Changing social inequalities in smoking, obesity and cause-specific mortality: Cross-national comparisons using compass typology

Andrea Teng; Tony Blakely; June Atkinson; Ramunė Kalėdienė; Mall Leinsalu; Pekka T Martikainen; Jitka Rychtaříková; Johan P Mackenbach

doi:10.1371/journal.pone.0232971

. 2020 Jul 10;15(7):e0232971. doi: 10.1371/journal.pone.0232971

Changing social inequalities in smoking, obesity and cause-specific mortality: Cross-national comparisons using compass typology

Andrea Teng ^1,^*, Tony Blakely ^2,^3,^*, June Atkinson ¹, Ramunė Kalėdienė ⁴, Mall Leinsalu ^5,⁶, Pekka T Martikainen ⁷, Jitka Rychtaříková ⁸, Johan P Mackenbach ⁹

Editor: Brecht Devleesschauwer¹⁰

PMCID: PMC7351173 PMID: 32649731

Abstract

Background

In many countries smoking rates have declined and obesity rates have increased, and social inequalities in each have varied over time. At the same time, mortality has declined in most high-income countries, but gaps by educational qualification persist—at least partially due to differential smoking and obesity distributions. This study uses a compass typology to simultaneously examine the magnitude and trends in educational inequalities across multiple countries in: a) smoking and obesity; b) smoking-related mortality and c) cause-specific mortality.

Methods

Smoking prevalence, obesity prevalence and cause-specific mortality rates (35–79 year olds by sex) in nine European countries and New Zealand were sourced from between 1980 and 2010. We calculated relative and absolute inequalities in prevalence and mortality (relative and slope indices of inequality, respectively RII, SII) by highest educational qualification. Countries were then plotted on a compass typology which simultaneously examines trends in the population average rates or odds on the x-axis, RII on the Y-axis, and contour lines depicting SII.

Findings

Smoking and obesity. Smoking prevalence in men decreased over time but relative inequalities increased. For women there were fewer declines in smoking prevalence and relative inequalities tended to increase. Obesity prevalence in men and women increased over time with a mixed picture of increasing absolute and sometimes relative inequalities. Absolute inequalities in obesity increased for men and women in Czech Republic, France, New Zealand, Norway, for women in Austria and Lithuania, and for men in Finland.

Cause-specific mortality. Average rates of smoking-related mortality were generally stable or increasing for women, accompanied by increasing relative inequalities. For men, average rates were stable or decreasing, but relative inequalities increased over time. Cardiovascular disease, cancer, and external injury rates generally decreased over time, and relative inequalities increased. In Eastern European countries mortality started declining later compared to other countries, however it remained at higher levels; and absolute inequalities in mortality increased whereas they were more stable elsewhere.

Conclusions

Tobacco control remains vital for addressing social inequalities in health by education, and focus on the least educated is required to address increasing relative inequalities. Increasing obesity in all countries and increasing absolute obesity inequalities in several countries is concerning for future potential health impacts. Obesity prevention may be increasingly important for addressing health inequalities in some settings. The compass typology was useful to compare trends in inequalities because it simultaneously tracks changes in rates/odds, and absolute and relative inequality measures.

Introduction

In many countries the prevalence of smoking is declining and obesity is increasing, but the social patterning of these risk factors has varied over time. There are important implications for inequalities in premature mortality, given the strong link of both smoking and obesity to premature mortality.

Smoking prevalence has been declining over time in many countries but still remains a leading cause of death. These declines have tended to be greatest in men and women who have a tertiary education [1–3], and future widening of social inequalities in smoking-related disease has been predicted [1]. In many settings, tobacco smoking is the major contributor to social inequalities in premature mortality [4, 5]. For example the Preston-Glei-Wilmoth method relies on lung cancer mortality as an indicator of the accumulated population exposure to smoking and was used to estimate the contribution of smoking to absolute educational inequalities in all-cause mortality in 14 European countries from 1990–2004 [5]. Between countries the contribution ranged from 19% to 55% in men, and −1% to 56% in women [5]. In Norway the contribution of smoking declined over time in men but increased in women [6]. In New Zealand smoking has been estimated to contribute to 21% and 11% of relative social inequalities in mortality in men and women respectively by education [7] with changes over time [8]. Tobacco control is considered one of the key entry points for addressing educational inequalities in mortality.

Recent decades have also seen ongoing increases in obesity prevalence in many countries. Obesity is significantly associated with all-cause mortality [9] for example by increased rates of diabetes, cardiovascular disease and cancer. Where there is strong social patterning of obesity it is likely to make an increasing contribution to health inequalities [10]. A European study used population attributable fractions and rate ratios from the literature to estimate how a scenario elimination of differences in obesity by education would decrease relative inequality in all-cause mortality. This resulted in average declines of 6% in men and 16% in women in relative inequalities [11]. There was variation between countries from 0% to 12% in men and 4% to 42% in women (eg lower in some Eastern European countries and higher in some Southern European countries) [11]. These figures were greater for causes of death such as diabetes and ischaemic heart disease. In another study, obesity was a less important contributor to income inequalities in health status in the UK because obesity was more equally distributed by income [12].

The world has witnessed dramatic declines in mortality in recent decades (albeit with slowing in some groups recently), however relative inequalities in premature mortality by socioeconomic position have continued to increase in many European countries [13], New Zealand [14] and elsewhere. There is variation between countries in the magnitude of these socioeconomic inequalities. Relative inequalities in mortality by education have continued to rise in most European countries but not in Southern Europe, where inequalities in mortality are smaller [13]. Eastern Europe countries have experienced larger increases in relative inequalities over time than other European countries [13]. At least part of the differences in these effects of education on mortality is likely to be due to differences in the prevalence rates of smoking and obesity [15].

This study’s authors have previously published comparisons of education inequalities in cause-specific premature mortality in Europe (1990–2010) [13, 16], and New Zealand (1981–2011) [14]. Given the potential confusion and challenge of understanding trends over time in average rates and both absolute and relative inequalities, the later study developed a compass typology. Specifically, graphs of average rates on the x axis and relative inequalities on the y axis are used. For any given point on this x-y plane, absolute inequalities are given by contour lines, according to the mathematical interdependence of these quantities. Country trends are plotted over time, applying the compass typology. The most desirable direction of travel is southwest, because it represents decreasing average rates and decreasing relative and absolute inequalities. The most undesirable direction is northeast, with all three quantities increasing over time. These graphs efficiently convey a lot of information without requiring several separate graphs of mortality rates and inequality trends. Empirically for all-cause mortality, the common pattern across countries is of decreasing mortality rates over time, with increasing relative inequalities but stable absolute inequalities: a northwest direction of travel [14]. In this current paper, we apply this compass typology for the first time to risk factors (smoking and obesity) and country-level comparisons of cause specific mortality rates.

Cross-country comparisons of trends provide feedback on comparative progress, insight into the potential malleability of these trends over time and they may stimulate further hypotheses about national level political, economic and social drivers of health and health inequalities. The compass typology can also predict the expected future direction of travel for both mortality and risk factors, given current settings and compass trajectories. For example inequality trends in smoking and obesity can help predict the likely future trends in smoking- and obesity- related mortality.

Aims and objectives

This study aimed to describe the simultaneous overall trends and trends in educational inequalities for smoking prevalence, obesity prevalence and cause-specific mortality, comparing adults in nine European countries and New Zealand using compass typology.

Methods

National trends in smoking and obesity prevalence over two decades in 1990–2010, and cause-specific mortality from 1980–2010 (with adjacent data from late 1970s or early 2010s if available), were used. Ten countries were included in this analysis based on the availability of smoking prevalence and mortality data by education. Countries were selected if, as a minimum, they had a smoking data point in both the 1990s and 2000s; and they had a mortality data point available for both the 1980s and 2000s (eg, mid date of a cohort study), ie, in the first and last decade of the study period 1980 to 2010 (Table 1). One country was missing obesity data (England) and one country had only 2000s data for obesity (New Zealand). Of the 10 countries selected, 4 were from Eastern Europe, 3 were from Western Europe, 2 were from Northern Europe, and 1 was from Australasia (New Zealand).

Table 1. Country data for socioeconomic inequalities in mortality and smoking/obesity prevalence by year.

Region	Country	Data	Years covered by the analysis
			1980s		1990s		2000s		2010s
Australasia	New Zealand	Mortality	1981–84	1986–89	1991–94	1996–99	2001–06	2006–11
		Smoking/obesity				1996-97^a	2002–03	2006–07	2011–12
Northern Europe	Finland	Mortality	1981–85	1986–90	1991–95	1996–00	2001–05	2006–10
Northern Europe		Smoking/obesity			1993/95	1997/99	2001/03	2005/07	2009/11
	Norway	Mortality	1980–85	1985–90	1990–95	1995–01	2001–06	2006–09
		Smoking/obesity				1998	2002	2005, 2008
Western Europe	Austria	Mortality	1981–82		1991–92		2001–02
		Smoking/obesity	1983^b		1991	1999		2006
	France	Mortality	1980–81, 1982–86	1987–89	1990–94	1995–98	1999–03	2004–07
		Smoking/obesity	1980^b		1991		2000	2005	2010
	England and Wales	Mortality	1976–81				2001–06	2006–09
	England	Smoking/obesity	1980 ^ab	1986 ^ab	1990 ^a	1996 ^a	2000 ^a	2005 ^a	2010 ^a
Eastern Europe	Czech Republic	Mortality	1982–85				1998–03
		Smoking/obesity			1993/99		2002/08
	Estonia	Mortality		1987–91			1998–02
		Smoking/obesity				1996		2006
	Hungary	Mortality	1978–81	1988–91			1999–02
		Smoking/obesity			1994		2000, 2003	2009
	Lithuania	Mortality		1988–90			2001–05	2006–09
		Smoking/obesity				1994/2000		2006/10

Open in a new tab

1. ^aSmoking data was available but not obesity data. ^bThis data was available but not presented in the typology plots because it did not align with time periods of data from other centres. ‘/’ indicates a combination of two different datasets from different years, and in this case the midpoint of the years was used to allocate the dataset to the right time period.

Datasets

Smoking and obesity prevalence data for men and women were collected from nationally representative surveys that also had measured highest educational qualification. Smoking was measured as current tobacco smoking except for Austria and New Zealand where it was daily smoking. This was done for 30–79 year olds, except for in some countries where upper age limits were lower (France 74 years old, and England 69 years old) [17]. Obesity was measured as a BMI of ≥ 30kg/m² calculated from self-reported height and weight except for New Zealand where these values were measured. This was done for survey participants 30–79 years old, with the exception of Hungary and Lithuania where the upper age limit was 64 years old. Standard survey weights were not available for all surveys. Therefore results from European countries present unweighted analyses and these are expected to be similar to survey weighted results [17, 18]. Survey-weighted prevalences were used for New Zealand data. Some survey results were combined to improve study precision, for example each pair of successive surveys in Finland was pooled to calculate the mean smoking and obesity prevalence.

