Trias‑Llimós et al. International Journal for Equity in Health (2022) 21:51 https://doi.org/10.1186/s12939-022-01650-3 RESEARCH Open Access Socioeconomic inequalities in physiological risk biomarkers and the role of lifestyles among Russians aged 35‑69 years Sergi Trias‑Llimós1,2* , Sarah Cook1,3, Anne Elise Eggen4, Alexander V. Kudryavtsev4,5, Sofia Malyutina6,7, Vladimir M. Shkolnikov8,9 and David A. Leon1,4 Abstract Background: Socioeconomic inequalities in cardiovascular (CVD) health outcomes are well documented. While Russia has one of the highest levels of CVD mortality in the world, the literature on contemporary socio-economic inequalities in biomarker CVD risk factors is sparse. This paper aims to assess the extent and the direction of SEP inequalities in established physiological CVD risk biomarkers, and to explore the role of lifestyle factors in explaining SEP inequalities in physiological CVD risk biomarkers. Methods: We used cross-sectional data from a general population-based survey of Russians aged 35-69 years liv‑ ing in two cities (n = 4540, Know Your Heart study 2015-18). Logistic models were used to assess the associations between raised physiological risk biomarkers levels (blood pressure levels, cholesterol levels, triglycerides, HbA1C, and C-reactive protein) and socioeconomic position (SEP) (education and household financial constraints) adjusting for age, obesity, smoking, alcohol and health-care seeking behavior. Results: High education was negatively associated with a raised risk of blood pressure (systolic and diastolic) and C-reactive protein for both men and women. High education was positively associated with total cholesterol, with higher HDL levels among women, and with low triglycerides and HbA1c levels among men. For the remaining risk biomarkers, we found little statistical support for SEP inequalities. Adjustment for lifestyle factors, and particularly BMI and waist-hip ratio, led to a reduction in the observed SEP inequalities in raised biomarkers risk levels, especially among women. High financial constraints were weakly associated with high risk biomarkers levels, except for strong evidence for an association with C-reactive protein (men). Conclusions: Notable differences in risk biomarkers inequalities were observed according to the SEP measure employed. Clear educational inequalities in raised physiological risk biomarkers levels, particularly in blood pressure and C-reactive protein were seen in Russia and are partly explained by lifestyle factors, particularly obesity among women. These findings provide evidence-based information on the need for tackling health inequalities in the Rus‑ sian population, which may help to further contribute to CVD mortality decline. Keywords: Cardiovascular risk, Health inequities, Health behaviours, Russia, Eastern Europe Background Cardiovascular disease (CVD) is the largest single *Correspondence: strias@ced.uab.cat 2 Centre d’Estudis Demogràfics, Centres de Recerca de Catalunya (CERCA), cause of death worldwide [1]. CVD mortality is par- Carrer de Ca n’Altayó, Edifici E2, Universitat Autònoma de Barcelona, ticularly high among middle aged individuals in Eastern 08193 Bellaterra, Spain Europe, and Russia in particular [1, 2]. For example, the Full list of author information is available at the end of the article © The Author(s) 2022. Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made. The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder. To view a copy of this licence, visit http://​creat​iveco​mmons.​org/​licen​ses/​by/4.​0/. The Creative Commons Public Domain Dedication waiver (http://​creat​iveco​ mmons.​org/​publi​cdoma​in/​zero/1.​0/) applies to the data made available in this article, unless otherwise stated in a credit line to the data. Trias‑Llimós et al. International Journal for Equity in Health (2022) 21:51 Page 2 of 11 age-standardized rates of 1212 per 100,000 among Rus- found clear educational gradients in established cardio- sian males and 756 per 100,000 among Russian females vascular factors and markers of inflammation [25], in in 2014 contrasts with the equivalent rate in the neigh- which biomarkers were found to explain 19 and 36% of bouring country of Norway of 287 and 209 per 100,000 educational differences in general health and physical observed among males and females, respectively (in function, respectively. A further analysis of SAHR data 2016) [1]. The potential contribution of lifestyle and soci- looked at educational impacts on all-cause and CVD oeconomic factors to CVD health has been described in mortality [19]. This confirmed the importance of estab- several reviews [3–7]. The elevated Russian CVD mortal- lished risk factors and markers of inflammation in medi- ity is driven by multiple factors including individual-level ating the relationship between level of education and determinants such as lifestyle and established cardio- death. However, despite these previous studies, there has metabolic factors [8], and levels of investment in the not been any focussed attempt to look at SEP inequalities health sector, particularly cardiovascular treatment tech- and physiological CVD risk factors with individual-level nology and staff training [9, 10]. lifestyle factors in Russia. The existence of potentially large socioeconomic ine- This paper aims to contribute to the understanding of qualities and high mortality in disadvantaged groups may SEP inequalities in CVD in Russia by addressing the fol- also contribute to overall levels of CVD mortality [11, lowing objectives: 1) To assess the extent and the direc- 12]. Socioeconomic position inequalities (SEP) refers to tion of SEP inequalities in established physiological CVD the social and economic factors that influence what posi- risk biomarkers, and 2) to explore the role of lifestyle fac- tions individuals or groups hold within the structure of a tors in explaining SEP inequalities in physiological CVD society [13]. Indeed, SEP inequalities in CVD incidence risk biomarkers. We hypothesize that SEP inequalities in and mortality are important in Europe across a wide risk biomarkers exist, and we will test the extent to which range of socioeconomic position measures [14, 15]. East- this is true for the biomarkers analyzed, using two alter- ern European countries, includin Russia, are known for native measures of SEP, and the extent to which behav- their large socioeconomic inequalities in all-cause and ioural factors mitigate theses associations. cardiovascular mortality [9, 11, 12, 15–20]. However, less is known about socioeconomic differences in biomarkers Methods and physiological pathways underlying the large inequali- To fulfill our objectives, we use two measures of SEP: ties in mortality and health in Russia. highest educational attainment and self-perceived house- There is consistent evidence that riskier lifestyles and hold financial constraints. We used data from two large risk factor profiles exist in Russia in comparison with