건강불평등측정 : 가치의문제

울산의대강영호

상대 vs 절대불평등

Relative Risk = Rate in Exposed / Rate in the Unexposed

Rate Ratio

Excess Risk = Rate in Exposed – Rate in the Unexposed

Risk Difference

예를들어,

3/100 vs. 6/100 : RR = 2, RD = 3/100

5/100 vs. 9/100 : RR = 1.8, RD = 4/100

우리나라 30-59세남성에서의교육수준별연령표준화사망률의변화(인구 10만명당)

1990-1991년 2000-2001년

초등학교이하 1383 1284

대졸이상 293 211

사망률비(초졸이하/대졸이상) 4.7 6.1

사망률차이(초졸이하 - 대졸이상) 1090 1073

자료원: Khang et al. J Epidemiol Community Health 2004 논문에서일부자료를발췌함.

유럽국가에서의 45-64세남성에서의직업계층간(육체직대비육체직) 사망률비와차이

국가 사망률비 사망률차이(100명당)

스웨덴 1.41 5.6

스페인 1.37 5.8

이탈리아 1.35 6.0

자료원: Mackenbach et al. Lancet 1997 논문에서일부자료를발췌함.

“… our data do not support the hypothesis that inequalities in

health are smaller in countries whose social, economic, and health

care policies are more influenced by egalitarian principles, such as

Sweden and Norway.

… our results challenge widely held views on the relations

between societal characteristics and the size of inequalities in

health.”

Mackenbach, et al. Lancet (1997)

Mortality for non-manual and manual

workers in nine European countries. Ranked by absolute level of mortality of manual workers; age groups 45–59.

Vagero & Erickson,

Lancet 1997

How to understand the large relative, but small

absolute inequalities in health in Nordic

countries?

“I wanted to raise the question of whether or not relative health inequalities

should be the yardstick against which social policies are evaluated. Ever since

the Black report there has been an assumption that the Nordic countries have, or

should have, smaller relative inequalities in health and mortality. While this

assumption seemed reasonable and in line with the Nordic countries having

lower income inequalities and poverty rates than other countries (as well as

being supported by some earlier studies), later and more comprehensive studies

did not find clearly lower relative inequalities in the Nordic countries. This, in

turn, sparked a discussion about relative versus absolute measures of inequalities.

However, any measure of inequalities, absolute or relative, will by definition

include the state of health among the privileged, and a positive development for

all social groups will often result in stable or even increasing inequalities.”

Olle Lundberg. Int J Epidemiol 2010

“Universal social policies, unlike means-tested ones, are designed for the

population at large and not only for the poorest. By including the broad middle

classes, universal programmes tend to be more sustainable, to have a higher

degree of generosity and a higher quality of services. While this will be

beneficial for the poorer segments of society, especially in comparison with

lean means-tested programmes, the middle classes are likely to benefit also. If

that is actually the case, the types of policies often associated with the social-

democratic type of welfare state will promote better health among lower and

middle classes alike, and hence better public health in general, but not

necessarily lower (relative) inequalities.”

Olle Lundberg. Int J Epidemiol 2010

사망률(인구 10만명당) 변화의가상적사례: 정책

시행전후사회계층별사망률의변화에따른사망률

차이와사망률비의양상

정책시행전사망률 사망률감소량 정책시행후사망률

높은사회계층 100 50 50

낮은사회계층 200 75 125

사망률차이 100 75

사망률비 2 2.5

Inverse Equity Hypothesis

“[F]ollowing the introduction of new

public-health interventions, inequities in

infant and child health status between

richer and poorer groups in society

usually widen before they get smaller and

improve. We have called this the “inverse

equity hypothesis”.

Victora et al. Lancet (2000)

Justification of relative measures

“In these analyses, we have used ratio measures of inequity, rather than

absolute or difference measures. Difference measures will almost

inevitably lead to an apparent reduction in equity gaps, because baseline

rates tend to be already low in absolute terms among the wealthiest.

Ratio scales, however, take baseline levels into account and are thus

more appropriate for deciding whether or not inequity is decreasing.

This is consistent with the use of ratio measures in epidemiological

research.”

Victora et al. Lancet (2000)

Do we need to implement public health programs

without concerns about increasing inequalities?

“The pessimistic conclusion is that public-health programmes specifically

targeted towards the poorest may not succeed in closing the overall gap in

child health within a reasonable time period. However, such programmes may

perhaps prevent the inequities from deteriorating even further.”

“We would argue that investment in public-health interventions is a priority in

order to prevent inequities becoming worse among poor people in the

developing world. … In most less-developed societies, the wealthiest are

likely to continue to benefit from the introduction of new health technologies.

