Bartoll et al. Archives of Public Health (2019) 77:12 https://doi.org/10.1186/s13690-019-0339-z RESEARCH Open Access Stable socioeconomic inequalities in ischaemic heart disease mortality during the economic crisis: a time trend analysis in 2 Spanish settings Xavier Bartoll1,3*, Mercè Gotsens1,3, Marc Marí-Dell’Olmo1,2,3,4, Laia Palència1,2,3, Montse Calvo5, Santiago Esnaola5 and Carme Borrell1,2,3,4 Abstract Background: Prior studies have identified a decrease in ischaemic heart disease mortality during the recent economic recession. The Spanish population was severely affected by the Great Recession, however, there is little evidence on its effects on socioeconomic inequalities in ischaemic heart disease mortality. This study examines trends in socioeconomic inequalities in mortality due to ischaemic heart disease (IHD). Methods: We used linked census records with mortality registers available from the Basque Country and Barcelona city for population above 25 years, between 2001 and 04, the accelerated economic growth period of 2005–08, and 2009–12, with the last period coinciding with the Great Recession. Applying Poisson models, we calculated relative and absolute indexes of inequalities by education level for each period, age group, gender, and site. Results: We found moderate age-adjusted inequalities in IHD with a gradient of increasing rates through less educational level, but no significant evidence of increasing trends in socioeconomic inequalities in IHD mortality, rather an inverted U-shape time trend in some groups below 75 years in relative inequalities. Absolute inequalities decrease in the last period except for women from 50 to 64 years. Conclusions: This study shows that the economic crisis has not increased socioeconomic inequalities in IHD mortality in two geographical settings in Spain. Keywords: Ischaemic heart disease mortality, Socioeconomic inequalities, Economic, Crisis, Spain Background evidence on the effects of the economic cycles on spe- An influential study in the late 70’s found that economic cific and all-cause mortality inequalities by socioeco- recessions were related to increased cardiovascular mor- nomic level. It is well reported that lower socioeconomic tality [1]. This result was soon challenged [2] by new evi- groups have higher mortality rates across European dence based on more sound panel methodology showing countries [4, 5]. One study for the period 1980–1990 in just the opposite, that the increased unemployment was USA reported that men with less education experienced associated with a decline in cardiovascular mortality [3]. higher all cause-mortality with increasing unemploy- The beginning of the Great Recession in 2008 has ment [6]. While in Nordic countries and Japan charac- brought renewed interest in analysing the effect of eco- terized with universal health care, the mortality in nomic cycles on mortality. However, there is little cardiovascular disease of lower socioeconomic group de- creased [7–9]. Another study for Northern Ireland found no significant increase in ischaemic heart disease mortal- * Correspondence: xbartoll@aspb.cat 1 Agència de Salut Pública de Barcelona, Plaça Lesseps 1, 08023 Barcelona, ity among the working population after the economic Spain crisis in 2008, but an increase for retired workers [10]. 3 Institut d’Investigació Biomèdica (IIB Sant Pau), Barcelona, Spain For Spain, decreases in all-cause mortality and Full list of author information is available at the end of the article © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Bartoll et al. Archives of Public Health (2019) 77:12 Page 2 of 9 cardiovascular disease during the Great Recession were record linkage between the mortality register and Barce- found larger for lower socioeconomic groups (proxied lona’s municipal census. Information on the at-risk by household floor space and car ownership) [11]. population, including age and sex and was obtained After the mild crisis of 2001, the Spanish economy grew from the municipal census. Due to a lack of linkage, we at a constant rate for the next years until the fourth quar- excluded 401 IHD deaths for men (4.4%) and 332 for ter of 2008 to enter in negative growth rates. Since then, women (4.7%). the economic recession has severely affected the Spanish For the Basque Country, we used a population-based population, with a rapid rise in unemployment, work pre- census-linked longitudinal mortality study. Individual cariousness and vulnerable population increasing socio- census records from the 2001 and 2006 population cen- economic inequalities [12]. Moreover, as in the European suses were linked with the mortality registry. Individuals context, the austerity measures imposed in Spain may from the 2001 census were followed up from November have negatively affected health and health services [13, 14]. 