Coronary heart disease is one of the leading causes of death worldwide and shows considerable geographical variation [1, 2]. This geographical variation can either be due to a true difference in cardiovascular risk factor levels [3–5] or to differences in cardiovascular risk factor management [6–9], although other causes have been suggested [10].

In Switzerland, health policies are decided at the local (canton) level. Further, with four official languages, Switzerland is a country of large socio- cultural diversity, which might also be reflected in health policies and health expenditures [11]. Indeed, smoking policies differ between cantons [12], and significant regional differences in glycemic control have also been reported [13, 14]. Hence, it is likely that these differing health policies might lead to differences in cardiovascular risk factor (CVRF) screening and management.

In this study, we used the data from the Swiss Health Survey 2007 to assess the prevalence and management of CVRFs according to geographical region.

To our knowledge, this is the first study to assess the geographical differences in cardiovascular risk factor screening and management in Switzerland. Our results indicate that CVRF screening and management differ between regions and that these differences cannot be accounted for by differences in population age, gender, educational level or migrant status, suggesting that other factors such as local habits might be at stake.

On bivariate analysis, small differences in the prevalence of overweight and obesity were found between regions, but these differences were no longer significant after multivariate adjustment. Overall, our results indicate that excess BMI is evenly distributed throughout Switzerland, although a trend reversal has been shown for Zurich [17]. Hence, it will be of interest to assess regional trends in CVRFs, to assess if there are any differences.

No regional differences were found regarding the reported prevalence of smoking. It is possible that differing smoking policies between cantons within a given region might have reduced the differences between regions but, as reported above, the number of participants for some cantons was too small to draw any valid conclusion. Conversely, current smokers from East and Central Switzerland had a higher likelihood of reporting they tried to quit. Although quitting smoking might be more due to personal motivations than to medical recommendations, our results suggest that some local anti-smoking policies might induce more smokers to try quitting than others. Still, as the prevalence of former smokers in these two regions (East and Central Switzerland) was somewhat lower than in the others, further studies are needed to better assess the actual impact of local policies on smoking prevalence and trends.

No significant differences were found between regions regarding hypertension screening and prevalence. Only two thirds of participants who had been told they were hypertensive reported being treated, a value lower than observed in another Swiss population-based study [18]. Treatment rates were somewhat comparable between regions, with the exception of East Switzerland, which showed significantly lower treatment levels. A possible explanation would be differing socio-economic characteristics of this region as it has been reported in the UK [19], but this hypothesis is rather unlikely as the differences persisted after adjusting for educational level. Another explanation is that health expenditures are lower in the cantons composing this region [11, 18], which could lead to lower screening and management efforts; nevertheless, more studies are welcomed to better assess this point. Overall, our data suggest that (i) hypertension screening and management are relatively homogeneous within Swiss regions and (ii) treatment rates could be improved.

Less than half of the participants reported having their cholesterol levels screened the year before, and screening was more frequently reported in French and Italian speaking regions. Similarly, less than half of the participants who had been diagnosed with hypercholesterolemia reported being treated, a finding in agreement with a previous study [20]. Again, most German-speaking regions had significantly lower treatment levels than French or Italian-speaking ones. Overall, our data suggest a clear socio-cultural cleavage regarding screening and management of hypercholesterolemia, practitioners living in French and Italian-speaking regions being more sensitized towards this risk factor. A possible explanation might be the fact that, in Switzerland, the most used equation to assess cardiovascular risk is PROCAM [21], a German-based equation that has not been validated in women neither calibrated for the Swiss population. Conversely, the SCORE equation recommended by the European guidelines [22] which has been calibrated to the Swiss population [23] and shown to present the best cost-effectiveness [24] is seldom applied [21]. Other reasons such as lower health expenditures in the German-speaking regions [11, 18] might also intervene, but further studies are needed to better assess the rationale for these regional differences in hypercholesterolemia management and their possible consequences in cardiovascular disease rates. Overall, our results indicate that in Switzerland, (i) the German-speaking regions present lower treatment rates for hypercholesterolemia, and (ii) less than half of the participants diagnosed with hypercholesterolemia actually benefit from drug treatment. It would be of interest to implement the current European guidelines in order to improve the management of hypercholesterolemia in Switzerland.