Education was selected as a proxy for socioeconomic position given its widespread availability; ease of harmonization across countries and lower sensitivity to health-related social mobility than other socioeconomic measures. The International Standard Classification of Education, 1997 was used to define the levels of education in all countries. Categories were 0–2 (pre-primary, primary, and lower secondary education) as “low,” 3–4 (upper secondary and post-secondary non-tertiary education) as “middle,” and categories 5–6 (first and second stage of tertiary education) as “high”. The proportion of the population with unknown education level varied between 0% and 11% across countries, and this group could not be included in this analysis.

Cardiovascular disease (CVD), cancer, external causes of death, other causes of death, and an overlapping group of tobacco-related mortality (lung cancer, chronic obstructive pulmonary disease [COPD] and laryngeal cancer, see S3 Table for ICD codes) were examined in men and women from nine European countries[19] and New Zealand (New Zealand Census-Mortality Study). Obesity related mortality was not examined because of the likely inconsistencies in diabetes as a coded cause of death over time and between countries. All mortality data covered complete national populations with the exceptions of England and Wales (1% representative sample) and France (1% sample of individuals born in and living in mainland France). The target age range for mortality was 35–79 years, with some slight differences in New Zealand (35–74 y), Norway (40–79 y), and Lithuania (35–69 y). Mortality studies were longitudinal in design with a population census linked to mortality during a follow-up period; with the exceptions of Czech Republic, Estonia, Hungary and Lithuania (in 1988–90 and 2000–02) which used a cross-sectional design where educational level of deceased individuals was taken from death certificates and person-years by education from the population census. For Lithuania, longitudinal data were used for 2001–2005, and 1988–1990 cross-sectional data were adjusted downwards to adjust for an overestimation of inequalities in mortality; details described elsewhere [19]. Laryngeal cancer was not included in tobacco-related mortality for New Zealand but is unlikely to have much impact given these cancers were equivalent to only 1.5% of the number of lung cancer deaths in 2014. For more information on the source mortality [13, 14, 16] and survey risk factor [17, 18] datasets please see previous publications and S5 Tables.

Analysis

Mortality rates and risk factor prevalences were age-standardised to the WHO European standard population (1976). Prevalences were converted to odds so they were on the same scale as the logistic regression used for calculating inequality measures. Average odds were calculated from the age-standardised prevalence of smoking (or obesity) divided by 1 –prevalence of smoking. Inequality estimates were based on national smoking and obesity prevalence odds (prevalence / [1—prevalence]) and mortality rates (per 100,000) by level of education.

Inequalities in mortality and prevalence were assessed with the relative (RII) and absolute (SII) indices of inequality. The relative index of inequality (RII) can be interpreted as the rate ratio of mortality/ratio of prevalence odds between those with the very lowest (rank = 1) and those with the very highest educational position (rank = 0) in the population. The slope index of inequality (SII) represents the absolute version of the RII; ie the difference in mortality rates/prevalence odds between those with the very lowest and the very highest educational position in the population. For smoking and obesity, RIIs were calculated on the odds scale using logistic regression analysis and exponentiation of the beta coefficient and its confidence intervals from the analysis (Eq 1). SIIs were calculated from the regression-predicted odds where x = 0 and x = 1.

L o g_{(o d d s p r e v a l e n c e)} = ß_{(c u m u l a t i v e s m o k i n g m i d p o i n t s b y e d u c a t i o n)} + a g e_{(5 y r g r o u p s)}

(1)

Compass plots have previously been used only for rates [14]. In this paper, the compass plots were extended to work with risk factors which have a prevalence bound between 0 and 1. To avoid modelling risks less than 0 or more than 1, we used logistic regression. To calculate the SII contours for the typology plots we derived the mathematical relationship between odds of smoking (or obesity), the SII of odds of average smoking in population and the RII (on the odds ratio scale) (Eq 2, see the Supplementary Material for derivation).

o d d s = \frac{S I I}{e x p (0.5 \times l n [R I I]) - \frac{1}{e x p (0.5 \times l n [R I I])}}

(2)

Mortality RIIs were calculated using Poisson regression with the observed number of deaths as the dependent variable, person-years of population as the offset variable, and educational rank and dummy variables for age (in 5 year age groups) as independent variables [20]. Educational rank was calculated as the cumulative midpoints of each education category by country, time period, sex and age group. Mortality RIIs from New Zealand were calculated slightly differently using weighted (by person time) linear regression with the standardised-mortality rate as the dependent variable, and educational rank as the independent variable, calculated separately by age group, but pooled across sex and ethnic groups [21–26]. The mortality SII was calculated as follows using formula (Eq 3), where ASMR is the age-standardised mortality rate [20].

S I I = \frac{2 \times A S M R \times (R I I - 1)}{(R I I + 1)}

(3)

Compass typology plots

For the risk factor compass plots, the log of the odds of smoking or obesity was presented on the x-axis, with the log of the RII (odds ratio scale) on the y-axis, and the SII as contours. Mortality compass plots were presented with the log of the age-standardised mortality rate on the x-axis, the log of the RII (relative scale) on the y-axis, and the SII as contours. The mathematical relationship between mortality rates, RII and SII has been reported elsewhere [14].

If compass typology trends continued on the same trajectory into the future then typology results could be used to predict future inequalities. Furthermore, given the strong relationship between smoking and smoking-related mortality, patterns in smoking inequalities may predict future smoking-related mortality inequalities for example in three decades time.

Results

Smoking

Fig 1 compares smoking and obesity trends in the 1990s and 2000s across ten countries by sex. The average smoking odds in men declined in most countries (less so in Czech Republic), accompanied by increased relative inequalities (RII, northwest direction of travel, except Estonia). At the end of the study period the highest smoking odds in men were in Lithuania, Estonia and Hungary and the lowest smoking odds were in New Zealand, Finland and Norway. The trend for women was generally more northward with increasing relative inequalities (except for Czech Republic), a mixed picture for absolute inequalities (which declined for England and Norway and increased in Austria France, Estonia, Hungary, and Lithuania), but generally few decreases (England, Finland, New Zealand, Norway) in smoking odds and therefore prevalence. This is possible where smoking rates in women were falling in high education groups and increasing in low educational groups. Smoking odds (and therefore prevalence) were consistently higher in men than in women. At the end of the study period the highest smoking odds in women were in Hungary, France then Norway and the lowest smoking odds were in Lithuania, Finland and New Zealand.

Obesity

Trends for obesity prevalence differed from smoking. In every country obesity odds increased in men and women (east) with a mixed picture of increasing absolute and sometimes relative inequalities. Absolute inequalities in obesity increased for men and women in Czech Republic, France, New Zealand, Norway, and for men in Finland and for women in Austria and Lithuania. Absolute inequalities decreased for men in Lithuania and Estonia where the more educated had greater odds of obesity. There were increases in relative inequalities in men in Norway, and women in France, Lithuania and New Zealand. Obesity prevalence appeared to be similar between men and women.By the end of the study period, the highest odds of obesity were in New Zealand and Lithuania and the lowest odds of obesity were in France, Norway and Austria.

Smoking-related mortality

Trends in smoking related mortality (eg lung cancer and COPD) were generally north/northwest in men (declining or stable [Estonia, Hungary and Norway] mortality rates and increasing relative inequalities) and north/northeast for women (stable or increasing mortality rates and increasing relative inequalities) (Fig 2). Eastern Europe countries tended to have lower relative inequalities for smoking mortality in women than other regions, crossing over from smoking mortality being greater in more educated women in the 1990s to being greater in the least educated women in the 2000s. The lowest smoking related mortality rates in men were in Norway and the highest were in Hungary. The lowest smoking related mortality rates in women were in Lithuania and the highest were in New Zealand which also experienced the highest absolute and relative inequalities. Women had lower smoking-related mortality rates than men, who also had much greater absolute and relative inequalities.

Fig 2 — Relative inequalities are measured by relative index of inequality (RII) on the log scale, and absolute inequalities by the slope index of inequality (SII) indicated by the contour lines.

Cause-specific mortality

Fig 3 compares the CVD, cancer, external injuries and other mortality rates and inequality trends from the 1980s to the 2000s. CVD mortality rates declined and relative inequalities increased substantially (northwest) for Western and Northern European countries and New Zealand. There was a decrease in absolute inequalities in men and women from England and Wales, Finland, and France and in women from Austria and New Zealand. Conversely countries from Eastern Europe (Hungary, Czech Republic, Estonia, Lithuania) generally followed a more northward pattern with limited declines in CVD mortality and increased absolute and relative inequalities. Mortality rates varied significantly over time and between countries ranging from 200 to 1200 per 100,000 in men and 80 and 660 in women. Mortality rates were 1.2 to 6 times greater (RII) in the least educated, with absolute inequalities (SII) of between 80 to 1280 deaths per 100,000 population. By the end of the time period, men and women in France had the lowest CVD mortality rates and absolute inequalities.

Cancer mortality trends and regional patterns were similar to CVD with widespread declines in cancer mortality. However, men and women in Hungary and Czech Republic; men in Estonia; and women in New Zealand and Norway had generally high and stable cancer mortality and increasing absolute and relative inequalities (northward trend). There was a decrease in absolute inequalities in cancer for men and women in England and Wales, and for men in France.

External injury mortality rates were lowest in England and Wales and highest in Eastern European men and women. There was a general northward trend with stable mortality and increasing relative inequalities. However there were some declines in external injury mortality in men and women in England and Wales and Austria, men in Finland and women in France, Czech Republic and Hungary. External injury rates increased in Estonia and Lithuania particularly in men. All mortality rates were greater in men than in women.

Discussion

Smoking prevalence was generally (northwest) decreasing in odds with increasing relative inequalities for men, but only a few countries had decreased odds in women (England, Finland, New Zealand and Norway). Trends in smoking-related mortality were consistent with a delay in the smoking epidemic in women and a lag compared to smoking prevalence trends. Namely among men there was a more stable smoking-related mortality (northerly) pattern compared to decreasing (northwest) smoking prevalence, and among women (northeast) there was increasing mortality compared to a generally stable (northerly) smoking prevalence. An approximately 30 year or more [27, 28] time lag from smoking to mortality (largely lung cancer and COPD deaths) has been described. Different countries appeared to be at different stages of progression the smoking epidemic, with almost all heading towards lower smoking prevalence but high relative inequalities. If existing trends continue then smoking-related mortality in men and then women will shift northwest in the decades ahead, with decreased smoking-related mortality rates and increased relative inequalities.