provincial Russian cities. Our focus on working age Western European countries, at least in high alcohol use groups is consistent with the common finding that SEP [21], smoking [22] and high blood pressure levels [23]. differentials for CVD are particularly pronounced at this Lifestyle factors, such as smoking or alcohol, and estab- younger age. lished CVD risk factors, such as raised blood pressure and cholesterol levels, predict CVD health outcomes, and Data provide potential pathways along with socioeconomic We used data from the population-based cross-sectional circumstances impact on CVD. For example, results survey Know Your Heart (KYH) study carried out in men derived from the HAPIEE study found a clear educational and women aged 35-69 years in the cities of Arkhangelsk gradient in CVD mortality in Eastern Europe, which and Novosibirsk (Russia) between 2015 and 2018 [26]. diminished after adjusting for established CVD risk fac- The KYH Study is one of the major components of The tors [24]. A small number of other studies have focused International Project on Cardiovascular Disease in Rus- on links between socioeconomic inequalities and health sia (IPCDR), which aimed to provide a more compre- outcomes in Russia. Several of these have explored the hensive understanding of the reasons for the extremely mediating role of established cardiometabolic risk factors high cardiovascular mortality in Russia, and which pro- using data from the Lipid Research Clinics (LRC) and vides one of the most recent and detailed health survey MONICA studies aged 35 to 64 years, and they found and biomarker information for Russian populations. that established risk factors can potentially explain 20 KYH consisted of a baseline interview completed in par- to 30% of the educational differences in the risk of CVD ticipants’ homes by a trained interviewer and a health mortality [18, 20]. check at a polyclinic (a health centre providing outpatient More recent studies have focused on older ages using care) completed by medical professionals (both doctors data from the Survey on Stress, Ageing and Health in and nurses). Detailed standard operating procedures Russia (SAHR) collected in 2006-2009 (1800 Moscow were followed to ensure standard collection of the physi- residents aged 55 and older, 68 years on average). This ological parameters in each site. Review of the quality of Trias‑Llimós et al. International Journal for Equity in Health (2022) 21:51 Page 3 of 11 data from both sites was conducted centrally monthly alcohol use (non-drinkers, and two drinking catego- throughout the study and feedback provided to the clini- ries using the Alcohol Use Disorders Identification Test cal sites. Within each of the two cities, four districts were (AUDIT) score [29] cut point of ≥8 to indicate hazardous selected purposefully to represent a range of socio-demo- or harmful drinking [30, 31]); physical activity (floors of graphic and mortality levels in each city. The sampling stairs climbed per day over the last 12 months) [32, 33]; frame was provided by the regional health insurance body mass index (BMI, in kg/m2: < 25, > = 25 and < 30, funds and included addresses of all residents having com- and > =30); waist-hip ratio (in cm, dichotomous, cut- pulsory health insurance, supplemented by information off 0.85 for women and 0.90 for men) [34]; one item on on the age and sex of occupants at individual addresses. diet from the Dietary Quality score (fruit intake) [35]. In Overall, 4560 individuals completed the interview and addition, we adjusted for medications, again to see how took part in the health check: a response rate of 51% out far these factors might drive the observed inequalities of the total number of those invited. Further details con- in the CVD biomarkers (blood pressure lowering medi- cerning response rates, and organization of interview- cation (Anatomical Therapeutic Chemical Classification ing and medical examination can be found in the KYH System codes [36] C02-C03, C07-C09), lipid lowering descriptive study by Cook et al. [26]. medication (statins C10) and blood sugar lowering medi- Dependent variables included in this study were cation (A10A (insulin), A10B (oral blood sugar lowering derived by dichotomising levels of physiological CVD drugs)); visits to the doctor over the last 12 months; and biomarkers. The risk biomarkers cut off points were self-reported familiarity with arterial hypertension (high those used in a previous SAHR study conducted in Mos- blood pressure). cow [25], except for LDL which was not included in We used complete case data for each (physiological) SAHR: systolic blood pressure (SBP, cut off ≥140 mmHg), risk biomarker. Individuals with missing values in any of diastolic blood pressure (DBP, cut off ≥90 mmHg), total the variables of interest were excluded from the analysis cholesterol (cut off ≥6.216 mmol/L), LDL cholesterol (410 (9.0%) blood pressure, 140 lipids (3.1%), 195 HbA1c (cut off ≥3.8 mmol/L corresponding ot), HDL choles- (4.3%), and 84 CRP (1.9%)). terol (cut off < 1.036 mmol/L), and triglycerides (cut off To explore whether our findings were influenced by ≥2.26 mmol/L), glycated hemoglobin (HbA1C, cut off the individuals with pre-existing diseases, we perform a > 6.5%), and C-reactive protein (CRP, cut off > 3 mg/L). sensitivity analysis by excluding individuals who reported These risk biomarkers are all associated with higher mor- having had a myocardial infarction or stroke and or had tality risks, and they were selected based on the robust- angina on the Rose Angina Questionnaire (short version) ness and availability in the KYH. Details on the KYH [18] (see Table S1). sample, including the methods behind the risk biomark- ers measurements, can be found elsewhere [26]. Analytical approach We conducted parallel analyses using two alternate All analyses were stratified by gender. We used direct measures of self-reported SEP: educational attainment age-standardization to estimate prevalences of high-risk (education) and household financial constraints (finan- profiles by SEP (including 95%CI and p-values of the dif- cial constraints). Education was dichotomized as high ference between educational groups) using the Russian SEP (higher or incomplete higher education vs lower population from 2014 as the standard from the Human SEP (secondary or lower education) [27]. Financial con- Mortality Database (HMD). straints were dichotomized as follows: high SEP for Logistic regression models were run to assess the asso- those able to buy non-essential products (no or mini- ciations between SEP measures (level of education and mal constraint of not being able to buy large domestic financial constraints) and physiological risk biomark- appliances), and lower SEP for those with constrains to ers (prevalence of raised biomarker levels). In both sets buy essential products (constrained in buying food or of models we adjusted for age, city (Model 1), and a set clothes). We acknowledge that the dichotomized