Portugal has seen significant improvements in the health status of its citizens over the last 25 years. Since 1980, when the country had some the worst health indicators in Europe, there has been continuous improvement, closing the gap in relation to other countries. Indicators that have improved include perinatal and infant mortality, mean life expectancy at birth, potential years of life lost before age 65 and mortality before age 65 from the most common causes, including ischemic heart disease (IHD), stroke and motor vehicle accidents.1

Cardiovascular disease (CVD) is among the principal causes of morbidity, mortality and disability in Portugal, particularly stroke and IHD, which are the third and fourth cause of potential years of life lost and the leading cause of death in Portugal in both sexes.2 The increasing societal burden of these diseases has social, economic and cultural impacts and highlights the need for health promotion and disease prevention strategies that take account of different populational characteristics, such as gender, ethnicity and socioeconomic status (SES).2

The etiology of CVD is atherosclerotic in most cases. Various cardiovascular risk factors (CVRFs), both modifiable (including hypertension [HTN], diabetes mellitus [DM], hypercholesterolemia, dyslipidemia, obesity, smoking, physical inactivity, diet, alcohol abuse and stress) and non-modifiable (gender, age, and personal and family history), contribute to its onset, clinical course, complications and prognosis. It is estimated that 75% of cases of CVD can be attributed to modifiable CVRFs and are thus preventable, treatable and/or controllable.3

The 2004–2010 and 2012–2016 Portuguese National Health Plans2,4 established various priorities, including CVD prevention, treatment and rehabilitation and promotion of healthy lifestyles, as well as equality in health care.

Various theories have been proposed to explain the association between SES and CVD, in most of which exposure to triggering factors, patient needs and access to resources are important variables, which demonstrates the crucial role SES plays in health and how it interacts with other factors. What is certain is that differences in health between socioeconomic groups are partly due to their different social and cultural experiences. Differences in access to health care can lead to inequity and injustice, and should therefore be identified and corrected.5–7

According to the World Health Organization (WHO), in spite of the progress made there are still significant differences in the health status of Portuguese citizens according to gender, geographical region and SES.1 The health system currently faces the challenge of consolidating and improving the population's health status, reducing inequalities between groups and meeting the expectations of the people, while dealing with the problems of risk behaviors including dietary habits and smoking and of the financial sustainability of the health system.1,8

Despite the importance of these issues, particularly with regard to CVD, there has been little research on the subject. There is thus a need to conduct studies analyzing the relation of socioeconomic inequalities to CVD in Portugal, in order to fill the gaps in our knowledge and provide data on which to base policies for reorganizing health care provision to respond to citizens’ needs.

The present study sets out to analyze the distribution of CVD using data from the Fourth National Health Survey (NHS), classifying the population according to household equivalent income and determining the distribution of CVD, IHD, stroke and CVRFs using odds ratios (OR) in the adult population aged 35–74 years.

We conducted an exploratory descriptive cross-sectional study with the overall objective of analyzing the association between CVD, its risk factors, and use of health services according to SES in the Portuguese population. Specific objectives were:

• •

  to analyze the association between IHD and SES;

• •

  to analyze the association between stroke and SES;

• •

  to analyze the association between CVRFs (DM, HTN, obesity, smoking, psychological distress and physical inactivity) and SES; and

• •

  to analyze the association between number of medical visits and SES.

The Fourth NHS was conducted under the aegis of the National Institute of Statistics and the Dr. Ricardo Jorge National Institute of Health (NIH).9

The population covered by the survey consisted of individuals living in family households in Portugal at the time of the survey; those living in collective or other non-traditional types of housing were excluded. In each region, the areas selected were distributed in a relatively uniform manner by three-monthly and weekly periods in order to minimize seasonal effects on the results. The interviews were conducted in person between February 2005 and February 2006 by specially trained interviewers of all individuals residing in each household included in the selected sample and data were recorded in digital format. Full interviews were achieved in 76% of cases.9

The sample consisted of 15 239 households, corresponding to 41 193 individuals (15 457 families).9

The study population consisted of 21 807 adults (aged 35–74 inclusive) from the Fourth NHS database for whom information was available on household income.

Excel 2007© and SPSS Statistics 17.0 were used to analyze the data.

Absolute, relative and, when applicable, cumulative relative frequencies were used to characterize the sample, and minimum and maximum values, means and standard deviations were determined for quantitative variables.

ORs and 95% confidence intervals (CI) were used for analysis of the association between CVD, CVRFs and SES. ORs for SES were calculated by comparing population groups with higher and lower scores on the OECD equivalence scale: the top 50% of the sample with the bottom 50% and the top 10% with the bottom 10%. The ORs between SES were taken as the ratio between the likelihood of having CVD or CVRFs among the poorer households (comparing the poorest with CVRFs or CVD with those without CVRFs or CVD) in relation to the better-off (comparing the better-off with CVRFs or CVD with those without CVRFs or CVD), a ratio of >1 reflecting an association between morbidity and lower SES, a ratio of <1 showing an association between morbidity and higher SES, and a ratio of 1 (or the CI containing 1) indicating no significant association.18

The aim of this study was to analyze the distribution of CVD according to SES in the Portuguese population. The most common forms of CVD were stroke and IHD (37.31% and 34.26%, respectively). Stroke, IHD and CVD were associated with lower income groups, rates of stroke showing the greatest socioeconomic inequality; inequalities tended to increase at the extremes of household income distribution.