Obesity prevalence increased over time similarly in men and women. Indeed, no country recorded a decline in obesity prevalence from the 1990s to 2000s. This was combined with increasing absolute inequalities in many countries (eastward trend) with two countries reporting a decrease in absolute and relative inequalities where obesity became more prevalent in men with higher education (Lithuania and Estonia) but this pattern was not noted for woman. This finding has been seen previously in Estonia [29] and the authors’ hypothesise physical activity from manual occupations as a contributing factor. It also may be related to being on an earlier stage in the nutrition transition [30], where energy dense food became more accessible to the most educated first after the dissolution of the Soviet Union, then later became more concentrated in the least educated. Men in Norway and women in France, Lithuania and New Zealand had a pattern of worsening absolute and relative inequalities. These trends are an increasing concern for countries and a warning about the future obesity-related disease burden and the likely increasing absolute inequalities in obesity-related mortality (such as diabetes, obesity-related cancers and cardiovascular disease). There is a concern that increasing obesity will hinder future progress in reducing obesity-related mortality and impact on inequalities in obesity-related mortality. Authors have suggested that the lag time between obesity prevalence and mortality may be decades [31–33], however there is also evidence that sudden changes in nutrition interventions can reduce CVD events within months [34]. If the greatest mortality impacts are after a long period of obesity exposure, then the trends here in obesity prevalence and absolute inequalities are not yet evident in the reported mortality rates.

Clear northwest trends emerged from the cause-specific mortality analysis, with decreasing rates of CVD, cancer, external injury, other mortality rates, while at the same time there were almost ubiquitous increased relative inequalities as described previously [13, 14, 16] but that are likely to worsen if current trends are maintained. Cause specific mortality rates in men and women from Eastern Europe countries (Hungary, Estonia, Lithuania and Czech Republic) tended more northward recording much small declines in average mortality if any and larger increases in relative inequalities than other countries.

Mortality trends reflect the social drivers of health and progress taken to reduce mortality, and how almost universally the most educated have benefited more proportionally from these improvements, thus increasing the relative gap. This is possible even when absolute mortality gains were the same for both groups, which was the case in countries that tracked parallel to SII lines on the compass plot.

Despite the large differences in starting and finishing positions, the similarities in trajectories identified in the compass plots between countries were striking. The generally north/northwest smoking prevalence trends, the eastward obesity prevalence and the northwest mortality patterns suggest global drivers are important such as the globalisation of the obesogenic environment. The trends presented here may be similar to those found in other developed countries.

Strengths and limitations

For the first time the compass typology is applied to country-level comparisons of smoking and obesity prevalence, cause-specific mortality, and their corresponding relative and absolute inequalities. The compass approach goes a step further than conventional ways of analyzing trends in health inequalities. For example, this paper shows the differences between countries in the relative positioning of obesity, obesity inequalities and trends, and the possible future direction of those trends, which are not clearly demonstrated using conventional methods [18].

There are however some limitations. We did not have the required data to include any countries from Southern Europe or any other regions, where trends may differ. Similarly we did not have data for the precise corresponding timespans for included studies. This was largely managed by ensuring all countries had data available in the first and the last decade of each timeframe. Also for very complex and large international comparisons, there was no possibility to collect more recent data.

Differences between data sources may have impacted some results. For example cross sectional data is used from Eastern European countries, which may have introduced bias eg, if education was used from the death certificate in the numerator and from the population census in the denominator there is the risk that numbers are from somewhat different populations. This may have for example overestimated education inequalities for some Eastern European countries as found in Lithuania [35]. There were differences the degree of missing education data, which may lead to different levels of underestimating inequalities. The older age range of obesity data is likely to have overestimated New Zealand’s obesity rates compared to other countries. Small differences in disease definitions and age span were expected to have small effects.

Random error appeared to have contributed to instability in the inequality trends (See supplementary results files for confidence intervals). To limit instability, for some countries risk factor results were averaged across adjacent time periods. Confidence intervals in the Supplementary Files show that in some countries there was reasonably wide uncertainty in the RIIs and SIIs at each time point. However we were interested in the statistical significance of inequality trends. These can be tested with trend statistics, which work for linear (in the chosen outcome scale (e.g. log)) trends, but not otherwise without extension. The compass plots, however, provide a way to see inequality trends. Future developments could include methods to convey statistical certainty (e.g. 90% uncertainty zones around the arrows, or perhaps even a 95% confidence interval about the degrees of travel).

With the exception of smoking-related mortality, it was impossible to directly compare cause-specific mortality trends with those of the related factors given the variety of underlying risk factors contributing to mortality type. This paper therefore focusses on trends and future predictions.

Education is not the only cause of these patterns of social inequality. Children from different socioeconomic positions have very different chances of obtaining a tertiary education, and this might be related to parental education, financial opportunities, racism and other factors. Thus education is likely also a proxy for these factors as well.

Although the RII and SII are superior to more simple measures of inequalities, the interpretation can be complex, as illustrated by Renard, Devleesschauwer [36] who show that under certain conditions improvements in the education distribution can increase the SII/RII.

Implications practice policy and research

The compass typology provides a useful visual tool for monitoring simultaneous trends in cause specific mortality rates and inequalities, to visualise progress on social health inequalities. Countries who want to address social inequalities can examine their compass typology to examine their progress relative to other countries, the rate of change of these trends over time, and possible future trends—and perhaps even set targets using the compass typology. Risk factor compass typologies can help predict likely changes in future smoking- and obesity- related mortality and social inequalities.

Breaking the link between education and health or mortality outcomes is a policy imperative. Tobacco control should be meeting the needs of the least educated. A greater focus on preventing obesity in the least educated is also important.

Conclusions

Increasing relative inequalities in smoking prevalence also are suggestive of further increases in relative smoking-related mortality inequalities. Obesity in men and women increased in all countries with an increase in absolute inequalities as well in several countries. There is a concern that these trends will not only hinder future progress in reducing obesity-related mortality (such as CVD and cancer mortality) but also will impact inequalities in obesity-related mortality, although this was not measured here. The compass typology is a useful tool in monitoring country comparisons of trends and inequalities in risk factor prevalence and cause-specific mortality.

Supporting information

S1 File. Further methods.

(PDF)

Click here for additional data file.^{(307.8KB, pdf)}

S1 Spreadsheet. Risk factor prevalences (subsequently converted into odds), SIIs and RIIs and confidence intervals.

(XLSX)

Click here for additional data file.^{(30.7KB, xlsx)}

S2 Spreadsheet. Mortality rates, SIIs and RIIs and confidence intervals.

(XLSX)

Click here for additional data file.^{(76.5KB, xlsx)}

S1 Table. Countries excluded because mortality data was not available.

(DOCX)

Click here for additional data file.^{(15.2KB, docx)}

S2 Table. Countries excluded because mortality inequality data was not available from both the 1980s and 2000s decades.

(DOCX)

Click here for additional data file.^{(17.2KB, docx)}

S3 Table. ICD coding for the cause-specific mortality groups.

*162–163, 165 (ICD-9) and C33–C34, C39 (ICD-10) in most Europe studies in the 1990s, ~not included for NZ smoking related mortality, ICD coding based on that in: Mackenbach JP, Kulhanova I, Menvielle G, et al. Trends in inequalities in premature mortality: a study of 3.2 million deaths in 13 European countries. J Epidemiol Community Health. 2014;69(3):207–217.

(DOCX)

Click here for additional data file.^{(15.3KB, docx)}

S4 Table. Compass trajectories for general mortality cause specific inequality trends between 1980 and 2010, and smoking and obesity inequalities between 1990 and 2010.

North is increasing relative inequalities, south is decreasing relative inequalities, west is decreasing prevalence/mortality, and east is increasing prevalence/mortality. South West corresponds with decreasing absolute and relative inequalities and decreasing rates/prevalence. It is the most preferable trend. All countries were required to have data from the first and last decade of the prescribed time periods. The trajectory is taken to be the average linear trend over the whole time period. Where the trajectory was borderline, both possible directions were recorded in the table. Note the distance of the trajectory is not presented in this table and some changes were large and some were very small.

(DOCX)

Click here for additional data file.^{(16.8KB, docx)}

S5 Table. Characteristics of mortality, smoking and obesity data.

(DOCX)

Click here for additional data file.^{(39.6KB, docx)}

S6 Table. Source mortality and risk factor datasets, where available.

(DOCX)

Click here for additional data file.^{(20.1KB, docx)}

S1 Fig. Trends in age-standardised smoking prevalence, all education groups combined, 30–79 year olds 1980–2010.

(PDF)

Click here for additional data file.^{(241.8KB, pdf)}

S2 Fig. Trends in age-standardised obesity prevalence, 30–79 year olds 1980–2010.

(PDF)

Click here for additional data file.^{(223.2KB, pdf)}

S3 Fig. Age-standardized mortality rates of CVD, cancer, external mortality, smoking-related mortality, in men (left) and women (right) 35–79 years.

(PDF)

Click here for additional data file.^{(275.4KB, pdf)}

S1 Data

(XLSX)

Click here for additional data file.^{(76.5KB, xlsx)}

S2 Data

(XLSX)

Click here for additional data file.^{(30.7KB, xlsx)}

Acknowledgments

Thank you Associated Professor James Stanley for statistical advice calculating weighted age-standardised prevalence from New Zealand health survey data. Thank you co-authors and Katalin Kovács, Gwenn Menvielle, Bjorn Heine Strand, Chris White, Johannes Klotz; who supplied the European data.

Data Availability

Study findings data are available in S2 and S3 supplementary files. Underlying datasets may be available on request depending on individual country requirements. Data cannot be shared publicly because of individual country restrictions. Data are available from each country for researchers who meet the criteria for access to confidential data. Contact information is in Table S5d. The authors had no special access privileges to the data others would not have.

Funding Statement

AT was funded for this study by the University of Otago, Wellington (Dean’s Grant). This study was supported by a grant (FP7-CP-FP grant no. 278511) from the European Commission Research and Innovation Directorate General, as part of the “Developing methodologies to reduce inequalities in the determinants of health” (DEMETRIQ) project. Funders played no role in study design, data collection and analysis, decision to publish or preparation of the manuscript.

