Nutrient analysis using computer programs: comparison of a Canadian and an American database
1994
Barr, S.I. | Kwan, S. | Janelle, K.C.
This study was conducted to assess the comparability of results obtained when food intake records were analyzed using a Canadian and an American database. Two-hundred twenty-two diet records were obtained from 40 women aged 20-40 years, half of whom were vegetarians. Each record was coded and analyzed twice: initially using a Canadian database (derived from the Canadian Nutrient File, 1992), which contains variable proportions of missing values, and then using the ESHA II American database, which has 1% missing values. Differences in results appeared to be due to differences in nutrient content of similar foods in Canada and the U.S. and/or to the presence of missing values in the Canadian database. Results revealed: 1) for energy, proteins, total lipids, carbohydrates, cholesterol, calcium, vitamin C, pantothenate, sodium, potassium, and phosphorus, there were no differences in the results, or the differences were so small as to be of little biological relevance; 2) for thiamin, riboflavin, niacin and iron, results from the American database appeared to falsely overestimate Canadian values by 5-12%; 3) for nutrients such as pyridoxine, folacin, copper, magnesium and zinc, missing values in the Canadian database appeared to result in underestimates of 6-14%; and 4) for dietary fibre and vitamin E, the proportion of missing values in the Canadian database was so high that use of the results would be inappropriate. In conclusion, dietetic practitioners need to be aware of the advantages and limitations of using databases, be they Canadian or American. The appropriate choice will depend on the nutrients of greatest interest
显示更多 [+] 显示较少 [-]