To examine relationship between the family composition and medical expenses for circuration diseases in the elderly people who were 65 or more, we analysed the 1982 and 1983 data of medical expenses of national health insurance in the rural area of Fukuoka Prefecture.
The elderly people were classified into five groups according to whether they were living with their children and whether their spouse was alive.
The results were as follows:
1) In the elderly people who were living with their children, the medical expenses were low among those living with their spouses and high among those living without.
2) In the elderly people who were not living with their children, the medicalexpenses were low among the elderly men with their spouses and high among the eldrly women without.
3) In the elderly people who were living alone, the elderly men and women had low medical expenses in hospital services and ambulatory services.
4) The medical expenses for circulation diseases were more influenced by the presence of spouses than living with their children.

We have been delivering health care to the residents and investigating about their health administration in Yasu-machi of Fukuoka prefecture, since 1982. This study has suggested that the results of three examinations are useful to establish in primary health care of the rural area.
First, examination was carried out to evaluate their health consciousness and their attitude about treatment in 1982. Then we conducted their health needs from the data, which were gathered from questionaires of 1259 adults who were between 20 and 74 years of age.
Second, health statuses of 498 adults in objective group were certificated by medical examination including the data used for electrocardiograph and ophthalmofundoscope.
Third, we investigated both the case history and the medical cost utilizing the details of the medical expenses of national health insurance in the all area (26, 103 cases in 1982 and 27, 437 cases in 1983). For these data, we examined the difference in medical costs before the Health Service Law for the Aged was enforced and the medical costs after that. Further correlations between medical expenses and social factors such as family pattern and spouse presence were devised.
We conclude that the results of these three examinations provide basal materials which are useful to proceed further with the policy of primary health care in this area.