References

1.Giskes K, Kunst AE, Benach J, et al. Trends in smoking behaviour between 1985 and 2000 in nine European countries by education. J Epidemiol Community Health. 2005;59(5):395–401. Epub 2005/04/16. 10.1136/jech.2004.025684 . [DOI] [PMC free article] [PubMed] [Google Scholar]
2.Huisman M, Kunst AE, Mackenbach JP. Educational inequalities in smoking among men and women aged 16 years and older in 11 European countries. Tob Control. 2005;14(2):106–13. Epub 2005/03/26. 10.1136/tc.2004.008573 . [DOI] [PMC free article] [PubMed] [Google Scholar]
3.Huisman M, Kunst AE, Mackenbach JP. Inequalities in the prevalence of smoking in the European Union: comparing education and income. Prev Med. 2005;40(6):756–64. Epub 2005/04/27. 10.1016/j.ypmed.2004.09.022 . [DOI] [PubMed] [Google Scholar]
4.Jha P, Peto R, Zatonski W, et al. Social inequalities in male mortality, and in male mortality from smoking: indirect estimation from national death rates in England and Wales, Poland, and North America. The Lancet. 2006;368(9533):367–70. 10.1016/s0140-6736(06)68975-7 [DOI] [PubMed] [Google Scholar]
5.Gregoraci G, Lenthe FJv, Artnik B, et al. Contribution of smoking to socioeconomic inequalities in mortality: a study of 14 European countries, 1990–2004. Tobacco Control. 2017;26:206–68. 10.1136/tobaccocontrol2015-052766 [DOI] [PubMed] [Google Scholar]
6.Martikainen P, Ho JY, Preston S, et al. The changing contribution of smoking to educational differences in life expectancy: indirect estimates for Finnish men and women from 1971 to 2010. J Epidemiol Community Health. 2013;67(3):219–24. 10.1136/jech-2012-201266 . [DOI] [PMC free article] [PubMed] [Google Scholar]
7.Blakely T, Wilson N. The contribution of smoking to inequalities in mortality by education varies over time and by sex: two national cohort studies, 1981–84 and 1996–99. Int J Epidemiol. 2005;34(5):1054–62. 10.1093/ije/dyi172 . [DOI] [PubMed] [Google Scholar]
8.Teng A, Atkinson J, Disney G, et al. Changing smoking-mortality association over time and across social groups: National census-mortality cohort studies from 1981 to 2011. Sci Rep. 2017;7(1):11465 10.1038/s41598-017-11785-x . [DOI] [PMC free article] [PubMed] [Google Scholar]
9.Aune D, Sen A, Prasad M, et al. BMI and all cause mortality: systematic review and non-linear dose-response meta-analysis of 230 cohort studies with 3.74 million deaths among 30.3 million participants. BMJ. 2016;353:i2156 10.1136/bmj.i2156 . [DOI] [PMC free article] [PubMed] [Google Scholar]
10.Bono F, Matranga D. Socioeconomic inequality in non-communicable diseases in Europe between 2004 and 2015: evidence from the SHARE survey. Eur J Public Health. 2018. Epub 2018/09/01. 10.1093/eurpub/cky165 . [DOI] [PMC free article] [PubMed] [Google Scholar]
11.Hoffmann R, Eikemo TA, Kulhanova I, et al. Obesity and the potential reduction of social inequalities in mortality: evidence from 21 European populations. Eur J Public Health. 2015;25(5):849–56. 10.1093/eurpub/ckv090 . [DOI] [PubMed] [Google Scholar]
12.Vallejo-Torres L, Morris S. The contribution of smoking and obesity to income-related inequalities in health in England. Soc Sci Med. 2010;71(6):1189–98. Epub 2010/07/16. 10.1016/j.socscimed.2010.06.008 . [DOI] [PubMed] [Google Scholar]
13.Mackenbach JP, Kulhanova I, Menvielle G, et al. Trends in inequalities in premature mortality: a study of 3.2 million deaths in 13 European countries. J Epidemiol Community Health. 2014;69(3):207–17. 10.1136/jech-2014-204319 . [DOI] [PubMed] [Google Scholar]
14.Blakely T, Disney G, Atkinson J, et al. A typology for charting socioeconomic mortality gradients:" Go south-west". Epidemiology. 2017;28(4):594–603. 10.1097/EDE.0000000000000671 [DOI] [PubMed] [Google Scholar]
15.Mackenbach JP, Valverde JR, Bopp M, et al. Determinants of inequalities in life expectancy: an international comparative study of eight risk factors. The Lancet Public Health. 2019;4(10):e529–e37. 10.1016/S2468-2667(19)30147-1 [DOI] [PubMed] [Google Scholar]
16.Mackenbach JP, Kulhanova I, Artnik B, et al. Changes in mortality inequalities over two decades: register based study of European countries. BMJ. 2016;353:i1732 10.1136/bmj.i1732 . [DOI] [PMC free article] [PubMed] [Google Scholar]
17.Hu Y, van Lenthe FJ, Platt S, et al. The Impact of Tobacco Control Policies on Smoking Among Socioeconomic Groups in Nine European Countries, 1990–2007. Nicotine Tob Res. 2017;19(12):1441–9. Epub 2016/09/11. 10.1093/ntr/ntw210 . [DOI] [PubMed] [Google Scholar]
18.Hoffmann K, De Gelder R, Hu Y, et al. Trends in educational inequalities in obesity in 15 European countries between 1990 and 2010. Int J Behav Nutr Phys Act. 2017;14(1):63 Epub 2017/05/10. 10.1186/s12966-017-0517-8 . [DOI] [PMC free article] [PubMed] [Google Scholar]
19.Mackenbach J, Kulhánová I, Artnik B, et al. Changes over two decades in inequalities in mortality: a registered-based study of 11 European countries. BMJ. 2016;353:i1732 10.1136/bmj.i1732. [DOI] [PMC free article] [PubMed] [Google Scholar]
20.Mackenbach JP, Kulhanova I, Bopp M, et al. Inequalities in Alcohol-Related Mortality in 17 European Countries: A Retrospective Analysis of Mortality Registers. PLoS Med. 2015;12(12):e1001909 10.1371/journal.pmed.1001909 . [DOI] [PMC free article] [PubMed] [Google Scholar]
21.Blakely T, Tobias M, Atkinson J, et al. Tracking Disparity: Trends in ethnic and socioeconomic inequalities in mortality, 1981–2004. Wellington Ministry of Health, 2007. August 2007. Report No. [Google Scholar]
22.Fawcett J, Atkinson J, Blakely T. Record linkage of census and mortality 2001–04 records Wellington Health Inequalities Research Programme, University of Otago, Wellington, 2008. [Google Scholar]
23.Tobias M, Blakely T, Matheson D, et al. Changing trends in indigenous inequalities in mortality: lessons from New Zealand. Int J Epidemiol. 2009;38(6):1711–22. 10.1093/ije/dyp156 [DOI] [PubMed] [Google Scholar]
24.Blakely T, Tobias M, Atkinson J. Inequalities in mortality during and after restructuring of the New Zealand economy: repeated cohort studies. BMJ. 2008;336:371–5. 10.1136/bmj.39455.596181.25 [DOI] [PMC free article] [PubMed] [Google Scholar]
25.Blakely T, Fawcett J, Hunt D, et al. What is the contribution of smoking and socioeconomic position to ethnic inequalities in mortality in New Zealand? The Lancet. 2006;368(9529):44–52. [DOI] [PubMed] [Google Scholar]
26.Sloane K, Atkinson J, Bastiampillai N, et al. Record linkage of 2006 census records and 2006–2011 mortality and cancer registration records: unlock ratios and linkage weights. Wellington: 2015. [Google Scholar]
27.Lopez AD, Collishaw NE, Piha T. A descriptive model of the cigarette epidemic in developed countries. Tob Control. 1994;3:242–7. [Google Scholar]
28.Lynch J, Smith GD. A life course approach to chronic disease epidemiology. Annu Rev Public Health. 2005;26:1–35. 10.1146/annurev.publhealth.26.021304.144505 . [DOI] [PubMed] [Google Scholar]
29.Tekkel M, Veideman T, Rahu M. Changes over fourteen years in adult obesity in Estonia: socioeconomic status and use of outpatient services. Cent Eur J Public Health. 2010;18(4):186–91. 10.21101/cejph.b0413 [DOI] [PubMed] [Google Scholar]
30.Popkin BM. The nutrition transition and its health implications in lower-income countries. Public health nutrition. 1998;1(1):5–21. 10.1079/phn19980004 [DOI] [PubMed] [Google Scholar]
31.Capewell S, Buchan I. Why have sustained increases in obesity and type 2 diabetes not offset declines in cardiovascular mortality over recent decades in Western countries? Nutr Metab Cardiovasc Dis. 2012;22(4):307–11. Epub 2012/03/13. 10.1016/j.numecd.2012.01.005 . [DOI] [PubMed] [Google Scholar]
32.Cecchini M, Sassi F, Lauer JA, et al. Tackling of unhealthy diets, physical inactivity, and obesity: health effects and cost-effectiveness. The Lancet. 2010;376(9754):1775–84. 10.1016/s0140-6736(10)61514-0 [DOI] [PubMed] [Google Scholar]
33.Okunade AA, Rubin RM, Okunade AK. Delayed Effects of Obese and Overweight Population Conditions on All-Cause Adult Mortality Rate in the USA. Front Public Health. 2016;4:212 Epub 2016/10/14. 10.3389/fpubh.2016.00212 . [DOI] [PMC free article] [PubMed] [Google Scholar]
34.Capewell S, O'Flaherty M. Rapid mortality falls after risk-factor changes in populations. The Lancet. 2011;378(9793):752–3. 10.1016/s0140-6736(10)62302-1 [DOI] [PubMed] [Google Scholar]
35.Shkolnikov VM, Jasilionis D, Andreev EM, et al. Linked versus unlinked estimates of mortality and length of life by education and marital status: evidence from the first record linkage study in Lithuania. Soc Sci Med. 2007;64(7):1392–406. Epub 2007/01/02. 10.1016/j.socscimed.2006.11.014 . [DOI] [PubMed] [Google Scholar]
36.Renard F, Devleesschauwer B, Speybroeck N, et al. Monitoring health inequalities when the socio-economic composition changes: are the slope and relative indices of inequality appropriate? Results of a simulation study. BMC Public Health. 2019;19(1):662 Epub 2019/05/31. 10.1186/s12889-019-6980-1 [DOI] [PMC free article] [PubMed] [Google Scholar]

PLoS One. doi: 10.1371/journal.pone.0232971.r001

Decision Letter 0

Brecht Devleesschauwer

Transfer Alert

This paper was transferred from another journal. As a result, its full editorial history (including decision letters, peer reviews and author responses) may not be present.

