Cardiovascular disease is the leading cause of death worldwide. Elevated low-density lipoprotein (LDL) cholesterol levels, hypertension, diabetes and smoking are key modifiable risk factors.1,2 Published practice guidelines3–10 recommend aggressive treatment of individual risk factors and have defined thresholds for the initiation of and goals for treatment (see Table 1). These treatment recommendations are based on the results of rigorously controlled clinical studies, with often perfect study logistics and infrastructure and with multiple control visits of the patients. Under these circumstances, aggressively treating these risk factors has resulted in significant reduction of risk. However, whether these ambitious treatment goals can be transferred into ‘real life’ is unclear.
Both physician surveys and studies in clinical practice suggest that first, the prevalence of cardiovascular disease and often multiple risk factors is high, and second, in primary care adherence to guidelines and achievement of treatment goals are low.11–34 The EUROASPIRE II survey showed a high prevalence of uncontrolled hyperlipidaemia and hypertension (54 and 59%, respectively).11 Approximately 20% of patients with coronary artery disease (CAD) had a diagnosis of diabetes in this survey, and the prevalence of concurrent risk factors was high: smoking 17%, obesity 43%, hypertension 57% and elevated total cholesterol levels 55%. The adverse lifestyle trends observed were accompanied by a similar lack of improvement in blood pressure management and the observation that the majority of patients with CAD did not achieve the cholesterol treatment goals. A recent survey of European primary care physicians also showed that treatment guidelines are incompletely implemented in primary care, even though the physicians know about the guidelines and agree with them.18
In the Minnesota Heart Survey, the mean prevalence of hypercholesterolaemia in 2000–2002 was 54.9% for men and 46.5% for women, although lipid-lowering drug use significantly increased compared with previous years. In addition, the percentage of patients achieving cholesterol control (defined as a concentration <5.18mmol/l) was low for both men and women.19 Even in high-risk patients, screening for treatable risk factors is rare (e.g. plasma lipid levels are frequently not determined),20,23 risk factor management is poor in patients receiving treatment, most patients do not achieve recommended treatment goals for any of the known risk factors26 and patients with more risk factors are less likely to achieve treatment goals.
A high prevalence of uncontrolled hyperlipidaemia in patients with known CAD has been described previously despite widespread use of multiple lipid-lowering agents.13,32 In some of these studies, baseline levels of LDL cholesterol were an independent predictor of achieving target lipid levels.32 In a cross-sectional study in primary care, the prevalence of hypertension was 35%, hyperlipidaemia 29%, diabetes 14% and coronary heart disease (CHD) 12%. Only 11% of all dyslipidaemic patients achieved LDL cholesterol treatment goals. In patients with diabetes, glycated haemoglobin (HbA1c) values ≥7.0% were recorded in 40%, and hypertension was present in 45%.33 In one German study in patients with primary hypercholesterolaemia starting therapy in clinical practice,34 35% of men and 0.5% of women without manifest atherosclerosis had a high (>20%) 10-year CHD risk using the Prospective Cardiovascular Munster Study (PROCAM) algorithm,35 and only 7% of the high-risk men and 5% of the high-risk women achieved the LDL cholesterol target goals with treatment. In addition, 65% of men and 72% of women treated for secondary prevention had LDL cholesterol levels >4.2mmol (100mg/dl), and after nine months of statin therapy only 21% of the men and 17% of women reached the target LDL cholesterol levels.
We recently showed in a large study36 in 110 primary care (general practitioner and internal medicine) offices all over Germany that routinely documented data from the physician patient management systems – the International Classification of Diseases (ICD)-10 diagnostic codes, measurements of systolic and diastolic blood pressure, total cholesterol, LDL cholesterol, high-density lipoprotein (HDL) cholesterol, triglycerides, HbA1c and information regarding smoking and use of antihypertensive, antidiabetic and lipid-lowering medication – can be retrieved and used to assess quality of care. Over a seven-year period, 284,096 patients with cardiovascular disease and/or diabetes were identified (39.7% of the total patients seen during this time period). Mean age was 56±17 years, 53% of the patients were female and the median observation period was 1,061 days. Of these 284,096 patients, 157,101 (55.3%) had hypertension, 64,205 (22.6%) had CAD, 83,005 (29.2%) had diabetes and 174,787 (61.5%) had hyperlipidaemia. There was significant overlap between individual risk factors: 48% of patients had at least one additional risk factor or co-morbidity, and 3% had all four.