2001 (date of the 2001 Census) to October 2006, and To analyse trends in socioeconomic inequalities of is- those from the 2006 census were followed up from chaemic heart disease (IHD) mortality we compared two November 2006 to December 2012. Deaths were distinct settings in Spain, for which data were available: weighted by the inverse of the proportion of linked Barcelona city and The Basque Country. Both areas have cases. The exact number of person-years of follow-up similar populations and are in the north of Spain. The was calculated by subtracting the date of the corre- Basque Country had the lowest unemployment rate (rose sponding census from the date of death (for deceased from 11.6% in 2001 to 14.9% in 2012) and poverty rates in persons) or the date of the end of follow up. The per- Spain (ranged from 5 to 10% even during the crisis period) centages of male and female IHD deaths for whom no and ranked first in health spending per capita (reached the information on educational level was available due to a peak of 1971€ in 2009). Conversely in Barcelona, the un- lack of linkage were 3.3 and 4.8% for the 2001–2004 employment rate rose from 10.5% to a 18.7% and poverty period, 2.8 and 3.3% for the 2005–2008 period, and 4.8 rate ranged from 15.7 to 18.3% in the same period, own and 2.7% for the 2009–2012 period, respectively. estimation of health spending per capita for Barcelona area are lower (reached the peak of 1398€ in 2009). The figures in Barcelona were closer to the Spanish average Variables and indicators analysed than those of The Basque Country [15–17]. We analysed deaths due to IHD included International The objective of this paper is to analyse trends in so- Classification of Diseases 10th (ICD-10) codes of the cioeconomic inequalities of IHD mortality before (pe- underlying cause of death: from angina pectoris (I20) to riods 2001–04 and 2005–08) and during the Great chronic ischaemic heart disease (I25). Our main inde- Recession (2009–12) according to education level. pendent variable was the highest level of education attained. Educational level was categorised in five Methods groups: (a) no formal education or pre-primary Design, study population and sources of information education (0–4 years of schooling), (b) primary educa- This study uses individual-level data from the Basque tion (5–6 years of schooling); (c) lower secondary educa- Country and Barcelona city, two settings where the in- tion (7–9 years of schooling); (d) upper secondary formation of the mortality register linked with the cen- education (10–14 years of studies), and (e) university sus is available in Spain. education (≥15 years of schooling). Although education The Basque Country population was 2,082,587 at the level from census records is not always an up-to-date end of 2001 and 2,179,815 at the end of 2011. Barcelona register during the life-course, it is attained relatively is the second largest city in Spain with 1,505,325 inhabi- early in life and it tends to be stable over the adult life tants in 2001 and 1,620,943 in 2012. We considered span, moreover it has been extensively used in mortality at-risk population to be all residents older than 25 years analysis as a proxy for socioeconomic position in most between 2001 and 2012. We choose 25 years as cut-off of the aforementioned literature. Other independent var- point because university studies are finished at this age. iables were sex, age, and year of death. Age of death was Barcelona and the Basque country have two different grouped in five categories: 25–49 years, 50–64 years, 65– register information. In Barcelona it is possible to link 74 years, ≥75 years, and an all-ages group. Year of death cases of the people who died, while in Basque Country was grouped into two periods before the beginning of uses a follow up linkage. For Barcelona, we only in- the economic recession (2001–04 and 2005–08) and one cluded IHD deaths in the mortality register that oc- period after the beginning of the recession (2009–12). curred among residents between 2001 and 2012, and for Although the economic recession actually started during whom complete educational level information was avail- the 4th term of 2008, unemployment levels did not start able. Data on subjects’ educational level was obtained by to rise noticeably until 2009. Bartoll et al. Archives of Public Health (2019) 77:12 Page 3 of 9 Data analysis and lower secondary above 75 years. For women, it de- All analyses were carried out separately for each region creases in the last period, except for women from 50 to (Barcelona and Basque Country), for men and women, 64 years that increases except for those in primary edu- for the four age groups and all-ages group, and for the cation level. three-time periods. We fit Poisson regression models to Regarding standardized rate deaths of IHD in all ages compute the Relative Index of inequality (RII) and its group in Table 2, we observe a decrease