Background
: This study was designed to analyze the pattern and appropriateness of the choice of antigypertensive drug and laboratory tests in hypertension by comparison between residents in family medicine before and after 1988, and between residents in family medicine and in internal medicine after 1988.

Methods : We reviewed medical records in Seoul National University Hospital and evaluated the age, past medical history, initial BP of the patients, the items of laboratory tests, the timing of medication, and drugs prescribed for 48 and 51 hypertensive patients prescribed by residents in family medicine from 1985 to 1987(Group 1) and from 1989 to 1991 (Group 2) respectively, and 38 hypertensive patients prescribed by residents in internal medicine from 1989 to 1991(Group 3).

Results : 1) The drugs mainly prescribed were diuretics and adrenergic blockers in group 1, ACE inhibitors, adrenergic blockers, and diuretics in group 2, and adrenergic blockers, calcium antagonists, and diuretics in group 3 in order of frequency. The prescribing pattern made little difference according to age except that diuretics were not prescribed for patients under 50 years old in group 2 and group 3.
2) In group 1, diuretics and adrenergic blockers were mainly prescribed irrespective of past medical history, but various drugs were prescribed according to past medical history in group 2 and 3.
3) Antihypertensive drugs were prescribed at first visit in above 40% of hypertensive patients irrespective of the degree of hypertension.
4) Among 10 laboratory tests, there were statistically significant differences in fasting glucose(P<0.001) and serum calcium(P<0.001) between group 1 and 2, and serum potassium(P<0.05) between group 2 and 3.
5) The funduscopic examination was ordered more frequently in group 1 than in group 2, and in group 3 than in group 2. The differences were statistically significant(P<0.001).

Conclusion : After introducing the individualized-care approach for hypertensive patients, the initial prescribing pattern for hypertensive patient was individualized according to the patient's age, past medical history.

Background
: Hypertension and hyperlipidemia are well known risk factor in atherosclerosis. But few studies about the relationship between hypertension and hyperlipidemia are available. Therefore, we carried out this study to determine the relationship between hypertension and hyperlipidemia.

Methods : Data gathered on epidemiologic status, body mass index, fasting blood sugar, blood cholesterol and triglyceride of 970 hypertensive group and 2994 normotensive group during the period from March 1988 to December 1990. By using these data, we attempted to compare cholesterol and triglyceride value in hypertensive group with those in normotensive group.

Results : Age, body mass index and fasting blood sugar were statistically significant association with hypertension(P<0.01, P<0.01). The cholesterol and triglyceride values of hypertensive group were significantly higher than those of normotensive group(P<0.01, P<0.01). The cholesterol value in hypertensive group were closely related to the several demographic factors, such as age, body mass index and fasting blood sugar, by which 8.8% of cholesterol value are explained. The triglyceride value in hypertensive group were closely related to the several demographic factors, such as body mass index, fasting blood sugar and age, by which 11.0% of triglyceride value are explained.

Conclusion : In this study, hypertension was associated with hyperlipidemia. The cholesterol and triglyceride value in hypertensive group were closely related to the several demographic factors.

Background
: Mortality survey for cause of death is very important, but basic statistical results of those are few. This investigation was carried out in a province of Kyonsangpook-do to estimate the mortality rates among Koreans.

Methods : The base population are the beneficiaries of Korea Medical Insurance Corporation(KMIC), Kyonsangpook-do area, among which the 1,767 deaths occurred from January 1, 1991 to December 31, 1991, were reviewed to confirm the death. These were based upon the death certificates and medical utilization records before dying which were available through the computerized database on medical care utility of KMIC. The survey was conducted along three stages. At first, death certificates were existed, as a second step medical utilization records were reviewed, and finally direct contacts to the family members of the deceased were done.

Results : As a result, 1767 persons were died (male 946, female 816 unknown 5). Cause of deaths was accidents, malignant neoplasm, poisoning and injuries among age group 1-39 years. Over age group 40 years, the orders of cause of deaths were malignant neoplasm, cardiovascular disease, hypertensive cardiac disease, and chronic liver disease. Estimated crude rate were stomach cancer, cerebrovascular disease, hepatoma, lung disease in females.
Overall cancer mortality rate in the area was estimated 149.3 per 100,000 person-years in males, and 73.8 in females. Overall standardized mortality rate to the world standard population was estimated 180.1 per 100,000 person-year in males, and 66.2 in females.

Conclusion : This results shows the correct estimated mortality rates and various cause of deaths in a province of Kyonsangpook-do.

Background
: Patients known as 'difficult' who make doctors get negative fillings can be products of 'difficult' doctor-patient relationship. This study describes the reasons for difficulty, demographic characteristics, and contents of medical care of 'difficult' patients as compared with control patients of nine third-year residents in hospital-based family practice center.