During the first office visit, treatment goals for total cholesterol, LDL cholesterol, blood pressure and HbA1c were achieved in 10, 24, 23 and 31%, respectively; these numbers were 9, 29, 28 and 36%, respectively, at the last visit (see Figure 1). Using the Framingham risk model,37 20% of the patients without known CAD had a high (>20%) 10-year risk for cardiovascular events based on the values during the last office visit. In contrast, the number of high-risk patients would have been <5% if the treatment goals for all risk factors had been achieved during the last office visit. Patients with only one diagnosis (e.g. hyperlipidaemia only or hypertension only) were seen quite commonly in this study.36 Even without concomitant CAD, these patients may still need aggressive treatment,38 and even high-normal blood pressure may be associated with a higher risk of cardiovascular disease.39,40 However, the high prevalence in our study of patients with more than one risk factor is particularly alarming because additive effects on outcome have been suggested.41,42 Low achievement of treatment goals was also seen in high-risk patients (i.e. those with CAD and/or diabetes), <10% and 14% of whom had total cholesterol and LDL cholesterol levels within the desired range at the first visit, respectively.
The reason(s) for not achieving treatment goals in primary care remain to be determined and most likely include suboptimal therapy (under-use of therapy, such as antihypertensive drugs and statins, and inadequate dose/lack of titration), concern regarding drug side effects, lack of time, prescribing costs and patient non-compliance.43–45 Prevention efforts include both lifestyle interventions (healthy diet, tobacco-free lifestyle, maintaining normal bodyweight and exercise) and pharmacological interventions, and medical intervention should not be limited to patients at high total risk. With any long-term treatment, but particularly in chronic disease with little symptoms (i.e. hypertension), patient compliance becomes key to the long-term success of therapy. Here, combination therapy to reduce the overall number of pills seems most promising. While some healthcare systems (e.g. Germany’s) put the emphasis on keeping (drug) therapy costs low, other systems (e.g. Great Britain’s) provide financial rewards when treatment targets are achieved, and the UK’s Pay-for-Performance Programme has been able to improve the management of patients with cardiovascular risk factors.46 These differences in healthcare systems and reimbursement need to be considered when European comparisons are being made.
Physicians may not be aware of or may ignore the treatment thresholds and goals set out in the guidelines. This is suggested by the observation in our study36 that more than half of the patients with hyperlipidaemia neither carried the appropriate diagnosis nor received treatment.
All of these components need to be considered before improvements in the percentage of patients achieving treatment goals can be expected.
Conclusions
- Almost 40% of patients in primary care are seen for cardiovascular disease or diabetes, and the majority do not achieve treatment goals, even if they are high-risk patients.
- Even with cautious interpretation of the data, the economic consequences of the observed low adherence to guidelines are most likely immense: for example, a one-fifth reduction of the five-year incidence of major coronary events, coronary revascularisation and stroke is observed per 1mmol/l reduction in LDL cholesterol.47
- Despite the known limitations in using the Framingham risk model,48–50 these calculations suggest that therapy targeted at threshold levels of blood pressure, LDL cholesterol and glucose would have a great impact on reducing cardiovascular disease,36 and this benefit would most likely not be limited to patients with extremely high risk levels.38–42
- Major changes (e.g. public education campaigns, screening programmes, collaboration of different healthcare disciplines [nurses, diet specialists, family practitioners and speciality physicians], reimbursement changes emphasising pay-for-performance and expansion of healthcare coverage) will be required to achieve better implementation of guidelines and achievement of their treatment goals.
- Primary care physicians are in a position to screen and identify high-risk patients and to start appropriate therapy: the majority (>75%) of patients with CAD are treated in primary care, and routine practice data can be used to assess achievement of treatment goals36,51 and may help to solve (some aspects of) the implementation problem.
Conflict of Interest
JCG has received speaker/consulting fees and/or study support from AstraZeneca, 3M Medica, MEDA Pharma, Merck, Sharp&Dome, Novartis, Pfizer pharmaceuticals and SanofiAventis.
Acknowledgements
The author would like to thank Dr Sven Cassens and Mathias Brosz for helpful discussions on the manuscript.