Unless investment is also made to make existing and new interventions more

widely accessible to the poorest, inequity gaps may widen rather than be

reduced.”

Victora CG. Lancet 2000

불평등평가에서상대비가더중요한가?

사망자수

Time 1 Time 2

높은계층 (10,000명), A 100 90

낮은계층 (10,000명), B 200 185

상대비(배), B/A 2 2.06

절대차(명), B-A 100 95

생존자Time 1 Time 2

높은계층 (10,000명), B 1000 1100

낮은계층 (10,000명), A 500 575

상대비(배), B/A 2 1.91

절대차(명), B-A 500 525

상대 vs 절대지표의차이는다른보건문제들에서도나타난다

• “The RR of myocardial infarction in the hour after coffee

intake was 1.49 (95% CI = 1.17-1.89).”

– Baylinet al. Epidemiol.(2006)

• “Average 10-year risk of MI is 10%, so the 1-hour risk is

approximately 1/1,000,000. The RR of 1.5 above implies 1

extra hourly MI for each 2,000,000 cups of coffee.”

• “researchers should routinely provide this kind of information

to readers: the absolute differences that are implied by the

ratios of risks, rates, and prevalences we typically estimate.”

– Poole, Epidemiol.(2007)

“Yet the evaluative judgments made about the distributional

changes associated with globalization may depend crucially on

whether one thinks about inequality in absolute or relative terms.

There is no economic theory that tells us that inequality is relative,

not absolute. It is not that one concept is right and the other

wrong. Nor are they two ways of measuring the same thing.

Rather, they are two different concepts. The revealed preferences

for one concept over another reflect implicit value judgment

about what constitutes a fair division of the gains from growth.

Those judgments need to be brought into the open and given

critical scrutiny before one can take a well-considered position in

this debate.”

Martin Ravallion, Competing Concepts of Inequality

in the Globalization Debate 2004

사회계층(집단)의수와크기

2005년도새국민건강증진종합계획상의건강형평성목표

목표 2001년현황 2010년목표

사망률목표

소득수준별상위 20%와하위 20% 간사망률차이를 25% 감소시킨다.

2.07배 1.80배

건강행태목표

소득수준별상위 20%와하위 20% 간건강행태의차이를 25% 감소시킨다.

- 20세이상매일흡연율 8.8% 6.6%

- 20세이상적절운동실천율 10.4% 7.8%

자료원. 2005년도 보건복지부와 한국보건사회연구원이 발간한<새국민건강증진종합계획 수립> 보고서에서 발췌함.

1.001.19

1.491.75

2.07

0.0

0.5

1.0

1.5

2.0

2.5

1분위 2분위 3분위 4분위 5분위

1.001.14

1.371.56

1.80

0.0

0.5

1.0

1.5

2.0

2.5

1분위 2분위 3분위 4분위 5분위

1.001.19

1.491.75 1.80

0.0

0.5

1.0

1.5

2.0

2.5

1분위 2분위 3분위 4분위 5분위

(가)

(나)

소득상위 20%와하위 20% 간사망률상대비감소(2.07배 --> 1.80배)의두가지가상적사례

서로다른인구구성비를지닌두국가에서의직업계층에따른사망률불평등: 가상적사례

사망률상대비(가) 국가

인구구성비(%)

(나) 국가인구구성비(%)

비육체직 1 30% 30%

육체직 2 65% 60%

실업자 3 5% 10%

직업계층에따른사망률불평등의시계열적변화: 가상적사례

인구구성비(%)시점 1에서의사망률상대비

시점 2에서의사망률상대비

비육체직 30% 1 1

육체직 65% 2 1.8

실업자 5% 3 3.5

기준집단

Definition of health inequalities and the

reference group

• “Health disparities occur when one group of people has a higher incidence

or mortality rate than another, or when survival rates are less for one group

than another.”—NCI Center to Reduce Cancer Health Disparities, 2003

(46)

• “A population is a health disparity population if...there is a significant

disparity in the overall rate of disease incidence, prevalence, morbidity,

mortality, or survival rates in the population as compared to the health

status of the general population.”—Minority Health and Health Disparities

Research and Education Act of 2000 (47, page 2498)

Definition of health inequalities and the

reference group

• “For all the medical breakthroughs we have seen in the past century, there

remain significant disparities in the medical conditions of racial groups in

this country.... [W]hat we have done through this initiative is to make a

commitment—really, for the first time in the history of our government—to

eliminate, not just reduce, some of the health disparities between majority

and minority populations.”—Dr. David Satcher, Former U.S. Surgeon

General, 1999 (48, page 18–19)

