Background
Cardiovascular disease (CVD) accounts for approximately one of every three deaths in the United States, and is estimated to cost $312 billion annually [
1]. The American Heart Association (AHA), in its recently released ‘Strategic Impact Goal Through 2020 and Beyond’ defined levels of four health behaviors (not smoking, engaging in sufficient physical activity, consuming a healthy diet, and body mass index (BMI) less than 25 kg/m
2), and three health factors (optimal total cholesterol, blood pressure (BP), and fasting blood glucose), to identify ideal cardiovascular health [
2]. A number of investigators have used this construct and reported low prevalences of ideal cardiovascular health in US samples [
3‐
5]. Furthermore, negative correlations between ideal cardiovascular health and all-cause and CVD mortality [
4,
6], and cardiovascular events [
4,
5,
7,
8] have been identified. Our literature search showed little evidence that attention has been paid to this concept outside of the USA. Moreover, comparisons between the prevalence of poor, intermediate and ideal cardiovascular health levels at study sites in the USA and other countries have not been undertaken.
In order to make a cross-national comparison, we used data collected from two prominent studies with similar data on CVD risk factors, health behaviors, and demographic variables: the Maine-Syracuse Longitudinal Study (MSLS) in the United States, and the Observation of Cardiovascular Risk Factors in Luxembourg (ORISCAV-LUX). The MSLS was conducted in Syracuse, New York (NY), USA and its catchment area (Central NY). ORISCAV-LUX was a nation-wide, population-based study. Several considerations make these specific comparisons important and meaningful. First, existing multinational data collection efforts (e.g., Health and Retirement Study in the USA/Survey of Health, Ageing and Retirement in Europe) do not measure key aspects of cardiovascular health (e.g., diet, total cholesterol, BP) and we know of no nationally representative data set that permit comparisons on these indices. Second, the MSLS and ORISCAV-LUX collect measures in essentially identical fashions, and both samples are representative of their respective geographic areas. Data collected from Luxembourg are of particular interest due to the multinational nature of its population, with a large number of individuals originating from neighboring European countries living in Luxembourg. As in the US, CVD is the number one cause of mortality in Luxembourg, with CVD related illnesses accounting for 33.8% of all deaths in 2011 [
9].
Our overall objective was to compare the pattern of cardiovascular health between two geographically and culturally distinct sites in the US and in Luxembourg. Specifically, the first objective was to compare the sites with respect to the total number of health metrics at ideal levels, indexed by a global Cardiovascular Health Score (CHS), derived from the AHA construct of cardiovascular health. The second objective was to examine the prevalence of poor, intermediate, and ideal health for each health behavior (smoking, physical activity, diet, BMI), and health factor (total cholesterol, BP, and fasting blood glucose) at each site. The third objective was to assess whether any observed differences in the cardiovascular health components between the two study sites remained after controlling for socioeconomic indicators. As European countries have higher rates of walking, cycling, and active transportation than in the United States [
10], and a number of studies have shown an inverse association between active transportation and overweight/obesity, BP, and triglyceride and fasting insulin levels [
10‐
15], we hypothesised that the CHS would be higher in the Luxembourg site than in the Central NY site. With regard to the individual health metrics, we postulated that the prevalence of ideal levels for physical activity and BMI would also be higher in Luxembourg than Central NY.
Discussion
The present study is, to our knowledge, the first to make cross-national comparisons using the AHA-defined components of cardiovascular health. Although we are unable to generalize beyond the two geographic study sites in Central NY, USA and Luxembourg, the study provides insight into how cardiovascular health differs between two sites in the US and Europe. Luxembourg, a centrally located European country with a large proportion of the population coming from Portugal and neighboring countries including Belgium, Germany and France, serves as a good representation of western Europe. Syracuse and its surrounding counties are ethnically diverse, and constitute the economic and educational hub of Central NY state.
The overall CHS, generated from the individual health metrics, was higher at the Luxembourg site than at the Central NY site. Ideal levels for BMI, smoking, physical activity, and diet were more prevalent in Luxembourg than in Central NY. However, ideal levels for BP were more prevalent in Central NY. Differences between the two sites with respect to BMI, physical activity, diet and BP cannot be attributed to age, gender, income and education as findings remained after adjustment for these potential confounders.
Importantly, the prevalence of overall ideal cardiovascular health (ideal levels for all components) was low at both the Central NY site (0.4%) and the Luxembourg site (1.0%). This is consistent with smaller state-based studies and national data in the USA, with prevalences ranging from 0–0.1% [
1,
3,
5] to 1.2% [
4]. Of concern are the national (US) data indicating that the prevalence of ideal cardiovascular health decreased from 2.0% in 1988–1994 to 1.2% in 2005–2010 [
4]. Over the same time, increases in physical inactivity and obesity, and decreases in fruit and vegetable consumption have been observed [
37]. The biggest prevalence differences between the two present studies were observed for the RFS. Poor health for this score in the Central NY sample (34.5% of participants) was nearly three times higher than in Luxembourg (11.8% of participants).
