Concepts and Indicators Used in Studies of Out-of-Pocket Health Expenditures
| . | Concept . | Indicator . | Interpretation . | Study introducing the indicator . | Comments . |
|---|---|---|---|---|---|
| 1 | Expenditure in absolute terms | Per capita annual out-of-pocket expenditures in monetary terms. | Absolute amount of expenditure per capita, adjusted for inflation and in international studies converted to a common currency. | Multiple studies in the United States report trends in per capita out-of-pocket expenditures in dollar terms (see, e.g., Banthin, Cunningham, and Bernard 2008). | |
| 2 | Dispersion (risk) | Coefficient of variation (CV) | Both interpreted as measures of out-of-pocket expenditure risk. | Gruber and Levy (2009) | Gruber and Levy actually use the variance, but this is not invariant with respect to the mean, quadrupling when everyone's out-of-pocket expenditures double (cf., e.g., Cowell 2011) |
| Q90/Q50. The ratio of the expenditures incurred by households at the 90th and 50th percentiles of the out-of-pocket expenditure distribution. | Gruber and Levy (2009) | Q90/Q50 is undefined if Q50 = 0. | |||
| 3 | Budget share | Budget share | Share of income or consumption spent on out-of-pocket health expenses. | ||
| 4 | Progressivity | Kakwani's (1977) index of progressivity (applied to out-of-pocket expenditures) is equal to twice the area between the Lorenz curve for income and the concentration curve for out-of-pocket expenditures, where the latter plots the cumulative share of out-of-pocket expenditures against the cumulative share of the population ranked by income, starting with the poorest. Equivalently, it is equal to the concentration index for out-of-pocket expenditures minus the Gini coefficient for income or consumption, where the concentration index is defined analogously to the Gini coefficient but with respect to the concentration curve for out-of-pocket expenditures, not the Lorenz curve for income.a | A negative value of Kakwani's index indicates that out-of-pocket expenditures are regressive, or equivalently, that the budget share declines with income or consumption. | Wagstaff et al. (1992) | Wagstaff et al. also compute progressivity indices for other financing sources, for example, social health insurance contributions, etc. They also use Suits’ (1977) progressivity index. |
| 5 | Catastrophic expenditures | CATA | Fraction of households whose out-of-pocket health expenditures exceed some pre-specified threshold (e.g. 10% or 25%) of their total income of consumption. | Berki (1986) and Wyszewianski (1986) | CATA (10%) is UN SDG indicator 3.8.2. Wagstaff and van Doorslaer (2003) add a variation that captures the “overshoot” of expenditures above the threshold, not just whether the threshold is exceeded. |
| 6 | Inequality in incidence of catastrophic expenditures | Concentration index (CI) of catastrophic expenditures – CI (CATA) | A negative value indicates catastrophic expenditures are more common among those at the bottom of the income or consumption distribution. | Wagstaff and van Doorslaer (2003) | |
| 7 | Impoverishment | The increase in the poverty headcount and mean poverty gap when out-of-pocket expenditures are subtracted from income or consumption – IMPOV. | Out-of-pocket expenditures are said to be “impoverishing” if they are sufficiently large to leave a household below the poverty line based on consumption or income net of out-of-pocket expenditures but above the poverty based on consumption or income gross of out-of-pocket expenditures. | Wagstaff and van Doorslaer (2003) |
| . | Concept . | Indicator . | Interpretation . | Study introducing the indicator . | Comments . |
|---|---|---|---|---|---|
| 1 | Expenditure in absolute terms | Per capita annual out-of-pocket expenditures in monetary terms. | Absolute amount of expenditure per capita, adjusted for inflation and in international studies converted to a common currency. | Multiple studies in the United States report trends in per capita out-of-pocket expenditures in dollar terms (see, e.g., Banthin, Cunningham, and Bernard 2008). | |
| 2 | Dispersion (risk) | Coefficient of variation (CV) | Both interpreted as measures of out-of-pocket expenditure risk. | Gruber and Levy (2009) | Gruber and Levy actually use the variance, but this is not invariant with respect to the mean, quadrupling when everyone's out-of-pocket expenditures double (cf., e.g., Cowell 2011) |
| Q90/Q50. The ratio of the expenditures incurred by households at the 90th and 50th percentiles of the out-of-pocket expenditure distribution. | Gruber and Levy (2009) | Q90/Q50 is undefined if Q50 = 0. | |||
| 3 | Budget share | Budget share | Share of income or consumption spent on out-of-pocket health expenses. | ||
| 4 | Progressivity | Kakwani's (1977) index of progressivity (applied to out-of-pocket expenditures) is equal to twice the area between the Lorenz curve for income and the concentration curve for out-of-pocket expenditures, where the latter plots the cumulative share of out-of-pocket expenditures against the cumulative share of the population ranked by income, starting with the poorest. Equivalently, it is equal to the concentration index for out-of-pocket expenditures minus the Gini coefficient for income or consumption, where the concentration index is defined analogously to the Gini coefficient but with respect to the concentration curve for out-of-pocket expenditures, not the Lorenz curve for income.a | A negative value of Kakwani's index indicates that out-of-pocket expenditures are regressive, or equivalently, that the budget share declines with income or consumption. | Wagstaff et al. (1992) | Wagstaff et al. also compute progressivity indices for other financing sources, for example, social health insurance contributions, etc. They also use Suits’ (1977) progressivity index. |
| 5 | Catastrophic expenditures | CATA | Fraction of households whose out-of-pocket health expenditures exceed some pre-specified threshold (e.g. 10% or 25%) of their total income of consumption. | Berki (1986) and Wyszewianski (1986) | CATA (10%) is UN SDG indicator 3.8.2. Wagstaff and van Doorslaer (2003) add a variation that captures the “overshoot” of expenditures above the threshold, not just whether the threshold is exceeded. |
| 6 | Inequality in incidence of catastrophic expenditures | Concentration index (CI) of catastrophic expenditures – CI (CATA) | A negative value indicates catastrophic expenditures are more common among those at the bottom of the income or consumption distribution. | Wagstaff and van Doorslaer (2003) | |
| 7 | Impoverishment | The increase in the poverty headcount and mean poverty gap when out-of-pocket expenditures are subtracted from income or consumption – IMPOV. | Out-of-pocket expenditures are said to be “impoverishing” if they are sufficiently large to leave a household below the poverty line based on consumption or income net of out-of-pocket expenditures but above the poverty based on consumption or income gross of out-of-pocket expenditures. | Wagstaff and van Doorslaer (2003) |
Source: Authors.
Note: aThe concentration index is twice the area between the Lorenz curve for income and the concentration curve for taxes or payments, while the Gini coefficient is equal to twice the area between the Lorenz curve and the line of equality. The Lorenz curve is formed by ranking households by their income and plotting the cumulative percentage of households so ranked on the x-axis against the cumulative percent of income on the y-axis. The concentration curve is formed by keeping households ranked by their income and plotting the cumulative percentage of households so ranked on the x-axis against the cumulative percent of taxes or payments on the y-axis.
Concepts and Indicators Used in Studies of Out-of-Pocket Health Expenditures
| . | Concept . | Indicator . | Interpretation . | Study introducing the indicator . | Comments . |
|---|---|---|---|---|---|
| 1 | Expenditure in absolute terms | Per capita annual out-of-pocket expenditures in monetary terms. | Absolute amount of expenditure per capita, adjusted for inflation and in international studies converted to a common currency. | Multiple studies in the United States report trends in per capita out-of-pocket expenditures in dollar terms (see, e.g., Banthin, Cunningham, and Bernard 2008). | |
| 2 | Dispersion (risk) | Coefficient of variation (CV) | Both interpreted as measures of out-of-pocket expenditure risk. | Gruber and Levy (2009) | Gruber and Levy actually use the variance, but this is not invariant with respect to the mean, quadrupling when everyone's out-of-pocket expenditures double (cf., e.g., Cowell 2011) |
| Q90/Q50. The ratio of the expenditures incurred by households at the 90th and 50th percentiles of the out-of-pocket expenditure distribution. | Gruber and Levy (2009) | Q90/Q50 is undefined if Q50 = 0. | |||
| 3 | Budget share | Budget share | Share of income or consumption spent on out-of-pocket health expenses. | ||
| 4 | Progressivity | Kakwani's (1977) index of progressivity (applied to out-of-pocket expenditures) is equal to twice the area between the Lorenz curve for income and the concentration curve for out-of-pocket expenditures, where the latter plots the cumulative share of out-of-pocket expenditures against the cumulative share of the population ranked by income, starting with the poorest. Equivalently, it is equal to the concentration index for out-of-pocket expenditures minus the Gini coefficient for income or consumption, where the concentration index is defined analogously to the Gini coefficient but with respect to the concentration curve for out-of-pocket expenditures, not the Lorenz curve for income.a | A negative value of Kakwani's index indicates that out-of-pocket expenditures are regressive, or equivalently, that the budget share declines with income or consumption. | Wagstaff et al. (1992) | Wagstaff et al. also compute progressivity indices for other financing sources, for example, social health insurance contributions, etc. They also use Suits’ (1977) progressivity index. |
| 5 | Catastrophic expenditures | CATA | Fraction of households whose out-of-pocket health expenditures exceed some pre-specified threshold (e.g. 10% or 25%) of their total income of consumption. | Berki (1986) and Wyszewianski (1986) | CATA (10%) is UN SDG indicator 3.8.2. Wagstaff and van Doorslaer (2003) add a variation that captures the “overshoot” of expenditures above the threshold, not just whether the threshold is exceeded. |
