한국의 노인인구 사망력추계 : 1970년, 1975년, 1980년

Abstract

[한글]

사망력 연구의 중요성은 보건학, 인구학적인 면에서 검토될 수 있다. 사망력 수준은 생명표를 통해 측정할 수 있으며, 이에 필요한 자료는 연령별 사망률이다. 그러나, 우리나라와 같은 개발도상 국가에서는 사망통계가 불충분하여 사망률 추정이 쉽지 않으며. 특히 사망질서가 급변하는 저연령층과 고연령층의 사망확률 추계가 문제되는데, 영유아 사망의 미비한 부분을 보정하는 많은 연구는 있었으나, 고연령층은 노인인구가 적고, 사망통계 자료가 미비하며, 불확실한 사인기록, 잘못된 연령기록 등으로 노인인구의 사망률 추정이 용이하지 않아 전체 평균여명 수준의 변동을 좌우할 수 있음에도 이에 대한 연구는

미비한 실정이다.

이 연구에서는 인구 및 주택 센서스보고, 인구동태통계의 65세이상 5세군 연령자료를 이용하였으며, Horiuchi 및 Coale모형에 따라 1970년, 1975년 및 1980년의 노인인구 사망력을 추계한 결과는 다음과 같다

1. 추정 사망률에서 사용되는 매개 변수값은 α=1.4, β=0.0951에서 매우 작은 비율의 오차를 나타냈으며, Horiuchi 및 Coale의 사망력 추계에 의한 한국의65세이상 노인인구의 사망률은 1970년, 1975년 및 1980년도에서 남자의 경우 0.15130, 0.12642 및 0.12942였고, 여자의 경우는 0.09688, 0.08417 및 0.09565였다.

2. 한국의 65세이상 노인인구의 65세 시점에서의 평균여명은 1970년, 1975년 및 1980년도에 남자의 경우 6.61세, 7.91세 및 7.73세였고, 여자의 경우는 10.32세, 11.88세 및 10.45세로 추계되었다.

3. Horiuchi 및 Coale방법에 의한 65세 평균여명과 한국인의 기존 생명표에서의 65세 평균여명을 비교하면 197O년 경재기획원 조사통계국의 보고보다 남자의 경우 3.0세, 여자의 경우 3.43세 낮은 수준을 보였고, 1975년 U.N.의 표준생명표에 의한 추계보다 남자의

경우 1.9세, 여자의 경우는 2.7세. 1980년 경제기획원 조사통계국의 65세의 평균여명보다는 남자의 경우 2.14세, 여자의 경우는3.72세의 낮은 수준을 보였다.

An Estimate of Old Age Mortality in Korea: 1970,1975 and 1980

Choi, Hye Lyung

Graduate School of Health Science and Management

Yonsei University Seoul, Korea

(Directed by Prof. Dongwoo Lee, M.Sc.)

From the point of health science and demography studies of mortality is

recognized an important subject. In most of countries, the level of mortality is

measured by constructing life tables based on converting age-specific death rates

into age-specific mortality rates. However, in the most of developing countries

age- specific death rates are uncertain because of the unavailability of reliable

mortality statistics. Specially, measuring early infant mortality and old age

mortality in developing countries is more difficult. The probability of dying at

early ages and old age are varied from time to time and errors in reporting ages

are often induced to estimate under or over estimates of the mortality concerned.

Recently, Horiuchi and Coale developed a new model of estimating old mortality

from the observed number of death at old ages.

Therefore, this study is aimed to measure levels of old mortality from 65 and

above by applying the Horiuchi and Coale's model for selected years in Korea.

Data used for this study is from the recent Korean national population and

housing censuses and the compiled statistics of vital registration published by the

National Bureau of Statistics, Korea Government.

Parameters and revels of Korean old population estimated by the model were as

follows:

1. From the Horiuchi and Coale's model two parameters, alpha and beta, estimated

for the Korean old population during the period of 1970 to 1980 1.4000 and 0.0951.

2. The estimated death rates by the model were, 0.15130 in 1970, 0.12642 in 1975

and 0.12942 in 1980 for male, and 0.09688 in 1970,0.08417 in 1975 and 0.09565 in

1980 for female.

3. The life expectancies at age 65 and over were, 6.61 for male and 10.32 for

female in 1970, 7.91 for male and 11.88 for female in 1975 and 7.73 for male and

10.45 for female in 1980.

4. Life expectancies estimated by the Horiuchi and Coale's model were compared

with other levels from life tables existing were, under estimated by 3 years for

male and 3.43 years for female from the life table of National Bureau of Statistics

in 1970, 1.9 years for male and 2.7 years for female from the standard life table

of the United Nations in 1975, and 2.14 years for male and 3,72 years far female in

1980 from the life table of National Bureau of Statistics.

[영문]

From the point of health science and demography studies of mortality is recognized an important subject. In most of countries, the level of mortality is measured by constructing life tables based on converting age-specific death rates

into age-specific mortality rates. However, in the most of developing countries age- specific death rates are uncertain because of the unavailability of reliable mortality statistics. Specially, measuring early infant mortality and old age

mortality in developing countries is more difficult. The probability of dying at early ages and old age are varied from time to time and errors in reporting ages are often induced to estimate under or over estimates of the mortality concerned.

Recently, Horiuchi and Coale developed a new model of estimating old mortality from the observed number of death at old ages.

Therefore, this study is aimed to measure levels of old mortality from 65 and above by applying the Horiuchi and Coale's model for selected years in Korea.

Data used for this study is from the recent Korean national population and housing censuses and the compiled statistics of vital registration published by the National Bureau of Statistics, Korea Government.

Parameters and revels of Korean old population estimated by the model were as follows:

1. From the Horiuchi and Coale's model two parameters, alpha and beta, estimated for the Korean old population during the period of 1970 to 1980 1.4000 and 0.0951.

2. The estimated death rates by the model were, 0.15130 in 1970, 0.12642 in 1975 and 0.12942 in 1980 for male, and 0.09688 in 1970,0.08417 in 1975 and 0.09565 in 1980 for female.

3. The life expectancies at age 65 and over were, 6.61 for male and 10.32 for female in 1970, 7.91 for male and 11.88 for female in 1975 and 7.73 for male and 10.45 for female in 1980.

4. Life expectancies estimated by the Horiuchi and Coale's model were compared with other levels from life tables existing were, under estimated by 3 years for male and 3.43 years for female from the life table of National Bureau of Statistics

in 1970, 1.9 years for male and 2.7 years for female from the standard life table of the United Nations in 1975, and 2.14 years for male and 3,72 years far female in 1980 from the life table of National Bureau of Statistics.