Mothers tended to overreport categorical infant birth weight of their children

Objectives: We validated infant birth weight by interview with mothers. Study Design and Setting: Infant birth weights obtained from maternal interview were compared with the data documented in birth certificates. Study subjects were mothers of 1,432 schoolchildren who participated in a diabetes mass screening program in 1992-1997, Taiwan. Data of infant birth weight obtained from telephone interview with mothers were compared with those from birth certificates to evaluate the accuracy of maternal report in eight categorized groups: <2,000, 2,000-2,499, 2,500-2,999, 3,000-3,499, 3,500-3,999, 4,000-4,499, 4,500-4,999, and >= 5,000 g. Results: The exact agreement was as low as 15.9% but increased substantially to 67.7% if maternal reports of weight one category higher than birth registry weight were considered to be indicative of agreement. Lower orders of birth weight were significantly associated with reports of higher category. But teen mothers and/or low-income mothers were associated with lower risks of overreport. Conclusion: Our data suggested that birth weight-associated studies in Taiwan should "round number" to an upper category rather than a lower category to avoid serious misclassification in birth weight. (C) 2006 Elsevier Inc. All rights reserved.