Baseline data of reproductive system in the JPHC study. Japan Public Health Center-based Prospective Study on Cancer and Cardiovascular Diseases.

Hormonal status in the body is closely related to the occurrence of estrogen-related cancers. Baseline survey data about the female reproductive system in JPHC study showed different gynecological and gestational profiles in each study area. Late menarche (15-16 y/o) was characteristic in the rural areas. Earlier gestational age and larger number of children were also more common in the rural areas. Baseline survey data, including gynecological past history, frequency of examination for uterine cancers, and so forth, showed some profile of middle aged women in the different areas in Japan.


INTRODUCTION
Japanese show the low incidence of breast cancer, endometrial cancer, and ovarian cancer compared to the Caucasians. Duration and amount of estrogen level is considered to be a risk factor for these cancers 1). Baseline survey of the JPHC study partially clarified the factors which being related to the geographic differences of the above described cancers in the female reproductive system.

METHODS
Information about the female reproductive system was obtained from the female session of two kinds of self-administered baseline questionnaires. Questions about age of menarche, age of menopause, regularity and days of menstruation cycle, any history of hormone therapy, number of pregnancy and children, age of gestation and childbirth, breast feeding, participation to the uterine and/or breast mass screening program, and past history of gynecological diseases were asked.
Several questions in the questionnaires were different in Cohort I and II. In terms of hormone treatment, "Have you ever used female sex hormones in the present or past?" in Cohort I was revised to "Have you ever used female sex hormones for contraception, dysmenorrhea or climacteric distress?" in Cohort II. Gynecological past history was selected from the list of diseases, such as "none, mastopathy, mastitis, mammary tumor, endometritis, uterine myoma, ovarian cyst, vaginitis, and others" in Cohort I, but vaginitis was deleted in Cohort II. The question about mass screening for uterine and mammary cancers was moved to the question for both sexes about an experience to participate to the mass-screening program performed by the local government, such as stomach, lung, colon, uterus and breast in Cohort II.

RESULTS
Age of cohort subjects was from 40 to 59 years in Cohort I and from 40 to 69 years in Cohort II. It may cause systematic bias about age of menarche and number of gestation and children due to the rapid change of society after the World War H.
Average age of menarche was 16.4 year and 15.0 year in both area of Okinawa, such as Miyako and Ishikawa, respectively. It was 13.2 years in Suita and 13.4 years in Katsushika (Table 1). In other areas, it was averagely 14 years. Age of menarche in metropolitan areas was earlier than other rural areas, and it was 2-3 years older in Okinawa. On the contrary, age of menopause did not show such clear trend, but almost 2 years difference was present in Cohort I (47.8 years) and Cohort II (49.4 years) ( Table 2).
Menstruation cycle was regular in about 80% women in all districts (Table 3a,         Only women who had ever experienced pregnancy Table 7. Number of child birth.

DISCUSSION
Active duration of estrogen secretion is closely related to th( occurrence of breast, endometrial, and ovarian cancer 1-3) Higher age of menarche among Japanese was considered to result in low incidence of estrogen-related cancers. Recent studies suggested that the dietary interaction like phytoestrogens modifies the effect of estrogen to reduce the risk of cancer and other gynecological distress [4][5][6] Proportion of many answers by women in Cohort I and I was different. The age of subjects in Cohort I was between age 40 and 59, while that in Cohort II was from age 40 to 69. This 10-year difference in age 60s reflected the dramatic change o: lifestyle among Japanese around the World War II. Number o gestation, pregnancy, and children had changed after the war Age of menarche and first pregnancy was different in rural and urban areas. Saku is known as an area of high education, so attitude of women seemed to be similar to those in urban city.
Mortality of uterine cervical cancer was steadly dropped since 1960. After the mass screening program was installed in 1983, high screening rate could decrease the mortality rate to 5/10,000 by early treatment 7). However, the participation rates to mass screening test was less than half in most areas. The education for hygiene and improvement of housing condition had contributed more to decrease the incidence rate. Rate of endometritis, vaginitis and uterine myoma in the past history varied much by area. The chance to go to gynecologists may be a factor to cause such difference, because the rural areas showed relatively high proportion to have had above diseases in the past history. Mammary tumor also showed the same trend, and only Suita subjects reported the tumor in more than 1%.
Cancer registration system has been made in each cohort area and works in high quality, so the risk and preventive factors could be more clearly shown based upon the histologically verified cases. Extension of the epidemiological study to osteoporosis and climacteric distress may be also possible in the JPHC Cohort in the future, if the proper diagnosis could be collected at the time of follow-up study.