6 Mar 2020

PONE-D-19-33671

Changing social inequalities in smoking, obesity and cause-specific mortality: cross-national comparisons using compass typology

PLOS ONE

Dear Dr. Teng,

Thank you for submitting your manuscript to PLOS ONE. After careful consideration, we feel that it has merit but does not fully meet PLOS ONE’s publication criteria as it currently stands. Therefore, we invite you to submit a revised version of the manuscript that addresses the points raised during the review process.

We would appreciate receiving your revised manuscript by Apr 20 2020 11:59PM. When you are ready to submit your revision, log on to https://www.editorialmanager.com/pone/ and select the 'Submissions Needing Revision' folder to locate your manuscript file.

If you would like to make changes to your financial disclosure, please include your updated statement in your cover letter.

To enhance the reproducibility of your results, we recommend that if applicable you deposit your laboratory protocols in protocols.io, where a protocol can be assigned its own identifier (DOI) such that it can be cited independently in the future. For instructions see: http://journals.plos.org/plosone/s/submission-guidelines#loc-laboratory-protocols

Please include the following items when submitting your revised manuscript:

A rebuttal letter that responds to EACH point raised by the academic editor and reviewer(s). This letter should be uploaded as separate file and labeled 'Response to Reviewers'.
A marked-up copy of your manuscript that highlights changes made to the original version. This file should be uploaded as separate file and labeled 'Revised Manuscript with Track Changes'.
An unmarked version of your revised paper without tracked changes. This file should be uploaded as separate file and labeled 'Manuscript'.

Please note while forming your response, if your article is accepted, you may have the opportunity to make the peer review history publicly available. The record will include editor decision letters (with reviews) and your responses to reviewer comments. If eligible, we will contact you to opt in or out.

We look forward to receiving your revised manuscript.

Kind regards,

Brecht Devleesschauwer

Academic Editor

PLOS ONE

Additional Editor Comments (if provided):

In your revision note, please include EACH comment of the reviewers, provide your reply, and when relevant, include the modified/new text (or motivate why you decided not to modify the text). Note that failure to do so may still result in a rejection of the manuscript.

Journal requirements:

When submitting your revision, we need you to address these additional requirements.

1) Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming. The PLOS ONE style templates can be found at

http://www.journals.plos.org/plosone/s/file?id=wjVg/PLOSOne_formatting_sample_main_body.pdf and http://www.journals.plos.org/plosone/s/file?id=ba62/PLOSOne_formatting_sample_title_authors_affiliations.pdf

2) Please include more information on the surveys and database used. For each country, please ensure that a link or a reference is provided to the relevant survey/ database used to collect data

3) Our internal editors have looked over your manuscript and determined that it is within the scope of our Health Inequities and Disparities Research Call for Papers. This collection of papers is headed by a team of Guest Editors for PLOS ONE: Clare Bambra, Hans Bosma, Diana Burgess, Joseph Telfair, Barbara Turner, and Jennie Popay. The Collection will encompass a diverse range of research articles on health inequities and disparities. Additional information can be found on our announcement page: hhttps://collections.plos.org/s/health-inequities

If you would like your manuscript to be considered for this collection, please let us know in your cover letter and we will ensure that your paper is treated as if you were responding to this call. If you would prefer to remove your manuscript from collection consideration, please specify this in the cover letter.

4) We note that you have indicated that data from this study are available upon request. PLOS only allows data to be available upon request if there are legal or ethical restrictions on sharing data publicly. For information on unacceptable data access restrictions, please see http://journals.plos.org/plosone/s/data-availability#loc-unacceptable-data-access-restrictions.

In your revised cover letter, please address the following prompts:

a) If there are ethical or legal restrictions on sharing a de-identified data set, please explain them in detail (e.g., data contain potentially identifying or sensitive patient information) and who has imposed them (e.g., an ethics committee). Please also provide contact information for a data access committee, ethics committee, or other institutional body to which data requests may be sent.

b) If there are no restrictions, please upload the minimal anonymized data set necessary to replicate your study findings as either Supporting Information files or to a stable, public repository and provide us with the relevant URLs, DOIs, or accession numbers. Please see http://www.bmj.com/content/340/bmj.c181.long for guidelines on how to de-identify and prepare clinical data for publication. For a list of acceptable repositories, please see http://journals.plos.org/plosone/s/data-availability#loc-recommended-repositories.

We will update your Data Availability statement on your behalf to reflect the information you provide.

5) PLOS requires an ORCID iD for the corresponding author in Editorial Manager on papers submitted after December 6th, 2016. Please ensure that you have an ORCID iD and that it is validated in Editorial Manager. To do this, go to ‘Update my Information’ (in the upper left-hand corner of the main menu), and click on the Fetch/Validate link next to the ORCID field. This will take you to the ORCID site and allow you to create a new iD or authenticate a pre-existing iD in Editorial Manager. Please see the following video for instructions on linking an ORCID iD to your Editorial Manager account: https://www.youtube.com/watch?v=_xcclfuvtxQ

6) We noted in your submission details that a portion of your manuscript may have been presented or published elsewhere. Please clarify whether this publication was peer-reviewed and formally published. If this work was previously peer-reviewed and published, in the cover letter please provide the reason that this work does not constitute dual publication and should be included in the current manuscript.

[Note: HTML markup is below. Please do not edit.]

Reviewers' comments:

Reviewer's Responses to Questions

Comments to the Author

1. Is the manuscript technically sound, and do the data support the conclusions?

The manuscript must describe a technically sound piece of scientific research with data that supports the conclusions. Experiments must have been conducted rigorously, with appropriate controls, replication, and sample sizes. The conclusions must be drawn appropriately based on the data presented.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

2. Has the statistical analysis been performed appropriately and rigorously?

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

3. Have the authors made all data underlying the findings in their manuscript fully available?

The PLOS Data policy requires authors to make all data underlying the findings described in their manuscript fully available without restriction, with rare exception (please refer to the Data Availability Statement in the manuscript PDF file). The data should be provided as part of the manuscript or its supporting information, or deposited to a public repository. For example, in addition to summary statistics, the data points behind means, medians and variance measures should be available. If there are restrictions on publicly sharing data—e.g. participant privacy or use of data from a third party—those must be specified.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

4. Is the manuscript presented in an intelligible fashion and written in standard English?

PLOS ONE does not copyedit accepted manuscripts, so the language in submitted articles must be clear, correct, and unambiguous. Any typographical or grammatical errors should be corrected at revision, so please note any specific errors here.

Reviewer #1: Yes

Reviewer #2: Yes

Reviewer #3: Yes

**********

5. Review Comments to the Author

Please use the space provided to explain your answers to the questions above. You may also include additional comments for the author, including concerns about dual publication, research ethics, or publication ethics. (Please upload your review as an attachment if it exceeds 20,000 characters)

Reviewer #1: Review of the manuscript: “Changing social inequalities in smoking, obesity and cause-specific mortality: cross-national comparisons using compass typology”

This is a remarkable paper, applying a typology previously build by the first author and aiming at integrating at a glance different components of health and inequalities and their evolution. In this paper, the authors use this typology to compare 10 countries qua health inequalities and trends for several health indicators.

Only minor changes are suggested. Results could be further developed and a bit restructured for the readability.

Introduction

L94: Please precise that it is the contribution to absolute inequalities

L200: why did you exclude laryngeal cancers from tobacco-related mortality in New Zealand? Did you receive only aggregated rates, or did you compute them yourself, and if so, why to exclude those cancers?

Methods:

It is not clear to me if you started from already computed values, or from the microdata originating from the surveys ?

L174: if you have the microdata, why not choose for the same age-ranges as for the other countries?

Results

As there are really plenty information, I think the readability would be higher is you started, for each topic, with the description of levels and trends of the health indicator, and afterwards the level and trends of inequalities. In addition, I would recommend to fully separating the genders, for each topic.

L290: “Eastern Europe countries tended…”: also Austria.

L295: not only rates in women, but also Rel and absolute ineq. are highest in NZ. Having high relative inequalities when the rate is also high is not so common and is worthwhile to mention, since it denotes very large absolute inequalities.

L307: for CVD mortality, some countries experiences a decrease of absolute inequalities, worthwhile to mention.

L311: please, comment also the decrease in cancer absolute inequalities in several countries for men, and in France, Austria, England for women, here or in the discussion part. Small inequalities in cancer in women can come from the inverse inequality observed in breast cancer.

Discussion

L335 : in Estonia and Lithuania, in men there is not only a decrease of absolute inequalities, but also in relative inequalities. The RII inversed, becoming lower than 1. Could you further discuss this ? Is this the reality, that obesity is more prevalent in higher educated men in those countries, and if so,why ? or could this be an artefact of the data ? the evolution of inequalities in obesity is different in women (decreases in ES, but increases in Lith).

Inequalities seem particularly high in New Zealand, at least for some heath topics. Could you further comment, and maybe provide hypothesis and clues for policies ?

Strength and limitations: the chosen inequality indicators, although include information available from all SE levels, do not adequately reflect shifts in the distribution of the socioeconomic status. Even if the distribution of the socioeconomic status affects the value of the RII and SII, an improvement of the SE distribution is not always (and even mostly not) translated into an improvement of the indicators. Please see discussion of this here https://bmcpublichealth.biomedcentral.com/articles/10.1186/s12889-019-6980-1

Implications practice policy

L405 : “to visualise progress on social inequalities”. I would rather say: “to visualize progress on social health inequalities” . Indeed, the typology cannot show the social inequalities themselves (that are difference in SES itself, for instance magnitude of income differences in, difference in education length/content, differences in wealth), rather it shows differences in health by SES . Social differences create differences in health by SE status.

Fig 1: title on the left side, I would keep ‘Smoking’ and ‘Obesity’ since you do not display the prevalence.

Fig 2: Why is X axis title called “average all exposure mortality rate” and not “smoking-related mortality rate” ?

Reviewer #2: This article uses a compass typology to show trends in rates (or prevalence) and absolute and relative inequalities. The authors examine smoking and obesity prevalence and smoking-related and cause-specific mortality outcomes, across 10 countries. The compass typology seems to me to be a useful way to summarise trends and inequalities simultaneously and possibly predict future trends. I have some specific comments:-

• Last sentence in the abstract – say why the compass typology is useful. Also in first key message. Could add the extension of work to include risk factors as that is new here

• Can you briefly describe the Preston-Glei-Wilmouth method

• Which European Standard Population was used?

• Can you say why education was used here as proxy for socio-economic status. Was this measure easiest to harmonise across countries?

• Which nationally representative surveys were used in each country?