Less than half of the participants reported having their blood glucose levels screened the year before, this percentage being higher in the South (Ticino). Similarly, less than two thirds of the participants who had been told they had diabetes reported being treated, a value lower than previously reported [25]. As for hypercholesterolemia, most German-speaking regions had significantly lower treatment levels than French or Italian-speaking ones. These findings are partly in agreement with a previous study, which showed significant regional differences in antidiabetic drug prescribing patterns and glycemic control among patients with type 2 diabetes mellitus [14]. Again, the rationale for such regional differences is not clear and might be due to differences in medical expenditures [11], or to differing practices. Overall, our data indicate that (i) management of diabetes varies according to region in Switzerland, the German-speaking regions presenting lower treatment rates, and (ii) less than two thirds of the participants diagnosed with diabetes actually benefit from drug treatment. Again, it would be of interest to apply the current guidelines on screening and management of diabetes [26–29] to optimize outcome and health expenditures.

The between-regional differences observed regarding CVD prevention might partly be due to differing local health policies or to differing health insurance systems. For instance, for most benefits covered by health insurance, tariffs are set at national level for medical goods or negotiated at cantonal level for services [30], leading to different insurance premiums between cantons and/or regions. Further, one third of the Swiss population also contracted a private supplementary health insurance, and over a thousand different supplementary health insurance products existed in 2011 [30]. Finally, a recent OECD report [30] concluded that the split governance between cantons and the federal level leading to a lack of political leadership and drive for reforms in the Swiss health system; the report also concluded that Switzerland should overcome co-ordination problems to define national policies for prevention and health promotion.

This study has some limitations. First, most data were self-reported, which could lead to an underestimation of the true prevalence of the main cardiovascular risk factors, as it has been shown that a significant percentage of the population is not aware of their status [8, 18, 20, 25]. Conversely, it is possible that more healthy-conscious subjects participated, which would increase screening and treatment rates. Nevertheless, this would not impact between-region differences and would suggest that true screening and treatment rates are actually lower than the ones presented, further increasing the urge of implementation procedures. The main strengths of this study is that it can be considered as representative of the Swiss population, it allowed to assess not only prevalence but also screening and management of the main CVRFs, and to adjust for a variety of co-factors, including educational level and medical consultations. It is also possible that some participants reporting high levels of cardiovascular risk factors might not justify being treated as their overall cardiovascular risk, as assessed by the common risk equations, might be below the treatment threshold. Still, it has been shown that non-calibrated CVD risk equations overestimate risk among Swiss [31], which would prompt Swiss GPs to treat their patients more frequently than actually needed. Further, a recent survey conducted in 66 general practices in 12 European countries showed that blood pressure, lipid and glucose control are completely inadequate with most patients not achieving the targets defined in the prevention guidelines [32]. Hence, it is likely that considerable progress can still be achieved regarding CVD prevention in Switzerland. Finally, although data obtained from Health Interview Surveys parallel those obtained using examination surveys [33], the best option would be to associate the results from a Health Interview and a Health Examination Survey, as the former allow the collection of objectively measured data. Furthermore, the implementation of a standardized Health Examination Surveys, based in questionnaires and measurements, would allow a better comparison between countries [34–36]. Still, the implementation of such a study is costly, usually leading to a smaller sample size (with possible biases regarding minorities or specific population groups), and despite recommendations [34] several European countries (Austria, Belgium, Portugal...) never conducted such a survey. In Switzerland, the ongoing 2012 National Health Survey follows a methodology similar to the previous ones² and the results will be compared to local or regional surveys based on objectively measured data.

In Switzerland, cardiovascular risk factor screening and management differ between regions and these differences cannot be accounted for by differences in populations' characteristics. The impact of these differing health strategies on cardiovascular disease incidence and mortality should be further assessed.

¹ http://www.bfs.admin.ch/bfs/portal/fr/index/themen/16/04/key/approche_globale.indicator.30101.301.html?open=2,1#1 (in French)

² http://www.bfs.admin.ch/bfs/portal/fr/index/infothek/erhebungen__quellen/blank/blank/ess/01.html, assessed 7 February 2012