The European Journal of Public Health Advance Access originally published online on August 10, 2005
The European Journal of Public Health 2006 16(2):217-222; doi:10.1093/eurpub/cki062
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Miscellaneous |
Trends in risk factor control in Germany 1984–1998: high blood pressure and total cholesterol
Ulrich Laaser and Juergen Breckenkamp** Section of International Public Health, Faculty of Health Sciences, University of Bielefeld, Bielefeld, Germany
Correspondence: Dr Juergen Breckenkamp, MPH, MSE, Section of International Public Health, Faculty of Health Sciences, University of Bielefeld, PO Box 10 01 31, D-33501 Bielefeld, Germany, tel: +49 521 106 5166, fax: +49 521 106 6009, e-mail: juergen.breckenkamp{at}uni-bielefeld.de
Received March 1, 2004, accepted October 12, 2004
 |
Abstract |
|---|
 Top Abstract MethodsResultsDiscussionReferences |
|---|
Background: Diagnosis and treatment of the two primary cardiovascular risk factors, hypertension and hypercholesterolaemia, are well established. Nevertheless, according to earlier analyses of representative health questionnaire and examination surveys in 1984, 1988 and 1991, control of risk factors in the sense of normalized values through drug therapy did not improve to any relevant degree in former West Germany. The National Health Survey of 1998 now allows the reconsideration of the hypothesis that medical treatment has been improving and lead to a reduction of risk factor values measured in the population. Methods: Datasets of independent cross-sectional studies in 1984, 1988, 1991 and 1998 with net random sample sizes between 3458 and 5335 were analysed for actual (persons with elevated values and persons successfully treated) and population (persons with elevated values) prevalence, awareness of the risk factors under question, treatment coverage (risk factor aware and treated) and effectiveness (risk factor aware, treated and normalized), and the resulting parameters of controlled (successfully treated persons among actual prevalence) and uncontrolled prevalence (persons with elevated values among actual prevalence), respectively. Thresholds chosen were blood pressure values
160/95 mmHg for hypertension and values 250 mg/dl for hypercholesterolaemia. Regarding medication, the answer of ‘one to two times weekly’ or more was considered to indicate a relevant drug intake. Results: For hypertension the population prevalence (population 30–69 years old) increased significantly (P < 0.0001) from 19.6% to 24.0% between 1984 and 1998, whereas the actual prevalence rose less steeply but still significantly (P < 0.0002) from 32.5% to 34.4%. For hypercholesterolaemia the population prevalence stagnated at 37.0% (1998), whereas the actual prevalence was 47.5% in 1998 (39.1% in 1984; P < 0.0001). For hypertension treatment, coverage improved from 45.4% to 63.0%, but treatment effectiveness decreased from 51.7% to 41.3%, both trends being highly significant. For hypercholesterolaemia, awareness increased from 18.3% to 57.6%, but treatment coverage decreased from 33.5% to 15.5%, whereas treatment effectiveness improved from 23.8% to 47.7%, all trends being highly significant (P < 0.0001). Conclusions: The results do not support the hypothesis that medical care for the large population at cardiovascular risk in (Western) Germany was adequate and successful in the 1980s and 1990s.
Keywords: hypercholesterolaemia, hypertension, medical prevention, national health surveys, risk factors
Elevated values of blood pressure and total cholesterol strongly contribute to the risk of cardiovascular diseases, as do some other risk factors such as increased body mass, diabetes mellitus and smoking.1–5 In almost all countries of the world, population samples have been examined to analyse the prevalence of cardiovascular risk factors, to determine the trends of their development over time and to evaluate the effects of intervention programs. With regard to hypertension and hypercholesterolaemia, the two most relevant pathophysiological factors, the regular health-care system has long had effective procedures for diagnosis and treatment (e.g.6–8). Therefore, given the enormous burden of morbidity and mortality resulting from elevated blood pressures and cholesterol, one should expect an almost complete coverage by and effective implementation of control schemes. In the Global Burden of Disease Study,9 hypertension was estimated to account for 4.7% of the total burden of disease in the developed regions of the world in 1990. For Germany, the burden due to stroke is estimated to be almost 3.7 million disability adjusted life years.10
In the general German population the awareness of risk seems to be high11 and the medical profession generally claims a successful approach to health care delivery. Nevertheless, the WHO World Health Report12 has placed Germany only 22nd in a ranking of countries. This has not lead to the expected major discussion in Germany.13 Given this situation, it seems to be more than appropriate to re-examine earlier analyses14–16 of the German National Health Surveys to find out whether at least a positive trend of risk factor control through medical treatment can be stated.