SEP of additional variables. Model 2 included the lifestyle variables represent a limitation in terms of assessing a full variables smoking, alcohol (AUDIT-8), diet (fruit intake), spectrum of educational inequalities. and physical activity (floors of stairs climbed). Model 3 We treated age (5-year age groups) and city (Arkhan- included the variables related to overweight and obesity: gelsk or Novosibirsk) as a priori potential confounders. BMI and waist-hip ratio. Model 4 included all variables However, we also included in models lifestyle related- included in previous models, whereas in model 5 we factors to see how far these might drive any observed additionally adjusted for risk factor specific medication inequalities in the CVD biomarkers: smoking behav- use. Finally, for blood pressure, an additional model was iour (never, former, up to 10 cigarettes per day, 10-20 added to account for health literacy (using blood pres- cigarettes per day and over 20 cigarettes per day) [28], sure knowledge as a proxy) although the results are not Trias‑Llimós et al. International Journal for Equity in Health (2022) 21:51 Page 4 of 11 shown as this had not effect. A sensitivity analysis using waist-hip ratio did not explain the associations between continuous risk factor biomarkers as dependent variables high-risk CRP levels and education (model 3, OR: 0.77 and linear regression models is included in Supplemen- (0.62-0.95). Results for household financial constraints tary Material. suggested weak SEP associations except for CRP. Among women, there were negative associations Results between higher education and raised blood pressure. Our sample consisted of 1901 men and 2637 women. For These were attenuated after adjusting for lifestyles, both men and women around 40% were classified as hav- BMI and waist-hip ratio (model 4, OR for SBP: 0.82 ing high education, whereas notable gender differences (0.66-1.02); OR for DBP: 0.81 (0.64-1.02)). Finally, the were observed when considering financial constraints associations between CRP and education weakened as SEP. Among men, 34% were classified as of low SEP after adjusting for obesity-related measures. Results based on financial constraints, whereas among women for household financial constraints suggested SEP dif- this figure was much higher at 72%. Table 1 provides the ferences in high-risk CRP levels (model 1, OR: 0.66 descriptive characteristics of our sample. (0.53-0.82), and the OR with statistical evidence for a difference in all models. Age‑standardized risk factor means Sensitivity analyses for continuous risk factor meas- Age-standardized prevalence of raised physiological risk ures derived from linear regression models were biomarkers by educational attainment suggest men and broadly consistent with our main analyses based on women without higher education have a higher preva- binary outcomes with lower mean levels of SBP among lence of high-risk blood pressure and CRP levels, and the higher educated group in both men (beta: − 2.38 lower prevalence of raised total cholesterol levels among (− 4.20, − 0.57) and women (beta: − 3.11 (− 4.61, women as compared to their counterparts with higher − 1.62) (Table S2). Similarly, inequalities were observed education (Table 2). Men with financial constraints had in DBP. These inequalities persisted in the multivariable a higher prevalence of raised triglycerides and CRP com- models, except for DBP levels among women, which pared with those free of financial constraints, while the declined after adjusting for lifestyles and obesity- results for women and for the other physiological risk related measures. However, in contrast to that observed biomarkers did not suggest the presence of inequalities in the main results, higher educated men had lower according to financial constraints. levels of HbA1c, and these inequalities weakened after adjusting for major lifestyles factors such as smoking Regression models and alcohol (model 2). The significance of CRP levels High level of education was negatively associated with was similar to the one presented in the main results. raised levels of blood pressure and CRP for both men Finally, higher educated women presented higher HDL and women, as well as for HDL for women (Tables 3 levels and lower HbA1c. In both cases, these differences and 4). For example, the odds ratios (OR) of high sys- weakened after adjusting for obesity-related measures tolic blood pressure were 0.78 (95%CI; 0.64-0.96) (model 3). Similar to the findings for the high-risk fac- among men and 0.73 (0.60-0.90) among women, while tor profiles, there was a lack of evidence supporting for high CRP the corresponding values were 0.78 (0.63- clear inequalities according to the financial status. 0.96) among men and 0.66 (0.55-0.79) among women. High education was associated with raised total choles- terol levels for women. Financial constraints were found Discussion to have weak associations with risk biomarkers preva- This study has found higher education to be associated lence, Among men, the associations between raised with a reduced probability of raised blood pressure (SBP blood pressure prevalence and education persisted and DBP) and CRP for both men and women, with raised after adjusting for all lifestyle-related variables and HDL levels and lower total cholesterol levels for women. blood pressure lowering medication (e.g. OR for SBP in For the remaining physiological CVD risk biomarkers model 2 was 0.76 (0.62-0.94)) (Tables 3 and 4). There analyzed, we found little statistical evidence of SEP ine- was only weak evidence of any associations of educa- qualities. High financial constraints were weakly associ- tion with HDL, LDL, triglycerides and HbA1c among ated with high risk biomarkers levels, except for strong men regardless of model. Finally, the associations evidence for higher levels of CRP in men. Therefore, our between high-risk CRP levels and education weakened initial hypothesis on the existence of SEP inequalities is after adjusting for lifestyles (model 2, OR: 0.91 (0.73- thus only partly confirmed. Lifestyle factors appear to be 1.14), which is explained by smoking (model 1b, OR: the driver of some of these observed SEP inequalities in 0.92 (0.74-1.14), results not shown in table). BMI and risk factor profiles. Trias‑Llimós et al. International Journal for Equity in Health (2022) 21:51 Page 5 of 11 Table 1 Descriptive characteristics of the sample, Know Your Heart, ages 35-69 Men (n = 1901) Women (2637) Variable Categories n % n % City Arkhangelsk 989 52% 1389 53% Novosibirsk 912 48% 1248 47% Age 35-39 160 8% 236 9% 40-44 225 12% 350 13% 45-49 255 13% 349 13% 50-54 280 15% 380 14% 55-59 291 15% 411 16% 60-64 344 18% 425 16% 65-69 346 18% 486 18% Education Higher (high SEP) 771 41% 1102 42% Lower (low SEP) 1130 59% 1535 58% Financial constraints Constrained (low SEP) 644 34% 1893 72% Unconstrained (high SEP) 1212 64% 707 27% Missing 45 2% 37 1% AUDIT 8 Non-drinker 284 15% 341 13% Lower than 8 1156 