• ‘Compass plots have previously been used only for rates’ – it would be worth citing ref[14] again here, alongside other references if applicable

• Define COPD abbreviation earlier

• Lines 293 and 421 – morality is written instead of mortality

• Figures (1-3b) are blurred and difficult to read

• Mortality data for most countries are presented until around 2011 at the latest, so that international comparisons can be made. There is some scope however, to update the analysis further by examining ‘future’ trends in mortality (deaths up to 2018?) for at least some of the countries. If so, you would be able to see if trends continue as ‘predicted’

• The results shown are for males and females, approximately aged 35-79. SIIs and RIIs vary by age. How are trends likely to differ when looking at different age ranges or groups?

• What are the suggested approaches for preventing smoking/obesity in the least educated groups?

• In the supplementary material, Figure S3 – the values on the y-axis are missing for some female plots

Reviewer #3: The proposed new kind of application of the compass methodology seems to be very promising also for further application. In this work there are some obvious limitations (accurately described) due to the availability of data across all the considered countries and time range. However, it is worth of publication, thanks to the accurate use of the methods and the obtained results.

Some minor revisions are suggested:

- the authors should better reproduce the formulas, maybe with an appropriate editor, only for reading reasons in the final version of the work;

- the authors should consider if a clearer version of the graphs could improve their understanding for the reader. They are clear for me, but I'm not sure that all the interested readers should make the same consideration.

After these revisions, the article can be published.

**********

6. PLOS authors have the option to publish the peer review history of their article (what does this mean?). If published, this will include your full peer review and any attached files.

If you choose “no”, your identity will remain anonymous but your review may still be made public.

Do you want your identity to be public for this peer review? For information about this choice, including consent withdrawal, please see our Privacy Policy.

Reviewer #1: Yes: Françoise Renard

Reviewer #2: No

Reviewer #3: No

[NOTE: If reviewer comments were submitted as an attachment file, they will be attached to this email and accessible via the submission site. Please log into your account, locate the manuscript record, and check for the action link "View Attachments". If this link does not appear, there are no attachment files to be viewed.]

While revising your submission, please upload your figure files to the Preflight Analysis and Conversion Engine (PACE) digital diagnostic tool, https://pacev2.apexcovantage.com/. PACE helps ensure that figures meet PLOS requirements. To use PACE, you must first register as a user. Registration is free. Then, login and navigate to the UPLOAD tab, where you will find detailed instructions on how to use the tool. If you encounter any issues or have any questions when using PACE, please email us at figures@plos.org. Please note that Supporting Information files do not need this step.

PLoS One. 2020 Jul 10;15(7):e0232971. doi: 10.1371/journal.pone.0232971.r002

Author response to Decision Letter 0

8 Apr 2020

Department of Public Health

University of Otago Wellington

Wellington, New Zealand

6 April 2020

Dear Editor / Brecht Devleesschauwer

Re: PONE-D-19-33671 Changing social inequalities in smoking, obesity and cause-specific mortality: cross-national comparisons using compass typology.

Many thanks for your request to submit a revised version of this manuscript to PLOS ONE. Please find our response to each of the Journal requirements and Reviewers’ comments below.

We would be happy for this paper to be considered for the Health Inequities and Disparities Research Call for Papers or as a general publication.

None of the datasets we used are publically available. There are ethical and legal restrictions on accessing the datasets, to maintain confidentiality for identifiable patient information. Further information on source datasets is now outlined in Table S5a to S5c which is referenced in the manuscript.

Similar data to that used in this manuscript have been published in peer reviewed journals elsewhere,1-5 however this work does not constitute dual publication. The novel contribution of this manuscript is the use a novel typology to compare inequality trends across cause specific mortalities and risk factors. The contribution that this paper makes is highlighted by all of the reviewers. Previous publications are clearly referenced in the manuscript, and none compared trends across Europe and New Zealand, or combined cause-specific mortality trends with inequality trends in risk factors smoking and obesity.

In another small change to the submission, I (AT) will be on maternity leave after the 3 April until 9 November. To cover this period, Prof Tony Blakely has agreed to also act as a corresponding author.

Kind regards,

Dr Andrea Teng

(on behalf of the co-authors)

1 Mackenbach, J. P. et al. Trends in inequalities in premature mortality: a study of 3.2 million deaths in 13 European countries. J Epidemiol Community Health 69, 207-217 (2014).

2 Mackenbach, J. P. et al. Changes in mortality inequalities over two decades: register based study of European countries. BMJ 353, i1732, doi:10.1136/bmj.i1732 (2016).

3 Blakely, T., Disney, G., Atkinson, J., Teng, A. & Mackenbach, J. P. A typology for charting socioeconomic mortality gradients:" Go south-west". Epidemiology 28, 594-603, doi:10.1097/EDE.0000000000000671 (2017).

4 Hu, Y. et al. The Impact of Tobacco Control Policies on Smoking Among Socioeconomic Groups in Nine European Countries, 1990-2007. Nicotine Tob Res 19, 1441-1449, doi:10.1093/ntr/ntw210 (2017).

5 Hoffmann, K. et al. Trends in educational inequalities in obesity in 15 European countries between 1990 and 2010. Int J Behav Nutr Phys Act 14, 63, doi:10.1186/s12966-017-0517-8 (2017).

Journal requirements

When submitting your revision, we need you to address these additional requirements.

1) Please ensure that your manuscript meets PLOS ONE's style requirements, including those for file naming.

Authors’ response: To align with journal requirements: we have updated the title page. Figure names in the manuscript have now been abbreviated to ‘Fig 1’ etc. Equations have been numbered. Supplementary files have been renamed and each table has been submitted separately.

2) Please include more information on the surveys and database used. For each country, please ensure that a link or a reference is provided to the relevant survey/ database used to collect data

Authors’ response: Table S5a to S5c has been added to provide more information on the surveys and databases that were used.

Authors’ response: We agree to this manuscript being considered for this collection and this is outlined in the cover letter.

Authors’ response: None of the datasets we used are publically available due to legal and ethical restrictions. Table S5a to S5c provides further details on the datasets that were used.

In your revised cover letter, please address the following prompts:

Authors’ response: A large number of datasets are used in this study. There are legal and ethical restrictions on sharing all of these de-identified datasets which differ between countries. Please see Table S5a to S5c for further information on the datasets that were used.

We will update your Data Availability statement on your behalf to reflect the information you provide.

Authors’ response: NA, as above.

Authors’ response: AT has added her ORCID to her PLOS ONE profile.

Authors’ response: Please see the cover letter.

________________________________________

Reviewer #1:

Review of the manuscript: “Changing social inequalities in smoking, obesity and cause-specific mortality: cross-national comparisons using compass typology”

Authors’ response: Many thanks for reviewing this paper and thank you for highlighting the value in the typology approach.

Only minor changes are suggested. Results could be further developed and a bit restructured for the readability.

Authors’ response: Please see responses below.

Introduction

L94: Please precise that it is the contribution to absolute inequalities

Authors’ response: ‘absolute’ has been added to the description.

Authors’ response: This was due to the way laryngeal cancers had been aggregated in the existing census mortality linked dataset. The original mortality ICD codes were not available in the linked datasets because they had been aggregated before data linkage was done, for consistency and confidentiality reasons with small numbers. Therefore, it would have been a slow and complicated process, and possibly not successful due to the length of time since data linkage, to have been able to get the external body, ie, Stats NZ, to put this information on the files for us, with little potential additional gain.

Methods:

It is not clear to me if you started from already computed values, or from the microdata originating from the surveys ?

L174: if you have the microdata, why not choose for the same age-ranges as for the other countries?

Authors’ response: NZ and European smoking and obesity analyses were largely done for the 30-79 year old age group. However some countries did not have data for the very oldest age groups. We have corrected the age ranges in the text for obesity because previously cited age groups as used in the Hoffman paper were incorrect. Sentence now reads; “This was done for survey participants 30-79 years old, with the exception of Hungary and Lithuania where the upper age limit was 64 years old.”

Results

Authors’ response: Please see Results for change in order of results description and clearer separation of men and women in descriptions. We believe the comparison between men and women is useful and therefore some of these comparisons have been left in.

Space limitations prevented a thorough description of levels and trends of every health indicator.

L290: “Eastern Europe countries tended…”: also Austria.

Authors’ response: Yes the relative inequalities were low in Austria but because there was no evidence they crossed over from RR<1 to RR>1 we have left this sentence as it stands.

Authors’ response: New sentence reads: “The lowest smoking related mortality rates in women were in Lithuania and the highest were in New Zealand which also experienced the highest absolute and relative inequalities.”

L307: for CVD mortality, some countries experiences a decrease of absolute inequalities, worthwhile to mention.

Authors’ response: Thanks, this sentence has now been added; “There was a decrease in absolute inequalities in men and women from England and Wales, Finland, and France.”

Authors’ response: Thank you, a new sentence has been added to the results: “There was a decrease in absolute inequalities in cancer for men and women in England and Wales, and for men in France.”

On our reading of the typologies there was no decrease in absolute inequalities (crossing of the contours) in France, Austria, England for women. We also note that the pattern of inequalities in breast cancer has reversed in some countries, to become disproportionately greater in low SEP/education groups, possibly affecting the differences between men and women.

Discussion

Authors’ response: We have added some further clarification to the Discussion, also referencing a previous study and hypotheses for these results. The sentence in the Discussion was revised to read: “This was combined with increasing absolute inequalities in many countries (eastward trend) with two countries reporting decreased a decrease in absolute and relative inequalities where obesity became more prevalent in men with higher education (Lithuania and Estonia) but this pattern was not noted for woman. This finding has been seen previously in Estonia [29] and the authors’ hypothesise physical activity from manual occupations as a contributing factor. It also may be related to being on an earlier stage in the nutrition transition [30], where energy dense food became more accessible to the most educated first after the dissolution of the Soviet Union, and then later became more concentrated in the least educated.”

Inequalities seem particularly high in New Zealand, at least for some health topics. Could you further comment, and maybe provide hypothesis and clues for policies ?

Authors’ response: NZ did have high absolute and relative inequalities in obesity prevalence, cancer in women and smoking-related mortality in women. NZ is known to historically have had high rates of smoking in women and high levels of inequality in this group (high Māori women smoking rates). These patterns may have contributed to the cancer and smoking-related mortality inequalities in NZ women. The impact of historical trends in smoking is discussed in the text. Also the worsening inequalities in obesity in NZ (and two other countries) are discussed from Line 338, 2nd paragraph in the discussion, which have also been linked to cancer inequalities (described in text).

Authors’ response: The interpretation of (changes in) the RII and SII is indeed complex. The Renard et al. paper focusses on a problem that is not directly relevant to our analysis: they warn against using the RII/SII when "improving the educational attainment is part of a policy", and highlight that under certain conditions an increase in the percentage of high educated will push the RII/SII up. We have however added this sentence to the limitations:

"Although the RII and SII are superior to more simple measures of inequalities, the interpretation can be complex, as illustrated by Renard, Devleesschauwer (35) who show that under certain conditions improvements in the education distribution can increase the SII/RII.

