|dc.description.abstract||Equity in health and health care distribution is a primary concern for welfare states. In countries that have adopted universal health care coverage, health care is considered a social service that ought to be distributed according to need; not based on the ability to pay or other socioeconomic characteristics.
To measure and examine inequity, it is important to distinguish between need and non-need factors that determine health care utilization. In the empirical analysis of this paper, variables that measure health status, age, gender and lifestyle are categorized as need factors. The socio-economic and other factors such as income, education, regional variation, occupation, civil status and place of origin are categorized as non-need. This categorization is a subjective value judgment. Variation in health care use due to need variables can be taken as legitimate. If the use of health care varies with the non-need variables significantly, controlling for differences in need, it shows that there is inequity in the utilization of health care. These are unfair inequalities because they are caused by the factors beyond the sphere of individual responsibility or characteristics (Fleurbaey and Schokkaert, 2009).
Conceptually, equity can be divided into horizontal and vertical equity. Horizontal equity exists when individuals with equal needs are treated equally. Vertical equity exists when individuals with different needs are treated in proportion to differences between them (Culyer, 2001). In O’Donnell et al. (2008) it is stated that horizontal equity principle is given more attention both in policy and research since a deviation from this principle has an implication on the distribution of health care in a system. Further, researchers assume vertical equity is satisfied on average.
To measure equity, the distribution of need for medical care in the population should be determined. Empirically need is defined as the estimated demand for medical care conditional on some determinant factors using indirect standardization method.
Norway has a universal health coverage system. As one of the components of the National Insurance Scheme (NIS) , the health care system is financed predominantly by the general tax system (Van Noord et al., 1998). Thus the health care use is expected to be distributed according to the equity principle.
The aim of this paper is to test and measure the degree of income-related inequality and inequity of medical care utilization, in the use of general practitioner (GP), private specialist and hospital specialist outpatient services in the Norwegian health care system. Further I will try to identify the major determinant factors that contribute to the income-related inequality. The research questions are:
Does the utilization of medical care services differ between lower-income and higher-income individuals?
What are the major contributing factors to income-related inequality?
Concentration index (CI) is a methodological tool that is commonly used to measure the degree of income-related inequality and inequity in health and health care utilization. In this paper the measurement of the degree of inequality and inequity are illustrated by two approaches; a geometrical approach and a statistical approach. The two approaches are interrelated and consistent to each other. The geometric approach provides a quick intuitive understanding of the implication of the magnitudes and signs of the indices that measure the degree of inequality and inequity. Without a geometric illustration it would be a demanding task to interpret scalar values of the estimated indices. The statistical approach, particularly the “convenient regression” method, makes it possible to estimate concentration indices and their standard error conveniently, and thus to conduct statistical inferences.
Geometrically, CI can be computed based on concentration curve. A concentration curve plots the cumulative proportion of health care use against the cumulative proportion of population ranked by income beginning with the lowest. If the concentration curve lies below (above) the equality (450) line, it indicates pro-rich (pro-poor) distribution of the medical care. If it coincides with the equality line, it indicates equal distribution of medical care use across income groups. The horizontal inequity index (HI) can also be computed using concentration curves. It requires first, to plot a need concentration curve to compute need CI. Then the HI is obtained as the difference between actual use CI and need CI. The magnitude of the indices range between -1 and +1, and the interpretations depends on the signs. Positive (negative) values of CI indicate inequality favoring the rich (poor). Similarly, positive (negative) values of HI indicate inequity favoring the rich (poor). A zero or insignificant value of CI (HI) indicates health care use is distributed fairly equally (equitably) across income groups. The method I use to identify the major factors contributing to overall income-related inequality is conceptually identical to the decomposition method used in Wagstaff et al. (2003), and it is based on the linear regression model. The decomposition method can also be used as a tool to identify policy relevant factors in a health care system.
To measure these indices the Norwegian level of living conditions cross-sectional survey data of the year 2005 are used. All the inequality and inequity indices and decomposed inequalities are estimated using STATA software.
The paper finds that the lower income groups are more likely to use GP and hospital specialist services intensively than the higher income groups. Moreover, lower income groups have higher need for the medical care. After controlling for need differences, no evidence of horizontal inequity is found in these two services. That is, GP and hospital specialist services are distributed fairly across income groups. However the result for the probability of private specialist indicates that there is horizontal inequity favoring the better-off.
Decomposition results show that the most important variables that contribute to the overall pro-poor income-related inequality in the probability of GP and hospital specialist visit are need variables. The contributions of non-need explanatory variables to the inequality are very low. The decomposition results support the findings that assert the absence of horizontal inequity in the distribution of these medical care services. The results for the probability of private specialist visit show different outcomes. The largest percentage share of the pro-rich inequality is caused by non-need variables, particularly income and education.
This study finds no evidence that the equity principle is violated in the hospital specialist and GP service utilization. However, the distribution of private specialist services favors the well-off. The fact that the lower income groups have higher need for medical care, and are more intensive users of hospital specialist and GP services suggest inequity in health in the society.||eng