|
Methods |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
Study population
Data from the National Health Surveys (NHS) of the German Cardiovascular Prevention (GCP) study were available for analysis. Methods and results of the GCP study have been described elsewhere.17 The NHS uses representative samples of the 25- to 69-year-old population with German citizenship and a permanent domicile in former West Germany and West Berlin in 1984, 1988 and 1991, respectively. The independent cross-sectional studies comprise net random samples of n = 4790 (1984), n = 5335 (1988) and N = 5311 (1991), collected from between 100 and 200 sample points. The corresponding response rates were 66.0, 71.4 and 69.0%, respectively.18 Data relating to the 30- to 69-year-old participants were used for analysis.
Extensive questionnaires (92 questions with 780 possible responses) form the basis of these datasets. These questionnaire data were supplemented by the measurement of physiological variables including blood pressure, total cholesterol, HDL cholesterol, body size, body weight, thiocyanate, glucose and uric acid.18,19
The objective of the Federal Health Survey (FHS) in 1998 was the establishment of a routine instrument for health monitoring at the federal level, which now also includes East Germany.20 Apart from the 18- to 79-year-old German population, the FHS also covered foreign citizens with a main domicile in Germany, who spoke German sufficiently. The net sample size of this cross-sectional study was 3458 (former West Germany) according to a response rate of 60.2%. Results are presented for a subset of the 1998 population, corresponding to the selection criteria of the previous surveys. Former East Germans and foreigners were excluded from analysis of the West German population.
The sample size of the 1998 subset is 
25% smaller than the three other samples (see table 2).
Assessment of risk factors and parameters of control
The technology of investigation was essentially identical for the first three surveys. The second measurement of blood pressure on the right arm (sitting position) and the fifth phase of diastolic blood pressure were used to determine hypertension.
The same methodology was adopted in the fourth survey in 1998. However, while in the earlier surveys two measurements were performed at an interval of 5 min,21 in 1998 three measurements were accomplished at intervals of 3 min. For analysis, we used the second measurement of the surveys in 1984, 1988 and 1991 and the average between the second and third measurement in 1998 in order to keep the time interval between the first measurement and the values used for analysis approximately equal (because of the well known observation that each subsequent measurement of blood pressure tends to be lower than the one before).22 Participants with missing measurements of blood pressure or cholesterol were excluded from the analysis.
Although the 1998 questionnaire was similar to those used in the earlier surveys, the formulation of some questions, especially regarding hypertension, was changed. In a previous publication15 only the answer ‘I have now’ was rated as hypertension or hypercholesterolaemia, whereas in the analyses presented here an extended definition is used in order to establish comparability across all four surveys: The answers ‘I don't have any more’ and ‘I don't know whether I still have’, given in the earlier surveys, are rated as positive answers assuming that these individuals would have answered ‘yes’ if confronted with the wording of the 1998 questionnaire. The answers ‘I don't have’ (1984, 1988, 1991) and ‘no’, respectively, ‘I don't know’ (1998), as well as missing values (with the exception of confirmed treatment, see below), are rated as negative answers, which is compatible with a conservative estimate of prevalences. Because the questionnaires merge hypercholesterolaemia and hyperlipidemia into one category, it had to be ignored that in some cases, elevated blood lipids may require treatment even without comprising hypercholesterolaemia.
