Bulletin of the World Health Organization

Estimating the cardiovascular mortality burden attributable to the European Common Agricultural Policy on dietary saturated fats

Ffion Lloyd-Williams a, Martin O’Flaherty a, Modi Mwatsama b, Christopher Birt b, Robin Ireland b, Simon Capewell a

Introduction

Cardiovascular diseases (CVDs) are the main cause of death in Europe, accounting for 49% of all deaths (and 30% of all premature deaths before the age of 65).1 Although age-specific mortality rates from CVDs have halved in western Europe in the last 20 years, the prevalence of CVD is actually increasing due to an ageing population.1 CVD is estimated to cost the European Union (EU) €169 billion annually.2 Apart from smoking, the main risk factors for CVD are raised cholesterol and blood pressure.3 Diet thus plays a dominant role in promoting or preventing CVD.

Policy decisions made at the European level can impact directly and indirectly on food availability and consumption at the national level. The EU Common Agricultural Policy (CAP) has had a major influence on agriculture and nutrition across Europe, not least through increasing the availability and consumption of products containing saturated fats. Currently, the CAP annual budget is approximately €45 billion, representing around 45% of the overall EU budget.4

The original CAP objectives are firstly to ensure an adequate supply of food to the population, and secondly to prevent rural poverty.4,5 However, direct financial support to farmers who produced milk and beef plus subsidies4 resulted in “mountains” and “lakes” of unsold food and drink, which the European Commission (EC) has subsequently been attempting to reduce through several CAP reforms. The EC then needed to dispose of this excess produce, principally as fats hidden in processed foods.6,7

EU support for the dairy industry exceeds €16 billion, including €500 million per year on domestic consumption aid for butter alone. This is equivalent to 1.5 kg per EU citizen per year, or 4 g per day.8 The school milk subsidy scheme introduced by the EC9 likewise means that a child drinking full-fat rather than skimmed milk will consume an additional 1.5 kg of saturated fat every year approximately 4 g per day.10 British children obtain 23% of their daily saturated fat intake from full-fat milk.11,12

These full-fat dairy products are a significant source of saturated fat to the population, potentially increasing coronary heart disease (CHD) and obesity. Although some studies have suggested that consumption of full-fat milk does not increase the risk of coronary death, these have methodological limitations.13,14

Therefore CAP, while established on the basis of sound public health principles, may now have become a hazard to public health throughout the EU and may be promoting inequalities in health through the types of food consumed. This might controversially be described as “a system designed to kill Europeans through CHD”.15

The large reduction in CHD mortality in North Karelia and throughout Finland principally reflected a decrease in mean population serum cholesterol level. Reduction in consumption of dairy fat in the Finnish population contributed substantially to this cholesterol reduction16 and the total energy intake from saturated fats fell from 21% in 1972 to 14% in 1997, with a compensatory increase in polyunsaturated fats from 3.5% to 5%.17 More recently, Poland reported a 7% reduction in saturated fat consumption in 10 years.18 Furthermore, Lock and Pomerleau estimated that everyone eating the minimum recommended level of 400 g of fruit and vegetables per person per day would prevent approximately 7% CHD and 4% of strokes, representing 50 000 deaths per year in the pre-2004 EU countries.19,20

We therefore aimed to estimate the burden of CVD as a result of excess dietary saturated fats attributable to CAP. We focussed on the 15 countries in the EU, prior to the 2004 enlargement, as the additional 10 European Urban Research Association (EURA) countries would not have been exposed to CAP.

Methods

Following Marshall,21 we developed a spreadsheet model to synthesize data on population, diet, cholesterol levels, and cardiovascular mortality rates and risk factor levels. Average consumption of saturated fat across the 15 EU countries is 13.1% (Table 1), higher than the population goals of less than 10% of energy consumption. Existing evidence indicates that dietary changes can result in a population reduction in saturated fat consumption. In Finland, saturated fat consumption fell by 5% in 15 years.22 Having observed the substantial 5% and 7% reductions in saturated fat consumption in Finland and Poland, we chose a conservative reduction of just 1%. We hypothesized that without CAP subsidies for dairy products, (e.g. butter, full-fat milk), per capita saturated fat consumption would have been 1% lower (2.2 g less), and that monounsaturate and polyunsaturate intake would each have been 0.5% higher (reflecting a compensatory increase in vegetable oils).

Using Clarke’s equation,23 this would decrease serum cholesterol by approximately 0.063 mmol/l. Law’s24 meta-analysis was used to estimate the resulting fall in CHD deaths, using age and sex-specific values. The most recent year for reporting the number of CHD and stroke deaths in the 15 EU countries was obtained from WHO.25

The number of cardiovascular deaths attributable to CAP was then calculated by multiplying the predicted change in the CHD death rate, for the 0.063 mmol/l cholesterol fall, by the actual number of CHD deaths in Europe. For example, the change in CHD death rate for cholesterol fall in men aged 65–74 years = 0.021 x 52663 = 1104 deaths.

A similar procedure was then followed for calculating stroke deaths. The stroke mortality burden attributable to CAP was calculated using the 10% fall per 1.0 mmol/l fall in cholesterol low-density lipoprotein (LDL)24 quantified in the Law 2003 meta-analysis.26 The overall change in stroke mortality for a 0.063 mmol/l fall in cholesterol was therefore approximately 0.6% in both men and women (min. 0.3%, max. 0.9%).26

A probabilistic sensitivity analysis was then conducted. We performed a Monte Carlo simulation allowing the parameters based on the Clarke equation and Law meta-analysis to vary stochastically. We generated 1000 iterations of the calculations for the numbers of deaths attributable to CAP. We then calculated point estimates and 95% confidence intervals. Results were stratified by age and sex, and individual EU country.