61% 2234 85% Equal or higher than 8 448 24% 52 2% Missing 13 1% 10 0% Smoking Never 523 28% 1807 69% Former 677 36% 399 15% Up to 10 140 7% 229 9% More than 10, up to 20 424 22% 175 7% More than 20 128 7% 25 1% Missing 9 0% 2 0% BMI kg/m 2 Normal (< 25.0) 554 29% 819 31% Overweight (25.0-29.9) 839 44% 848 32% Obesity (30.0+) 501 26% 959 36% Missing 7 0% 11 0% Waist-hip quartile Low 470 25% 1452 55% High 1427 75% 1183 45% Missing 4 0% 2 0% Physical activity: stairs (floors per day) Mean 8.12 6.15 Diet (fruit intake) None 149 8% 99 4% 1-2 per week 446 23% 347 13% 3-4 per week 382 20% 410 16% 5-6 per week 111 6% 208 8% 1-2 per day 685 36% 1244 47% 3-4 per day 93 5% 241 9% 5-6 per day 21 1% 56 2% More than 6 per day 12 1% 31 1% Missing 2 0% 1 0% Knowledge about Blood Pressure Not at all 162 9% 113 4% I have only heard the term before 236 12% 169 6% I know a little about it 631 33% 854 32% I am very familiar with it 870 46% 1500 57% Missing 2 0% 1 0% Blood pressure lowering medication Yes 646 34% 1060 40% No 1255 66% 1577 60% Trias‑Llimós et al. International Journal for Equity in Health (2022) 21:51 Page 6 of 11 Table 2 Age-standardized prevalence (in %) of raised risk dimensions [13] and have been shown to be differently physiological CVD biomarkers* by education** and household related to CVD health [37]. Education is usually com- financial constraints in the Know Your Heart Study pleted by early adulthood and therefore is an indicator Men of SEP that is unlikely to change in response to ill-health Low SEP High SEP p-value later in life [38]. Recent evidence from a Mendelian ran- Education SBP 38.1 (35.0-41.2) 31.9 (28.4-35.3) 0.018 domization study provided evidence of a causative role DBP 35.3 (32.2-38.4) 28.8 (25.4-32.3) 0.013 of education in CVD risks [39]. In contrast, household HDL 43.1 (39.8-46.3) 45.0 (41.4-48.7) 0.081 financial constraints reflect the current economic situ- ation of the household. The fact that education seems a LDL 19.2 (16.6-21.7) 16.7 (13.9-19.5) 0.527 stronger predictor of SEP inequalities in blood-pressure TCHOL 21.9 (19.3-24.5) 21.6 (18.6-24.6) 0.994 related risk factors as compared to household financial Triglycerides 19.5 (16.9-22.1) 18.7 (15.8-22.5) 0.803 constraints is in line with another study using the same HbA1C 4.9 (3.7-6.2) 3.9 (2.6-5.1) 0.483 dataset but focusing on primary care [40]. Therefore, it CRP 30.0 (27.0-32.9) 24.5 (21.3-27.6) 0.020 is plausible to think that the educational opportunities Financial SBP 37.1 (34.1-40.0) 34.5 (30.7-38.2) 0.433 constraints and family background, reflected in attained education, DBP 32.3 (29.5-37.1) 33.3 (29.5-37.1) 0.754 have a more direct role in determining both lifestyles and HDL 44.0 (40.9-47.1) 44.7 (40.7-48.7) 0.340 subsequent biomarkers profiles, particularly blood pres- LDL 18.8 (16.4-21.3) 17.5 (14.4-20.6) 0.676 sure and CRP, than the current financial situation as cap- TCHOL 22.1 (19.5-24.6) 21.4 (18.2-24.6) 0.953 tured by financial constraints. This implies differences are Triglycerides 21.2 (18.6-23.8) 15.5 (12.6-18.3) 0.015 driven throughout the life course, and that current cross- HbA1C 4.8 (3.6-6.0) 3.8 (2.4-5.2) 0.223 sectional SEP measures are less informative as regards to CRP 30.5 (27.7-33.3) 23.4 (20.0-26.8) < 0.001 inequalities in physiological risk biomarker levels. Women Strong SEP inequalities were observed for blood pres- Low SEP High SEP p-value sure measures (SBP and DBP) analysed as continuous Education SBP 22.9 (20.8-25.0) 18.7 (16.3-21.0) 0.004 variables. These educational inequalities were slightly DBP 19.8 (17.6-22.0) 14.6 (12.4-16.9) < 0.001 attenuated after adjustment for BMI and waist-hip ratio HDL 41.0 (38.4-43.5) 42.6 (39.7-45.4) 0.020 among women: for example the coefficient for DBP LDL 5.7 (4.4-6.9) 3.6 (2.4-4.7) 0.555 declined from − 1.53 (− 2.43 to − 0.63) to − 0.82 (− 1.69 TCHOL 23.5 (21.4-25.6) 27.6 (25.0-30.2) 0.021 to 0.05) on adjustment as shown in Table S2). However, Triglycerides 13.3 (11.5-15.1) 11.4 (9.4-13.3) 0.214 this was not seen for men. These results are in line with HbA1C 4.5 (3.5-5.5) 3.9 (2.8-5.1) 0.592 evidence on the importance of obesity in SEP differen- CRP 33.9 (31.2-36.5) 25.6 (22.9-28.2) < 0.001 tials among Russian women, but not among Russian men Financial SBP 21.2 (19.4-23.0) 19.1 (16.2-22.0) 0.201 constraints [41]. The fact that the other lifestyles (smoking, alcohol, DBP 17.7 (15.9-19.6) 16.6 (13.8-19.5) 0.448 fruit consumption and physical activity) contribute very HDL 41.9 (39.6-44.2) 41.6 (38.2-45.1) 0.198 little to the blood pressure profiles does not seem to be LDL 5.0 (4.0-6.1) 3.7 (2.3-5.1) 0.981 attributed to the categorization of the selected lifestyles, TCHOL 25.3 (23.4-27.3) 25.1 (22.0-28.2) 0.948 as different questions and categorizations were carefully Triglycerides 12.7 (11.2-14.3) 12.2 (9.8-14.6) 0.922 explored. For example, the frequency of alcohol con- HbA1C 4.3 (3.4-5.2) 3.9 (2.6-5.3) 0.930 sumed or AUDIT score in a continuous scale did not CRP 31.2 (29.0-33.4) 28.0 (24.6-31.4) 0.098 explain more than the dichotomous AUDIT score ≥ 8 * SBP stands for systolic blood pressure ≥ 140 mmHg, DBP for diastolic blood variable used in these analyses. pressure ≥ 90 mmHg, HDL for high density lipoprotein < 1.036 mmol/L, LDL for low density lipoprotein ≥3.8 mmol/L, TCHOL for total cholesterol in blood Alcohol has been established as a critically important ≥6.216 mmol/L, HbA1C for glycated haemoglobin > 6.5%, and CRP for C-reactive risk factor in Russia, so the fact that it did not contrib- protein > 3 mg/L ute to lowering the observed SEP inequalities in blood ** Education was categorized in: high SEP (higher or incomplete higher education), and lower SEP (secondary or lower education); and financial pressure levels deserves to be discussed. We need to constraints were categorized in: unconstained or high SEP (self-perceived acknowledge the important drops in harmful alcohol not to have any financial constraint or only constrained to by large domestic consumption and deaths from acute alcohol poisoning in appliances), and constrained or low SEP (constrained in buying food or clothes) Russia over the last 15 years [42]. These trends seem to be pointing towards important reductions in binge drink- Explanation of results ing. Previous studies on SEP in alcohol consumption in In this study we have used two different measure of SEP: Russia suggest that while low SEP groups are more likely education and household financial constraints. Each to engage in hazardous drinking they do not necessarily of these measures represent related but different SEP have an overall higher mean volume of alcohol consumed Trias‑Llimós et al. International Journal for Equity in Health (2022) 21:51 Page 7 of 11 Table 3 Odds ratios for raised risk physiological CVD biomarkers* by education and household financial constraints, men (n = 1901) Men Model 1: Age + city Model 2: Model Model 3: Model Model 4: Models Model 5: Model 1 + smoking 1 + BMI + waist-hip 2 + 3 4 + medication + alcohol (AUDIT8) + fruit + stairs Education (ref. Low) SBP 0.784 (0.638, 0.962) 0.774 (0.625, 0.958) 0.793 (0.645, 0.975) 0.784 (0.632, 0.971) 0.778 (0.628, 0.965) DBP 0.776 (0.630, 0.955) 0.748 (0.602, 0.929) 0.779 (0.631, 0.963) 0.758 (0.608, 0.945) 0.756 (0.607, 0.943) HDL 0.803 (0.628, 1.026) 0.870 (0.672, 1.126) 0.801 (0.623, 1.029) 0.879 (0.676, 1.144) 0.876 (0.673, 1.139) LDL 1.041 (0.862, 1.258) 1.011 (0.830, 1.232) 1.043 (0.861, 1.263) 1.028 (0.842, 1.256) 1.031 (0.843, 1.261) TCHOL 0.964 (0.770, 1.208) 0.927 (0.733, 1.174) 0.955 (0.761, 1.198) 0.926 (0.731, 1.174) 0.926 (0.730, 1.174) Triglycerides 0.951 (0.749, 1.207) 0.990 (0.771, 1.271) 0.948 (0.739, 1.216) 1.016 (0.784, 1.318) 1.014 (0.782, 1.315) HbA1C 0.846 (0.553, 1.293) 0.902 (0.580, 1.404) 0.825 (0.534, 1.276) 0.962 (0.611, 1.516) 0.954 (0.603, 1.511) CRP 0.778 (0.631, 0.959) 0.912 (0.731, 1.139) 0.768 (0.621, 0.949) 0.911 (0.727, 1.142) 0.908 (0.724, 1.138) Financial constraints SBP 0.910 (0.735, 1.126) 0.935 (0.750, 1.165) 0.930 (0.750, 1.154) 0.957 (0.766, 1.195) 0.971 (0.776, 1.214) (ref. Low) DBP 1.007 (0.814, 1.248) 1.034 (0.827, 1.292) 1.021 (0.820, 1.269) 1.056 (0.841, 1.326) 1.062 (0.845, 1.333) HDL 0.885 (0.685, 1.143) 0.923 (0.709, 1.203) 0.873 (0.672, 1.134) 0.932 (0.711, 1.223) 0.953 (0.726, 1.251) LDL 1.091 (0.895, 1.330) 1.078 (0.879, 1.322) 1.069 (0.873, 1.307) 1.060 (0.861, 1.305) 1.035 (0.839, 1.275) TCHOL 1.082 (0.855, 1.370) 1.079 (0.844, 1.378) 1.060 (0.835, 1.346) 1.050 (0.820, 1.345) 1.037 (0.809, 1.329) Triglycerides 0.754 (0.582, 0.977) 0.726 (0.555, 0.950) 0.738 (0.563, 0.967) 0.735 (0.556, 0.971) 0.748 (0.566, 0.990) HbA1C 0.773 (0.487, 1.229) 0.812 (0.503, 1.310) 0.757 (0.471, 1.217) 0.835 (0.511, 1.366) 0.900 (0.547, 1.479) CRP 0.657 (0.526, 0.822) 0.752 (0.596, 0.950) 0.658 (0.525, 0.826) 0.760 (0.599, 0.964) 0.773 (0.609, 0.981) * SBP stands for systolic blood pressure ≥ 140 mmHg, DBP for diastolic blood pressure ≥ 90 mmHg, HDL for high density lipoprotein < 1.036 mmol/L, LDL for low density lipoprotein ≥3.8 mmol/L, TCHOL for total cholesterol in blood ≥6.216 mmol/L, HbA1C for glycated haemoglobin > 6.5%, and CRP for C-reactive protein > 3 mg/L Table 4 Odds ratios for raised risk physiological CVD biomarkers* by education and household financial constraints, women (n = 2637) Women Model 1: Age + city Model 2: Model Model 3: Model Model 4: Models Model 5: Model 1 + smoking 1 + BMI + waist-hip 2 + 3 4 + medication + alcohol (AUDIT8) + fruit + stairs Education (ref. Low) SBP 0.733 (0.595, 0.903) 0.762 (0.615, 0.944) 0.790 (0.638, 0.977) 0.823 (0.661, 1.023) 0.853 (0.684, 1.064) DBP 0.678 (0.542, 0.848) 0.730 (0.580, 0.919) 0.757 (0.602, 0.951) 0.806 (0.637, 1.020) 0.826 (0.652, 1.046) HDL 0.623 (0.419, 0.925) 0.738 (0.492, 1.107) 0.749 (0.498, 1.125) 0.892 (0.586, 1.357) 0.896 (0.589, 1.363) LDL 1.035 (0.874, 1.226) 1.038 (0.873, 1.234) 1.082 (0.912, 1.284) 1.080 (0.906, 1.287) 1.071 (0.898, 1.276) TCHOL 1.213 (1.006, 1.463) 1.218 (1.005, 1.475) 1.231 (1.019, 1.487) 1.231 (1.014, 1.493) 1.221 (1.006, 1.482) Triglycerides 0.851 (0.667, 1.087) 0.867 (0.675, 1.114) 1.003 (0.778, 1.293) 1.013 (0.781, 1.313) 1.032 (0.795, 1.339) HbA1C 0.893 (0.609, 1.307) 0.900 (0.610, 1.327) 1.084 (0.730, 1.610) 1.066 (0.712, 1.597) 1.091 (0.727, 1.635) CRP 0.657 (0.550, 0.785) 0.680 (0.566, 0.816) 0.779 (0.643, 0.945) 0.798 (0.655, 0.972) 0.810 (0.665, 0.988) Financial constraints SBP 0.857 (0.679, 1.083) 0.902 (0.710, 1.145) 0.886 (0.699, 1.123) 0.932 (0.731, 1.187) 0.966 (0.756, 1.235) (ref. Low) DBP 0.885 (0.694, 1.129) 0.939 (0.732, 1.204) 0.926 (0.723, 1.186) 0.974 (0.755, 1.255) 0.997 (0.773, 1.287) HDL 0.752 (0.483, 1.171) 0.859 (0.545, 1.355) 0.821 (0.524, 1.287) 0.957 (0.601, 1.525) 0.968 (0.607, 1.544) LDL 1.034 (0.858, 1.246) 1.042 (0.862, 1.259) 1.043 (0.865, 1.258) 1.047 (0.866, 1.267) 1.038 (0.858, 1.257) TCHOL 1.054 (0.855, 1.299) 1.040 (0.841, 1.287) 1.055 (0.856, 1.301) 1.039 (0.839, 1.285) 1.028 (0.831, 1.273) Triglycerides 1.001 (0.766, 1.308) 0.999 (0.760, 1.313) 1.078 (0.818, 1.420) 1.065 (0.803, 1.411) 1.091 (0.822, 1.448) HbA1C 0.998 (0.650, 1.533) 1.023 (0.660, 1.585) 1.051 (0.677, 1.631) 1.069 (0.681, 1.678) 1.094 (0.695, 1.722) CRP 0.861 (0.708, 1.048) 0.888 (0.727, 1.085) 0.918 (0.742, 1.136) 0.946 (0.762, 1.174) 0.964 (0.776, 1.198) * SBP stands for systolic blood pressure ≥ 140 mmHg, DBP for diastolic blood pressure ≥ 90 mmHg, HDL for high density lipoprotein < 1.036 mmol/L, LDL for low density lipoprotein ≥3.8 mmol/L, TCHOL for total cholesterol in blood ≥6.216 mmol/L, HbA1C for glycated haemoglobin > 6.5%, and CRP for C-reactive protein > 3 mg/L Trias‑Llimós et al. International Journal for Equity in Health (2022) 21:51 Page 8 of 11 [31]. Nonetheless, our findings on the lack of contribu- that the biomarkers analyzed could be more driven by tion of alcohol to SEP inequalities in blood pressure is chronic inequalities (as captured by education) and less in line with recent research using US data that found by acute factors (as in part captured by current financial alcohol not to be a main predictor of SEP differences in constraints). hypertension, cardiovascular disease or diabetes [43]. In additional analyses, we also adjusted for other indi- Therefore, other factors, which we were not able to fully vidual factors unrelated to lifestyle such as visits to the adjust for, are likely to contribute to the remaining unex- doctor, blood pressure knowledge, and medication use. plained differences in blood pressure levels. For example, For blood pressure, only medication use among women psychological factors and strategies to cope with stress, substantially reduced the SEP-coefficient, and therefore which have been often postulated to differences in health contributed to explaining SEP inequalities, while for in Russia, or diet factors, which may mediate the associa- the remaining risk factors adding these adjustments did tion between socioeconomic status and risk factors [44, not substantially change our estimates. The relationship 45]. between SEP and medication use is complex given there For HbA1C, the slightly higher levels observed among may be SEP gradients both in indications for use and men from low SEP (see Table S2), particularly for educa- access to medication, and associated adherence to treat- tion, seemed to be explained by the combination of estab- ment when medication is needed. As suggested by the lished risk factors such as smoking or alcohol, and less results from this study when adding the multiple adjust- by diet, physical activity, or obesity indicators (BMI plus ments, our general conclusions would not change. We waist-hip ratio). Considering the elevated share of undi- have accounted for it by adding risk factor specific medi- agnosed diabetes in Russia [46], we could hypothesize cation as adjusted in Model 5. In doing so, the SEP coeffi- about higher SEP inequalities on both self-reported and cient remained generally similar for most of the analysed diagnosed diabetes. For