Concerning the question on drug intake, only a dosage of ‘one to two times weekly’ or more is considered as relevant medication and rated as positive in all four surveys.
Hypertension is defined as a blood pressure 160 mmHg systolic and/or 95 mmHg diastolic (fifth phase), and hypercholesterolaemia at values 250 mg/dl. For participants with positive answers regarding medical treatment, a known hypertension or hypercholesterolaemia is assumed in those cases where appropriate data in the questionnaire are missing (the percentage of this constellation, e.g. for hypertension in the 1984 survey is 2.1%).
The parameters were computed on the basis of an r x c table (see table 1), using identical arrangements for the four surveys and both risk factors. It is of particular importance to differentiate between the epidemiologically relevant ‘actual prevalence’ and the ‘population prevalence’. The actual prevalence is defined as the percentage of all persons with hypertension and hypercholesterolaemia, including those with successfully treated and therefore normalized values. Compared with this, the population prevalence describes the occurrence of elevated blood pressures or total cholesterol values within a population, including those who are treated ineffectively. Whereas the population prevalence is of importance with regard to the medical supply (ineffective prescriptions, lack of compliance), the actual prevalence is a result of all factors causally effective in the pathogenesis of hypertension or hypercholesterolaemia. This includes a genetic disposition as well as unhealthy ways of living.
|
|
In view of the conceptual definition of ‘treatment’ (category III: treatment coverage; see table 1), in principle both pharmacological prescription and behavioural changes have to be considered. However, the latter play a limited role. In all, 362 of 609 participants, aware of hypertension but not treated for it, show normal or borderline blood pressures in 1991. This corresponds to 20.0% of the actual prevalence of 1522 cases and describes the maximum potential of behavioural treatment, neglecting incorrect answers and spontaneous normalization of blood pressure. For the purpose of this analysis, we defined treatment as pharmacological treatment and excluded alternative behavioural interventions and spontaneous remissions from the definition (category IV and V: treatment effective and controlled prevalence).
Statistical analysis
As the age and sex distributions differ in the four surveys the age-adjusted prevalence of risk factors was calculated by weighted means of 5-year age groups and sex-specific proportions. Weights are chosen according to the age- and sex-distribution of the subset in 1998. Computations were marginally influenced.
The Cochrane Armitage Trend Test (asymptotic two-sided test statistics) was applied for trend analyses on presented prevalences.
Logistic regression analyses were performed to examine effects of possible confounders such as overweight [body mass index (BMI) 30], physical inactivity (<2 h of sports/week), currently smoking, diabetes mellitus, last physician attendance, and hypertension or hypercholesteriolaemia, respectively. After modelling the base models, a backward elimination with P = 0.175 was performed for the final models. Age, sex and phases of studies were included in each model (slstay = 5). All analyses were performed with the statistical software SAS 8.2.
|
Results |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
The actual prevalence of hypertension did increase significantly (see corresponding P value for trend) from 32.5% to 34.4% during the period of 1984 to 1998; for hypercholesterolaemia the actual prevalence also rose significantly from 39.1% to 47.5%. The population prevalence of elevated blood pressure, however, after a slight decrease from 19.6% to 19.1% between 1984 and 1991, rose significantly to 24.0% in 1998, whereas there was no significant trend during this period for hypercholesterolaemia. Altogether, the percentage of the actual prevalence effectively controlled stayed below one in five (17.6% in 1998) for hypertension and below one in 20 (4.3% in 1998) for hypercholesterolaemia.
These deteriorating or at best stagnating trends of the population prevalence can be explained for hypertension by the lack of improvement in detection of high blood pressure and treatment effectiveness, which offsets some improvement in the percentage of hypertensive patients treated (from 45.4% to 63.0%; P value for trend <0.0001); and for hypercholesterolaemia awareness and treatment effectiveness improved significantly (up to 57.6% and 47.7%, respectively, in 1998), but from a lower level (18.3% and 23.8%, respectively, in 1984). In contradiction to hypertension, for hypercholesterolaemia the treatment coverage deteriorated significantly (from 33.5% to 15.5%; P value for trend <0.0001).