Results

In 2000, the 15 EU member states reported 588 490 coronary heart disease deaths and 391 020 stroke deaths per annum.

The stated dietary intake assumptions (1% less saturated fat, 0.5% more monounsaturated and 0.5% more polyunsaturated fat) would have lowered blood cholesterol levels by approximately 0.06 mmol/l. This in turn would have resulted in some 9822 fewer CHD deaths (minimum estimate 1265, maximum estimate 11 050) and 3024 fewer stroke deaths (min. estimate 790, max. estimate 3794) each year. Of this total, 4388 (min. estimate 578, max. estimate 4939) were CHD premature deaths (under 75 years) and 607 (min. estimate 471, max. estimate 906) were stroke premature deaths (Table 2 and Table 3).

Table 3 provides an overview of CHD and stroke mortality across the 15 EU countries by age. Approximately half the specific mortality in men and women attributable to CAP was premature, occurring below the age of 75 years.

Fig. 1, Fig. 2, Fig. 3 and Fig. 4 show the excess deaths for each country; the burden was greatest in France, Germany, Italy, Spain and the United Kingdom. The data is presented in Table 4 (available at: http://www.who.int/bulletin/volumes/86/7/08-053728/en/index.html).

Fig. 1. Male CHD mortality attributable to CAP in 15 EU countriesa
CAP, Common Agricultural Policy; CHD, coronary heart disease
a Sensitivity analysis showing best, maximum and minimum estimates.
Fig. 2. Female CHD mortality attributable to CAP in 15 EU countriesa
CAP, Common Agricultural Policy; CHD, coronary heart disease
a Sensitivity analysis showing best, maximum and minimum estimates.
Fig. 3. Male stroke mortality attributable to CAP in 15 EU countriesa
CAP, Common Agricultural Policy; CHD, coronary heart disease
a Sensitivity analysis showing best, maximum and minimum estimates.
Fig. 4. Female stroke mortality attributable to CAP in 15 EU countriesa
CAP, Common Agricultural Policy; CHD, coronary heart disease
a Sensitivity analysis showing best, maximum and minimum estimates.

Discussion

This is perhaps the first study to quantify the impact of the CAP subsidies for dairy and meat commodities on cardiovascular mortality. The estimated mortality contribution attributable to CAP was approximately 9800 additional CHD deaths and 3000 additional stroke deaths within the EU, half of them premature. These results were robust in the sensitivity analysis. Furthermore, these were very conservative estimates, assuming a 1% reduction in saturated fat energy intake, rather than the 5% and 7% observed in Finland and Poland. By applying data from the Nurses Health Study conducted in the United States of America, the true mortality burden may be higher still.27 Reducing the number of CHD and stroke deaths by primary prevention at the population level is obviously preferential to measures targeted at individuals at high risk or secondary prevention therapies. The importance of reducing the consumption of saturated fat, leading to reduced cholesterol levels is also well established.24 This analysis contributes to the current wider debate concerning the relationship between CAP, health and chronic disease across Europe together with recent international developments and commitments to reduce chronic diseases.28–30 It complements the findings of Joffe & Robertson31 and Pomerleau & Lock19,20 on the role of fruit and vegetables in the diet. However, consideration of how policy may directly impact upon a reduction in CHD and stroke deaths is currently underresearched.

Recent evidence of the potentially powerful impact of reducing dietary saturated fats is graphically illustrated by the recent large falls in CHD mortality in Poland, between 1990 and 2002 (by 38% in men and 42% in women). This reduction across socioeconomic groups was attributed to the abolition of national food subsidies for saturated fats and the emergence of new, competitive markets, greatly increasing consumption of polyunsaturated vegetable oils. Ironically, this beneficial decline could now be threatened as Poland implements CAP after joining the EU in 2004.32

Our results suggest that changes in CAP subsides would particularly benefit France, Germany, Italy, Spain and the UK. Furthermore, the “French Paradox” of high saturated fat consumption but apparently low CVD levels may reflect both undercertification of ischaemic heart disease deaths, and a time lag effect due to previous low animal fat consumption.33 Eurohealth recently devoted a whole edition to CAP health issues, recommending CAP reforms to reduce consumption of saturated fats and increase consumption of vegetable oils, fruit and vegetables.34

There are limitations in this study. The data quality for CHD mortality and the year of latest available statistics from EU countries varied within countries. Furthermore, although CAP subsidies have a potentially powerful effect on markets, other factors may contribute, including the role of marketing and the globalization of food cultures (such as fast food consumption and supermarkets). Although we assumed that lag times were minimal, this is consistent with the very rapid changes seen in Poland, and also in statin trials. We also used a relatively simple methodology which would benefit from further refinements, for instance, to model socioeconomic variables. However, we based our calculations on very conservative estimates; the actual CVD mortality attributable to the CAP could be even greater. Furthermore, the methodology was transparent and easily replicated.

In conclusion, CAP reforms are urgently required. ■

Competing interests: None declared.

References

Affiliations

  • Division of Public Health, University of Liverpool, Whelan Building, Quadrangle, Liverpool L69 3GB, England.
  • Heart of Mersey, Liverpool, England.