• “Health disparities are differences in the incidence, prevalence, mortality,

and burden of diseases and other adverse health conditions that exist among

specific population groups in the United States.” —NIH Strategic Research

Plan and Budget to Reduce and Ultimately Eliminate Health Disparities,

Vol. 1, Fiscal Years 2002–2006

기준집단의종류

• 가장높은사회계층

• 전체평균

• “Better than best” method

– US Healthy People 2010

준거기준(Norm)을설정하는경우

• 형평하다고여겨지는수준에서최소기준, 표준, 또는목표치을설정하는방법

• 어린이예방접종률 : 95%를목표로설정

• 영아사망률기준 : 10/1000 미만

• 출생시기대여명 : 80세

• 미달치또는부족분(shortfall)이건강불평등의크기

SMR (Standardized Mortality

Ratio)의준거기준은?

인구규모가동일한 4개지역의사망률(인구10만명당) 변화에따른격차지표의변화양

상: 가상적사례

시점 1에서의사망률 시점 2에서의사망률

(가) 지역 300 300

(나) 지역 120 120

(다) 지역 150 200

(라) 지역 250 250

평균 205 217.5

최상값 120 120

평균값기준격차지표 34.1 26.4

최상값기준격차지표 94.4 108.3

비고: 격차지표(index of disparity)는 기준 집단의 건강 수준에서 개별 집단의 건강 수준 값의평균적인 차이를 의미하는데, 기준이 되는 집단(평균값 또는 최상값)의 건강 수준과 개별 집단의 건강 수준의 차이 값을 기준 집단의 건강 수준으로 나눈 후, 집단의 수(최상값을 기준으로 할 경우 집단의 수에서 1을 뺀 수)로 나눈 값의 100분율에 해당

지역별사망률(인구 10만명당) 변화의두가지가상적사례

지역별사망률 A 사례 B 사례

(가) 지역 300 300 250

(나) 지역 120 120 120

(다) 지역 150 150 150

(라) 지역 250 200 250

Prioritarianism vs Utilitarianism

Adapted from John Broome 2008

무엇을측정할것인가?

Total health inequality vs socioeconomic

health inequality

Inequality in life expectancy

• International comparison에서의어려움극복

• 소득에대하여 GINI를계산하는데, 왜건강

에대해불평등을계산하지못하느냐?

• 건강에서의차이는꼭사회경제적변수에

대해서만기술을하여야하는가?

• 경제학분야에서연구가이루어져옴.

Total Health Inequality

Mortality GINI

Rank of France by World Health Report 2000

육체, 비육체노동자간 국가별 총사망률의 차이

1

1.1

1.2

1.3

1.4

1.5

1.6

1.7

1.8

1.9

2

45 - 59 year old males

Rate

Rati

o

Mackenbach, et al. Lancet (1997)

Total inequality in mortality vs

Socioeconomic mortality

inequality

Houweling TAJ, Kunst AE,

Mackenbach JP. Lancet 2001

Total inequality in mortality vs

Socioeconomic mortality

inequality

Braveman P, Starfield B, Geiger HJ.

BMJ 2001

Asada Y, Hedemann T. Int J

Equity Health 2002

연구결과의차이를무엇으로

설명하여야하나?

Khang et al, J Prev Med Public Health, 2005

Changes in relative index of inequality (RII) for all-cause mortality in

relation to the economic crisis: use of area-based socioeconomic

position

85

90

95

100

105

110

115RII=1.29 RII=1.26 RII=1.31 RII=1.10 RII=1.10 RII=1.11

Males Females

1995-1997 1998-1999 2000-20011995-1997 1998-1999 2000-2001

Khang et al, J Prev Med Public Health 2005

SMR

P for trends = 0.67P for trends = 0.64

Murray CJL, BMJ 2001

Definition of “Equity in Health”

• International Society for Equity in Health (ISEqH)

• “Absence of potentially remediable, systematic

differences in one or more aspects of health across

socially, economically, demographically, or geographically

defined population groups or subgroups”

“Even if we take inequality as an objective notion, our

interest in its measurement must relate to our normative

concern with it, and in judging the relative merits of

different objective measures of inequality, it would

indeed be relevant to introduce normative

considerations. At the same time, even if we take a

normative view of the measures of income inequality,

this is not necessarily meant to catch the totality of our

ethical evaluation.”

Amartya Sen, On Economic Inequality, 1973

Key Reference

Harper S, King NB, Meersman SC, Reichman ME, Breen N, Lynch J. Implicit

Value Judgments in the Measurement of Health Inequalities. The Milbank

Quarterly 2010;88(1):4-29.