In addition to the deficits in medical treatment, lifestyle also does not seem to have improved during this period. The potential for behavioural and/or spontaneous remissions was defined by the difference of controlled + uncontrolled prevalence to 100%, which for hypertension remained constantly at a level between 23.0% and 24.0% between 1984 and 1991, and fell to 12.5% in 1998. For hypercholesterolaemia this difference improved up to 1991 from 5.1% to 5.7%, and increased in 1998 to 17.7%.
The computed odds ratios confirm the results described above (see table 3).
|
|
Discussion |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
The values of the population prevalence for hypertension and hypercholesterolaemia from 1984 to 1991 were almost identical to those published earlier,15 although the host of the database (Robert-Koch-Institute, Berlin, Germany) has revised the file accessible to the public in the meantime.
Possible inconsistencies between the surveys of 1984–1991 and the FHS of 1998 may be due to the different formulation of the relevant questions in the surveys of 1984–1991 as compared with 1998. If the different formulation is taken into account as in this analysis, the answers ‘no, do not have any more’ and ‘I don't know whether I still have’ are to be interpreted as indicating formerly elevated risk factor levels (extended definition). This inflates the weighted actual prevalence (see Methods) as compared with earlier publications without the corresponding weight,15 by e.g. 26.0% in the case of hypertension (from 26.9% originally to 34.0% in 1991), and therefore the subsequent percentages as calculated from this reference (table 3). No changes can be seen for hypercholesterolaemia (from 42.5% originally to 42.3% in 1991). However, the general trend of the actual prevalence as well as the population prevalence remains basically unchanged for both parameters, with a significant positive (increasing) tendency for the actual prevalence of both factors and for the population prevalence of hypertension.
The different combinations of the trends for awareness as well as treatment coverage and effectiveness can explain at least partly the unfavourable development of increasing population prevalences for both hypertension and hypercholesterolaemia. For hypertension, all gains in treatment coverage due to the decade long campaigns of organizations like the German League For Hypertension are offset by the loss of treatment effectiveness (see table 3), which indicates a serious deficit of continuing medical education that obviously has been focused instead mainly by the sponsoring by the pharmaceutical industry of the treatment of hypercholesterolaemia. Here, treatment effectiveness improved considerably, but as awareness jumped by a factor of two (57.6% in 1998) treatment coverage could not be maintained, which indicates a missing link between cholesterol screening (mainly in pharmacies) and medical care. As a consequence of these trends, which more or less neutralize each other, the population prevalence has been increasing, partly supported by the increase of the actual prevalence of both risk factors, indicating a lack of life style improvement and ineffective health education of the population.
Using the borderline definition of hypertension (140/90 mmHg) the results for actual prevalence (49.3%, 50.0%, 49.2% and 50.1%, respectively) and population (43.3%, 43.5%, 42.8% and 45.2%, respectively) prevalence differ from those described above. This is owing to the by definition higher number of participants with hypertension. For the new definition, in contrast to the definition used in our main analysis, no trend has been found for the actual prevalence. For the population prevalence a tendency of increasing values over time can be seen. Overall the analysis with border values of 140/90 mmHg shows the same tendencies as the analysis with our more conservative approach.
Modifications of lifestyle such as weight reduction, limitation of alcohol intake, increased physical activity, avoidance of tobacco and reduced sodium intake are recommended to reduce cardiovascular risk factors and to lower the incidence of cardiovascular disease.23 These risk behaviours also affect blood pressure24–28 and, partially, fat metabolism.29
It is therefore of interest to seek explanations looking at the trends for the associated factor of overweight (BMI 30). Not only was there no effect of the intervention programme in the GCP study, but in the reference population an almost continuous increase in the number of overweight individuals between 1984 and 1998, from 17.4% to 21.3% in males and from 17.9% to 22.8% in females, could be demonstrated (unpublished data). This probably does not explain trends in awareness, treatment coverage and effectiveness, but it is consistent with the trend of the actual prevalence of hypertension and hypercholesterolaemia.
