Obesity and type 2 diabetes are rapidly increasing all over the world and are lifestyle diseases with far-reaching health complications and reduced life expectancy [14-16]. The main cause of death in both sexes is cardiovascular diseases, but diabetes is often the unrecognized but underlying cause of vascular complications. In general, males are more likely to have myocardial infarctions compared to premenopausal women, but in the presence of prediabetes or diabetes the risk for women increases dramatically and cardiovascular mortality is very high in women with impaired glucose metabolism and in younger women with cardiovascular disease [17-20]. Cardiovascular mortality has improved less in women compared to males in the last decade and various studies have shown that women are treated differently than their male counterparts when seeking related health care, although the cause of such difference is still unclear and complex [21,22]. Blood pressure in women is higher at older ages as well as low-density lipoprotein (LDL) cholesterol despite antihypertensive and statin therapy [22,23].

Nutritional habits and the degree of physical activity play an important role in the development of overweight, diabetes and many chronic diseases [24-26]. Most evaluations of the effect of meat consumption on diseases such as obesity or cardiovascular disease differentiate between red meat and poultry, and sometimes further specify whether the meat was processed [25]. Physical activity has been associated with a graded, inverse relationship with body mass index (BMI), abdominal and visceral fat, and weight gain [27-30]. Physical inactivity is an important risk factor for all-cause and cardiovascular (CV) mortality in both men and women [31,32].

The selected health behaviours analyzed here include diet rich in meat, lifestyle low in physical activity, no regular intensive exercise, overweight and obesity, and daily smoking. The survey included the means to collect anthropometric measures, as well as questions regarding health status and health behaviours. Body Mass Index (BMI) was computed as kg/m² from self-reported height and weight measurements.

In both sexes older people tended to have to a higher proportion required schooling only and in younger subjects the proportion of subjects with high school or university degree was higher.

In women the ORs for overweight and obesity were inversely related to EL, with rates of overweight and obesity highest for EL0 (Table2). Such trends were attenuated in males. Similar results were also seen in both sexes for the health behaviours daily smoking, diet rich in meat and sedentary lifestyle, with EL showing a stronger influence on diet in women than men (Table3). Among men and women, significantly higher odds of daily smoking and diet rich in meat were seen at all levels of education lower than high school (Table3). Among both men and women the odds of having a sedentary lifestyle and doing no regular vigorous exercise were significantly higher at lower levels of education than among respondents with high school or university education (Table3).

However, in women only we also found a strong relationship between education and diabetes mellitus, and hypertension (Table2). The OR for diabetes was significantly higher for vocational school and required schooling with or without apprenticeship in women (Table2).

Overweight and obesity are dramatically increasing in all European countries and are associated with many health complications [15,26,37,38]. In comparison to other countries Austria shows a less prominent increase of overweight and obesity in both sexes [14]. In the present survey the difference between sexes was similar to data obtained in 2004 [39] with comparable rates of obese men and women and with 25 to 30% more overweight men than women.

A significant negative association between education and general obesity was also shown in other population-based studies in both sexes with a dose–response relationship from illiterate and primary education to high educational levels [2]. Furthermore, prevalence rates of both overweight and obesity were highest in women with low incomes, though such a relationship was not characteristic of the male population in Europe [14]. However, the inverse relation between social class and obesity in women in wealthy countries contrasts with the findings in low-income countries [40,41]. Overweight increased with increases in education levels in poor rural populations in India, where overweight is also seen as sign of wealth and health [41]. Overall, our findings are in agreement with previous reports confirming a stronger association between obesity and low socioeconomic status as proxy indicator of education or education levels in women in Europe.

As with obesity, diabetes is a growing problem worldwide with significant social and economic impact. In Austria at least 5% of the population is estimated to be affected [42]. In the U.S., in counties with a high risk of diabetes, thirty percent of excess risk was ascribed to obesity and sedentary lifestyle while 37% was attributed to non-modifiable risk factors such as age, gender, ethnicity and education [43]. The Multi-Ethnic Study of Atherosclerosis revealed inverse socioeconomic gradients in hypertension, diabetes, overweight and smoking in particular in white and black women and in white men, although in the latter the associations were weaker, but stronger for education than income [44]. In non-white men higher socioeconomic status was related to higher BMI, further supporting ethnic and gender differences in social patterning. A survey from Argentina reported that among women higher education was associated with better risk factor profiles including diabetes and hypertension and lower BMI in all areas but more strongly in urban areas [45]. Among men in less urban areas no association or even an adverse association was found between education and these risk factors. In Northern Italy, in men a low level of education (defined as falling into the lower of two educational categories) was related to higher BMI, prevalence of diabetes and smoking [46]. Less-educated women showed higher blood pressure and BMI and in both sexes of the low educational class a twofold increased incidence of stroke and cardiovascular disease (CVD) was observed at follow-up. However, in men CVD incidence alone was not related to education and in women higher ischemic stroke rates were observed in the more-educated group.