through the 95% confidence interval (95% CI) for age-adjusted edu- period 2001–2012 in Barcelona among all education cational level specific mortality. The dependent variable level men and women except for those men in upper was the logarithm of the mortality rate, and the inde- secondary level that increases in the last period. For in- pendent variables were educational level as a quantita- stance, men with no education rates ranges from 221.5 tive variable (with five values between 0 and 1, based on per 100.00 inhabitants in 2001–2004 to a 140.0 in 2008– the midpoint of the educational level group’s position in 2012, those in university from 197.2 to a 140.7 in the the cumulative educational level distribution of the same period. For women with no education, IHD rates population). The RII value can therefore be interpreted range from a 106.9 in 2001–2004 to a 66.0 in 2008– as the ratio of mortality rates between the two extremes 2012, and for those in university level from 76.5 to a of the educational spectrum (in our case, the lowest edu- 47.7 in the same period. In Basque Country, standard- cational level versus the highest, taking into account all ized death rates of IHD decreases among men and educational levels). We also fit models with the 3 periods women through the period. For instance, men with no together where the independent variables were age education rates range from 196.3 in 2001–2004 to a group, period, educational level and the interaction be- 130.0 in 2008–2012, those in university from 135.6 to tween period and educational level, to determine if the 107.1 in the same period. For women with no education, RII changed over time. IHD rates range from 75.4 to 65.0 and among those with Moreover, we estimated the age-adjusted Slope Index university degree from 62.3 to 29.7. of Inequality (SII). To estimate SII, simply adjusting for Figure 1 shows a moderate inequality gradient in mor- age may not be enough to obtain a truly comparable tality by education level for the all-ages groups, with the measure, so we used a novel approach [18]. First, differ- rate increasing with decreasing educational level. The ent SII were obtained for each age-group with an addi- pattern of inequality is quite stable across time. IHD tive Poisson model. Second, we obtained a single SII death rates for men in no education level decrease more performing a weighted sum of these SII. The weights than those of high educational level in both settings correspond to the relative sizes of the age groups in a while for women they are more stable. reference population (in our case both areas Barcelona To better analyse the evolution of these inequalities by and Basque Country). The regression models are adjusted education level across periods, Table 3 shows the relative by the 5-year variable. Poisson models were corrected for index of inequality (RII) and the slope index of inequal- overdispersion where necessary. Due to the increase of ity (SII), age group, sex, and study period. Most RIIs immigration to Spain in the last two decades, we have rep- were greater than 1, which means that those in lower licated all the analysis stratifying by country of birth, as educational level have higher risk of death from IHD the results were not significantly different, we have only than higher educational level ones. The most consistent reported the results for the overall population. finding is that there was no significant change in RII be- tween study periods as confidence intervals overlap and Results the interaction between period and educational level is Table 1 shows descriptive of the number and rate of not significant (interaction p-values not shown). How- crude IHD deaths, according to level of education, ever, for men and women in Barcelona the RII appears age-group, sex and period at the two sites. In Barcelona to show an inverted U-shape trend, that is, it is lower in we observe a decreasing trend among those at the ex- the first and last periods but higher in the middle one treme of the education level, no education and primary (except for women from 50 to 64 years that follows a as well as among those in university level, while for U-shape trend and for both sexes over 74 years). None- those in lower secondary level from 50 to 64 years that theless the time variations of the RII are of minimum increased across 2001–2012 and those above 75 years magnitude. For instance, the RII for men in all ages that increased in the last period. For women, rates de- group during 2001–2004 is 1.2 (95%CI:1.0–1.4), then in- crease for most groups except for those in lower second- creases to 1.3 (95%CI: 1.1–1.6) and finally decreases to ary from 50 to 64 years that increases in the last period, 1.1 (95%CI:0.9–1.3) in 2009–2012. For women in and university level from 65 to 74 years. For men in Barcelona the figures are, first 1.7 (95CI%:1.3–2.1), Basque