The same applies to the risk factors smoking and physical inactivity. While the rate of smokers decreased for males from 39.5% in 1992 to 37.4% in 1998, for females the rate increased from 26.7% to 29.0% during the same period.30 Generally, daily life in Germany can be described as life with movement-poor activities. Between 1992 and 1998 there was only a small rise of physical activity, defined as >2 h sports per week regularly.31 As it is the case for obesity, these risk factors are consistent with the trend for the actual prevalence of hypertension and elevated total cholesterol.
Altogether, these results do not support the hypothesis that medical care for cardiovascular risk in (Western) Germany is adequate and successful or that lifestyle has improved in the 1980s and 1990s. Neither the work of different organizations for health information or health education, nor the preventive work of public health insurance companies, public health offices and general practitioners has shown recognizable effects.
From the perspective of population-based monitoring the decline of cardiovascular mortality in former West Germany over the last decades, often cited as proof of success for primary and secondary prevention, may be better explained by other factors than effective reduction of manifest risk factors, i.e. of hypertension or hypercholesterolaemia.
Key points
|
|
Acknowledgments |
|---|
The authors are grateful to Dr Eleanor Hill (PhD, Med, BSocSci, SCM, SRN) for language editing.
|
References |
|---|
|
TopAbstractMethodsResultsDiscussionReferences |
|---|
1 Pooling Project Research Group. Relationship of blood pressure, serum cholesterol, smoking habit, relative weight and ECG abnormalities to incidence of major coronary events: final report of the Pooling Project. J Chron Dis 1978;31:201–306.[CrossRef][Web of Science][Medline]
2 Murabito JM, D'Agostino RB, Silbershatz H, Wilson PWF. Intermittent claudication. A risk profile from the Framingham Heart Study. Circulation 1997;96:44–9.
3 D'Agostino RB, Wolf PA, Belanger AJ, Kannel WB. Stroke risk profile: adjustment for antihypertensive medication. The Framingham Study. Stroke 1994;25:40–3.[Abstract]
4 Anderson KM, Wilson PWF, Odell PM, Kannel WB. An updated coronary risk profile. A statement for health professionals. Circulation 1991;83:356–62.
5 ERICA Research Group. The CHD risk-map of Europe. Eur Heart J 1988;9(Suppl 1):1–36.
6 Veterans Administration Cooperative Study Group on Antihypertensive Agents. Effects of treatment on morbidity in hypertension. I. Results in patients with diastolic blood pressures averaging 115 through 129 mmHg. JAMA 1967;202:1028–34.
7 Veterans Administration Cooperative Study Group on Antihypertensive Agents. Effects of treatment on morbidity in hypertension. II. Results in patients with diastolic blood pressures averaging 90 through 114 mmHg. JAMA 1970;213:1143–52.
8 West of Scotland Coronary Prevention Study Group. Prevention of coronary heart disease with pravastatin in men with hypercholesterolemia. N Engl J Med 1995;333:1301–7.
9 Murray CJL, Lopez AD. The global burden of disease. Geneva: WHO, 1996.
10 Yassin, KY. The burden of cerebrovascular disease in Germany. In: Laaser U, editor. The development of stroke morbidity and mortality and their most important determinants: a scenario analysis. Report to the German research council (DFG), University of Bielefeld, 2002 (unpublished).
11 Breckenkamp J, Kottmeier A. Früherkennung Schlaganfall. Early detection of stroke. First nation wide campaign. Verlag Hans Jacob, 1999.