Of note, in our analysis stroke only was found to be related to low education in men while no association was evident in women. In another study, more physical activity at work and during leisure time was found to be associated with a lower incidence of stroke [47]. Lower socioeconomic status and less education appear to be associated with less knowledge of risk factors of stroke in both sexes [48] but in general women seem to have better knowledge of warning signs than men. Also, men more frequently mentioned stress, physical activity and smoking as risk factors of stroke, while women more often reported diabetes and hypertension, which may also be attributed to the more frequent medical visits of women.

Smoking is a leading cause of morbidity and mortality in many industrialized countries, including Austria. Various studies reported a trend towards equalisation of smoking behaviour between the sexes, showing an increase in women but a decrease in men [11]. The higher rate of smokers in the low education group in our evaluation is in line with other reports.

In a large prospective study of elderly people, red and processed meat intakes were associated with modest increases in total mortality, cancer mortality, and cardiovascular disease mortality [49]. Data on education level and red meat intakes are scarce. However the relationship between low education level and high intake of red meat fits the overall finding of unhealthier lifestyles in people with lower socioeconomic status and educational levels.

In Austria all residents have health insurance coverage; the access to health care should be comparable among groups. In addition, education is free of charge, including university studies. Quality of life and standards of health care are high in Austria. The gender gap in salaries is, however, rather high, approaching 20 to 30%, although rates of tertiary education are nowadays comparable between women and men. This trend of equalisation in tertiary education between sexes has mainly been achieved in the last decade and higher-educated women can therefore be expected to be younger than males at present, a fact that might influence the prevalence rates of diseases like diabetes or cardiovascular disease in highly educated men and women. Nevertheless, in our analysis the impact of age is negligible because all data analyses have been corrected for differences in age.

Low educational level, in particular required schooling only, is usually associated with low income in both sexes and thus lower socio-economic status may be expected in men and women with required schooling only. In Austria, as in most countries, more women than men are living below the poverty line and it has also been shown that expenditures on health care and out-of-pocket healthcare payments constitute a much greater proportion of the household income among poor people compared to the better-earning population [50]. It was shown that in particular for older persons, the lower the education the greater the burden for medical services and the lower the awareness of how to lead a healthy lifestyle, and the lower the adherence to medication and the utilisation of preventive measures. Furthermore, in women the burden for medical services was even greater than in men including income-independent disadvantages such as sex-specific illnesses. In general, in Austria as well as in other European countries women participate more often in screening programs, are more interested in health prevention and visit their general practitioners more often. These attitudes may also relate to a higher rate of diagnosis of depression and anxiety disorders. Along with biological factors including sexual hormones, in particular oestradiol, psychosocial factors, culture and education may be responsible for the sexual-dimorph patterns of these mental disorders. A recent meta-analysis confirmed that less education is generally an important risk factor for late-life depression [51]. In addition, poor self-reported health status appeared to be more strongly associated with depression than the presence of chronic disease [52]. In the Korean Longitudinal Study of Aging, which used a nationally representative sample of community-residing adults aged 45 and older, the authors found that cognitive ability, economic resources, social status, social network, and health behaviour explained all of the education gradients associated with depression [53]. Epidemiological studies indicate that the lifetime risk of depression is twice as high in women as in men, and socioeconomic status, stressful life events and biological factors contributed to higher female vulnerability and predominance of the disease [54]. Also in our study the rate of depression and anxiety disorders was twice as high in females as males and related to poor education (EL0) in women, but not in men. Therefore, our results further support the hypothesis that education – potentially reflecting cognitive ability and socioeconomic status – more strongly influences mental health in female.

This study has several limitations. Based on the cross sectional design of the study no causal explanations are possible. In addition the validity of self-reported data regarding information on dietary habits, the degree of physical activity, and smoking, as well as self-reported data about chronic diseases may have a limited reliability.