Shimao Tadao
Japan Anti-Tuberculosis Association, Tokyo, Japan.
Kekkaku. 2009 Jan;84(1):23-9.
As mentioned in the previous report, PTB mortality by prefecture was printed in the "Statistics Annual", however, the population based on which the rate was calculated was not shown in the "Statistics Annual". In Japan, family register system was introduced in 1872, and every Japanese had to be registered at municipality where their family live. Based on this registry, statistics on registered population by sex and age, and by prefecture was available. In case when some family member(s) move to place other than registered place, he/she has to report to the local municipality in which they reside. When they move further to other place, the same procedure was required. Population based on this temporary registration system was named as A-type population, however, it did not show the population actually living in the municipality. Many persons newly register when they move in, however, forget to report to the previous residence when they move out. Overall numbers of move-in always larger than move-out. To adjust for this discrepancy and to estimate actually living population by each prefecture, the difference of move-in and move-out in a certain prefecture was multiplied by the ratio of total move-in and move-out in a certain prefecture and that of the whole country. Thus calculated population was designated as B-type population, and was proximate to actually living population in each prefecture. As B-type population was tabulated by sex in each prefecture, PTB mortality by sex could be calculated by using this B-type population, and the calculated figures were shown in Table 3 by sex for each prefecture. The correlation between PTB mortality by prefecture printed in the "Statistics Annual" and that calculated by using B-type population in 1886, 1892 and 1899 were shown in Fig. 4. Both coincide well nearly in all prefectures, except Tokyo and Hokkaido in 1886 and 1892, and in Tokyo in 1899. It was shown how difficult it was to estimate actually living population in an area where population move was very active, however, in general, in other prefectures both coincided well, and figures shown in Table 3 could be used to estimate PTB mortality in each prefecture by sex. Since 1899, ICD (International Classification of Diseases) was adopted in Japan as causes of death, and TB of all forms were divided into PTB, TB meningitis, intestinal TB and TB of other organs. As phthisis mortality was also available in 1899, correlation between PTB (Phthisis) mortality and that of TB of all forms was shown in Fig. 5 by age groups and by prefecture. The former was higher than the latter, and observing by age groups, correlation was not good in infants and elder population 60 years and over. In large cities like Tokyo and Oosaka, figures of the former were higher than the latter. It was indicated that it was not appropriate to combine and analyze the trend of PTB (Phthisis) mortality with the TB mortality statistics after 1899 adopting ICD.
如前一份报告所述，按县划分的肺结核死亡率刊登在《统计年报》中，但计算该比率所依据的人口数据并未在《统计年报》中显示。在日本，户籍制度于1872年引入，每个日本人都必须在其家庭居住的市町村进行登记。基于此登记册，可以获得按性别、年龄和地区划分的登记人口统计数据。如果某个家庭成员搬到登记地以外的地方，他/她必须向其居住的当地市町村报告。当他们进一步搬到其他地方时，也需要进行同样的程序。基于这种临时登记制度的人口被称为A型人口，但它并未显示实际居住在市町村的人口。许多人搬进来时会新登记，但搬出去时却忘记向前居住地报告。总体迁入人数总是多于迁出人数。为了调整这种差异并估计每个县的实际居住人口，将某个县的迁入和迁出差异乘以该县与全国的总迁入和迁出比率。这样计算出的人口被指定为B型人口，并且接近每个县的实际居住人口。由于B型人口按每个县的性别进行了列表，因此可以使用该B型人口计算按性别划分的肺结核死亡率，计算结果按性别显示在表3中每个县的数据。图4显示了《统计年报》中按县划分的肺结核死亡率与1886年、1892年和1899年使用B型人口计算的死亡率之间的相关性。除了1886年和1892年的东京和北海道以及1899年的东京外，几乎在所有县两者都非常吻合。这表明在人口流动非常活跃的地区估计实际居住人口是多么困难，然而，总体而言，在其他县两者吻合良好，表3中的数据可用于按性别估计每个县的肺结核死亡率。自1899年以来,日本采用国际疾病分类(ICD)作为死因，所有形式的结核病分为肺结核、结核性脑膜炎、肠结核和其他器官结核。由于1899年也有肺结核死亡率数据，图5按年龄组和地区显示了肺结核（痨病）死亡率与所有形式结核病死亡率之间的相关性。前者高于后者，按年龄组观察，在婴儿和60岁及以上的老年人口中相关性不佳。在东京和大阪等大城市，前者的数据高于后者。这表明将1899年采用ICD后的肺结核（痨病）死亡率趋势与结核病死亡率统计数据合并和分析是不合适的。