cholesterol-based measures, SEP risk biomarkers. Nonetheless, we acknowledge that more inequalities were only statistically significant for HDL detailed studies on the role of medication on these asso- among women and disappeared after adjusting for BMI ciations would be worthwhile. and waist-hip ratio. This, however, does not necessarily mean that SEP inequalities in cholesterol do not exist, Evaluation of data and methods but they may require larger sample sizes to be identified. This study has several limitations. First, we need to Finally, CRP was the only risk biomarker in which clear acknowledge that the sample comes from two Russian financial constraints inequalities were found, although cities (Arkhangelsk and Novosibirsk) and are not based only in men. It could be that CRP is more strongly on nationally representative samples. The participation influenced by acute factors (e.g. psychological distress) rate in KYH was 51% overall. This implies that we are than the other biomarkers analyzed. In addition, the subject to potential selection bias that may mean that we clearly observed educational inequalities in CRP among had lower power to detect true differences in the popu- women (model 1, OR: 0.66 (0.55-0.79)) seem to be largely lation if study participation was differential by SEP. Fur- explained by obesity-related measures (model 3, OR: 0.78 thermore, we dichotomized the educational variable for (0.64-0.93)), and are in line with previous studies sug- increasing the statistical power, but we acknowledge that gesting large SEP inequalities in obesity among Russian this implies a limitation in terms of assessing a full spec- women [41]. However, in both men and women inequali- trum of educational inequalities. Our other key socio- ties existed in the odds of raised CRP levels even after economic variable, self-reported financial constraints adjusting for lifestyles. In this case, the OR is substan- was previously used in other Russian studies [47] and tially attenuated after adjusting for smoking among men accounts for a more subjective and updated dimension and for obesity-related measures among women. Thus, of socioeconomic status as compared to education. The even after adjusting for lifestyles and medication educa- KYH appears to be robust as its final sample had similar tional inequalities persist in blood pressure among men age and educational profiles as the Russian urban popula- and in CRP among women. tion [26], and the risk factor data is consistent with that Overall, it seems that most of the potential explana- previously observed, including smoking and blood pres- tions of the SEP inequalities were captured by BMI/ sure levels [48, 49]. waist-hip ratio, particularly among women. Again, the This study used cross-sectional data of measured bio- fact that other lifestyle factors seem to contribute very markers levels and lifestyles (e.g. alcohol or tobacco use). little to the observed SEP inequalities in risk biomark- It may be that some individuals have changed their life- ers does not necessarily mean that inequalities do not style and improved their risk markers as a result of advice exist, but that we were unable to find statistical evidence from medical doctors due to a CVD event which would for them using cross-sectional data. This might indicate attenuate the strength of the underlying associations. Trias‑Llimós et al. International Journal for Equity in Health (2022) 21:51 Page 9 of 11 However, our sensitivity analysis for the sample free of Availability of data and methods: ta are available upon reasonable request. Requests to access the data set of KYH study from bona fide researchers CVD are in line with our main results, both in terms of may be sent to the International Project on Cardiovascular Disease in Russia: the strength of the statistical evidence and importance of https://​metad​ata.​knowy​ourhe​art.​scien​ce/. individual lifestyle factors (see Tables S1-S2), suggesting Authors’ contributions that this bias due to reverse causality may not be major. STL and DAL designed the study. SC, AEE and VMS contributed to the study A strength of this study is that we used actual individ- designed. STL carried out the empirical analyses and drafted the manuscript. ual risk factor measurements, as opposed to indicators AVK and SM were centrally engaged in organising the fieldwork that gener‑ ated the dataset. DL and SC aided in drafting the manuscript. All authors aided of overall risk factor risk [25], only self-reported elevated in interpreting the results, critically reviewed the manuscript, and approved prevalence of risk factors, as more commonly addressed the final manuscript. [43]. One of the exceptions to this found more consist- Funding ent association between SEP and biomarkers of CVD dis- The Know Your Heart study is a component of the International Project on Car‑ ease for women than for men using Australian data [50], diovascular Disease in Russia (IPCDR). IPCDR was funded by a Wellcome Trust which is in line with our findings. Strategic Award (100217), supported by funds from the University in Tromsø The Arctic University of Norway; Norwegian Institute of Public Health; and the Norwegian Ministry of Health and Social Affairs. The funders had no role in study design, data collection and analysis, decision to publish, or preparation Conclusion of the manuscript. VMS and DAL were also partly supported by funding from In this study, we have found important educational dif- the Basic Research Program at the National Research University Higher School of Economics (Moscow, Russia). SM was supported by Russian Academy of ferences in in risk biomarkers, particularly in blood pres- Science (AAAA-A17-117112850280-2). sure and CRP. These inequalities generally diminished once adjusting for obesity-related measures for women, Declarations but not for men. Educational inequalities in physiological risk biomarkers could aid in explaining SEP inequalities Ethics approval and consent to participate Ethical approval for the study was received from the ethics committees of the in CVD health outcomes. Our findings provide evidence- London School of Hygiene & Tropical Medicine (approval number 8808, Febru‑ based information on the need for tackling health ine- ary 24, 2015), Novosibirsk State Medical University (approval number 75, May qualities in the Russian population, which may help as 21, 2015), the Institute of Preventative Medicine (no approval number; Decem‑ ber 26, 2014), and Novosibirsk and the Northern State Medical University, well to further contribute to CVD mortality decline. Pub- Arkhangelskapproval number 01/01-15, January 27, 2015). lic health policies aimed at prevention should acknowl- edge the observed gender differences in risk factor Consent for publication Not applicable. profiles and the role of lifestyle factors in explaining some of them, to reduce inequalities in risk factors and subse- Competing interests quent CVD health outcomes. The authors declare no competing interests. Author details 1 Department of Non‑communicable Disease Epidemiology, Faculty of Epi‑ Abbreviations demiology and Population Health, London School of Hygiene & Tropical AUDIT: Alcohol Use Disorders Identification Test; BMI: Body mass index; CVD: Medicine, London, UK. 2 Centre d’Estudis Demogràfics, Centres de Recerca de Cardiovascular disesase; CRP: C-reactive protein; DBP: Diastolic blood pressure; Catalunya (CERCA), Carrer de Ca n’Altayó, Edifici E2, Universitat Autònoma de HbA1C: Glycated haemoglobin; HDL: High density lipoprotein; HMD: Human Barcelona, 08193 Bellaterra, Spain. 