Implications practice policy

L405 : “to visualise progress on social inequalities”. I would rather say: “to visualize progress on social health inequalities” . Indeed, the typology cannot show the social inequalities themselves (that are difference in SES itself, for instance magnitude of income differences in, difference in education length/content, differences in wealth), rather it shows differences in health by SES. Social differences create differences in health by SE status.

Authors’ response: Thank you, we agree and this change has been made.

Fig 1: title on the left side, I would keep ‘Smoking’ and ‘Obesity’ since you do not display the prevalence.

Authors’ response: We agree, this change has been made.

Fig 2: Why is X axis title called “average all exposure mortality rate” and not “smoking-related mortality rate” ?

Authors’ response: The mortality rate is the average across the three levels of education exposure.

Reviewer #2:

This article uses a compass typology to show trends in rates (or prevalence) and absolute and relative inequalities. The authors examine smoking and obesity prevalence and smoking-related and cause-specific mortality outcomes, across 10 countries. The compass typology seems to me to be a useful way to summarise trends and inequalities simultaneously and possibly predict future trends. I have some specific comments:-

Authors’ response: Many thanks for reviewing this paper.

• Last sentence in the abstract – say why the compass typology is useful. Also in first key message. Could add the extension of work to include risk factors as that is new here

Authors’ response: Thank you. Abstract and key message have been changed to read “The compass typology was useful to compare trends in inequalities because it simultaneously tracks changes in rates/odds, and absolute and relative inequality measures.” Additional sentence added to this key message “The typology was extended for the first time to include risk factors as well as cause-specific mortality.”

• Can you briefly describe the Preston-Glei-Wilmouth method

Authors’ response: This sentence in the Introduction was updated and now reads “For example the Preston-Glei-Wilmoth method relies on lung cancer mortality as an indicator of smoking prevalence and was used to estimate the contribution of smoking to absolute educational inequalities in all-cause mortality in 14 European countries from 1990-2004”

• Which European Standard Population was used?

Authors’ response: 1976. This is now indicated in the Method’s description of standardisation.

• Can you say why education was used here as proxy for socio-economic status. Was this measure easiest to harmonise across countries?

Authors’ response: This sentence was added to the methods: “Education was selected as a proxy for socioeconomic position given its widespread availability; ease of harmonization across countries and lower sensitivity to health-related social mobility than other socioeconomic measures.”

• Which nationally representative surveys were used in each country?

Authors’ response: Tables S5a to S5c hasve been added to the supplementary material to outline the datasets and surveys used in each country and this is now referenced in the methods; “For more information on the source mortality 1-3 and survey risk factor 4,5 datasets please see previous publications and Supplementary Tables S5a to S5c.”

• ‘Compass plots have previously been used only for rates’ – it would be worth citing ref[14] again here, alongside other references if applicable

Authors’ response: Ref 14 has now been cited.

• Define COPD abbreviation earlier

Authors’ response: Thanks, this has now been done.

• Lines 293 and 421 – morality is written instead of mortality

Authors’ response: This has now been corrected. Thanks.

• Figures (1-3b) are blurred and difficult to read

Authors’ response: Apologies, improved versions of these will be updated on the revised submission .

Authors’ response: This could be the focus of a second paper, perhaps when the next decade of data is available for these types of analysis. Delays between death and data availability can prevent earlier analyses.

• The results shown are for males and females, approximately aged 35-79. SIIs and RIIs vary by age. How are trends likely to differ when looking at different age ranges or groups?

Authors’ response: Age differences are something that was examined in a previous typology paper (Blakely, T., Disney, G., Atkinson, J., Teng, A. & Mackenbach, J. P. A typology for charting socioeconomic mortality gradients:" Go south-west". Epidemiology 28, 594-603 (2017).) Mortality trends were likely to occur at higher rates in older age groups/ranges, and this point is acknowledged in our limitations. Inequality trends in all-cause mortality in the above paper however were similar across age groups, so small age group variations may have a limited effect on the observed inequality trends.

• What are the suggested approaches for preventing smoking/obesity in the least educated groups?

Authors’ response: Ideally policy implications should follow from the data and analyses we have carried out. We have not evaluated smoking or obesity prevention policies here, so we have not made any recommendations about these. Also, further research is also needed to understand how nutrition policies such as fiscal measures, addressing availability and marketing restrictions may affect the least educated groups and health inequalities.

• In the supplementary material, Figure S3 – the values on the y-axis are missing for some female plots

Authors’ response: Thank you, this has now been corrected.

Reviewer #3:

The proposed new kind of application of the compass methodology seems to be very promising also for further application. In this work there are some obvious limitations (accurately described) due to the availability of data across all the considered countries and time range. However, it is worth of publication, thanks to the accurate use of the methods and the obtained results.

Authors’ response: Many thanks for reviewing this paper.

Some minor revisions are suggested:

- the authors should better reproduce the formulas, maybe with an appropriate editor, only for reading reasons in the final version of the work;

Authors’ response: Current equations are produced in word equation editor and font can be increased as necessary. Please let us know if there is a preferred format for submitting equations and we would be happy to revise.

Authors’ response: Please see response to reviewer above. Quality of figures has now been improved via the submission process.

After these revisions, the article can be published.

Attachment

Submitted filename: Response to reviewers FINAL.docx

Click here for additional data file.^{(60.6KB, docx)}

PLoS One. doi: 10.1371/journal.pone.0232971.r003

Decision Letter 1

Brecht Devleesschauwer

27 Apr 2020

Changing social inequalities in smoking, obesity and cause-specific mortality: cross-national comparisons using compass typology

PONE-D-19-33671R1

Dear Dr. Teng,

We are pleased to inform you that your manuscript has been judged scientifically suitable for publication and will be formally accepted for publication once it complies with all outstanding technical requirements.

Within one week, you will receive an e-mail containing information on the amendments required prior to publication. When all required modifications have been addressed, you will receive a formal acceptance letter and your manuscript will proceed to our production department and be scheduled for publication.

Shortly after the formal acceptance letter is sent, an invoice for payment will follow. To ensure an efficient production and billing process, please log into Editorial Manager at https://www.editorialmanager.com/pone/, click the "Update My Information" link at the top of the page, and update your user information. If you have any billing related questions, please contact our Author Billing department directly at authorbilling@plos.org.

If your institution or institutions have a press office, please notify them about your upcoming paper to enable them to help maximize its impact. If they will be preparing press materials for this manuscript, you must inform our press team as soon as possible and no later than 48 hours after receiving the formal acceptance. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information, please contact onepress@plos.org.

With kind regards,

Brecht Devleesschauwer

Academic Editor

PLOS ONE

Additional Editor Comments (optional):

Thank you for addressing the reviewer comments.

Reviewers' comments:

PLoS One. doi: 10.1371/journal.pone.0232971.r004

Acceptance letter

Brecht Devleesschauwer

25 Jun 2020

PONE-D-19-33671R1

Changing social inequalities in smoking, obesity and cause-specific mortality: cross-national comparisons using compass typology

Dear Dr. Teng:

I'm pleased to inform you that your manuscript has been deemed suitable for publication in PLOS ONE. Congratulations! Your manuscript is now with our production department.

If your institution or institutions have a press office, please let them know about your upcoming paper now to help maximize its impact. If they'll be preparing press materials, please inform our press team within the next 48 hours. Your manuscript will remain under strict press embargo until 2 pm Eastern Time on the date of publication. For more information please contact onepress@plos.org.

If we can help with anything else, please email us at plosone@plos.org.

Thank you for submitting your work to PLOS ONE and supporting open access.

Kind regards,

PLOS ONE Editorial Office Staff

on behalf of

Prof. Dr. Brecht Devleesschauwer

Academic Editor

PLOS ONE

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

S1 File. Further methods.

(PDF)

Click here for additional data file.^{(307.8KB, pdf)}

S1 Spreadsheet. Risk factor prevalences (subsequently converted into odds), SIIs and RIIs and confidence intervals.

(XLSX)

Click here for additional data file.^{(30.7KB, xlsx)}

S2 Spreadsheet. Mortality rates, SIIs and RIIs and confidence intervals.

(XLSX)

Click here for additional data file.^{(76.5KB, xlsx)}

S1 Table. Countries excluded because mortality data was not available.

(DOCX)

Click here for additional data file.^{(15.2KB, docx)}

S2 Table. Countries excluded because mortality inequality data was not available from both the 1980s and 2000s decades.

(DOCX)

Click here for additional data file.^{(17.2KB, docx)}

S3 Table. ICD coding for the cause-specific mortality groups.

(DOCX)

Click here for additional data file.^{(15.3KB, docx)}

S4 Table. Compass trajectories for general mortality cause specific inequality trends between 1980 and 2010, and smoking and obesity inequalities between 1990 and 2010.

(DOCX)

Click here for additional data file.^{(16.8KB, docx)}

S5 Table. Characteristics of mortality, smoking and obesity data.

(DOCX)

Click here for additional data file.^{(39.6KB, docx)}

S6 Table. Source mortality and risk factor datasets, where available.

(DOCX)

Click here for additional data file.^{(20.1KB, docx)}

S1 Fig. Trends in age-standardised smoking prevalence, all education groups combined, 30–79 year olds 1980–2010.

(PDF)

Click here for additional data file.^{(241.8KB, pdf)}

S2 Fig. Trends in age-standardised obesity prevalence, 30–79 year olds 1980–2010.

(PDF)

Click here for additional data file.^{(223.2KB, pdf)}

S3 Fig. Age-standardized mortality rates of CVD, cancer, external mortality, smoking-related mortality, in men (left) and women (right) 35–79 years.