12 World Health Organization. The World Health Report 2000. Health systems: improving performance. Geneva: WHO, 2000.
13 Möller J, Laaser U, Guentert B. Evaluation of health care systems, contributions made by ‘The WHO World Health Report 2000’. Z Gesundheitswissenschaften 2002;10:316–25.
14 Laaser U, Lemke-Goliasch P, Schumann V, Heuermann S. Effective treatment of hypertension and hypercholesterolemia still insufficiently: National Health Survey of the GCP. Z Präventivmed Gesundheitsförd 1991;3:21–4.
15 Laaser U, Lemke-Goliasch P, Schumann V, Kaftan S, Hellmeier W. Treatment and control of primary cardiovascular risk factors in the West German states. Results of the epidemiological monitoring through the National Health Surveys 1984/85–1988–1990/91. Z Gesundheitswissenschaften 1993;1:35–46.
16 Laaser U, Breckenkamp J, Allhoff PG. Epidemiological evidence in treatment of cardiovascular risk factors in Germany. Z ärztl Fortbild Qual sich 1999;93:395–402.
17 GCP Study Group. The German Cardiovascular Prevention Study (GCP): design and methods. Eur Heart J 1988;9:1058–66.
18 Forschungsverbund DHP, editor. The German Cardiovascular Prevention Study. Bern, Göttingen, Toronto, Seattle: Huber-Verlag, 1998.
19 Hoffmeister H, Mensink GBM, Stolzenberg H. National trends in risk factors for cardioavscular disease in Germany. Prev Med 1994;23:197–205.[CrossRef][Web of Science][Medline]
20 Bellach B-M, Knopf H, Thefeld W. The Federal Health Survey 1997/98. Gesundheitswesen 1998;60(Suppl 2):59–68.
21 Thamm M. Blood pressure in Germany—update review of state of trends. Gesundheitswesen 1999;61(Suppl 2):90–3.
22 Laaser U. Quality controls of blood pressure measurements in the Cologne study. Acta Cardiol 1978;XXXIII:80–2.
23 National Institutes of Health. The sixth report of the Joint National Committee on prevention, detection, evaluation and treatment of high blood pressure. Publication No. 98-4080. NIH, 1997.
24 Elliott P, Stamler J, Nichols R, et al. Intersalt revisited: further analyses of 24 hour sodium excretion and blood pressure within and across populations. Intersalt Cooperative Research Group. BMJ 1996;312:1249–53.
25 Greenberg G, Thompson SG, Brennan PJ. The relationship between smoking and the response to anti-hypertensive treatment in mild hypertensives in the Medical Research Council's trial of treatment. Int J Epidemiol 1987;16:25–30.
26 Paffenbarger RS Jr, Hyde RT, Wing AL, et al. The association of changes in physical-activity level and other lifestyle characteristics with mortality among men. N Engl J Med 1993;328:538–45.
27 Stamler J, Caggiula AW, Grandits GA. Relation of body mass and alcohol, nutrient, fiber, and caffeine intakes to blood pressure in the special intervention and usual care groups in the Multiple Risk Factor Intervention Trial. Am J Clin Nutr 1997;65(1 Suppl):338–65.
28 Whelton PK, Kumanyika SK, Cook NR, et al. Efficacy of nonpharmacologic interventions in adults with high-normal blood pressure: results from phase 1 of the Trials of Hypertension Prevention. Trials of Hypertension Prevention Collaborative Research Group. Am J Clin Nutr 1997;65(2 Suppl):652–60.
29 National Institutes of Health. Third report of the National Cholesterol Education Program (NCEP) Expert Panel on detection, evaluation, and treatment of high blood cholesterol in adults (adult treatment panel III). Executive Summary. Publication No. 01-3670. NIH, 2001.
30 Junge B, Nagel M. Smoking in Germany. Gesundheitswesen 1999;61(Suppl 2):121–5.
31 Mensink GBM. Physical activity. Gesundheitswesen 1999;61(Suppl 2):126–31.
CiteULike 
Connotea 
Del.icio.us What's this?
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||