The purpose of the study was to assess the prevalence of Fe deficiency and Fe-deficiency anaemia in a group of apparently healthy adolescents, and to assess the value of a food frequency and amount questionnaire as a screening tool to identify children at risk of Fe deficiency. White schoolchildren (399) aged 12-14 years living in a Southwest London suburb completed a food frequency and amount questionnaire to assess usual Fe and vitamin C intake, and provided a thumb-prick blood sample for analysis of haemoglobin (Hb), packed cell volume (PCV), and serum ferritin (SF). Children were classified as 'anaemic' if H b was below the Dallman 3rd percentile (girls: < 120 g/l; boys: < 122, < 124 and < 126 g/1 at ages 12, 13 and 14 years respectively); and 'low' or 'borderline' in Fe stores if SF was < 12 pg/l, or between 12 and 20 pg/l respectively. Of the boys and girls 3.5 and 10.5 YO respectively were anaemic; 1 YO of boys and 4% of girls had low ferritin values, and 14% of boys and 16% of girls were borderline, Fe intakes were significantly higher in boys than in girls (12.3 Y. 9.6 mg/d, P < 0.001). Prevalence of anaemia was 143% in the group with both low Fe intakes (< lower reference nutrient intake) and low vitamin C intakes (< median), compared with 2.3 YO in the group with both high Fe intakes ( > reference nutrient intake) and high vitamin C intakes (> median). Anaemia was three times more common in vegetarians than omnivores (25 v. 9 %), and in girls who had tried to lose weight in the last year compared with those who had not (23 v. 7 %). The questionnaire did not prove satisfactory as a screening tool for risk of Fe deficiency. The higher-than-expected prevalence of Fe deficiency in apparently healthy white girls suggests that other groups should also be investigated.

The response to the covert removal of around 500 kcal/d from the food of 10 healthy free-living male volunteers was measured over 10 days. A similar amount of energy was added to the food of four subjects. Normal food and drink was available ad libitum, and adjustments in food energy were made by the interchange of sugar and artificial sweeteners. Compensation for the theoretical energy deficit or surplus was incomplete, averaging around 50%. This, however, was very variable, some subjects achieving more than 90% recovery, whereas others failed completely to adjust energy intake within this timespan. It is concluded that the regulation of energy balance in the lean is a long-term phenomenon, conditioned by large day-to-day fluctuations in energy intake. The removal of sugar from the food raised the energy density of the diet. The substitution of artificial sweeteners for sugar caused an increase of 11% in total fat intake. The restoration of sugar had an opposite effect. This inverse relationship between dietary sugar and fat poses problems for those seeking to lose weight, and for the lean attempting to follow current guidelines for a 'healthier' diet without incurring weight loss.