Changes in the Mean Birth Weight by Gestational Week at an Obstetric Service in Naha During the Past Twenty Years

In our previous report we pointed out, by analysing the official health statistics, that the mean birth weight of the Ryukyu Islands (Okinawa prefecture, Japan) during the past decade was significantly lower than that of the rest of Japan and that the annual increase rate of the mean birth weight during these periods was markedly high (the highest of all prefectures). In this report the authors attempted to clarify these remarkable phenomena which were recognised in the Ryukyus in the recent postwar period, from the birth records of a private obstetric service in Naha, by computing the mean birth weight and its standard deviation by sex, parity (primiparous and multiparous groups) and gestational age in weeks. This obstetric service was established in July 1952 and the birth records are available since January 1955. The minimum unit of the balance used in its delivery room is 50 grams. The number of the single live births subjected to this study accounted for 9, 080 cases. The other 2, 460 cases were excluded on account of the incomplete description on sex, parity or the commencing date of the last menstrual period, multiple births, fetal deaths, early neonatal deaths or major congenital malformations (Tab. 1-2). The gestational age in weeks was calculated in the nearest week (for example 277-283 days as 40 weeks). The results were as follows: 1) Of the 9, 080 single live births between 1955 and 1974, the mean birth weights at 9 gestational weeks (35-43 weeks) and their standard deviations were calculated by sex and by parity in each year (Tab. 3, Fig. 1). In spite of the annual fluctuations due possibly to the small size of the sample, the mean birth weights by gestational week appear to increase gradually in each sex and parity. 2) Then the samples were gathered from 4-three year periods (1957-59, 1962-64, 1967-69, 1972-74) for the enlargement of the sample size. As a result of the similar analysis significant improvements were recognized especially between 38-42 weeks (Tab. 4-5, Fig. 2). 3) At the same time the areal distribution of the mother, the mean gestational weeks, the mean maternal ages and the number of the parities were investigated in each of the 4 periods (Tab. 6-9). In addition the differences of the mean birth weight by gestational week between males and females, and those between primiparae and multiparae were tested (Tab. 10). 4) Based on these results discussion is made from the epidemiological point of view upon a: the relationship of