There have been few studies on economic inequalities in cardiovascular morbidity, and no study of this kind in Portugal was found in the literature. One study of socioeconomic inequalities in cardiovascular mortality19 used indices for occupational (divided into manual and non-manual work) and educational groups based on mortality rates for stroke, IHD and CVD and ORs for risk factors according to SES, and concluded that mortality is higher due to stroke and CVD in those with lower SES and due to IHD in those with higher SES. These findings are not comparable with those of the present study since it used a different methodology, assessed mortality rather than morbidity and covered a different period (1980–1982). However, it is interesting to note that our study revealed higher morbidity due to IHD in the lower socioeconomic groups, with less inequality for stroke, in contrast to the above study based on mortality data.

Our study demonstrated socioeconomic inequalities in cardiovascular morbidity. Other studies have shown similar differences in Portugal for morbidity in general, based on SES indicators such as education, occupation and income and using various data sources, including the NIH.20–27

It should be noted that the present study did not set out to analyze cause and effect relations between CVRFs, CVD and SES.

With regard to CVRFs, smoking was more strongly associated with higher SES, as reported by various other authors.19,28,29

The ORs for psychological distress did not reveal significant inequality according to SES. A meta-analysis of 56 studies on socioeconomic inequality and depression30 showed a wide range of indicators, methodologies and results. In most studies, an OR of >1 for depression, and thus a greater frequency, was observed in those with lower SES. However, apart from the fact that SES was assessed by education, occupation, wealth or social class in most studies and by income in only six, they covered a long period of time (published between 1980 and 2001) and obtained conflicting results (ORs ranging between 0.53 and 7.09). One study of epidemiologic evidence for the relation between SES and depression31 found a social gradient, those with lower educational levels being more likely to be affected. These findings conflict with our results, but our decision to assess psychological distress through MHI-5 and SES through the OECD equivalence scale, together with the fact that no similar studies have been conducted in Portugal, makes it difficult to put our results into context.

In our study population 24.22% of individuals had an MHI-5 score of ≤52, reflecting psychological distress. Studies assessing mental health status in other countries have shown similar results to the present study, which indicates that while the distribution of mental health disorders may vary within a population, results are generally similar between populations with different characteristics. One study of the value of MHI-3 and MHI-5 in screening for depressive symptoms in Japan32 concluded that 23% of the general population had depressive symptoms. Another measuring the mental health of the Dutch population using the MHI-533 concluded that 21% had some form of mental disorder.

The OR values for the risk factors HTN, DM and obesity showed an association with lower SES, in line with the Eurothine Final Report,34 which found a greater prevalence of DM and obesity in groups with lower income and educational level, the study by Mackenbach et al.,19 which revealed a higher prevalence of obesity in these groups, a study by Hardy35 which found a greater prevalence of HTN in those with lower SES, and studies by Marmot and colleagues,36,37 which found a higher prevalence of physical inactivity and HTN in those with lower SES as assessed by employment grade.

The relative frequencies of HTN, DM and obesity were similar in the present study to those reported in the literature. HTN was found in 31% of the sample, in agreement with other studies in the Portuguese population38 which showed an overall prevalence of 24%, ranging between 22% and 36% according to sociodemographic characteristics and geographical location, while DM was found in 6.52% of individuals, which is also in agreement with other studies in the Portuguese population39 that revealed a prevalence of 8.9%, ranging between 6.5% and 17.4%.

In the present study, obesity was found in 19.65% of the sample and pre-obesity in 42.29%, which corroborates other studies40 that recorded levels of obesity and pre-obesity of 14% and 40%, respectively in the Portuguese population.

Physical inactivity did not show significant socioeconomic inequality for OR (50%), but OR (10%) did reveal significant inequality, being associated with lower SES. This finding is supported by studies such as Eurothine34 which found a greater prevalence of physical inactivity among the poorest sections of society. However, the low level of physical inactivity in the present study (36.03%) was not seen in various studies.2,41 This may be due to the fact that the NIH questions concerning physical activity were only applied to 24.40% of the total sample.

Analysis of the distribution of medical visits in the three months prior to the interviews showed that individuals with CVD or IHD were more likely to seek medical attention than those with stroke (OR 4.91, 3.91 and 2.81, respectively), which may indicate that less social importance is attributed to stroke.

No association was found between the distribution of medical visits and SES, which is paradoxical since diseases are more prevalent among the poor but the number of medical visits does not reflect this. This would indicate possible inequalities in the use or provision of health care services in favor of the better-off, as shown in a study by Pereira,42 who concluded that there was inequality in provision in favor of those with higher SES, irrespective of the morbidity indicator chosen.

The results obtained highlight the burden of cardiovascular morbidity and the complexity of the various health factors that affect it. The study also reinforces the findings of the WHO regarding the Portuguese National Health Plan.43 CVD and CVRFs related to lifestyle (healthy or otherwise) present significant socioeconomic inequalities in prevalence, reflecting the fact that health policies are still failing to recognize, monitor and correct socioeconomic inequalities in terms of behaviors and lifestyle, as well as in terms of treatment and rehabilitation. Besides effective medical intervention, comprehensive and targeted strategies are required for health promotion and disease prevention that take account of individual, cultural and socioeconomic characteristics.

We propose that the current National Health Plan should continue to consider equality and CVD among its priorities, but should establish quantifiable targets for evaluating, monitoring, detecting and correcting any departure from the strategies planned.

CVD and most CVRFs are more prevalent among the poor, which suggests that SES may affect the behaviors and lifestyle that are responsible for CVD. In general, the better-off are more likely to smoke, while the poor have a higher prevalence of HTN, DM and obesity. Given that modifiable CVRFs are mainly responsible for the onset, evolution and complications of CVD and that they are subject to socioeconomic inequalities, it is strange that current health policies do not address this issue and seem incapable of establishing specific programs according to individual socioeconomic characteristics.

The authors have no conflict of interest to declare.