3 National Heart and Lung Institute, Imperial Mortality Database; IPCDR: International Project on Cardiovascular Disease in College London, London, UK. 4 Department of Community Medicine, Faculty Russia; KYH: Know Your Heart Study; LDL: Low density lipoprotein; OR: Odds of Health Sciences, UiT The Arctic University of Norway, Tromsø, Norway. 5 Cen‑ ratios; SAHR: Survey on Stress, Ageing and Health in Russia; SBP: Systolic blood tral Scientific Research Laboratory, Northern State Medical University, Arkhan‑ pressur; SEP: Socioeconomic position; TCHOL: Total colesterol. gelsk, Russia. 6 Research Institute of Internal and Preventive Medicine - Branch of IC&G SB RAS, Novosibirsk, Russia. 7 Novosibirsk State Medical University, Supplementary Information Ministry of Health of Russia, Novosibirsk, Russia. 8 Laboratory for Demographic Data, Max Planck Institute for Demographic Research, Rostock, Germany. The online version contains supplementary material available at https://​doi.​ 9 International Laboratory for Population and Health, National Research Univer‑ org/​10.​1186/​s12939-​022-​01650-3. sity Higher School of Economics, Moscow, Russia. Additional file 1: Table S1A. Odds ratio of high physiological CVD risk Received: 18 August 2021 Accepted: 17 March 2022 biomarkers with education and household financial constraints in the sample free of CVD, Know Your Heart, men. Table S1B. Odds ratio of high physiological CVD risk biomarkers with education and household financial constraints in the sample free of CVD, Know Your Heart, women. Table S2A. Associations of physiological CVD risk biomarkers levels with References education and household financial constraints, Know Your Heart, men. 1. Townsend N, Kazakiewicz D, Lucy Wright F, Timmis A, Huculeci R, Torbica Table S2B. Associations of physiological CVD risk biomarkers levels with A, et al. Epidemiology of cardiovascular disease in Europe. Nat Rev Car‑ education and household financial constraints, Know Your Heart, women. diol. 2022;19:133–43. 2. Timmis A, Townsend N, Gale CP, Torbica A, Lettino M, Petersen SE, et al. European Society of Cardiology: cardiovascular disease statistics 2019. Eur Acknowledgements Heart J. 2020;41(1):12–85. We would like to acknowledge the contribution of the participants of the KYH study. Trias‑Llimós et al. International Journal for Equity in Health (2022) 21:51 Page 10 of 11 3. Ezzati M, Obermeyer Z, Tzoulaki I, Mayosi BM, Elliott P, Leon DA. Contribu‑ mortality in Eastern Europe: a multicentre prospective cohort study. PLoS tions of risk factors and medical care to cardiovascular mortality trends. Med. 2017;14(12):e1002459. Nat Rev Cardiol. 2015;12(9):508–30. 25. Glei DA, Goldman N, Shkolnikov VM, Jdanov D, Shalnova S, Shkolnikova 4. Harper S, Lynch J, Smith GD. Social determinants and the decline of M, et al. To what extent do biomarkers account for the large social dis‑ cardiovascular diseases: understanding the links. Annu Rev Public Health. parities in health in Moscow? Soc Sci Med. 2013;77:164–72. 2011;32:39–69. 26. Cook S, Malyutina S, Kudryavtsev A, et al. Know Your Heart: rationale, 5. Steptoe A, Marmot M. The role of psychobiological pathways in design and conduct of a cross-sectional study of cardiovascular structure, socio-economic inequalities in cardiovascular disease risk. Eur Heart J. function, and risk factors in 4,500 men and women aged 35–69 years 2002;23(1):13–25. from two Russian cities, 2015–18. Wellcome Open Res. 2018;3:67. https://​ 6. Méjean C, Droomers M, van der Schouw YT, Sluijs I, Czernichow S, Grob‑ doi.​org/​10.​12688/​wellc​omeop​enres.​14619.3. bee DE, et al. The contribution of diet and lifestyle to socioeconomic 27. Imahori Y, Frost C, Mathiesen EB, Ryabikov A, Kudryavtsev AV, Malyutina inequalities in cardiovascular morbidity and mortality. Int J Cardiol. S, et al. Effect of adiposity on differences in carotid plaque burden in 2013;168(6):5190–5. studies conducted in Norway and Russia: a cross-sectional analysis of two 7. Stringhini S, Carmeli C, Jokela M, Avendaño M, Muennig P, Guida F, et al. populations at very different risk of cardiovascular mortality. BMJ Open. Socioeconomic status and the 25 × 25 risk factors as determinants of 2020;10(5):e036583. premature mortality: a multicohort study and meta-analysis of 1·7 million 28. Levshin V, Slepchenko N. Determinants of smoking cessation and absti‑ men and women. Lancet. 2017;389(10075):1229–37. nence in a Russian smoking-cessation center. Tob Prev Cessat. 2017;3 8. Trias-Llimós S, Pennells L, Tverdal A, Kudryavtsev AV, Malyutina S, Hop‑ Available from: https://​www.​ncbi.​nlm.​nih.​gov/​pmc/​artic​les/​PMC72​ stock LA, et al. Quantifying the contribution of established risk factors to 32794/. [cited 2021 Dec 6]. cardiovascular mortality differences between Russia and Norway. Sci Rep. 29. Babor TF, de la Fuente JR, Saunders J, Grant M. The alcohol use disorders 2020;10(1):1–8. identification test: guidelines for use in primary care: World Health 9. Shkolnikov VM, Andreev EM, Tursun-Zade R, Leon DA. Patterns in the Organization, Department of Mental Health and Substance Dependence; relationship between life expectancy and gross domestic product 2001. in Russia in 2005–15: a cross-sectional analysis. Lancet Public Health. 30. Saunders JB, Aasland OG, Babor TF, De la Fuente JR, Grant M. Develop‑ 2019;4(4):e181–8. ment of the alcohol use disorders identification test (AUDIT): WHO 10. Kontsevaya A, Sabgaida T, Ivanova A, Leon DA, McKee M. How has the collaborative project on early detection of persons with harmful alcohol management of acute coronary syndrome changed in the Russian consumption-II. Addiction. 1993;88(6):791–804. Federation during the last 10 years? Health Policy. 2017;121(12):1274–9. 31. Cook S, De Stavola B, Saburova L, Kiryanov N, Vasiljev M, McCambridge 11. Shkolnikov VM, Leon DA, Adamets S, Andreev E, Deev A. Educational level J, et al. Socio-demographic predictors of dimensions of the AUDIT score and adult mortality in Russia: an analysis of routine data 1979 to 1994. in a population sample of working-age men in Izhevsk, Russia. Alcohol Soc Sci Med. 1998;47(3):357–69. Alcohol. 2011;46(6):702–8. 12. Shkolnikov VM. The changing relation between education and life expec‑ 32. Cust AE, Smith BJ, Chau J, van der Ploeg HP, Friedenreich CM, Armstrong tancy in central and eastern Europe in the 1990s. J Epidemiol Community BK, et al. Validity and repeatability of the EPIC physical activity question‑ Health. 