Country, IHD deaths rates decrease for most then increases to 1.8 (95CI%:1.5–2.3) and finally de- groups, except for those in university from 50 to 64 years creases to 1.5 (95%CI:1.2–1.9). This pattern is Bartoll et al. Archives of Public Health (2019) 77:12 Page 4 of 9 Table 1 Number and crude rate of deaths (per 100.00 population) due to ischaemic heart disease for Barcelona and the Basque Country by age group, 2001–2012 Region/Age Education level Men Women groups (years) 2001–2004 2005–2008 2009–2012 2001–2004 2005–2008 2009–2012 n rate n rate n rate n rate n rate n rate Barcelona 25–49 No education 5 22.0 2 9.3 1 3.4 2 10.5 1 7.0 0 0.0 Primary 36 18.9 26 11.3 19 8.7 3 1.9 9 5.3 2 1.3 Lower secondary 31 12.1 30 11.4 27 10.3 1 0.4 2 0.8 4 1.9 Upper secondary 31 7.3 27 6.1 25 6.2 4 1.0 4 0.9 4 1.0 University 16 5.9 12 3.8 12 3.1 4 1.2 1 0.3 1 0.2 50–64 No education 58 112.6 38 118.1 15 80.4 18 21.5 8 14.3 3 9.4 Primary 152 99.7 131 94.1 81 68.1 28 13.7 20 10.6 15 9.4 Lower secondary 61 65.7 74 74.1 85 80.0 6 4.9 8 5.8 13 8.8 Upper secondary 80 77.0 62 50.1 88 59.7 11 12.8 10 9.4 9 6.8 University 75 65.1 69 51.7 51 34.3 3 3.3 9 7.4 6 3.8 65–74 No education 207 290.9 143 271.1 76 218.7 125 98.0 78 82.8 47 73.2 Primary 295 284.9 217 235.6 167 197.5 109 70.2 68 49.4 33 26.1 Lower secondary 84 212.4 91 225.6 90 202.4 31 61.3 20 36.8 20 31.8 Upper secondary 127 294.2 86 197.2 95 203.0 17 47.1 15 39.8 9 20.8 University 107 246.2 75 164.6 71 127.9 13 46.9 7 22.9 12 30.0 > =75 No education 600 991.6 498 810.9 383 636.8 1154 677.4 839 506.9 720 461.8 Primary 717 974.6 651 790.6 642 714.8 828 509.6 804 448.0 714 372.2 Lower secondary 204 850.6 186 659.7 223 663.6 142 376.2 148 320.2 173 308.2 Upper secondary 268 952.4 224 690.4 269 717.4 123 448.8 107 312.7 124 304.8 University 259 928.5 268 801.3 297 739.0 107 496.7 83 321.6 101 320.5 Basque Country 25–49 No education 2 22.8 6 50.0 1 8.0 0 0.0 0 0.0 0 0.0 Primary 89 22.9 89 19.8 80 20.5 13 3.5 14 3.6 9 3.0 Lower secondary 22 48.1 10 19.9 2 4.5 7 13.3 2 3.8 2 4.9 Upper secondary 46 9.2 62 8.8 89 12.2 4 1.0 14 2.5 6 1.0 University 27 7.0 21 4.2 17 3.5 4 0.9 5 0.8 5 0.8 50–64 No education 28 112.5 37 161.4 21 132.2 7 20.4 5 15.7 6 28.2 Primary 285 91.6 300 80.6 266 78.1 63 15.7 44 9.2 37 8.7 Lower secondary 31 61.9 65 79.6 59 66.3 2 3.2 8 7.5 11 9.2 Upper secondary 91 75.7 136 69.2 146 58.5 2 3.3 6 5.1 11 6.2 University 62 69.9 77 57.5 96 59.0 5 9.4 8 8.2 11 7.6 65–74 No education 91 250.9 93 262.8 32 127.0 41 77.5 40 76.4 20 52.9 Primary 504 244.7 450 192.3 347 159.3 167 62.7 146 47.0 102 34.6 Lower secondary 26 206.1 39 179.9 44 141.5 10 63.8 12 40.2 11 24.4 Upper secondary 66 227.5 86 187.5 77 123.4 3 21.3 8 34.5 11 32.4 University 55 203.0 59 155.4 68 145.2 7 50.1 6 29.8 3 11.3 > =75 No education 310 993.7 308 752.4 297 710.3 423 586.0 509 556.1 451 498.1 Primary 1081 836.3 1266 657.7 1337 598.7 1100 459.3 1368 394.6 1268 320.0 Lower secondary 35 633.6 57 514.3 112 667.0 38 431.1 67 333.7 96 306.8 Upper secondary 88 758.0 140 619.8 176 532.4 27 302.9 45 265.3 63 260.5 University 91 647.5 148 641.4 154 514.3 52 460.8 58 325.1 54 238.2 Bartoll et al. Archives of Public Health (2019) 77:12 Page 5 of 9 Table 2 Number and age-standardized rate of deaths (per 100.00 population) due to ischaemic heart disease, Barcelona and the Basque Country, 2001–2012 Region/Age groups Education level Men Women (all ages) 2001–2004 2005–2008 2009–2012 2001–2004 2005–2008 2009–2012 n rate n rate n rate n rate n rate n rate Barcelona No education 870 221.5 681 185.8 475 140.0 1299 106.9 926 79.0 770 66.0 Primary 1200 222.1 1025 181.0 909 151.8 968 86.3 901 70.9 764 51.1 Lower secondary 380 185.3 381 156.9 425 153.3 180 67.2 178 53.9 210 50.3 Upper secondary 506 209.4 399 150.2 477 156.1 155 81.1 136 54.5 146 46.3 University 457 197.2 424 160.2 431 140.7 127 76.5 100 49.1 120 47.7 Basque Country No education 431 196.3 444 185.6 351 130.0 471 75.4 554 70.2 477 65.0 Primary 1.959 179.8 2.105 142.5 2.030 124.9 1.343 68.2 1.572 55.7 1.416 42.6 Lower secondary 114 149.7 171 125.7 217 118.9 57 69.4 89 48.0 120 44.5 Upper secondary 291 165.2 424 131.9 488 118.1 36 43.5 73 42.6 91 35.6 University 235 135.6 305 125.0 335 107.1 68 62.3 77 43.0 73 29.7 common across age groups for men and for women. (95%CI:1.5–2.4) in 2005–2008 and finally to a 2.1 The RII for men in the Basque Country also follows the (95%CI:1.5–2.8). This pattern repeats for each age group inverted U-shape across periods for each age group (ex- (except for those from 50 to 64 years that follows a cept for those over 74 years), while for women tend to in- U-shape trend). As said, none of these variations are sig- crease through time. For instance, in all age group the RII nificant so we must conclude stability in relative varies from 1.5 (95%CI:1.1–1.9) in 2001–2004 to a 1.9 inequalities. Fig. 1 Ischaemic heart disease age-standardized mortality rates and 95% confidence intervals. Barcelona and the Basque Country. 