean corpuscular volume they may be missing part of the problem.

Iron metabolism is a complicated business. Before frank anaemia develops with low haemoglobin, iron deficiency will occur, which may have problems of its own. How to measure iron deficiency is itself questioned, and it has been suggested that the US studies [1] may underestimate it.

In a closely argued but compelling study, a Swedish group [5] suggested using a serum ferritin cut-off of 16 ug/L. They found that 40% of 15-16 year old girls and 15% of boys were iron deficient, which is not out of line with studies in other countries.

Why should this be? Well, we lead low energy lives, and have a reduced food intake and different dietary balance from our ancestors. So we have less iron intake, and it may be less well absorbed. Bandolier would like to see a comprehensive review. Is there one we have missed?


  1. AC Looker, PR Dallman, MD Carroll, EW Gunter, CL Johnson. Prevalence of iron deficiency in the United States. Journal of the American Medical Association 1997 277: 973-6.
  2. SM Spiers, G Simmons. The Hunger Within. Milk for Schools, PO Box 412, Stafford ST17 9TF (fax +44 (0) 1785 248345), price £3.50.
  3. MA Aukett, YA Parks, PH Scott, BA Wharton. Treatment with iron increases weight gain and psychomotor development. Archives of Diseases in Childhood 1986 61: 849-57.
  4. J James, P Lawson, P Male, A Oakhill. Preventing iron deficiency in preschool children by implementing an educational and screening programme in an inner city practice. British Medical Journal 1989 299: 838-40.
  5. L Hallberg, L Hultén, G Linstedt et al. Prevalence of iron deficiency in Swedish adolescents. Pediatric Research 1993 34: 680-7.

previous or next story in this issue