Methods : Fifty-nine difficult patient sample was generated by asking residents to indicate patients whose care they considered difficult. And control sample was generated by random sampling after matching about duty doctor and being registered to Lifetime Health Maintenance Program in department of Family Medicine, Seoul National University Hospital. The control and difficult patient charts were audited and we compared two groups about various medical and demographic characteristics.

Results : Thirty-nine percent of difficult patients were selected due to medical problems and the others were selected due to psychosocial problems. There were no differences in demographic characteristics such as age, sex, marital status, educational level, and number of adults and children in family. Difficult patient group had more chronic problems, psychotropic drug use, provider continuity, and less total duration of care than control group. Two patient groups did not differ significantly in acute problems, chronic medications, samplings, radiological examinations, special tests, referrals, total visits to family practice center, and total duty doctors. After covariance analysis adjusting for total duration of care, significant differences remained between the two groups for chronic problems, psychotropic drug use. Also, after the covariance analysis the differences between difficult patient group and control group in number of acute problems and referrals became significant with the patient group having more acute problems and more referrals.

Conclusion : This study result suggest that residents experience a lots of difficulties in medical work-up and therapeutic plan of certain conditions, and that demographic characteristics and contents of medical care of difficult patients differ from previous studies.

Background
: Although headache is one of the most common complaints in primary care, the relationship between headache and depression and anxiety is controversial. Therefore we carried out a study to determine the relationship between heatache and depression, and between headache and anxiety.

Methods : Seventy new patients were selected from the patients who visited our family medicine clinic to receive periodic health examination from March 1 to July 31 1991. We took Zung's scales.

Results : There were significant differences in the 11 items of SDS among 20 items between patients and the control group. There were significant differences in the mean scores of SDS and SAS between patients and control groups when compared between males and females of the two groups(p<0.01, p<0.001 respectively). There were significant differences in the total mean scores of the SDS and SAS between patients and the control group(p<0.01, p<0.001 respectively).

Conclusion : There was a close relationship between headache and anxiety as well as with depression, so a screening test was needed to screen depression and anxiety in the patients with headache. Further studies, which can overcome for the limitation of this study, should be performed.

Background
: Death certificates are important data about death, which represent the changing pattern of disease and make it possible to compare the health status among other groups. But often there are error occurring tendencies which are due to the inaccuracy of diagnosis and inaccuracy of filling up death certificates, and which reduce the validity of death statistics. We reviewed the death certificates for the purpose of getting information about the incidence and the types of errors.

Methods : we reviewed a sample of 1047 death certificates collected at the Department of Statistics in Feb 1991. Seven items based upon WHO criteria were checked in the review, and we compared the error rate among to geographical districts, medical specialties, hospital size and year in which the physician who completed the death certificate obtain his/her medical license.

Results : One or more errors were found in 56.4% of death certificates. Of the total number of errors, 39% were due to listing the mechanism of death(such as cardiac arrest or respiratory arrest) as the immediate cause of death, 18.2% due to failing to state the immediate cause of death, 9.2% due to listing nonspecific disease entities instead of specific medical diagnosis, 8.3% due to listing illogical relationship between causes, 7.4% due to failing to state the underlying cause of death, 7.1% due to reversing the immediate and underlying cause of death, 5.6% due to part I containing conditions not or mannerly contributing to death in addition to the underlying cause of death, 2.5% to failing to certain information about E codes, 1.3% to part II containing either an underlying cause of death or a complication of the cause of death, and 1.5% to contain only the mechanism of death. We could not determine the precise underlying cause of death in 10.8%. Furthermore, up to 13.2% of death certificates may have involved inappropriate selection for the underlying cause of death. There are no differences in errors among districts, specialties, and hospital size. There are slightly more errors of death certificates recorded by the physician who obtains their licenses before 1960.

Conclusion : The accuracy of death certificates was low. So we need more attention in filling up death certificates.

Background
: As there are many concerns about prevention and control of chronic disability disease, it is emphasized to enhance quality of life and to evaluate functional response of treatment. So we need some index, which is useful in outpatient field easily and quickly. Using index at outpatient field we can readily evaluate and treat chronic patient. In this study, we used COOP chart method for evaluation quality of life, and other functional status.

Methods : We selected chronic renal failure patients undergoing hemodialysis. From 1, September 1991 to 26, September 1991 94 patients surveyed with 7 scales of COOP chart score method. Data was analyzed with ANOVA and Pearson Correlation Coefficient by PC/SAS(sas institute, 1985).

Results : Patients completed COOP chart within 3 minutes and the mean scores of each COOP charts appeared poor functional status mean scored 4.26±0.95. There were significant correlations between the scale of quality of life and other scales of functional status(p<0.001).

Conclusion : Using such an index in our outpatient field we have many opportunities for understanding emotional, social, physical status and planning treatment, establishing patient and doctor relationship also analyzing of functional effects of different treatment.