2001–2012 Bartoll et al. Archives of Public Health (2019) 77:12 Page 6 of 9 Table 3 Relative and absolute inequality measures of ischaemic heart disease mortality for Barcelona and the Basque Country, 2001–2012 Age group (years) 25–49 50–64 65–74 > = 75 All ages Region/Sex Period RII 95% CI RII 95% CI RII 95% CI RII 95% CI RII 95% CI Barcelona Men 2001–04 4.2 (1.8–10.0) 1.6 (1.2–2.3) 1.1 (0.9–1.5) 1.1 (0.9–1.2) 1.2 (1.0–1.4) 2005–08 4.9 (2.4–10.4) 2.3 (1.3–4.2) 1.6 (1.2–2.2) 1.1 (0.9–1.3) 1.3 (1.1–1.6) 2009–12 4.5 (1.9–10.1) 2.0 (1.2–3.5) 1.4 (0.8–2.5) 0.8 (0.7–0.9) 1.1 (0.9–1.3) Women 2001–04 1.8 (0.2–15.5) 3.7 (1.0–13.4) 2.2 (1.5–3.4) 1.5 (1.2–1.9) 1.7 (1.3–2.1) 2005–08 33.6 (3.5–322.2) 1.6 (0.6–4.2) 3.4 (1.7–6.6) 1.4 (1.2–1.7) 1.8 (1.5–2.3) 2009–12 7.9 (0.7–84.7) 2.0 (0.6–6.5) 3.2 (1.3–7.8) 1.4 (1.1–1.7) 1.5 (1.2–1.9) Basque Country Men 2001–04 3.3 (1.8–5.7) 1.4 (1.0–2.0) 1.2 (0.9–1.6) 1.4 (1.1–1.7) 1.5 (1.3–1.8) 2005–08 5.3 (2.7–10.3) 1.5 (1.0–2.3) 1.4 (1.1–1.9) 1.2 (0.9–1.4) 1.4 (1.2–1.7) 2009–12 4.1 (2.0–8.2) 1.5 (0.9–2.2) 1.1 (0.7–1.6) 1.2 (0.9–1.5) 1.3 (1.1–1.6) Women 2001–04 3.2 (0.7–14.4) 4.3 (1.2–15.1) 1.6 (0.9–2.9) 1.2 (0.9–1.5) 1.5 (1.1–1.9) 2005–08 4.3 (1.2–15.5) 1.4 (0.5–3.7) 2.1 (1.2–3.9) 1.6 (1.3–1.9) 1.9 (1.5–2.4) 2009–12 6.7 (1.2–38.7) 1.7 (0.4–6.8) 2.4 (1.2–4.8) 1.7 (1.4–2.1) 2.1 (1.5–2.8) Region/Sex Period SII 95% CI SII 95% CI SII 95% CI SII 95% CI SII 95% CI Barcelona Men 2001–04 14.9 (6.6–23.4) 39.7 (12.1–67.4) 38.7 (−32.4–109.9) 58.5 (−106.4–223.5) 29.0 (4.4–53.6) 2005–08 13.1 (7.6–18.5) 58.6 (27.0–90.1) 110.9 (49.4–172.5) 74.4 (−65.2–213.9) 40.6 (20.8–60.3) 2009–12 10.5 (5.4–15.6) 47.9 (20.0–75.8) 68.1 (−49.4–185.6) − 157.1 (− 280.7- -33.4) 9.2 (− 11.4–29.7) Women 2001–04 1.3 (− 1.1–3.8) 13.6 (1.1–26.0) 57.0 (27.5–86.5) 217.6 (132.9–302.2) 37.0 (21.9–52.1) 2005–08 2.9 (0.9–4.9) 3.2 (−4.9–11.4) 61.6 (33.4–89.8) 158.2 (78.3–238.1) 34.0 (23.9–44.0) 2009–12 2.0 (0.2–3.8) 5.4 (−3.2–14.0) 31.9 (−3.3–67.2) 121.3 (65.3–177.3) 20.4 (10.0–30.7) Basque Country Men 2001–04 15.3 (8.1–22.5) 27.7 (−0.4–55.7) 42.6 (−28.1–113.3) 338.1 (134.2–542) 69.2 (40.6–97.8) 2005–08 15.9 (9.7–22.1) 31.3 (0.7–61.9) 64.4 (9.3–119.5) 120.2 (−22.3–262.8) 38.1 (13.8–62.5) 2009–12 15.8 (9.9–21.7) 23.8 (−4.4–52.1) 16.5 (−47.9–80.9) 76.9 (−75.7–229.5) 26.4 (3.6–49.2) Women 2001–04 2.5 (0–5.01) 17.7 (7.23–28.2) 31.2 (−4.8–67.2) 103.7 (−2.7–210.2) 25.1 (7.9–42.3) 2005–08 3.2 (0.7–5.7) 2.8 (−5.0–10.6) 32.1 (5.6–58.6) 190.1 (114.3–266) 34.6 (22.6–46.6) 2009–12 1.8 (−0.4–4.1) 4.1 (−8.3–16.4) 27.2 (5.6–48.8) 157.9 (81.5–234.3) 29.1 (16.7–41.5) 95% confidence interval in parenthesis RII value can be interpreted as the relative inequalities in ischaemic heart disease mortality and SII as the absolute inequalities in ischaemic heart disease mortality comparing mortality rates between the two extremes of the educational spectrum (taking into account all educational levels) If we focus on SII in Table 3, we also observe that con- 2009–2012 holds for each age group except for women fidence intervals overlap across periods. It is worth men- from 50 to 64 years in both settings. tioning that SII decreases in the period of the economic crisis 2009–2012 for most groups. Comparing the two last periods, the SII in all age category for men in Discussion Barcelona decreases from 40.6 (95%CI:20.8–60.3) in We observed no significant trend in relative (RII) and 2005–2008 to a 9.2 (95%CI:-11.4–29.7) in 2009–2012, absolute (SII) inequalities in IHD mortality rates accord- from 34.0 (95%CI:23.9–44.0) to 20.4 (95%CI:10.0–30.7) ing to educational levels. Regardless the stability in time for women in Barcelona, from 38.1 (95%CI:13.8–62.5) to trend inequalities, data show a non-significant inverted 26.4 (95%CI:3.6–49.2) for men and from 34.6 U-shape trend for RIIs and for SII for some groups (95%CI:22.6–46.6) to 29.1 (95%CI:16.7–41.5) for women under 75, except for all ages women in the Basque coun- in Basque Country. The reduction in SII during period try and women from 50 to 64 years. Bartoll et al. Archives of Public Health (2019) 77:12 Page 7 of 9 The lowest relative and absolute inequalities observed to the economic downturn in period 2009–2012 may are among men. Cardiovascular disease epidemiology in- have avoided some of the IHD mortality mainly among dicates that unhealthy risk behaviours such as smoking lower socioeconomic groups contributing to a lowering and a high-fat diet, among others, may account for most in inequalities in IHD trends. of the socioeconomic gradient [19, 20], potentially affect- Our results differ from the increasing inequalities in ing at different degree the socioeconomic groups [20]. In mortality reported for USA during 1980 and 1990s eco- Spain, the transition of smoking from upper to lower so- nomic crisis, where individuals with low education ex- cioeconomic groups has been