(PDF)

Click here for additional data file.^{(275.4KB, pdf)}

S1 Data

(XLSX)

Click here for additional data file.^{(76.5KB, xlsx)}

S2 Data

(XLSX)

Click here for additional data file.^{(30.7KB, xlsx)}

Attachment

Submitted filename: Response to reviewers FINAL.docx

Click here for additional data file.^{(60.6KB, docx)}

Data Availability Statement

[pone.0232971.ref001] 1.Giskes K, Kunst AE, Benach J, et al. Trends in smoking behaviour between 1985 and 2000 in nine European countries by education. J Epidemiol Community Health. 2005;59(5):395–401. Epub 2005/04/16. 10.1136/jech.2004.025684 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0232971.ref002] 2.Huisman M, Kunst AE, Mackenbach JP. Educational inequalities in smoking among men and women aged 16 years and older in 11 European countries. Tob Control. 2005;14(2):106–13. Epub 2005/03/26. 10.1136/tc.2004.008573 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0232971.ref003] 3.Huisman M, Kunst AE, Mackenbach JP. Inequalities in the prevalence of smoking in the European Union: comparing education and income. Prev Med. 2005;40(6):756–64. Epub 2005/04/27. 10.1016/j.ypmed.2004.09.022 . [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref004] 4.Jha P, Peto R, Zatonski W, et al. Social inequalities in male mortality, and in male mortality from smoking: indirect estimation from national death rates in England and Wales, Poland, and North America. The Lancet. 2006;368(9533):367–70. 10.1016/s0140-6736(06)68975-7 [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref005] 5.Gregoraci G, Lenthe FJv, Artnik B, et al. Contribution of smoking to socioeconomic inequalities in mortality: a study of 14 European countries, 1990–2004. Tobacco Control. 2017;26:206–68. 10.1136/tobaccocontrol2015-052766 [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref006] 6.Martikainen P, Ho JY, Preston S, et al. The changing contribution of smoking to educational differences in life expectancy: indirect estimates for Finnish men and women from 1971 to 2010. J Epidemiol Community Health. 2013;67(3):219–24. 10.1136/jech-2012-201266 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0232971.ref007] 7.Blakely T, Wilson N. The contribution of smoking to inequalities in mortality by education varies over time and by sex: two national cohort studies, 1981–84 and 1996–99. Int J Epidemiol. 2005;34(5):1054–62. 10.1093/ije/dyi172 . [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref008] 8.Teng A, Atkinson J, Disney G, et al. Changing smoking-mortality association over time and across social groups: National census-mortality cohort studies from 1981 to 2011. Sci Rep. 2017;7(1):11465 10.1038/s41598-017-11785-x . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0232971.ref009] 9.Aune D, Sen A, Prasad M, et al. BMI and all cause mortality: systematic review and non-linear dose-response meta-analysis of 230 cohort studies with 3.74 million deaths among 30.3 million participants. BMJ. 2016;353:i2156 10.1136/bmj.i2156 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0232971.ref010] 10.Bono F, Matranga D. Socioeconomic inequality in non-communicable diseases in Europe between 2004 and 2015: evidence from the SHARE survey. Eur J Public Health. 2018. Epub 2018/09/01. 10.1093/eurpub/cky165 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0232971.ref011] 11.Hoffmann R, Eikemo TA, Kulhanova I, et al. Obesity and the potential reduction of social inequalities in mortality: evidence from 21 European populations. Eur J Public Health. 2015;25(5):849–56. 10.1093/eurpub/ckv090 . [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref012] 12.Vallejo-Torres L, Morris S. The contribution of smoking and obesity to income-related inequalities in health in England. Soc Sci Med. 2010;71(6):1189–98. Epub 2010/07/16. 10.1016/j.socscimed.2010.06.008 . [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref013] 13.Mackenbach JP, Kulhanova I, Menvielle G, et al. Trends in inequalities in premature mortality: a study of 3.2 million deaths in 13 European countries. J Epidemiol Community Health. 2014;69(3):207–17. 10.1136/jech-2014-204319 . [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref014] 14.Blakely T, Disney G, Atkinson J, et al. A typology for charting socioeconomic mortality gradients:" Go south-west". Epidemiology. 2017;28(4):594–603. 10.1097/EDE.0000000000000671 [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref015] 15.Mackenbach JP, Valverde JR, Bopp M, et al. Determinants of inequalities in life expectancy: an international comparative study of eight risk factors. The Lancet Public Health. 2019;4(10):e529–e37. 10.1016/S2468-2667(19)30147-1 [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref016] 16.Mackenbach JP, Kulhanova I, Artnik B, et al. Changes in mortality inequalities over two decades: register based study of European countries. BMJ. 2016;353:i1732 10.1136/bmj.i1732 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0232971.ref017] 17.Hu Y, van Lenthe FJ, Platt S, et al. The Impact of Tobacco Control Policies on Smoking Among Socioeconomic Groups in Nine European Countries, 1990–2007. Nicotine Tob Res. 2017;19(12):1441–9. Epub 2016/09/11. 10.1093/ntr/ntw210 . [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref018] 18.Hoffmann K, De Gelder R, Hu Y, et al. Trends in educational inequalities in obesity in 15 European countries between 1990 and 2010. Int J Behav Nutr Phys Act. 2017;14(1):63 Epub 2017/05/10. 10.1186/s12966-017-0517-8 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0232971.ref019] 19.Mackenbach J, Kulhánová I, Artnik B, et al. Changes over two decades in inequalities in mortality: a registered-based study of 11 European countries. BMJ. 2016;353:i1732 10.1136/bmj.i1732. [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0232971.ref020] 20.Mackenbach JP, Kulhanova I, Bopp M, et al. Inequalities in Alcohol-Related Mortality in 17 European Countries: A Retrospective Analysis of Mortality Registers. PLoS Med. 2015;12(12):e1001909 10.1371/journal.pmed.1001909 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0232971.ref021] 21.Blakely T, Tobias M, Atkinson J, et al. Tracking Disparity: Trends in ethnic and socioeconomic inequalities in mortality, 1981–2004. Wellington Ministry of Health, 2007. August 2007. Report No. [Google Scholar]

[pone.0232971.ref022] 22.Fawcett J, Atkinson J, Blakely T. Record linkage of census and mortality 2001–04 records Wellington Health Inequalities Research Programme, University of Otago, Wellington, 2008. [Google Scholar]

[pone.0232971.ref023] 23.Tobias M, Blakely T, Matheson D, et al. Changing trends in indigenous inequalities in mortality: lessons from New Zealand. Int J Epidemiol. 2009;38(6):1711–22. 10.1093/ije/dyp156 [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref024] 24.Blakely T, Tobias M, Atkinson J. Inequalities in mortality during and after restructuring of the New Zealand economy: repeated cohort studies. BMJ. 2008;336:371–5. 10.1136/bmj.39455.596181.25 [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0232971.ref025] 25.Blakely T, Fawcett J, Hunt D, et al. What is the contribution of smoking and socioeconomic position to ethnic inequalities in mortality in New Zealand? The Lancet. 2006;368(9529):44–52. [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref026] 26.Sloane K, Atkinson J, Bastiampillai N, et al. Record linkage of 2006 census records and 2006–2011 mortality and cancer registration records: unlock ratios and linkage weights. Wellington: 2015. [Google Scholar]

[pone.0232971.ref027] 27.Lopez AD, Collishaw NE, Piha T. A descriptive model of the cigarette epidemic in developed countries. Tob Control. 1994;3:242–7. [Google Scholar]

[pone.0232971.ref028] 28.Lynch J, Smith GD. A life course approach to chronic disease epidemiology. Annu Rev Public Health. 2005;26:1–35. 10.1146/annurev.publhealth.26.021304.144505 . [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref029] 29.Tekkel M, Veideman T, Rahu M. Changes over fourteen years in adult obesity in Estonia: socioeconomic status and use of outpatient services. Cent Eur J Public Health. 2010;18(4):186–91. 10.21101/cejph.b0413 [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref030] 30.Popkin BM. The nutrition transition and its health implications in lower-income countries. Public health nutrition. 1998;1(1):5–21. 10.1079/phn19980004 [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref031] 31.Capewell S, Buchan I. Why have sustained increases in obesity and type 2 diabetes not offset declines in cardiovascular mortality over recent decades in Western countries? Nutr Metab Cardiovasc Dis. 2012;22(4):307–11. Epub 2012/03/13. 10.1016/j.numecd.2012.01.005 . [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref032] 32.Cecchini M, Sassi F, Lauer JA, et al. Tackling of unhealthy diets, physical inactivity, and obesity: health effects and cost-effectiveness. The Lancet. 2010;376(9754):1775–84. 10.1016/s0140-6736(10)61514-0 [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref033] 33.Okunade AA, Rubin RM, Okunade AK. Delayed Effects of Obese and Overweight Population Conditions on All-Cause Adult Mortality Rate in the USA. Front Public Health. 2016;4:212 Epub 2016/10/14. 10.3389/fpubh.2016.00212 . [DOI] [PMC free article] [PubMed] [Google Scholar]

[pone.0232971.ref034] 34.Capewell S, O'Flaherty M. Rapid mortality falls after risk-factor changes in populations. The Lancet. 2011;378(9793):752–3. 10.1016/s0140-6736(10)62302-1 [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref035] 35.Shkolnikov VM, Jasilionis D, Andreev EM, et al. Linked versus unlinked estimates of mortality and length of life by education and marital status: evidence from the first record linkage study in Lithuania. Soc Sci Med. 2007;64(7):1392–406. Epub 2007/01/02. 10.1016/j.socscimed.2006.11.014 . [DOI] [PubMed] [Google Scholar]

[pone.0232971.ref036] 36.Renard F, Devleesschauwer B, Speybroeck N, et al. Monitoring health inequalities when the socio-economic composition changes: are the slope and relative indices of inequality appropriate? Results of a simulation study. BMC Public Health. 2019;19(1):662 Epub 2019/05/31. 10.1186/s12889-019-6980-1 [DOI] [PMC free article] [PubMed] [Google Scholar]

PERMALINK

Changing social inequalities in smoking, obesity and cause-specific mortality: Cross-national comparisons using compass typology

Andrea Teng

Tony Blakely

June Atkinson

Ramunė Kalėdienė

Mall Leinsalu

Pekka T Martikainen

Jitka Rychtaříková

Johan P Mackenbach

Roles

Abstract

Background

Methods

Findings

Conclusions

Introduction

Aims and objectives

Methods

Table 1. Country data for socioeconomic inequalities in mortality and smoking/obesity prevalence by year.

Datasets

Analysis

Compass typology plots

Results

Smoking

Fig 1. Trends and education inequalities in smoking prevalence (above) and obesity prevalence (below) 1990–2010.

Obesity

Smoking-related mortality

Fig 2. Trends and education inequalities in smoking-related mortality (lung cancer, COPD and laryngeal cancer), 1980–2010.

Cause-specific mortality

Fig 3. Trends and education inequalities in cause-specific mortality, 1980–2010.

Discussion

Strengths and limitations

Implications practice policy and research

Conclusions

Supporting information

Acknowledgments

Data Availability

Funding Statement

References

Decision Letter 0

Brecht Devleesschauwer

Roles

Transfer Alert

Author response to Decision Letter 0

Decision Letter 1

Brecht Devleesschauwer

Roles

Acceptance letter

Brecht Devleesschauwer

Roles

Associated Data

Supplementary Materials

Data Availability Statement

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases