Effects of tobacco price adjustments under different scenarios on the overall 10-year mortality risk at the whole population level, the relative index of inequality (RII) and slope index of inequality (SII) by income, and relative risk (RR) and absolute risk (AR) by income (income Q1 vs income Q5 presented), Korea
| Overall mortality | ||
| 10-year rate % (95% CI) | R % (95% CI) | |
| No change | 5.09 (5.00 to 5.16) | |
| 10% increase | 5.08 (4.99 to 5.15) | 0.2 (0.0 to 0.4) |
| 20% increase | 5.07 (4.98 to 5.14) | 0.5 (0.1 to 0.8) |
| 50% increase | 5.04 (4.95 to 5.11) | 1.1 (0.2 to 1.8) |
| 100% increase | 4.99 (4.89 to 5.09) | 2.0 (0.3 to 3.3) |
| Relative inequality measure | ||||
| RII (95% CI) | RIIR % (95% CI) | RR (95% CI) | RRR% (95% CI) | |
| No change | 1.68 (1.60 to 1.79) | 1.51 (1.46 to 1.59) | ||
| 10% increase | 1.68 (1.60 to 1.78) | 0.5 (0.1 to 0.9) | 1.51 (1.46 to 1.58) | 0.4 (0.0 to 0.8) |
| 20% increase | 1.68 (1.60 to 1.78) | 1.0 (0.1 to 1.7) | 1.51 (1.46 to 1.58) | 0.8 (0.1 to 1.5) |
| 50% increase | 1.67 (1.59 to 1.77) | 2.2 (0.2 to 4.0) | 1.50 (1.45 to 1.58) | 1.9 (0.2 to 3.5) |
| 100% increase | 1.66 (1.57 to 1.77) | 3.8 (0.4 to 6.8) | 1.50 (1.45 to 1.57) | 3.3 (0.4 to 6.0) |
| Absolute inequality measure | ||||
| SII per 100 000 (95% CI) | SIIR % (95% CI) | AR per 100 000 (95% CI) | ARR% (95% CI) | |
| No change | 2617 (2340 to 2913) | 2166 (1959 to 2426) | ||
| 10% increase | 2601 (2328 to 2899) | 0.6 (0.1 to 1.1) | 2153 (1943 to 2413) | 0.6 (0.1 to 1.1) |
| 20% increase | 2585 (2317 to 2884) | 1.2 (0.1 to 2.1) | 2141 (1934 to 2400) | 1.2 (0.1 to 2.0) |
| 50% increase | 2545 (2287 to 2843) | 2.7 (0.3 to 4.8) | 2108 (1887 to 2365) | 2.7 (0.3 to 4.7) |
| 100% increase | 2490 (2228 to 2797) | 4.8 (0.6 to 8.5) | 2064 (1833 to 2325) | 4.7 (0.6 to 8.1) |
The CI for 10-year mortality risk represents the 95% bootstrap CI estimated by the 2.5th and 97.5th percentiles of the distribution of the mortality risks calculated after applying the Cox regression model to 200 bootstrap samples from the National Health Insurance Service (NHIS) sample cohort data (2006–2015). The reduction % in the overall 10-year mortality risk was calculated as
, where
is the predicted risk in the whole population based on the current distributions of risk factors in the Korea National Health and Nutrition Examination Survey (KNHANES) 2015–2017, and
is the predicted risk in the whole population based on each of the scenarios.The SII is the difference in the predicted mortality risks at the top (range=1) and bottom (range=0) of the income distribution. The RII is the ratio of the predicted mortality risks at the top (range=1) and bottom (range=0) of the income distribution; the CI for SII and RII represents the 95% bootstrap CIs estimated by the 2.5th and 97.5th percentiles of the distribution of the SIIs and RIIs for the 10-year mortality risks calculated after applying the Cox regression model to 200 bootstrap samples from the NHISsample cohort data (2006–2015). RII reduction (RIIR, %) = [RIIC − RIIS]/[RIIC − 1]×100, SII reduction (SIIR, %) = [SIIC − SIIS]/SIIC ×100, where the subscript ‘C’ represents the current distributions of smoking-related variables in the KNHANES 2015–2017, and the subscript ‘S’ represents each of the scenarios (ie, % change in price).
The RR, AR, RR reduction (RRR) and AR reduction (ARR), for comparing income Q1 (the bottom 20% of income) and income Q5 (the top 20% of income), are defined as follows: RR=P_L/P_H, AR=P_L-P_H, RRR(%)=[(P_L/P_H)_C-(P_L/P_H)_S]/[(P_L/P_H)_C-1]×100, ARR(%)=[(P_L-P_H)_C-(P_L-P_H)_S]/(P_L-P_H)_C×100, where P_L is the mortality risk in income Q1, P_H is the mortality risk in income Q5, the subscript ‘C’ represents the current distributions of smoking-related variables in KNHANES 2015–2017, and the subscript ‘S’ represents each of the scenarios. The CI for RR and AR represents the 95% bootstrap CIs estimated by the 2.5th and 97.5th percentiles of the distribution of the RRs and ARs for the 10-year mortality risks calculated after applying the Cox regression model to 200 bootstrap samples from the NHIS sample cohort data (2006–2015).