The role of physical activity in weight control is well established. While the age-adjusted proportion of participants not engaging in any physical activity was similar in both study sites (10-12%), the mean time spent engaging in physical activity per week was over two times higher in the Luxembourg site than Central NY, equating to a difference of approximately 6.5 hours per week. Adjusting for BMI, waist circumference and waist/hip ratio in secondary analyses did not change the results (data not shown). This is consistent with data showing that Europeans walk and cycle over two and four times, respectively, the number of kilometres per person per year, than residents of the United States [
10]. Furthermore, active transportation via walking or cycling is also more common in Europe than in North America and the lowest estimates of adult obesity are found in countries that rely more upon active transportation and less upon automobiles [
10,
11]. As per intuition, higher rates of walking and cycling as a means of transport have also been associated with a higher percentage of adults meeting the recommended levels of physical activity, as well as lower estimates of diabetes [
11]. The infrastructure in Luxembourg supports walking and cycling for daily travel and may be one contributing factor that explains the difference observed in the present study. During the period of data collection for MSLS, the automobile was the predominant mode of transportation in Syracuse and Central New York with few cycling paths on city and town streets.
In contrast to these findings, average age and gender-adjusted BP was 133/84 mmHg in the Luxembourg sample compared with 123/70 mmHg in the Central NY sample. Epidemiological research utilizing national survey data from six European countries, the US and Canada, also found higher BP levels in Europe, and reported very similar levels to those in the present study: 136/83 mmHg in Europe and 127/77 mmHg in North America [
38]. Of note, less than 40% of those with diagnosed hypertension in the Luxembourg sample were being treated for high BP, compared with nearly 85% in the MSLS. Although hypertension has the same classification in the US and Europe (systolic BP ≥140 mmHg and/or diastolic BP ≥90 mmHg) [
39,
40], the initial approaches to BP control vary and we can only speculate that this may contribute to the BP differences observed [
39,
40]. Lifestyle modification and immediate initiation of antihypertensive drug therapy is recommended in the US [
40], whereas in Europe, the immediate initiation of drugs needs the presence of other symptomatic CVD risk factors [
39].
In addition, poor awareness of relatively ‘silent’, asymptomatic cardiovascular risk factors including hypertension and dyslipidemia has been demonstrated in this Luxembourg population, with 60% unaware of their diagnosed hypertension (diagnosis made from the ORISCAV-LUX survey) [
41]. This level of unawareness is two-fold that of the US; national estimates from the National Health and Nutrition Examination Survey 1999–2000 data indicate an unawareness level of 30% [
40].
There are several study limitations. ORISCAV-LUX was a community, national population-based study, whereas MSLS was a community-based sample restricted to Central NY. The MSLS sample is not nationally representative, but both Luxembourg and the Syracuse metropolitan statistical area (MSA) have similar population sizes of slightly more than half a million inhabitants. In 2008, Central NY state had an almost identical age-adjusted CVD mortality estimate as the US (244.0 versus 244.8 per 100,000 deaths [
42,
43]. In Onondanga County, NY, CVD mortality estimates have been decreasing, following a national trend [
42,
43]. While African Americans have been found to have significantly fewer ideal cardiovascular health components than whites [
3,
5], a sensitivity analysis excluding African Americans (9.7% in MSLS) did not affect the pattern of results. Smoking, diet and physical activity data were based on participant self-report and the same instruments were not used in both studies.
There are several study strengths. This is the first study to compare three levels of cardiovascular health in two studies from two different countries. We are not aware of published data using ideal cardiovascular health in countries outside of the USA. To compare prevalence estimates across these two samples with different age compositions, the effects of variation in age structure were removed by using a ‘world’ standard population to standardize age [
34]. The prevalence differences observed in these analyses were confirmed when comparisons between the two sites were made using the health metrics as continuous variables, with the added control of education and income.
The purpose of our study was solely to describe the epidemiological patterns of cardiovascular health in these two study sites. The contrasts found are of substantial magnitude, however the findings neither provide causal explanations nor effectiveness data on the health care system in either region.
Conclusions
The main finding of note in the present study is in the recognition that the majority of participants at both study sites had overall cardiovascular health scores that fell within the intermediate range, but the overall CHS was higher in Luxembourg. The differences found for the BP, diet, and physical activity metrics are particularly notable. As both Luxembourg and the New York State Department of Health have implemented public health policies to promote and maintain population cardiovascular health [
44,
45], it is difficult to attribute findings to specific health care policies, particularly as health care delivery is different in each region.
However the current findings suggest that different strategies for intervention will be important for different countries and underscore the need for cross-national comparisons. The continued improvement of education programs and focus on prevention measures may be helpful in both countries. Infrastructure to support active means of transportation may be an important consideration. Regardless of the approaches taken to achieve CVD reduction, even small health behavior changes at a population level would produce relatively large increases in the proportion of individuals in both ideal and intermediate categories [
1]. In a similar vain, small reductions in weight gain over decades may accumulate into meaningful reductions in risk for obesity-related disorders [
46]. The impact on US mortality from poor dietary habits is significant [
47], yet simple changes, such as dietary reduction of 3 g of salt per day, is projected to yield substantial reductions in mortality and health care costs [
48]. For example, the total costs of diagnosed diabetes in the US was estimated at $245 billion in 2012 [
49], and costs approximately €148 million per year in Luxembourg [
50]. Early intervention strategies to increasing physical activity, make healthier food choices, cease smoking, and lower blood sugar levels all seem important based on our findings, and may help to produce significant savings in health care costs in the long term. It may be particularly important for future studies to focus on children and young adults [
51] in relation to interventions designed to raise the AHA cardiovascular health metrics to higher levels.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
GC: conceptualization of the present study, data analyses and interpretation, manuscript drafting. ME: MSLS chief investigator, manuscript drafting and revision. NS: statistical analyses and interpretation, critical review of manuscript. AD: critical review and drafting of manuscript. JB, CD: ORISCAV-LUX investigators, critical review of manuscript. AA: ORISCAV-LUX investigator, study design, data analyses and interpretation, critical review of manuscript. All authors read and approved the final manuscript.