realized earlier for men perienced a declining health, while those with higher than for women, this may explain the lower inequalities education level presumably had some buffer-stock sav- observed in men due to the oldest only, especially in ings and prospects of avoiding long-term unemploy- Barcelona. In our study, the observed not decrease in ment [6]. A fact related to the meanstested provision of RII and SII among women from 50 to 64 years in both the public medical insurance program for Americans regions during the economic crisis, which may be related under age 65 years, leaving a large portion of the work- to the unfavourable trends in tobacco consumption for ing poor without medical insurance. Although women in the lower socioeconomic group [21]. Analo- non-significant, our results of a inverted U-shape in- gous dynamics of tobacco epidemics has been reported equalities found in some age groups in Spain gives a for the Basque Country [22]. Moreover, higher obesity weak support for this procyclical trend. In contrast, in- inequalities among Basque women compared to the equalities in Japanese males aged 30–59 during the eco- Spanish average may partially explain the observed in- nomic crisis of period 1980–2010 did not increased as creasing trend in inequalities in IHD [23]. A downward mortality rates among unemployed people steadily de- trend in fruits and vegetables intake has been reported creased for IHD perhaps because the basic properties in Spain [24] among the lowest socioeconomics groups of Japan’s social welfare system were maintained even and also in Greece [25] suggesting possible future in- during economic recession and mortality trends among creases in inequalities in cardiovascular mortality. the japanese managerial classes remained higher The SII decreased probably due to the improvement in relative to the other occupations due to overwork and access to clinical treatment and to primary care. In long working hours [9]. Again, in Norway and Finland Spain, the extensive use of drug treatment for vascular characterized with universal health care, disadvantaged so- risk factors during the last decades has been shown to cioeconomic groups were not hit harder by pro-cyclical be significantly associated with time decreasing rates of mortality than advantaged groups did [7, 8]. However in IHD hospitalisation [26]. The cuts on health spending New Zealand, the socioeconomic inequalities, due to the practiced during the economic crisis have affected strong restructuring of the economy, grew more than in mostly outpatient prescription drugs and patients in the Nordic as a consequence RII in cardiovascular mortal- long waiting visits time for health care, which is ex- ities increased [31]. pected to limit the accessibility to health services more For Spain, our results confirms previous research to those in the lower socioeconomic position. Certainly, reporting no changes in downward trends before and these austerity policies have not been so devastating as in during the economic crisis for cardiovascular disease the Greece [27] as to reverse the reductions in inequalities [11]. It has also been reported a non-significant increase in IHD trends achieved during the last decade [28]. For in inequalities in cardiovascular mortality during the instance, in Norway and Finland, both countries charac- economic crisis [32]. Although non-significant, our re- terized with universal health care, disadvantaged socioeco- sults of an inverted U-shape inequalities found in some nomic groups were not hit harder by pro-cyclical age groups gives a weak support for the procyclical mortality than advantaged groups did [7, 9]. trend. So that, the European Mediterranean countries Pollution could affect many individuals, but the effects most hit by the Great Recession, cardiovascular on vulnerable populations like the children, the elderly, mortality appears to follow different trajectories. In and most disadvantaged individuals are higher [29]. Greece, the downward trend in cardiovascular mortal- Although the inverted U-shape is not statically signifi- ity does not differ significantly between the pre- and cant, the decreasing of the absolute inequality in mortal- post-crisis periods, although the overall decline ended ity during the last period of economic crisis may be in 2011–2012 [33]. In Italy, cardiovascular mortality related to a decrease in vehicle traffic. In Barcelona area, was found to be positively associated with the un- the emission of nitrogen dioxide has been shown to be employment rate, and showed a significant increase in associated to IHD mortality in deprived areas for men 2010 [34]. In sum, our result of a stable trend in in- only [30]. Pollution may interact with basal worse health equalities in IHD mortality is akin with the recent status among the most disadvantaged population leading strand of research that stresses mild effects of eco- to higher mortality. A decrease in traffic congestion due nomic crisis on health outcomes [35, 36]. Bartoll et al. Archives of Public Health (2019) 77:12 Page 8 of 9 A limitation of our study is that we cannot determine 3. Ruhm CJ. A healthy economy can break your heart. Demography. 