The objective is to determine nutritional intake and vitamin status in a sample of homeless people who had been on the streets of Paris for more than 2 y. The nutritional status was evaluated by a 24 h recall questionnaire and by assay of circulating levels of vitamins A, B1, B6, B12, C and E. The study was conducted in four accommodation centres of the SAMU Social of Paris (CHUSI units) from July 1999 to May 2000.
A total of 329 homeless people were interviewed. Only 87 subjects, aged 26-76 y, were recruited based on seven selection criteria (over 18 y old, being homeless for more than 2 y, interviewed before taking a meal in the CHUSI units, good understanding of French, in a calm state, not coming from hospital or another CHUSI units, without any acute condition that might affect usual dietary habits), and completed the dietary questionnaire. Out of these 87, 71 agreed to blood sampling.
In ll 84% of the subjects declared that they drank alcoholic beverages regularly (175+/-167 g/day of alcohol), and 75.5% of the subjects also declared that they smoked. The mean total energy intake was 2111+/-1482 kcal/day for the men (8823.98+/-6194.76 kJ/day) and 1523+/-531 kcal/day for the women (6366.14+/-2219.58 kJ/day). Alcohol accounted for 44+/-30% and 19+/-17% of the total energy intakes (TEI) for the men and women, respectively. Proteins, lipids and carbohydrates accounted for 21+/-9, 23+/-14 and 57+/-16%, respectively, of the energy intake without alcohol for men, and 15+/-5, 22+/-11 and 62+/-14% EIWA for women. For all micronutrients, except for iron in the men, more than 50% of the population studied had intakes below the French recommendations for the adult population. All the women had calcium, vitamin E and vitamin B1 intakes below the recommendations for French women (900, 12 and 1.3 mg/day, respectively) in the general population. For the men, 98.7, 96.1 and 93.5% of the subjects had vitamin E, B1 and C intakes below the recommendations for French males in the general population (12, 1.5 and 80 mg/day, respectively). It is of note that subjects who ate in the CHUSI units more than once a day per week had significantly (P=0.018) higher intakes of vitamin C (27+/-38 mg/day) than did the subjects who ate in the CHUSI units less than once a day (8+/-12 mg/day). Measurement of serum vitamin levels demonstrated a deficiency in 95% of the subjects, mainly for vitamin C. The mean concentration in serum was 16+/-8 micromol/l, although 72% of the subjects had levels of 6 micromol/l. In contrast, 50.7, 42.2 and 35.2% of the subjects had high levels of vitamin B6, B12 and B1, respectively (100 nmol/l, 420 pmol/l and 40 micromol/l).
To conclude, alcohol takes a high place in the diet of these homeless people. Although the total energy intake was too low, macronutrient intakes seem to be satisfactory in quality (well-balanced). However, calcium and vitamin C intakes were too low, and these deficiencies favour the development of disease (scurvy, asthenia, ecchymosis), in addition to alcohol-linked disease. Furthermore, the overexposure to vitamin B6, among the other B vitamins, is a matter of concern since it has been shown to be toxic in high doses.