| 6 | Inequality in incidence of catastrophic expenditures | Concentration index (CI) of catastrophic expenditures – CI (CATA) | A negative value indicates catastrophic expenditures are more common among those at the bottom of the income or consumption distribution. | Wagstaff and van Doorslaer (2003) | |
| 7 | Impoverishment | The increase in the poverty headcount and mean poverty gap when out-of-pocket expenditures are subtracted from income or consumption – IMPOV. | Out-of-pocket expenditures are said to be “impoverishing” if they are sufficiently large to leave a household below the poverty line based on consumption or income net of out-of-pocket expenditures but above the poverty based on consumption or income gross of out-of-pocket expenditures. | Wagstaff and van Doorslaer (2003) |
| . | Concept . | Indicator . | Interpretation . | Study introducing the indicator . | Comments . |
|---|---|---|---|---|---|
| 1 | Expenditure in absolute terms | Per capita annual out-of-pocket expenditures in monetary terms. | Absolute amount of expenditure per capita, adjusted for inflation and in international studies converted to a common currency. | Multiple studies in the United States report trends in per capita out-of-pocket expenditures in dollar terms (see, e.g., Banthin, Cunningham, and Bernard 2008). | |
| 2 | Dispersion (risk) | Coefficient of variation (CV) | Both interpreted as measures of out-of-pocket expenditure risk. | Gruber and Levy (2009) | Gruber and Levy actually use the variance, but this is not invariant with respect to the mean, quadrupling when everyone's out-of-pocket expenditures double (cf., e.g., Cowell 2011) |
| Q90/Q50. The ratio of the expenditures incurred by households at the 90th and 50th percentiles of the out-of-pocket expenditure distribution. | Gruber and Levy (2009) | Q90/Q50 is undefined if Q50 = 0. | |||
| 3 | Budget share | Budget share | Share of income or consumption spent on out-of-pocket health expenses. | ||
| 4 | Progressivity | Kakwani's (1977) index of progressivity (applied to out-of-pocket expenditures) is equal to twice the area between the Lorenz curve for income and the concentration curve for out-of-pocket expenditures, where the latter plots the cumulative share of out-of-pocket expenditures against the cumulative share of the population ranked by income, starting with the poorest. Equivalently, it is equal to the concentration index for out-of-pocket expenditures minus the Gini coefficient for income or consumption, where the concentration index is defined analogously to the Gini coefficient but with respect to the concentration curve for out-of-pocket expenditures, not the Lorenz curve for income.a | A negative value of Kakwani's index indicates that out-of-pocket expenditures are regressive, or equivalently, that the budget share declines with income or consumption. | Wagstaff et al. (1992) | Wagstaff et al. also compute progressivity indices for other financing sources, for example, social health insurance contributions, etc. They also use Suits’ (1977) progressivity index. |
| 5 | Catastrophic expenditures | CATA | Fraction of households whose out-of-pocket health expenditures exceed some pre-specified threshold (e.g. 10% or 25%) of their total income of consumption. | Berki (1986) and Wyszewianski (1986) | CATA (10%) is UN SDG indicator 3.8.2. Wagstaff and van Doorslaer (2003) add a variation that captures the “overshoot” of expenditures above the threshold, not just whether the threshold is exceeded. |
| 6 | Inequality in incidence of catastrophic expenditures | Concentration index (CI) of catastrophic expenditures – CI (CATA) | A negative value indicates catastrophic expenditures are more common among those at the bottom of the income or consumption distribution. | Wagstaff and van Doorslaer (2003) | |
| 7 | Impoverishment | The increase in the poverty headcount and mean poverty gap when out-of-pocket expenditures are subtracted from income or consumption – IMPOV. | Out-of-pocket expenditures are said to be “impoverishing” if they are sufficiently large to leave a household below the poverty line based on consumption or income net of out-of-pocket expenditures but above the poverty based on consumption or income gross of out-of-pocket expenditures. | Wagstaff and van Doorslaer (2003) |
Source: Authors.
Note: aThe concentration index is twice the area between the Lorenz curve for income and the concentration curve for taxes or payments, while the Gini coefficient is equal to twice the area between the Lorenz curve and the line of equality. The Lorenz curve is formed by ranking households by their income and plotting the cumulative percentage of households so ranked on the x-axis against the cumulative percent of income on the y-axis. The concentration curve is formed by keeping households ranked by their income and plotting the cumulative percentage of households so ranked on the x-axis against the cumulative percent of taxes or payments on the y-axis.
This PDF is available to Subscribers Only
View Article Abstract & Purchase OptionsFor full access to this pdf, sign in to an existing account, or purchase an annual subscription.