2007;44: causality between the observed trends in IHD mortality 829–48 Available from: http://www.ncbi.nlm.nih.gov/pubmed/18232214. 4. Mackenbach, J. P., Kulhánová, I., Artnik, B., Bopp, M., Borrell, C., Clemens, T., De inequalities and the economic crisis. Caution is needed in Gelder, R. (2016). Changes in mortality inequalities over two decades: Register generalising these results to other regions, as the impact based study of European countries. BMJ, 353. https://doi.org/10.1136/bmj.i1732. of economic recession may be different in each region. 5. Mackenbach JP, Kulhánová I, Menvielle G, Bopp M, Borrell C, Costa G, Lundberg O. Trends in inequalities in premature mortality: a study of 3.2 million deaths in 13 European countries. J Epidemiol Community Health. Conclusions 2015;69(3) https://doi.org/10.1136/jech-2014-204450. Our data confirm the existence of moderate IHD mor- 6. Edwards R. Who is hurt by procyclical mortality? Soc Sci Med. 2008;67:2051–8. 7. Valkonen T, Martikainen P, Jalovaara M, Koskinen S, Martelin T, Makela P. tality inequalities, with a gradient of increasing rates Changes in socioeconomic inequalities in mortality during an economic through less educational level, but we did not find evi- boom and recession among middle-aged men and women in Finland. Eur dence of increasing trends in IHD socioeconomic in- J Pub Health. 2000;10:274–80. 8. Haaland VF, Telle K. Pro-cyclical mortality across socioeconomic groups and equalities during the economic crisis neither in relative health status. J Health Econ. 2015;39:248–58. nor in absolute terms. 9. Wada K, Gilmour S. Inequality in mortality by occupation related to economic crisis from 1980 to 2010 among working-age Japanese males. Sci Abbreviations Rep. 2016;6:22255 Available from: http://www.pubmedcentral.nih.gov/ CI: Confidence interval; ICD-10: International Classification of Diseases 10th; articlerender.fcgi?artid=4776242&tool=pmcentrez&rendertype=abstract. IHD: Ischaemic heart disease; OECD: Organization for Economic Co-operation 10. Osman M, Osman A. Cardiovascular mortality in Northern Ireland during the and Development; RII: Relative index of inequality; SII: Slope index of inequality 2008–2014 financial crisis years: who got the worst hit? Epidemiol Biostat Public Heal. 2017;14(3):1–6. Acknowledgements 11. Regidor E, Vallejo F, Granados JAT, Viciana-Fernández FJ, de la Fuente L, Barrio Not applicable. G. Mortality decrease according to socioeconomic groups during the economic crisis in Spain: a cohort study of 36 million people. Lancet. 2016;388:2642–52. Funding 12. Borrell C, Rodríguez-Sanz M, Bartoll X, Malmusi D, Novoa AM. El sufrimiento This article has been partially funded by the project “Efectos de la crisis de la población en la crisis económica del Estado español. Salud Colect. en la salud de la población y sus determinantes en España” (PI13/00897), 2014;10:95–8. funded by the Instituto de Salud Carlos III, and co-funded by European Union 13. Toffolutti V, Suhrcke M. Assessing the short term health impact of the great (ERDF/ESF, “Investing in your future”). recession in the European Union: a cross-country panel analysis. Prev Med (Baltim). 2014;64:54–62. Availability of data and materials 14. Karanikolos M, Mladovsky P, Cylus J, Thomson S, Basu S, Stuckler D, et al. Datasets are not publicly available but are available on reasonable request. Financial crisis, austerity, and health in Europe. Lancet. 2013;381:1323–31. 15. Fundación BBVA e Ivie (Instituto Valenciano de Investigaciones Económicas). Authors’ contributions Gasto sanitario público en España. 2013 [cited 2016 Nov 20]. Available from: XB, MG, MM, LP, MC, SE and CB designed the study in collaboration with the https://www.ivie.es/es_ES/bases-de-datos/sector-publico/gasto-sanitario-publico/ other authors. The data were prepared by MG, MM, LP, SE and MC. The 16. INE. EAPS. Economically Active Population Survey. 2016 [cited 2016 Nov 20]. analysis was conducted by MG. The first draft of the manuscript was written Available from: http://www.ine.es/dynt3/inebase/es/index.htm?padre= by XB. Redrafting was conducted with the participation of all authors. All 811&capsel=815 authors gave feedback on the final manuscript and approved it for submission. 17. IDESCAT. Poverty rates. Living conditions. 2016 [cited 2016 Nov 20]. Available from: https://www.idescat.cat/pub/?id=aec&n=416 Ethics approval and consent to participate 18. Moreno-Betancur M, Latouche A, Menvielle G, Kunst AE, Rey G. Relative No ethical approval and consent was necessary. index of inequality and slope index of inequality: a structured regression framework for estimation. Epidemiology. 