2006;60(10):875–81. naire: a validation study using accelerometers as an objective measure. 13. Krieger N, Williams DR, Moss NE. Measuring social class in US public Int J Behav Nutr Phys Act. 2008;5(1):1–13. health research: concepts, methodologies, and guidelines. Annu Rev 33. Wareham NJ, Jakes RW, Rennie KL, Mitchell J, Hennings S, Day NE. Validity Public Health. 1997;18(0163–7525):341–78. and repeatability of the EPIC-Norfolk physical activity questionnaire. Int J 14. Kunst AE. Socioeconomic inequalities in health in central and Eastern Epidemiol. 2002;31(1):168–74. Europe: synthesis of results of eight new studies. Int J Public Health. 34. World Health Organization. Waist circumference and waist-hip ratio: 2009;54(4):197–200. report of a WHO expert consultation. Geneva; 2008. p. 2011. 15. Di Girolamo C, Nusselder WJ, Bopp M, Brønnum-Hansen H, Costa G, 35. Toft U, Kristoffersen LH, Lau C, Borch-Johnsen K, Jørgensen T. The dietary Kovács K, et al. Progress in reducing inequalities in cardiovascular disease quality score: validation and association with cardiovascular risk factors: mortality in Europe. Heart. 2020;106(1):40–9. the Inter99 study. Eur J Clin Nutr. 2007;61(2):270–8. 16. Mackenbach JP, Valverde JR, Artnik B, Bopp M, Brønnum-Hansen H, 36. WHO Collaborating Centre for Drug Statistics Methodology [internet]. Deboosere P, et al. Trends in health inequalities in 27 European countries. World health Organization; 2019. Available from: https://​www.​whocc.​no/. Proc Natl Acad Sci. 2018;115(25):6440–5. 37. Winkleby MA, Jatulis DE, Frank E, Fortmann SP. Socioeconomic status and 17. Murphy M, Bobak M, Nicholson A, Rose R, Marmot M. The widening gap health: how education, income, and occupation contribute to risk factors in mortality by educational level in the Russian Federation, 1980–2001. for cardiovascular disease. Am J Public Health. 1992;82(6):816–20. Am J Public Health. 2006;96(7):1293–9. 38. Galobardes B, Shaw M, Lawlor DA, Lynch JW, Smith GD. Indicators 18. Malyutina S, Bobak M, Simonova G, Gafarov V, Nikitin Y, Marmot M. Educa‑ of socioeconomic position (part 1). J Epidemiol Community Health. tion, marital status, and total and cardiovascular mortality in Novosibirsk, 2006;60(1):7–12. Russia: a prospective cohort study. Ann Epidemiol. 2004;14(4):244–9. 39. Tillmann T, Vaucher J, Okbay A, Pikhart H, Peasey A, Kubinova R, et al. 19. Todd MA, Shkolnikov VM, Goldman N. Why are well-educated Muscovites Education and coronary heart disease: mendelian randomisation study. more likely to survive? Understanding the biological pathways. Soc Sci BMJ. 2017;358:j3542. Med. 2016;157:138–47. 40. Petersen J, Kontsevaya A, McKee M, Richardson E, Cook S, Malyutina S, 20. Dennis BH, Zhukovsky GS, Shestov DB, Davis CE, Deev A, Kim H, et al. The et al. Primary care use and cardiovascular disease risk in Russian 40–69 association of education with coronary heart disease mortality in the year olds: a cross-sectional study. J Epidemiol Community Health. USSR lipid research clinics study. Int J Epidemiol. 1993;22(3):420–7. 2020;74(9):692–967. 21. World Health Organization. Global status report on alcohol and health 41. Pikhart H, Bobak M, Malyutina S, Pajak A, Kubínová R, Marmot M, et al. 2018. [Internet]. 2018. Available from: http://​www.​who.​int/​subst​ance_​ Obesity and education in three countries of the central and Eastern abuse/​publi​catio​ns/​global_​alcoh​ol_​report/​en/. [cited 2019 Jun 10]. Europe: the HAPIEE study. Cent Eur J Public Health. 2007;15(4):140–2. 22. Giovino GA, Mirza SA, Samet JM, Gupta PC, Jarvis MJ, Bhala N, et al. 42. Danilova I, Shkolnikov VM, Andreev E, Leon DA. The changing relation Tobacco use in 3 billion individuals from 16 countries: an analysis of between alcohol and life expectancy in Russia in 1965–2017. Drug Alco‑ nationally representative cross-sectional household surveys. Lancet. hol Rev. 2020;39(7):790–6. 2012;380(9842):668–79. 43. Kino S, Kawachi I. How much do preventive health behaviors explain 23. NCD Risk Factor Collaboration. Worldwide trends in blood pres‑ education-and income-related inequalities in health? Results of Oaxaca– sure from 1975 to 2015: a pooled analysis of 1479 population-based Blinder decomposition analysis. Ann Epidemiol. 2020;43:44–50. measurement studies with 19· 1 million participants. Lancet Lond Engl. 44. Boylan S, Lallukka T, Lahelma E, Pikhart H, Malyutina S, Pajak A, et al. 2017;389(10064):37. Socio-economic circumstances and food habits in eastern, central and 24. Tillmann T, Pikhart H, Peasey A, Kubinova R, Pajak A, Tamosiunas A, Western European populations. Public Health Nutr. 2011;14(4):678–87. et al. Psychosocial and socioeconomic determinants of cardiovascular Trias‑Llimós et al. International Journal for Equity in Health (2022) 21:51 Page 11 of 11 45. Cockerham WC, Hinote BP, Abbott P. Psychological distress, gender, and health lifestyles in Belarus, Kazakhstan, Russia, and Ukraine. Soc Sci Med. 2006;63(9):2381–94. 46. Iakunchykova O, Averina M, Wilsgaard T, Malyutina S, Kudryavtsev AV, Cook S, et al. What factors explain the much higher diabetes prevalence in Russia compared with Norway? Major sex differences in the contribu‑ tion of adiposity. BMJ Open Diabetes Res Care. 2021;9(1):e002021. 47. Pridemore WA, Tomkins S, Eckhardt K, Kiryanov N, Saburova L. A case– control analysis of socio-economic and marital status differentials in alcohol- and non-alcohol-related mortality among working-age Russian males. Eur J Pub Health. 2010;20(5):569–75. 48. Balanova YA, Shalnova SA, Imaeva AE, Kapustina АV, Muromtseva GA, Evstifeeva SV, et al. Prevalence, Awareness, Treatment and Control of Hypertension in Russian Federation (Data of Observational ESSERF-2 Study). Rational Pharmacother Cardiol. 2019;15:450-466–466 Available from: https://​www.​rpcar​dio.​com/​jour/​artic​le/​view/​1996. [cited 2020 Mar 16]. 49. Shkolnikov VM, Churilova E, Jdanov DA, Shalnova SA, Nilssen O, Kudryavt‑ sev A, et al. Time trends in smoking in Russia in the light of recent tobacco control measures: synthesis of evidence from multiple sources. BMC Public Health. 2020;20(1):1–11. 50. Kavanagh A, Bentley RJ, Turrell G, Shaw J, Dunstan D, Subramanian SV. Socioeconomic position, gender, health behaviours and biomarkers of cardiovascular disease and diabetes. Soc Sci Med. 2010;71(6):1150–60. Publisher’s Note Springer Nature remains neutral with regard to jurisdictional claims in pub‑ lished maps and institutional affiliations. Ready to submit your research ? Choose BMC and benefit from: • fast, convenient online submission • thorough peer review by experienced researchers in your field • rapid publication on acceptance • support for research data, including large and complex data types • gold Open Access which fosters wider collaboration and increased citations • maximum visibility for your research: over 100M website views per year At BMC, research is always in progress. Learn more biomedcentral.com/submissions