2015;26:518–27. Consent for publication 19. Lynch JW, Davey Smith G, Harper S, Bainbridge K. Explaining the social Not applicable. gradient in coronary heart disease: comparing relative and absolute risk approaches. J Epidemiol Community Health. 2006;60:436–41. Competing interests 20. Harper S, Lynch J, Smith GD. Social determinants and the decline of The authors declare that they have no competing interests. cardiovascular diseases: understanding the links. Annu Rev Public Health. 2011;32:39–69 Available from: http://www.annualreviews.org/doi/10.1146/ Publisher’s Note annurev-publhealth-031210-101234. Springer Nature remains neutral with regard to jurisdictional claims in 21. Fernandez E, Garcia M, Schiaffino A, Borras JM, Nebot M, Segura A. Smoking published maps and institutional affiliations. initiation and cessation by gender and educational level in Catalonia, Spain. Prev Med (Baltim). 2001;32:218–23. Author details 22. Bacigalupe A, Esnaola S, Martín U. Evolución de la prevalencia y las 1 desigualdades sociales en el consumo de tabaco en la C.A. de Euskadi: Agència de Salut Pública de Barcelona, Plaça Lesseps 1, 08023 Barcelona, Spain. 2CIBER Epidemiología y Salud Pública (CIBERESP), Madrid, Spain. 1986-2007. Osagin Txostenak 2011-1. Vitoria-Gasteiz: Departamento de 3 Institut d’Investigació Biomèdica (IIB Sant Pau), Barcelona, Spain. 4Universitat Sanidad y Consumo; 2011. Pompeu Fabra, Barcelona, Spain. 5Departamento de Salud, Gobierno del País 23. Bacigalupe A, Eikemo TA, Borrell C, Regidor E, Kulha I, Esnaola S, et al. Why Vasco, Bilbao, Spain. does Spain have smaller inequalities in mortality? An exploration of potential explanations; 2014. p. 370–7. Received: 8 November 2018 Accepted: 20 February 2019 24. Bartoll X, Toffolutti V, Malmusi D, Palència L, Borrell C, Suhrcke M. Health and health behaviours before and during the great recession, overall and by socioeconomic status, using data from four repeated cross-sectional References health surveys in Spain (2001–2012). BMC Public Health. 2015;15(1):865. 1. Harvey Brenner M. Mortality and the National Economy. A review, and the Available from: http://bmcpublichealth.biomedcentral.com/articles/10.1186/ experience of England and Wales, 1936-76. Lancet. 1979;314:568–73. s12889-015-2204-5. 2. Wagstaff A. Time series analysis of the relationship between unemployment 25. Filippidis FT, Schoretsaniti S, Dimitrakaki C, Vardavas CI, Behrakis P, Connolly and mortality: a survey of econometric critiques and replications of GN, et al. Trends in cardiovascular risk factors in Greece before and during the Brenner’s studies. Soc Sci Med. 1985;21:985–96. financial crisis: the impact of social disparities. Eur J Pub Health. 2014;24:974–9. Bartoll et al. Archives of Public Health (2019) 77:12 Page 9 of 9 26. Medrano MJ, Alcalde-Cabero E, Ortíz C, Galán I. Effect of cardiovascular prevention strategies on incident coronary disease hospitalisation rates in Spain; an ecological time series analysis. BMJ Open. 2014;4:e004257 Available from: http://www.pubmedcentral.nih.gov/articlerender.fcgi?artid= 3927998&tool=pmcentrez&rendertype=abstract. 27. Simou E, Koutsogeorgou E. Effects of the economic crisis on health and healthcare in Greece in the literature from 2009 to 2013: a systematic review. Health Policy. 2014;115:111–9. 28. Lopez-Valcarcel BG, Barber P. Economic Crisis, Austerity Policies, Health and Fairness: Lessons Learned in Spain. Appl Health Econ Health Policy. 2017; 15(1):13–21. 29. Laurent O, Bard D, Filleul L, Segala C. Effect of socioeconomic status on the relationship between atmospheric pollution and mortality. J Epidemiol Community Health. 2007;61:665–75. 30. Barceló MA, Saez M, Saurina C. Spatial variability in mortality inequalities, socioeconomic deprivation, and air pollution in small areas of the Barcelona metropolitan region, Spain. Sci Total Environ. 2009;407:5501–23. 31. Blakely T, Tobias M, Atkinson J. Inequalities in mortality during and after restructuring of the New Zealand economy: repeated cohort studies. BMJ. 2008;336(7640):371–75. 32. Regidor E, Barrio G, Bravo MJ, De Fuente L. Has health in Spain been declining since the economic crisis? 2013. p. 4–6. 33. Vlachadis N, Iliodromiti Z, Vlachadi M, Xanthos T, Ktenas E, Vrachnis D, et al. Cardiovascular mortality and the financial crisis in Greece: trends and outlook. Int J Cardiol. 2014;176(3):1367–8. 34. Mattei G, Ferrari S, Rigatelli M. Short-term effects of the 2008 Great Recession on the health of the Italian population: an ecological study. Soc Psychiatry Psychiatr Epidemiol. 2014;49(6):851–8. 35. Ruhm CJ. Recessions, healthy no more? J Health Econ. 2015;42:17–28. 36. Tekin E, McClellan C, Minyard KJ. Health and Health Behaviors During the Worst of Times: Evidence from the great recession. National bureau of economic research working Paper No 19234. 2013.